  |
|
|
||||||||||||||||
|
||||||||||||||||||
| Journal of the New Zealand Medical Association,
01-June-2007, Vol 120 No 1255
|
||||||||||||||||||
|
Two cases of endocarditis due to Bartonella
henselae
James Fu, Sharmini Muttaiyah, Sushil Pandey, Mark
Thomas
Bartonella henselae—the slow-growing,
fastidious Gram-negative bacillus which causes cat scratch disease—is a
recently recognised cause of endocarditis.1 To
our knowledge, until now, there have not been any cases of endocarditis caused
by this organism recognised in New Zealand. We describe two recent patients with
endocarditis caused by B. henselae.
Case report 1A 67-year-old woman (with a past history of a mitral
valvotomy in 1968 for rheumatic mitral stenosis) presented to hospital with
symptomatic atrial flutter. She was febrile (38ºC), had multiple splinter
haemorrhages on her finger nails, and several Osler’s nodes on the dorsal
surface of her fingers (Figure 1) and had both a grade 1/4 early diastolic
murmur and a grade 1/4 ejection systolic murmur. She had not received any
medical treatment, and in particular had not received any antimicrobial
medications in the preceding months.
Figure 1. Right middle finger of Case 1 showing
multiple small splinter haemorrhages in nail-bed and an Osler’s node
adjacent to the PIP joint
 Initial investigations revealed a haemoglobulin level of 9.9
g/dL, a total leukocyte count of 9.1 ×
109/L, and an erythrocyte sedimentation rate
(ESR) of 64 mm/hr. A midstream specimen of urine contained 30 ×
106 WBC/L and >1000 ×
106 RBC/L with hyaline and cellular casts, and
her serum creatinine level was 255 μmol/L.
A transthoracic echocardiogram showed a 0.4 × 1.0 cm
vegetation on the anterior mitral leaflet with moderate mitral regurgitation,
and a 1.0 × 0.6 cm vegetation on the right cusp of the aortic valve without
significant valve dysfunction. An ultrasound of the renal tract showed a subtle
diffuse increase in renal echogenicity. Three blood cultures collected over 1
hour remained sterile after 10 days of incubation.
A diagnosis of culture-negative endocarditis was made and
the patient was treated with ceftriaxone 2 gm 24 hrly alone for 9 days, then
with benzyl penicillin 3 mU 4 hrly and gentamicin 120 mg 24 hrly.
On day 15 of medical treatment, a repeat echocardiogram
showed worsening valvular dysfunction and she proceeded to replacement of her
aortic and mitral valves with porcine bioprotheses. Both excised valves were
cultured but remained sterile after seven days incubation. However microscopy of
the mitral valve revealed small numbers of pleomorphic Gram variable organisms
and PCR amplification of bacterial 16S ribosomal DNA from the aortic valve
tissue produced an amplicon with a 100% match for the published sequence for
B. henselae.
A serum sample collected 30 days after presentation had
titres of IgM and IgG antibodies to B. henselae of >1:80 and 1:2048
respectively. Her postoperative antimicrobial treatment comprised another 10
days of intravenous penicillin 3 mU 4 hrly and gentamicin 120 mg 24 hrly
followed by doxycycline 200 mg daily for a further 28 days.
Two months after discharge from hospital she presented with
blurring of vision in the right eye and was found to have bilaterally swollen
optic nerves. A cranial magnetic resonance (MRI) scan showed minor inflammation
of both optic nerves.
A cerebrospinal fluid sample contained <1 ×
106 WBC/L. PCR amplification of blood and
cerebrospinal fluid failed to detect any B. henselae DNA, and two sets
of blood cultures remained sterile after 10 days of incubation. No further
antimicrobial therapy was given and her visual problems improved with
corticosteroid therapy. She remained well after a further 10 weeks of
follow-up.
Case report 2A 23-year-old man with known rheumatic valvular heart
disease presented to hospital in Tahiti with fever and was found to have
moderate aortic and mitral valve regurgitation with vegetations present on both
valves. There was evidence of embolic disease affecting his popliteal arteries
bilaterally. Blood cultures were sterile and he was treated with amoxicillin
plus clavulanic acid, an aminoglycoside and rifampicin.
