Rare case of Capnocytophaga canimorsus meningitis in a man without risk factors

Abstract

Capnocytophaga canimorsus meningitis is an uncommon but potentially serious cause of meningitis, which is considered particularly rare in healthy and immunocompetent individuals. We present a case of C. canimorsus meningitis in a young, immunocompetent patient which was acquired following a dog bite. We review the literature and propose that underdiagnosis of this condition is likely. To avoid misdiagnosis, and thus improper management, clinicians should ensure that they identify animal exposure in all meningitic patients, and adopt a higher clinical suspicion in the absence of classical risk factors.

Background

Here, we present a rare case of Capnocytophaga canimorsus infection manifesting as meningitis. Unusually, our patient is young, has no comorbidities and is presumed to be immunocompetent. We consider the following question—is this disease underdiagnosed in those without classical risk factors?

Case presentation

An immunocompetent and healthy 35-year-old man presented to the emergency department in April 2019 with a 5-day history of headache, neck stiffness and generalised myalgia. Examination revealed a temperature of 38.5°C, photophobia and meningism. The patient was immediately started on intravenous ceftriaxone 2 g two times per day and aciclovir 10 mg/kg three times per day to treat suspected meningitis.

The patient was subsequently referred to the medical team and a more detailed history was taken, revealing that he had sustained a dog bite from his Staffordshire Bull Terrier 6 days earlier. He confirmed that his dog had received the minimum standard of vaccinations as a puppy,¹ including protection against canine parvovirus, canine distemper virus, canine adenovirus and leptospirosis. The rabies vaccine was not given as this is not a requirement for dogs in the UK.² The patient had received all of his childhood vaccinations and the last tetanus dose he received was at 21 years of age.

Investigations

Initial blood tests showed evidence of inflammation with a white blood cell (WBC) count of 14.69×10⁹/L, neutrophil count of 11.94×10⁹/L and a serum C reactive protein (CRP) level of 78 mg/L. A lumbar puncture was performed revealing cerebral spinal fluid (CSF) with a turbid appearance. Bacterial meningitis was confirmed by the CSF microscopy, which showed a white cell count of 1450 cells/µL (of which 60% were polymorphs) and a positive gram stain revealing fusiform-shaped gram-negative rods. Similar gram-negative bacilli were subsequently identified in blood cultures. Intravenous ceftriaxone was continued and aciclovir was stopped after discussing these results with the microbiologist.

The patient’s blood investigations the following day showed a WBC count of 24.82×10⁹/L, a neutrophil count of 21.35×10⁹/L and a serum CRP level of 282 mg/L. A routine HIV test was performed which was negative. Blood and CSF were incubated on blood agar plates for a standard period of 2 days and 5 days, respectively, but no organism was grown. On the basis of advice of the microbiologist, broad-range 16s rRNA PCR was performed and C. canimorsus was detected, confirming a diagnosis of C. canimorsus meningitis.

Given that our patient was young and had no comorbidities, with no history of recurrent infections, functional asplenia was not suspected and screening tests were not performed.

Treatment

Intravenous ceftriaxone 2 g two times per day was continued and the patient clinically improved. On reviewing the case notes, we discovered that the patient did not receive prophylactic tetanus toxoid on admission. This would usually be part of standard care for any animal bite presenting to our emergency department. One possible explanation for this is that the clear history, signs and symptoms, and early CSF results confirming gram-negative meningitis, pointed towards a zoonotic bacterial infection such as Capnocytophaga or Pasteurella. This oversight is an important learning point for our department.

The patient did not receive rabies postexposure prophylaxis. The UK is considered ‘no risk’ for animals and ‘low risk’ for bats, meaning prophylactic treatment is not recommended after dog bites.³

Outcome and follow-up

The hospital at home team facilitated discharge on day 9 of admission to complete a 21-day course of intravenous ceftriaxone at home. The patient made a full recovery by the end of his antibiotic treatment, and there was no routine follow-up.

Discussion

C. canimorsus is a fusiform, slow-growing, gram-negative bacillus. It was first described by Bobo and Newton⁴ in 1976 as a cause of septicaemia and meningitis following a dog bite. Since then, eight further species of Capnocytophaga have been identified,⁵ of which six inhabit the oral cavity of humans and two inhabit the oral cavity of cats and dogs.

