STUDIES ONA VIRUS FROM APATIENT WITH FORTBRAGGFEVER (PRETIBIAL FEVER)

By HUGHTATLOCK1(From the Respiratory Disease Commission Laboratory,2 Regional Station Hospital, Section 2,

Fort Bragg, North Carolina; the Division of Virus and Rickettsial Diseases, ArmyMedical School, Washington 12, D. C.; the Children's Hospital Research

Foundation and Department of Pediatrics, University ofCincinnati College of Medicine, Cincinnati 29, Ohio 8)

(Received for publication October 8, 1946)

During the summer of 1942, an outbreak of adengue-like illness of unknown etiology occurredat Fort Bragg, North Carolina. The disease, de-scribed by Daniels and Grennan (1), was char-acterized by moderate prostration, fever, spleno-megaly, a rash localizing particularly on the an-terior aspects of the legs, and a short course. Thelocalization of the rash gave rise to the name"pretibial fever." Epidemiological informationindicated that all of the patients in this outbreakwere quartered in the same area of the post andthat the incubation period was 10 days or longer.Serological tests served to exclude endemic typhus,infectious mononucleosis, brucellosis, and the ty-phoid-paratyphoid group of diseases. A commis-sion was assigned by the Surgeon General of theUnited States Army to investigate the epidemi-ology and etiology of the disease. Working withfrozen material in the form of blood, nasal wash-ings, and urine from the patients, the membersof the commission concluded: "All tests on thismaterial proved negative as far as isolating or

1 Captain, MC, AUS. Part of this work was done bythe author while serving as a member of the Commissionon Acute Respiratory Diseases and part while Chief of theCommunicable Disease Section, Walter Reed GeneralHospital, Army Medical Center, Washington, D. C.

2 Work done at Fort Bragg, North Carolina, was sup-ported in part through the Commission on Acute Respira-tory Diseases, Army Epidemiological Board, PreventiveMedicine Service, Office of the Surgeon General, UnitedStates Army, and by grants from the CommonwealthFund, the W. K. Kellogg Foundation, the John andMary R. Markle Foundation, and the International HealthDivision of the Rockefeller Foundation to the Army Epid-emiological Board for the Commission on Acute Respira-tory Diseases.

8 The studies in Cincinnati were supported by the Com-mission on Neurotropic Virus Diseases, Army Epidemio-logical Board, Preventive Medicine Service, Office of theSurgeon General, U. S. Army.

determining the nature of the infectious agent re-sponsible for this disease was concerned" (2).

Similar outbreaks appeared during the sum-mers of 1943 and 1944. The present report de-scribes the recovery of an infectious agent froma patient having Fort Bragg fever in 1944 andsummarizes observations on the biological char-acteristics of this agent. In addition, it presentsthe results of neutralization tests which indicatethat the agent is related to the human infection;and finally, it describes the reproduction of theclinical disease picture in human subjects inocu-lated with the agent.

PART I

RECOVERYOF A VIRUS AND ITS CHARACTERISTICS

Recovery of the agent from a patientOn August 16, 1944, a 22-year-old white soldier noted

a sudden onset of headache, feverishness, and generalizedaching. In spite of these symptoms, he continued onduty' until August 19 when his symptoms increased fol-lowing a stimulating injection of tetanus toxoid, and hewas hospitalized. Physical examination showed an acutelyill patient with a fever of 1020 F., whose only other ab-normal physical finding was a questionably palpablespleen. The leukocyte count was 5,750 per cubic mm.An indistinct maculopapular eruption was noted over thepretibial areas on August 21. This rash faded after 2days. During his illness the patient showed a remittenttemperature which reached 1030 to 104.50 F. for 5 days.A blood culture taken shortly after the appearance of therash proved sterile. Two days after defervescence hewas completely recovered and was discharged from thehospital. The clinical diagnosis was Fort Bragg Fever.

