Murine typhus in Kuwait in 1978

Bulletin of the World Health Organization, 60 (2): 283 - 289 (1982) Murine typhus in Kuwait in 1978 ABDUL RAHMAN AL-AWADI,' NOURI AL-KAZEMI,2 GAAFAR EZZAT,3 ALFRED J. SAAH , CHARLES SHEPARD, TALAAT ZAGHLOUL, & BEATRIZ GHERDIAN Murine typhus occurred in 254 individuals in Kuwait between April and August 1978; 81N% ofpatients were aged between 15 and 44years, and 630% were male. The highest attack rate occurred among people in the lowest socioeconomic class. The outbreak coincided with a period of high rat andflea density. A study of the first 104 cases suggested that infection was acquired in the home, but a case -control study revealed no significant differences between patients and control subjects in terms ofexposure to rats or domestic animals, and other factors. This suggests that murine typhus is hyperendemic rather than epidemic in Kuwait. The disease is being controlled through reduction ofboth flea and rat populations. Murine typhus may be much more common in many areas than is generally realized, and its status should be re-evaluated regularly in all known and potentialfoci. Murine typhus occurs in many areas of the world but its importance is not generally realized (1). Burg- dorfer (2) estimated that unregistered cases out- numbered recorded ones by nearly 4 to 1, and there is increasing evidence from many countries that rickettsial diseases, especially murine typhus, consti- tute a large proportion of acute febrile illnesses. They should therefore be considered when diagnosing febrile disease, especially in areas with a large rodent population. STUDY AREA Kuwait is an Arabian Gulf State with a population of 1.2 million. The country covers 17 000 kM2, but almost all the inhabitants live in the 17 km2 that com- prise the city of Kuwait. The population can be classified in three distinct grades by socioeconomic status, living conditions, and sanitation, as follows: Grade 1 Those with a high standard of living who live in highly organized districts with good sani- tation. Grade 2-Labourers or unskilled workers of various nationalities, who live in semi-organized dis- tricts. The population consists largely of men, unmar- 'Minister of Public Health, Kuwait. 2 Director of Public Health and Planning, Kuwait. 3Head, Division of Preventive Medicine, Kuwait. 4WHO Temporary Adviser, Centers for Disease Control, Atlanta, GA, USA. ' Centers for Disease Control, Atlanta, GA, USA. 6 Head, Vector and Rodent Control Division, Kuwait. 7 Senior, Virology Laboratory, Public Health, Kuwait. ried or living apart from their family, and sharing crowded quarters. Grade 3 -This group has a mobile and changing population of Bedouin families who live in encamp- ments referred to as Asheesh complexes. The govern- ment of Kuwait has started to move this group into modern houses in new settlements that are provided with preventive and curative medical services. Typhus was first diagnosed in Kuwait in 1974 as an isolated case of the louse-borne variety. The close contacts of the patient were dusted for lice, and no other cases were reported. In the period 1975 - 77, 3 cases of typhus were identified; one was an imported case of louse-borne typhus, while the others were in- digenous cases of an unspecified type. In May 1978, typhus was diagnosed in an individual who had no history of exposure to the illness in the previous 2 weeks. The diagnosis was confirmed by Weil-Felix agglutination and complement fixation tests. A review of hospital records revealed that 7 cases of typhus had occurred in April 1978. The increased recognition of the disease was attributed to the recent acceptance of ampicillin or sulfamethoxazole and tri- methoprim for the treatment of patients thought to have typhoid fever. Previously, the use of chlor- amphenicol may have masked the symptoms of typhus. Subsequently, Weil-Felix agglutination tests were carried out on every patient admitted to hospital with a clinical history suggestive of typhus, and an out- break of the disease was confirmed. This report describes the outbreak, the clinical spectrum of illness, the results of a case-control study, and the control measures used. 4175 -283 284 A. R. AL-AWADI ET AL. MATERIALS AND METHODS Confirmation of diagnosis A total of 38 serum samples were taken from 24 subjects, comprising 13 controls (12 with measles and I with infectious