C. canimorsus is found in cat and dog saliva and causes clinically significant infections in humans.⁶ Disease can manifest as rash, septicaemia, endocarditis and occasionally meningitis.⁷ Mortality rates for C. canimorsus septicaemia have been found to be as high as 31%,⁸ though more recent studies demonstrate a lower figure.⁹ Fortunately, infection appears to be rare, with one survey from the Netherlands reporting an infection rate of 0.67 per 1 million population.⁹

Infection is thought to be uncommon in healthy and immunocompetent individuals⁷—a notion that is concordant with its low prevalence in the general population. Indeed, a major classical risk factor for disease is overt immunocompromise, with splenectomy⁸ and haematological malignancy¹⁰ being often reported. A background of alcoholism is also commonly seen.⁸

Diagnosing infection can be difficult due to the poor growth of C. canimorsus using standard culturing techniques. The use of a trypticase-soy base in routine sheep blood agar has been posited as one explanation for this.¹¹ More successful culturing techniques involve using media supplemented with cysteine¹² and incubating blood or chocolate agars in a carbon dioxide-rich environment at 35°C. Molecular testing using 16S rRNA PCR for the diagnosis of culture-negative bacterial infections can also be performed if available, as per recommendations made by the UK Standards for Microbiological Investigations.¹³

We reviewed the literature from 1986 to 10 June 2020,^{7 14–37} and identified 31 cases of confirmed C. canimorsus meningitis (table 1). Median age at presentation was 61 years (ranging from 12 days to 79 years), and 81% of patients (25/31) were male.

Table 1.

Cases of Capnocytophaga canimorsus meningitis ientified in the literature from 1986 to 10th June 2020

Case no.	Year of publication	Reference	Age/sex	Immunocompromised	Other risk factors	Animal exposure	Complications
1	1986	14	63 years/M	No	None	Dog bite	None reported
2	1987	15	4 months/M	No	None	Dog bite	None reported
3	1990	16	75 years/F	Yes—splenectomy	None	Dog bite	None reported
4	1994	17	57 years/M	No	None	Dog bite	None reported
5	1996	7	54 years/M	No	Alcoholism	Dog bite	None reported
6	1996	7	57 years/M	No	None	Dog bite	None reported
7	2003	18	45 years/M	No	Alcoholism	Dog bite	None reported
8	2003	19	12 days/M	No	None	Dog bite	None reported
9	2006	20	54 years/M	No	Alcoholism	Dog bite	None reported
10	2006	21	65 years/M	No	None	Dog bite	None reported
11	2006	22	65 years/F	No	None	Dog owner	None reported
12	2007	23	69 years/M	Yes—prednisolone	COPD	Dog bite	None reported
13	2007	23	58 years/M	No	None	None reported	None reported
14	2009	24	64 years/M	No	None	Dog bite	Hearing loss
15	2012	25	66 years/M	No	Alcoholism	Dog bite	Hearing loss
16	2012	25	67 years/F	Yes—splenectomy	None	Dog owner	Hearing loss
17	2012	25	79 years/M	No	None	Dog bite	Hearing loss
18	2013	26	56 years/M	No	Alcoholism	Dog bite	None reported
19	2015	27	78 years/M	No	None	Dog bite	None reported
20	2015	27	37 years/M	No	Alcoholism	Dog bite	None reported
21	2015	27	60 years/M	No	None	Unknown	None reported
22	2016	28	52 years/M	No	COPD	Dog lick	None reported
23	2017	29	49 years/M	No	None	Dog bite	None reported
24	2017	30	70 years/F	No	None	Dog bite	None reported
25	2018	31	60 years/M	No	Alcoholism	Dog lick	Hearing loss
26	2018	32	61 years/M	No	Alcoholism	Dog bite	None reported
27	2019	33	60 years/M	No	Alcoholism	Dog lick	Hearing loss
28	2019	34	71 years/F	No	Lung cancer	Dog owner	None reported
29	2020	35	67 years/M	Yes—chronic lymphocytic leukaemia on ibrutinib therapy	None	Cat owner	None reported
30	2020	36	77 years/F	No	COPD	Dog lick	None reported
31	2020	37	74 years/M	No	Type II diabetes	Dog bite	None reported

COPD, chronic obstructive pulmonary disease.