Two Syrian hamsters and two guinea pigs were in-jected intraperitoneally with blood freshly drawn fromthis patient on August 21, 1944, 5 days after onset, whenhis temperature was 104.50 F. Both hamsters werefound dead on the tenth day. No macroscopic abnor-malities were found at autopsy and the viscera werebacteriologically sterile. Passage to other hamsters was

287

2HUGHTATLOCK

not attempted. Both guinea pigs developed fever (105°-1060 F.) 8 days after inoculation. On the second day offever, one of the animals was sacrificed; citrated blood,suspensions of brain, pooled liver and spleen, and tunicavaginalis were obtained and each injected into two guineapigs by the intraperitoneal route. Cultures of the bloodand organs of the guinea pig showed no growth. Aftera 7-day incubatfli5wperiod, fever developed in guinea pigsof the groups which had received blood and pooled liverand spleen; but no disease appeared in the other animals.A febrile illness was regularly induced in passage animalsinoculated with citrated blood from affected guinea pigs.A similar febrile illness was induced and maintained bypassage in guinea pigs injected with suspensions of liverand spleen of such animals; after the fourth serial pas-sage, this line was abandoned for convenience. The stockcolonies of the guinea pigs and hamsters were free fromintercurrent disease at the time, as evidenced by severalnegative experiments employing intraperitoneal injectionsof blood from patients with other diseases.

Pathogenicity for animals and eggs

GUINEA PIGS: Guinea pigs developed fever fol-lowing intraperitoneal injection of citrated bloodor intracerebral injection of serum taken frompassage animals on their first day of fever. Theincubation period was from 4 to 8 days and thefebrile course was from 2 to 4 days. Youngguinea pigs (200 grams body weight) showed amore prolonged febrile reaction and a shorter in-cubation period than adult animals. Convalescentanimals remained afebrile when reinoculated withinfectious passage material. Fever was the onlyclinical evidence of infection in guinea pigs; noneof the several hundred animals used in the 90-odd_passages of the agent ever appeared sick. Nomacroscopic or microscopic pathological changeswere consistently noted in guinea pigs sacrificedat various stages of the infection. Inclusionbodies and rickettsiae were searched for in sec-tions and smears of organs stained by Giemsa'smethod and other techniques; none was found.

RABBITS: The disease induced in rabbits in-jected with infectious guinea pig material by theintraperitoneal or intracerebral route was essenti-ally identical with that seen in guinea pigs injectedwith passage material. This mild febrile illnesswas maintained during four serial passages inrabbits.

HAMSTERS: Adult hamsters which were inocu-lated intraperitoneally with blood from guineapigs of the sixth and subsequent passages died

after about 8 days. However, the disease was onlyirregularly reproduced in adult hamsters whichwere injected with blood from moribund animalsof this species. Young hamsters, 40 to 60 gramsbody weight, proved more susceptible to the ex-perimental infection since the agent was readilymaintained by serial intracerebral passage in them.Cerebral tissue from a young hamster used forthe fourth transfer titered 10-4 when tested inhamsters.

A characteristic, rapidly progressing disease de-veloped in inoculated hamsters following an incu-bation period of 7 to 16 days. Some hours pre-ceding death, these animals became abnormallyirritable and showed slight ataxia. They assumeda hunched position in the darkest corner of thecage, apparently having photophobia associated oc-casionally with a seropurulent discharge from theeyes. Generaiized tremors, but no true convul-sions, were occasionally noted. The initial ataxiarapidly progressed to complete paralysis of allfour extremities and death usually ensued withina few minutes. Gross and microscopic findings atautopsy were consistently normal except for areasof fresh hemorrhage in the lungs which were fre-quently observed and were regarded as agonalmanifestations. No inclusion bodies or rickettsiaewere seen in the histological sections of the cen-tral nervous system or viscera.

MIcE: Infectious material injected by variousroutes into mice of different ages and breeds in-duced no obvious disease. Furthermore, serial"blind" passages were unsuccessful.

