Journal of Medical Microbiology (2004), 53, 245–248
Short Communication Correspondence
DOI 10.1099/jmm.0.05274-0
Capnocytophaga canimorsus endocarditis Jonathan A. T. Sandoe Department of Microbiology, General Infirmary at Leeds and University of Leeds, Leeds LS1 3EX, UK
Jonathan A. T. Sandoe
[email protected]
Received 2 April 2003 Accepted 1 December 2003
Capnocytophaga canimorsus is a fastidious, Gram-negative rod that forms part of the normal oral flora of dogs and cats. Known for its ability to cause fulminant sepsis following dog bites, particularly in asplenic patients or alcoholics, this bacterium is also an uncommon cause of endocarditis. This article reviews 12 cases of endocarditis caused by C. canimorsus. Mean age of patients was 53 years, with 78 % of cases occurring in males. Overall, a history of dog-bite was documented in four cases (33 %) and a further four (33 %) reported contact with dogs. Four (33 %) of the endocarditis cases had underlying cardiological risk factors and two abused alcohol, but none had had a previous splenectomy. Subacute presentation, often involving more than one hospital admission, was common, as were initially negative blood cultures. A variety of antibiotics was used, but penicillins were the most common therapy. Three (25 %) of the 12 endocarditis patients died.
Introduction
Results and Discussion
Capnocytophaga canimorsus, formerly known as dysgonic fermenter 2 (DF2), is a fastidious, Gram-negative rod that forms part of the normal oral flora of dogs and cats (Westwell et al., 1987). This organism has a fusiform appearance on Gram-staining and exhibits characteristic gliding motility. C. canimorsus is known for its ability to cause fulminant sepsis following dog bites, particularly in asplenic patients or alcoholics (Lion et al., 1996), but its ability to cause endocarditis is less well appreciated. This article describes the clinical features, treatment and outcome of previously published cases of endocarditis caused by C. canimorsus.
Table 1 summarizes the findings for 12 cases of C. canimorsus endocarditis that were published between 1977 and 2002. Sufficient information was available to categorize nine of the reported endocarditis episodes by using previously published criteria (Durack et al., 1994; Li et al., 2000). Two cases were excluded, as the species was not known or was thought not to be C. canimorsus (Montejo Baranda et al., 1984; Roig et al., 1996). Patients had a mean age of 53 years (range, 24–69 years). Male : female ratio was 4.5 : 1. Four (33 %) of the endocarditis cases had underlying cardiological risk factors: one patient was known to have aortic stenosis and a permanent pacemaker in situ, a murmur had been detected previously in another, one patient had an atrial myxoma and the other had had a previous aortic valve replacement. Five patients had been in good health prior to presentation; the others had underlying risk factors such as chronic lymphocytic leukaemia (n ¼ 1) and alcohol abuse (n ¼ 2). Previous splenectomy was not reported in any of the patients. A history of dog-bite was documented in four cases (33 %) and a further four (33 %) reported contact with dogs; in one case, the dog had licked a lesion on the patient’s leg. Three (25 %) of the 12 endocarditis patients died. Initial clinical signs and laboratory findings are summarized in Table 2. A mixture of different antibiotic regimens had been used in treatment, as summarized in Table 1. It is not possible to draw any conclusions regarding optimal therapy from such a small number of cases.
Methods and Patients Possible cases of C. canimorsus endocarditis were identified by using Medline. Search terms used were ‘endocarditis’, ‘DF2’, ‘Capnocytophaga canimorsus’ and ‘Capnocytophaga’. Further cases were identified from references of published reports. Cases that were due to other Capnocytophaga species or where the species was not stated were not included. Where reported information allowed, cases of endocarditis were categorized as definite, possible or rejected by using the modified ‘Duke criteria’ (Durack et al., 1994; Li et al., 2000). Diagnosis of endocarditis by using these criteria relies on pathological findings, such as demonstration of micro-organisms in resected heart valve tissue, or clinical criteria. Definite diagnosis based on clinical criteria depends on the presence of two major criteria, of one major criterion and three minor criteria or of five minor criteria, where major criteria are based on blood-culture results and echocardiographic findings and minor criteria include fever, predisposing heart conditions and immunological and vascular phenomena. Possible cases were those with one major and one minor criterion or three minor criteria (Li et al., 2000). Cases were classified as ‘acute’ if symptoms had been present for up to 2 weeks prior to diagnosis and as ‘subacute’ for any period longer than 2 weeks.