mononucleosis), and 11 patients thought to have typhus on the basis of a clinically compatible history and a positive Weil-Felix (OX19) result (i.e., a 4-fold rise in titre or a single titre of 1:160). The serum samples were submitted blind to the Centers for Disease Control (CDC), Atlanta, Georgia, USA, for serological tests. The indirect fluorescent antibody (IFA) test (3) was performed the day after receipt of the specimens in Atlanta. The antigens used were rickettsia of murine typhus, epi- demic typhus, and Rocky Mountain spotted fever; anti-immunoglobulin (anti-Ig) and anti-IgM were used as conjugates. Based on the results obtained, antibody-inhibition tests to distinguish between murine and epidemic typhus (3) were done on the next day. In these tests the patient's serum was diluted with murine typhus, epidemic typhus, or normal yolk sac antigens and applied to fixed microdroplets of murine or epidemic typhus rickettsia; anti-Ig was used as con- jugate. Complement-fixation tests were later carried out with murine typhus, epidemic typhus, Rickettsia canada, Rocky Mountain spotted fever, Q fever, and control antigen. Clinical background Demographic information and clinical features of the illness were obtained from hospital records for the first 104 cases of typhus. Information was sought regarding age, sex, occupation, and home address of the patient, manner of onset of symptoms, occur- rence of fever, headache, myalgia, arthralgia, cough, rash (distribution and type), hypotension (defined as a pressure of less than 100/60 mm Hg), lethargy, delirium, conjunctival suffusion, uraemia (urea level > 5 mmol/litre), and the level of the leukocyte count. Patients with typhus were identified when they sought medical attention at the neighbourhood clinic. Subclinical illness was not evaluated. Medical care is free in Kuwait, and health workers are very well accepted by the people. Case - control study A case - control study was carried out to try to iden- tify risk factors for typhus and sources of infection. The study was conducted using a questionnaire designed to assess exposure to rats, by determining sleeping habits, frequency of rat bites, occupational exposure to rats, duration of occupancy of house, frequency of rubbish collection, and use of rat poison. A sanitarian also visited each house to investi- gate the type of house, presence and type of livestock in the household, the proximity of animal pens to the living quarters, the number of rat burrows, the amount of rat droppings, type of floor, presence of food on floor, manner of food storage, methods of rubbish disposal, the distance from the house to the rubbish tip, presence of rat harbourages, and method of sewage disposal. Controls from the same neighbourhood were iden- tified in a systematic fashion, and matched by sex and age. Altogether, 110 control subjects were selected, who had no history of fever or headache during the previous 3 months. Venous blood was obtained from 87 controls and assayed for complement fixation titre to eliminate those with a previous mild typhus illness or immunity due to an earlier infection. The antigens used were the R. prowazeki comple- ment-fixation (CF) antigen,a the typhus group CF bantigen, and Proteus OX antigen.' Positive human serum was used as a control.b The test itself was carried out by the standard microtitration technique, in U-shaped wells as described by Grist et al. (5). In the Weil-Felix test, both slide and tube methods were used, according to the manufacturer's recommendations. Rats and ectoparasites Live rats were collected by fitting cage traps against the entrances to the burrows, and passing the exhaust from a petrol engine into the burrows to expel the rats. Every cage was then placed in a polyethylene sack and transferred immediately to the laboratory. The rats were immobilized with ether and combed thoroughly. The ectoparasites were collected in a deep, glass bowl along with those found in the cage and the sack. A dissecting microscope was used to count and identify the ectoparasites recovered,d which were then preserved in 700 ml/litre alcohol in a labelled container. RESULTS Confirmation of diagnosis The IFA tests showed that the 13 control patients had titres against murine typhus antigen of 64 or less (10 had a titre of 16, 2 a titre of 32, and I a titre of 64). From the State