EMBRYONATEDEGGS: The intravenous injectionof 11-day embryonated eggs with infected serumfrom passage guinea pigs resulted in the death ofmost embryos after about 7 days of incubation at96° F. The agent was subsequently maintainedfor 23 serial transfers in embryonated eggs by theintravenous injection of suspensiong of infectedembryo liver diluted from 102 to 104. Theembryo livers were often dark green in color, butno other abnormalities were noted. No inclusionbodies or rickettsiae were seen in stained smearsof embryo viscera, yolk sac, or chorio-allantoicmembranes or fluids. Embryonated eggs survivedthe injection of infected guinea pig plasma intothe yolk sac or the chorio-allantoic sac, or ontothe chorio-allantoic membrane. However, it was

288

STUDIES ON THE ETIOLOGY OF FORT BRAGGFEVER

possible to demonstrate the presence of the agentin yolk sac tissue of embryos which had been in-jected intravenously with infected egg materialand to propagate the agent thereafter by the yolksac route. Titrations in hamsters of fresh suspen-

sions of infected embryo liver or yolk sac usuallyyielded end points of 1-4.

FilterabilityPlasma from febrile guinea pigs was employed

in a few preliminary filtration studies on theagent. For this purpose, the plasma was dilutedwith an equal quantity of physiological saline solu-tion and passed through Seitz or Corning UFfilters. Virus was demonstrated to be present inthe filtrate of a Corning fritted glass filter (UF)but not in that from a single Seitz pad. Materialswere tested for infectivity by intravenous injectioninto embryonated eggs, and the presence of thevirus in tissues of embryos which died was con-

firmed by passage to hamsters and guinea pigs.

The glass filter used in the experiment did notpermit theI passage of Staphylococcus albus whichwas added to the infected plasma.

Neutralization tests

Sera from patients convalescent from FortBragg fever and sera from recovered guinea pigswere tested for neutralizing antibody against thenew agent. Virus for these tests consisted ofundiluted guinea pig plasma or dilutions of in-fected chick embryo liver covering the range of1O1 to 10-. The technique was as follows: Equalamounts of undiluted serum to be tested and the'appropriate dilution of virus were mixed and in-cubated for 1 hour at'room temperature and theninjected into groups, of two or three adult ham-sters. Each animal received 0.5 ml. of the mix-ture intraperitoneally. The animals were observedfor a period of 21 days and deaths recorded.

The results of such neutralization tests carriedout on acute and convalescent sera from five cases

TABLE I

Results of neutralization tests in hamsters

Sera Source and dilution of virus

Guinea pig plasma Chick embryo liver

Source DayaferUndiluted 10-1 10-i 10>

5 D1O,* D1I D9, D1O, Dll D12, S, S D15, D16, SPatient A** 23 S, S S, S, S S, S, S S, S, S

39 S,S

4 D12, D14 D9, D9, S D17, D18, S D13, D13, SPatient B

28 S, S St S, S D20, S, S S, S, S

4 DlI, D1IPatient C

15 D15, S

4 DlI, DllPatient D

24 Dii, Dii

4 D9, D9, D1O D13, D16, D19 D1O, DI0, D12Patient E

24 D9, D9, D1O D10, S, S D12, Di3, D20

Guinea Pig 0 D9, D9, D9B99 24 S, S, S

Guinea Pig 0 D9, D9, D9BiOO ~~~~24D14, 5, 5

* DIO-Death on 10th day after inoculation. S-Survived.** The new virus was isolated from the blood of this patient.

289

HUGHTATLOCK

of Fort Bragg fever and two experimental guineapigs are summarized in Table I. It is evidentfrom the data that the convalescent sera from Pa-tients A and B and from both guinea pigs con-tained neutralizing substances which protectedthe hamsters from death.

StorageConsiderable difficulty has been encountered in

preserving the agent. Infectivity of blood, wholetissues, or tissues suspended in broth was lostafter storage for 24 hours at 200 C., 4° C., or at- 70° C. in sealed glass ampoules. However,viability has been maintained when 20 per centsuspensions of infected embryo tissue were pre-pared in sterile skimmed milk media, pH 7.2,frozen rapidly in glass sealed ampoules and storedat - 70° C.; in one instance the agent was activeafter 8 months storage under these conditions. Inone experiment, the infectivity of a chick-embryoliver suspension diminished after storage at - 70°C. for 24 hours from a titer of 10-4 to 10-2; bothtitrations were made in hamsters. The lability ofthe virus was further suggested in the tests withhuman subjects (see Part II). It was found thatplasma taken from these patients during the febrileillness and proved to be infectious (actually wholeblood was used here) by injection into other hu-man beings, was not infectious for man after stor-age at - 700 C. for 12 days.' Fresh plasma wasknown to be infectious from work with hamsters.

