Medical Science Publication No. 4, Volume II
THURSDAY AFTERNOON SESSION
29 April 1954
MODERATOR
LIEUTENANT COLONEL WILLIAM D. TIGERTT, MC
JAPANESE B ENCEPHALITIS-KOREA 1950*
COLONEL ARTHUR P.LONG, MC
COLONEL ROBERT L. HULLINGHORST,MC
ROSS L. GAULD, M. D.
Background
In this presentation, the scope of this symposium is taken literallyand the experience with Japanese B encephalitis is reviewed for the periodof the Korean war only. In fact, the discussion will be limited essentiallyto the experience of 1950 since time does not allow presentation of morerecent ecologic studies of the virus in avian reservoirs and arthropodvectors. Furthermore, 1950 was the only year when the disease occurredin appreciable numbers in U. N. Forces.
The general aspects of preventive medicine in the Far East were presentedat an earlier meeting of this symposium (1). The military and environmentalfactors involved were referred to briefly in that presentation. These undoubtedlywere of considerable epidemiological importance in the occurrence of JapaneseB encephalitis among U. S. forces in Korea.
Prior to 1950, Japanese B encephalitis had perhaps been considered mostprominently as a hazard in Japan and the islands of the Ryukyus chain,particularly Okinawa. It was known to be present in Korea, however, whereit was specifically demonstrated in 1949 (2). In that year alone,there were over 5,000 cases and more than 2,400 deaths reported among Koreannationals. This disease, then, was among the conditions considered as potentialspecial hazards to U. S. forces operating in Korea (3).
By 1950, a considerable weight of evidence had accrued indicating themost likely vector of Japanese B encephalitis to be the mosquito and themost probable species to be Culex tritaeniorhynchus. Continuingobservations and studies have added further confirmation to these earlierconclusions. Figure 1 shows the relationship of a mosquito population anda mosquito infectivity index to an epidemic of Japanese B encephalitis.These data were not collected in Korea but do serve to demonstrate an apparentrelationship between at least one outbreak of Japanese B encephalitis andthe occurrence of infected
*Presented 29 April 1954, to the Course on Recent Advances in Medicine and Surgery, Army Medical Service Graduate School, Walter Reed Army Medical Center, Washington, D. C.
318
Figure 1.
SOURCE: 406th Medical General Laboratory AnnualHistorical Report, 1950.
mosquitoes in large numbers. (Note: The figure here is takenfrom the Annual Report of the 406th Medical General Laboratory for 1950,page 199.) There is considerable evidence that a reservoir of the diseaseexists not only in the human population but also in domestic animals, particularlyhorses, and in some species of wild birds. A large proportion of the adulthuman population possess neutralizing antibodies against the virus of Japaneseencephalitis. Mosquito control, then particularly in the vicinity of nativehabitations, was accepted as the best means for protection in the Koreanoperation. Valiant efforts were made to achieve this protection but underthe conditions present in Korea, particularly during the early months ofthe campaign, the results attained were inevitably short of those desired.(The extent of malaria infection is further evidence of this fact.)
As an adjunct in the control of this disease, Japanese B encephalitisvaccine was administered, this with a killed virus vaccine produced inthe developing chick embryo (4). This type of vaccine had been inuse
319
in the Far East for American personnel since 1947, having that yearreplaced the mouse brain type of material which was used the two previousyears (5). Despite the 5-year experience with vaccination, however,a true evaluation of its efficacy had not been possible. Its use had beencontinued largely because there was no reliable evidence that it was noteffective. In view of the known hazards from Japanese B encephalitis inKorea, it was considered desirable to vaccinate all U. S. troops thereas well as those in Japan and Okinawa. However, since the supply of vaccinehad been based on normal requirements which did not contemplate war inKorea, there was insufficient material and the vaccination coverage oftroops in Korea was quite variable. This led to the presence of large numbersof both vaccinated and unvaccinated in an area where the disease occurredin appreciable numbers. There was thus provided a certain opportunity forthe evaluation of the vaccine. This will be discussed in some detail later.
