The study included 221 adults >14 years of age who were inpatients and outpatients at the Hospital Universitario Insular of Las Palmas (Las Palmas de Gran Canaria, Spain), who received a diagnosis of murine typhus during June 1, 1999–December 31, 2015. Epidemiologic, clinical, and laboratory data were retrospectively collected from medical records. Diagnosis of murine typhus was based on detection of antibodies against R. typhi by using an indirect immunofluorescence test and 2 criteria. Criterion 1 was titer >1:1,280 for IgM in 1 sample, and criterion 2 was a 4-fold increase in IgG titers between 2 consecutive samples. A total of 72 (32.6%) patients were given a diagnosis according to criterion 1, and 149 (67.4%) patients were given a diagnosis according to criterion 2. Clinical and laboratory data for both groups were analyzed separately.

Figure 1

Figure 1. Monthly distribution of cases of endemic murine typhus, Canary Islands, Spain, 1999–2015.

Figure 2

Figure 2. Annual distribution of cases of murine typhus, Canary Islands, Spain, 1999–2015.

Murine typhus was more frequent during July–November (Figure 1). The mean ± SD number of cases diagnosed per year was 18 ± 5.33. We provide the annual distribution of cases (Figure 2). Most (91.4%, 202/221) case-patients lived in urban areas; 73.3% (162/221) were male; and the median age was 40 years (interquartile range 28.5–52.5 years). Most (88.7%, 188/212) reported close contact with animals, especially dogs (66%, 140/212) and cattle (42%, 89/212). Arthropod bites were reported by 34 (19.5%) of 174 case-patients.

We provide the main clinical features recorded (Table 1). A total of 180 (82.95%) of 217 patients had high fever (median temperature 39.8°C) of >1 week duration. Pharyngitis was more frequently observed for younger (<20 years of age) patients (7/20, 35%) than for older patients (12/169, 7.1%) (p = 0.001). Rash was present more often in younger patients (8/19, 42.1%) than in older patients (42/190, 22.1%) (p = 0.05). We also provide laboratory findings (Table 2). Most (184/195, 94.4%) had urinalysis alterations in the form of microhematuria, proteinuria, or leukocyturia.

Complications developed in 31.6% (68/215) of patients, especially hepatitis (22/221, 10.0%), acute renal failure (21/215, 9.8%), meningitis (12/215, 5.6%), and pneumonia (9/215, 4.2%). No differences were found between patients given a diagnosis by using criterion 1 or 2 (Appendix). Cerebrospinal fluid samples from patients who had meningitis were characterized by a clear appearance, moderate mononuclear pleocytosis (range 6–43 cells), mild proteinorachia, and standard glucose levels. Round pneumonia developed in 2 patients and retinitis in 1 patient.

A total of 51 (22.6%) of the 221 patients were hospitalized: 29 (56.9%) had complications, 12 (23.5%) experienced vomiting, and 10 (19.6%) needed a diagnostic workup. The average length of hospital stay was short (median 6 days; interquartile range 4–9 days). Seven patients did not receive treatment because of spontaneous recovery. The remaining patients received doxycycline. Two patients required admission to the intensive care unit because of multiple organ failure, and no patients died. Low transient IgM titers and no IgG titers against other microorganisms were found in admission serum samples, especially against Coxiella burnetii (36/218, 16.5%) and Epstein-Barr virus (13/218, 6.0%).

Murine typhus was diagnosed primarily in middle-age men and showed a similar male:female ratio as in other clinical series (8). The seasonal prevalence of murine typhus during late summer and fall has been described (7). This temporal pattern seems to be related to the increased propagation activity of the vector linked to higher temperatures. The number of annual cases is similar to that reported by others (9–11), and diagnoses increased over the study period. However, these data probably underestimate the incidence of murine typhus because of the absence of clinical hallmarks and the fact that this disease is self-limiting.

The clinical features of murine typhus observed in this study are consistent with those reported by Tsioutis et al. (7); high fever and intense headaches were the most common clinical features. Most of the patients fulfilled the criterion for FID. There were differences by age groups. The presence of a rash was rare among elderly patients, as reported (12,13). This finding makes the diagnostic utility of the classic triad of fever, headache, and rash somewhat debatable, especially for older patients. Furthermore, patients <20 years of age sometimes showed a clinical profile indistinguishable from that for infectious mononucleosis associated with pharyngitis, visceromegaly, lymphadenopathy, and atypical lymphocytosis.

The most common finding for blood counts was thrombocytopenia (127/218, 58.3%). A prolonged prothrombin time was common. No association was observed between a prolonged prothrombin time and complications, which is in contrast to the results of Chang et al. (14). Hypertransaminasemia was the most common serum alteration, which reached values typical for viral, toxic, or ischemic hepatitis ​​in 10% of case-patients. However, clinical hepatitis, with the presence of hepatomegaly and increased levels of bilirubin, was much less frequent.

The higher incidence of renal damage for patients with murine typhus in the Canary Islands has been reported (15). This differential finding could be caused by specific strains of R. typhi that have a particular tropism, although there is no solid evidence to confirm this possibility.

Transient IgM titers against other microorganisms in admission serum samples are common. Obtaining 2 independent samples during an interval of 2 weeks is essential to avoid false-negative results or misdiagnoses.

A limitation of this study is its retrospective design, although based on an established protocol. Another limitation is the possibility of cross-reactivity; cross-reactivity is common in rickettsial diseases, and some cases diagnosed as murine typhus may have been caused by other rickettsial species. A third limitation is use of a single serum sample as a diagnostic criterion, which although used in most clinical case series is not rigorous, and previous exposure to pathogens as the cause of seroreactivity cannot be completely ruled out. However, the relatively high IgM cutoff point and the absence of relevant differences between patients given a diagnosis by using 1 sample and those with confirmed seroconversion support the data presented.

Murine typhus is a major cause of FID in the Canary Islands. Complications are frequent, especially in the elderly, usually with renal, hepatic, respiratory, or central nervous system involvement. These results should help raise awareness among physicians about the need to identify cases earlier, start treatment promptly, and thus improve clinical outcomes.