13 Scrub typhus

Bacteria of the genus Rickettsia are obligate intracellular rods that retain basic fuchsin when stained by the method of Gimenez. This genus has long been used as a generic term of small intracellular bacteria. However, taxonomic progress made over the last years has deeply modified the definition of ‘rickettsia’. As a result, in 1995 the position of R. tsutsugamushi has been reclassified from the genus Rickettsia into a separate new genus, Orientia (Tamura et al. 1995).

Scrub typhus, also known as ‘tsutsugamushi fever’, occurs only in Asia and is a chigger-borne zoonoses. The disease is acute, febrile, potentially fatal and has been known for centuries in China where it was probably described as early as the fourth century BC (Parola and Raoult 2006). In recent years this infection has been re-emerging because of descriptions of strains of O. tsutsugamushi with reduced susceptibility to antibiotics and of the surprising interactions between scrub typhus and the human immunodeficiency virus (HIV). It is estimated that more than a million cases of scrub typhus are transmitted annually in Asia and more than a billion people are at risk (Rosenberg 1997).

The presence of tsutsugamushi disease and its association with chigger bites has been recognized by some of the native populations of Japan and China for centuries. The term ‘akamushi’, the origin of the Japanese name for this rickettsiosis, means ‘red chigger’. Rural residents of these countries often knew that the best way to avoid being infected was to avoid areas infested by these arthropods (Weiss 1981; Walker 1991). Early Chinese and Japanese investigators suspected that the illness was related to small mites. In 1920 Hayashi isolated an agent from mites that he called ‘Theileria tsutsugamushi’. In retrospect, this agent was not the cause of scrub typhus, but the term ‘tsutsugamushi’ (for ‘noxious mite’) has persisted. In retrospect, the first identification of the causative agent of scrub typhus was by Nagayo and co-workers in 1930 (Nagayo et al. 1930). They called this organism Rickettsia orientalis but the name was changed to R. tsutsugamushi in 1948 (Bengston 1948) and then to Orientia tsutsugamushi in 1996. Prior to Nagyo’s isolation of O. tsutsugamushi, investigators working with what is now Malaysia used epidemiological data to classify locally occurring disease into urban (or ‘shop’) typhus or rural typhus. Rural typhus, which occurred predominately in grass or shrub land, later came to be called scrub typhus by the British and Amercians during the Second World War. The term scrub typhus is now used throughout the world except Japan where the name ‘tsutsugamushi disease’ is preferred. However, other symptoms have also been used, including chigger-borne rickettsioses, Kedani (hairy mite) fever, akamushi (red mite) fever, flood fever, Japanese river fever, and tropical typhus.

The interest of physicians and scientists in this disease increased during the Second World War, when more than 15,000 cases of infection were diagnosed among the allied forces, with a mortality rate varying from 1 to 35% (Weiss 1981). However, the disease cannot be associated with war conditions or natural disasters as is the case with epidemic typhus. The high incidence of scrub typhus during the Second World War and, to a lesser extent during the Vietnam War, can be ascribed to the fact that, during military field operations, larger numbers of non-immune individuals were introduced into ecological niches inhabited by trombidulid mites. Therefore, scrub typhus should not be directly associated to the lack of hygiene and health care, which are characteristic features of war conditions. The term ‘scrub’ typhus was adopted because it was thought that the vectors were mainly found on scrub vegetation.

Although the high incidence of scrub typhus among the allied troops may be partly due to false-positive serologiclal results, its occurrence led to a better description of the epidemiology and clinical features of the disease, and to the introduction of appropriate treatments. The disease shares common features with epidemic typhus, for example fever, headache, and rash, but the presence of an eschar and generalized lymphadenopathies is distinctive of scrub typhus.

In 1982, the WHO pointed out that, based on specific serological tests, a high proportion of fevers of unknown origin in endemic areas were probably undiagnosed scrub typhus cases even though the characteristic clinical signs, fever, eschar and adenopathies, are often lacking (Groupe de travail OMS 1982). In 1993, a WHO meeting on global surveillance of rickettsial disease reported that, if the amount of epidemiological data collected on rickettsioses was considered rather inadequate in developing countries, then the information on scrub typhus was downright non-existent (WHO 1993). Emergence of scrub typhus has recently been observed in Australia and Japan, proving that endemic foci of the disease persist and that the disease should not be underestimated (Yamshita et al. 1988, 1994; Currie et al. 1993).