The 1950 Experience
The Outbreak. Superimposed on the critical military and medicalsituation existing in the Pusan bridgehead in August 1950 there began toappear in our hospitals in ever increasing numbers a group of patientssuffering from an acute febrile illness with evidence of involvement ofthe central nervous system. In certain instances the clinical picture ofencephalitis with evanescent and changing neurologic abnormalities wasstrongly suggestive of Japanese B encephalitis. Laboratory confirmationof the etiology of a few sporadic cases was established shortly after theoutbreak proper began in mid-August (fig. 2). The peak was reached in Septemberand the epidemic continued into September. Figure 2 also indicates distributionby division which will be described later in the discussion. Approximately300 cases were reported from U. N. troops in Korea with 30 deaths, givinga case fatality of 10 percent. For comparison the time distribution of26 cases occurring in our forces in Japan in 1950 is shown.
Reporting and Diagnosis. The approximately 300 cases presentedin the foregoing figure and forming the basis for the discussion were selectedthrough the utilization of the best reporting and diagnostic proceduresavailable. In accordance with FEC policy, all patients admitted to hospitalswith or subsequently developing signs or symptoms of infectious encephalitiswere reported by radiogram. Acute and convalescent blood specimens weresubmitted to the 406th Medical General Laboratory for serological diagnosison patients so reported. In addition, appropriate specimens obtained atnecropsy on fatal cases were submitted for virus isolation and histologicalexamination.
320
A careful review of some 350 abstracts prepared from clinical recordsresulted in the selection of the approximately 300 referred to. These wereconsidered to have a disease clinically consistent with Japanese B encephalitis.*The greatest problem in this selection was presented by the necessity fordifferentiation from poliomyelitis. Typical cases of encephalitis withthe usual prodromal fevers, mental confusion and coma, coupled with a pictureof changing abnormal reflexes and recovery without paralysis, gave little
FIGURE2.
SOURCE: 406th Medical General Laboratory AnnualHistorical Report, 1950.
difficulty. Borderline cases, however, were troublesome and it is notunlikely that a few errors were made. This likelihood appears even greaterwhen it is realized that about 70 percent of the reported cases could beclassified as moderate or mild. The error here, however, was probably thattoo many cases were excluded. In fact, it appears more than likely thatan appreciable number of individuals diagnosed as having "fever ofundetermined origin" during the period under consideration may, infact, have been experiencing mild infections with the virus of JapaneseB encephalitis.
*Dr. Grant Taylor, Associate Member of the Neurotropic Virus Commission of the Army Epidemiological Board, assisted in this and other aspects of the study.
321
In the group selected on clinical grounds as compatible with encephalitis,adequate serum specimens for serological study were received in 237 instances.(This is rather remarkable and speaks well for the interest and cooperationof the various medical officers caring for those patients.) One hundredand twenty-one of these showed complement-fixing antibody titer rises ofsignificant degree (fourfold or more) and 63 showed rises of questionablesignificance (twofold.) Of those with no rises in titer, 15 maintainedantibody titers of one to four or higher and 38 showed no significant complement-fixationreactions. Thus, in about 85 percent of the group there was specific laboratoryevidence of experience with the virus of Japanese B encephalitis otherthan through vaccination.*
Twenty-eight of the thirty fatalities in the group presented histopathologicalpictures typical for Japanese B encephalitis. One, while not typical, wasconsistent with such diagnosis and the other was considered to be histologicallypoliomyelitis. Eleven strains of Japanese B encephalitis were isolatedfrom central nervous system tissue obtained at necropsy from fatal casesoriginating in Korea.
The most noteworthy clinical laboratory findings were those in the spinalfluid and peripheral blood. Of 211 spinal fluid cell counts on 194 patientsbetween the first and fifth days of the disease, the total count variedfrom 0 to 3,350 WBC/cu. mm. The mean total count of the group was 277 WBC/cu.mm. with a mean of 65 percent lymphocytes. These counts tended to be somewhatlower as the disease progressed and showed an increase in the proportionof lymphocytes to a mean of about 85 percent between the sixth and tenthdays. Remembering that this is a virus disease it is of some interest thatthe white blood count in the peripheral blood was over 12,000/cu. mm. in45 percent and over 15,000 in 28 percent of 260 patients.
