Rat bite fever (RBF) is a rare zoonotic infection. In North America, RBF is caused primarily by Streptobacillus moniliformis while Spirillum minus infection, also known as sodoku, is more common in Asia.1 Other Streptobacillus species such as S. felis, S. ratti and S. notomytis can cause symptoms similar to RBF.2S. moniliformis is a pleomorphic, Gram-negative rod that requires microaerophilic conditions to grow, often making microbiologic diagnosis difficult.1 Rats are the dominant reservoir and reportedly S. moniliformis can colonize the oro-respiratory tract of 10%–100% of domesticated and laboratory rats and 50%–100% of wild rats.1,3,4 Historically, RBF has affected children and young adults in poverty.1 However, recently, there has been a shift in demographics with more cases reported in rat handlers (pet stores, lab personnel) or pet rat owners.3,5–7 The true incidence of RBF is unknown.5
The classic manifestations of RBF are a triad of fever, rash and joint involvement. Rash usually begins 2–4 days after the onset of fever and can be variable in appearance such as maculopapular, petechial or purpuric and favors extremities.1,6 Polyarthritis and septic arthritis can develop in up to 50% of patients.1,5,8 Complications can include endocarditis, myocarditis, pyogenic arthritis, acute kidney injury, soft tissue or brain abscesses, and hemophagocytic lymphoproliferative (HLH) disease.9,10 Untreated, the mortality rate can be up to 10%.1,8,11 Variable presentations of RBF outside of the classical triad have been reported in the literature in both children and adults.7,11–14 Hence, a high degree of suspicion and careful history taking are warranted in these cases.
In this report, we describe the clinical and laboratory characteristics of pediatric patients with classic and nonclassic presentations of RBF at a free-standing, quaternary care children’s hospital to elucidate the variable presentations of RBF in children.
We conducted a retrospective chart review of patients admitted with RBF from 2008 to 2020 at our institution. Cases were identified using ICD-9 and ICD-10 codes for RBF. Pediatric patients 0–18 years of age with confirmed and suspected RBF were included in this study. Confirmed cases were defined as patients presenting with signs and symptoms consistent with RBF along with isolation of S. moniliformis from either blood, synovial fluid cultures, identification by 16S ribosomal RNA gene sequencing (PCR) performed on synovial fluid or by plasma metagenomic next generation sequencing (pmNGS) testing. S. moniliformis isolated from blood culture (BCx) grows fine on routine blood agar; colonies are visualized after 24–28 hours of incubation at 35°C under aerobic conditions. It is further confirmed by 16S pyrosequencing. Suspected cases were defined as cases with exposure history, signs and symptoms consistent with RBF but without microbiologic confirmation. Classic cases presented with the triad of clinical findings, including fever, rash and joint involvement, while nonclassic cases presented with 1 or 2 of these clinical findings. We reviewed demographics, clinical features, hospital course, laboratory and imaging results, management and follow-up encounters when available within our electronic medical record system. The Institutional Review Board approved the study.
Twelve patients were identified during the study period: 8 with confirmed and 4 with suspected RBF. The median age at presentation was 5.5 years (range: 1.2–17.7 years). Seven (58.3%) were male. All patients were admitted to the inpatient ward. Only 4 of the 12 patients (33.3%) presented with the classic clinical triad of symptoms (Table 1, cases 1–4). Of these 4, 2 were confirmed cases and 3 had rat exposures. Eight patients (66.6%) did not present with the clinical triad. Of these, 6 (75%) were confirmed cases (Table 1, cases 5–10), and 6 (75%) had known rat exposure (Table 1, cases 6–10, 12).