Relation to other infectious agentsThe available data are insufficient to identify this

agent with any known pathogen. Sera of guineapigs convalescent from the experimental diseasefailed to fix complement with the soluble antigenof lymphocytic choriomeningitis. Recoveredguinea pigs were fully susceptible to infection withthe Balkan Grippe strain of the rickettsia of Qfever (3). Convalescent sera from three of thepatients described in Part II failed to fix comple-ment with the antigen of Rocky Mountain spottedfever; in addition, no significant titers developedin the Weil Felix reaction with OX19, OX2, and

4This might explain the failure in 1942 of the groupwho investigated Fort Bragg fever to recover the agentfrom the blood of patients. All of the materials werestored in the frozen state for some days before injectioninto animals (2).

OXK. The new agent was clearly different fromthe rickettsia-like organism which had been pre-viously encountered in guinea pigs used in studieson Fort Bragg fever (4). Subsequent unpub-lished work indicated that the infection caused bythe rickettsia-like agent was enzootic in the guineapig colony and bore no relation to Fort Braggfever in human beings.

PART II

PATHOGENICITY OF THE AGENTFOR

HUMANBEINGS 5

In order to test the disease-producing capacityin man of the virus which had been maintained inguinea pigs and embryonated eggs, the inocula-tion of human subjects was next undertaken. Theplan was to determine whether or not any diseasecould be produced by the injection of a suspensionof infected egg embryo liver and, if infection oc-curred, to study the resistance to the new agent ofindividuals known to have recovered from denguefever and from sandfly fever.

METHODS

Preliminary physical examinations were performed onall subjects; in addition, the results of x-ray examinationsof the lungs, routine blood counts, urinalyses, and dailyrectal temperatures indicated that they were, with fewexceptions, in good general physical condition; all werefree of intercurrent infection. Particular attention wasgiven to examination of the skin, where long-standingblemishes might be confused later with the appearance ofspecific rashes. No attempt was made to isolate the pa-tients because of the lack of epidemiological evidence forcross infection in naturally occurring Fort Bragg fever.It may be mentioned now that subsequent experience inthe present studies likewise failed to suggest that the dis-ease is contagious.

The first group of three patients were injected with a10 per cent suspension of infected embryonated chick liverin saline. The material for this purpose was harvestedfrom living embryos on the sixth day after intravenousinjection with the 23rd egg passage, at a time when 50per cent of the embryos had already died. This egg linehad been started from the infectious serum of a guineapig of the 80th serial passage in this host. Aerobic andanaerobic cultures of the inoculum for patients showed nogrowth on the usual bacteriological media. This sameembryo liver emulsion was infectious at a dilution of 10-4when titrated intracerebrally in hamsters. Mice given a

5 The human subjects used in this study were part of agroup undergoing fever therapy at the Longview StateHospital in Cincinnati.

290

STUDIES ON THE ETIOLOGY OF FORT BRAGGFEVER

combined intracerebral and intraperitoneal injection re-mained well. Guinea pigs which were also injected intra-peritoneally with this inoculum developed the characteris-tic febrile reaction and the agent was readily transmittedto a second group of guinea pigs by inoculation of blood.It is evident that the infectious agent injected into the firstgroup of human subjects had characteristics typical ofthose previously described for the virus.

The three individuals who made up the first group ofsubjects received 3.0 ml. intramuscularly and 0.4 ml. in-tracutaneously of the chick embryo liver suspension de-scribed above. Subsequent transfers of the agent to otherindividuals were made with blood which was drawn fromthe patients within 24 hours after the onset of fever. Thiswas defibrinated, pooled, and immediately injected intothe new subjects who received 5.0 ml. intramuscularlyand 0.4 ml. intracutaneously.