Management of Cases. It had been well learned by Far East CommandMedical Services that maximal rest and minimal exertion are of paramountimportance in the management of central nervous system infections, particularlythose involving the higher centers. This concept while given full considerationin 1950 was not completely compatible with the urgent need for the limitedmedical facilities present in Korea at the time. Accordingly, the policyof early and rapid evacuation in all clinically suspected cases of encephalitiswas adopted together with the convention that those patients with markedsigns and symptoms of central nervous system disease were to be retainedand treated in facilities in Korea. Thus, the intensive treatment for severecases such as those with coma was made available to
*Experience has shown that vaccination with the available Japanese B vaccine does not stimulate detectable complement-fixing antibodies unless the individual concerned has had prior experience with the living virus.
322
the fullest extent possible with no more trauma to the patient thanwas absolutely necessary in the face of the situation. This practice provedto be reasonably successful as evidenced by the case fatality rate referredto previously (about 10 percent).
The effects of this practice on the duration of illness and the lengthof convalescence could not be measured. It was seen that under almost anycircumstance of treatment and handling, those with moderately severe andsevere cases required a long period of convalescence for complete recovery.With such convalescent periods, however, serious nervous system residualswere relatively uncommon and complete recovery was seen in many patientswith what appeared to be a very extensive involvement of the central nervoussystem.
No specific value of antibiotic therapy was demonstrated. However, expectanttreatment with these agents undoubtedly saved lives otherwise consignedto demise through intercurrent infections. As already indicated, rest assoon and as complete as possible, plus comprehensive supportive and nursingcare were the principal and most effective therapeutic procedures. Veryimportant among these were measures for relieving pharyngeal pooling ofsecretions, respiratory failures and severe circulatory embarrassment.It is believed that this watchful, careful, sympathetic treatment and managementsaved lives and restored useful function to many severely ill patientswith potentially severe impairments.
Inapparent Infections. The occurrence and significance of inapparentinfections with the virus of Japanese B encephalitis has been referredto and discussed by a number of authors (6-8). The great variabilityin the demonstrable severity of infections in the 1950 outbreak and thelarge numbers of fevers of undetermined origin observed during that timesuggested the occurrence to a significant degree of unrecognizable infectionwith the Japanese B encephalitis among Army troops.
Demonstration of such infections was attempted late in the epidemic.For this purpose, a group of patients evacuated from Korea to Japan forreasons other than central nervous system infections were studied serologically.All of these individuals had been in Korea at least 1 to 2 weeks duringthe period 1 August to 30 September 1950. In addition, only those of knownimmunization status as determined by examination of individual immunizationrecords were selected. Forty-nine percent of 90 completely vaccinated andexposed patients and 57 percent of 149 who had received no vaccine showedappreciable amounts of antibody as determined by the neutralization test.In support of the resultant deduction that roughly half of all troops inKorea during the epidemic period were infected with the virus, it
323
should be mentioned that no neutralizing antibodies could be demonstratedin 140 American soldiers newly arriving in Japan the following year.
Evaluation of Vaccine
Reference has already been made to the matter of evaluation of JapaneseB encephalitis vaccine. For obvious reasons, such an evaluation was ofextreme importance and considerable efforts were made to glean from thesituation the best information available. As indicated before, those forcesalready in Japan and Okinawa were vaccinated early in the season. Thesevaccinations started in Okinawa about 1 May, in southern Japan about 15May, and in the remainder of Japan about 1 June. For those previously unvaccinated,three doses of 1 cc. each of the vaccine were administered, the secondinjection being given a week after the first and the third 3 weeks later.In accordance with the recommendations of the Neurotropic Virus Commission(12 April 1949) those arriving in Japan after 15 June and in Okinawa after1 June, were given the full course of three doses of vaccine at intervalsof 2 to 4 days between doses. Individuals who had received as much as twodoses of the Japanese B encephalitis vaccine during a previous season weregiven a single booster dose.
The "lead time" for procurement of Japanese B encephalitisvaccine was several months. Hence, the unexpected onset of the Korean campaignallowed insufficient time for the procurement of adequate quantities ofthe material for administration to all troops sent into the area. Thus,the early forces dispatched from Japan to Korea were reasonably well vaccinatedwhile additional troops who were committed, particularly those sent directlyfrom the United States, did not receive vaccine. Units and replacementsprocessed through Japan were in many instances either not vaccinated oronly partially so. Such vaccination of these personnel as was accomplishedwas done on an accelerated schedule, some receiving the agent on alternatedays. There resulted a force of American troops some of whom were vaccinatedand some of whom were not.
