high-dose intravenous benzylpenicillin and genta- micin in a synergistic dose for suspected infective endocarditis. Transthoracic echocardiography (TTE).
Q J Med 1996; 89:855-858
Native-valve endocarditis caused by Staphylococcus lugdunensis M.P.A. LESSING, D.W.M. CROOK, I.C.J. BOWLER and B. GRIBBIN 1 From the Department of Medical Microbiology and10xford Heart Centre, The John Radcliffe, Oxford Radcliffe Hospital, Oxford, UK Received 8 August 1996 and in revised form 9 September 1996
Summary Coagulase-negative staphylococci cause about 5% of native-valve endocarditis. Staphylococcus lugdunensis, a recently-described species of coagulasenegative staphylococci, has been reported to cause destructive native-valve endocarditis with a high mortality. We report four consecutive cases of definite Staphylococcus lugdunensis native-valve endo-
carditis by the Duke criteria over a 4-year period. All patients required urgent aortic valve replacement 1-5 days after admission, and recovered. An intriguing aspect in the presentation of these patients was a history of vasectomy and inguinal skin breaks in the immediate period preceding the occurrence of endocarditis.
Introduction Coagulase-negative-staphylococcus native-valve endocarditis (CNS-NVE) comprises about 5% of cases of native-valve infective endocarditis.1'2 The coagulase-negative staphylococci consist of 30 species currently.3 Of these, Staphylococcus epidermidis is an established cause of both native-valve endocarditis1'2 and, particularly, prosthetic-valve endocarditis.1 Staphylococcus lugdunensis, first delineated by Freney,4 has been recently reported to cause destructive native-valve endocarditis.5"7 We describe four cases of definite Staphylococcus lugdunensis native-valve endocarditis by the Duke criteria,8 with an age range of 32-45 years, presenting 21-30 days following an inguinal skin break.
Methods Microbiology The strains were identified by the following methods: (i) a negative tube coagulase test,6'7 (ii) recognition as Staphylococcus lugdunensis by the API Staph Identification system for Staphylococci (Bio Merieux SA),4'6'7 (iii) a positive ornithine decarboxylase
test, and (iv) a positive plate-agar DNase test after > 9 6 h incubation. 6 The minimum inhibitory concentrations (MICs) to penicillin of three strains were determined by the E-test method (AB Biodisk).
Patients We managed four consecutive cases of definite Staphylococcus lugdunensis native-valve endocarditis, by the Duke criteria, over a 4-year period ending November 1994 (Table 1). Case 1 A 34-year-old male toolmaker felt unwell 30 days after a vasectomy, with onset of malaise, sweats and fever. He was admitted 3 weeks later following transient pain and weakness of the right hand and found to have clinical evidence of severe aortic reflux. He was febrile (38 °C) and blood cultures were drawn before starting empiric antibiotic therapy high-dose intravenous benzylpenicillin and gentamicin in a synergistic dose for suspected infective endocarditis. Transthoracic echocardiography (TTE) showed evidence of acute, severe aortic reflux with
Address correspondence to Dr M.P.A. Lessing, Department of Medical Microbiology, Level 6, The John Radcliffe, Oxford Radcliffe Hospital, Oxford OX3 9DU © Oxford University Press 1996
856 Table 1
M.P.A. Lessing etal. Patient characteristics
Patient
Age Sex
Valve
Source?
