RELAPSING FEVER.

BY WILLIAM PEPPER, M.D., LL.D.

SYNONYMS.—Febris recidiva, vel recurrens; Fièvre a rechutes; Fièvre recurrente; Typhus icterodes, vel recurrens; Bilious Typhoid Fever; Rückfall's Typhus; Tifo recidivo; Famine Fever, Hunger-pest, Armentyphus, Hunger-typhus, Spirillum Fever.

DEFINITION.—Relapsing fever is an epidemic contagious disease, the specific cause of which is not certainly known, although a peculiar spirillum appears to be constantly present in the blood. It occurs chiefly among the over-crowded and destitute, but may spread widely when introduced among more favorably situated populations. Its invasion is abrupt, and is marked by a distinct chill or rigor, followed quickly by high fever (104° to 106°), with severe headache and pains in the back and limbs. Delirium is comparatively rare. The tongue is heavily coated, and there are epigastric tenderness, vomiting, constipation, and enlargement of the liver and spleen, with frequent jaundice. There is no characteristic eruption. These symptoms cease abruptly from the fifth to the seventh day, with copious sweating; but after an apyretic interval of about a week's duration a relapse occurs similar to the first attack, but of less duration (three to five days). Second, third, or even more numerous relapses may subsequently occur at less regular intervals. One attack does not protect against a second one to the same extent as with other contagious diseases. The mortality is usually small.

HISTORY AND GEOGRAPHICAL DISTRIBUTION.—It is not important to consider here at any length the history of this disease. Allusions to it were made by Strother, 1729, and by Huxham, 1752, but the first reliable account on record is the description of an epidemic in the year 1739 by John Rutty.1 Relapsing fever undoubtedly occurred at different times and at various places during the next hundred years, although the records of it are scanty, and for the most part imperfect, owing chiefly to the want of a clear recognition of its essential difference from typhus and typhoid fevers.

1 A Chronological History of the Weather and Seasons, etc., London, 1770, pp. 75-90.

During the decade from 1842 to 1852 relapsing fever prevailed in a very active and widespread form. Epidemics occurred in England, Scotland, and Ireland, in various parts of Germany, and it was during this time that it was first observed and described in America. In June, 1844, an emigrant ship from Liverpool came to America with eighteen cases on board, which were taken to the Philadelphia and Pennsylvania Hospitals. In 1848 a few cases were imported by emigrants to New York, and in 1850 to Buffalo in the same way.2

2 See Fevers, their Diagnosis, Pathology, and Treatment, Meredith Clymer, Phila., 1846, p. 99; Clinical Reports on Continued Fever, A. Flint, Phila., 1855, p. 364; Dubois 1848.

The next great outbreak of relapsing fever began in Odessa in 1863 and lasted until 1872. It prevailed in various parts of Russia, in Germany, France, and Great Britain, and for the first time occurred extensively in the United States, especially in Philadelphia and New York. The present article is based largely on a study of this epidemic as it presented itself in Philadelphia during the years 1869-70, when the writer, in conjunction with the late Edward Rhoads, had the opportunity of observing about two hundred cases, in the wards of the Philadelphia Hospital. An admirable article on the same epidemic appeared from the pen of the late John S. Parry, in the Amer. Jour. Med. Sciences, N.S., vol. lx., Oct., 1870, p. 336.

Between the years 1877 and 1880 relapsing fever occurred quite extensively at Bombay, and was there studied by Carter3 and Lewis; and during 1879-80 it prevailed in Königsberg, an account of which epidemic has been published by Meschede.4

3 Spirillum Fever, by H. Vandyke Carter, M.D., London, 1882.

4 Virchow's Archiv, Bd. lxxxvii. p. 393.

The geographical distribution of relapsing fever is seen, therefore, to have been very extensive; and not only has it occurred in the above-mentioned localities, but there have also been less extensive outbreaks in France, India, Egypt, Algeria, South America, and elsewhere.

CAUSES.—In all probability the essential cause of relapsing fever is a specific poison, but we know nothing of its real nature nor of the precise conditions under which it originates. Recent investigations have shown that the spirillum discovered by Obermeier is constantly present during the febrile stages of relapsing fever, but it cannot yet be decided whether this minute organism is the actual cause or only an invariable accompaniment of the disease.

It appears that conditions of destitution, filth, and intemperance amongst an overcrowded population favor the development of the virus, and hence the epidemics have, as a rule, begun in towns, such as Dublin, Glasgow, Odessa, St. Petersburg, Breslau, etc., where such conditions prevail. Great importance has been attached, in particular, to the scarcity of food and to destitution as powerful factors in favoring the production of the disease. Some of its names (hunger-pest, hunger-typhus, famine fever) have been given with reference to this, and in the case of several outbreaks a careful comparison has been made of the decrease of the food-supply and the consequent advance in price of the staple commodities with the development and progress of the disease. Although this is in all probability true of those centres where relapsing fever originates, it has but a partial application to the secondary centres where the disease is imported and develops.

The presence of destitution and filth, enfeebling the vitality of a section of the community, would favor the spread of this as of any other specific fever, but there is considerable evidence to favor the view that the importance of starvation as a cause of the fever has been exaggerated. This was strongly urged by Parry5 as the result of his study of the Philadelphia epidemic of 1870, and our own more extended observation showed that the vast majority of the patients appeared to be well fed. On the other hand, the influence of overcrowding as favoring the development and spread of relapsing fever has been clearly established by the study of many epidemics, as in the Breslau attack of 1868, reported by Wyss and Bock, where single tenement-houses furnished as many as seventy-one cases; in the Edinburgh epidemic of 1869 and 1870, where Muirhead found the breathing-space allotted to each individual in the affected houses to vary from 250 to 400 cubic feet; and in the Philadelphia epidemic, where the observations of Parry and ourselves showed the presence of an extreme degree of overcrowding in most of the houses where the disease broke out.

5 Loc. cit., p. 339.

No age is exempt, but neither can it be said that age exerts any influence upon the occurrence or frequency of relapsing fever. Of 1164 cases in the Philadelphia epidemic of 1869-70 in which the age was noted, the result was as follows:

Males.Females.
Under 2014976
From 20 to 30220140
From 30 to 40143101
From 40 to 5013567
From 50 to 606034
From 60 to 70206
From 70 to 9067
Total733 431 = 1164

The youngest cases were in children two or three years old; the oldest patients were women over eighty-five years old.

Sex exerts no influence, though, on account of the larger proportion of males likely to be exposed to the specific cause, the results of nearly all epidemics show a preponderance of male patients in the proportion of 33 per cent., 66 per cent., or even 85 per cent. (Meschede).

Nationality does not act as a predisposing cause,6 except in so far as certain countries may present more frequently than others the conditions favorable for the development of this disease. Of 1170 cases in Philadelphia in which the nativity was noted, 219 were Irish, 61 English, 161 German, 729 American. Of the latter 729, about one-half, or nearly 28 per cent. of the whole number, were negroes, while the negro population of Philadelphia was only about 3.3 per cent. of the total. This excessive proportion of cases among the negroes was undoubtedly due in large part to the fact that in Philadelphia overcrowding is notoriously more common and extreme among them than in any other portion of the population, although it is also likely that they present an excessive susceptibility to the virus of this as of many other specific diseases.

6 Hirsch's Geog. and Hist. Pathology, New Syd. Soc. ed., 1883, vol. i. p. 615.

Attempts have been made to show some connection between the period of the year or the atmospheric conditions and the rise and spread of epidemics of relapsing fever; but, as Murchison clearly showed, these epidemics are wholly independent of such influences. In Philadelphia, of 1176 cases in which the date of occurrence is known, there occurred in September, 1869, 4 cases; December, 1869, 6 cases; January, 1870, 5 cases; February, 1870, 13 cases; March, 1870, 124 cases; April, 1870, 209 cases; May, 325 cases; June, 293 cases; July, 115 cases; August, 19 cases; September, 28 cases; October, 15 cases; November, 1 case; December, 2 cases; January, 1881, 2 cases; February, 1 case; March, 2 cases; May, 7 cases; June, 2 cases; September, 2 cases; October, 2 cases.

Occupation exerts no predisposing influence, but in all epidemics the great majority of cases occur among the vagrant classes, who lead a precarious life and commonly sleep in foul, overcrowded lodgings. Murchison noted that in the London epidemics a considerable proportion of cases occurred among recent residents, but he attributed this, correctly, not to any special local cause, but merely to the fact that this floating population is largely of the vagrant type. In Philadelphia a careful inquiry showed that recent residence produced no special predisposing influence, and a study of other epidemics confirms this view.

Contagion is, however, the essential cause of the spread of relapsing fever when the virus has once been developed. It seems clear from the distinct periods and from the widely-separated localities in which different outbreaks of relapsing fever have occurred that its special poison is capable of being called into existence or activity by favoring conditions. Murchison held the belief that it was very intimately connected with, if not generated by, destitution, and, as already stated, much evidence exists to show that the disease is most apt to break out after periods of scarcity; but no just and convincing proof exists that destitution, any more than over-crowding and other depressing influences, can actually engender a specific contagium capable of being transported to great distances and of originating widespread outbreaks of the specific disease among differently situated populations. It appears necessary to assume the existence of some unknown special virus which finds its suitable nidus for development in the conditions attendant on filth and overcrowding, and which attacks with greatest facility the systems of those who are enfeebled by want and depressed by vitiated air. When once this specific poison has been called into active existence, however, there can be no doubt as to the fact that it can be carried by fomites, and that it is given off from the bodies of relapsing-fever patients so as to affect any who may approach. Although a few observers have doubted this contagiousness of relapsing fever, the evidence in its favor is overwhelming. In many epidemics, as in Philadelphia in 1869, its contagiousness is at least as intense as that of typhus fever. A single case may, indeed, be admitted to a healthy family among the better classes or into the wards of a well-ventilated hospital without propagating the disease, although striking cases of contagion are on record where a patient has communicated the disease to all the members of a family favorably situated and living at a distance from any other possible source of contagion. On the other hand, if admitted to an overcrowded and filthy lodging the disease is apt to spread rapidly. Wyss and Bock report seventy-one cases as having occurred in a single lodging-house during the course of the Breslau epidemic of 1868, and in Philadelphia single houses in several instances furnished more than a score of cases, and several short streets more than one hundred cases each.

In the Philadelphia Hospital twenty-three persons lying sick in the wards with other affections contracted relapsing fever from the patients admitted with that disease; two of the visiting staff, five resident physicians, and nine nurses also suffered attacks of varying severity. This corresponds with the general experience of those connected with fever hospitals during the prevalence of relapsing fever.

As in the case of typhus and other contagious diseases, the distance at which relapsing fever can be contracted by direct contagion through the atmosphere is a very short one, not exceeding a few feet at most.

The poison may be carried by fomites. Instances are on record where persons having visited infected districts have conveyed the disease to others at a distance without contracting it themselves.

When rooms which have been occupied by relapsing-fever patients are subsequently occupied by other persons, these are very liable to acquire the disease. Parry relates two remarkable cases in which relapsing fever was transported to a distance by infected clothes; and it has been more than once observed that during epidemics of this disease laundry-women engaged in washing the clothes of fever patients, but without any means of more direct communication with the sick, were frequently attacked (Cormack, Wyss and Bock).

In connection with the etiology of relapsing fever it is necessary to consider the rôle played by a minute organism which has been frequently detected in the blood of patients suffering with this disease. This spiro-bacterium was first observed in relapsing fever by Obermeier7 in 1873, and has since been identified as a spirillum or spiroechete. The very numerous observations of Obermeier, Albrecht, H. V. Carter, Motschutkoffsky, Koch, Cohen, Holsti, Enke, Meschede, and others leave no doubt that this peculiar parasite does occur at least very frequently in the blood of patients with this disease. The failure to detect it, which has been reported by several good observers, may readily have been due to the extreme delicacy of the organism, or to the neglect of the proper method of preparing the slides of blood for examination, or to delaying the examination of the blood until after death, when it rapidly disappears. Thus no value can be attached to the negative observations of Rhoads and myself, made prior to Obermeier's discovery, since our method of examination was not sufficiently exact.

7 Centralbl. f. die med. Wissensch., 1873, No. 10.

The following description of the mode of examining the blood, and of the spirillum, is condensed from H. V. Carter's account: It is necessary to employ magnifying powers of not less than 500 diameters. The fresh blood may be examined immediately after obtaining it by pricking the washed finger of the patient. For preservation dried specimens are needed: a very thin layer of fresh blood is evenly spread with the needle over the glass cover, exposed to the weak fumes of a solution of osmic acid, and allowed to dry under protection from dust; the dried film of blood may then be treated with glacial acetic acid or may be stained.

FIG. 19.
Spirillum from the blood in a case of relapsing fever, X 700 (Koch).

The spirillum [See Fig. 19] is a colorless, slender, twisted filament, which when quiescent has a length of 2.66 times the diameter of a blood-disc (1/1500 to 1/500 inch = 0.012 to 0.043 millimetre). When unfolded they become distinctly elongated. They are very narrow (not more than 1/40000 inch), and present four to ten spiral turns; when fresh they are in active movement and unfold in part, becoming wavy or bent. They resist the action of concentrated acetic acid, and are readily stained by certain dyes. In number, five or ten may be visible in a field or they may be too numerous to count. They have not been detected either in the secretions or in the evacuations. Both Koch and Carter have succeeded in cultivating this special form of bacteria outside of the body.

To judge from the observations thus far made on this difficult question, the parasite is found first toward the close of the period of inoculation or soon after the beginning of the fever, or it may be detected throughout the febrile stage; but shortly before the cessation of the fever it quickly disappears, to reappear at the time of the relapse. There would seem, therefore, to be some close connection between the febrile paroxysms and this organism, and it is not remarkable that many observers have concluded that this spirillum is the essential and specific cause of the fever, and that it is impossible to have this disease present without the appearance of the parasite in the blood; nor that the name spirillum fever has been applied to the disease by Carter.

Such conclusions appear to be premature, however, and we prefer to regard the undoubted existence of the spirillum in the blood of relapsing-fever patients as at present only an important aid in diagnosis, and to await the occurrence of other epidemics and the repetition of careful studies upon this organism, both within and without the human system, before venturing to decide whether it is merely one of the phenomena of the disease or whether it is its true cause and specific contagious principle.

It must be added that both Carter and Koch have succeeded in inoculating monkeys with relapsing fever, and Motschutkoffsky8 of Odessa, who had the opportunity of inoculating a human being, asserts that he succeeded in producing the disease, and found the incubation period to be not less than five nor more than eight days. Carter also gives an interesting table9 of six instances of inoculation, four of them by cuts while making autopsies, with consequent development of relapsing fever in each instance. Some allowance must be made for the fact that in all the instances of this series there had been exposure to contagion by close communication with fever patients, though this exposure had existed for several months previously without leading to the development of relapsing fever.

8 Centralblatt f. d. med. Wissenschaften, 1876, No. 11, p. 194.

9 Op. cit., p. 403.

GENERAL CLINICAL DESCRIPTION.—After a period of not less than five or six days from the reception of the contagion the disease begins abruptly with a chill of variable severity, accompanied by headache and aching pains in the back and limbs. The patient feels weak and is often giddy, but is not always obliged to go to bed the first day. Nausea and vomiting are among the earliest symptoms, and distress at the epigastrium, with tenderness, may attend or even precede the chill. Fever quickly follows; the pulse runs up from 110 to 130 in a few hours; the temperature reaches from 103.5° to 106° by the end of twenty-four hours; the pains increase, and there are insomnia and great restlessness; appetite fails; thirst is extreme; the tongue is moist and furred, and the bowels quiet. During the subsequent six days these symptoms persist. The temperature presents a daily remission at some period of the twenty-four hours amounting to one or two degrees, the maximum reached in fully-developed cases varying from 104° to 108°. The pulse continues very rapid, and not rarely exceeds 140; the respirations are hurried and rapid, and cough attends many cases. Delirium is rare, but insomnia, restlessness, headache, and rheumatic pains in the back and limbs may prove constantly annoying. Appetite is variable, more frequently lost; nausea and vomiting are common; thirst is very troublesome; and the bowels are constipated or loose. No characteristic eruption appears, but sudamina are frequently present, since in a large proportion of cases there is more or less sweating, even during the continuance of high fever. Abdominal pain, tenderness in the epigastrium and hypochondria, and demonstrable enlargement of the liver and spleen are almost invariable. The urine is concentrated and dark or bile-stained. Jaundice is a common symptom, though its frequency varies greatly in different epidemics. The same may be said of epistaxis.

While these symptoms are at their height and the patient is suffering severely the paroxysm suddenly ceases, and in a few hours he is entirely relieved. This remarkable crisis occurs usually at the close of the seventh day, but may occur as early as the third or as late as the fifteenth day. It is attended with a critical discharge, copious sweating being by far the most common, though diarrhoea, free epistaxis, or hemorrhage from some other surface may replace it. The patient feels weak and languid; the temperature and pulse have fallen below the normal, and remain so for a day or two. Soon there is a rapid improvement in the appetite and the appearance of the tongue, and the patient regains strength day by day, and often feels so well that it is difficult to persuade him that he must avoid exertion and exposure. The enlargement of the spleen subsides rapidly, that of the liver more gradually; epigastric tenderness subsides, but in many cases some degree of it persists for several days. This interval or apyretic period lasts about a week, when, again without warning or provocation, the patient relapses, and is seized abruptly with the same set of symptoms which attended the first attack. This relapse does not usually last more than three days (one to five are the limits), and is terminated by a similar crisis, after which a slow convalescence is entered upon, or else after an apyretic interval of some days' duration a second relapse ensues, and this may, in rare cases, be in turn followed by a third, fourth, fifth, or even sixth similar relapse. In addition, it must be noted that many serious complications are liable to occur. The total duration of the disease thus varies from eighteen to ninety days. Convalescence is often tedious, and there are many troublesome sequelæ. The mortality, however, is not great, averaging 5 or 6 per cent. Death may occur suddenly from collapse at the close of the first paroxysm or from heart-clot; it may be produced by exhaustion in protracted cases; or be hastened by any serious complication; or the patient may sink into a typhoid condition, with low delirium, coma, and suppression of urine for several days before the fatal termination.

DETAILED STUDY OF SPECIAL CONDITIONS.—It is usually difficult to determine the period of incubation. In the unique case in which Motschutkoffsky is said to have produced relapsing fever by inoculation the initial symptoms occurred seven days after the inoculation. Wyss and Bock had several good opportunities of determining the minimum period of incubation, and found it to be six days. We may assume that the ordinary period is six to eight days, but that it varies, in accordance with the virulence of the virus or the susceptibility of the system, from four to fourteen days. During this time the patient feels as well as usual, or at most suffers for a day or two from slight malaise, with vague rheumatoid pains, headache, giddiness, and anorexia. In only 13 out of 181 of our cases in which this point is noted was the invasion gradual. Examination of the blood prior to the invasion does not discover any spirilla.

The invasion is usually abrupt and during the daytime; the patient can often fix the very hour of its occurrence, a severe chill attacking him while at work or at meal-time. This is the most common initial symptom (138 out of 168 our cases of sudden invasion); less commonly, obstinate vomiting and nausea or sudden vertigo are the first symptoms (each 8 times out of 168), or violent headache (14 times out of 168), or sharp epigastric pain. Parry also observed that the occurrence of obstinate and profuse vomiting as the initial symptom was especially frequent in children.

The physiognomy is carefully noted in one hundred and seventy of our records. The countenance is often flushed, with watery eyes and anxious, suffering expression. The flush is less dingy and dull than in typhus; the eye is comparatively rarely injected; and the expression is much less dull and stupid than in that disease. In cases where grave nervous symptoms supervene and the typhoid condition is developed the facies assumes all the characteristics of that state.

The livid bronzing of the face, described by Cormack in 1843 and by Carter (Bombay epidemic of 1877), was noticed in a moderate degree in only nine of our cases, and seems to be of infrequent occurrence. When we observed it it seemed due to an admixture of a faint jaundice tinge with a deep flush. Jaundice, as already stated, is of common occurrence, though its frequency varies greatly in different epidemics. It was present in 25 per cent. of our cases, rather more frequently in the negro patients than in whites, and in degree varied from a slight tinge of the conjunctiva and skin to the deepest staining of the entire body. The presence of jaundice in combination with the general features of high fever imparts a most peculiar and alarming appearance to such patients.

With the occurrence of the crisis the flush rapidly subsides and the face becomes pale, or, if the discharges have been profuse, it may appear sunken, haggard, and almost choleraic. Parry described a peculiar puffed, velvety look at this stage, as though the skin had been much thickened and softened at the same time.