TABLE 1. -
Cases of RBF at Texas Children’s Hospital, 2008–2020
||Month/Year of Infection
Classic triad cases (1–4) of RBF
||Petechial and pupuric rash to BL feet, soles, petechiae on L palm
||Shoulder, BL wrists
||IV PCN × 5 d + PO PCN × 7 d
||Petechiae below BL knees and in glove-sock distribution, pustular rash on R palm
||BL shoulders, BL knees, L wrist, R jaw, L ankle
||Unknown bite 2 wks prior
||XR, MRI; focal myositis on deltoid
||IV PCN × 10 d + PO PCN × 7 d, readmit: given additional PO PCN × 7 d
||XR, US: ankle effusion
||IV PCN × 7 d + PO PCN × 7 d
||Face, abdomen, back, extremities (per report)
||L knee, R shoulder, L hip
||MRI; pelvis/spine—abnormal signal intensity of vertebral bone marrow
||IV PCN × 10 d + PO PCN × 21 d
Nonclassic triad (5–12) cases of RBF
||XR, US, MRI; knee effusion, synovial thickening
||IV PCN × 7 d + PO PCN × 2 d
||R knee, L hip
||2 rats, bearded dragon, gecko, frog (pets)
||+ PCR (SF)
|XR, US, MRI; knee effusion
||Improvement on naproxen; PO amoxicillin × 10 d after PCR resulted
||+ PCR (SF)
|XR, MRI; knee effusion, diffuse synovitis
||PO amoxicillin × 30 d
(−) PCR (SF)
|XR, US, MRI; knee effusion, bursitis, myositis
||PO PCN × 21 d
||Pig and rat (pets)
||+ PCR (SF)
|CT, MRI; knee effusion, synovitis
||IV PCN × 4 d + PO amoxicillin × 21 d
||3 pet rats
||IV ciprofloxacin × 5 d + PO ciprofloxacin × 7 d
||BL ankles, BL wrists
||XR, MRI; synovitis and joint effusion of hand joints
||PO PCN × 7 d
||Erythematous macules with confluence on BL LE, petechiae over feet and thighs
||Rats seen inside the home
||IV PCN × 3 d + PO PCN × 7 d
*Microbiologic confirmation: isolation of S. moniliformis from BCx, PCR and pmNGS.
†Confirmed cases of RBF.
Abbreviations: BCx, anerobic blood culture; BL, bilateral; IV, intravenous; L, left; LE, lower extremities; MRI, magnetic resonance imaging; PCN, penicillin; PCR, 16S ribosomal RNA gene sequencing; pmNGS, plasma metagenomic next generation sequencing; PO, oral; R, right; RBF, rat bite fever.; SF, synovial fluid; SFCx, synovial fluid culture; US, ultrasound; XR, radiograph.
Duration of fever ranged from 5 to 11 days (mean: 7 days) in 3 patients with the classic triad, and the remaining 1 presented with 30 days of intermittent fever (Table 1). Five nonclassic cases had fever, ranging from 1 to 10 days (mean: 2.8 days). Rash at presentation was variable in appearance and location (Table 1). Of the 10 patients with joint involvement, knees were the most common joint involved, seen in 7 of 10 patients (70%), hips were involved in 2 of 10 patients (20%). Five of 10 patients (50%) had polyarthritis (Table 1, cases 1, 2, 4, 6, 11).
The average length of hospital stay for classic and nonclassic cases was 7.8 days (range: 4–11 days) and 4.4 days (range: 2–7 days), respectively.
Laboratory and Imaging Findings
Among patients with classic RBF, the mean white blood cell count (WBC) was 12.2 × 103/µL (range 13.5–16.5 × 103/µL) compared with those with nonclassic RBF, whose mean WBC count was 15.1 × 103/µL (range 10.2–23 × 103/µL). There was no difference in the mean erythrocyte sedimentation rate and C-reactive protein for those with classic versus nonclassic RBF. All 12 patients had at least 1 anerobic BCx drawn. Four of the 8 (50%) BCx in the confirmed cases yielded S. moniliformis, of which only 2 had the classic triad. The average time to BCx positivity was 67 hours (range: 40–126 hours).
Four of 10 patients with joint symptoms had synovial fluid obtained, all with confirmed RBF. PCR testing was sent on all 4 patients, 3 had a positive PCR for S. moniliformis (Table 1, case 6–9). Three patients had a WBC count of > 20,000/µL (range: 6500–94,020/µL, mean: 46,310/µL), all with a neutrophil predominance.
MRI was the most common imaging modality obtained (7/10 patients). Imaging was notable for joint effusions, synovitis and myositis (Table 1).
Treatment and Outcome
Antibiotic (Abx) choice and length of therapy (LOT) varied among patients (Table 1). All 4 classic cases received intravenous (IV) penicillin (PCN) and were discharged home on oral (PO) PCN. One classic case required readmission due to recurrence of polyarthritis and rash and was treated with additional doses of PCN. The mean total LOT for classic cases was 20.2 days (range: 12–31 days).