Institute of Hygiene, Warsaw, Poland. From the Centers for Disease Control, Atlanta, GA, USA. From Difco Laboratories, USA. d Taxonomic results were confirmed by Bernice Brewster, Department of Zoology, British Museum (Natural History), London, England. MURINE TYPHUS IN KUWAIT One of the control patients had a titre of 128 against spotted fever rickettsiae on the 12th and 25th day after onset of his disease, indicating a possible pre- vious infection with R. conori. All 11 presumed typhus patients had maximum murine typhus titres of at least 256. The anti-IgM conjugate gave equal or brighter staining with all positive typhus sera. There were 3 seroconversions (greater than 4-fold increase in titre between 2 specimens); 5 serum pairs did not show such an increase in titre, but in each case the first specimen had been collected 14 days or more after the onset of clinical disease. These results indicate primary infection with typhus fever rickettsiae. The murine typhus titre was 4 times higher than the epi- demic titre for 3 patients, while for the other patients the murine titre was less than twice the epidemic titre; these results indicate that the infection was caused by murine typhus rickettsiae. Antibody-inhibition tests showed murine-specific antibody in 5 cases, epidemic-specific antibody in 1 case, and uncertain differentiation in 5 cases. The patient with an epidemic-specific response had no recollection of immunization with epidemic typhus vaccine (4, 5), and may have had a previous infection with epidemic typhus. The proportion of serum samples with a doubtful result is higher than that observed previously in the CDC laboratory, where the antibody-inhibition method is in regular use for the differentiation of murine typhus from epidemic typhus of presumed flying squirrel origin in the USA (7). Nevertheless, on the whole, the IFA results indi- cated that the infection was murine typhus fever. The complement-fixation test showed antibodies to the typhus grpup in the expected titre range, with no significant difference between epidemic and murine typhus titres. Antibodies for spotted fever and Q fever were not demonstrated in the control serum samples. Clinical features and the case- control study The outbreak resulted in 254 cases of murine typhus over 14 weeks (16 April - 23 July 1978) (Fig. 1), none of which were fatal. All cases were confirmed in the laboratory by the Weil-Felix OX 19 test. Most of the clinically evident cases occurred in the lower socio- economic classes of the population, in persons aged between 15 and 44 years (Table 1); 160 cases occurred in men. A review of the first 104 cases found that in this group, there were 31 housewives (307o), 28 labourers (2707o), 19 children and students (18%IN), and 26 soldiers, police, drivers, servants, and clerical personnel (25%). The clinical features of the illness in those who sought medical attention are shown in Table 2. The onset of the outbreak coincided with a period of rat and flea proliferation. The case-control study 33 31 29 27 25 23 21 - 19 CU co 17 0 z 15 13 11 9 7 5 3 1 3 5 7 9 11 13 Week of onset Fig. 1. Incidence of typhus cases in Kuwait during summer 1978. Week 1 began on April 16. was designed to identify the factors predisposing to exposure to rat and other possible reservoirs of infec- tion. However, no significant difference was found between case and control households in data obtained from the questionnaire or in the results of the environmental inspection. A total of 110 controls were interviewed and investigated in the last week of June. Serum samples were obtained from 25 female and 62 male controls. The serum samples showed that 17 (2007o) had a CF titre . 1:4 (7 females, 10 males), while the remaining 70 (8007o) were seronegative. Rats and ectoparasites In 1976, studies were started to determine the magnitude of the rat population, and its susceptibility 285 Vt ao0 1 A. R. AL-AWADI ET AL. Table 1. Distribution of typhus cases by age group and socioeconomic class Age group (years) Socioeconomic class 0-4 5- 14 1 5-24 25-44 ) 45 Total Attack rate per 100 000 1 0 1 1 5 2 9 5.6 2 0 11 38 54 1 104 12.7 3 3 21 49 59 9 141 66.6 Total 3 33 88 118 12 254 21.2 Age-specific attack rate per 100000 1.4 11.6 41.6 34.8 10.6 to poison and to collect and identify their ectoparasites. The most prevalent commensal rodent in Kuwait is