CHAPTER VI
Scrub Typhus
Chris J. D. Zarafonetis, M.D., and Myles P. Baker, M.D.
Scrub (or miteborne) typhus is an acute febrile disease caused by infectionwith Rickettsia tsutsugamushi. Clinically, it resembles other rickettsialdiseases with an abrupt onset characterized by chilly sensations or rigors,followed by fever, headache, malaise, and later a rash. The differentialdiagnosis may be established early by finding the primary sore or eschar at thesite of infection, often with associated satellite or generalizedlymphadenopathy; at a later stage, some cases may be distinguishedserologically. Scrub typhus is also called tsutsugamushi disease, tropicaltyphus, rural typhus, Japanese river fever, and Kedani fever.1
The earliest description of the malady in the Japanese literature was writtenby Hakuju Hashimoto in 1810, while the first English report was made by Palm in1878. Mites were suspected as the probable vectors as early as 1879. The fieldmouse, Microtus montebelli, was implicated as an important naturalreservoir of infection in 1918.
During the 20 years before World War II, careful research work, notably inMalaya and Sumatra, had indicated that tsutsugamushi disease was not limited toNorthwestern Japan, Formosa, and the Pescadores Islands, where the Japanese hadby 1931 definitely established its rickettsial etiology. The clinical picture,pathology, and epidemiology were shown to be fundamentally the same in Sumatranmite fever, in the scrub typhus of Malaya, the endemic typhus of New Guinea, thecoastal fever of North Queensland, Australia, and in cases reported from Burmaand Indochina. The common pattern could be traced in spite of wide variations inmortality and frequent failures to find the primary eschar. Prior to theexperience of World War II, however, the only species of mite proved by animalexperiments to be the vector of tsutsugamushi disease was the trombiculid miteidentified by Japanese workers. Other mites had been highly suspected in Sumatraand Malaya, but conclusive proof had not been furnished. Similarly, theprincipal reservoir host of the trombiculid mite had not been incriminatedelsewhere than in Japan and Malaya. Definitive work on the mite-rat complex, inNew Guinea and Burma, remained to be done.
This identification of scrub typhus, as it was commonlycalled in the Armed Forces, with tsutsugamushi disease of the Japanese removesit from the list of strictly tropical diseases, for Japan can scarcely beconsidered
1Blake, F. G., Maxcy, K. F., Sadusk, J. F., Jr., Kohls, G. M., and Bell, E.J.: Studies on Tsutsugamushi Disease (Scrub Typhus, Mite-Borne Typhus) in New Guinea and Adjacent Islands; Epidemiology, Clinical Observations and Etiology in the Dobodura Area. Am. J. Hyg. 41: 243-373, May 1945.
112
within the tropics. It was, in fact, littleknown outside Japan as late as 1942. Its distribution isknown to us now only as indicated by the accident of human infection with R.tsutsugamushi. With the advent of large-scale jungle warfare, scrub typhusbecame an important medical problem to troops in the Far East. Indeed,approximately 6,000 cases were to appear in U.S. Forces alone during thecampaigns that followed the outbreak of war with Japan.
As military operations progressed in the Southwest Pacificand in Burma, U.S. Armed Forces passed into and through areas hitherto scarcelystudied as to prevalent diseases and lived there under field conditions veryunlike the customary sheltered life of the white planter of prewar years. Oneimmediate result was a wider appreciation of the geographic area of distributionof this disease. Until then, such places as Bougainville and Goodenough Islands,the Schouten Islands, and Netherlands New Guinea, or Luzon and Mindoro Islandsin the Philippines, had not been reported as sites of scrub typhus. Medicalofficers became acquainted at first hand with the clinical picture and pathologyof the disease and had opportunity to compare it with other rickettsialinfections, notably epidemic typhus. Opportunity was afforded for thorough studyof its epidemiological and entomological aspects in the field and laboratory andfor the institution of preventive measures.
CLINICAL EXPERIENCE
Southwest Pacific Area
In this region, Gunther had by 1940 collected 105 cases of adisease which he labeled "endemic typhus," chiefly from the Wau areain the mountains south of the Markham Valley in northeast New Guinea.2The clinical picture was described by him at length, the similarity totsutsugamushi disease emphasized. Diagnostic agglutinations with ProteusOX-K were reported. The patients were white men who had been clearing jungleareas. Scattered cases were reported from New Britain Island and the coastlineof northeast New Guinea west of Finschhafen. The disease was, then, well knownwhen American and Australian Armed Forces moved into New Guinea in the spring of1942.
THE FIRST PHASE: ORIENTATION (1942)
The first cases.-Typhus fever was firstrecognized among our troops in the Southwest Pacific as an isolated case,reported from northern Queensland in March 1942. Eight cases were reported fromthe Port Moresby area of Papua late in September 1942, and subsequently, a fewcases appeared in the Milne Bay area at the eastern extremity of Papua.3 Duringthe summer
2Gunther, C. E. M.: A Survey of Endemic Typhus in New Guinea. Med. J. Australia 2: 564-573, 30 Nov. 1940.
3Essential Technical Medical Data, Southwest Pacific Area, for February 1944.
113
and early autumn of 1942, combat operations in Papua wereprimarily the assignment of Australian forces. At the close of the year, Lt.Col. S. W. Williams, of the 2/9th Australian General Hospital at Port Moresby,was able to report a preliminary study of 300 cases of scrub typhus.4Within 10 miles of Port Moresby, the disease was rare, considering the largenumber of troops concentrated there; all but a handful of these cases developedunder combat conditions along the Moresby-Buna track, the so-called KokodaTrail. This early Australian experience tallied closely with the prewardescription of the clinical picture given by Gunther. Emphasis was laid on thefollowing points:
Among the first hundred patients there was only one death; among the secondhundred, exposed to the hardships of living conditions while following theJapanese retreat over the Owen Stanley Range, there were three deaths. In casesobserved during December 1942, chiefly from the Buna-Gona combat area, themortality was "nearly 10 percent." These soldiers, too, were spent byweeks of campaigning in the mountains and fighting around Buna.
Early diagnosis was not found a simple matter. Eschars were noted in only 60percent of the cases. As experience was to show, they should be sought incovered parts of the body. Similarly, the rash was observed in about 60 percentonly. Generalized lymph node enlargement was the most useful diagnostic sign indistinguishing the disease, in the early stages, from malaria. Even so, thediagnosis, in absence of eschar and rash, might well remain in doubt until theappearance of blood agglutinins for Proteus OX-K on about the 9th or 10thday of fever. It was noted that patients with scrub typhus fever may appeardeceptively well during the first week, only to develop serious pulmonary,circulatory, and neurological symptoms in the second week of fever.
American troops were soon presenting examples of those diagnostic problemsthat proved to be scrub typhus. The majority of men who were to engage in combatin the Buna-Gona area were flown over "The Hump" of the Owen Stanleys,were set down at Dobodura airstrips, and went into battle late in November 1942.Patients with scrub typhus began to appear in December in Port Moresbyhospitals, evacuated thereto by air. In January 1943, there was a sharp rise inthe number of reported cases of scrub typhus in the Southwest Pacific Area(table 12). The case fatality rate for 1943 was 5.5 percent per 100 cases. In1944, over 4,000 cases of scrub typhus were reported with a case fatality rateof 3.8 percent. During the first 8 months of 1945, the number of cases was lessthan those reported for the same period in 1943, with a case fatality rate of5.4 percent. Owing to the exigencies of the military situation, soldiersconvalescing from scrub typhus were evacuated to general hospitals on theAustralian mainland, with the result that any compilation of case material wasrendered impracticable. Medical officers in hos-
4Williams, S. W.: Scrub Typhus in Papua. Preliminary Report. A Description Founded on Observation of 300 Cases by the Medical Staff of an Australian General Hospital, 5 Jan. 1943. [Official record.]
114
[Preliminary data based on unit summaryreports]
| Cases | Deaths |
| 1943 |
|
|
January | 92 | 4 |
February | 62 | 1 |
March | 45 | 5 |
April | 55 | 2 |
May | 64 | 0 |
June | 79 | 5 |
July | 83 | 6 |
August | 81 | 3 |
September | 95 | 5 |
October | 93 | 4 |
November | 67 | 3 |
December | 119 | 13 |
|
| 935 | 51 |
| 1944 |
|
|
January | 75 | 9 |
February | 104 | 16 |
March | 184 | 8 |
April | 75 | 12 |
May | 130 | 4 |
June | 212 | 15 |
July | 647 | 9 |
August | 1,759 | 32 |
September | 757 | 38 |
October | 251 | 10 |
November | 113 | 6 |
December | 89 | 6 |
|
| 4,396 | 165 |
| 1945 |
|
|
| January | 100 | 5 |
| February | 66 | 2 |
| March | 45 | 4 |
| April | 32 | 2 |
| May | 15 | 2 |
| June | 43 | 2 |
| July | 20 | 1 |
| August | 11 | 0 |
|
| 332 | 18 |
|
| 5,663 | 234 |
115
pitals of the Advance Base, becoming familiar with clinicalfeatures of the textbook picture of the disease, cultivated a high index ofsuspicion of scrub typhus in the case classified as fever of undetermined originwith no evidence of malaria, no leukocytosis, no clinical response to Atabrine(quinacrine hydrochloride) or quinine, and with generalized lymph nodeenlargement. Eschars were more carefully sought for, and ulcers, without a scab, were found in the groins and axilla (fig. 8). By the time that agglutininsfor Proteus OX-K appeared in the blood, the temperaturecurve had become diagnostically characteristic in cases without otherdiagnostic signs. One temperature chart (often making up the bulk of the recordevacuated with the convalescent) could be superimposed on another, so alikewere they.
In one series of cases,5 the temperaturewas usually 99? to 102? F. on the first day of illness. There was a rise intemperature within 24 hours to 100? to 103? F., and during the days thatfollowed the 4-hourly chart showed a swinging temperature with daily remissions from99? to 104? F. throughout the illness, subsiding by lysis onthe 14th to 17th days. In the more severe cases, for the first 3 to 5 days aremittent type of fever was seen, followed by a sustained fever up to 105? F.until recovery or death. In some fatal cases a remittent fever rangingfrom 99? to 106? F. was recorded. A temperature which did not rise above103? F. was seen in a few fatal cases.
The first necropsies-Inexperience with the disease and lack of anyspecific treatment lent particular interest to the pathologist's findings in the first
5Williams, S. W., Sinclair, A. J. M., and Jackson, A. V.: Mite-Borne (Scrub) Typhus in Papua andthe Mandated Territory of New Guinea; Report of 626 Cases. Med. J. Australia2: 525-539, 18 Nov. 1944.
116
seven necropsies performed in December 1942 and January 1943at Port Moresby by Capt. (later Maj.) Austin J. Corbett, MC.6 Hereported three groups of findings, as follows:
1. An acute diffuse myocarditis, with perivascular and interstitialinfiltration with plasma cells, macrophages, and lymphocytes, as a constant andcharacteristic finding. Damage to the myocardial fibers varied in degree.Arteriolar walls showed thickening, degeneration, and intramural round-cellinfiltration.