There is no characteristic eruption in relapsing fever. In 150 out of 180 cases where the condition of the skin was carefully noted there was no eruption of any kind; in 4 cases there were small roseolar spots, with peculiar subcuticular mottling, which resembled the early stages of typhus eruption, but soon faded away without becoming petechial. A similar eruption was noticed by Murchison in 8 out of 600 cases. It appears from the third to the seventh day of the first paroxysm; it may or may not recur in the relapse, or it may occur then only. Eruptions apparently similar to this have been described by others as quite common in certain epidemics. Carter describes minutely an eruption which was noted in at least 10 per cent. of his Bombay cases, the spots of which were at first small, slightly raised, and pinkish or rose-colored, and which either faded away soon or changed into purplish, more persistent stains. In a valuable report on the Königsberg epidemic of 1879-80, Meschede10 remarks that roseola was observed in cases complicated by exanthematic typhus, which prevailed simultaneously, but in no case of uncomplicated relapsing fever. While, however, this suggestion may apply to some few of the cases of eruption observed by others, it is certainly inapplicable to the vast majority of them. We also noticed an eruption of pale-reddish, slightly elevated papules in seven cases. It must be borne in mind that persons of such a low class as are the great majority of relapsing-fever patients would naturally be expected to present a variety of cutaneous eruptions from filth or vermin, and that in consequence some of the appearances above described may have been of such origin. It is certain that the bites of either mosquitoes, fleas, or bedbugs may in this disease be followed by persistent reddish papules passing into petechiæ. Apart from this, however, true petechiæ have been quite common in some epidemics, while very rare in others. Parry saw "small spots of purpura" once only, in a delicate girl; and we did not observe petechiæ once in several hundred cases, many of which had extensive internal ecchymoses. On the other hand, they have been found in as much as 30 per cent. of all cases (314 out of 1000 cases, Smith at Glasgow). They do not appear on any fixed day, but are more common in the first paroxysm than in the relapses; and although sometimes associated with a tendency to hemorrhages from other surfaces, they have been so often observed in cases of ordinary severity that scarce any unfavorable prognostic value can be attached to them.

10 Virchow's Arch., Bd. lxxxvii., p. 405.

Vibices and extensive ecchymoses of the surface are of much more grave import, and in cases where fatal sinking is threatened they may appear accompanying a purplish lividity of the countenance.

Herpetic eruptions about the mouth or nostrils were observed in 20 out of 181 of our cases in which this point is noted. They appeared usually toward the close of the febrile stage, and their development was found to have value in determining the approach of the crisis. Bärensprung mentions especially the occurrence of herpes labialis in cases of irregular relapsing fever which bore considerable resemblance to typhus. Sudamina are, as might be expected in a disease attended with so much sweating, of quite common occurrence, though much more so in some epidemics than in others, unless searched for with greater care by the one set of observers. Desquamation was noted in 42 out of 181 of our cases, and invariably at the close of the relapse. It was usually confined to the hands and face, and occurred in the form of comparatively small flakes. This is more frequent than has been the case in most epidemics. Murchison quotes a case in which a piece of epidermis ten inches square separated from the body of a lad convalescent from relapsing fever.

A peculiar odor exhaling from patients with relapsing fever has been repeatedly noticed. A description of this unpleasant symptom, given by Kelly, as quoted by Murchison,11 accords closely with what was frequently manifest in our own cases: "The smell was peculiar, not fetid or heavy, but somewhat like burning straw with a musty odor." Carter, in describing a similar odor in some of his cases, notes that the skin was not in these instances in a particularly foul state.

11 Op. cit., p. 346.

From what has already been said, it will be anticipated that the variations of the temperature in relapsing fever constitute the most peculiar and characteristic feature of that disease. A careful study of the accompanying charts will convey a more accurate impression than can be given by any description. The temperature begins to rise before the chill is fully developed, and when there is no initial chill the patient may be found within a few hours of the appearance of giddiness and headache with a temperature of 102.5° to 103.5°. Before twenty-four hours have passed it has risen to from 104° to 106°. During the paroxysm the febrile movement is continued, presenting merely a diurnal variation of one to two degrees, sometimes attended with sweating and partial relief of distressing symptoms, the minimum being observed at different hours in different cases, or even in the same case, though more frequently it occurs in the morning.

In a case reported by Parry a chill recurred at the same morning hour on three successive days. Wyss and Bock report some unusual cases in which a brief intermission occurred, with a fall of pulse and temperature to the normal, most frequently on the day before the real termination of the paroxysm. The highest temperature varies from 104.5° to 108.75°; in our cases the highest observed was 107.5°. This occurs, as a rule, on the last day or the day before the last of the initial paroxysm, and Obermeier has observed a sudden rise of four degrees in half an hour just before the crisis. Meschede,12 however, found the highest temperature on the corresponding days of the first relapse.

12 Loc. cit.

The duration of the primary paroxysm is usually six or seven days; but this is subject to considerable variations, as will be seen from the following table of 160 cases in which the duration was accurately ascertained: Initial paroxysm lasted—2 days in 1 case; 3 days in 2 cases; 4 days in 10 cases; 5 days in 19 cases; 6 days in 40 cases; 7 days in 58 cases; 8 days in 18 cases; 9 days in 2 cases; 10 days in 5 cases; 11 days in 2 cases; 14 days in 2 cases; 15 days in 1 case; and Parry, observing the same epidemic, found the duration of the first paroxysm to vary from 4 to 11 days. It is, however, rare for the duration to exceed ten days unless some complication be present.

FIG. 20.
Typical case of relapsing fever, with three relapses, terminating in recovery. (From Motschutkoffsky)

With the beginning of the crisis there is a prodigious and sudden fall of temperature, unequalled in any other condition of disease. Within a few hours it may fall six or eight degrees (going down at the rate of 1.5° or 2° an hour); and falls of 12°, 13°, or even 14.4° (Murchison), in the course of twelve hours have been noted. In our own cases the greatest fall was from 107.2° to 95°, or 12.2°; and this is as low a point as is usually reached, though temperatures of 94°, 93°, or even 92°, have repeatedly been observed. Murchison refers to one case in which collapse supervened, where the rectal temperature was 90.6°. In nearly all of our cases a subnormal temperature occurred at the crisis, and lasted for a day or two subsequently, when it gradually rose and remained normal until the relapse, unless some transient complication caused a temporary rise in the interval.

FIG. 21.
Typical case of relapsing fever (Mary Collins, aged 32), terminating in recovery. One relapse, with slight post-critical rise of temperature.

Occasionally, there is no relapse whatever, but convalescence follows the initial paroxysm. This occurred in 10 out of 181 of our cases, and Murchison found that of 2425 cases reported by various authors no relapses occurred in about 30 per cent. Carter describes these under the name of the abortive form, and found them to constitute 23.8 per cent. of all his cases. It is probable, however, that in many cases so regarded either a relapse of very transient duration has been overlooked, or else that an attack of ephemeral fever has been regarded as of specific nature. In ordinary cases the duration of the intermission averages six or seven days, but here, again, considerable variation occurs. In 139 of our cases where its duration could be accurately determined it was as follows:

3 days in 4 cases.7 days in 64 cases.11 days in 1 case.
4 days in 3 cases.8 days in 22 cases.12 days in 1 case.
5 days in 12 cases.9 days in 9 cases.13 days in 1 case.
6 days in 12 cases.10 days in 9 cases.20 days in 1 case.

Despite these variations in the duration of the initial paroxysm and of the first intermission, the average date of the occurrence of the relapse in any large series of cases is about the twelfth day from the primary chill.

The relapse is ushered in with the same striking abruptness as the initial attack. The temperature again rises rapidly to 104° or 106°, and then pursues a continuous course resembling ordinarily that of the primary paroxysm. The difference between the maximum of the two paroxysms is rarely more than 1.5° or 2°, though either may be much milder than the other; as a rule, the highest temperature is attained on the last or penultimate day of the first attack. The duration of the relapse averages three or four days, though it may last but a few hours or a single day, and yet exhibit a rise of 5°, 6°, or 7°; or, on the other hand, it may be prolonged to six, seven, or even more days. Lyons, observing the disease in the Crimea, reports some relapses as having lasted twenty-one days, though it is improbable that a greater duration than seven days occurs without the presence of some complication. The relapse usually terminates by crisis, with an abrupt fall to an abnormally low temperature; though we observed at this time, much more frequently than at the close of the first paroxysm, a gradual subsidence of fever, or lysis. Again the patient regains strength and appetite, but in a considerable proportion of cases subsequent relapses ensue. As a rule, the second, third, and later relapses are attended with a febrile movement of shorter duration and of less severity than the first two paroxysms, and are also separated by intermissions of increasing length. Meschede13 found from a study of 360 cases that the average duration was for the first paroxysm six or seven days; second paroxysm, four or five days; third paroxysm, three or four days; fourth paroxysm, one or two days; fifth paroxysm, one day.

13 Loc. cit.

In a remarkable case given in full [below], the duration of the paroxysms and intermissions were as follows:

First paroxysm, 8 days; first intermission, 9 days.
Second paroxysm, 5 days; second intermission, 1 day.
Third paroxysm, 1 day; third intermission, 6 days.
Fourth paroxysm, 6 days; fourth intermission, 8 days.
Fifth paroxysm, 5 days; fifth intermission, 9 days.
Sixth paroxysm, 4 days; sixth intermission, 10 days.
Seventh paroxysm, 3 days; seventh intermission, 11 days.
Eighth paroxysm, 3 days; followed by convalescence.

The proportion of cases in which more than a single relapse occurs appears to vary in different epidemics. Murchison found that in 1500 cases reported by various authors a second relapse occurred 109 times (1 out of 14); a third relapse, 9 times (1 out of 166); and a fourth relapse, once. Of 182 cases noted carefully by ourselves, a second relapse occurred 24 times (1 out of 7½); a third relapse, 5 times (1 out of 36); a fourth relapse, once; and in the above-mentioned case six or seven relapses.

It follows that the total duration of the morbid process varies from the average of about eighteen or twenty days, in cases with a single relapse, to forty, sixty, or even ninety days. Of course the occurrence of complications may lead to very great modifications of the febrile movement and of the total duration of the disease.

There are several additional points about the febrile process requiring mention. In all the paroxysms there is a greater tendency to local or general perspirations than is met with in other continued fevers, and occasionally there are rigors or slight chills about the same hour on several days after the invasion or on the day preceding the crisis. It has been noted also that, even when the temperature is very high, the quality of the heat, as judged by the feeling of the skin, is different from that in typhus fever, and that the peculiar pungent irritating sensation known as calor mordax is rarely marked. But a more important peculiarity is the fact that the extreme temperatures (106°, 107°, or 108°) that are frequently observed in relapsing fever for several days in succession do not appear to involve any great increase of danger, and in particular are not attended with the production of the grave nervous symptoms so often met with in connection with hyperpyrexia in typhus and typhoid, and often regarded as the direct result of the exalted temperature itself. This striking fact is of much interest in its bearing on the theory of hyperpyrexia, and may possibly be explained by some marked difference in the conditions of heat-dispersion in these different diseases.

The pulse in relapsing fever is very rapid, and on the whole the rate corresponds with the movement of the temperature. It usually rises above 110, the limits being 90 and 140, the lower rate being noticed in the milder and uncomplicated cases and in subjects of phlegmatic constitution. The pulse rises rapidly at the invasion, and may reach 120 in the course of a few hours. Its maximum is usually noticed when the temperature is highest, shortly before the crisis; and when this actually begins the pulse may fall with a rapidity as remarkable as that of the decline of the temperature. Thus, within twenty-four hours it may fall from 152 to 80, or in even a shorter time from 140 to 54, or even as low as 48 (Obermeier) or 44 (Muirheid), or even 30 (Stillé). While this great fall is often noted, it is by no means constant. In our own cases it was frequently observed that the critical fall in temperature was not accompanied by a commensurate fall in pulse. Thus, at the close of a very severe initial paroxysm lasting nine days the temperature was 107°, and fell in the course of twenty-four hours to 99°, and in twenty-four hours more to 96°; during the first day of this fall the pulse was from 96 to 100, and during the second it fell to 76.

This want of correspondence was more marked at the close of the relapse than of the primary attack; thus, in a well-marked case, where the maximum temperature (105.4°) occurred eighteen hours before the crisis of relapse, the temperature fell in four hours from 104.4° to 96.2°, while the pulse, which was 130, fell in twelve hours to 108, and in twelve more to 92. In another case, in a man aged twenty, the temperature at the close of the second relapse was 106.4°, with a pulse of only 100; after the crisis, as the temperature fell, the pulse rose to 120, and did not descend until the end of twenty-four hours; and later, at the close of thirty-six hours, the temperature was 98° and the pulse 72, lower than which it did not go. Carter14 states that in the Bombay epidemic it was invariably the case that the pulse did not decline to an extent corresponding with the temperature.

14 Op. cit., p. 140.

During the remainder of the intermission the pulse may be normal, or it may continue accelerated in consequence of some irritative condition; as the time for the relapse approaches it frequently again becomes abnormally slow. In either event it is found that any muscular exertion causes marked acceleration of the pulse.

During the paroxysm the character of the pulse is full and bounding, and there is considerable arterial tension. This is well shown in some of the sphygmographic tracings by Carter;15 while in one of our tracings from the right radial of a man æt. 32, taken on the fourth day of a severe initial paroxysm, the line of ascent is steep and the summit sharp. During the crisis, and for a day or two thereafter, the pulse may be weak, compressible, and dicrotic, and occasionally irregular.

15 Op. cit., p. 103.

The sounds of the heart and its impulse are weakened, except possibly during the first few days of the primary paroxysm. Blood-murmurs over the base of the heart and along the great vessels in relapsing fever were first noticed by Stokes, and have been frequently observed in subsequent epidemics. They were found in a large proportion of our cases, not rarely in both paroxysms, and during the early stage of convalescence when anæmia was marked; but during the intermissions they are rarely audible, and when the action of the heart was slow they were replaced by prolongation of the first sound.

It must be further noted that the pulse-rate is not a reliable indication of the danger in this disease, since, just as is the case with the hyperpyrexia, extreme rapidity of pulse may be present when the general symptoms denote no unusual danger, and when the patient ultimately recovers most satisfactorily.

There is a remarkable disproportion and dissimilarity between the cerebral and peripheral nervous phenomena in relapsing fever and those familiar to us in typhus and typhoid fevers. We have seen that patients almost invariably complain of headache. When prodromes are present it is commonly among them, and it may be the initial symptom to usher in each paroxysm. When the attack is fully developed headache is usually very severe, and no symptom is more bitterly complained of. It varies in seat and character. More commonly it is frontal or general; occasionally we found it occipital, and still more rarely it was unilateral, constituting hemicrania. It rarely continues during the relapse. Headache of an equally acute and violent character may be present in typhoid, but the headache of typhus is much more dull and contusive.

The mental condition is only exceptionally affected, a circumstance which greatly increases the patient's perception of his sufferings. Delirium is not present in ordinary cases, even though very severe and attended with hyperpyrexia; or if present is limited to the period immediately preceding the crisis, when there may be violent and noisy delirium of transient character. In some of our cases forcible restraint was necessary under these circumstances.

There are numerous instances on record showing the abruptness with which noisy, demonstrative, or even destructive delirium may appear, and the equal suddenness with which in the course of a few hours, or even of fifteen minutes, the patient may become rational and composed. Such attacks resemble hysteroidal spells, and probably occur more readily in patients of a nervous or hysterical temperament. They were certainly more common when the patients had been of intemperate habits; and, further, we had opportunities of noting that the occurrence of relapses in habitual drunkards who had previously suffered with delirium tremens was apt to develop a form of delirium which was to all appearance of that nature.

Delirium of a different and much more grave type may appear in connection with the symptoms of the typhoid state. In some cases this results from the presence of serious complications which induce a state of great prostration, while in others it is associated with great diminution or entire suppression of urine. The delirium under these circumstances is apt to be low and muttering, with a tendency to pass into stupor or profound coma.

Vertigo is present more frequently and in a more persistent form than in any other febrile disease. It was noticed as among the occasional prodromes, and was especially severe for the first few days of the initial paroxysm, though it often continued throughout this stage and recurred with the relapse. Occasionally it was complained of in the recumbent position, but usually it was excited only by a change of position.

Wakefulness was one of the most distressing symptoms in all cases, and appears to have been noted in all epidemics. Although the severity of the pain in various parts of the body and the absence of blunting of the perceptions would naturally cause much loss of sleep, the degree of the insomnia and the obstinate resistance it offers to the action of anodynes are apparently far in excess of what could thus be accounted for. Parry found that several of his patients could take as much as three grains of opium every second hour throughout the afternoon and night without either inducing sleep or causing contraction of the pupils.

Convulsions are rare and of very grave import. They may occur at the period just preceding crisis, when the nervous irritation is most intense, and are then somewhat less indicative of a fatal result than if occurring in the course of the paroxysm, when they are apt to be associated with extreme prostration of the nervous centres, with a tendency to subsequent fatal coma. No connection has been observed between their occurrence and the presence of albumen in the urine.

General tremor is rare, and was observed only in those of our cases where there had been habitual intemperance, with presumably a tendency to delirium tremens. Muscular rigidity was noticed occasionally, but may have been only apparent, being induced by the hyperæsthesia and soreness which were marked in some cases. The hyperæsthesia which was observed was both cutaneous and muscular, and was attended with tenderness of the body of the muscle, and also of the nerve-trunk supplying it. Meschede speaks of opisthotonos as a rare complication in his cases.

Motor paralysis involving single muscles or groups of muscles is occasionally noticed, as of the deltoid or of one arm (Meschede). Parry observed transient loss of power of the extremities in several cases, chiefly during the intermission or the period of convalescence. In one of our cases temporary hemiplegia occurred, with partial loss of sensation on the affected side.

The bladder and rectum are rarely affected, except in cases where the typhoid state with tendency to coma is present. Disorders of sensation are, however, much more common. When motor palsy occurs the affected part may also be the seat of impaired sensibility, while in a large proportion of all cases numbness of the extremities, with or without a sense of tingling, is complained of; out of 182 cases we noted this symptom in 94, affecting the fingers alone in 62, the feet alone in 6, and all the extremities in 25 cases. Cutaneous hyperæsthesia or partial anæsthesia are also occasionally observed. But the most noteworthy and constant symptom of this class are the pains in the muscles and joints which are bitterly complained of by nearly all patients with relapsing fever. They constitute, indeed, one of the highly characteristic features of the disease, and possess a diagnostic value. They may occur among the rarely present prodromes, but usually they appear with the chill and increase in intensity during the paroxysm; they may persist with even greater severity during the intermission, or, if they have then subsided, recur with the relapse, and may constitute one of the most troublesome hindrances to convalescence. It will thus be seen that in frequency, severity, and persistency they differ widely from the aching pains in the extremities complained of in typhus and other specific fevers. They are one of the most potent causes of the extreme insomnia, and are apt to dwell in the mind of the patient so vividly that he dreads each relapse on this account, and consequently looks back upon his attack of relapsing fever as a terribly painful experience. These pains are usually described as rheumatic in character, and several times patients presenting themselves at the hospital on the second or third day of the initial paroxysm stated that they had inflammatory rheumatism. As a fact, we observed the utmost intensity of these pains in a few cases where the patients were of marked rheumatic diathesis. The nape of the neck, the muscles of the trunk or extremities, or the large or small joints, or lower parts of the spinal region, may be the seat. At times they extend along the course of nerve-trunks. In character they are described as a deep intense aching, with occasional severe or excruciating, sharp, lancinating pains. Pressure or movement increases them. The joints are not red or swollen (though swelling may appear as a sequel), and the pains seemed to us rather to be referred to the joints than to be caused by any local irritation therein. As already stated, there is often tenderness of the body of the muscles, and this was especially marked in many of our cases on pressure along the course of the nerve-trunk.

Murchison suggests that they are due to the circulation in the blood of an abnormal substance, such as uric, lactic, or phosphoric acid; but it appears to us altogether probable that they are rather to be connected with states of congestive irritation of the sheaths of the nerve-trunks (early stage of perineuritis), or possibly in some cases of the spinal membranes also. It is true that they are sometimes shifting in their seat and fluctuating in their severity, but this is not inconsistent with the above suggestion, while the widespread irritative processes found in this remarkable disease, the resemblance of these pains and the frequently attendant numbness and tingling to the sensations caused by other forms of perineuritis, and the occasional development of local palsies of a single muscle or group of muscles, all are in its support.

The special senses are acute, sometimes painfully so. The eyes are watery and occasionally injected, but this latter condition is rare and slight in relapsing as compared with typhus fever. At the crisis and for a few days subsequently wide dilatation of the pupils is not infrequently observed. Dulness of hearing was present during the paroxysm in 14 of our cases, and a few patients complained of tinnitus; but these symptoms are not at all common in the disease, although it will be seen hereafter that affections of the middle ear are among its sequelæ.

