CHAPTER III
Dengue
Richard B. Capps, M.D.
INTRODUCTION
Dengue is an acute febrile illness caused by a filterable virus transmitted by mosquitoes. It was first described by David Bylon in 1779 under the name of joint fever.1 Since then, the disease has come to be recognized as common in many parts of the world, and a number of extensive epidemics have been described.2 Bancroft,3 in 1906, was the first to suggest that transmission might be due to Aedes aegypti. This was conclusively established by Cleland, Bradley, and McDonald in 1916 and 1919,4 Siler, Hall, and Hitchens5 in 1926, and Simmons, St. John, and Reynolds in 1931.6 Subsequently, it was demonstrated that Aedes albopictus, Aedes scutellaris, and Aedes hebrideus7 can also serve as insect vectors. That the etiological agent was a filterable virus was first proved by Ashburn and Craig in 1907.8 In 1929, Blanc, Caminop?tros, Dumas, and Saenz9 found that certain species of monkeys could be infected with the virus and could thus serve as a natural reservoir. This was confirmed by Simmons, St. John, and Reynolds in 1931.
Although dengue is a nonfatal disease, it may assume considerable militaryimportance because of its tendency to occur in massive outbreaks resulting inincapacity of large numbers of men. This type of epidemic is favored by theintroduction of nonimmunes into an endemic area as so often occurred
1Pepper, O. H. Perry: A Note on David Bylon and Dengue. Ann. M. Hist. 3:363-368, September 1941.
2Sabin, Albert B.: Dengue. In Viraland Rickettsial Infections of Man. 2d edition. Philadelphia: J. B. LippincottCo., 1952, pp. 556-568.
3Bancroft, T. L.: On the Etiology ofDengue Fever. Australas. M. Gaz. 25: 17, 1906.
4(1) Cleland, J. B.,Bradley, B., and McDonald, W.: On the Transmission of Australian Dengue by theMosquito Stegomyia Fasciata. M.J. Australia 2: 179-184, 1916. (2) Cleland, J.B., Bradley, B., and McDonald, W.: Further Experiments in the Etiology of DengueFever. J. Hyg. 18: 217, October 1919.
5Siler, J. F., Hall,M. W., and Hitchens, A. P.: Dengue: Its History, Epidemiology, Mechanism of Transmission, Etiology, Clinical Manifestations, Immunity and Prevention.Philippine J. Sc. 29: 1-304, January-February 1926.
6Simmons, J. S., St. John, J. H., andReynolds, F. H. K.: Experimental Studies of Dengue. Philippine J. Sc. 44: 1-251, January-February 1931.
7(1) Daggy, R. H.: Aedesscutellaris hebrideus Edwards; AProbable Vector of Dengue in the New Hebrides. War Med. 5: 292-293, May 1944.(2) Mackerras, I. M.: Transmission of Dengue Fever by Aedes(Stegomyia) scutellaris Walk. in New Guinea. Tr. Roy. Soc.Trop. Med. & Hyg. 40: 295-312, December 1946. (3) Fairley, N. Hamilton:Medicine in Jungle Warfare, 26 Feb. 1945. [Official record.] (4) King, Willard V.: Notes on the Vectors of Dengue inNew Guinea, February 1946. [Official record.]
8Ashburn, P. M., and Craig, C. F.:Experimental Investigations Regarding the Etiology of Dengue Fever. J. Infect.Dis. 4: 440-475, 1907.
9Blanc, G., Caminop?tros, J., Dumas,J., and Saenz, A.: Recherches Exp?rimentales sur la Sensibilit? des Singes Inf?rieurs au Virus de la Dengue. Compt. rend. Acad. d. Sc. 188: 468-470,4 Feb. 1929.
60
during World War II. Since prophylactic measures were limitedto mosquito control and since this was difficult to accomplish during combat,the disease was a definite military hazard throughout World War II inpractically all areas of the Pacific and Asiatic theaters.
INCIDENCE
The incidence of dengue in the U.S. Army during World War II,by theaters of operations, is shown in table 3. It is evident that the diseasewas largely restricted to the Pacific and Asiatic theaters, although scatteredcases were reported from each of the other theaters. The lack of a specificdiagnostic test raises a question as to the validity of the diagnosis,especially where sporadic cases were reported. On the other hand, many caseswere undoubtedly not recognized and were reported under the diagnosis of"fever of undetermined origin." It seems probable that the actualcases exceeded those reported. Finally, it should be noted that the incidenceremained low throughout 1945. This was presumably due to improved mosquitocontrol measures, although in certain areas an increased percentage of immunesmay have also been a factor.
