Murine Typhus in 5 Children Hospitalized for Multisystem Inflammatory Syndrome in Children

We describe a case series of 5 children managed in a single institution in Southeast Texas during May 2020. All 5 presented with fever, were admitted to the pediatric hospital medicine service for further evaluation for MIS-C and were discharged with a diagnosis of murine typhus. The local institutional review board approved the study, and parents provided verbal consent to be included. Basic demographic information, clinical features, laboratory and imaging findings at time of presentation, and clinical course were extracted from the electronic medical record. The postdischarge course was obtained from phone calls to parents and the electronic medical record when available.

A 17-year-old boy presented with a 5-day history of fever associated with rash (Table 1). Because of high inflammatory markers (Table 2), the rheumatology department was consulted to evaluate for MIS-C and agreed an infectious etiology was more likely. His fever resolved and he clinically improved with empirical doxycycline.

A 16-year-old girl presented with 10 days of fever accompanied by headache and abdominal tenderness (Table 1). In addition to significant elevation of liver enzymes (Table 2), she had elevated prothrombin time, partial thromboplastin time, and creatinine kinase. She showed clinical and laboratory improvement with doxycycline.

A 12-year-old boy presented with 6 days of fever associated with headache and rash (Table 1). The initial examination was notable for dry lips, conjunctival injection, and an eczematous rash. Laboratories revealed mildly elevated C-reactive protein (CRP) as well as hyponatremia and thrombocytopenia (Table 2). Additionally, he had sterile pyuria and mildly elevated prothrombin time. He was treated with empirical doxycycline, and the fever resolved over the next 24 hours.

An echocardiogram performed as part of his initial evaluation revealed mild dilation of the left anterior descending and right coronary arteries, which, despite resolution of fever by the time it was resulted, prompted treatment with intravenous immunoglobulin and aspirin for atypical KD versus MIS-C. He developed a doxycycline-associated rash and completed treatment against R typhi infection with azithromycin. His coronary-artery dilation resolved on subsequent outpatient echocardiograms.

A 10-year-old girl presented with 10 days of fever associated with headache, rash, and abdominal tenderness on examination (Table 1). Laboratory findings are noted in Table 2. Her fever and headache resolved with empirical doxycycline treatment.

An 8-year-old girl presented with 9 days of fever accompanied by headaches. Her examination was largely normal aside from fever. Her mother had known coronavirus disease 2019 exposure. Laboratories revealed leukocytosis with an elevated absolute neutrophil count (Table 2). Multiple SARS-CoV-2 polymerase chain reaction results were negative and antibodies were nonreactive. Her fever resolved with empirical doxycycline.

Murine typhus, also known as endemic typhus, is an inflammatory vasculitis caused by infection with R typhi. The disease is present along the southeastern Gulf Coast, Southern California, and Hawaii and is transmitted to humans via bites from fleas with rodent reservoirs. However, many patients will not report known flea exposure,^8,9  as seen in 2 children in our series. Although thought of as a rare disease, murine typhus incidence is underestimated⁸  and rising in South Texas.⁹  Anecdotally, we have noted a particular increase in hospitalized children in spring 2020, including the period of this study. Although it is yet unclear whether this represents a true increase in prevalence, we postulate that school closures and social distancing may have increased exposure because of time spent outdoors or with pets. Alternatively, parental concern for SARS-CoV-2 infection could mean children with symptoms are presenting to care and being referred or admitted to the hospital more frequently because of provider concern for MIS-C.

Murine typhus disease is classically described as triad of fever, headache, and rash, but this triad only occurs in 41.6% to 42.7% of children.^8,9  Instead, there is a broad spectrum of disease, as demonstrated in this series; 3 of 5 had the triad, and other symptoms included abdominal pain, vomiting, cough, myalgias, fatigue, and malaise, all of which have been previously described.^7–10  The rash in murine typhus can vary.^8–10  Laboratory findings are nonspecific, as is reflected in this case series, and most commonly include transaminemia, lactate dehydrogenase elevation, hypoalbuminemia, elevated inflammatory markers, hyponatremia, anemia, and thrombocytopenia.^7–10  The pattern and degree of derangements also vary. For example, although transaminases are usually mildly elevated, in a review of cases, one-quarter of the patients had normal levels, and one-quarter had a significant (greater than fourfold) increase in levels,⁸  as seen in cases 2 and 4.

After presenting with fever, a range of laboratory tests was undertaken in these children, in accordance with Centers for Disease Control and Prevention recommendations to evaluate for MIS-C.¹  In these 5 children, CRP and D-dimer were elevated in all 5, the erythrocyte sedimentation rate and lactate dehydrogenase were elevated in 4, ferritin was elevated in 1, and procalcitonin and fibrinogen were not elevated. Interleukin-6 was not assessed. This series serves as a reminder that these laboratory findings are nonspecific and, although rarely obtained, would likely be abnormal in a variety of non–MIS-C infectious and inflammatory conditions.

Additionally, we postulate that mild cardiac involvement in non–MIS-C and non-KD diagnoses may also be underappreciated. Although troponin and B-type natriuretic peptide levels in our selected cases were within normal range, the patient in case 3 had mild coronary-artery dilation. Although KD cannot be definitively ruled out in this child, his fever resolved on doxycycline alone, and he had hyponatremia and thrombocytopenia, features not typical in KD. Indeed, although true coronary aneurysms are specific to KD, there is evidence that coronary-artery dilation can result from other febrile illnesses because of interleukin-6 cytokine storm.^11,12  Again, because cardiac enzymes and imaging have, until now, not been routinely included in the evaluation of febrile illness, the specificity of mild abnormalities is unclear. During this time of uncertainty and heightened awareness of SARS-CoV-2, interpretation of results may contribute to premature closure and, possibly, overtreatment for MIS-C.

Untreated, the fever in murine typhus persists for 12 to 24 days.⁸  However, most patients defervesce within 72 hours on appropriate antibiotics.^8,9  In describing our cases, it is important to note that the gold standard to diagnose murine typhus is convalescent antibody titers 2 to 4 weeks apart to show a fourfold increase in anti–R typhi immunoglobulin G (IgG), which differentiates present and past infection.⁷  Because antibody testing takes several days to result, providers at our institution rely on surrogate factors for diagnosis, namely, rapid response to doxycycline, as seen in all 5 of these cases. Convalescent titers were not feasible because of the retrospective nature of this study. However, elevated immunoglobulin M (IgM) in all 5 children, which resulted after discharge, supports acute infection with R typhi, and none had a recurrence of symptoms. Brief observation periods on empirical antibiotics were sufficient to make a presumptive diagnosis of murine typhus, avoiding intravenous immunoglobulin and other anti-inflammatory therapies in 4 out of 5 cases.

With this case series of children evaluated for MIS-C, we emphasize the wide differential diagnosis hospitalists must maintain when approaching a child with prolonged fever and evidence of systemic inflammation. In addition to common viral, bacterial, and noninfectious diagnoses, a range of regional endemic rickettsial and parasitic infections must be considered as alternative diagnoses to MIS-C, including murine typhus. Many of these diseases cannot be reliably differentiated from MIS-C on presentation, and, as community exposure to SARS-CoV-2 grows, hospitalists should be prepared to admit febrile children with evidence of systemic inflammation for brief observation periods to evaluate for MIS-C. When appropriate, empirical treatment of common or even uncommon infectious diseases, such as murine typhus, may avoid overdiagnosis and overtreatment of MIS-C as well as improve patient outcomes.