With the appearance of Japanese B encephalitis in this population, attemptswere made to determine, if possible, the immunogenic efficacy of the vaccine.Here, only a brief résumé of the findings and conclusionscan be presented. To do this, however, it is necessary to indicate theseveral categories of vaccination status which were established by definition.These were:
Vaccinated. Those personnel who had been vaccinated in accordancewith theater regulations. This included those who had received the vaccineon an accelerated schedule as well as those who had received an initialcourse previously and whose vaccination had been kept current by one ormore booster injections.
324
Partially vaccinated. Those individuals who had received twoor more doses of Japanese B encephalitis vaccine regardless of time relationbut whose immunization record showed that their vaccination was incompleteor not kept current.
Unvaccinated. Those personnel who had never received JapaneseB encephalitis vaccine and those whose immunization record showed thatthey had at some time received a single dose of vaccine only.
The determination of vaccination status was made from a review of theindividual immunization record. Unless this record was reviewed, the personwas classified as vaccination status unknown. Despite the diligent effortsof a number of workers, complete vaccination information was unobtainable.For example, 40 percent of the immunization records of the cases and deathswere never located. It was, therefore, not possible to determine the proportionof troops in the area who were vaccinated nor to compare the incidenceof disease in vaccinated and unvaccinated troops. Thus, the early hopesfor a well defined and valid statistical evaluation of the vaccine werenot realized. It was possible, however, to estimate within certain limitsthe vaccination status of major organizations. With this as backgroundin considering the cases and fatalities in these organizations, certaincomparisons were possible.
Case Fatality
Table 1 presents data concerning case fatalities. It is noted that fourorganizations-24th Infantry, 25th Infantry, First Cavalry Di-
Table 1. Japanese Encephalitis-Korea 1950 CaseFatality by Unit
Unit | Total cases | Confirmed cases | ||||
Cases | Deaths | Percent fatality | Cases | Deaths | Percent fatality | |
Part of personnel vaccinated: | ||||||
24th Infantry | 36 | 5 | 14 | 26 | 5 | 19 |
25th Infantry | 68 | 0 | 0 | 58 | 0 | 0 |
1st Cavalry | 35 | 5 | 14 | 21 | 5 | 24 |
F. E. A. F. | 7 | 1 | 14 | 4 | 1 | 25 |
Personnel unvaccinated: | ||||||
2d Infantry | 104 | 12 | 12 | 76 | 12 | 16 |
1st Marines | 27 | 4 | 15 | 15 | 4 | 27 |
5th R. C. T. | 14 | 1 | 7 | 13 | 1 | 8 |
27th Brigade (British) | 4 | 1 | 25 | 2 | 1 | 50 |
Miscellaneous | 16 | 1 | 6 | 7 | 1 | 14 |
|
|
|
|
|
|
|
325
vision, and the Far East Air Forces-were considered to have been atleast partially vaccinated. The Second Infantry and First Marine Divisions,as well as the Fifth Regimental Combat Team and the 27th Brigade (British)plus certain miscellaneous troops were known not to have been vaccinated.The uniformity of case fatality rates among the divisions is striking,except for that of the 25th Infantry Division in which there were no deathsamong 68 cases although 7 to 10 fatalities might have been expected basedon the experience in the other divisions. This cannot be attributed tovaccination since both the 24th Infantry and the First Cavalry Divisionshad large proportions of their personnel vaccinated. Also, it was determinedthat about one-half of the patients of the 25th Infantry Division whosevaccination status was known were unvaccinated. The data presented here,then, indicate only that there is no evidence that the case fatality washigher in nonvaccinated units than in those who were at least partiallyvaccinated.
Vaccination Status of Cases and Deaths
Table 2 presents available but admittedly incomplete data on the vaccinationstatus of cases and deaths. (As indicated above, only about 60 percentof the immunization records were obtainable on these cases.) Relativelylittle is shown here other than that approximately 20 percent of the patientswhose vaccination status was known had been vaccinated. This does not speakwell for the potency of the antigen used.