Abbreviation: DF2, dysgonic fermenter 2.
05274 & 2004 SGM Printed in Great Britain
C. canimorsus has been associated with a variety of conditions, including meningitis, fulminant septicaemia, cellulitis and endocarditis (Pers et al., 1996). There are well-described risk factors for infections caused by this bacterium, including dog-bite, previous splenectomy and alcohol abuse. A history of dog-bite has been reported in 43–47 % of cases (Brenner 245
J. A. T. Sandoe
246
Table 1. Summary of the main clinical features of 12 episodes of endocarditis caused by C. canimorsus Abbreviations: A, aortic valve; AVR, aortic valve replacement; CLL, chronic lymphocytic leukaemia; COPD, chronic obstructive pulmonary disease; E, endocarditis; M, mitral valve; MI, myocardial infarction; P, prosthetic; PPM, permanent pacemaker; T, tricuspid valve; NK, not known; NS, not stated. Case
Journal of Medical Microbiology 53
Age/sex
Previous cardiac Associated pathology conditions
Previous splenectomy
Valve affected
Animal exposure
Surgery
Antibiotics (duration, days)
Diagnosis
Outcome
Reference
1 2 3 4
NK
NK
NK
NK
NK
NK
NK
NK
NK
NK
64M
NK
NK
No
Dog bite
Yes No No No
NK
NK
A A M T+A
Dog bite
NK
E E E E
Died Recovered Recovered Died
Butler et al. (1977) Butler et al. (1977) Butler et al. (1977) Shankar et al. (1980)
5
59F
Atrial myxoma
CLL, steroids
No
T
NS
Yes
E
Died
6
39M
NK
Alcohol abuse
No
M
Dog contact
No
E
Recovered
Worthington et al. (1984) Archer (1985)
7
24M
Heart murmur
NK
No
A
Dog bite
No
MI + E
Recovered
8
47M
NK
Alcohol abuse
No
T
Dog contact
Yes
E
Recovered
9
56M
NK
NK
No
T
Dog contact
No
E
Recovered
10
52M
Aortic stenosis, PPM
NK
No
A
Dog bite
No
E
Recovered
Andersen & Pedersen (1992) Decoster et al. (1992)
11
69F
NK
COPD
No
T
None
No
E
Recovered
Kooter et al. (1999)
12
63M
AVR
NK
No
PA
Dog contact
Yes
E
Recovered
Ngaage et al. (1999)
NK NK
NK NK
Penicillin (5), erythromycin (5) Cephalothin + gentamicin (14) Ampicillin (42) + tobramycin (NS) Penicillin (28) Vancomycin (14) + gentamicin (14), penicillin (42) Penicillin (42) + gentamicin (NS) Penicillin (NS), aztreonam (35) Cefuroxime (7) + gentamicin (7), + flucloxacillin (7), then penicillin G (42) Ceftriaxone (28) + gentamicin (28), then penicillin G (28)
Newton & Sharma (1986) Niefield & Young (1988)
Definite Definite Definite Possible Definite Definite Definite Possible Definite Acute Subacute Subacute Acute Subacute Subacute Acute Acute Subacute NS
Although endocarditis caused by C. canimorsus is reported rarely in the literature, there are many reports of bacteraemia and sepsis in which a focus for the infection was never found. As with other fastidious, Gram-negative infections, the published record of cases may underestimate the true incidence of C. canimorsus infection. Since 2000, the Laboratory of Healthcare-associated Infections, Colindale, London, has received (on average) one isolate of C. canimorsus for identification per year, one of which was from a patient with endocarditis (M. E. Kaufmann, personal communication). Prolonged periods with negative blood cultures and previous hospital admission occurred in several of the previously reported cases of endocarditis in this review (Shankar et al., 1980; Worthington et al., 1984; Archer, 1985; Newton & Sharma, 1986). Blood cultures may be negative in the early stages of infection by this organism and subcultures can take up to 7 days incubation in CO2 before visible growth is apparent. Such slow growth explains how C. canimorsus may be a cause of apparently culture-negative endocarditis. C. canimorsus sepsis has also been associated with myocardial infarction (Newton & Sharma, 1986; Ehrbar et al., 1996) and post-mortem examination of two cases in a study from Denmark revealed myocarditis (Pers et al., 1996).
