Journal of the Royal Society of Medicine Volume 71 July 1978
507
Recent studies in scrub typhus: a review' Major G W Brown MRCP DTM&H Royal Army Medical College, Millbank, London SWI
Introduction Scrub typhus is a febrile illness, endemic in much of the roughly triangular area bounded by Japan, Pakistan and Australia. The causative organism, Rickettsia tsutsugamushi, is transmitted by the larvae (or chiggers) of several species of trombiculid mites. The name was coined to stress the association of the disease with scrub, or wasteland, the habitat in which the vector mites and their rodent hosts are frequently found. The Japanese synonym, tsutsugamushi disease, clearly had precedence, but the name, scrub typhus, became fully established through widespread use in World War II, when tens of thousands of cases occurred in European and Asian soldiers. The disease continued to be of military significance during the Malayan Emergency (McCrumb et al. 1957) and the Vietnam War (Berman & Kundin 1973), and still troubles Taiwanese soldiers on the Pescadores (Fang et al. 1975). Clinical features
The 'text-book' presentation, with a dusky red maculopapular rash and an eschar marking the site of the infecting chigger bite, is easily recognized. Many workers, however, have emphasized the infrequency of these features (Lewthwaite & Savoor 1940, Brown et al. 1976). If present, the eschar and rash are easily recognized on light skins, but are less noticeable on darker pigmented patients. Eschars may also be mistaken for the ulcers seen in other conditions. Common clinical features, such as headache and cough do not distinguish scrub typhus from other infections. Studies during the Vietnam War (Berman et al. 1973) demonstrated that the diagnosis was often missed, many cases being categorized as 'pyrexia of unknown origin (PUO)'. The spectrum of illness is wide, from severe and prostrating to mild and, probably, subclinical (Shishido 1962). Laboratory diagnosis Isolation of the organism is usually achieved by intraperitoneal mouse inoculation of a fresh specimen of a patient's blood (Elisberg & Bozeman 1969). Infected mice usually become sick 810 days after inoculation and organisms may be demonstrated in peritoneal mononuclear cells, by immunofluorescence. Usually, however, tissues from the first set of mice are inoculated into further mice which, after a period on chloramphenicol, are challenged with a lethal dose of R. tsutsugamushi. Survival of this challenge, over six weeks after the original isolation attempt, is positive evidence of infection. This process is obviously of no use in the immediate management of the patient. Recently, it has become possible to demonstrate (by immunofluorescence) the organisms growing in monocytes cultured from the peripheral blood of infected subjects, a week after 1 Paper read to United Services Section, 30 June 1977. Requests for reprints to: Commander, US Army Medical Research Unit, Institute for Medical Research, Jalan Pahang, Kuala Lumpur, Malaysia
508
Journal of the Royal Society of Medicine Volume 71 July 1978
their being harvested (D L Huxsoll, personal communication). This technique, although relatively simple will, like mouse inoculation, remain beyond the capability of a routine laboratory. A serological diagnosis can be made by the demonstration of a significant rise in antibody titre, either by indirect immunofluorescence (IFA) or complement fixation (CF). The IFA test, introduced by Bozeman & Elisberg (1963) has recently been modified by Robinson, Brown et al. (1976) to enable simultaneous titration of a specimen against 9 antigenic strains of the organism. The IFA test can also be used on specimens of capillary blood collected on blotting paper (Gan et al. 1972). As with isolation procedures, however, few laboratories maintain the capability to do these specific serological tests. An inherent disadvantage, too, of a diagnostic test which relies on the demonstration of antibody, is the lag time for a positive result. The low titres occurring in the first four to five days of illness are common in normal subjects from
Fgr
1i. O
p ashowin p a
l
i
Figure 1. Oil-palm plantation showing leaf-litter
Journal of the Royal Society of Medicine Volume 71 July 1978
509
