12 Epidemic and murine typhus

Epidemic typhus is now a rare disease, but previously it was worldwide in distribution. Typically epidemic typhus occurred during instances in which humans were forced to live in crowded, cold, and unhygienic conditions (e.g. aboard ships, in jails, and during military operations). Until recently man was considered to be the only reservoir for R.prowazekii, but in 1975, a new sylvatic cycle involving the flying squirrel and its ectoparasites was discovered in the eastern USA.

Murine typhus occurs throughout the world. Its epidemiology is primarily linked to the distribution of rats and the rat flea, Xenopsylla cheopis. However, recently both a new reservoir (opossums in southern California) and a new potential vector (the cat flea) have been discovered.

Control or avoidance of the vectors are the cornerstones of strategies to prevent morbidity and mortality.

Bacteria of the order Rickettsiales were first described as short Gram-negative bacillary microorganisms that retained basic fuchsin when stained by the method of Gimenez and grew in association with eukaryotic cells (Raoult and Roux 1997). In 1993, Rickettsiales was divided into three families, namely, Rickettsiaceae, Bartonellaceae, and Anaplasmataceae (Raoult and Roux 1997). To date the Rickettsia genus is currently made of 24 recognized species, and also contains several dozens of as yet uncharacterized strains or tick amplicons (Fournier and Raoult 2007). These 24 species are classified into three groups:

The life-threatening louse-borne epidemic typhus, also named ‘jail fever’, is caused by a Rickettsia of the typhus group, R. prowazekii, with the human body louse as vector. The first description of epidemic typhus was in the sixteenth century in the Mediterranean area and the name typhus was first used in 1760 (Andersson and Andersson 2000). From the sixteenth century until the Second World War, the disease is mentioned to afflict each army moving up through Europe caused by the low hygienic conditions which normally exist during wars. It is estimated that 30 million cases occurred in the Soviet Union and Eastern Europe between 1918 and 1922, with an estimated 3 million deaths (Saah 1995). After the Second World War, foci of the disease remained restricted to the cooler mountainous countries of Africa and epidemic typhus was considered a disease of the past (Parola and Raoult 2006). However, cases of epidemic typhus are still reported in situations of poverty, lack of hygiene (Mokrani et al. 2004; Zanetti et al. 1998), and among homeless people who are particularly exposed to ectoparasites—they often present with the hallmarks of epidemic typhus and relapsing fever (Badiaga et al. 2005).

R. prowazekii belongs to the alpha subgroup of Proteobacteria and is short (0.8–2 μm long and 0.3–0.5 μm diameter), Gram-negative bacillary organism, classified as a category B bioterrorism agent. The genome of R. prowazekii (1.1 Mb) consists of a single circular chromosome (Andersson et al. 1998), contains a high proportion of non-coding DNA (24%) (Andersson et al.1998). R. prowazekii is related antigenically to R. typhi (Baxter 1996).

The disease is transmitted by the body louse (Pediculus humanus corporis) which lives in clothes (Gross 1996) (Fig. 12.1) but R. prowazekii has been also found in African-Hyalomma ticks (Reiss-Gutfreund 1966), Amblyomma ticks from Mexico (Medina-Sanchez et al. 2005) and in flying squirrels (Bozeman et al. 1975). Lice live in clothing and their prevalence is determined by the weather, humidity, poverty, and lack of hygiene. As a result, R. prowazekii is transmitted to people when the infected faeces of lice contaminate their feeding sites or when conjunctivae or mucous membranes are exposed to the crushed bodies or faeces of infected lice. Infected lice usually die within 1 to 3 weeks due to obstruction of the alimentary tract and do not transmit the organism to their offspring. Transmission might also result from the inhalation of infected faeces and this is thought to be the main route of infection for health workers attending patients (Parola and Raoult 2006). Infection through aerosols of faeces-infected dust has been reported and provides the main risk of contraction of typhus for the physicians who are in contact with the patients infected with the infected lice (Houhamdi and Raoult 2007).

After inoculation, R. prowazekii spreads throughout the body via the bloodstream and enters the endothelial cells of capillaries and small blood vessels producing vasculitis usually in the skin, heart, central nervous system, skeletal muscle, and kidneys (Baxter 1996). In severe cases, endothelial damage results in permeability changes and the passage of plasma and plasma proteins from the intravascular compartments to the interstitium. As a result, tissue biopsy reveals perivascular infiltration with lymphocytes, plasma cells, polymorphonuclear leukocytes, and histiocytes, with or without necrosis of the vessel (Houhamdi and Raoult 2007).

