Background
Rickettsial disease is caused by several species within the order of Rickettsiales, which are obligate intracellular arthropod-borne gram negative bacteria and divided into the Rickettsiaceae and Anaplasmataceae families [
1]. Within the Rickettsiaceae family, there are two genera: the genus
Rickettsia which is subdivided into four subgroups, i.e. spotted fever group (SFG) rickettsiae, typhus group (TG) rickettsiae, the
Rickettsia bellii group and the
Rickettsia canadensis groups [
2], and the genus
Orientia (i.e. scrub typhus group, STG) with
Orientia tsutsugamushi as type species
.
Murine or endemic typhus is caused by
R typhi, which belongs, together with
R prowazekii to the TG rickettsiae. The main vector of
R typhi is the widely spread rat flea
Xenopsylla cheopsis, which also feeds on peridomestic animals such as cats and dogs. Murine typhus has been mainly described in travelers returning from Southern Asia [
3,
4].
Scrub typhus is caused by
Orientia tsutsugamushi, and is transmitted by trombiculid mites (chiggers), primarily living on rodents in tall grasses and scrub vegetation. The disease is endemic in large parts of the world, including the Far East, South-East Asia, Russia, India and Australia [
5]. Travel-related scrub typhus has been almost exclusively reported in travelers returning from South-East Asia [
6].
Rickettsial disease classically presents with fever, rash and in case of scrub typhus, an inoculation eschar [
7]. Its course can be severe and sometimes life threatening, highlighting the need for prompt diagnosis and treatment. Clinical diagnosis is sometimes suspected on pattern recognition, but is hardly possible in the substantial proportion of cases with undifferentiated presentation. Definitive diagnosis is still usually based on antibody detection in serum requiring comparison between acute- and convalescent-phase samples [
8]. Negative results during the acute febrile phase, cross reactivity between
Rickettsia species (that may have different prognosis) and very long antibody persistence are well-identified problems of serological assays [
9]. In addition, they are often performed in batch for evident cost-effectiveness purposes and therefore of little utility for immediate case management. Molecular techniques, which may circumvent most of these limitations, are increasingly used but availability remains limited in many clinical settings [
10].
In this report, we describe two cases of murine typhus and two cases of scrub typhus diagnosed by real-time polymerase chain reaction (PCR) on serum or whole blood in Belgian travelers admitted for severe febrile illness after a stay in the tropics.
Discussion and conclusion
We report on four travel-related cases of severe rickettsial disease, in whom PCR on serum or whole blood allowed a definitive species-specific diagnosis.
Rickettsial disease is not unusual in travelers but is predominantly due to
R.africae from the spotted fever group [
6]. In one large prospective study conducted before PCR assays became available, rickettsiosis was diagnosed by means of serology in 3.3% of febrile returning travelers, while both murine and scrub typhus were found in only 0.2% of all cases [
14]. In a 12-year survey conducted by the GeoSentinel network, rickettsioses accounted for 1.5% of all etiologies of travel-related fever. Of the 280 patients with rickettsiosis, 16 (5.7%) were diagnosed with scrub typhus and 10 (3.6%) with TG rickettsiosis [
6]. In the most recent GeoSentinel publication, murine and scrub typhus similarly accounted for 10% of all rickettsial infections in travelers [
4]. Both conditions however ranked in the top 10 of potentially life threatening disease in travelers [
15].
