A 71-year-old Caucasian female was found altered at home and transferred to an outside hospital. On presentation, she complained of fevers and severe headaches; however, she was uncertain of the length of her illness. Vitals signs included a temperature of 38.8 °C, pulse of 100 beats per minute, respiration 20 breaths per minute, and blood pressure of 182/86 mmHg. She had an altered mental status and meningismus including neck stiffness. There were no focal neurologic deficits or other examination findings.
Past medical history was remarkable for recently diagnosed lung cancer status-post lobectomy; she did not require adjunctive chemoradiation therapy. She also had a history of hypertension and chronic subdural hematoma. She denied diabetes, alcohol abuse, or prior splenectomy. She lived in Southern California and reported no recent travel history. She owned a dog and frequented a dog park with contact with several canines on a regular basis; she reported no dog bites.
Laboratory data on presentation were notable for a white blood count of 19,900 cells/mm3 (92% neutrophils), hemoglobin 13 g/dl, and platelets of 196 × 103/mm3. Lactate level was 2.4 mmol/l, creatinine was 1.0 mg/dl, and glucose was 109 mg/dl. Liver function tests and urinalysis were within normal limits.
A magnetic resonance image (MRI) with and without gadolinium showed left occipital and parietal acute infarcts without mass effect and stable small bilateral frontal subdural hematomas. A lumbar puncture was performed during the first 24 h of admission and revealed neutrophilic pleocytosis (590 white cells/mm3; 95% polymorphonuclear cells), red cell count of 2600 cells/mm3, low glucose of 12 mg/dl (reference range, 40–70 mg/dl), and elevated protein level of 413 mg/dl (reference range, 15–59 mg/dl). Gram stain did not show any organisms.
The patient was started on empiric antibiotic therapy with intravenous vancomycin and piperacillin-tazobactam prior to the lumbar puncture, but antibiotics were changed after the lumbar puncture revealed meningitis to intravenous vancomycin 500 mg every 8 h, ampicillin 2 g every 4 h, and ceftriaxone 2 g every 12 h; no steroids were administered.
Blood cultures (two sets) were drawn on admission using BD Plus Aerobic/F and BD Lytic/10 Anaerobic/F media. The first set was positive at approximately 4 days (98 h) and the second set at 4.6 days (110 h), in both the BD Lytic/10 Anaerobic/F media. The gram stain was reported as gram-negative rods with bacterial growth on plates by day 6. Matrix Assisted Laser Desorption/Ionization Time of Flight Mass Spectrometry (MALDI-TOF) was performed, but no identification was obtained. A Remel RapID ANA II test system was used for a biochemical identification, and Capnocytophaga was identified. The organism could not be successfully grown on culture for susceptibility testing.
Cerebrospinal fluid (CSF) cultures obtained on admission were sterile; however, antibiotics had been given prior to the lumbar puncture. Due to high suspicion of Capnocytophaga canimorsus as the causative organism given the patient’s exposure to dogs, a CSF specimen was sent to the University of Washington for multiplex broad-range bacterial polymerase reaction (PCR) testing, which was positive for C. canimorsus.
Antibiotic therapy was modified to meropenem 2 g IV q8 h. A transthoracic echocardiogram did not show vegetations and a chest, abdominal, and pelvic CT scan was unremarkable except for post-surgical finding consistent with prior lung lobectomy. The spleen appeared within normal limits. Repeat blood cultures after the initiation of antibiotics showed no growth.
The patient’s meningismus resolved during her hospital stay, and at the time of discharge her headache was significantly improved. She completed a 21-day total course of antibiotics and subsequently made a full recovery. She was educated on the potential infectious risks associated with dog ownership/exposure.
Review of the Literature
A total of 37 cases of Capnocytophaga canimorsus meningitis were identified including the current case (Table 1) [2, 4, 8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28]. Median age at presentation was 63 years (range, 12 days to 83 years) with a male predominance (28/37, 76%). While C. canimorsus meningitis has classically been characterized as a disease of immunocompromised patients, particularly patients who are asplenic, only 16% (6/37) of published cases occurred in patients with splenectomies. Five percent (2/37) had active steroid use. When summating all possible immunosuppressive states (e.g., medications or the presence of splenectomy or hematologic/autoimmune condition that impairs the immune system), 24% (9/37) of cases had one of these conditions: 5 splenectomy, 1 splenectomy and lymphoma, 2 steroid use, and 1 rheumatoid arthritis. Additionally, alcoholism was noted in 19% (7/37) and was the most common single medical condition identified.
Regarding animal exposure history, the majority of patients (59%; 22/37) reported a recent dog bite. The timing between the dog bite and presentation was a median of 6 days (range, 3 to 14 days). A smaller proportion reported non-bite dog exposures (22%; 8/37) and single cases of non-bite exposures to both dog and cat (3%, 1/37) and a cat bite (3%; 1/37). In addition, there was one reported case of indirect contact through two health providers who owned dogs. Overall, 11% (4/37) indicated no known animal contact prior to development of meningitis.
Presenting symptoms often included fever, headache, neck stiffness, altered mental status, and photophobia (Table 1). Other manifestations included rash in six cases, which was described as macular/papular in most cases; two cases had a rash that had a petechial/purpuric appearance. Other symptoms included seizures, myalgias, vomiting, fatigue, and hearing loss. Serum white blood cell (WBC) count was reported elevated in 11 (69%) of 16 cases that reported these data with a median WBC count of 13,500 cells/mm3 (range, 8000–25,000 cells/mm3).
CSF studies demonstrated a pattern consistent with bacterial meningitis with elevated CSF white counts of a median of 1024 cells/mm3 (range, 0–15,630 cells/mm3) with only one case with a normal white count. Most cases had neutrophilic predominance (81%, 26/32 cases with data). In addition, an elevated CSF protein was noted in 92% (22/24) of cases with a median value of (190 mg/dl) and low CSF/serum glucose ratio in all cases (median, 0.23). In cases that reported brain imaging, it was typically normal but in three cases (19%; 3/16 with imaging data), including the present case, acute infarcts were noted, and one additional case had cerebritis.
Blood cultures were found to be positive in 20 (54%) cases with a median growth time of 4 days (range, 2–9 days); 9 (24%) had negative blood cultures and 8 (22%) did not report results. CSF cultures were positive in 26 (70%) of cases with a median growth time of 5 days (range, 1–9 days). Due to the relatively long time for cultures to become positive, PCR is emerging as an important diagnostic technique. PCR was utilized to establish the diagnosis in the current case and overall in eight (22%) cases in this review.
Treatment involved extended courses of antibiotics with median treatment duration of 15 days (7–42 days). The most commonly utilized antibiotics were penicillin, ampicillin, and cephalosporins. Antibiotic susceptibilities were presented in eight cases in the literature with many cases lacking data, often because of insufficient growth of the bacteria on culture media. Overall, penicillins (including ampicillin), second- and third-generation cephalosporins, carbapenems (e.g., imipenem), and fluoroquinolones were susceptible in all cases reporting data for these antibiotics (Table 2).
Outcomes were generally favorable with only one (3%) death observed in the published cases; this death may have been unrelated to the infection as the patient died of a cardiac arrest 10 days after discharge and had known coronary artery disease. Overall, 19% of survivors (7/36) had chronic sequelae of the disease including four with hearing loss, one with chronic headaches/disorientation, one with extremity amputations and chronic neurologic abnormalities, and one with extremity amputations. Both patients who underwent amputations also had brain imaging showing acute infarcts. Of the seven patients with chronic sequelae, three had a history of alcohol abuse and one was asplenic. No clear relationship between antibiotic duration and adverse outcomes was noted.