Of the 31 cases reviewed, only 13% (4/31) were overtly immunocompromised; two cases with splenectomies, one case with haematological malignancy and one case on immunosuppressant therapy. The most prevalent risk factor was alcoholism, observed in 29% (9/31) of cases. A further 16% (5/31) had some other underlying health condition; three cases with chronic obstructive pulmonary disease, one case with diabetes and one case with lung cancer. The remaining 42% (13/31) had no reported risk factors (figure 1).

Figure 1.

Risk factors identified in the reviewed literature.

A dog bite or lick precipitated infection in 81% (25/31) of cases. Of the remainder, 13% (4/31) did not recall any such contact, but did have regular exposure to cats or dogs; one of these cases developed as an iatrogenic complication of routine myelography in which both the patient and the performing clinicians owned several cats and dogs as household pets.²² In 6% (2/31) of cases exposure to a cat or dog was either unknown or unreported (figure 2).

Figure 2.

Causes of infection identified in the reviewed literature.

All 31 cases reviewed survived infection, giving a 0% mortality rate. The only long-term sequela reported was hearing loss, seen in 19% (6/31) of patients. When interpreting these reported outcomes, it is important to consider the limitations of our literature search—specifically, that cases with poor outcomes are often not published. These figures may not accurately reflect clinical practice, in which C. canimorsus meningitis, and disseminated infection in particular,⁸ may have a poorer prognosis.

Overall, our review of the literature demonstrates the following: in 87% of cases of C. canimorsus meningitis patients were immunocompetent, in 71% of cases there was no history of alcohol excess and in 42% of cases no risk factors were identified at all.

These findings were not what we expected. Atypical infections tend to present in immunodeficient or comorbid patients, and C. canimorsus has classically been considered no exception. However, these statistics suggest that this condition is not limited to asplenics and alcoholics, and infection among the immunocompetent and healthy may be more prevalent than previously thought.

In the context of its apparent rarity, this raises an obvious question—are we underdiagnosing C. canimorsus meningitis? There are certainly a number of factors that support this possibility. First, given the slow-growing and fastidious nature of C. canimorsus, diagnosis is difficult using routine culturing. Thus, unless C. canimorsus meningitis is suspected, the additional appropriate tests may not be done and the pathogen may not be identified. Indeed, in our case, CSF and blood were only sent for molecular studies after a second and more thorough history was taken to elucidate a recent dog bite. Second, C. canimorsus is well-treated by current empirical antibiotic regimens for meningitis, as suggested by the 0% mortality rate in the literature we reviewed. If CSF samples are not obtained before or shortly after the initiation of antibiotic therapy, cultures may come back falsely negative.³⁸ Third, regular exposure to cats and dogs is very common; the Pet Food Manufacturers’ Association estimated that 39% of UK households owned a cat or dog in 2020.³⁹

It seems that underdiagnosis of C. canimorsus meningitis is a distinct possibility. These missed diagnoses may be contributing towards the large proportion of cases of meningitis—almost 70% in one study⁴⁰—in which no aetiological agent is identified.

We make the following suggestions to avoid underdiagnosis in future clinical practice:

1. A thorough history should be taken in all meningitic patients to identify possible animal exposure.

2. C. canimorsus meningitis is considered in all patients with a history of animal exposure, regardless of the presence of classical risk factors.

Patient’s perspective.

I had a good experience in hospital and appreciate the care I got. I was particularly happy that I was able to be discharged early to complete my antibiotics at home.

Learning points.

• Prophylactic tetanus toxoid should be given as part of animal bite management.

• Underdiagnosis of Capnocytophaga canimorsus meningitis is likely.

• C. canimorsus meningitis should be suspected in all meningitic patients with a history of animal exposure. A thorough history should be taken to identify this.

• C. canimorsus meningitis should be considered even in the absence of classical risk factors.