RESULTS

A short febrile illness developed in all threeof the subjects following injection with the sus-pension of infected chick embryo liver. The threeindividuals of the second- group, who receivedpooled defibrinated blood of the first patients, de-veloped a similar short febrile illness. Accord-ingly, the plan to carry out immunity studies wasput into effect in the third group which consistedof two normal individuals, two persons recoveredfrom sandfly fever, and four recovered from den-gue fever. All of the patients, with one exception,in this third group developed the febrile reaction,and the majority exhibited the clinical picture ofFort Bragg fever with rash and leukopenia. Thesingle exception was a normal subject who failedto show any fever over a 23-day period of observa-tion. He had not been out of the state of Ohio,as far as could be determined, and his failure ofresponse remains unexplained. The results ofthese studies in man are summarized in Figures1 and 2. Figure 3 illustrates the types of rashwhich were seen. More detailed information con-cerning the various clinical features of the diseaseproduced in man is given below.6

Intcubation periods. The three individuals in thefirst group, inoculated with infected embryo liversuspension, came down sharply with fever on the

6 One additional individual, not included in the 14 men-tioned above, was injected intracutaneously with only 0.4ml. of pooled serum from the first group of patients. Hefailed to develop fever during the following 20 days.Whether this person, like patient 7, was resistant to thevirus or whether larger doses of virus were required fortransmission remained undecided.

ninth day after inoculation. The second groupdeveloped fever from the 11th to the 14th day afterinoculation with defibrinated blood from the firstgroup. The third and largest group, consisting ofeight individuals, had incubation periods varyingfrom 8 to 14 days, averaging 11. These were quitein accord with the estimate of 10 days or longer inthe naturally occurring disease at Fort Bragg (2).

Fever. The temperature curves of the patientswere spiking, such as occurs in the natural disease(1). Daily physical examinations failed to revealevidence of the occurrence of intercurrent disease.The clearcut temperature elevations, ranging from1010 to 105.50 F. (rectal) and extending over aperiod of 1 to 6 days, average 3.2, stood out sharplyfrom the preliminary temperature base lines. Infour patients, a slight rise in temperature appeared4 days preceding the main bout of fever. Whileno isolation of virus was attempted at this time, itwill be noted in a subsequent paragraph that thisagent circulates at least 2 days prior to the onsetof the typical disease. The possibility was con-sidered that this mild fever might represent theonset of viremia. Figures 1 and 2 show the varia-tions in febrile response in the group of fburteenpatients.

Other clinical signs and symptoms includingrash. An evaluation of symptomatology in thesepatients was not too reliable. During the first 2days of fever the majority complained of mildto moderate headache and aching in the back andthighs. Three individuals had mild but definitechills. There was loss of appetite, and occasionalvomiting occurred. None of the patients appearedcritically ill at any time, and soon after deferves-cence all were out of bed and had a rapid returnof appetite and strength. During the course of ill-ness, physical examination was not remarkable ex-cept for fever, splenomegaly, and rash. Three ofthe thirteen febrile individuals developed a defi-nitely palpable spleen, and in two more there wasquestionable enlargement. A few cases developedinjection of the sclerae with photophobia. Therewere no signs of respiratory or of meningeal in-volvement, and there was no significant lymph-adenopathy.

No unequivocal rash was seen in any of the pa-tients in the first two groups. However, five ofthe eight members of the third group developed

291

2HUGHTATLOCK

lqgKhttl~~~~~~~~~JtzLI\ 1 1

I HIoIztwI~I *

4. S~~~~~~~~~~~IC,.~~~~~~~~~~~~~~~~~~~~~~~~~~~~I

0~~~~~~~~~~~~~N 0 -~~~~~~~~~~~~~~~~~~~~~~~0+

1

In

-+0+O0 44...~~~~~~~~~+1~~~~~~~~~~~~~~~~~~~4

4 0 = ~~~~~~~~~~~~~~~~~

4. 0~~~~~~~~~~~~~00

2 C/)~~~~~~~~~~~~0CC,)~~~~~~~~~~~~4.~~~~~~CO 4( L 2 RR4 +5 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ U~~~~~~~~~~~~6

z~~~~~~~~~~~~~~~~~~~o

a.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~%0. ; ota(,)~uOe.~ 0 *v~-*

292

STUDIES ON THE ETIOLOGY OF FORT BRAGGFEVER 293

THIRD PASSAGE- MAY 17, 1946103102

1o0100

99

989796

W8C 9550 14300RASH _ I_SPLEEN _ 11 1IT r _

104

13 Pt. 9 - SANDFLY FEVER102 IMMUNE |

99 \

..a ...