TABLE 3.-Incidence of dengue in the U.S. Army, by area and year, 1942-45
[Rate expressed as number of cases per annum per 1,000 average strength]
Area |
| 1942 | 1943 | 1944 | 1945 | |||||
| Rate | Number of cases | Rate | Number of cases | Rate | Number of cases | Rate | Number of cases | Rate | |
Continental United States | 35 | 0.00 | (1) | (1) | 12 | 0.00 | 14 | 0.00 | 9 | 0.00 |
Overseas: |
|
|
|
|
|
|
|
|
|
|
| Mediterranean | 97 | 0.07 | (1) | (1) | 82 | 0.19 | 13 | 0.02 | 2 | 0.01 |
| Africa-Middle East | 8 | .05 | 4 | .70 | 1 | .02 | 2 | .04 | 1 | .02 |
| China-Burma-India | 7,753 | 17.92 | 143 | 25.14 | 1,150 | 24.96 | 4,050 | 25.39 | 2,410 | 10.84 |
| Southwest Pacific | 48,632 | 28.45 | 3,923 | 58.51 | 5,778 | 29.60 | 26,580 | 47.96 | 12,351 | 13.78 |
| Pacific Ocean Area | 27,365 | 23.67 | 12 | .08 | 11,931 | 40.42 | 14,200 | 35.98 | 1,222 | 3.76 |
| North America | 1 | .01 | 1 | .02 | (1) | (1) | (1) | (1) | (1) | (1) |
| Latin America | 203 | .53 | 117 | 1.14 | 39 | .32 | 33 | .38 | 14 | .19 |
| Total overseas | 84,059 | 8.65 | 4,200 | 8.18 | 18,981 | 12.15 | 44,878 | 12.82 | 16,000 | 3.86 |
| Total Army | 84,094 | 3.46 | 4,200 | 1.38 | 18,993 | 2.82 | 44,892 | 5.98 | 16,009 | 2.25 |
1Troops present in the area; no cases reported.
NOTE.-0.00 indicates a rate of less than 0.005 per annum.
SPECIFIC OUTBREAKS
Australia
In March, April, and May of 1942, an extension epidemic of dengue occurredamong U.S. Army troops stationed in Northern Territory and Queens-
61
land (fig. 2). Approximately 80 percent of all U.S. personnel in this areawere attacked within a period of about 3 months.10 Epidemics were reportedduring January, February, and March of 1943 at Rockhampton and in the Brisbanearea. Four hundred and sixty-three cases occurred among U.S. military personnelin the former outbreak.11 A survey of Rockhampton during this period showedthat 80 percent of more than 6,000 dwellings examined were breedingdengue-carrying mosquitoes. U.S. Army personnel required for mosquito controlvaried from 15 to 55 men; oil was supplied by the Rockhampton City Council (fig. 3).12 The vector in Australia was A. aegypti.13
10Letter, Chief Surgeon, U.S. Army, Services of Supply, Southwest PacificArea, to The Surgeon General, 15 Dec. 1942, subject: Medical Service inAustralia, Section I: Sanitation and Vital Statistics.
11Quarterly Report, Surgeon, BaseSection No. 3, U.S. Army, Services of Supply, Southwest Pacific Area, 27 Apr.1943.
12Quarterly Report, Surgeon,I Corps, U.S. Army Forces in the Far East, 1 Jan. 1943-31 Mar. 1943.
13Essential Technical Medical Data,U.S. Army, Services of Supply, Southwest Pacific Area, for March 1944.
62
New Hebrides and New Caledonia
An extensive epidemic of dengue occurred at Esp?ritu Santo between Februaryand August 1943. Over 5,000 cases were reported in military personnel,representing approximately 25 percent of the base strength (table 4).14 Forseveral months prior to the onset of the epidemic, there had been widespreaddumping of tin cans over the base without regard to sanitary regulations. Thishad resulted in heavy breeding of A. aegypti and Aedes scutellarishebrideus. In June, with the epidemic still continuing, a complete mosquitosurvey of all camp areas and all territory within 500 yards of camp wasinstituted. All possible water containers, including tin can dumps, storedtires, oil drums, machinery, and tarpaulins were spotted on maps. A cleanupcampaign was started employing approximately 300 men, 40 trucks, and other heavyequipment, and by August the epidemic was under control. It is noteworthy thatvery few cases occurred at this base during the succeeding rainy season in 1944.
14(1) Malaria News Letter No. 3,Headquarters, Malaria and Epidemic Control, South Pacific Area, September1943. (2) Stevens, Frank W.: Medicine-South Pacific Area. [official record.]
63
[Rate expressed as number of cases per annum per 1,000 average strength]
Month |
| Esp?ritu Santo | ||
| 1944 | 1943 | 1944 | |
January | 1 | 1 | --- | 1 |
February | 65 | 15 | 441 | 1 |
March | 186 | 120 | 1,095 | 0 |
April | 645 | 56 | 1,713 | 0 |
May | 317 | 16 | 1,531 | 0 |
June | 66 | 5 | 909 | 0 |
July | 30 | 1 | 245 | 0 |
August | 3 | 1 | 82 | 0 |
A less severe epidemic occurred at New Caledonia in 1943, as shown in table4. The wide distribution of breeding places for A. aegyptiand the lack of preventive measures by the resident population contributed tothe persistence of the outbreak. The incidence of infection in militarypersonnel would undoubtedly have been much higher if it had not been forintensive mosquito control measures carried out by the base malaria controlunit. An epidemic was avoided, although cases developed during the 1944 rainy season.