Among the 8 nonclassic, only 3 received IV PCN as an empiric therapy during hospitalization. Four of the 8 suspected cases were discharged home on PO PCN, amoxicillin (n = 2), and ciprofloxacin (n = 1, known PCN allergy). One nonclassic case presenting with arthritis alone did not receive Abx at discharge, the patient improved on naproxen, was later found to have PCR + for S. moniliformis and was prescribed amoxicillin for 10 days. The mean total LOT for nonclassic cases was 15.5 days (range: 1–30 days).
Nine (70%) patients had documented full recovery as documented in the electronic medical record either at discharge or at a follow-up visit within a month after discharge. One patient had a residual limp 3 weeks after discharge, 1 patient presented to the ER with nonspecific complaints that were thought to be unrelated to RBF, and no follow up was available for 1 patient. In 4 patients presenting primarily with joint complaints, a rheumatology evaluation was performed before the diagnosis of RBF was confirmed.
Our retrospective review highlights challenges in diagnosing and managing RBF in children. First, the presentation can be variable and nonspecific since only few children might present with classic symptoms of fever, rash and polyarthritis. Furthermore, the growth and isolation of S. moniliformis are difficult using routine culture medium; hence, delayed diagnosis and treatment can occur. Finally, emerging literature shows that other streptobacillus species other than S. moniliformis can cause clinical findings consistent with RBF; hence, a thorough animal exposure history is vital when managing these patients.
Nine of our patients (75%) had a known rat exposure, but only 3 patients specifically reported a rat bite preceding the onset of symptoms. Rat bites are not always documented in confirmed cases, but infection can occur from scratches and mucous membrane exposure (kissing)—which is a similar finding in our study.6,15 Cases of RBF have also been reported without an identified rat exposure; as was seen in 2 of our 12 patients with confirmed RBF.6,15
Although most pediatric studies report the classic presentation of RBF with fever, rash and polyarthritis, presentation with 1 or 2 of those symptoms were more common in our study among confirmed and suspected cases.3,12,13,16–18 None of our patients had complicated disease (endocarditis, acute kidney injury, HLH, abscesses) as reported in other studies.9,10
As with most pediatric studies, knees were the most commonly involved joint in our population, followed by wrists and ankles.15,17,19 It is important to note that hips are more commonly involved in RBF in children compared with adults.15,17,19 Two (20%) of our 10 patients with joint symptoms had hip involvement. Elevated WBC count and neutrophilic predominance in the synovial fluid of RBF patients have been described in studies with negative cultures.15,17 With molecular diagnostic testing, the confirmation of RBF diagnosis is now higher and timelier than in the past.15 BCx generally have a lower diagnostic yield, especially if only aerobic bottles are collected. In this review, only partial 16S sequencing for PCR results was available. This is a limitation because depending on the partial sequence, S. moniliformis can be indistinguishable from other streptobacillus species, such as S. notomytis. However, S. notomytis was first described in 2015 to our knowledge and would not been identified in our cases before this year.20 We highly recommend PCR testing of synovial fluid in pediatric patients who present with classic and nonclassic signs of symptoms of RBF when there is a history of rat exposure. If joint effusion is not present in these patients and synovial fluid cannot be obtained, pmNGS along with BCx (sheep blood agar plates after 24–48 hours of incubation at 35°C) should be considered.
The variability in empiric Abx choice and LOT in our study is worth noting. We believe the low occurrence of RBF, delay in microbiologic confirmation, and broad differential diagnosis played a role in the variability and LOT. However, once the diagnosis of RBF was established in our population, all but 1 patient was transitioned to oral PCN or amoxicillin as definitive therapy. Clinicians should be familiar with the variable presentation of RBF and consider appropriate therapy when rat exposure is elicited even without the classic triad of symptoms.
PCN is the treatment of choice for RBF. For uncomplicated cases of RBF, children should initially receive IV PCN and transition to oral therapy once clinically improved to complete a 14-day course. In children who are PCN-allergic, a 14-day course of doxycycline is recommended. For complicated disease, children should receive IV PCN or IV ceftriaxone for 4 weeks.21
RBF is a rare but serious zoonotic infection that requires a high degree of suspicion and prompt diagnosis. Our review highlights the variability in the presentation and treatment of RBF in children. Due to low diagnostic yield of BCx, PCR testing of synovial fluid and pMNGS testing should be sent when possible. Treatment with PCN should be initiated as soon as clinical suspicion of RBF is made while awaiting microbiologic confirmation.