2. In the lungs, alveolar septa were infiltrated with mononuclear cells in aperivascular arrangement, and the vascular endothelium within the septa wasswollen and fragmented.
3. In the brain, perivascular proliferation of glial cells and infiltrationwith lymphocytes were observed; the endothelium of smaller blood vessels wasswollen, and several vessels showed intramural round-cell infiltration.
Cardiac findings and prognosis-The finding of a diffuse myocardiallesion at autopsy, and the occasional finding of gallop rhythm and"soft" heart sounds on physical examination of the precariouslybalanced patient in the second and third weeks of fever, gave rise to muchconcern about the convalescent's prognosis. In the rear hospitals, in Australia,convalescents were examined carefully for evidence of residual cardiac damageand insufficiency and were found to have none. Pending the accumulation ofenough data to support a general policy of sending all but the most hard-hitcases back to duty, the majority of convalescents were returned to dutyfollowing maximum hospitalization benefit.
THE SECOND PHASE: INVESTIGATIONS (1943)
A second phase in U.S. Army experience with scrub typhus began in the summerof 1943. The occurrence of cases from month to month, without seasonalincidence, was unremitting among combat and service troops, which wereincreasing in number as the bases were developed along the north coast of NewGuinea to Finschhafen. The New Guinea hospitals had become more stabilized,facilitating the study of cases. This period saw clinical and epidemiologicalstudies carried out by a group of investigators, headed by Dr. Francis G. Blake,working under the auspices of the United States of America Typhus Commission,and it saw also the development of preventive measures to protect troops fromscrub typhus.
Clinical study-It will be noted in table 12 that caseswere diagnosed without letup during the spring and summer of 1943. Base B, whichincluded Dobodura and Oro Bay, Papua, became a hive of activity withever-increasing numbers of troops moving into camp areas in the base. Between 1August and 1 December 1943, 248 patients with scrub typhus were admitted to BaseB hospitals. The mortality in these cases was 2.4 percent. From a study of agroup of 20 cases, most of them at the 363d Station Hospital, Dr. Blake and
6Corbett, A. J.: Scrub Typhus. Bull. U.S. Army M. Dept. No. 70, pp. 34-54, November 1943.
117
his collaborators constructed an exemplary word picture of the disease, asfollows:7
The characteristic early symptoms and physical signs of mite typhus at onsetand during the first 5 to 7 days of the disease are headache, apathy andgeneralized malaise, fever, relative bradycardia, anorexia, conjunctivalcongestion, lymphadenitis, often regional, and eschar. Although the diagnosismay be suspected from the general symptoms and physical signs together with anormal or low leukocyte count, only the presence of a typical eschar issufficiently pathognomonic to establish the diagnosis at this stage of thedisease. Consequently, the whole body surface should be scrutinized in detailwith adequate light for, unless this is done, the eschar may be easilyoverlooked if it is located in the less readily accessible areas.
Commonly, between the fifth and eighth days a characteristic macular ormaculopapular rash appears on the trunk and may later spread to the extremities.It should be borne in mind, however, that although the development of a rashtends to confirm the suspected diagnosis, rashes sufficiently similar to that ofmite typhus to cause confusion may at times occur in dengue and otherinfections.
The subsequent course of tsutsugamushi disease after the first weekmaybe relatively mild, with no recognizable evidence of the more serious aspects ofthe disease, and with a total febrile period of 12 to 14 days; or moderatelysevere with signs of pneumonitis and encephalitis and recovery by lysis earlyin the third week; or severe to extremely severe with febrile course ofapproximately 3 weeks' duration. Symptoms and signs of more severe pneumonitisand encephalitis are constant. Evidence of peripheral circulatory collapse arecommon and signs of myocarditis may appear. Hepatitis and nephritis sometimescomplicate the picture. Thromboses and cerebral or gastrointestinal hemorrhagemay take place. Death is not infrequent in these severe cases, particularly inolder individuals, but the majority recover by lysis during the latter half ofthe third week.
Laboratory studies have confirmed the already known fact that theWeil-Felixtest becomes positive for Proteus OX-K about the 12th to 14th day of thedisease in the majority of cases but that not all patients develop agglutininsfor this organism in a diagnostic titer of 1 in 160 or higher. Rickettsiaorientalis (tsutsugamushi) may readily be recovered from the blood ofpatients during the acute stage of the disease by the intraperitonealinoculation of 0.2 to 0.3 ml. of blood into white mice. Both of theselaboratory procedures, if positive, establish the diagnosis, but a negativeWeil-Felix test for Proteus OX-K does not exclude miteborne typhus.
The total leukocyte count, usually within the normal range during the firstweek but sometimes showing a leukopenia, tends to increase during the secondweek. In cases progressing favorably, this leukocytosis is commonly due in largepart to a progressive increase in the absolute number of lymphocytes.
Limited observations on the blood chemistry of the miteborne typhus of NewGuinea have shown that hypochloremia, a moderate but not critical reduction inserum albumin, a rise in serum globulin, and a diminution in plasma fibrinogentend to occur during the height of the disease.
The treatment of mite typhus consists primarily of completebed rest, good nursing care, adequate diet, fluid and salt intake, and thejudicious and selective use of symptomatic drug, supportive, and oxygentherapy only when specifically indicated. Penicillin is ineffective.
Blake and his coworkers particularly emphasized thelocalization above thigh level of the eschar, or primary ulceration, present inall 20 of their cases. This is in contradistinction to the bites of mitescausing scrub itch, on the
7See footnote 1, p. 111.
118
ankles and shins. This observation tallied with that of theAustralians at Port-Moresby (65 percent on arms, axilla, neck, and trunk). Theyemphasized further (1) the absence of signs of right-sided heart failure, (2)the picture of a true rickettsial pneumonitis, with which physical signs andfindings at autopsy are in keeping, and (3) the relation of cyanosis andpulmonary congestion to pneumonitis, rather than to myocardial insufficiency.
During the spring and summer of 1943, Australian patientswere held at the general hospitals in Port Moresby, as American patients werenot, making possible certain observations on 626 cases8 that addedto the clinical picture.
Significantly, as regards course, a few cases were recognized inambulatorypatients not sick enough to go off duty, but later showing agglutinins to Proteus OX-K of 1:125 or higher (taken as a minimum diagnostic titer inthis series). No relapse of fever attributable to the rickettsial infection wasobserved. Among symptoms, macules were noted on the soft palate, in associationwith the typical rash over the body.
In the severe cases, cardiovascular abnormalities such as faint systolicmurmurs, tic-tac sounds, gallop rhythm, tachycardia, hypotension, and pulsusalternans were occasionally observed. Measurements of venous pressure andelectrocardiographic study did not speak for myocardial insufficiency orclinically demonstrable myocarditis. Hearts examined microscopically inpatients dying in the sixth week of complications not referable to circulatoryapparatus showed no infiltration or fibrosis.
Cyanosis and increased respiratory rate accompanied the development ofabnormal physical signs in the lungs, which did not include those ofconsolidation unless there was secondary infection, atelectasis, or infarction.Pulmonary infarction was a late complication to be reckoned with.
In the severely ill, mental symptoms uniformly appeared from about the 6th tothe 10th day; apathy was more common than restless irritability and insomnia.Nerve deafness was present in 90 cases, and could be expected to be transient,lasting only a week. Paresis of the shoulder girdle muscles and sensory changesin arms and legs were observed during convalescence. Tremors of the handsappearing in the difficult second week were rare but striking, in one instancesimulating the tremor of parkinsonism. Examinations of cerebrospinal fluid (60)revealed an increase in pressure (about 160 mm.) in less than half theindividuals tapped; increase in lymphocyte count was infrequent (7 cases).
Mortality continued to vary much during the period of investigation; it rosefrom 1 to 2 percent in the first 200 cases to 25 percent in the next 150cases, comprising soldiers exhausted by weeks of jungle combat, and handicappedby slow evacuation. For the next 276 cases, mortality was 7.2 percent. Thisvarying mortality gives a fairer picture than the overall figure for 626 cases,9.7 percent.
If, prior to evacuation, the soldier was seriously ill by the sixth day, orhad already had 6 days of fever, it was thought wiser, if possible, to treathim "where he lies," rather than subject him to evacuation. Mostpatients were treated in convalescence in the expectancy that they would be fitto return to duty within 12 weeks from onset of their illness.
Lipman, Casey, Byron, and Evans reported a study of 200 patients observed atthe 362d Station Hospital in Base B between February 1943 and February 1944.9The clinical picture in their cases conformed with Blake's. Their extensiveexperience permitted observation of complications that should
8See footnote 5, p. 115.
9Lipman, B. L., Casey, A. V., Byron, R. A., and Evans, E.C.: Scrub Typhus; Results of a Study of the Cases of Two Hundred PatientsAdmitted To and Treated At a Station Hospital Between Feb. 9, 1943, and Feb.4, 1944. War Med. 6: 304-315, November 1944.
119
be borne in mind; that is, pleural effusion, 5 percent;peripheral neuritis, ulnar, 5 percent; thrombophlebitis, 5 percent; andpulmonary and splenic infarction, each 2.5 percent. The mortality in their 200cases was 10 percent. They emphasize the point made by the Australians at PortMoresby, that combat troops who will or must disregard early symptoms of scrubtyphus are those with whom it goes hard. The older age group tolerated theonslaught of scrub typhus less well than did those younger. Six of the thirteenofficers in this series died; four of the six were over 35 years of age. Intherapy, these authors recommend the following: (1) A lumbar puncture for reliefof meningismus and cerebral symptoms associated with an elevated spinal fluidpressure, (2) treatment of incipient and established peripheral vascular failurewith oxygen and intravenous plasma, and (3) penicillin for secondarybronchopneumonia complicating the rickettsial pneumonitis.
Irons and Armbrust, at the 3d Medical Laboratory in Base B, extended theirobservations on the Weil-Felix reaction to a series of 74 cases of scrub typhus.10
Of this series, 86 percent developed blood agglutinins for Proteus OX-K.Controls with other febrile diseases gave negative reactions. In general,however, the incidence, maximum titer, and duration of the reaction were inversely proportional to the severity of the disease. Only 2 of the 7 fatal casesshowed a positive Weil-Felix reaction during their clinical course; the maximumtiter was 1:80. Patients with less than 18 days of fever were more likely todevelop a titer of 1:160 or over than were those with longer febrile periods.The initial Weil-Felix reaction occurred at approximately the 17th day of thedisease, after the peak in severity of clinical manifestations, or duringobvious clinical improvement.
Accordingly, the authors concluded that the Weil-Felix reaction is of novalue in prognosis. For diagnosis, no absolute level could be given, but a riseand fall in serial tests proved helpful in some cases. That a negative reactiondoes not exclude scrub typhus was indicated by isolation of the causativemicro-organism in some cases with no OX-K titer.