Epidemic typhus occurs in two clinical forms: the primary febrile illness and recrudescent infection (Brill-Zinsser disease). The primary febrile illness has an incubation period from 10 to 14 days. Most patients develop malaise and vague symptoms before the abrupt onset of nonspecific symptoms of high fever (100%), headaches (100%), and severe myalgias (70–100%) (Parola and Raoult 2006). In a recent investigation in Burundi, a crouching attitude due to myalgia, named ‘sutama’, was reported (Houhamdi and Raoult 2007). A petechial rash may appear in 20% to 60% of cases (Parola and Raoult 2006). In Africa the rash is observed more rarely (20 to 40%) (Perine et al. 1992). Typically, the rash begins in the axillary folds and upper trunk on about the fifth day of illness and spreads to the extremities. The rash initially appears as non-confluent erythematous macules that blanch on pressure, but after several days becomes maculopapular and petechial, affecting the trunk and extremities and sparing the face, palms, and soles (Baxter 1996). Other manifestations include nausea or vomiting (42%–56%) and coughing-pneumonia (38%–70%) (Parola and Raoult 2006). Many patients manifest various abnormalities of the central nervous system (CNS), ranging from confusion to stupor (18%–80%), drowsiness, hearing loss, and coma (4%) (Parola and Raoult 2006). Myocarditis, pulmonary involvement, mild thrombocytopenia, jaundice, and abnormal liver function test may occur in severe cases. In uncomplicated epidemic typhus, fever usually resolves after 2 weeks of illness if untreated, but recovery usually takes 2 to 3 months (Baxter 1996). Without treatment, the disease is fatal in 13 to 30% of the cases (Raoult and Roux 1997).

People who survive epidemic typhus remain infected with R. prowazekii for life and under conditions of stress or of a waning immune system may experience a recrudescence, known as Brill-Zinsser disease. Brill-Zinsser disease is generally milder but the patients may become the source of a new epidemic if they become infected with body lice (Weissmann 2005). Patients with Brill-Zinsser has a fatality rate of 1.5% (Houhamdi and Raoult 2007).

Careful clinical examination and epidemiologic investigation of patients with potential rickettsioses is critical. Typical findings of epidemic typhus such as fever, headaches and skin rash in patients infected with body louse or in persons who are living in poverty and lack of hygiene, can suggest the diagnosis. Thrombocytopenia and an increase of the hepatic enzymes may be observed particularly in severe cases.

As with other rickettsial diseases the diagnosis of epidemic typhus can be confirmed by culture on to shell vials containing human embryonic lung (HEL) fibroblasts grown on coverslips, in a biosafety level 3 containment laboratory (Birg et al. 1999). Immuno-detection, in blood or other tissues, allows the confirmation of infection in patients before their seroconversion and thus permits early prescription of specific treatment. R. prowazekii can be detected by PCR amplification from an array of samples that include eschars, paraffin-embedded tissues, slide-fixed specimens, peripheral blood mononuclear cells, and arthropod tissues (Parola and Raoult 2006). If PCR-based diagnosis is delayed for more than 24 hours, the samples should be stored at –20°C or lower. Blood should be collected before antimicrobial therapy either in a citrate-containing vial for culture or in EDTA. Recently a nested PCR assay, 2.2 times more sensitive than culture and 1.5 times more sensitive than regular PCR using single-use primers targeting single-use gene fragments present in the genomes of R. prowazekii, was proposed by our laboratory and named ‘suicide’ PCR (Fournier and Raoult 2004).

The Weil-Felix test is the oldest serological assay for rickettsioses test but it has poor sensitivity and specificity (Ormsbee et al. 1977). Patients with Brill-Zinsser disease usually have no agglutinating antibodies detectable by the Weil-Felix test (La ScoLa and Raoult 1997). The rickettsial IFA adapted to a micromethod format is the test of choice for the serodiagnosis of rickettsial diseases. The diagnosis of recent epidemic typhus can be established by demonstrating a fourfold or greater rise in titer of antibody in acute and convalescent serum samples (Houhamdi and Raoult 2007). However, R. prowazekii’s antibodies cross-react with those of R. typhi and their differentiation is difficult by serology (Raoult and Roux 1997). As a result epidemic and murine typhus cannot be differentiated by serology. Cross-adsorption followed by IFA and Western blotting can increase the identification of the etiological agent but the high costs

of such studies limit their use (La ScoLa et al. 2000).