Debility is not such a prominent symptom as in typhus and typhoid fevers. Patients manage to drag themselves about for several days during the initial paroxysm with all the symptoms fully developed, and after admission to the hospital will often be able to help themselves, or even to rise from bed, unless prevented by the severe pains or the vertigo. Still, there are many cases, not necessarily of very grave type, in which there is a marked sense of weariness and exhaustion, and of course in all cases of typhoid character the prostration is great. It must constantly be borne in mind that even when the patient feels or seems able to sit up he must on no account be permitted to do so, since the occurrence of sudden and fatal syncope is one of the accidents constantly to be apprehended. It is not only during the pyrexia that this precaution must be enforced; we meet with extreme debility during the intermission in some cases, and syncope has followed exertions made at that period as well as at others.

During the paroxysms the respirations are much accelerated, at times to a greater degree than would correspond with the pulse-rate, while at others extreme rapidity of pulse may be associated with moderate elevation of the rate of respirations.

As examples of the relation between temperature, pulse, and respirations we quote the following from our records of adult cases:

(a) Temperature, 108°; pulse, 124; respiration, 40. In the relapse; no chest trouble.

(b) Temperature, 107.5°; pulse, 120; respiration, 28; falling to temperature, 96°; pulse, 68; respiration, 18, within twelve hours, during which crisis occurred.

(c) Temperature, 107°; pulse, 144; respiration, 31. In the relapse.

(d) Temperature, 107°; pulse, 108; respiration, 44. Initial paroxysm; no pulmonary congestion.

Temperature, 106°; pulse, 116; respiration, 28. Relapse; no pulmonary congestion.

Temperature, 97°; pulse, 76; respiration, 24. Critical fall; cough, congestion of lungs posteriorly, and left one relatively dull on percussion, but pneumonia did not develop.

In many epidemics bronchitis, hypostatic congestion, and pneumonia are of rare occurrence, while in others, as in Philadelphia in 1870, they are comparatively frequent and lead to serious respiratory symptoms. While the pyrexia was high there was very frequently an irritative dry cough, with the fine crepitant and subcrepitant râles attending congestion and imperfect expansion of the lungs heard at the middle and lower portions of the chest posteriorly. In numerous instances the râles would disappear entirely after a few full inspirations in the sitting posture, just as in the corresponding condition in typhoid fever. But in a considerable proportion of all the cases (fully 35 per cent.) there was more troublesome bronchial cough, associated with sonorous, sibilant and subcrepitant râles, with mucous or muco-purulent expectoration.

Bronchitis of this character was a source of serious annoyance to many patients. In several cases there was impaired resonance at the lower margins of the lungs posteriorly, with imperfect bronchial respiration, but without the symptoms of fully-developed pneumonia. Such conditions were regarded as due to hypostatic congestion, and proved amenable to treatment. Pneumonia occurred in eleven cases out of 200 recorded with reference to this complication. It will be more fully discussed under the head of Complications. It was attended with the usual physical signs, and gave rise to extremely rapid and labored breathing, especially when associated with painful enlargement of the liver and spleen. In a case of double pneumonia, with enlarged and ruptured spleen, the respirations were from 80 to 90 for two days, the pulse being 130 to 136. It was a very fatal complication, death resulting in all but two instances.

Leyden16 has shown that though the percentage of carbonic acid in the air expired during the pyrexia is diminished, the total quantity exhaled is increased, the proportion being as 1.5 to 1 in the non-febrile state.

16 "U. d. Resp. in Fieber," Deutsch. Arch. f. klin. Med., 1870, 536, quoted by Murchison.

Elaborate investigations have been made of the condition of the urine in relapsing fever by numerous observers, and in the Philadelphia epidemic of 1870 we had the great advantage of being assisted by the distinguished chemist, the late Horace B. Hare, who conducted an extensive series of analyses in our cases. In a number of cases quantitative analyses were continued daily throughout the entire course of the disease.

As a rule, the quantity of the urine is comparatively free during the febrile periods, very scanty at the time of crisis, except in the cases where critical discharges of urine occur, and excessive for some days after the crisis.

Still, there were not rare exceptions, especially to the first of these statements. Thus on four successive days of the relapse of a severe case with delirium, but without albumen, and which ultimately recovered, the analysis gave—

Temperature.Amount in ccm.Sp. gr.Urea in Grm.Na. Cl.
103400102423.82.64
105300102515.271.95
106500102424.74.3
106 to 97850102124.7355.525

And in another severe case, also resulting in recovery, the analysis was, for two days preceding the crisis of the initial paroxysm—

Amount. Sp. gr. Urea. Na. Cl.
500 1014 12.9 Traces of albumen.
650 1014 15.85 1.365
After the crisis:
2250 1004 18.9 15.75 No albumen.

And again, in another case at the height of the initial paroxysm, within twenty-four hours of the crisis, no vomiting, purging, or epistaxis being present; temperature 105°; only 500 ccm. was passed of dark reddish colored urine, non-albuminous, and with sp. gr. 1011.

In a fatal case there was total suppression of urine for three days, the catheter drawing off only a few drops of almost pure liquid blood.

When crisis occurs by copious urination the discharges are frequent, large, and of light color and low specific gravity.

The urine of the intermissions is of similar character, and for several days after crisis it is not rare to have 2000 to 2500 ccm. passed. The largest amounts we noted were in a man who recovered, and who passed at the crisis of the relapse and during the following days the amounts here given.

Amount.Sp. gr.Urea.Na. Cl.
1000 ccm.101014.92.6
2000 ccm.100320.242.8
3550 ccm.100226.625130.995
2600 ccm.100219.2427.30
2800 ccm.100524.9622.66
2500 ccm.101347.2511.25
2700 ccm.101459.137.29

Carter reports a case where the patient continued for two weeks after the relapse to pass 130 oz. of sp. gr. 1002.6.

The amount of urea varies considerably, and is evidently under the influence of complicated conditions. The rule appears to be that it increases during the paroxysms, diminishes during the crisis, increases during the few days following crisis, and then falls off again. These results are stated upon the authority of Murchison, quoting from Pribram and Robitschek, Wyss and Bock, and others. Our own observations, however, while agreeing in the main with these, show that there are numerous and important exceptions, especially to the occurrence of the post-febrile increase in the elimination of urea.

The largest amount of urea excreted in twenty-four hours by any of our patients was 59.13 grammes, or 912 grains, on the sixth day after the end of the relapse, but as much as 74 grammes (1142 grains) have been found.

Deposits of urates were very common in the urine of the paroxysms and of the crisis. The uric acid has been found increased, and so also have the phosphates, crystals of which are frequently found mixed with the urates.

The chlorides diminish during the paroxysms, until just before the crisis their amount is very small, or they may even have disappeared. Immediately after the crisis they reappear slowly or quickly, and even very large amounts may be discharged, as seen in the figures given by Hare's analyses: 2.6 grm. on day of crisis, 42.8 grm. the following day, and the enormous amount of 130.995 grm. on the next day. A copious flow of urine corresponds with great augmentation in the amount of the chlorides.

Bile-pigment was constantly present in jaundiced cases, the amount being proportioned to the depth of the jaundice and the quantity of the urine. Bile-acids have been detected (Carter and Schmidt), and also leucin and tyrosin (Pribram and Robitschek).

Albumen, with or without tube-casts, is not uncommonly found, and traces of sugar have been detected in a few cases. More careful consideration will be given to these under the head of Complications.

The following appearance of the tongue has been repeatedly described, and when present may be regarded as possessing some diagnostic value: The body of the tongue slightly swollen, so as to show the impressions of the teeth, and by the second day the central part of the dorsum covered with a peculiarly white fur, while the edges and a small triangular space at the tip are clean and red. Such a tongue was seen in many cases at the beginning of the Philadelphia epidemic, but later it was present in but a small proportion. We find it specially mentioned in 97 of our recorded cases, or about 50 per cent., the general description being given that it was moist, rather large, with pink, clear edges, and a triangular clear space at the tip, and with heavy white fur in the centre.

Some accurate observers, as Wyss and Bock, did not notice anything peculiar about the tongue, but merely described it as moist and coated with a thick white fur. The tongue often remains moist throughout the case, the coat becoming yellowish, and later brownish. Of course if there is nasal obstruction from epistaxis or catarrh, and the patient breathes through the mouth, the tongue will soon become dry and brown; but in addition, this state of the tongue with sordes on the teeth and lips, appears in a small proportion of cases (3 per cent., Zuelzer; 12 per cent. of our own patients) in conjunction with grave typhoid symptoms.

During the intermissions the tongue clears off quite rapidly, unless marked gastric disturbance persists, but regains its former state as soon as the relapse occurs.

In rare cases the tongue is red and glazed, and Parry and ourselves observed peculiar painful cracks continuing obstinately after the relapse. It is apparent, therefore, that the tongue presents evidences of vitiated secretions, of local catarrh of the buccal mucous membranes, and of the high grade of gastric irritation so constantly attendant on this disease.

As a rule, there is complete anorexia during all of the febrile paroxysm, while in the intermission the appetite soon returns, and is sometimes truly ravenous. We did not, however, observe in any case a voracious appetite during the febrile paroxysms, such as was very often present during the London epidemic of 1843 and the Irish epidemic of 1847, and is particularly mentioned by Murchison.17

17 Op. cit., p. 360.

Thirst is constant and intense, and is excited not only by the high temperature, but by the irritation of the stomach; it may continue through the intermission, when natural appetite and the power of digesting solid food have returned.

Nausea and vomiting are always prominent symptoms, and most especially so in children. In some cases nausea occurs among the prodromes; and occasionally the attack is ushered in by profuse and uncontrollable vomiting instead of by a chill, and the stomach continues entirely non-retentive throughout the paroxysm. Vomiting is not usually so obstinate and severe, however, and with extreme care in feeding and medication it will often be allayed after two or three days. It occasionally recurs profusely immediately before the crisis, as in the case given in full [below], where after a violent attack of vomiting the patient fell asleep, and awakened in a profuse sweat.

This symptom was present in 146 out of 182 of our cases, was usually confined to the febrile stages, and was, as a rule, worse in the initial paroxysm.

The matters vomited consist of the ingesta colored with bile, of glairy mucus tinged with bile, or of green bile, sometimes in considerable quantity. Small particles of blood may occasionally be noticed in the matters vomited, and in rare instances true hematemesis occurs. Judging from the frequency with which in fatal cases we find ecchymoses of the gastric mucous membranes with blood-stained mucus in the cavity of the stomach, we should expect black vomit to be more often observed than is the case. Murchison (p. 361) states that it was not noted in any British epidemic except that of 1843, and then it occurred in only a few cases, although it seems to have varied in frequency at different places. Arrott at that time described the symptoms as "quite common" in the fever at Dundee; and W. Reid of Glasgow recorded the case of a girl in the same epidemic who vomited large quantities of clotted blood, and who also had hemorrhages from the bowels and from the ears. It has occasionally been observed in the continental epidemics. It was observed in four of our cases. By all who have observed blood-vomiting in relapsing fever it is recognized as a symptom of almost invariably fatal import. Three of the four cases in which we observed it proved fatal, but one patient, who had copious hematemesis, both at the close of the first relapse and during the second relapse, recovered after a desperate and protracted struggle.

The bowels are not so often constipated as in typhus, and it is not rare for diarrhoea and constipation to alternate, or for the bowels to be loose throughout the paroxysms. They are noted in 181 of our cases as regular in 32, loose in 61, and constipated in 88 instances. Meschede states that diarrhoea was present in nearly one-half the cases of the Königsberg epidemic of 1879, though usually as a late symptom, the early stage being marked by constipation, which in a few cases persisted throughout. The stools may be consistent and dark or thin and bilious, or occasionally, when gastric or intestinal hemorrhage has occurred, they contain black coffee-ground matter. Occasionally, the diarrhoea has a critical character, and occurs at the close either of the initial paroxysm or of the relapse, though it may not entirely substitute sweating. This mode of crisis occurred in two of our cases, but Douglas observed it in 6 out of 33 cases.

The abdomen may appear enlarged, but this is as much the result of the enlargement of the liver and spleen as of gaseous distension, which is rarely present in a high degree. Abdominal pain is almost constant, and may be very severe. It is especially mentioned as having been present in 148 out of 182 of our cases. It commonly extends throughout the epigastrium and both hypochondria, but may be present on one or the other side, while, on the other hand, there may be general abdominal soreness. It is associated with tenderness on pressure, which may be so great as to hinder the movements of the trunk and to render the descent of the diaphragm in breathing painful. This may be the first symptom to usher in the attack, and it occurs at an early stage in most cases. Many of our patients when admitted to the hospital had already been cupped or blistered over the region of the liver or spleen. This distress was greatest in cases attended with jaundice and marked gastric irritation; and Parry reports that in his cases (occurring in the early part of the epidemic which we studied) jaundice was rare (4 out of 37), and abdominal tenderness was not present. It is not difficult to explain its almost universal presence in view of the severe lesions of the substance of the liver and spleen, the distension of their capsules from the acute swelling of the organs, and the implication of the coats of the stomach.

Enlargement of the liver and spleen probably exists to a greater or less degree in every case of relapsing fever without exception. This statement is based on the concurrent testimony of accurate observers in all epidemics and upon the evidence of post-mortem examinations.

The enlargement of the liver can be demonstrated in nearly all instances by careful percussion. It varies greatly in its degree, however; in mild cases it may be slight, while in severe ones the liver may be found extending at least three inches below the margin of the ribs within three or four days from the initial symptom. In our own fatal cases the weight of the liver averaged between four and four and a half pounds.

The spleen enlarges even more rapidly and to a greater degree than the liver. In fact, its enlargement in relapsing fever is greater than in any other acute disease. It may be detected by percussion by the first or second day, and may then continue to rapidly increase until by the fifth or sixth day a large painful mass is readily recognized by palpation and percussion, or even by inspection. The organ often weighs twelve or sixteen ounces, not rarely twenty to twenty-five, and, as an instance of the extreme limit that may be reached, Küttner reports sixty-eight ounces in one case. This enlargement is greatest toward the close of the first or second paroxysm, and subsides quite rapidly in most cases during the intermissions and as convalescence progresses; we have, however, known a moderate degree of enlargement of the spleen to persist for some weeks after the crisis of the last paroxysm.

The occurrence of jaundice in a considerable proportion of cases of relapsing fever is a clinical fact of much interest. Its frequency varies greatly in different epidemics, and even at different stages of the same epidemic. At times it is rarely met with (1 out of 14, 20, or 35 cases), while in other epidemics it is present in 1 out of every 6, 5, or even 4 cases. Of 182 of our own cases jaundice is recorded in 45, or exactly in 1 out of 4. According to our observation, it occurred in a larger proportion of cases among negroes (14 out of 32) than in whites, and Stillé states that it occurred in nearly every such case that came under his observation. When present it usually occurs during the first paroxysm, and may be limited to that stage; or, again, it may be present in each of three or four successive paroxysms in the same case; or, finally, it may first appear in the relapse. As a rule, it subsides speedily after the crisis, though Carter states that in two or three cases the symptom made its first appearance just after the crisis. It varied from the slightest yellow tinge of the conjunctiva to the deepest staining of the whole surface. The urine is discolored in proportion to the intensity of the jaundice, and the serum of a blister will be deeply tinged. It must be carefully noted, however, that the feces are not decolorized, but, as already described, contain fully a normal amount of biliary coloring matter. This fact has been relied on by Murchison and others to prove that the jaundice in relapsing fever is purely dependent on the morbid state of the blood, and is not due to obstruction of the biliary passages; and we are prepared to admit that the element of blood-dyscrasia may play a part in the production of the jaundice. The anatomical evidence, however, given [below], renders it probable that in many cases at least the essential cause is to be sought in an obstructed state of the minute gall-ducts of certain areas of the liver. If the main hepatic duct or the common duct were obstructed, there would of course be paleness of the feces, as the bile would be prevented from entering the duodenum. But when a large amount of highly-colored bile is being secreted, as in relapsing fever, it seems clear that the obstruction of a certain number of minute ducts would cause sufficient resorption of the bile to induce jaundice of varying degrees of intensity, while at the same time allowing a flow of bile through the patulous ducts.

Jaundice must be regarded as an unfavorable or even a grave symptom in relapsing fever, but not to the extent that would be the case were it directly connected with the intensity of the blood-dyscrasia. Many of the most violent cases in all epidemics have been unattended with jaundice, while, on the other hand, many cases in which jaundice has been marked "have had not a single symptom that made them differ from ordinary cases excepting the yellowness" (Henderson). It follows, therefore, that the gravity of a certain proportion of the jaundiced cases does not follow directly from the presence of bile in the blood and tissues, but from the lesions of the liver of which the jaundice is a symptom, or from the existence of widespread irritation of many parts of the body. Thus jaundice is present in an unusually large proportion of the cases attended with marked enlargement and tenderness of the liver and spleen, whether vomiting is also present in extreme degree or not. It was noteworthy that it was disproportionately frequent in negroes, and that in these patients the lesions of the liver and spleen were also unusually pronounced. Again, jaundice is present in an unusually large proportion of the cases attended with low delirium, extreme prostration, defective secretion of urine, and the other features of the typhoid state—so much so that such cases have been described by various writers under the name of bilious typhoid fever.

But, as already stated, it is not legitimate to consider the gravity of these cases as the result of the jaundice, but rather that the jaundice is merely a symptom of the widespread irritative lesions, which in such cases not only involve the liver and spleen, but the kidneys, the lungs, the marrow of the bones, the muscle of the heart, and occasionally the membranes or substance of the brain and cord.

The true prognostic value of jaundice in relapsing fever would then seem to be, that of itself it indicates merely an obstructed state of a certain number of minute bile-ducts, but that its presence justifies the apprehension that the local lesions of the liver may become excessively developed, or that there is a tendency to widespread tissue-changes which at a later stage of the disease may lead to the appearance of grave constitutional disturbance of a typhoid type.

Hemorrhage in relapsing fever is not uncommon, and may occur from various surfaces. Epistaxis is, however, the only form which is frequent enough to justify being regarded as a symptom. It usually occurs in from 5 to 15 per cent. of cases of relapsing fever, but in the Philadelphia epidemic it was much more frequent than this, occurring in not less than 83 out of 182 of our cases. It was not more frequent or profuse in grave cases than in those of ordinary severity, and consequently could not be regarded as a reliable indication of the intensity of the blood-dyscrasia. Although ordinarily moderate in amount, it was occasionally so copious and persistent as to require prolonged plugging of the nostrils, and in at least one case contributed chiefly to cause an intense anæmia, which long delayed convalescence. It occurs at all periods of the paroxysms, but more commonly toward the close. In fifteen of our cases extraordinarily profuse epistaxis attended the crisis, and evidently replaced in part the copious sweating by which the paroxysm more commonly terminates.

SYMPTOMS ATTENDING THE CRISIS.—We have already described the aggravation of all the symptoms which immediately precedes the crisis in typical cases of relapsing fever, and the abrupt fall of temperature, and usually of the pulse, that follows. But this extraordinary change is nearly always attended with some profuse critical discharge, of which sweating is by far the most common, though copious epistaxis, metrorrhagia, diarrhoea, or vomiting may also occur, and to a greater or less degree, but seldom entirely, replace the sweating. In 182 cases in which we carefully noted the mode of termination of the paroxysm there was no definite crisis (termination by lysis or gradual and irregular defervescence) in 76; profuse sweating, 89; profuse epistaxis, 15; profuse diarrhoea, 2.

In most epidemics the proportion of true crises is greater than in the above table—a fact dependent upon the unusually severe and complicated form of the disease which we were studying. The beginning of the sweat may be preceded by chilliness or rigors, by extreme and dangerous prostration, or by violent nervous disturbances; or there may be an attack of profuse vomiting, followed by sleep, during which sweating begins. The sweat may be moderate in amount, but is often extraordinarily copious; the patient is literally bathed in it, the bed- and body-clothing is saturated, and we have seen the mattress saturated. It has an acid reaction, but we do not know of any accurate analyses of it. Some writers have attributed to it a characteristic disagreeable odor, but we did not notice any in our cases that could be considered peculiar to this disease.

CONVALESCENCE.—We have already stated the average duration of relapsing fever to be eighteen or twenty days, while the extreme limits are from eighteen to ninety days. Despite the fact, however, that the mortality is in most epidemics only about 5 or 7 per cent.—greatly less, therefore, than in typhus fever—the convalescence from relapsing fever is frequently slow and protracted. The obvious cause is, just as in the case of typhoid fever, the existence of numerous and serious lesions of the solids and the tendency to many troublesome complications and sequelæ. We have, however, seen many instances of rapid recovery of strength and health, even after prolonged attacks with several successive relapses.