. .

DAY

0 1 2 34- S 6 7 8 9 10 11 12 13 14 15 16 170 1 2 3 4 5 6 7 8 9 10 ItIt1 13 14 15 16 17

WBC 5L S700 e20CRASH T I I T T I +SPLEEN T 1

._. ~~~~~~TRANSFER103 Pt. 11- DENGUEFEVER IMMUNE

102 "NEW GUINEA C |

101

97-961 DAYS,O I 2 3 4 5 6 7 8 9 10 II 12 13 14 IS 16 17

103

102

101

100

99

98

97

RASH [I 1 I' F00i[SPLEEN I

0 1 2 3 4 5 6 7 8 9 10 It 12 13 14 IS 16 17W8C I 901 1111I 10900 39SO 720bRASH +i+[it+I

SPLEENII II

-103102101

100

999897

WBG 8400_ 900 5450 4950[RASH 11 1 II I+I+ 4I-SPLEEN I1 0 1 I IIVIRUS 0- 0 _ 400

104103 - Pt. 10-SANDFLY FEVER102 IAMUNE \ \t

96 ,DAkYS .2 3 4 5 6 7r 8 9 10 II 12 13 14 15 16 17

WBC 1130 1 13650 6550 3350 10300

RASH | | |I I +1+1 +SPLEEN II

03 Pt. 14- DENGUEFEVER IMMUNE

l01

1000 9I

99E97L96 1 .... BYOI 2 3 I

10 11 12 13 14 15 16 17Iw:C 11850IIII I14 50 10000RASH II

Arrows pointing downward indicote doy of -inoculation

FIG. 2. TEMPERATURECHARTS-EXPERIMENTALPRETIBIAL FEVER IN MAN

Pt. 7-CONTROL

. D4YS. . . . . .0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17I

Pt. 8-CONTROL

. . ,, DYS0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Il

W"I4

11

I

I

I

.Pt. 13- DENGUEFEVER IMMUNE"HAWAII"

DAYS

HUGHTATLOCK

TABLE II

Average total and differential leukocyte counts in the five cases with leukopenia

Days of fever Base - | 2 3 4 6 7line* ~ *

Total WBC 9,531 11,775 10,950 11,017 6,500 5,400 5,940 6,775 9,400Per cent granulocytes 62 73 79 79 64 61 55 50 65Adult forms 57 67 73 74 52 56 43 47 61Young forms 5 6 6 5 12 5 12 3 4Per cent lymphocytes 36 26 20 19 34 38 43 47 27

* Average of counts taken well before onset of fever; at least 2 counts per subject.** Indicates the day before the onset of fever; subsequent numbers indicate the days after the onset of fever.

rashes of varying extent which appeared from thesecond to the fifth day after the onset of fever. Forthe most part these were limited to the anteriorand lateral surfaces of the legs. The lesions weretransitory and faded rapidly after the second daywith evidence of residual pigmentation and smallhemorrhagic areas in two of the cases.

Some of these patients were seen by Dr. AshleyA. Weech, Professor of Pediatrics, University ofCincinnati, who described their skin lesions in thismanner: "The following description of the rashin experimental pretibial fever is the result of in-specting three patients, one with presumably earlylesions on the fifth febrile day, another with betterlesions on the third day, and the third with lesionson the seventh day. The three cases are suffi-ciently similar to permit one description. Lesionswere present in the skin over the lower half of thetibiae in all cases; in one subject, similar lesionswere present on the dorsal surface of the leftforearm. The involved areas show an erythema-tous blush and vary in diameter from a few milli-meters to a few centimeters. The borders arewell defined, but the shapes are irregular. Theareas are slightly elevated above the surroundingunaffected skin and on palpation feel mildly in-durated. In two lesions over the tibia of one pa-tient there was a small ecchymosis in the center.The appearance was that of fresh hemorrhage."