endemic areas (Robinson, Gan & Donaldson 1976), whilst titres high enough to be diagnostic are seldom obtained soon enough to influence management. The Weil-Felix reaction, which is the only generally available diagnostic test, failed to detect over 50% of proven cases in several series (Doherty 1956, Berman & Kundin 1973). Being a nonspecific test it may also be positive (and show a significant rise in titre) in leptospirosis, relapsing fever and other infections. Further work is needed to develop methods for rapid diagnosis, such as detection of antigen in blood or urine. Epidemiology Serological surveys of groups of rural Malaysians have shown a prevalence of antibody to R. tsutsugamushi of up to 64% (Cadigan et al. 1972). In a recent study of aborigines, we observed an attack rate, based on serological changes, of 3.9% per month (Brown et al. 1978). The type of shifting cultivation practised by most of these aborigines, and many rural Malays, produces suitable ecological conditions for both mites and rodents, and similar conditions exist around villages in much of South East Asia. Large areas of Malaysian rain forest are being cleared and the land is being brought under cultivation with oil-palm and rubber. Oil-palm fruit provides a superb food supply for rodents, and the leaf litter a good chigger microhabitat (Figure 1). Perhaps unsurprisingly then, an ongoing study of febrile patients from rural areas is showing scrub typhus to be a very common cause of illness (Brown et al. 1976). This is particularly so amongst the tens of thousands of newly settled oil-palm labourers and their families, and it may well be that they all become infected within the first few years of arrival. Prospective studies are in progress to determine this. It is interesting that in 1927 Fletcher & Field recognized the first Malayan cases in three plantation managers who had been working in an area of overgrown oil-palm. A rapid means of estimating the prevalence of infection in a specific area is now available: unengorged chiggers are collected on black plates and their internal tissues examined for R. tsutsugamushi by immunofluorescence (Dohany, Shirai et al.). There is a growing body of evidence for the mite being both the vector and reservoir which increases the validity of this survey method. Naturally infected mites, reared in the laboratory, have now transmitted the infection through greater than twenty generations, whilst uninfected chiggers, if fed experimentally on infected mice, take up R. tsutsugamushi, but fail to transmit the infection transovarially to the next generation (Walker et al. 1975). Treatment Following the use of a single, oral dose of a long-acting tetracycline, doxycycline, in the treatment of epidemic typhus in Africa (Krause et al. 1975), this regime has now been tested in scrub typhus. A single dose of doxycycline is as effective as the standard 7 day course of tetracycline (Brown, Saunders et al. 1978). This should soon lead to its use in field conditions. Prophylaxis Smadel et al. (1951) inoculated volunteers with a high egg-passaged Gilliam strain of R. tsutsugamushi, followed by a suppressive course of chloramphenicol. They also tried a number of chemosuppressive regimes and showed that 4 g of chloramphenicol at weekly intervals during, and for four weeks after exposure, prevented significant illness, though the subjects did become infected (Smadel, Traub et al. 1950). A more acceptable, and probably more efficient, regime would use doxycycline, at intervals of perhaps a week, and could be justified in a high risk area. A further attraction of chemoprophylaxis is that it would almost certainly reduce the incidence of another disease of great military and civilian importance in the area, leptospirosis. However, the opportunity to test such a regime has not yet occurred. One difficulty in producing a vaccine is due to the existence of multiple serotypes of R. tsutsugamushi within a small area (Miesse et al. 1950). In addition to the three prototype strains, Karp, Kato and Gilliam, Elisberg et al. (1967) distinguished at least five antigenically
510
Journal of the Royal Society of Medicine Volume 71 July 1978
separate serotypes in Thailand, for example. Whilst most of the several hundred human isolates recently obtained in Malaysia seem related to the Karp strain, some are apparently new (D L Huxsoll, personal communication). As Smadel, Ley et al. (1950) demonstrated the transient nature, in man, of the cross-immunity between strains, this means that unless it can be shown that one or two serotypes are mainly responsible for serious illness - and this seems uncertain - a practicable vaccine is still not in sight. As an adjunct to other physical methods of vector control, systemic acaracides, such as dimethoate, are useful over small areas, such as around a camp-site. The chemical is mixed with bait, eaten by rodents, and kills chiggers feeding on them, thus slowly reducing the mite population (Dohany, Cromroy & Cole 1977). Comment Clearly, the number of cases of scrub typhus notified to the Malaysian health authorities (average 55 per annum from 1967-74) understates the magnitude of the problem. Whilst it is likely that there is under-reporting of diagnosed cases, most of the discrepancy must be due to lack of recognition. It is apparent from both the prevalence of antibody and the incidence of disease, that the infection is of considerable socioeconomic importance in Malaysia and, probably, other Asian countries.