Enzyme-linked immunosorbent assay (ELISA) has been introduced for detection of R. prowazekii antibodies. The use of this technique is highly sensitive and reproducible, allowing the differentiation of IgG and IgM antibodies (La ScoLa and Raoult 1997).

Lice may be tested by molecular biology and used as an epidemiologic tool. Lice are easy to collect and to transport to reference laboratories where suitable molecular biological approaches can be used. For example, R. prowazekii detected in lice collected from refugees in Burundi was sent to our laboratory in Marseille to confirm the presence of epidemic typhus (Roux and Raoult 1999).

Cells models presented the efficacy of chloramphenicol, tetracycline, doxycycline, minocycline, and rifampin against R. prowazekii (Raoult and Drancourt 1991). Erythromycin, rifampin and the new ketolide, telithromycin were also effective (Rolain 2007). However, new quinolones were found only to be moderately active and beta-lactams and aminoglycosides were not effective (Raoult and Drancourt 1991).

Because antibiotic therapy does not eradicate rickettsia in lice-infested patients, delousing is essential in the management of a typhus outbreak. If appropriate treatment is begun promptly, complications, including mortality, can be avoided in most cases. Doxycycline remains the treatment of choice for epidemic typhus. A single dose of oral doxycycline 200 mg usually leads to defervescence within 48 to 72 hours in most cases. Chloramphenicol is widely used as an empirical treatment in places where diagnostic facilities are unavailable but it can be also administered in cases of allergy to tetracyclines. Fluoroquinolones should be avoided for the treatment of epidemic typhus as a patient who, misdiagnosed as having typhoid, died from typhus after treatment with ciprofloxacin (Houhamdi and Raoult 2007). Finally, co-trimoxazole is reported to be ineffective for the treatment of epidemic typhus (Huys et al. 1973). Table 12.1 presents the guidelines for the treatment of epidemic typhus.

Prevention efforts aimed at reducing conditions associated with contact with the human body louse and at minimizing the risk of arthropod bites should be taken. These measures include regular bathing and washing of clothes and the use of long-acting insecticides. Dusting of all clothing with 10% DDT, 1% malathion, or 1% permethrin is a rapid and effective method of killing body lice and reduces the risk of reinfestation (Raoult and Roux 1999). Doxycycline as chemoprophylaxis to visitors at high risk areas can also provide protection. Vaccines using crude antigen and/or inactivated rickettsia are partially protective against epidemic typhus but have been accompanied with undesirable toxic reactions and difficulties in standardization (Coker et al. 2003).

Murine typhus (also called endemic typhus) is caused by R. typhi (formely R. mooseri). It is classified as a typhus-group rickettsia, transmitted by fleas and has rodents as its main reservoirs. The disease was probably reported in Mexico in 1570 by Bravo (Parola and Raoult 2006). Paullin made the first clinical description of a ‘milder form of typhus’ without mortality in 1913 in Atlanta (Tselentis and Gikas 2007). In 1917, Neil noted that male guinea pigs inoculated with the blood of typhus patients in south Texas often developed a scrotal swelling and inflammation, along with haemorrhage beneath the tunica, similar to the lesions elicited by R. rickettsii (Tselentis and Gikas 2007). His results confirmed in 1928 by Mooser, showed that the American and European variety could be clearly differentiated by their reactions in the guinea pigs (Tselentis and Gikas 2007). These studies contributed in distinguishing the endemic from the classic typhus. In 1923, in the Annual Reports of the Department of Health of Palestine, there was a reference to a mild course of typhus fever, similar to Brill’s disease and different from the classic form of typhus (Tselentis and Gikas 2007). In 1925, an outbreak of 200 cases of endemic typhus occurred in Australia. Plazy, Marcandier, and Pirot observed the first Mediterranean cases of murine typhus on sailors on the warships of Toulon (Tselentis and Gikas 2007). In the period 1931–1946 in the USA, 42,000 cases were reported. Late reports established the worldwide distribution of endemic typhus. In 1940, Lewthwaite and Savoor reiterated the important distinction between the Xenopsylla cheopsis borne murine typhus (‘shop typhus’) and the chigger-borne scrub typhus (‘tsutsugmushi’ or ‘rural typhus’) (Lewthwaite 1952).