Table 2. Japanese Encephalitis-Korea 1950 VaccinationStatus of Cases and Deaths
| Cases |
| |||
Total | Confirmed | ||||
Number | Percent | Number | Percent | ||
Vaccinated | 36 | 19.1 | 31 | 21.4 | 1 |
Partially vaccinated | 9 | 4.8 | 7 | 4.8 | 0 |
Unvaccinated | 143 | 76.1 | 107 | 73.8 | 19 |
Status unknown | 123 | ----- | 77 | ----- | 10 |
Incidence Rates of Units According to Vaccination Status of Personnel
In attempting to compare the incidence of Japanese B encephalitis amongthe various units concerned, many complicated factors were immediatelyinvolved. Among the most important of these was that of the military situationat the time. It is believed that all or nearly
326
all of the cases occurred among troops in the Pusan bridgehead. (Therewas never evidence that those who first entered Korea through Inchon on15 September were exposed to infection.) In preparation for the Inchonmaneuver, some units were withdrawn from the Pusan bridgehead just priorto the new assault, re-enforced to several times their previous strengthand recommitted. These, then, were removed from exposure to encephalitisfor a time. Considering these factors and taking into account the incubationperiod of Japanese B encephalitis (7 to 14 days), a comparison has beenmade based on the strength of units in Korea between 10 August and 10 Septemberand the cases which occurred between 17 August and 30 September. If thisintroduces a bias, it is felt to be favorable to the units at least partiallyvaccinated.
Table 3. Japanese Encephalitis-Korea 1950 Incidenceby Major Units
| Man weeks in Korea (10 Aug. to 14 Sept.) | Cases Japanese encephalitis (17 Aug. to 30 Sept. 1950) | |||
All cases | Confirmed cases | ||||
Number | Rate* | Number | Rate* | ||
Part of personnel vaccinated: | |||||
1st Cavalry | 62,837 | 31 | 49.3 | 21 | 33.4 |
24th Infantry | 65,090 | 31 | 47.6 | 25 | 38.4 |
25th Infantry | 70,268 | 63 | 89.7 | 53 | 75.4 |
F.E.A.F. (Korea) | 16,060 | 6 | 37.4 | 4 | 24.9 |
Personnel unvaccinated: | |||||
1st Marines | 21,327 | 27 | 126.6 | 15 | 70.3 |
2d Infantry | 73,150 | 99 | 135.3 | 73 | 99.8 |
5th R.C.T. | 16,566 | 14 | 84.5 | 13 | 78.5 |
27th Brigade (British) | 3,770 | 4 | 106.1 | 2 | 53.1 |
*Per 100,000 man weeks.
Table 3 shows these comparisons. These findings indicate that, in general,the rate in unvaccinated units was higher than that in those units whowere partially vaccinated. The experience of the 25th Infantry Division,however, stands out sharply. This organization with at least part of itspersonnel vaccinated had rates nearly twice those of units similarly conditionedand slightly higher than that of the Fifth Regimental Combat Team whichwas not vaccinated. A major factor in the inconclusiveness of these datamay well have been the difficulty in measuring or even indicating the degreeof exposure in the various organizations. In an effort to demonstrate thepossible effect of geographic locations (which might influence degree ofexposure) a comparison was made of the incidence of Japanese B
327
encephalitis in seven infantry regiments located in proximity to eachother in the Naktong River Valley and along the seacoast near Masan.* Theapproximate location of these units together with their vaccination statusand cases of encephalitis experienced are indicated in figure 3.**
The exact strength of these organizations was not obtainable but theywere known to be sufficiently alike to make the direct comparison of casenumbers valid. It appears from this information that location rather thanvaccination may well have played the major role in the determination ofthe number of cases which occurred in these organizations. Certainly allof the units in the Naktong River Valley had similar disease experiencesregardless of vaccination status as did those nearer the coast with somesuggestion of a lower degree of exposure among the latter group.
From the study on the evaluation of the vaccine, it appears necessaryto conclude that the evidence obtained pointed neither to the presencenor absence of the ability of the vaccine to protect against Japanese Bencephalitis.