The following case is quoted partly on account of the numerous relapses, and the long duration of the sickness:

B. B. Y., medical student, was much exposed to the contagion of relapsing fever in the wards of the Philadelphia Hospital during the spring of 1870, and in May had an attack apparently of this disease, which, however, subsided in four or five days and was followed by no immediate relapse. He continued his attendance at the hospital during the remainder of May and the whole of June; in July took a trip to the South, where there was no relapsing fever prevailing, and after exerting himself for several days during intensely hot weather, he became sleepless and much prostrated. He returned home, and after recovering from the fatigue felt quite well for about a week, until 3 A.M., August 1st, when he was attacked with a severe chill, followed by great insomnia, obstinate vomiting, intense headache, especially in the back of the neck, occasional sweating, violent fever, recurrence of very severe chill the following day at 11 A.M., epigastric and hypochondriac tenderness, decided jaundice, costive bowels, and scanty, high-colored urine. This paroxysm lasted till the morning of August 9th, when severe vomiting took place, followed by sleep, during which crisis occurred by drenching sweat lasting several hours. Appetite and strength soon began to return, though some jaundice persisted, and by August 17th he felt able to drive out a short distance, and retired feeling somewhat fatigued. He awoke with pain in the back of the neck, which continued increasing till 11 A.M., August 18th (second paroxysm), when a severe chill occurred, lasting three hours and followed by the same train of symptoms, including jaundice, which persisted five days, till Aug. 23d, when crisis again occurred by sweating. On the 24th he felt well enough to use slight exercise, which was followed by prostration and by a return of chill (third paroxysm) the next day at 11 A.M., with subsequent headache, fever, irregular sweats, etc., lasting but one day. Again felt well until Aug. 30th, when he was attacked (fourth paroxysm) at 11 A.M. with severe chill, lasting three hours, followed by severe paroxysm, lasting six days, till Sept. 5th, when crisis again occurred by sweating. Again felt well for eight days, until Sept. 13th, when the fifth paroxysm occurred, lasting five days, ending Sept. 18th by critical sweating. This was followed by an intermission of nine days, until Sept. 27th, at 11 A.M., when the sixth paroxysm occurred, lasting four days, and less severe than the preceding ones. This was followed by an intermission of ten days, till Oct. 11th, when the seventh paroxysm occurred at the same hour of the day, and lasted three days. He then went sixty miles from home to a fine, pine-bearing district, and enjoyed an intermission of eleven days, when the eighth and last paroxysm occurred at the same hour, and lasted three days, until Oct. 25th. His convalescence was very satisfactory, and he was enabled to resume his studies by the middle of November. No sequelæ occurred. In 1878 Dr. Y., who had been working very steadily with a rapidly-growing practice, was attacked with severe typhoid fever, with grave nervous symptoms and with albumen and tube-casts in the urine, and died on the twelfth day.

It will thus be seen that in this unusually protracted case there were seven distinct relapses, one of which was brief and interrupted one of the regular intermissions, while the rest were all severe.

Duration of 1st paroxysm, violent, 8 days. 1st intermission, 9 days.
Duration of 2d paroxysm, violent, 5 days. 2d intermission, 1 day.
Duration of 3d paroxysm, less violent, 1 day. 3d intermission, 6 days.
Duration of 4th paroxysm, severe, 6 days. 4th intermission, 8 days.
Duration of 5th paroxysm, severe, 5 days. 5th intermission, 9 days.
Duration of 6th paroxysm, less severe, 4 days. 6th intermission, 10 days.
Duration of 7th paroxysm, less severe, 3 days. 7th intermission, 11 days.
Duration of 8th paroxysm, mild, 3 days, followed by convalescence.

The total duration of the case, which was entirely free from complications, was therefore ninety days.

VARIETIES.—The foregoing clinical description prepares us to appreciate the varieties of relapsing fever that may be said to exist. They consist of—

The abortive form, in which a single paroxysm of variable length and severity occurs, terminating in a critical fall of temperature and usually with some critical discharge, but not followed by any relapse. There can be no doubt of the existence of such cases, although they are not common; and at times the paroxysm is so slight that were it not for the known exposure of the individual to the prevalent epidemic influence, in the absence of any other adequate cause, the case might readily be regarded as one of non-specific febricula. The caution must, however, be borne in mind as to the occurrence of relapses of such extreme shortness of duration (less even than twenty-four hours) as to readily escape notice unless a careful watch be kept for their detection.

The ordinary or typical form, including the cases with one or two relapses, presenting the usual variations in the severity of the symptoms and in the duration of the paroxysms and of the intermissions.

The multiple or protracted form, if it be thought desirable to thus particularize cases presenting an excessive and unusual number of relapses, as three, four, five, six, or even seven.

The grave or subintrant form, which is designed to include the highly congestive form of Cormack and the bilious typhoid of Griesinger and Lebert.

Under another heading (see relations to other diseases, [below]) we shall give reasons for regarding the bilious typhoid fever of Griesinger and Lebert as merely a form of relapsing fever, with which a certain proportion of cases of true typhoid fever complicated with hepatic catarrh may have been included.

The characteristics of this grave subintrant form are as follows: Jaundice, occasionally absent, but usually present in an intense degree; marked enlargement of the liver and spleen; a tendency to hemorrhage from various mucous surfaces; extreme prostration; defective or suppressed secretion of urine; hypostatic congestion or inflammation of the lungs in a large proportion of cases; dry brownish tongue; low muttering delirium, often passing into stupor or coma; hiccough; imperfect crisis; and a continuance of some morbid phenomena, so that merely a remission occurs to separate the paroxysms; and a high percentage of mortality. The great modification of the intermission which is so highly characteristic of typhoid relapsing fever is doubtless due in chief part to the serious local lesions developed, and seems to justify the name of subintrant as above suggested. The course of such fever is well illustrated by the following case, in which the characters of typhoid relapsing fever were present in the highest degree, death occurring on the fifteenth day:

Charles Hood, colored, æt. 28, of temperate habits, was taken ill on April 5, 1870, after malaise lasting thirty-six hours, with fever, nausea and vomiting, headache, and general aching throughout body; and was admitted to the hospital April 6th. There was already marked jaundice, and epistaxis had occurred; there were also insomnia; wandering delirium; extreme tenderness over the liver and spleen, both of which were enlarged; dryness of tongue, vomiting, and distension of the abdomen. These symptoms continued, his condition becoming daily more aggravated. Restless delirium alternated with heavy sopor. The jaundice grew deeper. Marked digital formication existed, but the arthritic pains were not so severe as in ordinary cases. The tongue was dry and of a red orange color. Profuse epistaxis occurred on the seventh day of the disease, requiring plugging of both anterior and posterior nares, and followed by great prostration. A gradual fall in the temperature occurred during the sixth, seventh, and eighth days, reaching 99° on the latter day. During this decline the delirium ceased and the mind remained merely dull; the jaundice decreased, as did also the tenderness of the hypochondriac zone. The pulse and respirations improved, and diarrhoea ceased. The improvement was but brief; for about eighteen hours he lay apyretic, with cool hands and feet, and with eyes closed and mind dull but free from delirium. Fever then reappeared and with the ascent of the temperature the unfavorable symptoms recurred. The relapse lasted but two days, and was followed by irregular decline of fever till death occurred on the fifteenth day of the disease. Obstinate hiccough appeared on the eleventh day, and continued, accompanied with occasional vomiting on the fourteenth day. Delirium alternating with sopor reappeared. Jaundice again became marked, and again there was extreme tenderness over the liver and spleen. The pulse grew small and feeble, the respirations shallow and labored, with an expiratory moan. Cough began on the twelfth day, and was soon followed by the physical signs of pneumonia of the lower lobe of both lungs. The urine continued free from albumen. The patient sank into deeper coma, and died on the fifteenth day. Post-mortem examination showed highly-developed characteristic lesions of the spleen and liver, with red hepatization of lower lobe of both lungs. There was no affection of the glands of Peyer. The course of the fever is shown in the following tracing (see Fig. 22).

FIG. 22.
From a case of the bilious typhoid or grave subintrant form of relapsing fever.

COMPLICATIONS AND SEQUELÆ.—As would be anticipated from what has been said of the wide range of the symptoms and of the remarkable course of the temperature in relapsing fever, there are many complications and sequelæ liable to occur, and which require special consideration. They may be classified according as they affect the febrile movement, the state of the blood, or one or other of the groups of organs.

We have already described the various irregularities presented by the febrile paroxysms and the intermissions, and no further allusion need be made to mere variations in length, severity, or number of the former. In rare cases, however, a peculiarity is presented, usually in the first intermission, which is difficult of explanation. About twenty-four hours after an apparently complete crisis, with a fall of temperature to a subnormal point, there may be a sudden and rapid rise or rebound of temperature to 104° or 105°, attended with distressing symptoms of high fever, but lasting only twenty-four or forty-eight hours. A good example of this is given in the case described [above] and Carter18 cites several examples of it terminating either in recovery or in rapid death. He asserts that examinations of the blood during such post-critical febrile rebounds invariably showed an absence of spirilla, so that in his opinion such fever must be considered non-specific. Their explanation seems difficult, since the pyrexia is too brief to be associated with any local inflammatory complication.

18 Op. cit., p. 172.

More frequent and serious is the protracted post-critical pyrexia which we have already described as modifying the interval, so as to produce a subintrant type by maintaining continuous though irregular fever until the accession of the relapse, unless cut short by death. This post-critical fever is non-specific, is unattended with spirilla in the blood, and is to be associated with the extensive irritative processes in the liver, spleen, kidneys, lungs, and other parts that are present in these grave and complicated cases. It is to be noted that the course of those paroxysms which terminate in lysis indicates that they may represent a milder type of the above process.

The peculiarities of the delirium, amounting sometimes to maniacal excitement, which attends some cases of relapsing fever, has been fully described.

Less common are the following: mental hebetude, lasting some days or even weeks after the close of the last paroxysm, or, as in a case of Carter's, gradually increasing mental feebleness, terminating in imbecility. In such cases suspicion must arise of the occurrence of some local lesion of the membranes or substance of the brain.

Partial palsy is mentioned by numerous authors as occurring during or shortly after attacks of relapsing fever. Paralysis of one or both deltoids has been noted, the latter by Cormack, who saw it continue ten days after the patient was well in all other respects. Temporary paralysis of the forearm (Douglas) or of the whole arm (Parry, Meschede) has been observed; and Parry also describes loss of power in the legs lasting for one week. In one of our cases temporary loss of power of the left arm and leg occurred, attended with such impairment of sensibility that the woman had to feel for the fingers of the left hand to assure herself of their existence. This loss of power occurred during the initial paroxysm, and gradually passed away, but she was unable to stand alone on the thirty-first day of the disease. In a case reported by Tennent19 facial palsy was developed six days after the second crisis.

19 Glasgow Med. Jour., May, 1871, p. 379.

Various explanations have been offered for these local palsies, but, as already stated (see [above]), it seems probable that they are referable to morbid conditions of the nerve-trunks, or, less commonly, of the spinal cord. It must be noted, however, that in a certain number of autopsies serious intracranial lesions are found, which are evidently the results of the attack of relapsing fever. These consist of abscess of the brain, meningitis, and specially cerebral hemorrhage. This was present in one of our cases, but Carter found copious hemorrhage in no less than 8 out of 54 autopsies, and in 5 others there were minute capillary cerebral hemorrhages. Still, in nearly all the cases of large hemorrhage we have found recorded the effusion was upon the surface of the brain, and this, combined with the absence of true hemiplegia from the forms of paralysis noted in relapsing fever, and the transient character of these palsies, makes it clear that they are not to be explained by any considerable cerebral hemorrhage. On the other hand, however, it must be admitted that an additional possible cause of them is to be found in minute hemorrhage into small areas known to govern the movements of certain groups of muscles. Again, we have had occasion to note the occurrence of both thrombosis and embolism among the lesions of relapsing fever, and it is evident that either of these accidents, if involving a comparatively small branch of a cerebral vessel in certain motor areas, might cause transient paralysis, such as has been described. Nor can we fail to see that, while such symptoms as the delirium, mania, coma, or subsequent mental impairment may receive other explanations, it is possible that they may arise from similar processes of minute hemorrhage, thrombosis, or embolism involving other parts of the brain.

The frequent occurrence of severe rheumatic pains in the muscles and joints during the course of the disease has been dwelt upon ([above]); but in some cases they persisted during the intermissions and for a considerable time after all other symptoms of disease had passed away. Occasionally they greatly retarded convalescence by interfering with exercise and sleep. These pains were mostly in the legs, and were increased by exercise, and also seemed to be influenced by changes of weather. Patients who suffered thus were also liable, after exposure or in consequence of severe atmospheric changes, to sharp attacks of similar pains elsewhere, and especially in the course of the intercostal nerves. Occasionally violent and persistent headache follows the disease, not improbably associated with changes in the membranes of the brain, although in other cases severe neuralgia occurs in consequence of the anæmia which may remain in an intense degree after the fever. Troublesome numbness and soreness of the soles of the feet and of the palms of the hands, increased by pressure, has been noted as a sequel persisting for several days or weeks.

Affections of the special senses are not rare. The most remarkable among these is the affection of the eyes, which is apt to occur far more frequently in connection with relapsing fever than with typhus or typhoid. The proportion of cases in which this sequel appears varies greatly in different epidemics. In the British epidemics of 1826 and 1843, when this form of post-febrile ophthalmia was first accurately described by Mackenzie of Glasgow, it was very frequent; and it was equally so in Finland in 1867-68, when Estlander20 again carefully studied it.

20 "U. Choroiditis nach Febris Recurrens," Arch. f. Ophth., 1869, Bd. xv., Abth. ii., 108.

On the other hand, so far as can be stated in regard to a sequel which may appear after convalescence is far advanced and the patient discharged from medical care, it was very uncommon in the Philadelphia epidemic of 1869-70. This ophthalmia may occur during the course of the fever, but more frequently it begins during convalescence, and even some months after convalescence has been established. It occurs in patients of both sexes and at all ages. Usually it affects but one eye, but both may be attacked simultaneously or consecutively. Patients who were very ill-nourished and debilitated were most apt to present this sequel, and Murchison regards previous starvation as one of its main causes. The exciting cause and true pathology appear obscure as yet, however, and the existence of a neural origin is not improbable. In some cases the ophthalmia has seemed to result directly from exposure to cold. Among our own patients, as already stated, eye symptoms were less common and severe. A careful record of 184 cases was kept in reference to this question. Several patients complained of diplopia during the febrile stage, and one asserted that every object appeared fourfold to him. Conjunctivitis of moderate severity, usually associated with otorrhoea, occurred in about 5 per cent. of our cases; it generally affected only one eye, and occurred in a few instances as late as the third week after the relapse. In a few cases (four) also there was dulness of vision in one eye, noted during the course of the disease and persisting for some time after convalescence began. In only one instance, however, did permanent impairment of vision ensue, and this man had passed through a violent attack of the fever with unusually grave nervous symptoms. It left him with optic neuritis on the right side, which induced partial atrophy of the nerve and great limitation of the field of vision. Meschede reports intraocular affections in 6 cases out of 180 specially examined, though it is not certain that such affections were directly connected with the febrile process. Ocular ecchymosis occurs in a small proportion of cases, especially of the graver types.

Dulness of hearing is not so common in relapsing fever as it is in typhoid. It was present in 14 out of 184 of our cases during the course of the disease, and in a few instances partial or almost complete deafness in one ear persisted after convalescence, owing doubtless to a slight affection of the middle ear. In one case marked deafness appeared suddenly on the day after the termination of the relapse by crisis. Meschede21 found disease of the middle ear in no less than 8 per cent. of his cases.

21 Loc. cit.

Purulent otorrhoea from one or both ears is of more frequent occurrence, and without any special exciting cause may present itself at any time during the course of the disease or more commonly after the relapse. In the same manner purulent coryza may occur.

The eruptions occasionally present during the fever have been described. Bed-sores from pressure are much less common than in typhus, but are met with in a small proportion of cases. As a rule, they are of moderate size and heal quickly. Superficial gangrene of the lips, nose, and ears has also been noted in rare cases (Zuelzer) in connection with gangrene of the extremities, probably from embolism. The occasional occurrence of painful boils, of abscesses in the cellular tissues (Wyss and Bock), and the more rare occurrence of erysipelas may be mentioned among the sequelæ.

As already stated, the severe pains in the joints and members which so frequently occur during relapsing fever are, as a rule, unattended by any redness or swelling of the joints. In rare cases, however, there is effusion into the joints during the fever, or more commonly there are attacks during convalescence which simulate subacute rheumatic arthritis. Such attacks may last but a few days, but in several of our cases there was painful swelling of the knees, wrists, and fingers which persisted for several weeks after the fever, being attended with slight crepitation on motion, and altogether behaving like subacute rheumatism.

As would be expected from the severity of the fever, the marked disorder of digestion, and the lesions of the spleen and liver in relapsing fever, anæmia is a common sequel. In cases where there has also been free hemorrhage, usually in the form of epistaxis, the anæmia may indeed reach an intense degree.

The cardiac murmurs which have been described as present in a certain proportion of cases are dependent upon the blood-changes, and when the anæmia is extreme these murmurs are also audible over the large veins and the pulmonary artery, and persist after convalescence is fully established.

Oedema of the lower extremities occurs in a considerable number of cases. It is clearly due in part to the anæmia, but the cardiac debility which follows the fever is also largely concerned in its production. It was, indeed, marked in some of our cases where no anæmic murmurs existed, but where there was great nervous and muscular debility. Usually limited to the feet and ankles, it occasionally extended above the knees, and in one case, where great anæmia and debility from fever and over-exertion coexisted, there was oedema of the hands and wrists, with great distension of the legs up to the hips. It is not associated with albuminuria as a rule, and yields readily to treatment and rest, in the course of a few weeks.

Hemorrhages from various surfaces have already been mentioned, and a full account given of epistaxis, which is by far the most common form. Bloody vomiting has been noticed in a small proportion of cases in various epidemics. It varies in amount, but is always attended with great gravity of the attack, and usually is followed by fatal results. It occurred in four of our cases, two of which presented also black stools containing altered blood, and suppression of urine; while in another it occurred at the close of the first relapse, and during the second relapse was copious and repeated. In this case it was attended with alarming symptoms of collapse, from which the patient rallied, and after a desperate struggle recovered.

Blood may also be discharged from the bowels in such large amount as to constitute actual hemorrhage—a symptom of great gravity; or in small quantity and completely altered, so as to impart an inky black color to the stools—a condition not necessarily attended with urgent danger; or, finally, there may be frequent bloody dysenteric stools.

Hemorrhage has also been observed from the uterus, from the kidneys, from the ears, and from the old cicatrix of a syphilitic chancre. Hemorrhage occurred in 87 out of 183 of our cases, or in nearly 50 per cent. It was from the nostrils in 82 cases, from the uterus in 1 case, from the stomach in 4 cases, and from the cicatrix of a chancre in 1 case.

Sudden collapse occurs with such comparative frequency in relapsing fever as to require special attention as one of its complications. It may occur at any period of the disease, but it is most common at the crisis of the first paroxysm or of the relapse. The symptoms are usually those of cardiac failure, with rapid, small, and feeble pulse; shallow and hurried, or slow, labored, and imperfect respiration; coldness of the extremities, while the central temperature may remain elevated; muttering delirium, rapidly passing into unconsciousness. Occasionally almost instantaneous death occurs from syncope induced by some muscular exertion, as standing up or even rising in bed. In other cases the symptoms indicate the development of cardiac thrombosis, and subsequent examination has verified this opinion. In still other cases the symptoms resemble those which occur in extreme hyperpyrexia dependent upon overwhelming and paralysis of the nervous centres. Copious hemorrhage from the stomach and nose may also induce syncope of alarming and even fatal severity. When from the latter cause, reaction may be induced and the patient may ultimately recover, as we saw in a case where after repeated hematemesis the patient sank into profound collapse. In all of its forms, however, this complication is of extreme and imminent danger, and death follows, as a rule, in a few hours. The cases in which it occurs are usually of severe type, occurring in persons who have previously been in poor health or intemperate, or who have been subjected to privation and improper exposure previous to and during the early stages of their attack. Still, collapse may occur in mild cases also, and whatever the type of the disease there may be no special indication of approaching trouble, when the patient rapidly passes into collapse, to be followed by death in a few hours. It occurred in nine of about two hundred cases under our observation. In one it was the result of hemorrhage from the stomach, and ended in recovery; in one, at the close of the initial paroxysm the patient, who was stupid, with muttering delirium, sank into collapse as the temperature rapidly fell from 105° to 97°, and died in a few hours; in one, on the fourth day of the relapse the temperature suddenly fell from 102° to 96°, with free sweating, but suddenly rebounded to 102°, with very rapid, feeble pulse, distinct basic cardiac murmur, constriction of chest, restlessness and delirium, slight convulsions, and death in eight hours; in one, a man at the end of the initial paroxysm, immediately after his admission to the hospital in apparently fair condition, became violently delirious, with bounding pulse, soon grew comatose, and died in one hour; in one, a man who was in feeble condition, on the nineteenth day, with irregular persistent fever (he had splenic abscess), sat up on the edge of the bed, sank back in syncope, and died in less than an hour; in one, a man who did well until the second day of the relapse, when pleuro-pneumonia and pericarditis were developed, died suddenly four days later: there was considerable pericardial effusion; in one, sudden death from syncope or cardiac thrombosis occurred on the twelfth day in a man who had suppurative parotitis and metastatic abscesses of the lungs; in one, sudden collapse and death occurred in one and a half hours at the end of the initial paroxysm; in one, a drunkard with large fatty liver had pyrexia continuing after the initial paroxysm, and on the ninth day, while in a state of hebetude, with mild delirium and a pulse of 112, coma suddenly occurred, and death followed in two hours.