Epidemiological study-In addition to delineating theclinical features of scrub typhus, Blake and his coworkers clarified theepidemiological background of the disease, as follows:
An analysis of the occurrence of cases among American forcesin an advanced base area near Buna on the northeast coast of New Guineaindicated that, although the disease was contracted in widely scatteredlocalities, some were more dangerous than others. It also indicated that theexposure to infection was associated with activities which brought men intointimate contact with field conditions, such is those which prevail in bivouacsor in establishing camps. During the period of this study in the Dobodura area,the environments in which human infections were known to have originated werekunai grass fields, in which natural conditions were undisturbed at the time ofoccupation. From the time distribution of onset of cases of the disease it wasclear that maximum risk was experienced during the first week or two followingarrival of a unit in a new area. This
10Irons, E. N., and Armbrust, C. A.. Jr.: Relation of the Weil-FelixReaction to the Clinical Course of Tsutsugamushi Disease. Bull. U.S. Army M.Dept. 5: 85-95, January 1946.
120
risk decreased progressively so that after the fourth or fifth week theincidence became sporadic or ceased entirely, and the site could be occupiedwith impunity thereafter.
Decline in the attack rate could not be explained asdue to the accumulation of immune individuals. It therefore was probably due todecreased exposure to bites of some species of larval mites that served asvectors. Decrease in exposure was correlated with progressive changes producedby development and use of the campsite. Evidently, the conditions thus createdwere unfavorable to continued activity and survival of the vector species.
In attempting to identify this vector species, collections oflarval mites were made not only in localities where cases had originated andextensive environmental changes taken place, but also in localities where theenvironment appeared to be similar but where the natural conditions remainedundisturbed.
* * * * * * ** * * the observations made are consistent with thehypothesis that either Trombicula fletcheri or Trombicula walchi, or both, mayserve as vectors in this area. Furthermore, by mouse inoculation, R.orientalis was recovered from two pools of mites (T. fletcheri) collected from twobandicoots. Brain tissue from these same animals, injected into white micefailed to produce a rickettsial infection. In view of this negative result andthe fact that generation-to-generation transfer of R. orientalis occursin mites, the evidence does not incriminate the bandicoot as a reservoir host.
This concept of bivouacs and the establishment of campsitesas the background of scrub typhus was entirely in keeping with U.S. Armyexperience during 1943, elsewhere than at Dobodura. On 5 September 1943, the503d Parachute Infantry Regiment jumped in the Markham Valley terrain whichincluded much open grassy country with kunai grass prevailing. The regimentremained in jungle combat for about 12 days. By October, 37 cases of scrubtyphus had appeared.11 For the next 5 to 6 months, cases continued to occuras areas were developed around the Nadzab airstrips northwest of Lae innortheast New Guinea.
Outbreaks-In November 1943, an epidemic of scrub typhusbroke out on Goodenough Island, north of Papua, chiefly in a hospital areasituated in kunai grass. For the first few nights after arrival in the area,personnel lived under conditions prevailing in the field. This was a seriousoutbreak. Seventy-five cases occurred on the island over the subsequent 4months, most of them in December. The mortality, 25 percent, was uniquely highand cannot be adequately explained with the knowledge at hand. The patients werenot in combat and were not exhausted when stricken. With each successivebeachhead taken, cases of scrub typhus developed. The landing at Cape Gloucesterin the Bismarck Archipelago was made in December 1943. At the 30th EvacuationHospital, 104 cases were reported in February with 5 deaths; in March, therewere 51 cases with 8 deaths. The mortality was 8.3 percent for these New Britaincases, more in keeping with the Dobodura series.12It is well to note, however,that in the first 3 months of 1944 the 363d Station Hospital at Dobodura had 60cases of scrub typhus, of which 9 ended fatally, a mortality of 15 percent, incontrast to an average of 7.1 percent for the
11See footnote 3, p. 112.
12Quarterly Report, Surgeon, 30th Evacuation Hospital (Semimobile),Southwest Pacific Area, 1 Jan.-31 Mar. 1944.
121
Dobodura region.13 Eight of the deaths occurredin the first 35 cases admitted after transfer from the combat areas severalhundred miles to the west. The patient sick with scrub typhus did not tolerateevacuation well.
Following the landing at Saidor in northeast New Guinea inJanuary 1944, 61 cases developed in March, 59 of them among troops on patrolsor bivouacked in a combat area. At Aitape, on the northwest coast, the samestory held true: Landings on 22 April and 33 cases of scrub typhus in May, 52in June, and continued incidence in July and August. Headquarters, 32d InfantryDivision, proffered the comments that the cases developed in fairly wellidentified areas, or in troops on patrol, that withdrawal from the perimeterto cleared areas caused a dwindling of cases, and that uncleared kunai grass andlow sites where kunai grass and underbrush are mingled are dangerous. AtHollandia in Netherlands New Guinea, there were scattered cases; 43 of the 140reported for the Sixth U.S. Army in June originated there. It is worthy of notethat the new staging area for the 32d Division at Hollandia was burned overbefore troops were allowed to move into it in September. During the ensuingmonth, there were no cases reported from this area. An engineer aviationbattalion that moved into Hollandia late in April reported 24 cases with 7deaths in May and June, "due to sleeping in kunai grass, since no tentsor cots were available the first month." The small Wakde Task Force thatcaptured this island with its important airstrip in May reported a few casesin the third and fourth weeks of June.14
It is to be noted that task force units were becomingfamiliar with the type of terrain to be avoided and the dangers of sleeping onor near the ground in uncleared areas likely to harbor infected mites.Meanwhile, active measures for personnel protection of the soldier were inpreparation.
THE THIRD PHASE: PEAK OF INCIDENCE AND CONTROL MEASURES(1944)
The subsequent epidemics at Owi and Biak (in the SchoutenIslands) and at Sansapor (in Netherlands New Guinea) mark a third period in Armyexperience with scrub typhus, in which (1) the disease first assumed proportions serious from a military point of view, (2) new ecological types of fociappeared, and (3) procedures, long-planned, were directed at energetic controlof the epidemics.
Outbreaks-Cases first developed on Owi, a small island off the southern shore of Biak, about 10 to 14 days after arrival of units of theArmy AirForces early in June. During July and early August, as much as one-fourth toone-third of the effective personnel of two air force squadrons on Owi werehospitalized for scrub typhus. On nearby Biak, there were 107
13Kohls, G.M., Armbrust, C. A., Irons, E. N., and Philip,C. B.: Studies of Tsutsugamushi Disease ( Scrub Typhus, Mite-Borne Typhus) inNew Guinea and Adjacent Islands; Further Observations on Epidemiology andEtiology. Am. J. Hyg. 41: 374-396, May 1945.
14(1) Quarterly Report, Surgeon, Sixth U.S. Army, SouthwestPacific Area, 1 Apr.-30 June 1944. (2) Quarterly Report, 32d Infantry Division,plus attached units, Southwest Pacific Area, 1 July-30 Sept. 1944. (3) QuarterlyReport, 1913th Engineer Aviation Battalion, Southwest Pacific Area, 1 Jan.-31 Mar. 1944.
122
cases in the 186th Infantry within 41 days of D-day, 32 ofthem from one company engaged in patrol in the island hills.15Certain units were unable to function effectively until replacements weresecured. Owi and Biak are coralline, with bare ridges and porous outcrops, verylittle swampy land, no kunai flats, and few open grassy areas. Even where thereis little topsoil, as on Owi, dense underbrush, ferns, and many immense rottinglogs provide cover and moisture for mites. With New Guinea experience asprecedent, scrub typhus was not expected in this type of terrain. The firstcases were relatively mild, eschars were infrequent, and these fevers ofundetermined origin were not directly recognized as scrub typhus. They werethought on Owi to be dengue fever until typical cases of scrub typhus appeared,and Culex mosquitoes were found to be scarce. The risk on Owi lay in theslowness in clearing of campsites. A contributing factor was the necessity forusing all available bulldozers for the construction of airstrips. Men leavingclothes on rotting logs, when swimming in the ocean, may also have contributed.On Biak, severe fighting in uncleared areas, seeking cover in thick underbrush,and patrolling, all contributed to exposure to the infected mites.
Members of the U.S.A. Typhus Commission arrived at Owi soonafter scrub typhus was recognized, and encouraged the speedy clearing ofcampsites and sanding of tent floors, together with impregnation of clothing.Emulsions of dimethyl phthalate were made available and all personnel wereequipped with treated clothing within 2 weeks of arrival of the repellent.16The fall in incidence in the various units expressed the sum of the effect ofcontrol measures. On Biak, similarly, the 41st Infantry Division was inimpregnated clothing by the second week in August and camp areas were cleared.The scrub typhus rate fell appreciably in the latter part of August.17The absorbingly interesting story of the development and institution ofpreventive measures against scrub typhus are recounted in another volume in thehistory of the Medical Department in World War II.18
The Owi-Biak outbreak included at least 1,080 cases, which passed throughthe 92d Evacuation Hospital. The mortality, 0.7 percent, was low, and many caseswere milder than had been the experience in Papua.19 No ill effectswere noted from the air evacuation to Nadzab or Finschhafen of convalescents 5days or more after defervescence.
The Sansapor task force landed at the western extremity of NewGuinea
15Essential Technical Medical Data, U.S. Army, Services of Supply, Southwest Pacific Area, forSeptember 1944.
16Letter, Lt. Col. Cornelius B. Philip, SnC, to Surgeon, U.S. Army Forces inthe Far East, 5 Aug. 1944, subject: Scrub Typhus on Owi and Biak Islands.
17Monthly Sanitary Report, 41st Infantry Division, Office of Surgeon, Southwest Pacific Area,October 1944.
18Philip, Cornelius B.: Scrub Typhus and Scrub Itch. In Medical Department, United States Army. Preventive Medicine in World War II. Volume VII.Communicable Diseases : Arthropodborne Diseases Other Than Malaria. [In preparation.]
19Quarterly Report, Surgeon, 92d Evacuation Hospital (Semimobile), Southwest Pacific Area,1 July-30 Sept. 1944.
123
on 30 July 1944.20 On 6 August, the first case ofscrub typhus developed, an unusually short incubation period. The outbreak wasexplosive (135 cases by D+13) and a cause of serious depletion in strength. Inone battalion, 17 percent (125 cases) were stricken. Impregnated clothingappeared to be definitely protective, when used. In all, about 931 cases (dataare incomplete) were reported in the course of 53 days at Sansapor, with amortality of 3.4 percent. Attempt was made to avoid a long 1,200-mile airevacuation until patients were afebrile. This led to much congestion at the onefunctioning evacuation hospital but was doubtless the better part of wisdom inview of the earlier experience in the Cape Gloucester and Aitape operations. AtSansapor, it was noted that areas of infection were found particularly inabandoned plantations or native gardens filled with a rank growth of grass.These areas, as at Owi-Biak, represented a different type of focus from thoseseen theretofore, and apparently in this instance a highly infected type, tojudge from the high incidence of eschars and the recovery of R. tsutsugamushifrom white mice injected with material from each of three lots of rat mites.
Convalescence and disposition-The base hospitals towhich the New Guinea cases were evacuated continued, over the course of the 20months from January 1943 to August 1944, to determine the disposition of convalescent cases. Many ambulatory convalescents complained on admission of rapidheart rate and shortwindedness on slight exertion. Some litter cases were slowto mobilize themselves. At the 105th General Hospital in Australia, Levinecarried out observations on 130 convalescents and found no evidence, by physicalexamination, electrocardiogram, roentgen examination, or by measurements ofvital capacity or venous pressure, of persistent myocardial damage.21
Exercise tolerance tests gave mediocre results, which couldbe contributed to, commonly, by a prevalent anxiety state. Patients whose heartrates were rapid prior to a road march returned from a 5-mile march with normalheart rates. Such a functional test appeared more reliable than a"Schneider index" determination. Of this series of 130 patients, 117were returned to full duty.