Hawaiian Islands
In 1943, dengueappeared in Honolulu, T.H., in epidemic form for the first time in over 30 years. The evidence suggested that the disease wasimported from Suva, Fiji Islands, where an epidemic was in progress. Twocommercial airline pilots were hospitalized with dengue in Honolulu early in July1943, shortly after arrival from Suva. One of the pilotswas ill upon arrival whereas the other did not develop symptoms for several daysand was not isolated by hospitalization until he had passed through theinfectious period.15 Threeweeks later, two civilian cases appeared in the Waikiki Beach area of Honolulu,and, 12 days later, two cases occurred in Army personnel in the same section.16 Measures were taken immediately to prevent an explosiveoutbreak, consisting chiefly of an extensive program of mosquito control. Also,proper screening of patients in hospitals and in homes was made mandatory,and large areas of the City of Honolulu were placed off limits to troops.Although 1,355 civiliancases were reported through 31 December 1943, only 56cases occurred in military personnel.
15Gilbertson, W. E.: Sanitary Aspects of the Control ofthe 1943-1944 Epidemic of Dengue Fever in Honolulu. Am. J. Pub. Health 35: 261-270,March 1945.
16History of Preventive Medicine,United States Army Forces, Middle Pacific, pp. 231-235. [Official record.]
64
In order to render effective measures for control of mosquitoes in militaryareas, it was necessary to have adequate control in the surrounding civilianareas. Toward this end, the Army gave all possible assistance to civilianagencies. A medical officer was attached to the territorial board of health tomake an epidemiological study of all new cases. Fifty enlisted menwere assigned to spray the buildings and to eliminate breeding places ofmosquitoes in homes where there were cases of dengue. Trucks, ladders, andspraying equipment were made available for use by civilian agencies. InSeptember 1943, it became necessary to extend the program for mosquito controlto include the entire City of Honolulu. The program was supervised by the U.S.Public Health Service, and labor was provided by a medical service company.Honolulu was divided into 3 districts and subdivided into 77 inspection zones,each of such size that one man could thoroughly cover his zone every 10 days.
Aedes mosquitoes were found breeding in all varieties of containers thatcould hold water, such as tin cans, bottles, barrels, jars, flower vases andcups, tanks, tubs, tires, storm drains, catch basins, unstocked fishponds,abandoned cesspools, and cisterns. Breeding places were also encountered inwaterholding plants, such as spider lilies, pineapple lilies, and ape plants andin rotted-out holes and crotches in poinciana, algarroba, haole koa and guavatrees, bamboo and banana stumps, and the larger water-holding pockets intraveler's palms. Other unusual breeding places were in fallen palm fronds andthe holes of lava-formed rocks and pockets in emerged formations of coral reef.It is striking that in over one million inspections, on only four occasions wereground pools found to be breeding places for Aedes mosquitoes. Since bothA. aegypti and A. albopictus have short flight ranges (up to 200yards), it was only necessary to extend control operations to the fringes of theinhabited areas.
The effectiveness of these measures is shown in table 5. Itwill be noted that the breeding indices of Aedes mosquitoes weresatisfactorily reduced. It
TABLE 5.-Breedingindex of Aedes mosquitoesin Honolulu, T.H., from August 1943 to August 1944
Year and month | Index1 | Year and month | Index1 |
1943 |
| 1944 |
|
August | 5.7 | January | 1.0 |
September | 1.7 | February | 1.9 |
October | 1.1 | March | 3.5 |
November | .9 | April | 1.8 |
December | 1.2 | May | 1.0 |
|
| June | 1.1 |
|
| July | .7 |
|
| August | .8 |
1Percentage of premises inspectedin which Aedes larvaewere found.
65
is interesting that the critical index or threshold ofimportance for dengue was 3.0 or less, which is considerably lower than thevalue of 5.0 which is generallyaccepted as the critical point for yellow fever.17
New Guinea and the Philippine Islands
From the onset of operations in New Guinea and adjacentislands, dengue was an important cause of noneffectiveness of troops.18Table 6 shows the case rates for the years 1944 and 1945; separate rates for this areaare not available for the year 1943. It will be noted that the incidence washighest during January and February, which are the months of heavy rainfall.During the first 6 months of 1944, thecase rates for dengue exceeded those for
[Rate expressed as number of cases per annum per 1,000 average strength]
| Number of cases | Rate |
|
|
|
January | 3,137 | 197.6 |
February | 2,849 | 164.1 |
March | 2,469 | 90.6 |
April | 1,848 | 68.0 |
May | 1,970 | 66.1 |
June | 2,756 | 61.2 |
July | 2,613 | 60.1 |
August | 1,571 | 33.6 |
September | 1,828 | 30.5 |
October | 1,102 | 21.5 |
November | 935 | 23.5 |
December | 1,001 | 23.2 |
|
| 24,079 | 54.2 |
| 1945 |
|
|
January | 640 | 24.6 |
February | 576 | 25.8 |
March | 597 | 25.5 |
April | 468 | 30.5 |
May | 353 | 27.3 |
June | 208 | 15.3 |
July | 48 | 6.9 |
August | 222 | 9.2 |
|
| 3,112 | 21.5 |
17Soper, F. L., and Wilson,D. B.: Species Eradication; A Practical Goal of Species Reduction in theControl of Mosquito-borne Disease. J. Nat. Malaria Soc. 1 (No. 1): 5-25, 1945.