Pericarditis is a rare complication, and is apt to coexist with pleuro-pneumonia. This combination occurred in one of our cases where pleuro-pneumonia and pericarditis were developed on the second day of relapse, and proved fatal by sudden collapse on the fifth day, with the pericardial sac distended with serum and its layers coated with plastic lymph.

Thrombosis of veins, as in phlegmasia alba dolens, occurs much more rarely than after typhoid fever. Arterial embolism, on the other hand, is not uncommon. Murchison22 reports a case in which gangrene of the left foot from obstruction of the left femoral artery, together with cerebral softening from obstruction of the left middle cerebral artery, occurred in connection with cardiac thrombosis. Zuelzer alludes to similar cases in the St. Petersburg epidemic of 1865-66, where, in addition to the extremities, the nose, ears, and lips became gangrenous. Other examples of embolism are found in lesions of the spleen and kidneys, where infarctions are of frequent occurrence.

22 Op. cit., p 384.

Heart-clot, or cardiac thrombosis, appears to occur more frequently than in any other acute zymotic disease, with the exception of diphtheria. Even when the occurrence of passive hemorrhages and of ecchymoses of various tissues indicates marked dyscrasia of the blood, there will not rarely be found firm white clots in one or other of the cavities of the heart. These frequently present unmistakable evidences of ante-mortem formation, and, as already stated, there is a certain proportion of the cases of rapid and unexpected death where the fatal result is directly due to cardiac thrombosis, attended with the usual symptoms.

The constant affection of the spleen has been fully described; it is not therefore surprising that both complications and sequelæ arise in connection with it. At times, in cases which ultimately recover, the pain in the splenic region is so violent and continuous, and is attended with so much tenderness over the enlarged organ, that localized peritonitis is undoubtedly present. Occasionally this perisplenitis persists, and in conjunction with the inflammatory changes in the substance of the spleen maintains an irregular fever after the specific pyrexia has run its course. This was noticed in several of our cases, but especially so in a case where, after the initial paroxysm, an irregular fever was kept up, obscuring the relapse, until the nineteenth day, when death occurred suddenly from syncope on rising on the edge of the bed, and where examination showed splenic peritonitis, with a splenic abscess as large as a pigeon's egg.

The enlargement of the spleen usually subsides during the intermission, and disappears speedily or in the course of a few weeks after convalescence is established. Occasionally, however, it persists, and is attended with marked anæmia. In one case, where death occurred from pneumonia, the sequel of relapsing fever, at about the thirtieth day, the spleen weighed twenty-nine ounces; and in another case, where death occurred from gangrenous pleuro-pneumonia, at the fortieth day, the spleen was still enlarged and presented characteristic changes in its pulp. On the other hand, in a case where death occurred on the twelfth day of typhus, occurring forty-four days after recovery from a very bad case of relapsing fever, making it altogether the one hundredth day, none of the lesions of the first disease were discoverable.

Rupture of the spleen occurs occasionally, and is usually attended with sudden pain, collapse, and speedy death. Murchison refers to two examples recorded by Zuelzer and one by Hudson; Petersen reports fifteen cases, in seven of which sudden rupture occurred with speedy death, while in the other eight the rupture followed local softening from infarction, and resulted in death in a few days from purulent peritonitis.

In one of our cases, where death occurred on the sixteenth day, apparently from double pneumonia and heart-clot, it was found that there was a rupture in the enlarged spleen near its upper end, recent plastic peritonitis in the region of the spleen, and a moderate amount of bloody pulpy fluid throughout the peritoneal cavity.

As we have seen, disturbances within the respiratory tract occur with very different frequency in different epidemics. In many they are rare, while in 1870 we noticed cough and other evidences of respiratory trouble in no less than 90 out of 200 cases.

Severe catarrhal laryngitis is a rare and dangerous complication. It did not occur in our cases, but both Begbie and Paterson report cases of it which required tracheotomy, and Wyss and Bock met with ulcerative laryngitis with perichondritis.

Bronchitis of moderate severity, although rare in many epidemics, occurs so frequently in others, as in Philadelphia in 1870, as to rank as a symptom of the disease.

Pneumonia is one of the most fatal complications. The results of our own observations agree with the statements of Jenner and of Carter, that it is the next most common lesion after enlargement of the liver and spleen. On the other hand, Murchison noted it only in 4 or 5 out of 600 cases. It occurred in at least 11 of our cases, 8 of which were fatal; and unquestionably less extensive inflammation was present in other cases which recovered, in view of the marked respiratory disturbances frequently present. Both lungs were involved in 4 cases; of the remainder, the right and left were about equally divided. Out of 23 autopsies, the lesions of pneumonia were found 8 times. The lower lobes were affected in every case. The form of this disease was croupous in 9 cases; in 1 it was that of metastatic suppuration, and in 1 it was more properly described as splenification. The amount of plastic pleurisy associated with it was usually great, and in one case there was also severe pericarditis. In another case the disease advanced to the stage of gangrene of a circumscribed area of the pleura and of the superficial layer of the lung. In only one instance was albuminuria present. In two cases the pneumonia occurred so late in the course of the disease that it might be regarded as a sequel. Death occurred in one of these on the thirtieth day, and in the other (that in which gangrene ensued) it ran a subacute course, and death did not take place until the fortieth day. In the other cases the disease began at the close of the initial paroxysm, during the intermission, or early in the relapse. As would be expected, the sympathetic fever due to this complication modified and obscured the characteristic course of the specific pyrexia.

This rare termination in gangrene has been noted by other observers; in all five or six times. Parry met with a truly remarkable case of double pneumonia, followed by gangrene, and yet resulting in recovery. Jaundice is apt to attend cases of relapsing fever which are complicated with pneumonia.

Pleurisy is an almost constant accompaniment of pneumonia, and frequently occurs in marked degree. It may also be present in cases of severe splenic inflammation. In all probability, localized plastic pleurisy is not infrequent, and may cause some of the severe thoracic pains so frequently present.

Metastatic abscesses of the lung occur occasionally as a result of the profound toxæmia, and are apparently preceded by patches of infarction, which soften in the centre, as in the usual development of pyæmic abscesses. This condition was found in one of our cases in conjunction with suppurative parotitis. It has been included among the instances of pneumonia.

Acute miliary tuberculosis, involving chiefly the lungs and intestinal canal, occurred as a sequel in one case under our observation, and phthisis has been found to follow by other observers (Carter). It is to be expected that if the patient did not so quickly pass from under observation it would be found that an affection so gravely complicating nutrition as does relapsing fever is frequently followed by serious organic disease.

Parotitis is mentioned by so few authors as to show that it is a rare complication in most epidemics, varying from 1 in 600 to 1 in 50 cases. One gland only is affected at a time as a rule, though both may be involved successively. The inflammation begins either during the intermission or the relapse, and may terminate by resolution or by suppuration. Although a painful and severe complication, it is followed by recovery in a considerable proportion of cases. Carter23 states "that in some degree it was noted in 2 or 3 per cent. of all cases, and nearly as often amongst survivors as in the casualties." It occurred in three of our cases (185); once it underwent resolution; once suppuration occurred in the parotid and in the masseter muscle, with metastatic abscesses in the lungs, and death; and once the patient, who had previously existing amyloid degeneration of liver and spleen without albuminuria, had severe relapsing fever with two relapses, in the first of which parotitis occurred in both glands, successively terminating in suppuration, after which he did well through an apyretic period of six weeks, when sudden high fever appeared, followed by speedy death.

23 Op. cit., p. 210.

Pharyngitis and tonsillitis of mild grade occur in from 3 to 25 per cent. of the cases in different epidemics.

Hiccough deserves to be ranked among the complications, because it is of frequent occurrence, obstinate and annoying. It occurred in a considerable proportion of our cases, and much more frequently in those who had jaundice. It was often present both in the initial paroxysm and in the relapse, but disappeared soon after the end of the pyrexia. It bore no constant relation to the severity of the vomiting. Not rarely it lasted several days and nights, causing exhaustion and interference with sleep and proving rebellious to treatment. Hypodermic injections of morphia and atropia, chloroform internally, and extremely careful alimentation proved most serviceable.

Hemorrhage from the stomach has already been spoken of (see [above]).

Diarrhoea, as already stated (see [above]), occurs much more frequently than in typhus fever, varying from 1 per cent. (Murchison) to 15 per cent. (Scotch epidemics) or 33 per cent. (Philadelphia), or even 50 per cent. (Königsberg). It is usually of moderate severity, but occasionally is so profuse and intractable as to constitute the main cause of death. In some epidemics the attacks of looseness occur almost exclusively after the relapse, but in others the bowels are frequently loose during the febrile stages. In our cases there were not infrequently from three to eight thin, dark, bilious or light yellowish stools daily after the second or third day of the initial paroxysm, and then the looseness would stop during the intermission, probably to recur in the relapse. Occasionally diarrhoea with very frequent liquid stools occurs at the close of one or both of the febrile stages, assuming a critical character, and substituting more or less of the sweating which is the common mode of crisis, although in several such cases quoted by Murchison from Douglas the sweating, despite the critical diarrhoea, was usually profuse. It can scarcely be said that there is any relationship between diarrhoea and vomiting; both are frequently present, and may even be severe and persistent in the same case, though either may be marked while the other is moderate or slight. Abdominal pain and tenderness in the epigastrium and hypochondria are constant symptoms, but when diarrhoea is marked there are apt also to be griping pains and tenderness in the lower segment of the abdomen. When diarrhoea occurs as a sequel, either beginning after the close of the relapse or continuing in cases where the bowels have been loose during pyrexia, it is apt to prove obstinate and intractable, or even to lead to a fatal result.

The character of the stools varies much; usually thin and dark, they may be light yellowish or even whitish. Thus, in a severe case with deep jaundice we observed seven liquid and decidedly whitish stools in twenty-four hours. In such instances there is undoubtedly more or less complete closure of the biliary ducts by plugs of mucus or by swelling of the mucous membrane. On the other hand, the stools may be inky black from admixture with altered blood, or, lastly, they may consist of mucus and blood, in which event the complication assumes the form of actual dysentery and is attended with increased abdominal pain and with tenesmus. Dysentery was, as would be expected, quite frequent in the Indian epidemics studied by Carter.24 It is usually of moderate severity, but occasionally it runs into gangrenous inflammation, is attended with perforation of the bowel, or is followed by hepatic abscess. In one instance we noticed a peculiarly fetid puriform discharge from the anus, which occurred during the relapse and persisted for several weeks, gradually subsiding, as though from some unhealthy ulceration which slowly healed.

24 Op. cit., p. 218.

Jaundice is of frequent occurrence, but has been sufficiently discussed [above].

Peritonitis is not rare in its circumscribed form. This statement is based on the comparative frequency with which localized splenic peritonitis, of varying degrees of severity, is found after death in relapsing fever from various causes, and from the great frequency of severe pain and tenderness in the region of the enlarged spleen in favorable cases. In its lesser degrees it may not add materially to the danger of the patient, but in more severe forms, associated with serious splenic lesions, it may run a protracted subacute course and maintain irregular fever.

General peritonitis is, on the other hand, a rare complication, occurring not more than once in several hundred cases. It results from dysenteric perforation of the bowel, from rupture of a splenic abscess, or from rupture of the spleen itself. An example of this latter accident which occurred under our observation has already been given. Speedy death invariably follows, though in the case just referred to the symptoms of peritonitis were totally masked by those of the coexisting double pneumonia, which seemed to be the immediate cause of death.

Suppuration of the mesenteric glands is a rare complication, mentioned especially by Wyss and Bock. As these glands are not usually found enlarged, there being no irritative lesion of the intestines of common occurrence in relapsing fever, it is probable that the collections of pus which have been found were metastatic in origin.

Dyspepsia is not an infrequent sequel, as would necessarily be the case after a disease characterized by so much gastric irritation and by such serious lesions of the liver and spleen. As a consequence, care in diet is often required for a considerable period after the course of the disease has ended; dyspeptic symptoms are frequently complained of, and marked emaciation and anæmia often protract convalescence.

It may be observed that a striking appearance of emaciation is often developed shortly after the crisis of the first paroxysm, or, more particularly, of the relapse. It is partly due to the actual loss of weight during the high pyrexia, but even more to the abrupt transition from a state of extreme febrile turgescence to one of equally extreme relaxation and maceration of the surface.

The amount of urine has been seen ([above]) to vary greatly in cases distinguished by no special disorder of the kidneys; the extremes in ordinary cases being from twelve or fifteen ounces just before the crisis to from eighty to one hundred and twenty within forty-eight hours after the crisis. Suppression is, however, sometimes noted, and is always a grave symptom, though Parry25 reports more than one case in which on several successive days there was not more in twenty-four hours than one fluidounce of non-albuminous urine, and in which no symptoms of uræmia occurred, and the sweat had no urinous odor. In one of our fatal cases, with intense jaundice, hematemesis, inky black stools, and oedema of the feet and of the lungs, there was not a drop of urine secreted during the last four days of the initial paroxysm; death occurred on the eighth day, and the kidneys were found intensely engorged, of a deep blackish-blue color, with numerous ecchymoses in the cortex, due to impaction of the convoluted tubules with blood, while the renal epithelium was granular and swollen, and many tubules were filled with epithelial cells and granular matter. At the autopsy the urinary bladder was firmly contracted and contained a very small amount of bloody liquid.

25 Op. cit.

More frequently, incontinence of urine, with or without retention, occurs during the febrile stages—according to our observation, most commonly in cases attended with mental disturbance and tending to a typhoid condition. The symptom was not of very grave significance, however, and after the use of the catheter for a few days the bladder regained its tone.

Albumen is quite frequently present in small amounts during the pyrexia of relapsing fever. Thus, in 18 cases of ordinary severity, which all recovered, and in which the urine was carefully examined daily, a trace of albumen was found in 5; in 2 cases it appeared both in the initial paroxysm and in the relapse, but in all instances its presence was of brief duration. In one of these five cases the albumen appeared at both critical periods, when the amounts of urine in twenty-four hours were respectively 150 ccm. and 250 ccm.; but in the other cases the transient albuminuria coincided with free secretion of urine (1250 ccm., 1850 ccm.). It is probable that were the same careful search to be made in all cases the presence of albumen would be detected in fully 20 to 25 per cent. On the other hand, in fatal cases the occurrence of albuminuria is by no means constant, although undoubtedly it is present in a larger proportion of such cases than of those of ordinary severity.

Our experience does not confirm that of Murchison, who states that he never met with typhoid symptoms in relapsing fever without albuminuria or some other evidence of retarded elimination by the kidneys. In several of our cases where the typhoid state was developed in the highest degree repeated examination of the urine failed to discover albumen.

Most observers have been struck with the comparative immunity of the kidneys from serious disturbance in a disease presenting such complicated morbid processes and widespread lesions as relapsing fever. To show, however, that these organs suffer specially in certain epidemics, it may be mentioned that Obermeier26 reports having found albumen with tube-casts of various kinds in 32 out of 40 cases of relapsing fever, thus showing that, in the particular epidemic he was studying, catarrhal nephritis was of almost uniform occurrence. It is true that serious interference with the elimination of urea and other nitrogenous matters may occur without the coexistence of albuminuria, so that it is impossible to deny that severe nervous symptoms may result from impaired renal activity even when the urine contains no albumen.

26 "U. d. wiederkehrende Fieber," Arch. f. path. Anat. u. klin. Med., Bd. xlvii. p. 170.

Attention has already been called to the variations presented in the amounts of urea, but more extended observations are required to show the precise relations of these variations to the graver nervous phenomena. It will be found, we venture to opine, that, while in one group of relapsing-fever cases of grave type, cerebral symptoms are dependent upon the retention and accumulation in the system of urea and other effete nitrogenous products, owing to interference with renal activity from pre-existing organic disease of the kidneys or from an exceptional degree of congestion of those organs, there are other groups where similar typhoid cerebral symptoms are more directly dependent upon the specific toxæmia, upon the hyperpyrexia, upon exhaustion of the nerve-centres by intense peripheral irritation, or upon congestion or other morbid conditions of the nerve-centres themselves.

In all cases where cerebral symptoms manifest themselves in relapsing fever the daily examination of the urine—which here, as in other zymotic diseases, is a duty in all cases—becomes of extreme importance. Three conditions should be borne in mind in such examinations. In the first place, the attack of fever may have occurred in one already the subject of organic kidney disease, and, considering the classes from which the majority of the cases of relapsing fever are drawn, this possibility cannot be of rare occurrence. Out of eighteen post-mortem examinations in which the kidneys were studied with especial care we found positive evidence of pre-existing organic disease four times. In these cases the albuminuria was marked and persistent, though tube-casts were rarely found, and severe cerebral symptoms of typhoid type were prominently present. In another highly interesting case the patient, who had amyloid disease of the liver, spleen, and kidneys, contracted severe relapsing fever; he had increased albuminuria during both febrile stages, suppurative parotitis, but no grave cerebral symptoms, and apparently recovered. After an apyretic period of six weeks, during which the symptoms of the amyloid visceral disease persisted, a sudden and rapidly fatal pyrexia occurred. Unfortunately, the existence of spirillar infection of the blood was not known at the time.

In the second place, the attack of fever may become complicated with acute nephritis from special localization of the poison, as in Obermeier's cases, or from vulnerability of the kidneys. In such cases careful study of the urine should indicate the event, and the prognosis, though grave, is not so hopeless as in the first instance. An interesting example of this occurred under our observation, where the patient, who had apparently an ordinary attack, was seized with acute catarrhal nephritis, with temporary uræmia, during the relapse, but after a dangerous illness recovered without any organic renal disease as a sequel.

In the third place, may be found the more usual and more readily-determined condition of slight and transient albuminuria (with variations in urea excretion) which has already been discussed, and which has no serious prognostic significance.

The following very interesting case deserves special mention: The patient, a man aged thirty-six, was admitted on the fifteenth day of an attack of acute catarrhal nephritis, with slight ascites, marked oedema of the feet and legs, and highly albuminous urine. In the course of ten days the oedema and albuminuria were much diminished, when on the thirteenth day after admission he was attacked with relapsing fever, the ward in which he lay containing a number of persons ill with that disease. The initial paroxysm was severe, but without any grave cerebral symptoms; the urine grew scanty, dark, and bloody, and the oedema increased and invaded the pelvis. Crisis occurred on the fifth day, temperature falling 9°, sweating copious, urine 473 ccm. in twenty-four hours, color of porter, highly albuminous, and depositing blood, renal epithelium, hyaline, granular and epithelial casts, all stained reddish. Two days later, urine 1600 ccm., light colored, with only a small amount of albumen.

A slight and brief relapse (101° for two days) occurred after an interval of four days; a second imperfect relapse (100.5° for three days) after a further interval of six days; and finally, after a further interval of only two days, a violent relapse (temperature rising rapidly to 106°) with crisis (fall of 8° in twelve hours) at close of fifth day. The oedema gradually diminished from the time of the first crisis, did not increase in the relapses, and disappeared completely and finally about ten days after the last relapse. The urine was very free after the first paroxysm, averaging from 2000 to 2300 ccm. During the subsequent febrile periods it did not decrease, and indeed on the second day of the last relapse, with the temperature at 105°, the amount in twenty-four hours was 3200 ccm. Four days subsequently, during crisis, the amount was only 350 ccm.

The albumen disappeared entirely from the urine in two weeks from the close of the last relapse; there had then been no tube-casts for some days, and the patient was discharged entirely well a short time afterward. The treatment consisted of hot vapor-baths, repeated dry cupping over the kidneys, infusion of digitalis with acetate of potash during pyrexia, and Basham's iron mixture in the intermissions. It seemed that the occurrence of the relapsing fever interfered wonderfully little with the recovery from nephritis.

Hematuria is a comparatively rare and very grave complication. It may occur as an additional evidence of the dyscrasia of the blood in connection with hemorrhages from other surfaces, or as in the case we have before referred to or in that reported by Murchison,27 it results from intense engorgement of the kidneys. In Murchison's case hematuria, with much albumen and tube-casts, occurred in both paroxysms without any uræmic or typhoid symptoms, and was followed by satisfactory recovery.