A critical evaluation by Howell of the 233d Station Hospital, at Nadzab, ofelectrocardiograms in 190 convalescents from scrub typhus (92 percent within 1to 4 weeks after subsidence of fever) and 10 acute cases revealed nosignificantly abnormal electrocardiographic patterns.22 There wereminor deviations from normal, to be sure, and these were all too oftenexaggerated in importance by the unwary observer. But they were no morefrequent than
20(1) See footnote 13, p. 121. (2) Quarterly Report,Surgeon, Headquarters, Sixth U.S. Army, Southwest Pacific Area, 1 July-30 Sept.1944.
21Levine, H. D.: Cardiac Complications of Tsutsugamushi Fever (ScrubTyphus); An Investigation of Their Persistency. War Med. 7: 76-81, February1945.
22Howell, W. L.: Absence of Electrocardiographic Changes in TsutsugamushiFever (Scrub Typhus). Report of 200 Consecutive Cases. Arch. Int. Med. 76:217-218, October 1945.
124
have been found in a series of tracings from approximately 500 healthy youngadults.
In the summer of 1944, at the instigation of Col. Bruce P. Webster, MC, thenSurgeon, Base E, a reconditioning program for scrub typhus convalescents was setup at the 90th Station Hospital. Three hundred and twelve unselected andsuccessive patients were classified on admission with regard to their capacityfor exercise.23 Only 8 percent were considered to be, properly, bed cases. Thesepatients had been evacuated chiefly from Owi-Biak, Sansapor, and Aitape. They hadaveraged 4 weeks prior hospitalization. In rare cases, there was residualhypesthesia and deafness. Tachycardia, tremors of the hands, and giddiness werefrequent. At the completion of a purposeful, graded reconditioning program, 303,or 97 percent, were returned to full duty, after an average of 1 month in thishospital. These patients were symptom free on discharge, the majority withweight regained. Nine patients were evacuated to the rear because of associateddisabilities, not referable directly to scrub typhus.
This policy of returning the scrub typhus convalescent to duty was thussupported as experience with the disease increased in 1944. Few such soldierswere subsequently evacuated for inability "to stand the gaff" of fieldduty. Technical Memorandum No. 10, Headquarters, United States Army Forces inthe Far East, dated 29 August 1944, declared it official policy (1) that inpatients who have safely reached convalescence from scrub typhus, completerecovery is to be expected, (2) that permanent physical defects resulting fromthe disease are exceptional, (3) that physical reconditioning, with unequivocalreassurance that there is no residual heart damage, will expedite recovery ofstamina, and (4) that patients who require a month or more of full physicalreconditioning should be transferred for that purpose to the newly establishedtraining center at Oro Bay.
The concept of the part that the myocardial lesion plays in scrub typhus hadthus changed materially from that which prevailed early in 1943 when the firstcases were observed, precariously ill, and at post mortem. Berry, Johnson, andWarshauer,24 reporting experience with 85 cases and 110 convalescentsin two New Guinea station hospitals between 20 December 1943 and 20 June 1944,drew the following conclusions:
1. Clinically, there is no severe irreversible myocardial damage.
2. Peripheral vascular collapse in the severely ill patientposes the therapeutic problem, not myocardial failure; the clinical picture isanalogous to that of shock and should be treated accordingly.
3. Circulatory symptoms in convalescence are no more pronounced than afterany severe infection.
23Romeo, B. J.: Convalescence From Scrub Typhus. Bull. U.S. Army M. Dept. 6: 167-173, August 1946.
24Berry, M. G., Johnson, A. S., Jr., and Warshauer, S. E.:Tsutsugamushi Fever; Clinical Observation in One Hundred and Ninety-five Cases. War Med. 7: 71-75, February1945.
125
4. The patient should not be given the impression that he has any heartdisease.
The soundness of this policy was borne out by the report25 from one generalhospital in the Zone of Interior that examination of such patients as werereturned from overseas with the diagnosis of myocarditis following scrub typhusfever definitely indicated their disability to be that of neurocirculatoryasthenia.
Settle, Pinkerton, and Corbett26 reviewed thepathological changes from 55 fatal cases of scrub typhus occurring in Americantroops in Papua, Netherlands New Guinea, and adjacent islands. They reassertthat generalized acute vasculitis was a constant finding. But they note thatalthough myocarditis was present in all cases, it was severe in only about halfthe patients, more commonly in those dying relatively early, between the 9th and12th days of illness. They suggest that in these severe cases myocardial failuremay contribute to fatality; yet they elect to emphasize the generalized vascularlesions and the concept that patients with rickettsial vasculitis may die ofperipheral circulatory collapse. They draw attention to a similar conclusion ofWoodward and Bland27 from their cases of epidemic louseborne typhusin the Mediterranean Theater of Operations, U.S. Army.
IN THE PHILIPPINES (1945)
In the Philippine Islands, scrub typhus appeared sporadicallyduring U.S. Army operations28 but never to the degree of militaryimportance it had assumed at Owi and Biak, and might have done at Sansapor hadenemy resistance been more taxing. Scrub typhus had not been identified in thePhilippines prior to World War II. Toward the close of the Leyte campaign, latein 1944, 3 cases were reported from the northern end of the island. Fifty-onecases were reported from Samar, where units of the Army did extensive patrolduty after the invasion of nearby Leyte. On Mindoro, experience with scrubtyphus followed the typical course: Invasion in December 1944, the first caseadmitted to hospital 12 days after the initial landing, and the bulk of the 100cases reported in January and early February. Sporadic cases continued for thenext 3 months.
On Luzon, rare cases were identified from the mountainous country east ofManila during the fighting in February 1945. Diagnosis was based on the findingof classical eschars and a positive Weil-Felix reaction with the OX-K strain of Proteus. A cluster of three cases appeared early in May in one in-
25Benjamin, J. E., Porter, R. R., and Dreisbach, R. H.: Sudden Death inPatient Supposed To Have Myocarditis Following ScrubTyphus. Bull U. S. Army M. Dept. 4: 235-238, August 1945.
26Settle, E. B., Pinkerton, H., and Corbett, A. J.: APathologic Study of Tsutsugamushi Disease (Scrub Typhus) With Notes onClinicopathologic Correlation. J. Lab. & Clin. Med. 30: 639-661, August1945.
27Woodward, T. E., and Bland, E. F.: Clinical Observations in Typhus Fever,With Special Reference to the Cardiovascular System. J.A.M.A. 126: 287-293, 30Sept. 1944.
28Essential Technical Medical Data, U.S. Army Forces, Pacific (for WesternPacific Area), May and June 1945.
126
fantry regiment engaged in battle in the mountains west of Baguio. A risingtiter of OX-K agglutination was present in all three cases. All had a typicalrash and clinical course, and one had an eschar. Beginning on 9 May 1945, casesof scrub typhus were recognized in a regiment engaged in a beachhead landing inthe Bicol Peninsula in the southern end of Luzon. Six cases developed in thecourse of a week and four more in the following fortnight. Eschars were presentin only 2 of the 10 cases, but clinical course, rash, and rising OX-K titerswere diagnostic. During the combat of Negros in late April and May 1945, sixcases appeared in two infantry units, all with eschars.
These reported data on approximately 180 cases are inevitably incomplete, forthey do not include patients evacuated from Luzon for other medical reasons whodeveloped scrub typhus en route to or after arrival at New Guinea basehospitals. Statistical reports on Southwest Pacific Area scrub typhus for 1945include 258 cases up to 1 June (table 12) with a mortality of 5.8 percent. Therewas nothing unusual in the clinical picture of the cases in the PhilippineIslands. Plasma and blood transfusions were used more frequently than in NewGuinea days, to forestall or alleviate peripheral circulatory collapse. Thenumber of cases so treated, however, did not permit any comparative study. Inthe focal areas of infection studied, as on Mindoro and Negros, cogon grass wasprevalent. This was probably not so near Baguio. Bearing in mind the Owi-Biakexperience, the U.S.A. Typhus Commission warned that any environment harboringrats infested with Trombiculae that will accept man as a casual host can be apotential source of human infection.
The Philippine experience with scrub typhus made it clear that when manexposes himself to field conditions wherein he may become an accidental host,scrub typhus will appear sporadically, and unpredictably, in hitherto unreporteddistricts.
South Pacific Area
Before the war, scrub typhus had not been recognized in the Solomon Islands.The first case was identified on Bougainville in December 1943. Nine more casesdeveloped in January 1944. All had been exposed in a bivouac area on the bank ofthe Laruma River. Sporadic cases, 11 in number, each after exposure in the samearea, developed in the course of the next 6 months. In all but two of thesecases, agglutinins for Proteus OX-K developed in a titer of 1:160 or higher.
An interesting outbreak of mild scrub typhus developed on Bougainville in aFijian force of 750 who were bivouacked in a meadow of coarse grass betweenriverbank and jungle at the former site of a native village.29 From14 to 22 days later, 49 patients were in hospital with a fever that averaged 11days' duration. An eschar typical of scrub typhus was found in all cases, andwas above the legs in 32 (65 percent) of them. Generalized lymph node
29Anderson, W. L., and Wing, W. M.: Tsutsugamushi Disease (Scrub Typhus); A Clinical Study of 49 Cases. War Med. 8: 163-166, September 1945.
127
enlargement was constant; the eruption was present in 45percent (22 patients). Bradycardia was the rule. Tachypnea was not observed. Inthese Fijians, in whom malaria could be excluded, the splenic enlargement wasmade out in the course of the second week in approximately 40 percent. ProteusOX-K agglutinins were present in 24 cases, or about 50 percent; only 5 hadtiters of 1: 1,200 or over; 19 ranged from 1: 50 to 1: 200. Two weeks later,another patrol of the same strength wearing clothing impregnated with dimethylphthalate bivouacked in the same area with no resultant cases of scrub typhus.Two months later, an unprotected patrol went into the same area. After they cameout, 23 cases of scrub typhus appeared. The available evidence suggests thatthis was a localized focus of mild scrub typhus.
India-Burma Theater
Before the outbreak of hostilities in 1941, there had beenconsiderable interest in Indian Army Medical Services in cases diagnosed as"fevers of the typhus group." It had become clear that there were twotypes in the Simla hill country: (1) The murine or flea typhus case,developing agglutinins for Proteus OX-19, occurred under urbanconditions, such as the bazaar sections of the cantonments, and (2) scrub ormite typhus, with agglutinins for Proteus OX-Kingsbury strain, occurredsporadically or in explosive focal outbreaks in units living under fieldconditions, particularly in scrub jungle. Attempts to establish the presence ofrickettsial infection in a larval mite had not been made, nor had it beenpossible to isolate a strain of OX-K typhus in a wild rat. The rodent reservoirsof mite typhus in India were not established. The disease had been known toexist in Burma since 1932.