R. typhi is a small (0.4 × 1.3 μm), Gram-negative, obligate intracellular bacterium which belongs to the alpha subgroup of Proteobacteria. The genome of R. typhi (1,111,496 bp) is nearly identical to its close relative R. prowazekii and highly similar to R. conorii and other spotted fever group Rickettsia. A 12-kb insertion in the genome of R. prowazekii, a large inversion close to the origin of replication with no loss of genes in the region, several pseudogenes for which functional homologs are found in R. prowazekii and the fact that R. typhi has lost the complete cytochrome c ox

i dase system are some of the few differences between the two rickettsiae (McLeod et al. 2004).

Murine typhus is one of the most prevalent rickettsial diseases. R. typhi strains have a worldwide distribution, but the number of reported cases does not reflect the current prevalence. The disease occurs on every continent except Antarctica, in a variety of environments, ranging from hot and humid to cold and montane or semi-arid (Traub and Wisseman 1978). Cases are regularly documented in the USA, Mexico, and Europe. Murine typhus have also been reported in tourists returning from countries including China, Indonesia, India, Morocco, Canaries Isles, Africa, Malaysia, Southeast Asia, and Thailand (Parola and Raoult 2006). Recently, murine typhus, which had never been reported in Japan since the 1950s, re-emerged in that country (Parola and Raoult 2006). Although murine typhus is most prevalent in warmer countries, the fact that the disease is mild and non-specific suggests that its incidence is probably largely unrecognized or misdiagnosed in these.

The rat flea X. cheopsis and the rat louse (Polyplax spinulosa) are the principal vectors of murine typhus (Azad et al. 1997). Occasionally, other flea species or arthropods vectors have been reported to transmit R. typhi, including the cat flea Ctenocephalides felis, the mouse flea Leptopsyllia segnis, and lice, mites, and ticks (Boostrom et al. 2002). The fleas remain infected for life, but neither their lifespan nor their reproductive activity are affected (Tselentis and Gikas 2007). Rats belonging to the subgenus Rattus, mainly R. norvegicus and R. rattus are the primary reservoirs (Azad 1990) and the widespread distribution of murine typhus in many coastal areas is attributed to the introduction of infected rats and their fleas from ships. However, various rodents and other wild and domestic animals, such as house mice, cats, opossums, shrews, and skunks can also act occasionally as hosts (Azad 1990). The classical cycle of infection is rat-to-rat flea after a rickettsemic blood meal from an infected rat. Beside the classical cycle of infection, R. typhi is rarely transmitted transovarially from flea to uninfected flea (Azad 1990). Rickettsiae can persist into the flea faeces for several years and they infect humans by feeding or more rarely via inhalation or contamination of the conjuctiva (Azad 1990).

Infection of endothelial cells lining vessel walls, and the resultant vascular inflammation and haemostatic alterations are salient pathogenetic features of R. typhi (Walker et al. 1989). In fatal cases, organ damage secondary to vasculitis has been described in the lungs, kidneys, myocardium, brain, and liver parenchyma. Interstitial myocarditis is believed to be a major risk factor for death in these patients (Baxter 1996). Obliterative thrombovasculitis and perivascular nodules of the skin at autopsy resemble lesions of murine typhus. Widespread infection of the hepatic sinusoidal lining cells and the endothelium of vessels in portal regions results in injury to adjacent hepatocytes and may cause localized symptoms or liver function abnormalities (Baxter 1996).

The disease is usually mild with a group of symptoms that is shared with an array of other infectious diseases, including several bacterial and viral infections. As a result, many cases of murine typhus can be overlooked without a laboratory-confirmed diagnosis (WHO Working Group on Rickettsial Diseases 1982). Usually, after an incubation period of 6 to 14 days, patients with scrub typhus usually present an abrupt onset of symptoms like fever, rash, cough, headaches, maculopapular exanthema on the trunk to the half patients, chills, as well as with myalgias and hepatomegaly (Raoult and Roux 1997). Less common manifestations of murine typhus are lymphadenopathy (4%) (Gikas et al. 2002), splenomegaly (5%) (Silpapojakul et al. 1993). The rash is non-specific and its reported prevalence varies from 20% of patients from Thailand, 38% of patients from Laos, 49% of patients from Texas and 80% of patients from Greece (Silpapojakul et al. 1993; Phongmany et al. 2006; Gikas et al. 2002; Whiteford et al. 2001). Most cases of murine typhus are mild, and signs in untreated patients last for 7 to 14 days, when there is usually a rapid return to health, but aseptic meningitis, deafness, deep venous thrombosis, and even death have been reported with a fatality rate which may be as high as 4% (Dumler et al. 1991).