27 Op. cit., p. 370.

Sugar is sometimes present in small quantity as a transient symptom; and diabetes has been observed as a sequel.28

28 Tyson, Phila. Med. Times, 1871, i. 418.

Metastatic inflammation of the kidneys, with centres of suppuration, was observed by Wyss and Bock.

When menstruation occurs during relapsing fever, as it may do at any time, it is apt to be excessive, and may amount to severe hemorrhage. Crisis has been known to occur in this manner.

The numerous cases reported by various observers of relapsing fever occurring in pregnant women establish the rule that abortion almost invariably occurs, whatever may be the stage of the pregnancy. In a large majority of cases the mother recovers, but the child, if viable, is stillborn or dies in a few hours. Only two of our patients were pregnant women, and the result in each was unusual. In one, the patient, already the mother of several children, was in the fifth month of gestation; the initial paroxysm was severe, with delirium, but no symptoms of abortion occurred; the intermission lasted six days, during which she felt very well; the relapse was also severe, and crisis occurred on the fifth day, the temperature falling below normal, and the case promising to do well; but on the following day there was a sudden rebound of temperature, pulse 140, severe præcordial pain, and death occurred in twenty-four hours, the contents of the uterus being partially expelled during the act of dying. In the other case, a girl of eighteen years, who had aborted at the third month of gestation eight months previously, and who was again three months advanced in pregnancy when attacked with relapsing fever, went safely through a bad attack and carried her baby successfully to full term.

MORBID ANATOMY.—The surface of the body often presents patches of livid discoloration, and jaundice persists in cases where it has been present during life. There is but little appearance of emaciation, except in cases where it has been present before the attack.

When death occurs while the temperature is high the body remains warm an unusual length of time. Thus, in one case where death occurred at 11.30 P.M., the temperature at 12 was 103°, and at 1 A.M. it was 101.6°, that of the room being 73°; at 6 A.M. it remained at 93°, the room being at 73°; between 9 A.M. and 2 P.M. the room was kept at 55°, but the body was still at 82° at the latter hour.

The voluntary muscles are often jaundiced, and in prolonged cases they may be found flabby and having undergone marked granular degeneration. In many cases, however, they remain quite dark and firm. Ecchymoses of the muscular substance are met with occasionally.

In one case, where during life there had been painful swelling of the left parotid region, with fistulous openings on the cheek, and where death occurred on the twelfth day of the disease, the masseter muscle was swollen, with patches of dark, almost black, discoloration from ecchymosis, and was studded throughout with small collections in its substance. The fluid from these contained very numerous cells indistinguishable from leucocytes. The muscular fibrils were friable and granular, and there was multiplication of the nuclei of the sarcolemma. These unusual lesions seemed to have originated in interstitial disintegrating thrombi, with consequent inflammation of the muscle.

The muscle of the heart is more frequently affected, and in the fatal cases our attention was particularly drawn to those lesions. Ponfick29 has also described them minutely. The degree of change varies from a partial loss of transverse striation, with slight granular appearance, up to a very high degree of granulo-fatty degeneration. The organ is then flabby, its substance pale gray or brownish, either wholly or in streaks, and microscopic examination shows an extreme degree of fatty granular change. It must not be forgotten, however, that many of the subjects of relapsing fever have been leading irregular and dissipated lives, and that in some instances the lesions of fatty degeneration detected in their organs may have been the result of their previous habits.

29 Virchow's Archiv. f. path. Anat., Bd. lx. Hft. 2, p. 162.

Lesions of the cardiac muscle were most marked in those of our patients who had been intemperate, and in whom fatty degeneration of the viscera (chiefly liver and kidneys) was also found. They were most fully developed in cases where death occurred at a comparatively late period, while in some very severe cases, in which death occurred as early as the fifth day, the cardiac fibre presented merely faintness of striation without actual granular degeneration.

Ponfick in particular notes that the great majority of the bodies he examined were of persons who had been habitual drunkards.

Pericarditis is occasionally present, and is marked by the usual lesions. In a very severe case in which it contributed largely to the production of the fatal result it was associated with pneumonia. In addition to this, effusions of blood beneath the endocardium and pericardium are not rare; and we have seen them quite large and numerous in cases where the muscular fibre was firmly contracted and the cavities contained quite firm decolorized clots.

Thus in our case No. 62, Series C., "the heart was normal in size, with no appearances of previous disease. There were numerous ecchymoses of both layers of the pericardium. The right cavities contained large, firm, yellowish, fibrous clots, forming a cast of the upper part of the ventricle and of the auricle, and extending both into the pulmonary artery and back into the veins, and so firm that by gentle traction a complete cast of these vessels was drawn out. The clot in the pulmonary artery was throughout firm, fibrous, and yellowish. There were numerous ecchymoses of the pleura and of the mucous membranes of the stomach and urinary bladder, hemorrhagic infarctions in the kidneys and lungs, and granulo-fatty degeneration of the cardiac muscle." Death had occurred in this case about the close of the third week, and was preceded by hematemesis and suppression of urine. We must note in this connection the tendency to embolism that exists in this disease.

Especial interest attaches to the condition of the blood in relapsing fever. Usually it presents no abnormal appearance if drawn during life, though in grave cases it may coagulate imperfectly. We have no knowledge of its minute chemical characters, save that in several cases where there was great diminution in the amount of urine, with uræmic symptoms, urea has been found in considerable amount in the blood (Murchison, p. 368). The red globules present no definite or characteristic changes. In some of our examinations they appeared of light color and became crenated very quickly on exposure. On the other hand, the white corpuscles have repeatedly been observed to be increased in number, at times considerably so (Cormack, Thompson, Zuelzer, Carter, Boeckmann, and ourselves), though this change is not regarded as constant or essential. It has, however, a very great interest in connection with the characteristic lesions of the spleen which will be described hereafter. In several cases we observed that many white corpuscles were small and apparently imperfectly developed. Boeckmann30 concludes that they increase in number during the febrile paroxysm, reaching their highest number at the crisis, and then diminishing gradually to the normal. The red globules are much decreased during the fever, and return to the normal slowly during convalescence.

30 Deutsch. Arch. f. klin. Med., Sept. 1881, p. 513.

In addition to these changes, various abnormal elements have been observed more or less constantly. By far the most important of these is the spirillum or spirochete of Obermeier, which has been already carefully described. In proportion as this organism has been carefully looked for it has been found constantly, so that the evidence has become very strong in favor of its uniform presence in the blood of relapsing-fever patients during the febrile stage of the disease.

Ponfick in 187431 called attention to the occurrence of large granule-cells in the blood in this disease. They are found during life as well as after death, when they exist in largest proportion in the blood of the splenic, hepatic, and portal veins. Their shape is spherical, ovoid, or elongated; the basis of the cells is a delicate, translucent, albuminous substance; and the granules are of a fatty nature, as shown by the action of reagents. These cells have been found by other observers, and the view is generally received that they are derived from the lymphoid elements of the spleen, and perhaps of other portions of the lymphatic system; and Carter, who has studied them carefully, is inclined to think there is some connection between them and the development of the spirillum.

31 Centralbl. f. d. med. Wissensch., 1874, p. 25.

Ponfick also first described32 certain other large, irregularly-shaped, pale, granular, nucleated cells, which occur in smaller number in the blood in relapsing fever, and which he regarded as altered endothelium, derived from the lining of the blood-vessels, of the lymphatics, or of the lacunar spaces of the spleen. Occasionally these cells are found with such highly granular contents as to make them closely simulate the large granule-cells described above. These results of Ponfick have been confirmed by other observers.

32 Loc. cit.

In several of our reports of examinations of blood there is mention made of quite abundant, free granular matter—an appearance also observed by Carter. Finally, the latter describes the occurrence of thread-like filaments and of short, rod-like bodies.

There are no characteristic lesions connected with the gastro-intestinal canal. The mucous membrane of the stomach may be normal or merely injected, though where there has been much vomiting, and especially bloody vomiting, there is marked injection, and not rarely ecchymosis and submucous extravasations of blood, with softening of the membrane. These extravasations are usually small, but Cormack reports a case where one-third of the mucous membrane of the stomach was the seat of ecchymosis and extravasation. In one of our own cases the extravasations occupied an area of four inches square.

The small intestines exhibit patches of congestion or ecchymosis less frequently than the stomach, though it is usual to find injection of the mucous membrane, especially of the lower portion, in cases where there has been diarrhoea. Carter, observing the disease in India, found in one-half of all autopsies some amount of congestion, hemorrhage, or inflammation of the ileum. In two instances he found a layer of diphtheritic deposit over the mucous membrane of the lower part of the ileum.

There are no special alterations of the solitary or agminated glands, and ulceration never occurs. Even in cases where the constitutional infection is severe, whether diarrhoea has been present or not, it is noteworthy that there is rarely any swelling of the solitary glands or Peyer's patches, such as is met with in many other acute specific diseases. It was not present in any of our autopsies.

The large intestine in like manner exhibits no characteristic lesions. Patches of congestion and occasionally submucous ecchymoses may be observed, and croupous exudation occurs here somewhat more frequently than in the small intestine.

Wyss and Bock33 speak of enlargement of the mesenteric and retroperitoneal glands as of frequent occurrence, but we did not observe it, and Murchison states that these glands present no abnormal appearance.

33 Op. cit., p. 223.

Alterations of vascularity of the brain or its membranes are met with, but they are variable and bear no definite relation to the precedent symptoms. Ecchymoses of the membranes are occasionally observed, and in one of our cases extensive meningeal hemorrhage was found. Murchison reported a case in which embolism of the left femoral artery occurred, and subsequently of the left middle cerebral artery, inducing death. The suggestion may be hazarded that in some of the cases where there is severe delirium ending in stupor and death there has been multiple capillary embolism of the cerebral vessels.

There is occasionally the evidence of catarrhal inflammation of the upper air-passages, and in some epidemics diphtheritic exudation in the pharynx and larynx has been noted (Wyss and Bock); and Ponfick found acute oedema of the glottis in a considerable proportion of the fatal cases at Berlin. The lesions of pleurisy are met with in a small proportion of cases; in our own autopsies this complication was more frequent than in most epidemics.

The lungs may be normal, and Murchison concludes that they are more frequently so than in typhus. Still, they often present congestion or oedema, and subpleural ecchymoses, hemorrhagic infarctions, and pneumonic consolidation are not rare. Lobar pneumonia was present in 33 per cent. of our own autopsies, in 28 per cent. of Carter's, and in 20 per cent. of those conducted by Ponfick. The inflammation usually presents the regular stages, and is associated with a moderate degree of plastic pleurisy; but occasionally, as in one of our cases, it terminates in gangrene. In the instance referred to there was an area of gangrene about three inches square and one inch in depth, involving the pleura and a superficial layer of lung on the antero-lateral aspect of the left lower lobe. In another remarkable instance, already referred to on account of the suppurative inflammation of one masseter muscle, the lungs, which were stained yellow throughout, presented numerous deep purplish patches, which on section altogether resembled the secondary metastatic deposits of pyæmia, with yellowish softening or even puriform centres surrounded by a rim of purplish livid discoloration. Very numerous similar patches, varying from the size of a pea to that of a hazel-nut, and presenting every stage of development, were found throughout both lungs. In a few instances we found the lesions of chronic phthisis, which had, of course, existed before the attack of relapsing fever. The bronchial glands were found swollen and infiltrated in cases where inflammatory processes in the lungs have existed.

Much interest attaches to the state of the genito-urinary organs in relapsing fever, but caution is required to distinguish lesions that have existed prior to the attack from those properly referable to it.

Owing to the intemperate and exposed lives of many of the patients, renal lesions might reasonably be expected in no small proportion. The comparative rarity of albuminuria (see [above]), even in severe cases, is suggestive of the view that when it is present it may at least sometimes be due to pre-existing lesions aggravated by the acute infectious process, and further that the extreme gravity generally presented by such cases may be in part due to the impaired condition of the kidneys.

The morbid changes most frequently referable to the fever are moderate enlargement and congestion, occasionally very intense so that we find it described in our notes as deep blackish-purple or blue; ecchymoses of the capsule or of the mucous membrane of the pelvis; small hemorrhagic infarctions, usually in the cortex; and cloudy swelling of the glandular cells. Less commonly are found hemorrhagic infarctions, or small embolic patches advanced to various stages of disintegration, even to the formation of small puriform collections. In quite rare cases the lesions of acute nephritis are present, while caution must be used in interpreting other changes occasionally met with, such as pallor with granulo-fatty degeneration or other advanced alterations of the glandular cells, or hyperplasia of the intertubular connective tissue, with or without contraction of the kidneys.

The mucous membrane of the bladder, as already mentioned, may present ecchymoses, or, more rarely, croupous exudation (Wyss and Bock). The urine contained may be bloody, or, as in one of our cases where there had been total suppression of urine for over seventy-two hours before death, there may be but a small amount of almost pure blood, containing a few phosphate crystals, but no tube-casts. In this case there were also ecchymoses of the bladder and of the pelvis of the kidneys, with intense congestion and numerous small hemorrhagic infarctions of the kidneys.

The liver is constantly though variously affected. It is found enlarged in the great majority of cases, especially if death has occurred during the febrile stage. The ordinary degree of enlargement in our cases was from four to four and a half pounds, but in a few instances the liver weighed one hundred or one hundred and two ounces, though in most of these extreme cases the patients had been drunkards, and there was such advanced fatty alteration of the liver as to make it probable that the organ had been diseased previously. These figures correspond with the results of other observers.

In many cases, especially when death occurs early and during the febrile stage, the capsule and substance of the liver are congested, at times intensely so; and when ecchymoses are found elsewhere they are apt to be present here also, appearing as purplish patches dotted over the capsule and extending into the superficial layer of hepatic tissue. Not rarely, however, the liver substance is paler than normal, and presents a yellowish tinge, apart from the decided yellowish staining present in cases attended with jaundice. Carter describes a partial mottled paleness of the liver as having been frequently observed in his cases, the circumscribed pale areas presenting a corresponding localized degeneration of the cells, as though from some local interruption of circulation.

Cloudy swelling and fatty degeneration of the liver-cells are indeed very often present, and in some epidemics with preponderance of bilious symptoms are constantly found (Ponfick). The degree of the cell-alteration varies from a slight granulo-fatty change to an advanced fatty degeneration, even with a marked tendency, in rare cases, to disintegration of the cells, so as to produce lesions analogous to those of acute yellow atrophy (St. Petersburg epidemic).

The whitish deposits described by Küttner as due to albuminous or fibrinous infiltration are probably referable to transformed hemorrhagic infarctions, and the minute puriform collections that have been observed at the centre of the acini (Wyss and Bock) may have been metastatic in origin, or attributable to the disintegration of minute thrombi associated with irritative hyperplasia of the adjacent lymphoid elements. The consistence of the liver varies: when death occurs early and bilious symptoms have not been marked, it may be even firmer than normal, but more frequently it is softer, and it may be relaxed, flabby, and friable.

The condition of the bile-ducts is of great interest in view of the frequency of jaundice as a symptom in relapsing fever, and most authorities unite in saying that they present no lesions capable of explaining it.

The gall-bladder is usually found full of dark bile, but there is no such degree of inspissation, except in rare instances, as could interfere with its passage through the ducts. Murchison quotes the statement of Peacock that in some instances the bile was thick and viscid, so as apparently to cause obstruction, but all observations agree in showing that this is exceptional. The mucous membranes of the larger ducts may present evidences of slight catarrhal inflammation, but in nearly all cases where they have been carefully examined, even when jaundice had been marked, they have been found patulous and free, so that the jaundice cannot be regarded as due to obstruction of the larger ducts save in rare instances (Pastau). In further confirmation of this may be stated the fact that there is no want of bile in the duodenum and feces.

On the other hand, a careful consideration of the lesions of the substance of the liver will show that it would be most improbable that the minute biliary ducts in the areas most affected should escape implication. Münch, who investigated this subject carefully, found that there was a catarrhal state of the fine bile-ducts in every case of relapsing fever with jaundice; and Litten found the smallest ducts plugged with bile-stained pellets of mucus. It would appear, therefore, that in many cases at least the jaundice is really obstructive in its origin, the seat of the obstruction being in the too-rarely examined minute bile-ducts, though further investigation of this interesting question is required.

The clinical bearing of these conditions has been fully discussed in the appropriate section.

The changes in the spleen are constant, and even more remarkable than those in the liver. It is enlarged with rare exceptions, and especially so if death has occurred during the febrile stage. Upon the subsidence of the fever the spleen probably returns to its normal size more rapidly than the liver. The more common extent of the enlargement in our own cases was from ten to eighteen ounces, though we found the spleen in one case weighing twenty-nine and a half ounces and in another forty-four and a half ounces. In neither of the latter instances was there any reason to suspect malarial complication. The most extensive enlargement we have found recorded is sixty-eight ounces in a case reported by Küttner.34

34 Schmidt's Jahrb., 1865, vol. cxxvi.

There is usually a correspondence between the stage and extent of the splenic and hepatic lesions, but this is not invariable, and one or the other organ may present a far higher degree of enlargement or much more intense interstitial changes. It may be mentioned, moreover, that in some unusual cases the lesions of the lungs, such as ecchymoses and hemorrhagic infarctions, may be disproportionately marked as compared with those of either the liver or spleen.

The capsule of the spleen often presents a mottled look, with at times large purplish ecchymoses; it is apt to be more or less opaque, and local peritonitis, with thin layers of plastic exudation often forming friable adhesions with the abdominal wall, may exist.

In one of our cases the capsule presented a small perforation or rupture, with an exudation of plastic lymph over an area of four by six inches, and diffuse peritonitis, with effusion of bloody liquid with shreds of lymph throughout the abdominal cavity. This fatal termination is fortunately rare, but there are several other instances on record. The splenic pulp may retain its consistency and firmness, even in cases that have run a long course; but more frequently it is softened, and may be almost diffluent. The pulp is often swollen, so that when cut it projects above the section. The color is darker than normal, and often is of a deep maroon color. This swelling is due to enlargement of the blood-vessels, associated with great increase of the cellular elements of the pulp and with enlargement of the Malpighian corpuscles.

When death occurred early in the disease we found these bodies grayish or grayish-yellow in color and of the size of hempseed, so that the section very thickly studded with them closely resembled shad-roe, and this stage of the lesion is frequently described in our notes as the shad-roe spleen. Subsequently, the Malpighian bodies enlarge still more, and stand out above the section a line or more in diameter, and of a lighter color; not rarely, several of them come in contact, and thus form a considerable mass of irregular shape, resembling the infarctions described below.35 It is probable that central softening may occur later in the Malpighian bodies, though we are inclined to regard the puriform collections frequently found as chiefly due to the disintegration of hemorrhagic infarctions or of embolic patches. Of these, hemorrhagic infarctions are by far the most common and present the familiar appearances. They may be quite numerous, superficial, or deep-seated, and of variable shape and size. At first dark reddish, firm, and sharply separated from the surrounding pulp, they grew reddish-yellow or yellowish later, softened in the centre, and eventually were transformed into puriform collections. Doubtless, in a large proportion of cases that recover such infarctions exist and are slowly absorbed. Ponfick has shown that these are venous infarctions, the arterioles leading to them being patulous. True arterial embolism does, however, occur, though much more rarely (Ponfick, Murchison), giving rise to firm, wedge-shaped infarctions at the periphery of the spleen, which may undergo degenerative changes similar to those above described. The resulting abscesses may burst into the peritoneum, pleura, lung, or bowel. The microscopic appearances have been most fully described by Ponfick, our own comparatively meagre observations having accorded entirely with his subsequent accurate description. The cells of the swollen pulp contain red blood-discs and pigment, and some present collections of bright granules. The lymphoid cells of the Malpighian corpuscles are at first in a state of cloudy swelling with multiplication of their nuclei, and later show marked granular fatty degeneration.

35 Thus, Wyss and Bock describe "multitudes of minute abscesses as large as poppy or hempseed, and containing a single drop of pus."

The lymphatic glands present no lesions, and the pancreas is normal.

The peritoneum is not affected as frequently as other serous membranes in this disease. Superficial ecchymoses are, however, quite common, especially so over the solid viscera; and more rarely effusions of blood have been found in the subperitoneal connective tissue, involving the muscular or glandular tissues beneath. We have already mentioned ([above]) the occasional occurrence of local peritonitis, most frequently of the splenic capsule, and also the rare accident of diffuse inflammation from rupture of the spleen.

The marrow of the bones was carefully examined by Ponfick, who first called attention to the presence of important changes in relapsing fever, which have since been confirmed by other observers. These changes consist in proliferation and subsequent degeneration of the lymphoid cells of the marrow, with multiplication of the nuclei in the walls of the minute vessels and fatty degeneration of their coats. As a result of these changes, spots of puriform softening may form, chiefly in the cancellous tissue of the extremities of the long bones, with the production of localized necrosis, and possibly with extension of inflammation to the neighboring articular cavity.