18See footnote 7 (3), p. 52.
66
malaria. It will be noted that in 1945 the rates remainedlow, even during the rainy season. This presumably was due to improved mosquitocontrol and to an increased percentage of immunes among military personnel. Asoccurred elsewhere, striking outbreaks appeared in certain units, notably in theHollandia and Biak areas. In the Biak area, the ratio of dengue to malaria was4:1.19
In New Guinea, the vector appeared to be A. scutellaris.20A. aegypti are rare in this area. This increased the problem of mosquito controlbecause of the greater variety of breeding places employed by A. scutellaris(fig. 4). In addition, the day-biting habits of this species made individualprotective measures necessary at all hours, especially in shaded jungle areas.
The situation in the Philippines is of particular interest because itillustrates the effectiveness of measures for controlling mosquitoes. Priorto World War II, dengue had always been a problem among Army forcesstationed in the Philippines. Replacements frequently contracted the diseasewithin a few months after arrival. Consequently, with the opening of thecampaign for reoccupation, a serious situation was anticipated, especially
19Quarterly Report, Surgeon, I Corps, Southwest Pacific Area, 1 Apr.1944-30 June 1944.
20See footnote 7 (4), p. 52.
67
during the rainy season in areas of dense population.However, in spite of the presence of large numbers of nonimmunes and theappearance of cases throughout the Philippine Islands, no real epidemicoutbreaks developed. The case rates are shown in table 7.
[Rate expressed as number of cases per annum per 1,000 average strength]
| Number of cases | Rate |
| 1944 |
|
|
November | 652 | 39.8 |
December | 1,360 | 49.1 |
| 1945 |
|
|
January | 1,033 | 32.1 |
February | 893 | 24.8 |
March | 1,214 | 25.7 |
April | 857 | 19.9 |
May | 913 | 19.6 |
June | 839 | 13.4 |
July | 930 | 16.4 |
August | 1,140 | 15.2 |
September | 439 | 7.9 |
October | 294 | 6.3 |
November | 285 | 6.3 |
December | 89 | 4.1 |
|
| 8,926 | 15.7 |
|
| 10,938 | 17.9 |
Effective mosquito control was achieved by attaching malariaunits to all forces operating in the Philippines. In addition to the usualcontrol measures which were carried out on an intensive scale, area spraying withDDT (dichlorodiphenyltrichloroethane) from airplanes was carried outextensively over Manila (fig. 5) and other populated centers on Luzon Islandduring the early months of 1945.21 The houses of natives adjacent toconcentrations of troops were also sprayed with DDT.
It is of interest that reports from this area indicated considerablevariation in the clinical picture. Thus, many cases showed only a single peak intemperature, and in many instances the disease was quite mild. The difficultiesin diagnosis, particularly during the first few days of illness, were
21Essential Technical Medical Data, U.S. Army Forces in the Far East, forMarch and April 1945, Inclosure 13, subject: DDT Spraying in Luzon.
68
repeatedly pointed out. The disease was most commonly confused with malaria,acute infectious hepatitis, and scrub typhus.
Saipan
Perhaps the most extensive outbreak of dengue during WorldWar II occurred in the Marianas Islands in the late summer of 1944. Relativelycomplete records are available only for Saipan. Shortly after the assault onthis island, on 15 June 1944, dengue made its appearance among the troops. Atfirst, the incidence was low, probably because the rainy season did not beginuntil the first of August and mosquitoes were not abundant. However, by 11August, mosquitoes had become plentiful and the dengue rate had reached 300. Theincidence continued to rise rapidly, and by 8 September the rate had reachedapproximately 3,500 per 1,000 per annum.22
This outbreak was obviously caused by the presence of largenumbers of mosquitoes and the ineffectiveness of ordinary measures of control.Thus, there were innumerable breeding places provided by "an unbelievableamount
22Essential Technical MedicalData, U.S. Army Forces, Pacific Ocean Areas, for September 1944. Inclosure 4thereto.
69
of rubble resulting from the total destruction of villagesand scattered dwellings, a multitude of wells, cisterns, vats, troughs andrainwater-collection facilities as well as an immense quantity of tins, shellcases, et cetera." The difficulties encountered in controlling mosquitoesduring and immediately following the assault phase of a campaign aresufficiently great under ordinary circumstances, but in this instance theyproved to be insurmountable. This is illustrated by counts made of the number ofmosquitoes found biting a single human during 10-minute periods. Thus, between 16August and 10 September 1944, countsmade in the late afternoon near the vicinity of towns or villages showed from 5to 36 specimens of A. aegypti andfrom 2 to 16 specimens of A.albopictus as well as 1 to 7 specimens of other species. Counts made at nightshowed as many as 42 specimens ofstill other species.