References Berman S J, Irving G S, Kundin W D, Gunning J J & Watten R H (1973) American Journal of Tropical Medicine and Hygiene 22, 796-801 Berman S J & Kundin W D (1973) Annals of Internal Medicine 79, 26-30 Bozeman F M & Elisberg B L (1963) Proceedings of the Society of Experimental Biology and Medicine 112, 568-573 Brown G W, Robinson D M & Huxsoll D L (1978) American Journal of Tropical Medicine and Hygiene 27, 121-124 Brown G W, Robinson D M, Huxsoll D L, Ng T S, Lim K J & Sannasey G (1976) Transactions of the Royal Society of Tropical Medicine and Hygiene 70, 444 448 Brown G W, Saunders J P, Singh S, Huxsoll D L & Shirai A (1978) Transactions of the Royal Society of Tropical Medicine & Hygiene 72, (in press) Cadigan F C, Andre R G, Bolton M, Gan E & Walker J S (1972) Transactions of the Royal Society of Tropical Medicine & Hygiene 66, 582-587 Dohany A L, Cromroy H L & Cole M M (1977) Journal of Medical Entomology 14, 79-81 Dohany A L, Shirai A, Robinson D M, Ram S & Huxsoll D L Journal of Medical Entomology (in press) Doherty R L (1956) Medical Journal of Australia 2, 212-219 Elisberg B L & Bozeman F M (1969) In: Diagnostic Procedures for Viral and Rickettsial Infections. American Public Health Association, New York; pp 849-851 Elisberg B L, Sangkasuvana V, Campbell J M, Bozeman F M, Bodhidatta P Rapmund G (1967) Acta medica et biologica (Niigata) 15, pp 61-67 Fang R C Y, Lin W P, Chao P S, Kuo N T & Chen C M (1975) Tropical and Geographical Medicine 27, 143-150 Fletcher W & Field J W (1927) Bulletinsfrom The Institutefor Medical Research, Federated Malay States, No 1 Gan E, Cadigan F C & Walker J S (1972) Transactions of The Royal Society of Tropical Medicine and Hygiene 66, 588593 Krause D W, Perine P L, McDade J E & Awoke S (1975) East African Medical Journal 52, 421-427 Lewthwaite R & Savoor S R (1940) Lancet i, 255-259, 305-311 McCrumb F R, Stockard J L, Robinson C R, Turner L H, Levis D G, Maisey C W, Kelleher M F, Gleiser C A & Smadel J E (1957) American Journal of Tropical Medicine and Hygiene 6, 238-256 Miesse M, Diercks F H & Danauskaus J (1950) Bacteriological Proceedings pp 90-91 Robinson D M, Brown G, Gan E & Huxsoll DL (1976) The American Journal of Tropical Medicine and Hygiene 25,900905 Robinson D M, Gan E & Donaldson R (1976) Tropical and Geographical Medicine 28, 303-308 Shishido A (1962) Japanese Journal of Medical Science and Biology 15, 330-336 Smadel J E, Ley H L, Diercks F H & Traub R (1950) Archives of Pathology 50, 847-861 Smadel J E, Ley H L, Diercks F H, Traub R, Tipton V J & Frick L P (1951) American Journal of Hygiene 53, 317-325 Smadel J E, Traub R, Frick L P, Diercks F H & Bailey C A (1950) American Journal of Hygiene 51, 216-228 Walker J S, Chan C T, Manikumaran C & Elisberg B L (1975) Annals of the New York Academy of Sciences 266, 80-90
507
Recent studies in scrub typhus: a review' Major G W Brown MRCP DTM&H Royal Army Medical College, Millbank, London SWI
Introduction Scrub typhus is a febrile illness, endemic in much of the roughly triangular area bounded by Japan, Pakistan and Australia. The causative organism, Rickettsia tsutsugamushi, is transmitted by the larvae (or chiggers) of several species of trombiculid mites. The name was coined to stress the association of the disease with scrub, or wasteland, the habitat in which the vector mites and their rodent hosts are frequently found. The Japanese synonym, tsutsugamushi disease, clearly had precedence, but the name, scrub typhus, became fully established through widespread use in World War II, when tens of thousands of cases occurred in European and Asian soldiers. The disease continued to be of military significance during the Malayan Emergency (McCrumb et al. 1957) and the Vietnam War (Berman & Kundin 1973), and still troubles Taiwanese soldiers on the Pescadores (Fang et al. 1975). Clinical features