Considerable space has been devoted to the detailed consideration of the pathological changes in relapsing fever, partly because we believe the fact has not been sufficiently recognized that the disease is constantly attended with important and characteristic lesions. These consist, in brief, of remarkable changes in the blood; of widespread ecchymoses and infarctions, which not rarely undergo puriform disintegration; of hyperplasia and subsequent degeneration of the Malpighian corpuscles of the spleen, with changes in the cellular elements of the splenic pulp; of cloudy swelling of the gland-cells of the liver and kidneys, with a marked tendency to fatty degeneration; of changes in the marrow of the long bones; and, finally, of granulo-fatty degeneration of the muscles, and especially of the heart.

DIAGNOSIS AND RELATION TO OTHER DISEASES.—The entire question of the diagnosis of relapsing fever is dominated by that of spirillar infection. Before Obermeier's discovery the differential diagnosis of the initial paroxysm, and to a less extent that of the subsequent events of a case of relapsing fever, was attended with considerable difficulty. But if, as now seems established, immediately before and throughout the initial paroxysm and subsequent relapses a characteristic spirillum is to be detected in the blood upon proper examination, while it rapidly disappears after the crisis, it is evident that as soon as a suspicion is aroused as to the possible presence of relapsing fever the question may be settled conclusively by the microscope.

None the less is it important to consider carefully, but briefly, the symptoms by which relapsing fever is to be distinguished from various affections which may simulate it, because even the most experienced observers admit that the spirillum cannot be invariably detected; because it is not yet known that a similar organism may not be found in some other affections; and, finally, because on the outbreak of an epidemic of relapsing fever, especially in America, where its occurrence has hitherto been so rare, there is strong probability that the nature of the early cases will not be even suspected until the relapse occurs.

Typhus fever often prevails in an epidemic form simultaneously with relapsing fever, so that it was inevitable they should have been for a time confused. Their essential non-identity is, however, now too well recognized to require any lengthy demonstration. The following statement of the heads of the argument may therefore suffice.

In typhus there is no characteristic spirillum, and the lesions which are truly characteristic of relapsing fever are totally wanting. There are convincing differences in the symptoms, course, and results of the two diseases. There is no evidence to show that when fever has been imported into a locality by a single case, typhus fever has ever produced other than typhus, or relapsing other than relapsing fever. The two diseases often prevail together, and may coexist in the same house, each preserving its own distinct characteristics; and persons exposed to the double contagion may contract one or the other, or first one and then the other at a shorter or longer interval, so that an attack of either exerts no protective power against the other. It must be noted, however, that in a large majority of such cases of successive contagion it is relapsing fever which has been followed by typhus, while the reverse has been observed much more rarely.

In 1869-70 the two diseases were prevalent in Philadelphia, and the wards of the municipal hospitals constantly contained a considerable number of cases of both. Three instances came under our care in which after recovery from relapsing fever the patient contracted typhus. All of these patients were employed as assistant nurses, and were continuously under observation from the early part of their attack of relapsing fever to the end of the attack of typhus. In one case the interval of health between the close of the relapse and the onset of typhus was forty-four days; in the second it was thirteen days. In both cases the original disease was thoroughly characteristic and the subsequent attack of typhus was typical. In both death followed, and the post-mortem examination verified the above statement. The third patient had severe relapsing fever, from which he recovered and returned to work, though with pains in the legs, shoulders, and forehead. After an interval of apparent health of eleven days he developed a well-marked attack of typhus, which terminated on the twelfth day in recovery. It may be added that although typhus is not of frequent occurrence in any portion of North America, there have been a number of epidemics unattended with a single case presenting the features of relapsing fever.

Between well-marked cases of the two diseases there should be no difficulty in making a prompt diagnosis. Relapsing fever is distinguished from typhus clinically by the severity of the initial chill; the rapid elevation of the pulse and temperature; the comparative infrequency and mildness of cerebral symptoms, despite the intense fever; the severity of the gastric symptoms, nausea and vomiting; the enlargement of the liver and spleen, with marked abdominal pain and soreness; the frequency of jaundice, of epistaxis, and of other hemorrhages, and of anæmic murmurs over the heart and large vessels; obstinate insomnia; vertigo; peculiar rheumatoid pains and perversions of sensation; the frequency of sweating during the high pyrexia; by the occurrence of crisis, subnormal temperature, apyretic interval, and relapse; the rarity of measly eruption and of bed-sores; the frequency of pneumonia, diarrhoea, ophthalmia, oedema, and desquamation as complications and sequelæ; the usual occurrence of abortion in pregnant females; the protracted course of the disease, and its remarkably low mortality despite the severity of the symptoms, except in cases of complicated or typhoid type; and, finally, by the modes in which death occurs. Of course to this must be added the specific result of examination of the blood in relapsing fever.

Doubt will arise only in very rare cases where a measly eruption appears on or before the fifth day of relapsing fever, with headache and mild delirium, but without severe gastric symptoms, epistaxis, or jaundice. If no relapsing fever were prevalent at the time, such a case might well be regarded as one of mild typhus until the crisis and the relapse disclosed its real nature. But if the two diseases were known to be prevalent in the community, examination of the blood would properly be made at once and the diagnosis be established.

The diagnosis between ordinary cases of relapsing fever and typhoid is readily made by the gradual onset and peculiar course of the pyrexia in the latter disease, as well as by the frequency of delirium, of abdominal distension, and of diarrhoea, and by the characteristic eruption. The occurrence of epistaxis, bronchial irritation, and splenic enlargement is common to both, and an eruption of small rose-pink spots has been noted by some observers (Carter, pp. 194, 317). But jaundice, enlargement of the liver, hypochondriac pain and soreness, excessive nausea and vomiting, severe rheumatoid pains, and numbness and tingling of the extremities, are very significant symptoms of relapsing fever. Attention has already been called to the grave type of relapsing fever in which the typhoid state is fully developed, and to the fact that in such cases the pyrexia is often modified, the onset less abrupt, the crisis imperfect, and the interval occupied by an irregular post-critical symptomatic fever. It is altogether probable that such cases have not rarely been regarded as of true typhoid character; and indeed the attempt has been made by Griesinger to establish as a separate and independent affection, under the name of bilious typhoid fever, a group of cases which close examination seems to show to be chiefly composed of grave complicated relapsing fever with a certain proportion of true typhoid fever, complicated with jaundice.

The recognition of the bilious typhoid type of relapsing fever is based upon the history of the case; the mode of onset; the greater severity of the pains, arthritic and abdominal; the early appearance and intensity of the jaundice; the more marked enlargement of the liver and spleen; the marked tendency to hemorrhages from various surfaces; the peculiarities which careful study of the temperature curve will show, especially about the time of crisis; the rarity of eruption; the characteristic spirillum;36 and the totally different anatomical lesions, which are, unfortunately, often demonstrable, as this form of relapsing fever is fatal in from 33 to 50 per cent. of cases.

36 As first demonstrated by Motschutkoffsky.

Since the discovery of the spirillar test for relapsing fever it may be said that Griesinger's bilious typhoid must be stricken from medical nosology as an independent affection.

The case of Charles Hood, [above], is a good example of the bilious typhoid form which occurred not rarely in the Philadelphia epidemic.

Murchison points out that, owing to the frequent occurrence of jaundice in relapsing fever, this disease has been mistaken for yellow fever by such good observers as Graves, Stokes, and Cormack. Difficulty in diagnosis would be likely to arise only in regard to the bilious typhoid type of relapsing fever, and since its clinical history has become so well known, a mistake is not likely to occur. The geographical distribution of the diseases is widely different. Yellow fever is influenced powerfully by season and temperature, while relapsing fever is independent of both. Negroes are but slightly liable to yellow fever, while relapsing fever attacks them with special violence. Yellow fever is not contagious, but infectious, and second attacks are extremely rare; relapsing fever is one of the most contagious of the zymotic diseases, but one attack does not protect against a subsequent one. The mortality, the anatomical lesions, the course of the pyrexia, the leading clinical symptoms, are all widely distinct in the two affections; and, finally, no spirillum has been found in the blood in yellow fever. Yellow fever is an extremely fatal disease; the ordinary form of relapsing fever has a mortality of 2 to 10 per cent.; the bilious typhoid form, one of 33 to 50 per cent. In yellow fever the spleen is but slightly enlarged, and the liver is pale and softened; in relapsing fever the liver and spleen are greatly enlarged, and there is great tenderness over the hypochondriac region. In yellow fever albuminuria is much more common, and the urine more frequently suppressed, than in relapsing fever.

The sudden onset, the severe headache and pains in the limbs, the vomiting, jaundice, epigastric tenderness, enlargement of the liver and spleen, occasional epistaxis, hematemesis, or hematuria, absence of characteristic eruption, liability to herpes facialis, pneumonia, and diarrhoea; the occasional occurrence of remissions in the pyrexia, and even of more or less fully-developed chills for several successive days during the initial paroxysm or the relapse, suffice to explain the difficulty which may arise in distinguishing the bilious form of relapsing fever from bilious remittent fever. But the latter disease arises exclusively from malaria, and is therefore powerfully influenced by season and locality; is not contagious; does not present anything approaching to the crisis, the apyretic interval, or the abrupt relapse of relapsing fever; presents pigmentary changes in the blood, instead of the spirillum; and lesions of the spleen and liver totally unlike those characteristic of relapsing fever; can be promptly controlled by antiperiodic doses of quinine, and therefore should have a mortality far less than that of the grave form of relapsing fever. It is not necessary to pursue this subject further, but a reference to the temperature charts of Carter37 or of Litten38 will show that in some epidemics single paroxysms resembling those of quotidian ague might occur during the interval between the initial paroxysm and the relapse, or a series of two, three, or more such paroxysms of quotidian or tertian type might represent an entire relapse. Such phenomena are wholly uncontrollable by quinia, and are presumably dependent upon irregularities in the specific infection, instead of upon a blending of malaria with the poison of relapsing fever. There is some ground for believing, however, that those who have recently passed through an attack of the latter are highly, perhaps unusually, susceptible to malarial infection, as we have already seen they are liable to contract typhus.

37 Op. cit.

38 Deut. Arch. f. klin. Med., xlii. 1874.

The chill, the sudden and high fever, the acid sweat, the high-colored urine, the intense pains and soreness, and the occasional murmur over the heart, will in some cases of relapsing fever suggest the idea of severe rheumatic fever, with illy-developed articular inflammation and with a tendency to hyperpyrexia. The urgent danger presented by the latter condition and the necessity for immediate recourse to cold baths and large doses of quinine or of the salicylates, render it highly important that no such error of diagnosis should be made. It will usually be avoided readily by observing that in relapsing fever there are great nausea, repeated vomiting, insomnia, peculiar formication of the extremities, jaundice, early enlargement of the liver and spleen, with abdominal pain and soreness, and a tendency to epistaxis; and, further, that despite the high temperature, cerebral symptoms such as result from rheumatic hyperpyrexia are not threatened, except in grave typhoid cases or just preceding the crisis.

The onset of relapsing fever may suggest forcibly the invasion period of small-pox, with its marked rigors, high fever, lumbar pain, aching in the head and limbs, nausea and vomiting, and if the patient is known to have been exposed to the contagion of both diseases a diagnosis would be impossible until the third day. But such a dilemma can rarely occur, and under ordinary circumstances the patient's antecedents will enable a correct opinion to be formed.

Severe cases of simple febricula with marked gastric disturbance may, as remarked by Jenner, closely simulate relapsing fever; and the same is true of attacks of acute gastro-hepatic catarrh, with severe headache, sharp fever, cholæmic eye, epigastric tenderness, and frequent vomiting. Of course there is no danger under ordinary circumstances of these simple conditions being regarded as relapsing fever, but when the latter is prevalent in epidemic form it is probable that the mistake is frequently made. Although an immediate diagnosis might be possible only by microscopic examination of the blood, the peculiar clinical symptoms of relapsing fever would soon be found wanting, and suitable treatment would bring the simpler affection under control.

Acute yellow atrophy of the liver occurs chiefly in pregnant women, though it is also met with in men and children; but it is so rare that should a case of it come under observation during the prevalence of relapsing fever there is considerable danger that its nature would be overlooked. It resembles relapsing fever in the occurrence of jaundice and other signs of hepatic disorder, of delirium, and of a tendency to hemorrhage from various surfaces. The temperature, however, is more moderate, and does not exhibit the sudden remission of relapsing fever; the liver is usually demonstrably diminished in size; severe nervous disturbances, such as convulsions followed by stupor and then by coma, are more constant; while the occurrence of spirilla in the blood of relapsing fever and of leucin and tyrosin in the urine of acute yellow atrophy serves to distinguish completely the two diseases. Acute yellow atrophy is, moreover, invariably fatal.

With ordinary care there is but little danger that any of the local complications of relapsing fever will so absorb attention as to lead to a neglect of the specific general disease, so that the cerebral symptoms should be readily distinguished from the onset of any acute intracranial affection; the parotitis which occasionally appears early in the disease should not be confounded with idiopathic mumps; and so for other complications. There is far more danger, indeed, lest some of the complications may be overlooked; and this is especially true of pneumonia, one of the most frequent and most important of them all. Its occurrence is the cause of the supervention of grave typhoid symptoms or of the modification of the normal course of the pyrexia in so many cases that nothing but a systematic daily examination of the lungs will avert serious oversights.

MORTALITY AND PROGNOSIS.—The rate of mortality has varied in different epidemics from 2 or 3 to 24 per cent. Murchison shows that out of 2115 cases admitted to the London Fever Hospital during a period of twenty-two years, and embracing two distinct outbreaks, only 39 proved fatal, making 1.84 per cent. mortality. Adding to these the results of Scotch and Irish epidemics, a total of 18,859 cases, with 761 deaths, is reached, giving the rate of mortality for Great Britain as 4.03 per cent. The great Indian epidemics studied by Carter gave 111 deaths out of 616 cases, equal to 18.02 per cent. Recent German epidemics have given from 5 to 10 per cent. The above rates are obtained where all the cases observed during an epidemic are included. If, however, the mortality of the ordinary form of relapsing fever is computed separately from that of the bilious typhoid form, it does not exceed 2 to 5 per cent., whilst the mortality of the latter form rises to from 33 to 50 per cent., or even higher.

In the Philadelphia epidemic, out of a total of 1174 cases there were, as nearly as can be ascertained, 169 deaths, giving a rate of mortality of 14.4 per cent. Taking all the cases admitted to the hospital under our observation, many of which entered at a late period of the disease and not a few when moribund, the mortality was not less than 13 per cent. The mortality among the negroes who were attacked with the disease was considerably greater than among the whites. Finally, if the mortality of the bilious typhus form be considered separately—although from the frequency of jaundice in this epidemic and the numerous gradations of severity presented it is difficult to form a sharply defined group of this character—it was certainly not less than 50 per cent.

The date of death varies with the epidemic, the form of the disease, and the previous condition of vitality of those attacked. Ordinarily, by far the larger proportion of deaths occur during the first relapse or the second interval, but in bilious typhoid cases, presenting grave complications, especially pneumonia or severe hemorrhages at an early date, or in cases occurring in intemperate subjects, or in those previously in impaired health, the mortality is much heavier in the initial paroxysm or the first interval than at later periods.

Youth exerts the same favorable influences upon the result of relapsing fever as it does in the case of typhus and typhoid. Murchison states that of 717 male patients under twenty-five years of age admitted into the London Fever Hospital, not one died, and in most epidemics similar, though not equally marked, results have been noted. In some epidemics the mortality among young children has been considerable. As a rule, the percentage of deaths increases with each decade after thirty years.

Sex does not exert any definite or constant influence upon the mortality. The number of males affected is far greater; they are liable to be exposed to the contagion in its most concentrated form; a larger proportion of them are probably the subjects of intemperance than in the case of females; and thus most statistics agree in making the mortality somewhat greater in the male sex; but, all things being equal, there is no good reason for holding that sex itself has any value in determining the result.

As in other zymotic diseases, the mortality from relapsing fever is highest during the early period of an epidemic, and the type of the disease grows milder as the epidemic declines. Cases of the bilious typhoid form have become notably less frequent during the later stages of some epidemics than at an earlier period.

Marked difference has been observed also as to the action of remedies at different stages of epidemics, the early cases exhibiting an extraordinary resistance to remedies, and especially to anodynes, which passes away later. When typhus and relapsing fevers have prevailed together, and a clear discrimination between the two sets of cases has not been made, it has appeared that the mortality increased as the epidemic advanced, but this apparent exception has been due to the fact that at first the cases of relapsing fever were in the majority, while later those of typhus, the much more fatal disease, preponderated.

Epidemics of relapsing fever prevail at all seasons, but more commonly they are at their height during the colder months of the year. The total mortality will of course correspond, but the actual percentage is not constantly greater during any one season, although it is probable that the greater liability to chest complications during the colder months will render the disease more fatal then.

The gravity of relapsing fever has varied so greatly in different epidemics that it is very difficult to determine what influence upon the mortality has been exerted by mere difference of race. A further source of difficulty is found in estimating the differences in the physical conditions of the poorer classes in the various communities affected. The mortality has been exceptionally high in the Russian and Indian epidemics and in some of the German ones, while in the British epidemics it has uniformly been light. It is interesting to note that in the Philadelphia epidemic, where the great majority of patients were Irish or negroes, the mortality was high, over 14 per cent. The previous condition of the Irish patients must certainly have contrasted favorably with that of the individuals attacked in the Dublin and Belfast epidemics, so that the difference in result seems attributable only to a greater virulence of the disease. As an ample opportunity was here afforded to judge of the relative severity of relapsing fever in the negro and white races when the cases occurred at the same season, at the same stage of the epidemic, and in individuals living under nearly similar conditions, it may be stated that the conclusion of all who studied the question closely was that the disease was much more severe among negroes, and in particular that they displayed a greater tendency to serious complications and to the bilious typhoid form.

Although the degree and virulence of the infection undoubtedly constitute the most important elements in determining the mortality, the previous health and habits of those attacked with relapsing fever exert an influence upon the result. This is especially true of habitual intemperance, which, by disposing to disease of the liver and kidneys, greatly increases the liability to a fatal result. It has been seen ([above]), however, that even when acute catarrhal nephritis existed at the time of the attack severe relapsing fever might terminate favorably. Another observation which we made frequently, and which coincides with what is well known in regard to typhoid and typhus, is that improper exertion and exposure during the stage of incubation and immediately after the invasion produced a highly unfavorable effect on the subsequent course of the disease, and seemed in particular to dispose to dangerous or fatal collapse at the critical periods.

Apart from these general considerations, there are many special points to be considered in regard to the prognosis of relapsing fever:

If after the crisis of the invasion there is not rapid and decided improvement, complications should be suspected.

A sharp rebound of temperature quickly following crisis may be followed by speedy death.

Mere elevation of temperature during the invasion and the relapse, even though to an extreme height, is not attended with the danger which even a somewhat lower degree would indicate in other zymotic diseases.

Increased elevation toward the expected time of crisis should arouse anxiety, as sudden and dangerous cerebral symptoms may occur.

Prolonged duration of the pyrexia, or the substitution of irregular gradual defervescence (lysis) for the characteristic crisis often associated with typhoid symptoms as are these conditions, is significant of complications and of danger.

Wild delirium during the pyrexia, or transient active delirium about the time of crisis, is not necessarily unfavorable, but continuous low delirium, with disposition to stupor, is associated with a typhoid tendency and is frequently followed by death. Excessive muscular tremor or convulsions are highly unfavorable, but not necessarily fatal, symptoms.

Cardiac murmurs are not of serious import. The pulse is not usually as rapid in proportion to the temperature as in typhus or typhoid, and an excessively rapid pulse toward the expected time of crisis, especially if associated with feebleness of the heart's action, points to the danger of sudden collapse at or soon after that time. Previous cardiac disease, especially fatty degeneration in habitually intemperate persons, increases this danger. Continued frequency of pulse after the crisis indicates some complication or the danger of some accident.

Cough of a bronchial origin is not a specially unfavorable symptom, but if associated with the physical signs of pneumonia and with marked disturbance of respiration it indicates extreme danger.

Epistaxis, even when copious, often occurs in favorable cases, but hemorrhage from the stomach or the kidneys is usually, though not invariably, followed by death.

An eruption, measly or of pink spots, with or without minute petechiæ, is rare, and usually occurs in severe cases, but is not of specially unfavorable significance unless associated with the typhoid state or with patches of purpura.

Hiccough is a much less unfavorable symptom in relapsing fever than in typhoid or typhus, and vomiting, even frequent and persistent, may occur in cases of ordinary severity.

Enlargement of the liver and spleen indicates special risk only when persistent for some time after the relapse, in connection with persistent irregular fever. Jaundice has no necessarily unfavorable signification, is frequent in ordinary cases in some epidemics, but when it is associated with the other features of the bilious typhoid form the danger is extreme, at least 33 per cent. of such cases proving fatal.