Ten weeks after his initial presentation he was transferred
to Auckland City Hospital and underwent aortic valve replacement and mitral
valve repair. A serum sample collected 3 days after arrival at Auckland City
Hospital had titres of IgM and IgG antibodies to B. henselae of
<1:20 and 1:16384 respectively. No organisms were seen in the mitral valve
vegetations or in the aortic valve tissues, but PCR amplification of bacterial
16S ribosomal DNA from the aortic valve produced an amplicon with a 100% match
for the published sequence for B. henselae.
He was treated for a further 2 weeks with an aminoglycoside
and for 6 weeks with doxycycline 100 mg BD. He has remained well 9 weeks after
completion of this treatment.
DiscussionThe overwhelming majority of organisms that cause infective
endocarditis are readily isolated from blood cultures incubated for 10
days. 2 However in a small minority of patients
with infective endocarditis no organism is isolated, commonly either because
previous antimicrobial therapy has partially cured the infection, or because the
infection is due to an unusually fastidious or slow-growing organism (Table
1).
Table 1. Aetiological agents in infective
endocarditis2
Bartonella (previously known as Rochalimaea)
henselae was first identified as a cause of endocarditis in
1993.1 Subsequent investigations have shown
that B. henselae is responsible for approximately 6% cases of culture
negative endocarditis and that it can be diagnosed either by detection of serum
antibodies to B. henselae (sensitivity >90%) or by detection of
B. henselae DNA in excised valve tissue (sensitivity approx
90%)—or less reliably by isolation of B. henselae from blood
cultures incubated for 3 weeks (sensitivity approx
20%).3
Other important infective causes of culture negative
endocarditis include C. burnettii (approx 50%), B. quintana
(approx 20%), and Tropheryma whipplei, Mycoplasma
spp, Chlamydia spp, and Legionella spp (all very
rare).3
Patients with culture-negative endocarditis comprised 21/189
(11%) episodes of endocarditis reported from Green Lane Hospital between 1959
and 1976;4 5/102 (5%) episodes reported from
Auckland Hospital between 1979 and 1986;5 but
none of 78 episodes reported from Middlemore Hospital between 1976 and
1986.6
Each of these case series were collated before the
identification of B. henselae as a cause of endocarditis and before the
introduction of modified diagnostic criteria which would be expected to increase
the rate of diagnosis of culture negative
endocarditis.7
Two cases of endocarditis due to B. henselae have
been reported from Australia, one in a 50-year-old Queensland man and the other
in a 44-year-old man transferred from New
Caledonia.8,9 B. henselae DNA has been
detected in 11% of 114 cat fleas collected from cats presenting to veterinary
clinics in three New Zealand towns.10
B. henselae is likely to have been the unrecognised cause of
occasional cases of endocarditis in New Zealand in the past.
The optimum antibiotic regimen for B. henselae
endocarditis remains unclear. In vitro, B. henselae are highly
susceptible to beta-lactam agents, aminoglycosides, macrolides, tetracyclines,
and rifampin.
A review of 101 cases of Bartonella endocarditis
found that the outcome was improved in patients who received at least 2 weeks of
treatment with an aminoglycoside.11 The current
recommendation of the American Heart Association is for 2 weeks of gentamicin 3
mg/kg daily plus 6 weeks of doxycycline 100 mg
BD.12 Valvular surgery is commonly required in
patients with Bartonella endocarditis.
We hope that these two cases will stimulate clinicians to
consider the diagnosis of endocarditis due to B. henselae in patients
whose blood cultures remain sterile. Failure to identify the cause of
endocarditis in such patients is likely to result in inappropriate treatment
with a reduced likelihood of cure.
Author information: James Fu, Infectious
Diseases Registrar; Sharmini Muttaiyah, Microbiology Registrar; Sushil Pandey,
Scientific Officer, DNA Microbiology; Mark G Thomas, Infectious Diseases
Physician; Department of Infectious Diseases and Microbiology Laboratory,
Auckland City Hospital; Auckland.
Correspondence: Dr Mark Thomas, Infectious
Diseases Physician, Auckland City Hospital, Private Bag 92024, Auckland. Fax:
(09) 307 4940; email: mthomas@adhb.govt.nz
References:
|
|
||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|