†x/7, no. days; x/12, no. months.
NS
NS
NS
NS
+
NK
*CRP, C-reactive protein; ESR, erythrocyte sedimentation rate.
NS
0 6/6 4/6 5/8 2/38 6/6 1/3 2/5
+ (+) + + + +
+ (+) (+) + +
NS
(+) + (+) (+) + + + +
NS
+ (+) (+) + + + +
NS
1, 2, 3 4 5 6 7 8 9 10 11 12
http://jmm.sgmjournals.org
et al., 1989; Gill, 2000). In this study, there was a history of dog-bite in 33 % of cases, but previous ‘contact’ with dogs in a further 33 %. However, given the number of dog owners in the population, it is uncertain whether ‘contact’ with a dog is a useful pointer for infection caused by this bacterium. Although Capnocytophaga infections can occur in previously healthy adults, immune-compromised, asplenic patients and alcoholics are at particular risk of severe infection (Lion et al., 1996). None of the patients with definite endocarditis in this study had a history of splenectomy, which may have contributed to the relatively high survival rate (75 %) in this group. It is noteworthy that previous splenectomy did not increase the incidence of infected vegetations in a rabbit model of DF2 endocarditis (Butler et al., 1985). Capnocytophaga infections have been described as predominantly occurring in men of 50–70 years (Lion et al., 1996; Pers et al., 1996); endocarditis would appear to be no exception.
10/7 12/7 14/7 5/12 5/7 7/7 4/12
NK NK NS
NS
NS NS NS
NS
(+) +
NS NS
(+D7) + (+) + (+D5) +
(+) (+) (+)
Acute/subacute Time to first positive culture† No. positive blood Petechial cultures haemorrhages on legs Haematuria Raised CRP/ESR* Murmur Anaemia Leukocytosis Pyrexia (>38 8C) Case
Abbreviations: , absent; +, present; D, day; NK, not known; NS, not stated. Parentheses indicate negative findings that subsequently became positive later in presentation.
Table 2. Summary of clinical features of C. canimorsus endocarditis
Duke classification (modified)
Capnocytophaga canimorsus endocarditis
Susceptibility testing is difficult and standardized methods are not available. Penicillin is considered to be the treatment of choice, but C. canimorsus has also been reported to be susceptible to imipenem, clindamycin, erythromycin, vancomycin, chloramphenicol, third-generation cephalosporins, rifampicin, quinolones and doxycycline in vitro (Gill, 2000). A variety of antibiotic regimens have been used successfully to treat C. canimorsus endocarditis (Table 1). It is clear that C. canimorsus can be a cause of acute or subacute, ‘culture-negative’ endocarditis and myocarditis. Clinical findings that are usually associated with endocarditis, such as heart murmur, fever and raised C-reactive protein, may be absent at presentation and infection may occur in individuals without previous cardiac pathology. A history of recent dog-bite in a patient presenting with clinical features of endocarditis should highlight the possibility of C. 247
J. A. T. Sandoe
canimorsus infection. Although cases are few, prolonged use of a penicillin appears to be appropriate therapy. Perhaps C. canimorsus should be considered among the HACEK (Haemophilus spp., Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella corrodens and Kingella spp.) group of fastidious, Gram-negative organisms that are capable of causing endocarditis. References Andersen, H. K. & Pedersen, M. (1992). Infective endocarditis with
involvement of the tricuspid valve due to Capnocytophaga canimorsus. Eur J Clin Microbiol Infect Dis 11, 831–832. Archer, S. L. (1985). Dysgonic fermenter 2 infection resulting in chronic
Kooter, A. J., Derks, A. & Vasmel, W. L. (1999). Rapidly progressive
tricuspid valve endocarditis caused by Capnocytophaga canimorsus infection in an immunocompetent host. Clin Microbiol Infect 5, 173–175. Li, J. S., Sexton, D. J., Mick, N., Nettles, R., Fowler, V. G., Jr, Ryan, T., Bashore, T. & Corey, G. R. (2000). Proposed modifications to the Duke
criteria for the diagnosis of infective endocarditis. Clin Infect Dis 30, 633–638. Lion, C., Escande, F. & Burdin, J. C. (1996). Capnocytophaga canimorsus
infections in human: review of the literature and cases report. Eur J Epidemiol 12, 521–533. Montejo Baranda, M., Acha Arrieta, V., Hernandez Almaraz, J., Perez Rodriguez, M., Antonana Oraeta, M. & Aguirre Errasti, C. (1984). Two
glomerulonephritis. Can Med Assoc J 132, 657–660.