Relation of Prior Injury to Onset of Encephalitis
An interesting and perhaps irrelevant observation was that approximately14 percent of 261 cases give a history of wound or injury in a 2-week periodprior to onset. This contrasts sharply with the highest rate of woundedin action in Korea during any 14-day period of the summer of 1950. Thisrate was approximately 6 percent. This relationship has suggested two hypotheses:
1. A wound or injury favors the development of clinically apparent encephalitis,or
2. Early encephalitis may reduce normal caution or agility thus resultingin increased battlefield exposure and likelihood of being wounded.
Conclusions
1. The greatest problem posed by Japanese B encephalitis in the experienceof United States troops in the Far East was presented during the 1950 summercampaign of the Korean conflict. During that period, there were approximately300 cases of the disease with 10 fatalities.
2. The best known methods of control were practiced very well underthe existing circumstances but were not effective.
*Lieutenant Colonel Wallace gave valuable information with respect to the location of these units at the time.
**Lieutenant Colonel Floyd Berry, MSC, gave valuable assistance in the collection of this information.
328
3. No evidence was adduced demonstrating conclusively either the effectivenessor non-effectiveness of Japanese B encephalitis vaccine as an immunizingagent. (The practice of the administration of this vaccine in the Far Eastwas discontinued after the 1951 season.)
4. No specifically effective therapy was demonstrated; though conservationof the patient's energy and strength, detailed supportive
329
and nursing care and expectant antibiotic therapy for intercurrent infectionswere all unquestionably of extreme value.
5. Only a relatively small proportion of cases of Japanese B encephalitisdevelop permanent central nervous system residual impairment.
6. As many as 50 percent of troops in the Pusan bridgehead of Koreain August and September of 1950 probably developed inapparent infectionswith Japanese B encephalitis virus and, hence, a resistance to the clinicaldisease. It appears that this natural process is at least as effectiveas immunization with vaccines now available.
7. The hazard from Japanese B encephalitis in areas where Americanshave had experience lies in great part with the fear and concern over thedisease as well as from the disease itself.
8. Continued study and search for improved control and management methodsfor Japanese B encephalitis and related central nervous system disordersare indicated. The true significance and importance of these conditions,however, should be recognized and maintained within their proper perspective.
Note. Grateful acknowledgment is made to the manyworkers, both military and civilian, whose contributions form the basisof this discussion. Much of the material presented here is drawn from theAnnual Historical Reports for the 406th Medical General Laboratory forthe years 1950, 1951 and 1952.
References
1. Long, A. P.: General Aspect of Preventive Medicinein the Far East Command. This Symposium, page 247.
2. Hullinghorst, R. L., Burns, K. F., et al.: JapaneseB encephalitis in Korea. J. A. M. A. 145 : 460-466 (17 Feb.), 1951.
3. Estimate of Principal Health Hazards to Troops Operatingin Korea. Far East Command Surgeon's Letter, Vol. 5, No. 7, July 1950.
4. Warren, J., and Hough, R. G.: A Vaccine Against JapaneseEncephalitis Prepared from Infected Chick Embryos. Proc. Soc. Exp. Biol.and Med. 61 : 109-113, 1946.
5. Sabin, A. B.: The St. Louis and Japanese B Types ofEpidemic Encephalitis. J. A. M. A. 122 : 477-486 (19 June), 1943.
6. Bawell, M. B., Deuel, R. E., Jr., et al.: Statusand Significance of Inapparent Infection with Virus of Japanese B Encephalitisin Japan in 1946. Am. J. Hyg., Vol. 51, No. 1, pp. 1-12 January 1950.
7. Deuel, R. E., Jr., Bawell, M. B., et al.: Statusand Significance of Inapparent Infection with Virus of Japanese B Encephalitisin Korea and Okinawa in 1946. Am. J. Hyg., Vol. 51, No. 1, pp. 13-20, January1950.
8. Tigertt, W. D., Hammon, W. McD., et al.: JapaneseB Encephalitis: A Complete Review of Experience on Okinawa 1945-1949. Am.J. Trop. Med., Vol. 30, No. 5, pp. 689-722, September 1950.