Effective control of mosquitoes only became possible when asupply of DDT arrived on 3 September 1944. Areacontrol was employed by spraying 5 percent DDT and kerosene from airplanes. Itwas found that small planes were inadequate and that it was necessary to use C-47'sbecause of the size of the area involved. Between 12 and 22 September, 8,600 gallons of the mixture wassprayed over a total of approximately 15,000 acres, an average of approximatelytwo-tenths of a pound of DDT per acre. In addition, DDT residual spray was usedin all tents and living quarters of hospitals.23 Thiswas accomplished with a truck-mounted power spray unit for chemicaldecontamination provided by the Chemical Warfare Service.
The effectiveness of these measures is shown in table 8. The number of new cases began to decrease significantlyabout 1 week after the aerial spraying of DDT was started. After the first ofOctober, the number of new cases was less than 10 percent of the number whichoccurred at the height of the epidemic. Although the preliminary summaryreports show only 10,834 cases ofdengue for the entire Pacific Ocean Area during August, September, and October, 1944,it is reliably estimated that there were 20,000 cases on Saipan alone.24 This discrepancy was probably dueto a high percentage of cases that were cared for in quarters and were notofficially reported.
The effectiveness of these measures in controlling themosquito population was clearly demonstrated by observations on the "bitingrate" per minute. Thus, surveys made before and after spraying by airplaneindicated a decrease up to 98 percent.In addition, it was generally agreed by troops that there had been a tremendousreduction in the mosquito population. This was further substantiated by surveysof breeding places of mosquitoes made before and after DDT spraying. Finally,the effectiveness of these measures is indicated by the marked decrease in newcases of dengue which occurred before the end of the rainy season and at a timewhen susceptible troops were still arriving on the island.
23Letter, Deputy Surgeon, Headquarters, U.S. ArmyForces, Pacific Ocean Areas, to Surgeon, Pacific Ocean Areas, 30 Sept. 1944,subject: Measures Used for Control of Dengue Fever on Saipan.
24Annual Report, Eighteenth Medical General Laboratory,Pacific Ocean Areas, 1944. Inclosure 5 thereto.
70
| Date | Number | Date | Number |
| 1944 | 1944 | ||
| September 14 | 393 | September 26 | 62 |
| 15 | 426 | 27 | 87 |
| 16 | 294 | 28 | 79 |
| 17 | 306 | 29 | 71 |
| 18 | 289 | 30 | 44 |
| 19 | 275 | October 1 | 36 |
| 20 | 230 | 2 | 33 |
| 21 | 137 | 3 | 27 |
| 22 | 137 | 4 | 28 |
| 23 | 112 | 5 | 32 |
| 24 | 93 | 6 | 23 |
| 25 | 81 |
China-Burma-India
Although dengue was endemic in most of the China-Burma-Indiatheater, the majority of cases among U.S. troops occurred in the region ofCalcutta, India. The highest incidence appeared between July and October withthe peak varying according to the dates of the monsoon.25 During1942, 1943, and 1944, the dengue rates were approximately the same; namely, 25per 1,000 per annum. However, in 1945, the rate dropped to less than half.Although the explanation for this is not entirely clear, it is felt that theimprovement was largely due to the work of malaria control detachments whosemeasures were directed towards Aedes mosquitoes as well as the Anophelesmosquitoes. In addition, antimosquito supplies and equipment were morereadily available in 1945 and individual protective measures were betterenforced (fig. 6).
One small but sharp outbreak of dengue which occurred in theChina theater illustrates the military importance of this disease as follows:During September 1945, after V-J Day, an epidemic was reported in Hankow, China,which was said to have affected 80 percent of the population of the city. WhenAmerican forces occupied the airport, 40 of the first 48 men to arrivecontracted dengue within 5 to 10 days.26 Becauseof this situation, it was first recommended that operations from Hankow besuspended. Subsequently, however, they were considered essential. Intensivemeasures for
25Van Auken, H. A. : AHistory of Preventive Medicine in theUnited States Army Forces of the India-Burma Theater, 1942 to 1945, p. 317.[Official record.]
26Essential Technical Medical Data,U.S. Forces, China Theater, for October 1945.
71
mosquito control were undertaken, the city of Hankow wasdeclared out of bounds, and personal protective measures were rigorouslyenforced. The effectiveness of these steps was indicated by an absence offurther cases among U.S. military personnel.
CLINICAL MANIFESTATIONS
Variations in the character and severity of dengue have long been recognized as dependent upon differencesin particular outbreaks and uponinherent and acquired degrees of individual resistance.27 Thevariabilities of this disease in civilian populations have been discussed indetail elsewhere28
27Lumley, G. F.: Dengue. Service Publication No. 3, Commonwealth of Australia. Department of Health, 1943.