O. tsutsugamushi is an obligate intracellular Gram-negative bacterium that has a different cell wall structure and genetic make-up from those of rickettsiae. Compared to other species belonging to the genus Rickettsia, O. tsutsugamushi presents a lack of peptidoglycan and lipopolysaccharide (Ohashi et al. 1990) that make a thicker outer leaflet of the cell wall (Silverman et al. 1978). This difference makes the bacteria very soft, fragile and more resistant to growth in penicillin (Miyamura et al. 1989). Orientia has also a unique profile of antigenic variation and this heterogeneity among strains is greater than that encountered in other Rickettsiales. O. tsutsugamushi has a variable 56-kDa protein as a major surface protein antigen, which accounts for 10 to 15% of its total protein. As a result there are three classical antigenically distinct prototype strains of O. tsutsugamushi, (Karp, Kato and Gilliam) isolated from New Guinea, Japan, and Burma respectively (Chattopadhyay and Richards 2007). However, other strain types have been found in Thailand and it has also reported that Shimokoshi, Kawasaki, and Kuroki strains, which were isolated from patients in Japan, were antigenically distinguishable from the prototype strains of Gilliam, Karp, and Kato (Tamura et al. 1999). This antigenic variation depends largely on the diversities of the immunodominant 56-kDa type-specific antigen located on the surface of this microorganism.

Recently, the complete genome of O.tsutsugamushi strain Boryong was sequenced by Cho et al. (2007) which is a single circular chromosome consisting of 2,127,051 bp with an average G+C content of 30.5%. The genome size and estimated number of genes are the largest among the currently sequenced genomes in the order Rickettsiales. The repeat density of the scrub typhus pathogen is 200-fold higher than that of its close relative R. prowazekii, the agent of epidemic typhus. A total of 359 tra genes for components of conjugative type IV secretion systems were identified at 79 sites in the genome. A unique feature is the presence of 4,197 identical repeats >200 bp, which represents 37.1% of the O. tsutsugamushi genome. Additionally, the O. tsutsugamushi genome contains >400 transposases, 60 phage integrases, and 70 reverse transcriptases.

Scrub typhus and is one of the most common infectious diseases of rural south, south-eastern Asia and the western Pacific. The endemic region of scrub typhus is often referred as the ‘tsutsugamushi triangle’ bounded to the north by Siberia and the Kamchatka Penisula, to the south by Australia, to the east by Japan and to the west by Afghanistan and India (Fig. 13.1). Endemic areas range from typical tropical secondary growth (scrub) vegetation to temperate zones and even the Himalayas (Jensenius et al. 2004).

The disease is transmitted by the bites of larval trombiculid mites (chiggers) of the genus Leptotrombidium which live in forests and areas with tall grass. These mites only feed on mammalian tissue fluid once in their lifetime and constitute the reservoir of infection through transovarial transmission. Chiggers usually feed on rats but may readily bite humans on any part of the body and feed for 2 to 10 days. Several species can act as vectors, which are also known as reservoirs of the disease with Leptotrombidium deliense to be the

most important vector species in Southeast Asia and southern China, whereas L. akamushi, L. scutellare and L. pallidum are main vectors in Korea and Japan and L. chiangraiensis was found in cultivated rice fields in Thailand (Watt and Kantipong 2007).

Humans usually become infected when they accidentally encroach on a zone in which there are infected chiggers. That is the reason that disease transmission has been reported in suburban area during clearing of land, logging, and road building, when residents of city centers are not at risk. The transmission depends on the seasonal activities of both chiggers and humans with most scrub typhus cases occurring during the rice-planting and rice-harvesting seasons, although infected chiggers and rodents can be found in rice fields all through the year (Watt and Kantipong 2007). Co-infection by leptospirosis and scrub typhus is also possible and rice farmers in north-eastern Thailand are commonly infected with both diseases (Watt and Parola 2003). Military personnel are also at risk and it is estimated that during the Second World War, there were 18,000 cases of scrub typhus and during the Vietnam war, the ratio was one case of scrub typhus to 50 to 100 cases of malaria (Baxter 1996; Bavaro et al. 2005). Travellers visiting rural areas in order to engage in activities like camping, trekking, or rafting may be also in danger. Since 1986 about 20 cases of travel-associated scrub typhus have been reported to people returning from scrub typhus endemic areas and mainly from Thailand (Jensenius et al. 2004).