The late months of 1943 marked a period of orientation inthis disease for U.S. Army medical officers. During November, a group of 22febrile patients were admitted to the 20th General Hospital at Ledo, Assam.30The fever was intense, associated with generalizedlymphadenopathy, mild conjunctivitis, and on occasion a nonpetechial rash,appearing on the fifth to eighth day. Small ulcers or crusted papules were foundand thought to mark the bite of the infecting insect. The fever lasted 8 to 20days. The first Weil-Felix reactions were negative, save for a few agglutinationsof OX-K in low titer. However, these patients, mostly Chinese troops in activetraining for jungle warfare, were considered likely cases of scrub (mite)typhus. In Shingbwiyang, Burma, on the Stilwell Road, 50 cases, entirelysimilar, had appeared in November. By 1 February 1944, 352 cases had beenreported, 35 of them Americans, the remainder Chinese, all from units livingunder field conditions along the Stilwell Road. Two centers of infection wheretroops had been in training were incriminated, one of them the center of thejungle training area of the 22d Chinese Division, the other a campsite much fre-
30Letter, Maj. D. S. Pepper, MC, Assistant Chief, MedicalService, 20th General Hospital, to Surgeon, Base Section 3, China-Burma-IndiaTheater, 9 Dec. 1943, subject: Report on Investigation of "C.B.I.Fever."
128
quented by Chinese troops. All American soldiers with scrubtyphus had been living in the jungle. By December, the clinical picture hadtaken on a definite and characteristic pattern, in every way conforming to thepicture of scrub typhus already familiar in the Southwest Pacific Area and tothe descriptions of tsutsugamushi fever in Japan. Eschar, rash, fever curve,circulatory symptoms and signs, pneumonitis, and encephalitic manifestationswere observed. Agglutinations in high titer for Proteus OX-K were found.
It was noted that the focal areas from which most of theseChinese cases came were not occupied by an appreciable number of troops untilOctober 1943. During each of the last 2 weeks of November, there were 40 cases,and this rate rose to 60 per week through December. Incidence dropped offsharply in early January, following rather closely the movement of troops out ofthe area.
Earlier, in the autumn of 1943, a British unit in the India-Burma theater had121 cases in an outbreak that began 9 days after moving into an area fortraining exercises.31 The epidemic subsided directly after the areawas left, with cases continuing in the unit for about 2 weeks. The descriptionof their cases was in all respects similar to the Chinese and American groups.
Captured Japanese reports revealed that the enemy in Burma was encountering thesame disease, which they called eruptive fever. They recognized similarity totsutsugamushi fever but did not find the eschar frequently enough to warrantmaking this diagnosis in Burma. Their mortality rate is not given save in thecomment-"very low." This incidence among the Japanese is all the moreinteresting in the light of our repeatedly negative reports about scrub typhusin enemy troops in New Guinea.
The subsequent story of scrub typhus among American troops inBurma is described in a comprehensive report by Sayen, Pond, Forrester, and Woodfrom the 20th General Hospital.32 Up to July 1945, there were 726cases reported from the India-Burma theater in U.S. Army personnel with 52deaths, a mortality of 7.2 percent. Among U.S. Army troops, 535 cases werecarefully studied at the 20th General Hospital; 472 of these Americans wereacutely ill on admission. There were 27 deaths, giving a mortality of 5.7percent.
Sayen and his coworkers point out that a more intelligentunderstanding of the severity of scrub typhus is gained from grouping theircases according to the circumstances surrounding the successive outbreaks. Among113 American soldiers from Services of Supply units installed along the LedoRoad who developed the disease sporadically, the mortality was approximately 4 percent. Among 105 patients who were evacuated by air from active jungle combatin the mountains north of Myitkyina, Burma, from March to May1944, the mortality was 16 percent. This group, exhausted by the stress of
31Blumgart, Herman L., and Pike, George M.: History ofInternal Medicine in India-Burma Theater. Chapter on Scrub Typhus, inclosure 5 thereto. [Official record.]
32Sayen, J. J., Pond, H. S., Forrester, J. S., and Wood, F. C.: Scrub Typhus in Assam andBurma; A Clinical Study of 616 Cases. Medicine 25: 155-214, May 1946.
129
fighting, and stricken with scrub typhus, was in an area fromwhich evacuation was difficult. The mortality of only approximately 1.5 percentamong 177 cases evacuated in the autumn of 1944 from troops undergoing jungletraining north of Myitkyina must be related to speedy hospitalization in theforward area, and early evacuation to a general hospital. This group was in farbetter physical condition than the combat troops of the second group. In afourth group of 77 cases, evacuated from combat south of Bhamo, Burma, inJanuary and February 1945, the fatality rate rose again to 7 percent.Evacuation for this group was as difficult as for the Myitkyina combat group,marching after onset of fever again being unavoidable. The patient's pretyphusstate of health and quality of early care were vital in prognosis.
The typical march of symptoms developed in all of thesecases. Effort on the part of Sayen and his coworkers to classify the casesaccording to severity warrants quotation, as follows:
In the average case of scrub typhus, during the first weekthere were no signs by which the probable outcome could be determined.Occasionally a high fever at this time indicated a severe course. During thesecond and third weeks, however, certain phenomena appeared which justifiedplacing the patient in "severe" or "grave" groups withrespective mortalities of approximately 25 and 50 percent. However, until thedisease was definitely on the decline one could not be sure that these featureswould not appear. Consequently, the diagnosis of "mild" or"moderate" scrub typhus, which carried practically no mortality, wasnot justified until defervescence was established. * * *
A case of scrub typhus was assigned to the "severe" group on thebasis of any one of the following:
(1) An alarming increase of the general evidences of illness inthe second week.
(2) More than the usual amount of fever: Over a week of peaks to 104?, or105? for more than 2 days.
(3) Frank clinical signs of dysfunction of an important organ:
(a) Signs of "typhus pneumonitis": Respirations 36 per minute for2 days; cyanosis of the skin.
(b) Signs of meningoencephalitis: Severe delirium; meningismus.
(c) Signs of nephritis: Azotemia over 50 mgm. percent, isosthenuria, heavyalbuminuria and cylindruria.
(d) Marked enlargement of the heart or sharp T-wave inversion in theelectrocardiogram.
(e) Pitting edema with or without ascites.
(f) Multiple hemorrhagic phenomena.
The appearance of any of these phenomena indicated that statistically thepatient's chances of survival had been reduced from 19 in 20 to about 3 in 4.
A severe case of scrub typhus was considered in the "grave" groupon the basis of one of the following findings:
(1) A steady increase of illness during the third week.
(2) Very high fever: 105? for 5 days or 106? for 2 days.
(3) Signs of severe dysfunction of more than one vital organ,or evidence of severe inflammation of the central nervous system and the lungs.Of these, the most common were the following:
(a) Extensive pneumonitis: Persistent cyanosis out of oxygen [tent];respiratory rate 50 per minute for 2 days, or over 36 per minute for a week.
(b) Encephalitis: Malignant restlessness, Cheyne-Stokesrespiration, a convulsion, coma.
130
(c) Severe nephritis: Anuria.
(d) Tachycardia exceeding 130 per minute.
The appearance of these phenomena indicated that,statistically, the patients' chances for survival were 25 to 50 percent. We havenever seen a patient recover whose cyanosis was unrelieved by oxygen or who hadpersistent hyperpnoea, pulmonary edema, or coma lasting 24 hours.
In their discussion of the classical signs, laboratoryfindings, and treatment of scrub typhus, Sayen and his coworkers bring out thefollowing interesting points that tally with or supplement findings recorded inthe Southwest Pacific Area.
1. The finding of an ulcer or papule that later became a typical mite ulcer 1to 14 days prior to onset of symptoms was not uncommon. The presence (60 percentof 200 cases) or character of the eschar or ulcer had no relation to theseverity of the disease, or to the OX-K titer. In only 27 percent was theprimary lesion on the thighs or below.
2. Increasing generalized enlargement with, usually,tenderness of the lymph nodes was an early development in all but 6 of the 200patients, and was, in fact, the most constant of the main diagnostic clinicalsigns.
3. The eruption (71 percent of cases) involved the trunkalways, the face in 15 percent. Generally of 4 to 9 days' duration, it mightoutlast the fever. A few rashes were florid or purpuric. There was no relationof the rash to the severity of the disease.
4. Ophthalmoscopic examination revealed engorgement ofretinal veins in 67 percent of cases in the first and second weeks, progressingto bilateral edema of the retina and optic nerve head in 36 percent, withretinal hemorrhage and exudate in a few instances. Retinopathy of this sort wasabsent in other febrile disease, and was actually helpful in the early diagnosisof atypical cases, later confirmed by the OX-K agglutination reaction.33
5. Bronchial rales, changing in location, appeared in "mild" and"moderate" cases, did not of themselves indicate a rickettsialpneumonitis.
6. Brief periods of gallop rhythm were not evidence of a prognosticallyimportant degree of interstitial myocarditis.
7. Agglutination titers for Proteus OX-K of 1: 100 were not seen except inscrub typhus. About half of a series of 200 cases failed to have this"diagnostic" titer. It was helpful diagnostically in isolated atypicalcases. In epidemics, a "clinical" diagnosis was generally made beforeagglutination became positive (about the 14th day, on an average).
8. Cerebral malaria was the disease of chief importance to exclude indifferential diagnosis of the case without conclusive signs of scrub typhus.
9.Roentgen examination of the chest in scrub typhus cases often failed to indicatethe extent and severity of the rickettsial pneumonitis. Intermittent dyspnea atrest, then persistent tachypnea, and cyanosis relieved initially by oxygencomprised a march of signs more valuable than physical findings in
33For a further discussion of the ocular manifestations ofscrub typhus (new and original work) see Medical Department, United States Army.Surgery in World War II. Ophthalmology and Otolaryngology. Washington: U.S.Government Printing Office, 1957, pp. 141-144.-J. B. C., Jr.
131
the chest or roentgen examination in establishing the presence of pneumonitis.
10. Cardiac enlargement, sharp T-wave inversion with or without RS-Televation (11 percent of 61 cases suspected of cardiac complication), andpersistent gallop rhythm were not found to be necessarily evidence of ill omen.Dangerous rickettsial myocarditis could not be diagnosed, prior to finalcollapse, save by inference of precedent. No patient with T-wave inversiondied; three cases who had tracings before death showed no electrocardiographicabnormality.
11. Spinal fluids in 27 patients with signs of meningeal irritation wereunder increased pressure (average 230 mm. of water). Ten had increase in cells,generally lymphocytes; three had a polynuclear pleocytosis with negativecultures. Spinal fluid protein levels were elevated (68 to 156 mg. percent, infour cases).
12. Convulsions occurred in 12 patients, of whom 3 were not considereddangerously ill until the convulsive seizure.
13. Spontaneous diuresis late in the febrile course, or atthe onset of convalescence, was observed in 38 percent of 200 cases; the largestdiureses were seen in the sicker patients with edema (10 percent of face, hands,shins, and feet), and with ascites. It constituted a reassuring prognostic sign.