28(1) Simmons, James Stevens: Dengue Fever. In Virus and Rickettsial Diseases. A Symposium Held at the Harvard School ofPublic Health, June 12-June 17, 1939. Cambridge, Massachusetts: HarvardUniversity Press, 1940, pp. 349-364. (2) Simmons, J. S.: Dengue Fever. M. Clin. North America 27: 808-821, May 1943. (3) See footnote 2, p. 50.
72
and provide a comprehensive background for the military casesdescribed in different parts of the Pacific area. These were likewise of varyingseverity and symptomatology,29 and in some instances,because of lack of specific laboratory tests, the diagnosis could be onlydenguelike fever30 based onsymptomatology and epidemiological considerations.
Characteristically, the onset of dengue was sudden,accompanied by fever and sometimes by a chill. The fever was of two types-saddleback and single phase. Patients were usually admitted to thehospital within a few hours of onset of the disease, with a temperature rangingfrom 99? to 104? F. which persisted for at least 2 or 3days.31 Over half the patients then had a remission in which both thefever and symptoms practically disappeared. This remission lasted 1 or 2 daysand was followed by a second rise in temperature and return of symptoms. These,in turn, subsided rapidly on the fifth to seventh day of illness. In patientswho did not exhibit this saddleback type of fever, symptoms and fever werelargely the same but persisted for several days, regressing gradually from thethird to the eighth day of illness.
Dizziness, prostration, and extreme weakness in the legs werefrequent presenting symptoms. Frontal or occipital headache, generalized achingespecially in the back and joints, and pain in and around the eyes developedearly. Anorexia was common and was often accompanied by nausea and sometimes byvomiting. Occasionally, mild abdominal pain was noted. Many patients complainedof insomnia and restlessness, and these symptoms frequently persisted intoconvalescence.
Flushing of the face and neck and scleral injection werecommon. Two types of rash wereobserved. One, an enanthem, appeared within the first 12 hours, if at all, andconsisted of pinpoint-sized, discrete, glistening vesicles on the posterior halfof the soft palate. Star-shaped redness developed beneath these within 24 hours.This rash was morbilliform; it faded during the period of remission and did notrecur with the second rise in temperature. The other type of rash was anexanthem (fig. 7). Seventy-nine percent of the patients in one group had thisrash on admission to hospital. It could be mild (a few discrete, light pink,morbilliform spots on the sides of the thorax, inner surfaces of the upper arms,and in the lumbar region) or severe when it presented unbroken, erythematousareas covering the face below the forehead, neck, shoulders, and thorax. Themorbilliform charac-
29(1) Diasio, J.S., and Richardson, F.M.: Clinical Observations on Dengue Fever; Report of 100 Cases. Mil. Surgeon 94:365-369, June 1944. (2) Kisner, P., and Lisansky, E. T.: Analysis of an Epidemicof Dengue Fever. Ann. Int. Med. 20: 41-51, January 1944.
30(1) Fairchild, L. M.: Dengue-Like Fever on theIsthmus of Panama. Am. J. Trop. Med. 25: 397-401, September 1945. (2) Johnson,J. A., Jr., Martin, W. B., andBreslow, L.: Dengue-Like Fever on Okinawa. Bull. U.S. Army M. Dept. 5: 306-311,March 1946.
31(1) See footnote 29, p. 72. (2)Cavanagh, J. R.: Dengue; Observations on the Disease as Seen in the South Pacific Area. War Med. 4: 549-555, December 1943.
73
FIGURE 7.-Rash of dengue fever on chest andback.
74
ter became apparent at the edges of the confluent areas,notably the lower half of the upper arm, upper part of the abdomen, and thoracicextensions of the axillary spaces. This rash also faded during remission butrecurred, usually in less intense form, with recrudescence of fever. When itextended to the palms of the hands and soles of the feet, it was frequentlyfollowed by itching.
The incidence of adenopathy was highly variable and, whenpresent, it most commonly involved the cervical nodes,32 persistingthrough the second phase of fever and subsiding gradually during convalescence.Bradycardia was usually present and first appeared after the second day ofillness, sometimes lasting into convalescence. Leukopenia was present early,with an average reduction in number of white blood cells to 5,450 per cubicmillimeter. By the fifth day of illness, the average count numbered 3,500 cellsper cubic millimeter, with a relative lymphocytosis. Atypical lymphocytes withvacuolated cytoplasm and coarse granular inclusions were commonly seen. Thesymptoms and signs of disease in two groups of patients in the U.S. Army arerecorded in table 9.
32See footnote 29, p. 72.