Murine typhus should be considered for patients from places with high rat populations such as tropical countries, but also from northern countries late in summer or early in autumn. As a result murine typhus should be considered in patients with prolonged fever and rash with or without lymphadenopathy during the summer or early autumn months (Koliou et al. 2007; Nogueras et al. 2006).

During recent years, the development of cell culture systems for viral isolation has led to an increase in the number of laboratories suitably equipped to isolate rickettsiae. Murine typhus can be confirmed by culture onto shell vials containing Vero or L929 cells (La ScoLa and Raoult 1997). Detection of rickettsiae by using immunofluorescence allows the confirmation of infection in patients prior to their seroconversion. R. typhi has successfully been detected in the organs of a patient with a fatal case of murine typhus. Biopsy specimens of the skin with a rash around the lesion and preferably petechial lesions are the most common samples used for digital immunodetection (La ScoLa and Raoult 1997). PCR and sequencing methods are useful, sensitive, and rapid tools to detect and identify rickettsiae in blood and skin biopsies. Four genes have been proposed for use in the identification of rickettsia, namely, those encoding 16S rDNA, a protein of 17 kDa, citrate synthase, and OmpB (La ScoLa and Raoult 1997).

Because specific antibodies are frequently absent during the acute illness, serologic diagnosis can be made by obtaining acute and convalescent serum during the acute illness, serologic diagnosis can be made by obtaining acute and convalescent serum for specific rickettsial antibodies. The rickettsial IFA adapted to a micromethod format is the reference method for serodiagnosis of R. typhi in most laboratories. The micro-IFA has the advantage that it can simultaneously detect antibodies to a number of rickettsial antigens (up to nine antigens) with the same drop of serum in a single well containing multiple rickettsial antigen dots. An immunoperoxidase assay has been developed as an alternative to IFA R. typhi. The procedure is the same as IFA, but fluorescein is replaced by peroxidase (La ScoLa and Raoult 1997). The advantage of the immunoperoxidase assay is that the results can be read with an ordinary light microscope. In addition, the immunoperoxidase assay provides a permanent slide record. Cross-absorption studies are useful, especially if complemented by Western blotting. This is the case for typhus because in 50% of patients, the sera had the same level of antibodies to both R. prowazekii and R. typhi (Schriefer et al. 1994).

Enzyme-linked immunosorbent assay (ELISA) has also been introduced for the detection of antibodies against R. typhi. The use of this technique is highly sensitive and reproducible, allowing the differentiation of IgG and IgM antibodies (La ScoLa and Raoult 1997).

Fibroblasts and Vero cells have been used for the determination of the antibiotic susceptibility R. typhi in vitro testing. Chloramphenicol, tetracycline, doxycycline, minocycline, and rifampin were effective against R. typhi (Raoult and Drancourt 1991). R. typhi was also found to be susceptible to erythromycin, rifampin and the new ketolide, telithromycin (Rolain 2007). New quinolones were found to be moderately active when beta-lactams and aminoglycosides were not effective (Raoult and Drancourt 1991).

Complete clinical recovery is observed over a period of 15 days from initiation of symptoms, even in untreated murine typhus. Infection confers long-lasting immunity to reinfection (Tselentis and Gikas 2007). Doxycycline 200 mg/day is the recommended treatment for murine typhus and a single dose of oral doxycycline 200 mg usually leads to defervescence within 48 to 72 hours in most cases. However, of 7–15 day course is recommended. In patients with severe hypersensitivity to the tetracyclines, 2 gm per day of chloramphenicol has been considered an alternate therapy, but its use is limited by side effects. Chloramphenicol is the drug of choice for pregnant patients, except for the parturient, due to the danger of gray syndrome to the infant (Tselentis and Gikas 2007).Chloramphenicol relapses have been reported in patients with murine typhus (Shaked et al. 1989).

The control of the flea vector and mammalian reservoirs of infection is the main step for the prevention of murine typhus. As a result, the reduction of rat populations and the insecticide dusting campaigns has decreased the incidence of murine typhus. No effective vaccine is available for murine typhus but recovery from natural infection confers long-lasting immunity to reinfection (Baxter 1996).