In one individual (patient 12), a large erythe-matous blush first appeared on the right shin, fol-lowed two days later by erythema and edema at thesite of intracutaneous inoculation of the virus sus-pension. Two days after this, maculopapular ery-thematous lesions appeared over both shins. Noneof the other cases showed any reaction at the intra-cutaneous inoculation site. The rashes seen in the.five subjects were sufficiently similar in appear-

ance and distribution to the rashes described inthe naturally occurring disease to leave little doubtbut that the clinical picture of Fort Bragg feverhad been reproduced in these human subjects.

Five of the thirteen positive cases showed aslight leukocytosis with a relative increase ingranulocytes which appeared late in the incubationperiod and persisted for a day or so after the onsetof fever (see WBCin Figures 1 and 2). Begin-ning about the third day of the febrile phase, aslight leukopenia with a relative lymphocytosis ap-peared and persisted for a few days. Furtherinterpretation of the available data did not seemwarranted. Average daily values of the leukocytecounts for these five cases are shown in Table II.The hematological findings in these individualswere similar to those in patients with Fort Braggfever.

IMMUNOLOGICALRELATIONSHIP BETWEENFORT

BRAGGFEVER, SANDFLY FEVER, ANDDENGUE

FEVER

Sabin (6) found that there are distinct im-munological types of dengue virus which aresufficiently related to give rise to group specificcross immunity during the first 4 to 8 weeks, al-though at a later date the immunity is type-specific.Similarly in sandfly fever, one strain of virus iso-lated from an outbreak in Italy was completelyunrelated (i.e., there was no cross immunity evenduring the first weeks) to two other viruses whichwere isolated in the Middle East and Sicily. Thelast two agents were identical in so far as yieldingcomplete cross immunity even 2 years after a singleexperimental attack (6). The patients used inthese tests (see Figure 2) were selected by Dr.Sabin from among those he had previously inocu-

294

STUDIES ON THE ETIOLOGY OF FORT BRAGGFEVER 295

I IG. 3. ILLUSTRATION OF THE DIFFERENT TYPES OF RASH SEEN IN EXPERiMENTAL FORT BRAGGFEVER INHUMANBEINGS

Figures 3(1) and 3(2) show the early and late appearance, respectively, of the anterior sllin lesions inpatient 10. Figure 3(3) illustrates the raised macular lesions which appeared on the forearm of patient 13.

HUGHTATLOCK

lated with dengue or sandfly fever viruses. Thus,patiellts 11 and 12, who had had typical experi-miental attacks of dengue following inoculationwith the "New Guinea C" strain of virus 1 miionthbefore the present test, were selected since theywtould be expected to) be immiiiiunie to both "Hawaii"anid "New Guinea C" strainis of dengue. Patients13 and 14, who had had experimenital dengue feverfollowing inoculation witlh the "Hawaii" strainof virus, 7 and 6 montlhs, respectively, prior to thepresent test, would resist only an agent immuno-logically identical witlh the "Hawaii" type of virus.Similarly, patients 9 and 10 could indicate a rela-tionship only between the new agent and the"Middle East-Siciliaii" type of sandfly fevervirus (5) but not with sandfly fever in general.Patient 9 had lhad hiis first attack in November,1943 (21 2 years prior to the present test) and wasimmune when challenged with the homologousvirus in October, 1945. Patient 10 had had his ex-perimental attack in October 1945, or 7 monthsbefore the present test. Sinice all six patients de-veloped the experimental disease produced by thenew agent, it can be concluded that it possessesneither a group relationship with the known den-gue viruses nor any relationship with one strainof sandfly fever virus. It nmay be stressed, further-more, that neither the dengue nior the sandflyfever viruses possess the pathogenic properties oftlle new agelnt for lamiisters and guiniea pigs.