Much remains to be learned about the pathogenesis of scrub typhus. Several recent investigations have focused on the survival of O. tsutsugamushi in the intracellular milieu and on the host response to infection. O. tsutsugamushi invades host cells by induced phagocytosis and escapes from the phagosome to the cytosol (Watt and Kantipong 2007). The bacteria can induce apoptosis in a variety of host cells, including lymphocytes and endothelial cells by retarding the release of intracellular calcium (Watt and Parola 2003). Moreover in murine macrophages, O. tsutsugamushi produces inflammatory cytokines tumour necrosis factor (TNF)-alpha and interleukin-6. TNF-α production appears to be inhibited by the production of interleukin-10 (Watt and Kantipong 2007).

O. tsutsugamushi disseminates from the skin to target organs, and the bacteria can be demonstrated in peripheral white blood cells taken from patients presenting to hospital with acute scrub typhus (Watt and Parola 2003). O. tsutsugamushi produces vasculitis and perivasculitis of the small blood vessels of the skin, lungs, heart and brain. Endovasculitis and focal haemorrhage may be present but less prominent than in Rocky Mountain spotted fever and epidemic typhus (Watt and Kantipong 2007). The basic histopathological lesions, disseminated perivasculitis, and focal interstitial mononuclear infiltrations associated with oedema suggest that macrophages are a more important target cell than the endothelium (Watt and Kantipong 2007).

The onset of symptoms occurs usually 7 to 10 days after the chiggers bite and is usually abrupt, with fever, a drainage lymphadenopathy, chills, myalgia, a macular or maculopapular eruption, and headache. Rash, gastrointestinal manifestations, cough, and other respiratory symptoms are also frequent (Parola and Raoult 2006). In a recent study from Laos scrub typhus was the most common rickettsiosis identified and the patients presented with fever, headache, nausea, myalgia, lymphadenopathy, and a palpable liver (Phongmany et al. 2006). Moreover, in a review of 87 American soldiers who were infected with scrub typhus in South Vietnam, all were reported to have fever, headache, and anorexia (Berman and Kundin 1973). Other symptoms were chills (80%), cough (45%), myalgia (32%), and nausea (28%). The most common physical finding was generalized lymphadenopathy (85%), and an eschar was found in 46% of these. The eschar is primarily detected in males but it may be absent or innocuous (Kim et al. 2007a). The prevalence of eschars varies and patients from Laos with serologically confirmed scrub typhus had a prevalence of about 50% (Phongmany et al. 2006) but in patients from Japan the prevalence was 87% (Phongmany et al. 2006), in Korea, 92% (Kim et al. 2007a) and South Vietnam, 48% (Berman and Kundin 1973). The eschar is often hidden in skin folds, beneath the beltline, under the axilla, or around the buttock region. Most cases of scrub typhus are mild, but without treatment it may progress to severe organ dysfunction, meningismus, dyspnea, pneumonitis, acute respiratory distress syndrome (ARDS), Guillain-Barré, meningoencephalitis, disseminated intravascular coagulation, or renal failure (Pandey et al. 2006; Wang et al. 2007; Lee et al. 2007; Chen et al. 2006; Phongmany et al. 2006). The fatality rate of scrub typhus ranges from 1 to 35%, depending on the virulence of the infecting strain, host factors, and treatment (Silpapojakul 1997).

Although co-infection with the HIV-1 virus does not affect the clinical severity of scrub typhus, the discovery that acute O. tsutsugamushi infection was shown to suppress HIV-1 viral replication both in vivo and in vitro has spurned the development of new lines of research in HIV treatment (Parola and Raoult 2006).

The clinical and laboratory features of scrub typhus are notoriously non-specific. The eschar is the single most useful diagnostic clue but it may only appear with a minority of patients and can be overlooked by physicians. Infection with O. tsutsugamushi should always be suspected in patients who have visited or live in endemic areas and present with fever, acute hearing loss, lymphadenopathy with or without eschars. Unfortunately, scrub typhus is an often overlooked cause of both acute undifferentiated fever and of pneumonitis of undetermined etiology (Watt and Kantipong 2007). Rice farmers in north-eastern Thailand are commonly infected with both leptospirosis and scrub typhus and dual infections should be considered in patients at risk for both who have atypical clinical features of either disease alone, and in patients responding poorly to treatment.