75
TABLE 9.-Physical signs andsymptoms of 418 cases of dengue in two groups of U.S. Army patients, 1944
Condition |
| South Pacific |
Physical signs: |
|
|
| Fever | 100 | 97 |
| Saddleback | 55 | 66 |
| Single temperature rise | 45 | 30 |
| Intermittent temperature rise | 0 | 1 |
| Afebrile | 0 | 3 |
| Rash | 79 | 37 |
| Flushing of face (initial phase) | 33 | 26 |
| Tongue coated | 53 | --- |
| Bradycardia (after 2d day) | --- | 97 |
| Scleral injection | 89 | 26 |
| Eyeball tenderness to pressure | 45 | --- |
| Adenopathy | 100 | 17 |
| Cervical | 94 | 193 |
| Epitrochlear | 90 |
|
| Inguinal | 80 |
|
| Laryngeal or oropharyngeal vascular congestion | 17 | 11 |
| Leukopenia | --- | 100 |
| Red blood cell count | Normal | Normal |
| Febrile albuminuria |
| 8 |
Symptoms: |
|
|
| Feverishness | 100 | 97 |
| Chills or chilliness | 54 | 32 |
| Aches and pains | --- | 99 |
| Headache | 94 | 69 |
| General aching | 75 | 35 |
| Lumbar backache | 89 | 43 |
| Orbital pain | 74 | 25 |
| Anorexia | 85 | --- |
| Loss of taste | 55 | --- |
| Bitter taste | 45 | 4 |
| weakness (early) | 29 | 14 |
| Insomnia | 53 | 6 |
| Coryza (fleeting) | 24 | --- |
| Sore throat (mild) | 12 | 3 |
| Stiff neck | 14 | --- |
| Dizziness and nausea | 29 | --- |
| Constipation | 2 | 2 |
| Pruritus | 20 | 3 |
| Photophobia | --- | 5 |
|
| 100 | 318 |
1Of those with adenopathy; 16 percent of the total group (318 cases).
Source: (1) Diasio, J. S., and Richardson, F. M.: Clinical Observations onDengue Fever; Report of 100 Cases. Mil. Surgeon 94: 365-369, June 1944. (2)Kisner, P., and Lisansky, E. T.: Analysis of an Epidemic of Dengue Fever. Ann.Int. Med. 20: 41-51, January 1944.
The disease ran its course in 6 to 10 days (average hospitalstay 7? to 9 days) with complete recovery. In general, symptomstended to abate with the fever, but during convalescence some degree ofneurasthenia manifested by muscular weakness, lack of ambition, mentaldepression, insomnia, and anorexia was almost invariable. These usuallydisappeared in 7 to 14 days, but sometimes lasted much longer.33 Recurrenceswere rare and could probably be explained as re-infections.
Occasional complications of dengue were observed, includinghemorrhagic nephritis, trismus of the jaw, arthritis of hip, suppuration ofglands, persistent bradycardia, and purpuric manifestations.34Urogenital complications were also occasionally observed. Dull testicular painand impotence were not infrequent during convalescence, and the latter wasattributed to generalized weakness.35 In one group of 141 patients, 8men
33(1) See footnotes 29, p. 72; and 31 (2), p. 72. (2)Hyman, A. S.: The Heart in Dengue; Some Observations Made Among Navy andMarine Combat Units in the South Pacific. War Med. 4: 497-501, November 1943.
34See footnote 31 (2), p. 72.
35Weyrauch, H. M., and Gass, H.: Urogenital Complications of Dengue Fever.J. Urol. 55: 90-93, January 1946.
76
(5.7 percent) had late involvement of the urogenital tract,including orchitis, with subsequent atrophy of the testis and repeated bloodyseminal emissions. Rare neurological complications following dengue weredescribed by Kaplan and Lindgren36 who reported palsy of the facial, palatine,long thoracic, ulnar, peroneal, and sciatic nerves.
Dengue may be confused with rubella because of the rash andbecause of the type of cervical adenopathy observed. Other diseases frequentlyconsidered in differential diagnosis included scarlet fever, infectious mononucleosis, malaria, viral pneumonia, influenza, and occasionally meningitis. Aknowledge of the epidemiology of dengue and its clinical course was helpful inearly diagnosis.
Sufficient discrepancies in signs and symptoms existed incertain outbreaks of denguelike fever to question the diagnosis. In 32 casesfrom Panama,37 the symptoms and course of the disease were similar to dengueexcept that the typical rash was not seen, bradycardia did not occur, and theincidence of lymphadenopathy was low. Definite diagnosis was impossible becauseof lack of laboratory methods. Many of the patients observed in Okinawa in 1945likewise had most of the diagnostic signs of dengue.38 The clinicalpicture presented by these cases was remarkably uniform. Specifically noted wasthe sudden onset, with chilly sensations, rapid rise in temperature, headache,pain on movement of the eye, postorbital pain, photophobia, generalized aching,periarticular soreness, conjunctivitis, lymphadenopathy, bradycardia, andhematological changes. However, it was pointed out that the short duration,relative infrequency of recrudescence of fever and symptoms, and the absence ofrash in these patients were against the diagnosis of dengue. Later, Sabin,39on the basis of failure to produce dengue in four human volunteers inoculatedwith serum from five of these patients, suggested that this outbreak may havebeen leptospiral meningitis.
The treatment of dengue was symptomatic. Codeine (?to 1 gr.) and acetylsalicylic acid (10 gr.) usually sufficed for relief ofpain,40 and phenobarbital (1? gr.) was employed forinsomnia and restlessness. Morphine was rarely required.