cases of Capnocytophaga bacteremia, one with endocarditis. Can Med Assoc J 130, 1420.
Brenner, D. J., Hollis, D. G., Fanning, G. R. & Weaver, R. E. (1989).
Newton, N. L. & Sharma, B. (1986). Acute myocardial infarction
Capnocytophaga canimorsus sp. nov. (formerly CDC group DF-2), a cause of septicemia following dog bite, and C. cynodegmi sp. nov., a cause of localized wound infection following dog bite. J Clin Microbiol 27, 231–235. Butler, T., Weaver, R. E., Ramani, T. K., Uyeda, C. T., Bobo, R. A., Ryu, J. S. & Kohler, R. B. (1977). Unidentified gram-negative rod infection. A
new disease of man. Ann Intern Med 86, 1–5. Butler, T., Johnston, K. H., Gutierrez, Y., Aikawa, M. & Cardaman, R. (1985). Enhancement of experimental bacteremia and endocarditis
associated with DF-2 bacteremia after a dog bite. Am J Med Sci 291, 352–354. Ngaage, D. L., Kotidis, K. N., Sandoe, J. A. T. & Nair, R. U. (1999). Do not
snog the dog: infective endocarditis due to Capnocytophaga canimorsus. Eur J Cardiothorac Surg 16, 362–363. Niefield, S. & Young, E. J. (1988). Native valve endocarditis caused by dysgonic fermenter type 2 bacilli. Am J Med Sci 296, 69–70. Pers, C., Gahrn-Hansen, B. & Frederiksen, W. (1996). Capnocytophaga
caused by dysgonic fermenter (DF-2) bacterium after treatment with methylprednisolone and after splenectomy. Infect Immun 47, 294–300.
canimorsus septicemia in Denmark, 1982–1995: review of 39 cases. Clin Infect Dis 23, 71–75.
Decoster, H., Snoeck, J. & Pattyn, S. (1992). Capnocytophaga cani-
Roig, P. P., Lopez, M. M., Martin, C., Zorraquino, A., Sanchez, B., Navarro, V. & Merino, J. (1996). Bacteremia caused by Capnocytophaga
morsus endocarditis. Eur Heart J 13, 140–142. Durack, D. T., Lukes, A. S. & Bright, D. K. (1994). New criteria for
sp: presentation of 2 cases, one with endocarditis. Review of the literature. An Med Interna 13, 185–187 (in Spanish).
diagnosis of infective endocarditis: utilization of specific echocardiographic findings. Duke Endocarditis Service. Am J Med 96, 200–209.
Shankar, P. S., Scott, J. H. & Anderson, C. L. (1980). Atypical
Ehrbar, H. U., Gubler, J., Harbarth, S. & Hirschel, B. (1996). Capnocy-
endocarditis due to gram-negative bacillus transmitted by dog bite. South Med J 73, 1640–1641.
tophaga canimorsus sepsis complicated by myocardial infarction in two patients with normal coronary arteries. Clin Infect Dis 23, 335–336.
Westwell, A. J., Spencer, M. B. & Kerr, K. G. (1987). DF-2 bacteremia
Gill, V. J. (2000). Capnocytophaga. In Mandell, Douglas and Bennett’s
following cat bites. Am J Med 83, 1170.
Principles and Practice of Infectious Diseases, pp. 2441–2444. Edited by G. L. Mandell, J. E. Bennett & R. Dolin. Philadelphia, PA: Churchill Livingstone.
Worthington, M., Gleason, T., Pandian, N. G. & Daly, B. (1984).
248
Tricuspid valve myxoma infected with dysgonic fermenter-2. South Med J 77, 241–242.
Journal of Medical Microbiology 53