The 'text-book' presentation, with a dusky red maculopapular rash and an eschar marking the site of the infecting chigger bite, is easily recognized. Many workers, however, have emphasized the infrequency of these features (Lewthwaite & Savoor 1940, Brown et al. 1976). If present, the eschar and rash are easily recognized on light skins, but are less noticeable on darker pigmented patients. Eschars may also be mistaken for the ulcers seen in other conditions. Common clinical features, such as headache and cough do not distinguish scrub typhus from other infections. Studies during the Vietnam War (Berman et al. 1973) demonstrated that the diagnosis was often missed, many cases being categorized as 'pyrexia of unknown origin (PUO)'. The spectrum of illness is wide, from severe and prostrating to mild and, probably, subclinical (Shishido 1962). Laboratory diagnosis Isolation of the organism is usually achieved by intraperitoneal mouse inoculation of a fresh specimen of a patient's blood (Elisberg & Bozeman 1969). Infected mice usually become sick 810 days after inoculation and organisms may be demonstrated in peritoneal mononuclear cells, by immunofluorescence. Usually, however, tissues from the first set of mice are inoculated into further mice which, after a period on chloramphenicol, are challenged with a lethal dose of R. tsutsugamushi. Survival of this challenge, over six weeks after the original isolation attempt, is positive evidence of infection. This process is obviously of no use in the immediate management of the patient. Recently, it has become possible to demonstrate (by immunofluorescence) the organisms growing in monocytes cultured from the peripheral blood of infected subjects, a week after 1 Paper read to United Services Section, 30 June 1977. Requests for reprints to: Commander, US Army Medical Research Unit, Institute for Medical Research, Jalan Pahang, Kuala Lumpur, Malaysia
508
Journal of the Royal Society of Medicine Volume 71 July 1978
their being harvested (D L Huxsoll, personal communication). This technique, although relatively simple will, like mouse inoculation, remain beyond the capability of a routine laboratory. A serological diagnosis can be made by the demonstration of a significant rise in antibody titre, either by indirect immunofluorescence (IFA) or complement fixation (CF). The IFA test, introduced by Bozeman & Elisberg (1963) has recently been modified by Robinson, Brown et al. (1976) to enable simultaneous titration of a specimen against 9 antigenic strains of the organism. The IFA test can also be used on specimens of capillary blood collected on blotting paper (Gan et al. 1972). As with isolation procedures, however, few laboratories maintain the capability to do these specific serological tests. An inherent disadvantage, too, of a diagnostic test which relies on the demonstration of antibody, is the lag time for a positive result. The low titres occurring in the first four to five days of illness are common in normal subjects from
Fgr
1i. O
p ashowin p a
l
i
Figure 1. Oil-palm plantation showing leaf-litter
Journal of the Royal Society of Medicine Volume 71 July 1978
509
endemic areas (Robinson, Gan & Donaldson 1976), whilst titres high enough to be diagnostic are seldom obtained soon enough to influence management. The Weil-Felix reaction, which is the only generally available diagnostic test, failed to detect over 50% of proven cases in several series (Doherty 1956, Berman & Kundin 1973). Being a nonspecific test it may also be positive (and show a significant rise in titre) in leptospirosis, relapsing fever and other infections. Further work is needed to develop methods for rapid diagnosis, such as detection of antigen in blood or urine. Epidemiology Serological surveys of groups of rural Malaysians have shown a prevalence of antibody to R. tsutsugamushi of up to 64% (Cadigan et al. 1972). In a recent study of aborigines, we observed an attack rate, based on serological changes, of 3.9% per month (Brown et al. 1978). The type of shifting cultivation practised by most of these aborigines, and many rural Malays, produces suitable ecological conditions for both mites and rodents, and similar conditions exist around villages in much of South East Asia. Large areas of Malaysian rain forest are being cleared and the land is being brought under cultivation with oil-palm and rubber. Oil-palm fruit provides a superb food supply for rodents, and the leaf litter a good chigger microhabitat (Figure 1). Perhaps unsurprisingly then, an ongoing study of febrile patients from rural areas is showing scrub typhus to be a very common cause of illness (Brown et al. 1976). This is particularly so amongst the tens of thousands of newly settled oil-palm labourers and their families, and it may well be that they all become infected within the first few years of arrival. Prospective studies are in progress to determine this. It is interesting that in 1927 Fletcher & Field recognized the first Malayan cases in three plantation managers who had been working in an area of overgrown oil-palm. A rapid means of estimating the prevalence of infection in a specific area is now available: unengorged chiggers are collected on black plates and their internal tissues examined for R. tsutsugamushi by immunofluorescence (Dohany, Shirai et al.). There is a growing body of evidence for the mite being both the vector and reservoir which increases the validity of this survey method. Naturally infected mites, reared in the laboratory, have now transmitted the infection through greater than twenty generations, whilst uninfected chiggers, if fed experimentally on infected mice, take up R. tsutsugamushi, but fail to transmit the infection transovarially to the next generation (Walker et al. 1975). Treatment Following the use of a single, oral dose of a long-acting tetracycline, doxycycline, in the treatment of epidemic typhus in Africa (Krause et al. 1975), this regime has now been tested in scrub typhus. A single dose of doxycycline is as effective as the standard 7 day course of tetracycline (Brown, Saunders et al. 1978). This should soon lead to its use in field conditions. Prophylaxis Smadel et al. (1951) inoculated volunteers with a high egg-passaged Gilliam strain of R. tsutsugamushi, followed by a suppressive course of chloramphenicol. They also tried a number of chemosuppressive regimes and showed that 4 g of chloramphenicol at weekly intervals during, and for four weeks after exposure, prevented significant illness, though the subjects did become infected (Smadel, Traub et al. 1950). A more acceptable, and probably more efficient, regime would use doxycycline, at intervals of perhaps a week, and could be justified in a high risk area. A further attraction of chemoprophylaxis is that it would almost certainly reduce the incidence of another disease of great military and civilian importance in the area, leptospirosis. However, the opportunity to test such a regime has not yet occurred. One difficulty in producing a vaccine is due to the existence of multiple serotypes of R. tsutsugamushi within a small area (Miesse et al. 1950). In addition to the three prototype strains, Karp, Kato and Gilliam, Elisberg et al. (1967) distinguished at least five antigenically
510
Journal of the Royal Society of Medicine Volume 71 July 1978
separate serotypes in Thailand, for example. Whilst most of the several hundred human isolates recently obtained in Malaysia seem related to the Karp strain, some are apparently new (D L Huxsoll, personal communication). As Smadel, Ley et al. (1950) demonstrated the transient nature, in man, of the cross-immunity between strains, this means that unless it can be shown that one or two serotypes are mainly responsible for serious illness - and this seems uncertain - a practicable vaccine is still not in sight. As an adjunct to other physical methods of vector control, systemic acaracides, such as dimethoate, are useful over small areas, such as around a camp-site. The chemical is mixed with bait, eaten by rodents, and kills chiggers feeding on them, thus slowly reducing the mite population (Dohany, Cromroy & Cole 1977). Comment Clearly, the number of cases of scrub typhus notified to the Malaysian health authorities (average 55 per annum from 1967-74) understates the magnitude of the problem. Whilst it is likely that there is under-reporting of diagnosed cases, most of the discrepancy must be due to lack of recognition. It is apparent from both the prevalence of antibody and the incidence of disease, that the infection is of considerable socioeconomic importance in Malaysia and, probably, other Asian countries.