SCIENTIFIC INVESTIGATIONS
The military importance of dengue became evident early in World War II. As aresult, the Commission on Neurotropic Virus Diseases, Army EpidemiologicalBoard (Board for the Investigation and Control of Influenza and Other EpidemicDiseases in the Army), became interested in
36Kaplan, A., and Lindgren, A. J.: Neurologic Complications Following Dengue. U.S.Nav. M. Bull. 45: 506-510, September 1945.
37See footnote 30 (1), p. 72.
38See footnote 30 (2), p. 72.
39Sabin, A. B.: Research on Dengue During World War II. Am. J. Trop.Med. 1: 30-50, January 1952.
40See footnote 31 (2), p. 72.
77
promoting research in this field with the specific objectivesof producing a protective vaccine and, if possible, of developing a specificdiagnostic procedure. Neither of these was available before World War II, for upto that time the virus had not yet been definitely propagated in animals or intissue culture. Lt. Col. Albert B. Sabin, MC, was the central figure in theseries of studies which provided pertinent key information. Detailed reports ofhis findings are available elsewhere.41The following summarizesthose findings especially pertinent to military medicine:
1. Proof of the existence of multiple immunological types.Seven strains of dengue virus were isolated from patients who contracted theirillness in Hawaii, New Guinea, and India. Serum obtained during the first 48hours of the disease was shipped to the United States under refrigeration andinoculated into volunteers who had never resided in areas where dengue isendemic. Subsequently, studies of transmission by mosquitoes were conducted with A. aegypti. The existence of separate strains was demonstrated both bystudies of cross-immunity and through virus neutralization tests using immuneserum. At least two immunologically distinct types of virus were identified. Itwas shown that the Hawaiian strain, one of the four New Guinea strains, and thetwo Indian strains were identical. Thus, two separate strains were detected inNew Guinea.
2. The long persistence of immunity to homologous types of virus. It wasfound that homologous immunity persisted for at least 18 months under conditionswhich precluded reinforcement of immunity by subclinical reinfection.Heterologous immunity was observed but persisted for onlyabout 2 months.
3. The modifications of the clinical manifestations of thedisease which result from reinfection with a heterologous type of virus atvarious periods after the primary attack. It was shown that a superimposedheterologous infection produced marked variations in the course of the diseasedepending upon the time which lapsed from the primary infection. A short febrileillness of 2 days' duration with or without rash, lymphadenopathy, or othercharacteristic manifestations occurred under these circumstances. In theseexperimental cases, the virus, recovered from the blood showed conclusively thatthe observed disease actually was dengue. This knowledge readily explained themany cases reported of the transitory acute illnesses which could not bespecifically identified, particularly from New Guinea. This explanation isconfirmed further by the fact that two of the New Guinea samples of serum fromindividuals with this type of illness produced the
41(1) See footnotes 2, p. 50; and 39, p. 76. (2) Sabin, A. B., andSchlesinger, R. W.: Experimental Studies on Humanand Mouse Adapted Dengue Virus. Paper presented at joint meeting of AmericanSociety for Experimental Pathology and American Association of Immunologists,Atlantic City, N.J., 13 Mar. 1946. (3) Letter, Maj. Albert B. Sabin, MC, toPreventive Medicine Service, Office of the Surgeon General, 3 Aug. 1944, subject: Isolation ofSeveral Strains of Dengue Virus From Serum of Patients With Various Types of Fevers in New Guinea. (4) Sabin, A.B., and Schlesinger, R. W.: Production of Immunity to Dengue With Virus Modified by Propagation in Mice. Science 101:640-642, 22 June 1945.
78
classical picture of dengue when inoculated into nonimmune volunteers in theUnited States.
4. The demonstration that type-specific immunity to dengue is associated withneutralizing antibodies for the virus. It was shown that the demonstration ofneutralizing antibodies can be employed for diagnostic and epidemiologicalsurvey purposes. They should prove to be of great value in the future.
5. The propagation of dengue virus in mice. Sabin finallysucceeded in propagating the virus in mice. In the course of a number ofpassages, the virus underwent a mutation since it lost its capacity to producesevere illness in man but retained its capacity to produce a rash. A mousebrainextract of this modified virus was found to afford complete protection againstthe naturally acquired disease. A single mousebrain was shown to contain atleast 10,000 immunizing doses which led to the preparation of an effectivelyophilized vaccine. Unfortunately, large-scale field trials were notundertaken because the opportunities for testing the vaccine ceased with theending of the war.
6. Investigations of dengue, conducted by Colonel Sabin inPanama, showed clearly that dengue has occurred there since 1941 and suggestedthat the interior of Panama may be an endemic focus of the disease.
In conclusion, it should be pointed out that the history ofdengue during World War II reflects great credit on the U.S. Army MedicalDepartment. Like some other virus diseases, dengue was promoted to the categoryof an illness diagnosable by laboratory means and preventable.