O. tsutsugamushi can be grown in the yolk sac of 5–7 day old embryonated chicken eggs, in primary cultured cells of chicken embryos and established cell lines such as HeLa, Vero, BHK, McCoy and L929, and mice (Seong et al. 2001). PCR to peripheral blood and skin biopsy from patients with acute scrub typhus can be also used for the detection of O. tsutsugamushi. PCR must be performed before initiation of antibiotic treatment and before antibody becomes detectable. Recently, a real-time PCR assay has been introduced for the diagnosis of scrub typhus detecting major outer membrane antigen genes (Singhsilarak et al. 2005; Jiang et al. 2004).

A variety of serologic assays have been used in the diagnosis of scrub typhus such as the indirect immunoperoxidase and Weil-Felix assays. The Weil-Felix slide agglutination test lacks sensitivity but is easy to perform in less developed areas of the world (Isaac et al. 2004; Mahara 1984). IFA is the gold standard assay for the serologic detection of scrub typhus antibodies (La ScoLa and Raoult 1997) which has subsequently been modified to allow the use of smaller volumes of serum and antigens (Gan et al. 1972). Maximum titres of patients treated after 7 days of symptoms averaged between 1:640 and 1:1,280, compared to patients treated promptly who had maximum antibody titres between 1:40 and 1:160 (Berman and Kundin 1973). The immunoperoxidase assay has been developed as an alternative to IFA for the diagnosis of scrub typhus (La ScoLa and Raoult 1997). The advantage of the immunoperoxidase assay is that it provides a permanent slide record, and the results can be read with an ordinary light microscope. In cases of sera cross-reactions, Western blotting assay of the outer membrane proteins can be used for the detection of scrub typhus (Jiang et al. 2003). Rapid serological tests have also been developed and are positive in more than 90% of patients with scrub typhus infection during the first week of fever (Watt and Kantipong 2007).

In tissue cell cultures, O. tsutsugamushi are susceptible to tetracycline, demethylchlortetracycline, doxycycline, minocycline, chloramphenicol, and rifampin (Raoult and Drancourt 1991). Quinolones such as norfloxacin, ciprofloxacin and ofloxacin were only moderately active and beta-lactams and nalidixic acid were inactive (Urakami et al. 1989).

Therapy with anti-rickettsial drugs is recommended whenever a case of scrub typhus is suspected. Scrub typhus is said to respond even more promptly to antibiotics than do other rickettsial diseases, with patients generally becoming afebrile within 24 to 36 hours after beginning antibiotic therapy. Prompt antibiotic therapy generally prevents death, but good supportive care and early detection of complications are important in severe cases. To date, chloramphenicol (2 g/day) has been the first effective antibiotic for the treatment of scrub typhus. Currently, however, doxycycline (200 mg daily) for not less than 7 days is regarded the drug of choice. Shorter treatment courses are often curative, but may result in relapse (Sheehy et al. 1973). Physicians should be aware that scrub typhus cases from Chiangrai, in Thailand have been resistant to chloramphenicol and doxycycline (Watt et al. 1996). Roxithromycin was as effective as doxycycline and chloramphenicol in a trial of 39 Korean children (Watt and Kantipong 2007). Recently telithromycin was also found to be effective in 92 Korean patients (800 mg/day for 5 days) (Kim et al. 2007b). Quinolones have been used experimentally although fever usually subsides later when ciprofloxacin is used and its efficacy remains questionable (Mathai et al. 2003).

In cases of co-infection with leptospirosis, doxycycline was found to be highly effective (Phimda et al. 2007). However, azithromycin can be used as an appropriate alternative antimicrobial treatment in areas where doxycycline-resistant scrub typhus is prevalent (Phimda et al. 2007). Recently, a patient infected by leptospirosis and scrub typhus, was treated by early plasma exchange and a 7 day course of moxifloxacin therapy (Chen et al. 2007).

Chemoprophylaxis for persons with anticipated intense but transient exposure to O. tsutsugamushi with weekly doses of 200 mg of doxycycline has been proposed (Twartz et al. 1982). Soldiers and road construction crews are typically examples but chemoprophylaxis should also be considered in travellers at high risk areas. Contact with chiggers can be reduced by not sitting or lying directly on the ground and by applying repellent to the tops of boots and socks and to the hem of trousers, but this may be impractical in those exposed occupationally. The meal of mites lasts several days, but it seems that infectious organisms are not transmitted during the first 6–8 hours after attached. Thus rapid removal of attached chiggers may be helpful in avoiding infection. Because of the antigenic diversity of the various serotypes, an effective vaccine for prevention of scrub typhus has not been developed (Chattopadhyay and Richards 2007).