Rattus norvegicus. The study of the ecto- parasites revealed Xenopsylla astia fleas, Polyplax spinulosa and P. cenomudis lice, and Lelaps nutalli and Ornythonyssus bacoti mites. Rats infest about 80OVo of houses in the lower socioeconomic grades and 20- 30Wo of newly constructed houses (T. Zaghloul, unpublished data). Table 2. Clinical features of typhus in patients seeking medical attention Clinical feature No. positive/No. examined % Symptoms Gradual onset 38 / 65 58 Sudden onset 27 /65 42 Fever 104 / 104 100 Headache 97 / 101 96 Myalgia 64 / 88 73 Cough 46 / 96 48 Arthralgia 15/ 78 19 Signs Temperature > 390C 74 / 98 76 Conjunctival suffusion 65 / 98 66 Rash on trunk only 49 / 101 49 Generalized rash 6 / 101 6 Hypotension 14 / 82 17 Petechiae 20 / 73 27 Lethargy 8 / 93 9 Delirium 7 / 93 8 Laboratory tests Urea > 5mmol/litre 16 / 81 20 Leukocyte count < 5x109/litre 25/102 25 > 10 x 109/litre 9 / 102 9 The number of rats and fleas begins to increase during April, reaches a peak in May, and decreases again by September (Fig. 2); the outbreak of murine typhus in Kuwait in 1978 coincided with this seasonal increase. When murine typhus was diagnosed, intensive control measures were directed toward both ectopara- sites and rodents. Burrows were dusted routinely with a mixture of 100/o DDT and 2Glo lindane in talc one day before fumigation with carbon monoxide. It was important to reduce the flea population before killing the rats, or the fleas would have migrated to human hosts, which may have led to an increase in disease. The people in the lower socioeconomic grades were taught how to eliminate rat harbourages, housing projects were completed, and some of the population were moved to more modern housing. It was estimated that more than 25 000 burrows were treated between June and September 1978. The number of live rats per burrow during May was 1 and dropped to 0.2 during September 1978, while the number of fleas per live rat was 11.4 in April, 30.5 in May, and then dropped sharply to reach 0.1 by September 1978. The reduced flea density (Fig. 2) in August 1978 may have resulted from the increased efforts in flea control, although the hot and humid conditions pre- vailing in August are known to cause a marked reduc- tion in the population of X. astia (T. Zaghloul, unpublished data). In 1979, the average number of fleas per rat was sharply reduced from 1978 values to 3.6 in April and 8.7 in May. There were 151 cases of murine typhus in 1979. The greatest reduction in the number of typhus cases was found in the areas where flea and rat control were most intensive. The timing of the outbreaks in 1978 and 1979 was similar to that seen in Egypt where murine typhus was found to occur from April to June (8). 286 MURINE TYPHUS IN KUWAIT 460 J FMAMJ J AS ON DJ F MA MJ J AS OND J FM A MJ J ASO ND J F MA MJ J AS ON D 1977 1978 1979 1980 Fig. 2. Distribution of cases of typhus in Kuwait in comparison with the size of the rat and flea population. For 1980, the total number of rats is plotted, rather than the number of rats per burrow. 8 Z 1.; , 1.: 0 Z 0.l el co w o z 12 8 4 0 120 110 100 90 80 70 60 50 40 30 20 10 0 287 288 A. R. AL-AWADI ET AL. DISCUSSION Exposure to rats and fleas probably occurs mainly in the home since housewives, older children, and students made up almost 501o of the first 104 cases of typhus. The apparent low incidence of the disease in young children is not attributable to underdiagnosis because medical care is free and easily obtained in Kuwait. However, murine typhus is known to be mild in the very young, so that infection in this age group is probably much more common than the 3 patients in this study would suggest. We were unable to demonstrate any significant difference beween case and control households by the questionnaire or by home inspection. These negative findings could have resulted from prior infection and subsequent immunity in the control subjects. Although 8007o of the 87 controls who provided serum samples were susceptible, this sample of controls may have been biased. It is likely that there are factors that characterize those who become infected as distinct from those who do not; we suspect, however, that murine typhus has been hyperendemic in Kuwait, and that these distinctions are difficult to show quanti- tatively. In conclusion, it is emphasized that murine typhus may be