Clinical diagnosis may be rather straightforward for the spotted fever group (in particular African tick bite fever), but is usually much more challenging for the typhus group- and
Orientia infections since symptoms are most of the time non-specific and typical signs may often be absent. Incubation periods vary from 7 up to 14 (murine typhus) or to 21 days (scrub typhus), being in the same range as most other tropical conditions. Travel and exposure history can be helpful, in particular when there is a history of contact with rats or domestic animals for murine typhus- and hiking through bush wood for scrub typhus. The presence of an eschar, reflecting the local cutaneous vasculitis due to rickettsial multiplication, is usually missing in murine typhus. Albeit a key clinical finding in scrub typhus, it may be absent in up to 50% of the cases [
16] or may go completely unnoticed because of its small size, absence of pain, hidden localization, atypical aspect or self-healing character once systemic symptoms occur (as was possibly the case in Patient 4). The maculopapular rash, again a classic finding in the spotted fever group, is present in 20–80% of murine typhus- and in about 50% of scrub typhus patients [
16,
17]. As illustrated by our cases, disease course can be severe, and hospitalization is often required. Compared to a hospitalization rate of about 10% in SFG-patients, hospitalization rate in travelers with scrub typhus or murine typhus can reach 37 or 50% respectively [
6,
15]. Similarly to what is observed in the spotted fever group [
18], complications as diverse as hematophagocytic syndrome, myocarditis, shock, renal failure, acute respiratory distress syndrome or encephalitis have been reported in travelers presenting with murine [
1,
19‐
21] or scrub typhus [
22‐
24]. Mortality rates are generally low (0–6%) in high-resource settings [
6,
15], even in case of diagnostic delay, but remain substantial in most endemic areas [
25,
26]. Patient 1 developed transient sensorineural hearing loss during convalescence. This has been reported as a rare focal neurological manifestation of rickettsial disease, possibly due to immune-mediated mechanisms [
16,
27]. Patient 2 presented with multiple organ involvement and retinal lesions, probably secondary to the widespread vasculitis that is the hallmark of rickettsial infection [
28,
29]. Both patients 1 and 3 had severe hepatitis at presentation. This highlights the need for prompt diagnosis and empirical anti-rickettsial treatment in severely ill febrile travelers, especially when other relevant conditions have been ruled out.
The current gold standard for diagnosing murine as well as scrub typhus is the documentation of a seroconversion or of a 4-fold increase/decrease in antibody titer, which may take several weeks to occur, by the indirect fluorescent antibody (IFA) assay. Timely diagnosis is therefore difficult, as we have experienced in our patients. The initial serological testing would probably have missed patients 1 and 2 if they had presented earlier in the course of the disease. In addition, this gold standard is sometimes imperfect, as illustrated by patients 3 and 4 in whom
O. tsutsugamushi infection was diagnosed by PCR while IFA remained negative, as described in previous reports [
30]. The reasons for this surprising observation remain unclear; prompt antibiotic treatment may sometimes hamper antibody production possibly explaining the negative serology findings on convalescent serum in patients 3 and 4.
Real time PCR on blood and serum (or eschar, if present) offers the possibility of earlier diagnosis, by detecting DNA during the acute local and systemic rickettsial dissemination. It was particularly useful in patient 4 who had no “typical” clinical manifestations. PCR also permits genus- and species identification, which is not always possible with serological testing. The current limitations are its technical difficulties and relatively high cost, making this technique not suitable in its current design for poor-resource settings. Also, the diagnostic accuracy of PCR on blood/serum remains to be fully evaluated
. A prospective evaluation of real time PCR in early diagnosis of rickettsial disease in 180 febrile patients, living on the Thailand-Myanmar border, showed that sensitivity of PCR was rather low in serum and slightly better in whole blood (28.6% for
O. tsutsugamushi and 36.4% for
Rickettsia spp), when compared to IFA IgM seroconversion. On the other hand, PCR outperformed the acute phase detection of IgM (14.3% for
O. tsutsugamushi and 9.1% for
Rickettsia spp), but evaluation in larger cohorts are still needed [
31]. So far, only five cases of travel-related murine typhus have been diagnosed by PCR on blood, targeting different antigens such as the 17 kDa antigen, the citrate synthase gene (
gltA) or the
Rpr 247P gene of
R typhi, [
3,
6,
19,
20,
32]. Data on PCR diagnosis on blood of travelers with scrub typhus are even more scarce. Real-time PCR, targeting the 47 kDa and the 56 kDa antigens has been used on eschars in scrub typhus patients returning from respectively Cambodia and Vietnam [
33,
34], but we have no knowledge of any report describing PCR analysis on blood in travel-related scrub typhus. In patients 1 and 2 with murine typhus, the initial antibody titers were already high, triggering our PCR investigation in parallel. Definitive diagnosis would have been possible earlier if PCR had been requested and performed “immediately” at admission. And it is reasonable to consider that it would have allowed a much faster diagnosis if both patients had consulted earlier. On the other hand, diagnosis of scrub typhus would have been completely missed in patients 3 and 4 with the conventional serological methods only.
In conclusion, we emphasize the need for prompt recognition and treatment of travel-related murine and scrub typhus, which may both present as severe systemic illnesses. The use of PCR on whole blood or serum may become a valuable contribution to the acute-phase diagnosis at genus and species level, with important clinical and prognostic implications. Identifying the subset of febrile travelers most likely to benefit from this specific PCR testing remains an area for research.