References Berman S J, Irving G S, Kundin W D, Gunning J J & Watten R H (1973) American Journal of Tropical Medicine and Hygiene 22, 796-801 Berman S J & Kundin W D (1973) Annals of Internal Medicine 79, 26-30 Bozeman F M & Elisberg B L (1963) Proceedings of the Society of Experimental Biology and Medicine 112, 568-573 Brown G W, Robinson D M & Huxsoll D L (1978) American Journal of Tropical Medicine and Hygiene 27, 121-124 Brown G W, Robinson D M, Huxsoll D L, Ng T S, Lim K J & Sannasey G (1976) Transactions of the Royal Society of Tropical Medicine and Hygiene 70, 444 448 Brown G W, Saunders J P, Singh S, Huxsoll D L & Shirai A (1978) Transactions of the Royal Society of Tropical Medicine & Hygiene 72, (in press) Cadigan F C, Andre R G, Bolton M, Gan E & Walker J S (1972) Transactions of the Royal Society of Tropical Medicine & Hygiene 66, 582-587 Dohany A L, Cromroy H L & Cole M M (1977) Journal of Medical Entomology 14, 79-81 Dohany A L, Shirai A, Robinson D M, Ram S & Huxsoll D L Journal of Medical Entomology (in press) Doherty R L (1956) Medical Journal of Australia 2, 212-219 Elisberg B L & Bozeman F M (1969) In: Diagnostic Procedures for Viral and Rickettsial Infections. American Public Health Association, New York; pp 849-851 Elisberg B L, Sangkasuvana V, Campbell J M, Bozeman F M, Bodhidatta P Rapmund G (1967) Acta medica et biologica (Niigata) 15, pp 61-67 Fang R C Y, Lin W P, Chao P S, Kuo N T & Chen C M (1975) Tropical and Geographical Medicine 27, 143-150 Fletcher W & Field J W (1927) Bulletinsfrom The Institutefor Medical Research, Federated Malay States, No 1 Gan E, Cadigan F C & Walker J S (1972) Transactions of The Royal Society of Tropical Medicine and Hygiene 66, 588593 Krause D W, Perine P L, McDade J E & Awoke S (1975) East African Medical Journal 52, 421-427 Lewthwaite R & Savoor S R (1940) Lancet i, 255-259, 305-311 McCrumb F R, Stockard J L, Robinson C R, Turner L H, Levis D G, Maisey C W, Kelleher M F, Gleiser C A & Smadel J E (1957) American Journal of Tropical Medicine and Hygiene 6, 238-256 Miesse M, Diercks F H & Danauskaus J (1950) Bacteriological Proceedings pp 90-91 Robinson D M, Brown G, Gan E & Huxsoll DL (1976) The American Journal of Tropical Medicine and Hygiene 25,900905 Robinson D M, Gan E & Donaldson R (1976) Tropical and Geographical Medicine 28, 303-308 Shishido A (1962) Japanese Journal of Medical Science and Biology 15, 330-336 Smadel J E, Ley H L, Diercks F H & Traub R (1950) Archives of Pathology 50, 847-861 Smadel J E, Ley H L, Diercks F H, Traub R, Tipton V J & Frick L P (1951) American Journal of Hygiene 53, 317-325 Smadel J E, Traub R, Frick L P, Diercks F H & Bailey C A (1950) American Journal of Hygiene 51, 216-228 Walker J S, Chan C T, Manikumaran C & Elisberg B L (1975) Annals of the New York Academy of Sciences 266, 80-90