14. Hypochloremia was little modified by parenteral saline therapy. Return tonormal blood chloride levels occurred in convalescence, regardless of therapy.
15. In convalescence, thrombophlebitis and pulmonary embolism were occasionalcomplications. Pleurisy developed in 10 percent of 200 cases; effusions, ifpresent, were small, sterile; fever lasted only 4 days, and evidence of pleurisywas generally gone in a week.
16. In treatment.-(1) The mildest appearing case in the first week should beprepared by absolute bed rest for an unpredictably severe ordeal in the secondweek. (2) An adequate number of experienced nurses is of vital importance to seethat small frequent feedings are taken, fluids pushed (minimum of 3 litersdaily), and severely ill patients protected. (3) Salt administration can beoverdone; it may aggravate edema. (4) Oxygen should be given by mask preferably;when given by nasal catheter, it did not, in the average case, relieve cyanosis.(5) Rectal paraldehyde, 30 cc. in oil, was the best sedative for "malignantrestlessness," and might make it possible for the cyanotic patient totolerate oxygen by mask, and prevent exhausting exertion. (6) Digitalis was notused. (7) Penicillin was of no avail save in the presence of a complicatingbronchopneumonia. (8) The concept of peripheral circulatory collapse was notraised; intravenous plasma was not used with this in mind. (9) Intravenousfluids were given when indicated, carefully, and without untoward incident.
As experience with scrub typhus increased, confidence grew in theconvalescent's ability to cope with the demands of duty. Patients with mild and
132
moderate cases (less than 3 weeks of fever) were informedsoon after defervescence that they were returning to their units and weregradually reconditioned for discharge about 3 to 4 months from the onset ofillness. Patients who had been gravely ill were evacuated to the Zone ofInterior. The severe cases were appraised individually. Of the last 300 cases,82 percent were returned to full duty. No evidence of residual myocardial damagewas detected during reconditioning.
PATHOLOGY
Although a number of reports34 dealt with thepathology of scrub typhus, the most comprehensive study was performed at theArmy Institute of Pathology (now the Armed Forces Institute of Pathology),Washington, D.C., by Maj. Arthur C. Allen, MC, and Dr. Sophie Spitz.35These workers undertook to study the lesions not only of scrub typhus, but alsoof louseborne typhus, Rocky Mountain spotted fever, and Q fever in order todetermine whether the various rickettsial diseases could be differentiatedhistologically. After careful study of the histological preparations andprotocols of 78 cases of scrub typhus, 24 cases of epidemic typhus, 12 cases ofRocky Mountain spotted fever, and lung sections from 2 cases of Q fever, theauthors made the following observations:
l. The primary lesion, or eschar, is considered to beprovoked by the combined action of the secretion of the larval mite and theinoculated rickettsiae. It is suggested that the absence of the eschar incertain instances of scrub typhus may be due to variations in cutaneousimmunity.
2. Interstitial pneumonitis of a marked degree is common inscrub typhus in contrast with epidemic typhus and Rocky Mountain spotted fever.The histologic picture of the interstitial pneumonitis of scrub typhus isindistinguishable from that of Q fever, rheumatic fever, toxoplasmosis, andviral pneumonia.
3. It is concluded that the amount of hepatic damage as noted histologicallydoes not warrant the presumption that hypoproteinemia is due to hepaticinsufficiency.
4. Early, acute, diffuse glomerulonephritis is common inscrub typhus, epidemic typhus, and Rocky Mountain spotted fever. The indirectrole of the rickettsiae in the pathogenesis of the glomerulonephritis isindicated.
5. The focal encephalitis or nodule of scrub typhus is qualitatively similarto that of epidemic typhus and is in contrast to the "microinfarct" ofRocky Mountain spotted fever. The nodules of scrub typhus and epidemic typhusare practically limited to the gray matter, whereas the encephalitis of spottedfever involves the white matter preponderantly.
6. Contrary to the generally held impression, there is a sparsity ofhistologically evident vascular damage in scrub typhus. Arteritis is exceedinglyslight in scrub typhus in contrast with epidemic typhus and Rocky Mountainspotted fever. Accordingly, it is suggested that the designation "diffusevasculitis" when applied to scrub typhus represents an oversimplificationnot justified by the morphologic evidence.
7. It is concluded that the peripheral circulatory failure in patients withrickettsial diseases is a complex phenomenon which cannot be explained solely onthe basis of
34See footnotes 5, p. 115; and 26, p. 125.
35Allen, A. C., and Spitz, S.: A Comparative Study of the Pathology ofScrub Typhus (Tsutsugamushi Disease) and Other Rickettsial Diseases. Am. J.Path. 21: 603-681, July 1945.
133
morphologic damage of vessels. The contributory role of the adrenal gland inthe circulatory failure is suggested.
8. The evidence of lymphoblastic origin for the cellscharacterizing the interstitial infiltrate is presented. The identification ofthe large "basophilic macrophage" with the "acute splenic tumorcell" is suggested and the evidence pointing toward the association ofthese cells with an allergic response is given.
Finally, Allen and Spitz proposed a much broader pathologicalconcept of the rickettsioses than that generally held at the time of theirstudies, as follows:
From the pathologic point of view, the rickettsioses havelong been regarded as a form of diffuse vascular disease. Surely, thisimpression is almost inescapable after a study of epidemic typhus and spottedfever. However, the histology of scrub typhus may perhaps warrant a change inthe direction of emphasis. Although focal, more or less bland thrombophlebitisin scrub typhus is not uncommon, actual arteritis occurs rarely, and, in ourseries, was never of the fibrinoid variety seen in louse-borne or tickbornetyphus. Moreover, the arteritis of scrub typhus does not seem to be a lesion suigeneris, but, rather, appears to be secondary to an extension of theperiarterial infiltrate into the wall. This interpretation was made previouslyby Kouwenaar. Yet, notwithstanding the disparity in the histologic evidences ofvascular damage, there are basic clinical, etiologic, and, in many respects,immunologic similarities between scrub typhus and the other rickettsioses.Therefore, perhaps, a re-evaluation of the significance of the pathologicchanges is in order. A close analogy to this problem is found in anonrickettsial disease-acute disseminated lupus erythematosus (Libman-Sacksdisease). The prominence of the degeneration of vessels in many organs ledinitially to the concept that this entity was a diffuse vascular disease.However, further studies prompted a broader concept; namely, that disseminatedlupus erythematosus was in effect a disturbance of collagen, be it of a vessel,a cardiac valve, or a serous membrane. Moreover, the histologic and clinicalpictures were such as to suggest a hyperergic reaction. The analogy may beextended by reference to periarteritis nodosa and to the arteritis that followsadministration of sulfonamides. In other words, in the over-all view of thepathologist, the more remote, possibly hyperergic effects of the rickettsiae-theeffects of the adrenal gland, on the glomeruli, and on the production ofinterstitial inflammation-assume more importance than the direct damage wroughtby the localization of the rickettsiae.
LABORATORY AIDS IN DIAGNOSIS
The Weil-Felix reaction-In 1929, Fletcher, Lesslar, andLewthwaite,36 while studying two forms of tropical typhus, discovered aserological difference which proved to be of considerable diagnostic value. Theyfound that sera obtained from cases of rural typhus agglutinated in highdilution suspensions of the Kingsbury (K) strain of Proteus, in contrastto sera in urban typhus cases, which agglutinated the Proteus OX-19micro-organisms. Subsequent studies showed the rural form to be the mitebornescrub typhus and the urban form to be murine (fleaborne) typhus.
During World War II, the Weil-Felix reaction was extensivelyemployed in the laboratories of U.S. Army medical installations. Observations atthe Virus and Rickettsial Diseases Laboratory, Army Medical Center, Washing-
36Fletcher, W., Lesslar, J. E., and Lewthwaite, R.: The Aetiology of the Tsutsugamushi Disease and Tropical Typhus in the FederatedMalay States. Tr. Roy. Soc. Trop. Med. & Hyg. 23: 57-70, June 1929.
134
ton, D.C., clearly indicated the Proteus OX-Kagglutination was not elicited by sera from other rickettsial diseases, and was,therefore, a valuable diagnostic aid for scrub typhus.37
Meanwhile, Zarafonetis,38 working in the Cairo laboratory of theU.S.A. Typhus Commission, carried out a series of studies in typhus-vaccinatedindividuals to determine what serological effects may result from vaccinationalone. Only one instance of OX-K agglutination was found, and there was nochange in titer following booster vaccination. He concluded that typhusvaccination did not cause an increase in OX-K antibodies.
Proteus OX-K agglutination tests were also performed on serial bloodspecimens from 104 acutely febrile patients who had been previously vaccinatedwith Cox-type epidemic typhus vaccine. No anamnestic reactions were detected inthese studies.
The final phase of these observations was concerned with the serologicalfindings in typhus fever patients who developed their illness despite priorvaccination with Cox-type epidemic typhus vaccine.39 Again, such patients werefound not to develop agglutinins for suspensions of Proteus OX-K.
Thus, there appeared to be no need to modify the previously heldinterpretation of rising OX-K titers, that this finding was essentially specificfor scrub typhus. However, in tests on sera from 51 cases of lousebornerelapsing fever, Zarafonetis, Ingraham, and Berry40 found that all patients hadProteus OX-K titers of 1: 40 or more in at least one serum specimen. Thetiters ranged from 1: 40 to 1: 2,560. These workers suggested, therefore, that"since there is considerable overlapping in the geographic distribution oftsutsugamushi disease and louseborne relapsing fever, it becomes necessary tointerpret Weil-Felix OX-K results with caution, particularly when atypical casesof either disease are in question."
Complement fixation tests-At the outset and well intoWorld War II, the Weil-Felix Proteus OX-K agglutination test was the onlyavailable serological test of value in the diagnosis of scrub typhus. It wasfortunate that this nonrickettsial antigen appeared to be specific for scrubtyphus with the single exception of louseborne relapsing fever that has beennoted. In view of important progress in the development of specific rickettsialantigens for use in complement fixation and agglutination tests for otherrickettsial diseases, it was anticipated that similarly successful results wouldfollow with
37Plotz, H., Wertman, K., and Bennett, B. L.: TheSerological Pattern in Epidemic Typhus Fever. II. The Weil-Felix Reaction.Division of Virus and Rickettsial Diseases, Army Medical School, Army MedicalCenter, Washington, D.C., 1944. [Official record.]
38(1) Zarafonetis, C. J. D.: Serologic Studies in Typhus-VaccinatedIndividuals. I. The Effect of a Stimulating Dose of Typhus Vaccine on theWeil-Felix and Complement-Fixing Antibodies. J. Immunol. 51: 365-374, November1945. (2) Zarafonetis, C. J. D.: Serologic Studies in Typhus-VaccinatedIndividuals. II. The Effect of Non-Typhus Fevers on the Weil-Felix andComplement-Fixing Antibodies. J. Immunol. 51: 375-388, December 1945.
39Zarafonetis, C. J. D., Ecke, R. S., Yeomans, A., Murray,E. S., and Snyder, J. C.: Serologic Studies in Typhus-Vaccinated Individuals.III. Weil-Felix and Complement-Fixation Findings in Epidemic Typhus FeverOccurring in the Vaccinated. J. Immunol. 53: 15-30, May 1946.