Valve replacement
1* 2 3 4
32M 37M 42 M 45 F
AR(tri) AR(bi) AR(bi) AR(tri)/TR
Vasectomy 30d Vasectomy 21 d Vasectomy 29d Inguinal furuncle 30d
AVR 1d AVR 1d AVR 5d AVR/TVR 2d
Time quoted for Source is that between procedure and first symptoms; that for Valve replacement is the interval between admission and operation. NVE, native valve endocarditis; AR/TR, aortic/tricuspid valve reflux; tri/bi, tri/bicuspid; AVR/TVR, aortic/tricuspid valve replacement. * Patient previously reported.6
premature closure of the mitral valve leaflets and vegetations. The valve appeared to be tricuspid. Flucloxacillin was added after Gram-positive cocci resembling staphylococci were seen on Gram stain from 8/8 blood culture bottles. These were identified as 5. lugdunensis susceptible to penicillin, and the flucloxacillin was discontinued. An aortic valve replacement was performed 5 days after admission for worsening aortic reflux and concern about left ventricular function. On inspection, the valve was covered in a thick layer of vegetations, with evidence of a root abscess. A total of 38 days of intravenous antibiotic therapy was given. He was still well at follow-up 36 months later. Case 2 A 37-year-old male computer programmer reported flu-like symptoms 21 days after vasectomy. Three weeks later, he was admitted with acute dyspnoea due to left ventricular failure. He was febrile (38.1 °C) and blood cultures were taken. He was commenced on empiric high-dose intravenous benzylpenicillin and gentamicin in synergistic dose for infective endocarditis. Twenty-four hours later, TTE imaging showed prolapse of an aortic cusp with severe reflux and a small vegetation. Gram staining of the blood cultures demonstrated Gram-positive cocci typical of staphylococci, and flucloxacillin was added to the patient's treatment. Surgery was carried out 2 days after admission, when the aortic valve was found to be largely destroyed with an aortic root abscess requiring an aortic valve replacement. Blood culture bottles (4/4) and the valve grew S. lugdunensis, susceptible to penicillin, minimum inhibitory concentration (MIC) 0.016 mg/l. He was transferred back to the referring hospital 10 days post-operatively, and remains well 2 years later. Case 3 A 42-year-old male computer programmer reported flu-like symptoms 29 days after vasectomy. One
week later, he presented with embolic occlusion of his left femoral artery which required embolectomy. Successful embolectomy was complicated by the onset of pulmonary oedema. He was febrile (38.2 °C). Blood was cultured, and empiric therapy with intravenous vancomycin and gentamicin was started because of a vague history of penicillin allergy. The vancomycin was discontinued and intravenous benzylpenicillin commenced when 3/4 blood culture bottles grew S. lugdunensis susceptible to penicillin, MIC 0.032 mg/l. On admission to this hospital there was clinical evidence of severe aortic reflux with pulmonary oedema, and the electrocardiogram showed prolongation of the PR-interval consistent with an aortic root abscess. Large aortic valve vegetations were shown on TTE. Urgent surgery revealed a disintegrating aortic valve, from which 5. lugdunensis was grown, and a root abscess. The valve was replaced and 41 days of antibiotic therapy administered. He was well at follow-up 8 months later. Case 4 A 45-year-old unemployed female presented with malaise, left-sided chest pain, haemoptysis and dyspnoea at rest. She gave a history of an inguinal furuncle 30 days earlier. On examination she was morbidly obese, febrile (37.6 °C) and had clinical evidence of aortic reflux. Transoesophageal echocardiography showed vegetations on the aortic and tricuspid valves, with a flail aortic cusp associated with severe reflux. Blood cultures were taken, and empiric intravenous therapy with vancomycin and gentamicin was commenced. Urgent surgery was performed, and inspection confirmed the echocardiographic findings with the addition of a large root abscess cavity and almost complete destruction of the tricuspid valve. Blood culture bottles (8/8) and samples of both valves grew penicillin-susceptible 5. lugdunensis, MIC 0.023 mg/l. Her antibiotic therapy was changed to intravenous benzylpenicillin
S. lugdunensis endocarditis and gentamicin which was given for 43 days. She was well 9 months post-operatively with a mild paraprosthetic leak.