Slight transient albuminuria may exist without special danger, but if associated with evidences of catarrhal nephritis, or if extreme diminution of urine, with or without albuminuria, exists, cerebral symptoms are apt to ensue, with a high degree of danger.

All serious complications—parotitis, erysipelas, dysentery, abortion, pneumonia, and, above all, peritonitis—greatly increase the risk.

It is not possible to determine in what cases the relapse will fail to occur. Motschutkoffsky's statement, that when a slight post-critical rise occurs a relapse will follow, must be applicable only to a limited number of cases.

In all cases at least one relapse must be expected; the patient in the interval must be regarded as still sick, and after the close of the relapse he must still be treated with rigid care until convalescence is permanently established. It must be remembered in hospital practice that many patients enter toward or after the crisis of the first paroxysm, so that caution is needed in estimating the effect of remedies and the period of the disease.

The undue prominence of certain conditions during the course of the disease is apt to be followed by corresponding sequelæ, and emaciation, anæmia, dyspepsia, diarrhoea, dysentery, enlargement of the spleen and rheumatoid pains may then be anticipated. The liability to ophthalmia and affections of the middle ear is not to be forgotten.

CAUSES OF DEATH.—In fatal cases death occurs from exhaustion dependent on the protracted and severe sufferings of the patient; from cerebral symptoms; from hyperpyrexia; from the virulence of the toxæmia; from uræmic poisoning; from sudden collapse; or from some complication, such as hemorrhagic meningitis, hemorrhages, pneumonia, dysentery, rupture of the spleen, peritonitis, or abortion.

TREATMENT.—The indications for treatment presented by regular cases of relapsing fever seem to be—to moderate the pyrexia; to relieve distressing symptoms, especially pain, insomnia, and gastric irritability; to sustain the strength of the system; to prevent or modify the relapses; and to avoid complications and sequelæ.

It is needless to observe that until the nature of the specific cause of relapsing fever is fully determined, whether the spirillum occupy that relation or not, it is impossible to direct our efforts rationally toward its neutralization or elimination. The various remedies which have been employed for these special purposes have no clinical support to recommend them. And while experiment has shown that the activity of the spirillum is readily destroyed by the direct action of various weak solutions, as of quinine, carbolic acid, iodine, and mineral acids, no special curative effect follows the internal administration of these remedies, even in the largest doses consistent with safety. In fact, there can scarcely be any disease in which treatment is less satisfactory or its results more difficult to estimate. The marked difference between various epidemics, and the wide variation presented by the development of individual symptoms in different cases of the same epidemic, fully account for this.

Quinine, as might be expected, has been largely used, in the hope that it might control the pyrexia or prevent the relapse. Murchison39 quotes a considerable amount of evidence from various sources to show that it does not possess either of these powers. It was administered to a considerable number of our cases, either in small and frequently repeated doses during the pyrexia or the intermission, or else in large doses repeated several times in immediate anticipation of the expected time of the relapse. Thus in some cases three grains of sulphate of quinia were given every two or three hours until tinnitus was produced, and then this was maintained during the remainder of the pyrexia and of the intermission. The amount given daily was from thirty to forty-two grains. It seemed to rather increase the discomfort in the head, and in some cases it aggravated the irritability of the stomach. The pyrexia was certainly not controlled by it. Given in the same manner during the intermission, it was usually well borne, but was not effectual in preventing the relapse. It is true that in some cases the subsequent relapse seemed to be somewhat modified.

39 Op. cit., p. 408.

Thus in one case 30 grains were given on the 6th of April; 39 grains on the 7th; 39 grains on the 8th; 42 grains on the 9th; and 60 grains on the 10th; the critical fall had occurred during the night of the 7th, and the relapse began on the evening of the 9th, but the rise in temperature was less abrupt than usual, and the relapse lasted less than five days. It was quite severe, however, so that it is doubtful whether the apparent modification was anything more than is frequently observed in cases where no quinine has been administered.

In another case the fall in temperature at the end of the first paroxysm was from 105.5° to 97° on March 26th: 35 to 40 grains of sulphate of quinine were given daily on April 4th, 5th, 6th, 7th, and 8th; the temperature began to rise on the 3d, but the severe pyrexia and the usual symptoms of the relapse were limited to a period of less than thirty-six hours. This is a less common irregularity, and yet does not afford sufficient evidence of the efficiency of quinine. In other cases, however, as already stated, no appreciable effect followed its administration in this manner.

To illustrate the other method of giving quinia, a case may be quoted in which 20-grain doses every three or four hours were given from April 25th to April 29th, so that in four days 575 grains were taken. The initial paroxysm was of average severity, and terminated at the end of the seventh day, April 20th. The quinine did not postpone the relapse, which occurred on April 28th, but was of much less than the usual duration.

In no other case in which these large doses were given was there even as much reason as in the above instance to attribute to quinine any positive influence upon the course of the disease.

In order to demonstrate that the failure of quinine was not dependent upon a want of absorption, Muirhead injected large amounts subcutaneously with no better results.

In conclusion, it may be said that the evidence shows positively that quinine possesses no specific influence whatever upon relapsing fever; that in only occasional cases, if at all, will even enormous doses given during the intermission postpone or modify the subsequent relapse; and that it is not effective in reducing the temperature. In view, therefore, of the usual gastric irritability and tendency to vertigo and headache, which seem to be increased by large doses of quinine, and, further, in view of the small mortality, and of the fact that when death occurs it usually comes from causes over which large doses of quinine could exert no influence, it seems clear that this drug should be prescribed only in tonic doses and only in cases where it is well tolerated by the stomach.

Arsenic was used in a considerable number of our cases with the view of determining if it possessed any power of relieving the severe pains or of influencing the relapse. It was administered in the form of Fowler's solution (Liq. potassii arsenitis), and was given exclusively by the mouth. If given during the intermission, it was well borne in doses of five to ten drops every four or even every three hours, given freely diluted with water and immediately after food. In several cases it quickly induced puffiness about the eyes, but no effect whatever was produced on the pains or on the succeeding relapse. In more than one such case there was an unusually profuse crop of sudamina during the relapse, many of the vesicles breaking and being followed by brownish stains. When given during the pyrexia it aggravated the nausea and vomiting, so that it had to be suspended. In one unfortunate case, indeed, although promptly suspended, the arsenical solution seemed to have assisted in the establishment of vomiting and purging, which proved uncontrollable and contributed greatly to the fatal result. Hypodermic injections of arsenic have been used considerably with no better results. There seems, therefore, to be no reason whatever for any further use of this drug in relapsing fever.

The high pyrexia and the severe rheumatoid pains have naturally suggested the use of salicylic acid and the salicylate of soda. We were not sufficiently aware of their antipyretic properties in 1869-70 to have recourse to them, but in more recent epidemics Unterburger40 and Riess41 have found that large doses of the latter substance (one hundred grains or more daily) will reduce the temperature either in the initial paroxysm or in the relapse, but that the disease is not cut short nor are the lesions of the blood or solids prevented.

40 Jahrb. f. Kinderheilk., v. x., 1876.

41 Deutsch. Med. Wochnsch., Dec., 1879.

It must be borne in mind here, as in connection with the action of quinine, that apparent modifications of the relapse are to be viewed with great distrust, since such great irregularities therein naturally present themselves. Care must further be taken lest such attempts to reduce the temperature aggravate the irritation of the stomach, and by lessening the power of taking food induce more serious exhaustion than would have resulted from the unchecked pyrexia. The evidence in our possession is not sufficient to justify a positive decision as to the therapeutic value of the salicylates in relapsing fever, but, apparently, they are applicable to only a portion of the cases, and in these are of but limited utility.

The same failure which has followed the use of quinine, of arsenic, and of salicin and the salicylates has attended the effort to prevent the relapse by berberine, benzoate of soda, tincture of eucalyptus, and other reputed antiperiodics.

Digitalis, veratrum viride, and aconite were used by us quite freely as antipyretics. The first two of these were often suspended on account of the irritability of the stomach, and no valuable results followed their use when well tolerated. Aconite in small doses, frequently repeated, as one drop every two hours, seemed to aid in allaying nausea and to exert some slight influence upon the fever. In cases where there was a distinct tendency to heart-failure, digitalis was given freely with advantage.

Cold baths were not used to reduce the temperature in any of the cases under our observation. They have been employed in other epidemics, but, as far as we know, with no other effect than to cause merely temporary lowering of temperature, without any decided relief to the other symptoms and without any apparent influence upon the course of the disease. Frequent spongings with cool water and the application of ice to the head gave only slight and temporary relief.

Simple febrifuge remedies, such as effervescing draught or spirit of nitrous ether with solution of acetate of ammonium, were well received by the stomach, and appeared to promote perspiration and the more free secretion of urine.

Finding all our efforts to control the pyrexia so unsuccessful, recourse was had in a large proportion of our cases to the hyposulphite of soda, given, dissolved in two ounces of water, in doses of twenty grains every two or three hours. In two cases it seemed to increase nausea, and at times it caused some purging, but otherwise it was well borne by the stomach, and, indeed, frequently appeared to aid in controlling vomiting. The records show that this drug was given in only two or three of the fatal cases, so that although the patients who took it regularly presented every grade of severity of the disease, they did well uniformly. It is certain, however, that the hyposulphite of soda exerted no specific effect upon the disease; it did not reduce temperature, it did not prevent or modify the relapses nor relieve the severe pains; it may have promoted more free and healthy secretions, and, by tending to prevent vomiting, may have aided in maintaining nutrition; but, on the whole, it may fairly be doubted whether this remedy merits any more extended trial.

One chief reason of the failure of antipyretics in relapsing fever is to be found in the existence of widespread irritative lesions of the glandular and mucous tissues, which combine with the specific blood-changes in causing and maintaining the high temperature. It is not surprising, therefore, that the remedies which afford the greatest relief in this disease are opiates and sedatives to the gastro-intestinal mucous membrane. Opium, or morphia, must indeed be regarded as the basis of the rational treatment of relapsing fever. It is called for by the insomnia, the severe headache and the pains in various parts of the body, the nausea and vomiting, and the pyrexia. It does not appear to have been as prominent a feature in the treatment of other epidemics as we found it necessary to make it in Philadelphia. Parry42 used it very freely, chiefly in the form of opium, by the mouth, and found a singular tolerance exhibited by his patients, several of whom took as large a dose as three grains every two hours during the afternoon and night without producing any sleep or even any contraction of the pupils. This resistance to the action of opium was observed chiefly in the early part of the epidemic, and we may add that it was exhibited chiefly when opium was given by the mouth. When morphia was used hypodermically we found that one-fourth of a grain, given at intervals of six to twelve hours, afforded very great relief to the pains, aided and relieved vomiting, and often induced quiet, refreshing sleep. Its use was not contraindicated by jaundice, by cough or pulmonary congestion, or by moderate contraction of the pupils. It was frequently given so as to maintain decided drowsiness throughout the pyrexia. When the pains persisted during the intermission the morphia was continued in smaller doses or at longer intervals. It occasionally happened that when patients were thus kept continuously under opium influence no relapse occurred; but here, as in regard to the action of quinine, it may safely be asserted either that what was regarded as the initial paroxysm was in reality the relapse, or else that the absence of a relapse was a mere irregularity, and in no way to be attributed to the action of the opium. On the other hand, in cases presenting a tendency to the typhoid state, with a disposition to stupor, or where the urine was scanty and albuminous, no opiate was administered.

42 Loc. cit.

We have already stated that in our cases quinine in acid solution was frequently ordered, and it answered very well to add to each dose of this a suitable amount of morphia.

Atropia, in the dose of gr. 1/60 to gr. 1/40, was usually associated with the hypodermic injections of morphia. This was done particularly in cases where the pains were very severe, when the pupils were disposed to be contracted, or when there was continued profuse sweating. In addition to this, atropia was continued without morphia during the intermission in a few cases. The patients proved susceptible to its influence, and dryness of the mouth with dilatation of the pupils was readily produced by gr. 1/60 every six hours. In one case gr. 1/40 every four hours for two days caused delirium, with the usual symptoms of belladonna action, all of which passed away quickly after withdrawal of the drug. But in none of these cases was the relapse influenced in the least.

Other remedies may be used for the relief of the insomnia, which is always one of the most distressing symptoms. Chloral and bromide of potassium have been found serviceable in various epidemics, and some observers have preferred them to opium for the relief of headache and insomnia. They did not prove reliable in the Philadelphia epidemic of 1869-70. Bromide of potassium, even in large doses, produced scarcely any effect, and, while in a few cases chloral in doses of gr. xx. gave positive relief, in the majority of instances 40 grains failed to cause sleep or relieve suffering. It must not be forgotten also that, as there is a special tendency to cardiac failure in this affection, the action of chloral must be closely watched.

In a small series of our cases where muscular pains, hyperæsthesia, and twitching were marked succus conii was given quite freely, but without any apparent benefit.

The condition of the stomach required attention in almost every case. Nausea, vomiting, and epigastric and hypochondriac soreness were the prominent symptoms. Anorexia was usually complete during pyrexia, and not rarely patients were admitted to the hospital who asserted that for one or more days they had not taken any nourishment whatever. Under such circumstances, and in a disease where the tendency to prostration and cardiac failure calls for stimulants and food, it is evident that strict care must be given to the diet. In many cases skimmed milk with lime-water, meat broths, arrowroot, or gruel, could be taken in small amounts at short intervals, and retained. But whenever these are rejected, no attempt should be made to persist in their use, but koumiss, whey, or chicken-water should be substituted, and continued until the stomach grows retentive. Equal care must be paid to the selection of a suitable form of stimulus. It may be proper to employ a mild and relaxing emetic if the patient be seen at the onset of the disease and if there is reason to suspect the presence of indigested food in the stomach, but under any other circumstances there seems no reason for its use in a disease where vomiting is so common and gastric irritability one of the most troublesome symptoms. Nor should purgatives be given save when very positive indications exist for their use.

Constipation is rarely obstinate; the amount of nourishment taken is very small; in a considerable proportion of cases there is diarrhoea, or at least a sensitive state of the bowels; and as a consequence it is preferable in nearly every case to dispense with laxatives entirely, and, if the bowels must be opened by assistance, to administer a simple enema.

When irritability of the stomach is marked, benefit may be derived from very small doses of calomel frequently repeated, as, for example, gr. 1/8 or 1/4 every one or two hours. Subnitrate of bismuth may be used in combination with this or as a substitute for it. In several instances more prompt relief was obtained from nitrate of silver given in the dose of gr. 1/12 every three or four hours, dissolved in thin mucilage of acacia.

Stimulants were remarkably well borne, and their administration in such form as was acceptable to the stomach was clearly of service, even from an early period of the disease. As a rule, whiskey was employed, given in the form of milk punch. By carefully graduating the amount of alcohol, and when necessary diluting the milk freely with lime-water, the stomach usually received it well. If circumstances favored, dry champagne, or brandy or sherry in carbonated water would often prove preferable. The exhausting nature of the disease, the marked tendency to cardiac failure, and the inability to digest an adequate amount of nourishment, all indicate the early use of stimulants. In cases where a tendency to the development of the typhoid state existed alcohol was freely given, even to the extent of sixteen ounces of whiskey in twenty-four hours. Other stimulants were usually given in these cases, such as carbonate of ammonium, especially if pulmonary congestion existed; turpentine, especially if tympany was marked; or Hoffmann's anodyne or spirit of chloroform, if muscular twitchings, hiccough, or insomnia with wandering delirium were prominent symptoms. In all cases of severity the use of tonics and stimulants should be maintained in reduced doses during the intermission and for some days after the final fall of temperature.

It remains to allude briefly to certain special remedies and to certain symptoms requiring special treatment. Formerly, much diversity of opinion existed as to the propriety of venesection or local depletion in relapsing fever, but Murchison concluded, after a careful examination of the evidence, that it had not been shown to be of service; and certainly the disease as it occurred in Philadelphia in 1869-70 presented no indication whatever for even the mildest depletory measures. This corresponds with the recognized plan of treatment in all the specific fevers.

Blisters are not so objectionable in relapsing fever as in either typhus or typhoid, and there are several conditions in which they have been found decidedly useful. In cases where the headache has obstinately resisted cold applications, bromide of potassium, and opiates, a blister to the back of the neck has afforded marked relief, with no unfavorable result. Again, in cases where the vomiting and epigastric distress were severe and obstinate the application of a blister three inches square to the epigastrium is to be recommended.

Chloroform has proved of value for the relief of various symptoms in relapsing fever. As already stated, it was found the most useful remedy for the hiccough which was so troublesome in a number of our cases, and especially in those where jaundice was pronounced. It also seemed serviceable in controlling the peculiar chills which in varying degrees of severity were present in a few cases, recurring at about the same hour on successive days. These rigors or chills were uninfluenced by very large doses of quinine or other antiperiodics, but were apparently controlled by full doses of chloroform given in advance of the expected hour of recurrence.

Jaundice, which, as has been stated, is partly of hæmic origin, but is probably also due in part to obstruction from catarrhal swelling of the mucous membrane of the bile-ducts, is not influenced by mineral acids, and still less should mercurials or purgatives be administered for its relief. It would seem proper, in cases where this symptom is marked, to observe special care in diet and the use of stimulants, and to employ local sedative astringents, such as small doses of nitrate of silver combined with opium and belladonna.

Muscular soreness, pains, and tremor may call for special treatment on account of their severity. The only remedy which has proved useful in relieving the first two of these symptoms is opium, conjoined with the external use of anodynes. Iodide of potassium fails even in doses as large as can be borne, and the same is true of muriate of ammonium and cimicifuga, which we used thoroughly without any effect. In the muscular pains, however, which torment the patient during convalescence, the ammoniated tincture of guaiacum was found of service. Atropia hypodermically and chloroform internally have been found useful for the relief of severe muscular twitchings.

Upon the whole, therefore, it will be seen that in ordinary cases a supporting and expectant plan of treatment is all that is required. Abandoning the idea of forcibly controlling the fever or of preventing the relapse, care should be given in the first place to the diet and to judicious stimulation.

Opium or morphia should be used to control pain, excitement, and insomnia, aided, as far as the latter is concerned, by bromide of potassium or the cautious use of chloral. Cooling drinks should be allowed, cool applications made to the head, and the body should be repeatedly sponged with cooling and disinfecting lotions. If the stomach is retentive, quinine in moderate doses may be given in acid solution, alternating with a simple fever mixture; but if nausea and vomiting are present, the first purpose will be to allay them by the appropriate measures already discussed.

Epistaxis is a frequent symptom, but usually requires no special attention. Occasionally it is profuse, and then should be promptly checked, since serious exhaustion may follow its continuance. If, therefore, mild astringent applications do not arrest it, recourse must be had to the tampon saturated with diluted Monsell's solution.

The urine must be closely watched and frequently analyzed in relapsing fever. In some epidemics serious alterations in this secretion are rare; in others it is not uncommon for the urine to be scanty, and to contain albumen or blood. When this latter condition is presented, especially if at the same time uræmic symptoms exist, dry cups should be applied over the kidneys, to be followed by the use of dry heat, and free perspiration should be promoted by hot-air baths or by the hot wet pack. It is probable that jaborandi given in repeated small doses, so as to avoid any depressing effect on the heart, will be found valuable in such cases. Infusion of digitalis, with spirit of nitrous ether or with acetate of potassium, may also be used with advantage.

Absolute rest must be insisted on throughout the entire period of paroxysm and relapse. The records of every epidemic present instances of sudden death from cardiac syncope following trifling exertions. The patients should therefore be kept strictly quiet in bed from the initial rigor until their strength is fully restored after the relapse. As the danger of collapse is especially great at the time of the critical fall in temperature, the patient should be closely watched as the end of the initial paroxysm and of the relapse approaches. If there is any sudden rise of temperature, with head symptoms due to hyperpyrexia, large doses of quinine, ice to the head, cold spraying, or the cold bath must be promptly used. As sweating begins the body must be covered with a warm blanket and warm stimulating drinks be administered. If any marked tendency to collapse is observed, the subcutaneous injection of strychnia or of ether and digitalis, conjoined with diffusible stimulants internally and hot applications externally, are to be employed immediately. The special remedies required for the various complications and sequelæ have already been sufficiently indicated.

I desire in conclusion to acknowledge the important assistance received from Drs. Geo. S. Gerhard, Louis Starr, Charles Shaffner, and R. G. Curtin, who, under the supervision of my colleague, the late Dr. Edward Rhoads, and myself, recorded the histories of most of the cases which serve as the basis of this article, and also tabulated them for statistical purposes.43

43 Reference must also be made to the interesting observations on spirilla published by Mülhaüser in Virchow's Archiv for July 9, 1884, after this article had been printed. His results go to confirm the view that the spirillum of Obermeier is the essential cause of relapsing fever.