40Zarafonetis, C. J. D., Ingraham, H. S., and Berry, J. F.:Weil-Felix and Typhus Complement-Fixation Tests in Relapsing Fever, With SpecialReference to B. proteus OX-K Agglutination. J. Immunol. 52: 189-199, March 1946.
135
scrub typhus antigens. Bengtson,41 at the NationalInstitute of Health, U.S. Public Health Service, was the first to preparesatisfactory complement fixing antigens from infected eggs by a techniquesimilar to that used for epidemic typhus. Additional experience, however,revealed that there were significant antigenic differences between variousstrains of R. tsutsugamushi. Because of these differences, diagnosticcomplement fixation tests on human sera require the use of several antigensprepared from different strains of R. tsutsugamushi. Thus, while the complementfixation test became available for scrub typhus, its use was limited essentiallyto research laboratories during World War II.
Isolation and identification of strains -Diagnosis ofscrub typhus by isolation of strains is, of course, absolute in contrast to thepresumptive nature of serological tests. Isolation of strains is also desirablefor laboratory comparison of immunity relationships between strains, forpossible vaccine production, and for the preparation of antigens for serologicaltests. In addition, strains may be used in the laboratory evaluation ofchemotherapeutic agents proposed for the treatment of scrub typhus. To theseends, therefore, strain isolations were carried out by several groups of workersin World War II and returned to adequately equipped laboratories for pertinentstudy.42
Prior to these efforts, attempts to isolate and seriallytransmit tsutsugamushi disease in laboratory animals gave negative results inthe hands of many investigators. Japanese workers had been successful withrabbits, using intraocular injections. Mice were later found to be readilysusceptible but were, unfortunately, not available for use in the fieldlaboratories. Guinea pigs were susceptible to many tsutsugamushi strains, butapparently not to others. This was the situation in January 1944, when theImphal, Ceylon, and Calcutta strains of scrub typhus were received in the Cairo,Egypt, laboratory of the U.S.A. Typhus Commission. These strains had beenmaintained at the district laboratory in Calcutta and at the Haffkine Institutein Bombay, India, by British workers and were forwarded to Brig. Gen. Leon A.Fox, Field Director, U.S.A. Typhus Commission, by Lt. Col. M. H. P. Sayers,RAMC. The strains were received in rabbits, infected intraocularly, and weremaintained for several passages in rabbits. The infectivity of these strains fortwo desert rodents, namely, Gerbillus pyramidum and Gerbillus gerbillus, wasthen tested.43 They were found to be highly susceptible to infectionwith R. tsutsugamushi and were, therefore, a suitable substitute in theabsence of a supply of white mice. Since these rodents were plentiful, it
41Bengtson, I.A.: Complement Fixation in Tsutsugamushi Disease (ScrubTyphus). Pub. Health Rep. 61: 895-900, 14 June 1946.
42(1) See footnote 1, p. 111; and 13, p. 121. (2) Philip, C. B., Woodward,T. E., and Sullivan, R. R.: Tsutsugamushi Disease (Scrub or Mite-Borne Typhus) inthe Philippine Islands During American Reoccupation in 1944-45. Am. J. Trop.Med. 26: 229-242, March 1946. (3) Letter, Lt. Col. M. H. P. Sayers, R.A.M.C.,Assistant Director of Pathology, 14th Army, Calcutta, India, to Brig. Gen. L. A.Fox, U.S.A. Typhus Commission, Cairo, Egypt, 4 Jan. 1944.
43Zarafonetis, C. J. D.: The Susceptibility of the Rodents,Gerbillus pyramidum and Gerbillus gerbillus, to ExperimentalTsutsugamushi Infection (Scrub Typhus). Proc. Soc. Exper. Biol. & Med. 59:113-116, June 1945.
136
was possible to accelerate a number of studies in scrub typhus at the U.S.A.Typhus Commission laboratories at Myitkyina as well as in Cairo.
Growth of the etiological agent of scrub typhus in the yolksac of developing chick embryos was accomplished by a number of workers.44This was an important method of experimental vaccine and antigen production.
SPECIFIC TREATMENT
The treatment of scrub typhus includes both general andspecific measures. The general supportive measures consist of nursing care,diet, maintenance of fluid and electrolyte balance, and the management ofcomplicating diseases or conditions. These aspects of therapy have beenadequately indicated in the reports that have been cited.45 Here,consideration will be limited to specific measures, such as the use of immuneserum, antibiotics, and chemotherapy.
Serotherapy-Hyperimmune rabbit serum was found to reducethe mortality of experimental scrub typhus in mice.46 Humanconvalescent serum, however, was without effect on the clinical course of scrubtyphus even when given during the first week of illness.47 Sincethere are wide differences in the antigenic pattern of various strains of R.tsutsugamushi,48 it may be that only homologous antiserum would beeffective therapeutically.
Antibiotics-Although penicillin and streptomycin proved to be valuablefor certain complicating bacterial infections, these antibiotics had nosignificant specific effect on the etiological agent of scrub typhus. Since thewar, the newer broad-spectrum antibiotics have been used with signal success inrickettsial infections.
Chemotherapy-In a report to the Division of MedicalSciences, National Research Council, dated 26 December 1942, Drs. John C.Snyder, J. Maier, and C. Russell Anderson, first demonstrated an antirickettsialeffect of PABA (para-aminobenzoic acid) in mice experimentally infected withmurine (fleaborne) typhus. This led to extensive clinical and laboratory studiesand to experimental trials of PABA in animals infected with R. tsutsugamushi.Gerbilles proved to be of great value in this connection. Snyder andZarafonetis,49 and Murray with these authors50 demonstratedconclusively that,
44(1) See footnote 39, p. 134. (2) Bengtson, I. A.: Apparent SerologicalHeterogeneity Among Strains of Tsutsugamushi Disease (Scrub Typhus). Pub. HealthRep. 60: 1483-1488, 14 Dec. 1945. (3) Lewthwaite, R., and O'Connor J. L.:Prophylactic Vaccine Against the Tsutsugamushi Disease. Second Report on anAttempt to Prepare a Vaccine From Hens' Eggs Experimentally Infected. VirusLaboratory, Commonwealth Serum Laboratories, Melbourne, Australia, 1943.
45See footnotes 1, p. 111; 9, p. 118; and 32, p. 128.
46Topping, N. H.: Tsutsugamushi Disease (Scrub Typhus); The Effects ofImmune Rabbit Serum in Experimentally Infected Mice. Pub. Health Rep. 60:1215-1220, 12 Oct. 1945.
47(1) See footnote 9, p. 118. (2) Hay, C. P.: Scrub Typhus at Port"X." J. Roy. Nav. M. Serv. 30: 127-135, July 1944.
48See footnote 44 (2).
49Snyder, J. C., and Zarafonetis, C. J. D.: Effects of Para-AminobenzoicAcid in Experimental Tsutsugamushi Disease (Scrub Typhus). Proc. Soc. Exper.Biol. & Med. 60: 115-117, October 1945.
50Murray, E. S., Zarafonetis, C. J.D., and Snyder, J. C. : Further Report on Effect of Para-Aminobenzoic Acid inExperimental Tsutsugamushi Disease (Scrub Typhus). Proc. Soc. Exper. Biol. &Med. 60: 80-84, October 1945.
137
properly administered, PABA significantly reduced themortality of experimental scrub typhus in gerbilles. PABA was effective againststrains of R. tsutsugamushi that had come from widely separated regionsincluding India, Ceylon, and New Guinea. On the basis of their observations,these workers strongly recommended a clinical trial of PABA therapy in humanbeings.
Such an opportunity presented itself to Tierney,51 of the CairoUnit of the U.S.A. Typhus Commission, in 1945. Working at the 20th GeneralHospital, in Ledo, Tierney carried out a controlled study on 18 patients withscrub typhus. He administered sufficiently large doses to obtain adequate bloodconcentrations of PABA and found that the treated patients had fewercomplications and shorter fever than a comparable group of untreated subjects.
Although there are now improved means of administering large doses of PABA,52and PABA has itself been supplanted by the newer antibiotics as the treatment ofchoice in scrub typhus, these early studies were of great significance to thoseconcerned with the scrub typhus problem. Lack of an effective vaccine, theinsidious vector, the widespread distribution, and a significant mortality rate,all made scrub typhus an important medical problem affecting morale in thefield. Furthermore, several prominent investigators succumbed to infectionsacquired during the course of laboratory investigations with R. tsutsugamushi.The search for an effective therapeutic agent for scrub typhus during World WarII was stimulated by an awareness of these factors.
Immunization-Information regarding the degree and duration of immunityfollowing an attack of scrub typhus was fragmentary, even at the close of WorldWar II. From the available data, it appeared that a strong and lasting immunitywas induced by the disease, and considerable support was added to this belief byanimal studies in several laboratories.53 Indeed, these observationsindicated that animals experimentally infected with one strain of R.tsutsugamushi are resistant for some months, at least, to inoculation withboth homologous and heterologous strains. Studies in progress at the end ofhostilities, however, revealed that a number of strains of R. tsutsugamushi hadimportant antigenic differences as judged from cross-neutralization tests54 andtoxic neutralization tests,55 as well as from study of the com-
51Tierney, N. A.: Effect of Para-Aminobenzoic Acid in TsutsugamushiDisease. J.A.M.A. 131: 280-285, 25 May 1946.
52Zarafonetis, C. J. D.: Clinical Use of Para-Aminobenzoic Acid. Texas J.Med. 49: 666-672, September 1953.
53(1) See footnotes 1, p. 111; and 13, p. 121. (2) Bell, E. J., and Plotz,H.: Infection and Immunity Following the Intracutaneous Inoculation of ScrubTyphus. Proc. Soc. Exper. Biol. & Med. 59: 143-144, June 1945. (3)Zarafonetis, C. J. D., Snyder, J. C., and Murray, E. S.: Immunity FollowingPara-Aminobenzoic Acid Therapy in Experimental Tsutsugamushi Disease (ScrubTyphus). Proc. Soc. Exper. Biol. & Med. 61: 240-242, March 1946.
54Bell, E. J., Bennett, B. L., and Whitman, L.: AntigenicDifferences Between Strains of Scrub Typhus as Demonstrated byCross-Neutralization Tests. Proc. Soc. Exper. Biol. & Med. 62: 134-137,June 1946.
55Smadel, J. E., Jackson, E. B., Bennett, B. L., and Rights, F. L.: A ToxicSubstance Associated With the Gilliam Strain of R. orientalis. Proc. Soc. Exper.Biol. & Med. 62: 138-140, June 1946.
138
plement fixation antigen.56 From theobservations and from efforts to produce an effective vaccine, it becameevident that much remained to be learned regarding the immune response to thevarious scrub typhus strains.