The authors would like to acknowledge Dr. James Dunn for his microbiologic input in the manuscript.
1. Elliot S. Rat bite fever and Streptobacillus moniliformis
Clin Microbiol Rev. 2007;20:13–22.
2. Matt U, Schmiedel J, Fawzy A. Infection in a young immunocompetent male caused by Streptobacillus felis
, a putative zoonic microorganism transmitted by cats. Clin Infect Dis. 2020;72:1826–1829.
3. Miraflor AP, Davallow Ghajar L, Subramaniam S, et al. Rat-bite fever: an uncommon cause of fever and rash in a 9-year-old patient. JAAD Case Rep. 2015;1:371–374.
4. Andre JM, Freydiere AM, Benito Y, et al. Rat bite fever caused by Streptobacillus moniliformis in a child: human infection and rat carriage diagnosed by PCR. J Clin Pathol. 2005;58:1215–1216.
5. Kache P, Person M, Seeman S. Rat-bite fever in the United States: an analysis using multiple national data sources, 2001-2015. Open Forum Infect Dis. 2020;7:ofaa197.
6. Khatchadourian K, Ovetchkine P, Minodier P, et al. The rise of the rats: a growing paediatric issue. Paediatr Child Health. 2010;15:131–134.
7. Graves MH, Janda JM. Rat-bite fever (Streptobacillus moniliformis
): a potential emerging disease. Int J Infect Dis. 2001;5:151–155.
8. Glaser C, Lewis P, Wong S. Pet-, animal-, and vector-borne infections. Pediatr Rev. 2000;21:219–232.
9. Heaton P, Schultz NH, Helseth PH, et al. Streptobacillus moniliformis
with features of hemophagocytic lymphohistiocytosis identified by direct sequencing. Pediatr Infect Dis J. 2020;39:1131–1133.
10. Carr JP, McCloskey KM, Campbell J, et al. Fever, rash and acute kidney injury in a 10-year-old girl. Pediatr Infect Dis J. 2014;33:227, 231.
11. Stehle P, Dubuis O, So A, et al.Rat bite fever without fever. Ann Rhuem Dis. 2003;62:894–897.
12. Flannery DD, Akinboyo I, Ty JM, et al. Septic arthritis and concern for osteomyelitis in a child with rat bite fever. J Clin Microbiol. 2013;51:1987–1989.
13. Mackey JR, Melendez EL, Farrell JJ, et al. Direct detection of indirect transmission of Streptobacillus moniliformis
rat bite fever infection. J Clin Microbiol. 2014;52:2259–2261.
14. Adam JK, Varan AK, Pong AL, et al.; Centers for Disease Control and Prevention (CDC). Notes from the field: fatal rat-bite fever in a child—San Diego County, California, 2013. MMWR Morb Mortal Wkly Rep. 2014;63:1210–1211.
15. Dendle C, Woolley IJ, Korman TM. Rat-bite fever septic arthritis: illustrative case and literature review. Eur J Clin Microbiol Infect Dis. 2006;25:791–797.
16. Costa-Pinto J, Morley C, Hauser S. A case of rate bite fever in a 12-year-old boy. J Pediatr Child Health Care. 2017;53:84–86.
17. Wegner A, Look N, Haus B. Surgical management of multijoint septic arthritis due to rat-bite fever in a pediatric patient: a Case Study. Case Rep Orthop. 2017.
18. Theodoro MF, Morrison J, Hill B, et al. Arthralgia and vesicular rash in a toddler. Clin Pediatr (Phila). 2017;56:1365–1367.
19. Wang TK, Wong SS. Streptobacillus moniliformis
septic arthritis: a clinical entity distinct from rat-bite fever? BMC Infect Dis. 20077.
20. Eisenberg T, Ewers C, Rau J, et al. Approved and novel strategies in diagnostics of rat bite fever and other Streptobacillus infections in humans and animals. Virulence. 2016;7:630–648.
21. King K, Sexton D, Edwards M, et al. Rat Bite Fever. Available at: www.uptodate.com
. Accessed January 26, 2021.