DE)"TECTION OF VIRUS IN TILE BLOOD OF PATIENTS

Information on the time of appearance and dura-tion of viremia in experimiientally infected humansubjects was obtained in the following manner.Blood was drawn from the patients at varioustilmes during the disease, immediately defibrinated,and injected into young hamsters. In certain in-stances the animals received 2.0 ml. of undilutedblood intraperitoneally. In otlhers, serial ten-fold dilutions of serumii Mvere injected ilntracere-brally in order to determine the infective titer.Those hamsters which died between the seventhanid sixteenth days were assumed to lhave suc-cumiibed to infection with the virus under investiga-tioln. The results of these tests in patients 1, 2,aind 3, Figure 1, clearly slyo)ve(l that virus wasp)reseilt at least 48 lhoturs before the oniset of feveranid was still detectable on the second day of fever,

but not on the fourth day or thereafter. Patient8, Figure 2, had detectable virus circulating 24hours before and 24 hours after the onset of fever,but not 4 days before nor 3 days after the onset offever. In two instances, undiluted 1)lood frompatienit 8 was lethal for hamiisters but failed to killin dilutions of 1:10 or greater. It may be conl-cluded that the virus appeared in the blood of thesepatients shortly before the onset of fever and disap-peared promiiptly; furthermiiore, it did lnot occuir inlarge amounts.

SUMMARYAND CONCLUSIONS

An inifectious agent was recovered from guinieapigs injected with blood from a soldier sufferilngfrom Fort Bragg fever in August, 1944 at FortBragg. The agent iniduced a febrile nioni-fatal ill-ness in guinea pigs and ralibits anid a letlhal diseasein hamsters. It was maintained by passage in anii-mals or embryonated eggs, but storage eveni at- 700 C. was unsatisfactory unless special precau-tions were taken. The agent was filterable unlderconditions whicl were adequate to retaini staphyl-ococci.

The virus, after prolonged passage in guinleapigs anid embryonated eggs, was able to induce theclinical picture of Fort Bragg fever in some of theinoculated human beings, while the miiajority ex-hibited only fever for 1 to 3 days.

The niew virtus appears to lie uniirelated in itsiroperties or immiiunologically to the ageiits oflymphocytic choriomeningitis, Q fever, RockyMountain spotted fever, sandfly fever, and deniguefever. It does not resemble the rickettsia-likeagent previously enicountered in work oln FortBragg fever and subsequently show, n to le enzooticin guinea pigs.

ACKNOWLEDGMENTS

The author wishes to thank the Commission on Neuro-tropic Virus Diseases and particularly one memiiber of theCommission, Dr. Albert B. Sabin, for his help in carryingout the human transmission studies. Grateful acknowl-edgment is made to Brig. General George C. Beaclh, USA,Commanding General, Army Medical Center, for grantingthe author detached service for the work on human sub-jects, to Brig. General S. Baynie-Jonies, USA, and Col-o0iel Worttl l'. Daniiels, MC, for their interest and en-eoniragement, and(l to lt. Colonel Joseph El. Smadel, MIC,for his hell) with the manuscript. The cooperation of Dr.Douglas Goldman, Medical Director of the Longview

296

STUDIES ON THE ETIOLOGY OF FORT BRAGGFEVER

State 'Hospital in Cincinnati, and of Dr. E. A. Baber, Su-perintendent of the institution, is gratefully acknowledged.

BIBLIOGRAPHY

1. Daniels, W. B., and Grennan, H. A., Pretibial fever,an obscure disease. J. A. M. A., 1943, 122, 361.

2. Topping, N. H., Philip, C. B., and Paul, J. R., Reportof the Commission for the Study of an UnidentifiedDisease at Fort Bragg, N. C. (Sept. 3, 1942-Sept.11, 1942). Preliminary report submitted to theSurgeon General, U. S. Army, 15 Oct. 1942; finalreport 31 March 1943.

3. Commission on Acute Respiratory Diseases, Identifica-tion and characteristics Qf the Balkan grippe strainof Rickettsia burneti. Amer. J. Hyg., 1946, 44,110.

4. Tatlock, H., A rickettsia-like organism recovered fromguinea pigs. Proc. Soc. Exper. Biol. & Med., 1944,57, 95.

S. Sabin, A. B., Philip, C. B., and Paul, J. R., Phleboto-mus (Pappataci or sandfly) fever, a disease of mili-tary importance, summary of existing knowledgeand preliminary report of original investigations.J. A. M. A., 1944, 125, 603, 693.

6. Sabin, A. B. Personal communication.

297