much more abundant in many areas than is generally realized. Accurate diagnosis, substantiated by suitable laboratory tests, is essential. This infec- tion is likely to persist wherever the Rattus population remains high, and therefore it is very important to carry out serological surveys for rickettsioses in patients with acute febrile disease. ACKNOWLEDGEMENTS We thank Dr Charles Wisseman and Dr Robert Traub of the Department of Microbiology, University of Maryland School of Medicine, Dr Philip Brachman, Director, Bureau of Epidemiology, CDC, and Dr J. Lyli Conrad, Director, Field Services Division, Bureau of Epidemiology, CDC for their recommendations and editorial assistance. We also acknowledge the participation of Mr Kamal El-Saleh, Head of Vital and Health Statistical Division, Kuwait, in analysing the case-control study, and medical staff in the State of Kuwait who carried out the surveillance and control measures. We also thank Dr Alexander D. Langmuir for his constructive and kind advice and help in establishing the concept of epidemiology among the workers in the State of Kuwait. RESUME TYPHUS MURIN AU KOWEIT EN 1978 Une poussee de typhus murin au cours de laquelle ont et denombres 254 cas dont aucun mortel s'est produite au Koweit entre les mois d'avril et aout 1978. La majorite des malades avait entre 15 et 44 ans et appartenait aux classes socio-economiques les plus defavorisees; 160 etaient de sexe masculin. Cette poussee a coincide avec une periode de forte densite de rats et de puces dans la ville. Le diagnostic de typhus a et confirme par la reaction de Weil-Felix et l'epreuve des anticorps fluorescents, 1'epreuve d'inhibition des anticorps ayant servi a operer une distinction entre le typhus epidemique et le typhus murin. Une etude cas-temoins a et entreprise dans le but de determiner les facteurs pouvant predisposer a l'infection mais aucune difference significative n'a et observee entre les menages des cas et ceux des temoins du point de vue des conditions environnementales ou du niveau d'exposition aux rats. Lorsque le diagnostic de typhus murin a et confirme, des mesures intensives de lutte ont e instaurees contre les rats et les ectoparasites. Tous les trous de rats ont e systema- tiquement traites a l'aide d'une poudre de talc contenant 10% de DDT et 2% de lindane, un jour avant des fumiga- tions au monoxyde de carbone. On est ainsi parvenu a reduire sensiblement le nombre de puces par rat entre 1978 et 1979 de meme que l'incidence du typhus. Le declin de l'inci- dence a et particulierement marque dans les zones oii les operations de lutte contre les puces et les rats avaient e les plus intensives. 11 est a noter que le typhus murin peut etre beaucoup plus repandu qu'on ne le pense generalement. L'infection peut persister partout ou la population de Rattus est importante et il faut donc prevoir une recherche serologique des rickett- sioses chez les sujets presentant des infections febriles aigues. REFERENCES 1. TRAUB, R. ET AL. The ecology of murine typhus-a critical review. Tropical diseases bulletin, 75: 237 - 317 (1978). 2. BURGDORFER, W. In: Hunter, G. W. et al, ed. Tropical medicine, 5th ed., Philadelphia, Saunders, 1976, pp. 110-111. MURINE TYPHUS IN KUWAIT 289 3. NEWHOUSE, V. F. ET AL. A comparison of complement fixation, indirect fluorescent antibody, and micro- agglutination tests for the serologic diagnosis of rickettsial disease. American journal of tropical med- icine and hygiene, 28: 387 - 395 (1979). 4. GOLDWASSER, R. A. & SHEPARD, C. C. Fluorescent antibody methods in the differentiation of murine and epidemic typhus sera, and specificity changes resulting from previous immunization. Journal of immunology, 82: 373 - 380 (1959) 5. GRIST, N. R. ET AL. Diagnostic methods in clinical vir- ology, 2nd ed., Oxford, Blackwell Scientific Publi- cations, 1974, p. 80. 6. GOLDWASSER, R. A. ET AL. The specificity of antibody response in typhus fever. Its alteration during murine typhus infection as a result of previous exposure to epi- demic typhus antigen. Journal of immunology, 83: 491 (1959). 7. McDADE, J. E. ET AL. Evidence of Rickettsia prowaceki infections in the United States. American journal of tropical medicine and hygiene, 29: 272 - 284 (1980). 8. IMAN, 1. Z. E. ET AL. Murine typhus in Egypt, U.A.R. Journal of the Egyptian Public Health Association, 37: 179- 186 (1962).