Vaccine studies-In view of the apparent immunityconferred by an attack of scrub typhus in man and confirmed in the animal workjust noted, it was hoped that a satisfactory vaccine could be developed forscrub typhus as had been done for epidemic typhus. Several laboratories turnedtheir attention to this problem. Fulton57 at the National Institute for MedicalResearch in London prepared vaccines from the lungs of infected mice and cottonrats. He found some protection was afforded by intraperitoneal vaccination tointraperitoneal challenge. Similar observations were made by Smadel, Rights, andJackson58 at the Virus Division of the 1st Medical GeneralLaboratory, 814th Hospital Center, Paris, France, and by Plotz, Bennett, andReagan59 at the Medical Department Professional Service Schools, Washington.D.C. The subcutaneous administration of these vaccines induced no resistance toinfection.
Vaccine suspensions were also prepared from infected yolksacs of developing chick embryos.60 However, these preparations failed toconfer immunity in a wide series of tests. It appears that the ether-extractionmethod, which was so valuable in the production of epidemic typhus vaccine, hadsome inexplicably deleterious effect when applied to suspensions of R. tsutsugamushi.
POSTWAR STUDIES
Immunity-In the period immediately following the war,important work was done on unsettled problems bearing on prophylaxis andtreatment. In studies conducted by Smadel and his associates at the Army MedicalDepartment Research and Graduate School, Washington, D.C.,61 it wasfound, briefly that in human volunteers immunity to the homologous strain of R.tsutsugamushi persisted for at least 1 year, and in some for longer periods.In contrast, resistance to heterologous strains of R. tsutsugamushi was of atransient nature. Even within a month after inoculation with one strain, anappreciable number of persons became ill following injection with a heterologous strain. By the end of a year, all were again susceptible toheterologous
56See footnote 44 (2), p. 136.
57Fulton, F.: Methods For the Study of Mite Typhus-A Progress Report.National Inst. Med. Res., Hampstead, London, 21 June 1944.
58Smadel, J. E., Rights, F. L., and Jackson, E. B.: ScrubTyphus Vaccines Prepared From Formalinized Suspensions of Tissues of White andCotton Rats. Report dated 19 June 1945, Headquarters, First Medical GeneralLaboratory, 814th Hospital Center, U.S. Army.
59Plotz, H., Bennett, B. L., and Reagan, R. L.: Preparation of an Inactivated Tissue Culture Scrub Typhus Vaccine. Proc. Soc. Exper. Biol. & Med. 61: 313-317, March 1946.
60See footnotes 44 (3), p. 136; and 59.
61(1) Smadel, J. E., Ley, H. L., Jr., Diercks, F. H., and Traub, R.:Immunity in Scrub Typhus; Resistance to Induced Reinfection. Arch. Path. 50:847-861, December 1950. (2) Smadel, J. E., Ley, H. L., Jr., Diercks, F. H.,Paterson, P. Y., Wisseman, C. L., Jr., and Traub, R.: Immunization AgainstScrub Typhus; Duration of Immunity in Volunteers Following Combined LivingVaccine and Chemoprophylaxis. Am. J. Trop. Med. 1: 87-99, January 1952.
139
infection. During the period of waning immunity, the disease which resultedfrom the heterologous strain was modified from the classical picture. Indeed,the illness observed in tests shortly after recovery was so mild that it wouldprobably not have been recognized as scrub typhus if rickettsemia had not beendemonstrated. On the other hand, the disease which resulted when the individualswere tested after 1 year generally presented the typical picture of scrubtyphus.
Of further interest is the observation that only about 40 percent of thesubjects who developed illness following reinoculation with scrub typhus after 1year displayed a significant (fourfold) rise in titer of Proteus OX-K agglutinins.
Treatment-In postwar studies, dramatic results in treatment of scrubtyphus and other rickettsial infections were achieved with broad-spectrumantibiotics, such as Chloromycetin (chloramphenicol), Aureomycin (chlortetracycline), and Terramycin(oxytetracycline).62 Aprogram making use of these for chemoprophylaxis under special circumstanceswould not prevent infection but would prevent clinical illness. Field trialssuggest that, if prophylactic medication is continued for 4 weeks or longerafter exposure, it is unlikely that signs of scrub typhus will appear after thedrug is discontinued.63
Vaccines are still (1957) in the experimental stage, but, inview of the immunity factors discussed, it is not anticipated that effectiveprotection will be afforded in this manner.
SUMMARY
During World War II, scrub typhus was found to be endemic infoci distributed throughout a large triangular region of Asia and the Pacificwhich lies between Japan, India, and Australia. Outbreaks are more common intypes of terrain that provide suitable cover for reservoir hosts as well asmoisture conditions favorable to the growth and activity of the vector mites.Wild mice and rats and jungle tree squirrels are known to be naturally infectedwith R. tsutsugamushi. Larval mites of the genus Trombicula transmitthe disease from animal to animal or from animal to man. Since infection of themites can be transovarially acquired, the vector also serves as an importantnatural reservoir of scrub typhus.
62(1) Smadel, J. E., Woodward, T. E., Ley, H. L., Jr., and Lewthwaite, R.:Chloramphenicol (Chloromycetin) in the Treatment of Tsutsugamushi Disease(Scrub Typhus). J. Clin. Investigation 28: 1196-1215, September (pt. 2) 1949.(2) Smadel, J. E., Jackson, E. B., and Ley, H. L., Jr.: Terramycin as aRickettsiostatic Agent and Its Usefulness in Patients With Scrub Typhus. Ann.New York Acad. Sc. 53: 375-384. 15 Sept. 1950. (3) Prezyna, A. P., Teh-Ling, C., Tsu-Lin, W., Dougherty, W. J., and Bond, H. B.: Treatment ofScrub Typhus in the Pescadores Islands With Chloramphenicol, Aureomycin, andTerramycin. Am. J. Trop. Med. 3: 608-614, July 1954.
63Smadel, J. E., Traub, R., Frick, L. P., Diercks, F. H.,and Bailey, C. A.: Chloramphenicol (Chloromycetin) in the Chemoprophylaxisof Scrub Typhus (Tsutsugamushi Disease). III. Suppression of Overt Disease byProphylactic Regimens of Four-Week Duration. Am. J. Hyg. 51: 216-228, March1950.
140
The desert rodents, G. pyramidum and G. gerbillus, were shownto be susceptible to infection with scrub typhus and were valuable experimentalanimals in studies on this disease during World War II.
Clinical picture-After an incubation period of from 6 to18 days, the illness begins abruptly with chilly sensations or rigors, followedby fever, headache, malaise, and anorexia. The face is flushed, and there isconjunctival injection. A primary lesion or eschar is present in nearly all casesat the time of onset. It may have been present for as long as 5 days earlier,and usually persists throughout the active phase of the disease. The escharrepresents the site of infection by the mite. Occasional patients exhibit morethan one eschar. This lesion may be up to 1.0 cm. in diameter and consists of a central tough black scab surroundedby a slightly elevated dull red areola (fig. 9). In moist areas of the body, such as the axilla,groin, and scrotum, the scab is often lacking, and the lesion appears as ashallow punched-out ulcer. It is neither painful nor pruritic. There is oftenregional or generalized lymphadenitis. The spleen may become palpable. Thefever rises in stepwise fashion, reaching 102? to 105?F. by the end of the first week. It usually remains elevated until the thirdweek at which time it subsides by lysis. As the disease passes into its secondweek, the general symptoms are increased except in mild cases.
On about the fifth to the eighth day, a dull red macular eruption appearsfirst on the trunk (fig. 10) and later spreads to the arms, legs, and face. Itvaries in intensity and extent and lasts from 1 to 10 days. Nonproductive coughis common, and in severe cases, bronchitis and pneumonia may appear. Roentgenexamination of the chest may reveal changes similar to those of primary atypicalpneumonia. Indeed, pathological studies suggest that these findings represent arickettsial pneumonitis. Hypotension, tachycardia, and cyanosis may occur at theheight of the disease and are attributed more to peripheral vascular collapsethan to heart failure.
The clinical features presented by the scrub typhus patient reflect thehistopathology of the disease, which is basically a disseminated, focalvasculitis and perivasculitis especially of the vessels of the skin, lungs,heart, and brain.
Leukocyte counts are usually within normal limits, but a moderateleukocytosis may develop during the second week of illness.
Diagnosis-The typical case of scrub typhus is readily diagnosed on theclinical findings. Prior to the appearance of the cutaneous eruption, however,early diagnosis depends upon finding an eschar, since dengue, relapsing fever,malaria, infectious hepatitis, typhoid fever, epidemic typhus, and murinetyphus exhibit many of the same clinical features.
The diagnosis may be confirmed by serological tests. A significant rise intiter in the Weil-Felix Proteus OX-K agglutination testdifferentiates scrub typhus from other rickettsial infections. However, patientswith relapsing fever also develop OX-K agglutinins. Complement fixation tests
141
FIGURE 9.-Eschar on ankle in tsutsugamushi disease.
FIGURE 10.-The rash of tsutsugamushi disease is a faint macularone, frequently transient in nature.
142
with purified suspensions of R. tsutsugamushi may be usefulin the diagnosis, but the sera must be tested with antigens prepared fromseveral different strains of the micro-organism. This is necessary because thereare important antigenic differences between various strains. Although recoveryand identification of the micro-organism proves the diagnosis, it is notfeasible for routine use.
Natural course-After the first week, thesubsequentcourse of scrub typhus may be relatively mild, with a total febrile period of 12 to 14 days, or it may be moderately severe with signs of encephalitis andpneumonitis and fever up to 3 weeks. In severe cases, patients are febrile forabout 3 weeks, have signs of pneumonitis, encephalitis, and often of circulatorycollapse and myocarditis. Thromboses and cerebral or gastrointestinal hemorrhagemay occur. Death is not infrequent in severe cases. However, the mortality ofuntreated cases varies from 1 to 25 percent or more in different series. Thedeath rate rises sharply after the age of 40. There is evidence thatthe course of scrub typhus is likely to be more severe in men who have beenunder the strains of combat or have lacked early treatment.
Recovery from scrub typhus is ultimately complete. Many patients, however,have a prolonged convalescence and present a picture of neurocirculatoryasthenia with sleeplessness, tremulousness, lack of ability to concentrate,easy fatigability, excessive sweating, palpitation, dyspnea, and sense ofintrathoracic pressure.
Relapse is rare in cases that have run their natural course.
Following an attack of scrub typhus, immunity to thehomologous strain appears to persist for at least a year. Resistance toheterologous strains of R. tsutsugamushi, however, is transient.
Therapy-The patient should be placed at bed rest, avoid overexertion, and receive frequent small feedings and adequate fluidintake. Specific therapy should be instituted as early as possible. Scrub typhusduring World War II was found responsive to high doses of PABA(best administered as a 10-percent chilled aqueous solution of the sodium or,preferably, the potassium salt);64 since World War II, chemotherapy has beensuccessful.
Prophylaxis-Prevention of scrub typhus in areas whereit is endemic is difficult, at best, and may be impossible under combatconditions. When feasible, preventive measures should include appropriateclearing and oiling of campsites to change the ecology of the area so that it isunfavorable to rodents and mites. Impregnation of clothing with dimethylphthalate, or with an emulsion of benzyl benzoate and dibutyl phthalate,affords considerable individual protection as mites which come in contact withtreated cloth are killed. In view of the immunity factors that have beenreferred to, it is not anticipated that effective protection will be afforded bythe development of a vaccine.
64See footnote 52, p. 137.