Discussion Smyth first described a case of aggressive communityacquired 'new type' CNS-NVE requiring aortic valve replacement in 1988.5 The species Staphylococcus lugdunensis was described in that year (Lugdunum [Lat]: Lyon; French National Reference Centre for Staphylococci.4 In the largest study of Staphylococcus lugdunensis native-valve endocarditis and prostheticvalve endocarditis/ a series of 11 patients, 63% (5/8) of patients with native valve endocarditis had a known valvular abnormality, and the mortality of all Staphylococcus lugdunensis NVE cases reviewed in the literature by that report7 was 64% (9/14). A further feature of the native-valve-endocarditis group described in that series was the aggressive course of the disease—8/9 patients who did not have valve replacement died. The four cases presented in our series exhibited dramatic rapidly progressive valve destruction necessitating surgery. All of our patients recovered. An intriguing aspect in the presentation of our series of patients is a history of inguinal skin breaks in the immediate period preceding the occurrence of infective endocarditis. The three men had preceding vasectomies. Whether the inguinal skin breaks served as the source of these infections is uncertain, and can only be inferred from the incubation period. An 'expected' temporal relationship between two events is often used as circumstantial evidence for a causal link. 9 Our patients became symptomatic within a median time period of 27.5 days (range 21-30) after the skin lesion. This interval is within that described by Van der Meer, who reported a median interval of 72.5 days (range 3-170) between a risk-prone procedure and symptoms of infective endocarditis (IE).10 It is also consistent with the 'incubation period' of IE reported by Durack (14 days in 50%; 35 days in 80% of episodes.9'11 However, none of the cases reported by Vandenesch's paper had inguinal lesions recorded.7 It is tantalizing that Etienne and Herchline speculated from limited investigation that the normal habitat of 5. lugdunensis was the perineum. 7 ' 12 This being so, it would be the simple explanation for the occurrence of these cases following skin breaks in the inguinal region. The question as to which special niche, if any, is occupied by 5. lugdunensis can only be resolved by studies scrutinizing the human microflora. Vasectomy is a common minor operation and widely perceived to be safe and effective.13'14 Few
857
serious infective complications are recognized.13 Nevertheless, there is one description of native-valve endocarditis following vasectomy in 1984. 15 That infection was caused by Staphylococcus warneri and occurred in a 32-year-old pilot with no previous valvular lesions, who presented with embolic phenomena 30 days after vasectomy. He required aortic valve replacement 19 days after admission. As here 3/4 cases had a preceding vasectomy, we would consider Staphylococcus lugdunensis native-valve endocarditis as a possible rare infective complication of vasectomy. However, further cases exhibiting this association would have to occur before it could be confidently stated that this type of infection is indeed linked to vasectomy.
Acknowledgements Data from this study were presented by MPAL at the 33rd Annual Meeting of the Infectious Diseases Society of America (IDSA), 16-18 September 1995, San Francisco, USA.
References 1. Whitener C, Caputo GM, Weitekamp MR, Karchmer AW. Endocarditis Due to Coagulase-Negative Staphylococci: Microbiologic, Epidemiologic, and Clinical Considerations. Infect Dis Clin North Am 1993; 7:81 -96. 2. Etienne J, Eyken SJ. Increase in Native Valve Endocarditis caused by Coagulase Negative Staphylococci: An AngloFrench Clinical and Microbiological Study. BrHeartJ\990; 64:381-4. 3. Kloos WE, Bannerman TL. Update on Clinical Significance of Coagulase Negative Staphylococci. Clin Microbiol Rev 1994; 7:117-40. 4. Freney J, Brun Y, Bes M, Meugnier H, Grimont F, Grimont PAD, ef al. Staphylococcus lugdunensis sp nov and Staphylococcus schleiferi sp nov, Two Species from Human Clinical Specimens. IntJ Syst Bacteriol 1988; 38:168-72. 5. Smyth EG, Wright ED, Marples RR. New type of Staphylococcal Endocarditis. J Clin Pathol 1988; 41:809-10. 6. Walsh B, MounseyJP. Staphylococcus lugdunensis and Endocarditis. J Clin Pathol 1990; 43:171. 7. Vandenesch F, Etienne J, Reverdy ME, Eyken SJ. Endocarditis Due to Staphylococcus lugdunensis: Report of 11 Cases and Review. Clin Infect Dis 1993; 17:871-6. 8. Durack DT, Lukes AS, Bright DK and the Duke Endocarditis Service. New Criteria for Diagnosis of Infective Endocarditis: Utilization of Specific Echocardiographic Findings. Am JMed 1994; 96:200-9. 9. Durack DT. Prevention of infective endocarditis. New Engl JMed]995; 332:38-43. 10. Van der Meer JTM, Van Wijk W, Thomson J, Vandenbrouck JP, Valkenburg HA, Michel MF. Lancet 1992; 339:135-9. 11. Durack DT, Kaplan EL, Bisno AL. Apparent failures of endocarditis prophylaxis. JAMA 1983; 250:2318-22.
858
M.P.A. Lessing et al.
12. Herchline TE, Ayers LW. Occurrence of Staphylococcus lugdunensis in Consecutive Clinical Cultures and Relationship of Isolation to Infection. J Clin Microbiol 1991; 29:419-21. 13. Sapire KE, Levy S. Vasectomy as Outpatient Procedure. S Afr MedJ 1979; 55:10-14.
14. Massey FJ, Bernstein CS, O'Fallon WM, Schuman LM, Coulston AH, Crozier R, et al. Vasectomy and Health. JAMA 1984:252:1023-9. 15. Dan M, Marien CJR, Coldsand C. Endocarditis caused by Staphylococcus warneri. Can Med Assoc J 1984; 131:211-13.
