Infective endocarditis at a large community teaching hospital, 1980-1990. A review of 210 episodes

We studied 210 episodes of infective endocarditis in 204 patients. The prevalence of this disease in our series ranged from 0.32 to 1.30 (mean, 0.75) episodes per 1000 admissions per year. There were 115 male and 89 female patients, whose ages ranged from newborn to 91 years (median, 60-70). One-hundred-and-forty-eight episodes involved host valves and another 33 episodes occurred in intravenous drug users. There were 2 episodes of early and 27 episodes of late prosthetic valve endocarditis. Staphylococcus aureus accounted for 99 episodes (47.1%), alpha-hemolytic streptococci for 29 episodes (13.8%), enterococci for 11 episodes (5.2%), culture-negative endocarditis for 11 episodes (5.2%), and other organisms for 60 episodes (28.6%). Severe back pain was the chief complaint in 15 patients. 2-D echocardiography was performed in 164 episodes, results in 67 (40.9%) of which were positive. Valve surgery was performed in 29 episodes (23 host valves and 6 prosthetic valves). The overall mortality was 21.4%. Autopsy was performed in 22 of the 45 patients who died (48.9%). The mortality rate increased with age, (10.1% and 31.5% for patients < 60 years old and 60 years or older, respectively (p < 0.001).     

Infectious endocarditis, risk factors, prevention. Research group for infectious endocarditis and jury of the consensus conference

Infective endocarditis, resulting from deposition of circulating microorganisms during a period of bacteremia on damaged endothelial heart valves, remains a serious disease. Its overall incidence did not decline in recent years, 24 cases/year/million inhabitants, in France. This can be explained by a modification of the type of underlying cardiac conditions with regression of rheumatic fever and increase of degenerative heart diseases, prosthetic valves and mitral valve prolapse. The risk of bacterial seeding on a damaged valve remains difficult to evaluate, the highest risk being for patients with prosthetic valve, previous infective endocarditis and cyanotic congenital heart disease. A case-control study, done in 1991, confirmed that procedures are risk factors for infective endocarditis and that the multiplicity of procedures increases the risk. A French consensus conference on the prophylaxis of infective endocarditis has updated the recommendations for antibiotic prophylaxis. Two groups of cardiac patients were identified, based on the incidence and the severity of endocarditis occurring in patients with these conditions, 1) patients considered at risk which require specific prophylaxis, and 2) patients considered non at risk which do not require antibiotic prophylaxis. Procedures which require antibiotic prophylaxis are dental procedures and specific gastrointestinal and urologic procedures. As complex protocols are associated with poor compliance by practitioners and patients, the jury has aimed for simplicity and feasibility.     

Physiopathology of leg ulcers]

Polymicrobial endocarditis (PE) is uncommon, whether in series of cases of polymicrobial bacteriemia or of endocarditis. Among the 201 cases of infective endocarditis seen between 1986 and 1995 by an infectious diseases service, 12 patients had PE (6%). Nine were males, mean age was 28 years and ten were active intravenous drug users. All of them were HIV (+) and 50% had AIDS. Eleven subjects had infection of the tricuspid valve and 58% developed septic pulmonary emboli. The most common organism encountered was Staphylococcus aureus in 8 patients followed by Streptococcus viridans and S. pneumoniae in three. The most common combinations of organisms were S. aureus and S. pneumoniae in 3 cases and S. aureus and Pseudomonas aeruginosa in two. Two patients died, one with Xantomona maltophilia and another with Candida albicans. The symptoms of PE were usually indistinguishable from endocarditis caused by a single organism and the prognosis depended on the species rather than the number of organisms isolated.     

Multiple promoters direct stimulus and temporal specific expression of brain-derived neurotrophic factor in the somatosensory cortex

Patients operated on for infective endocarditis (n = 69) at two regional hospitals between 1988 and 1994 are reviewed. 70% had a known valvular heart disease and 16% had prosthetic valve endocarditis. In 28% the offending microorganism was Staphylococcus aureus; in 26% Streptococcus viridans. Therapy was intended to be a six-week antibiotic course before operating, but 55% of the patients had to be operated on earlier. The postoperative course was uncomplicated in 59%, mortality was 16% and one-year survival 81%. Increased risk of death was associated with operating before the six-week course of antibiotics was completed (p = 0.005), with preoperative renal failure (p = 0.006) or lung failure (p = 0.008), with the growth of microorganisms from tissue samples extirpated during the operation (p = 0.01), with additional surgical procedures concomitant to valvular replacement (p = 0.02), S. aureus endocarditis (p = 0.03), and with the presence of paravalvular abscesses or intracardial fistulas (p = 0.03). The study shows that infective endocarditis is a serious disease. Wherever clinically feasible, all patients should be given antibiotics for six weeks before evaluating surgery. However, close surveillance of infection and haemodynamics is necessary to allow for the possibility of acute surgery before the development of organ failure. Special attention must be paid to cases of S. aureus endocarditis.     

A review of infectious endocarditis due to Candida

OBJECTIVE: As fungal endocarditis is a serious disease, frequently requiring cardiac surgery, a review was made of the experience of our Departments in this pathology. DESIGN: A retrospective analysis of clinical, echocardiographic and surgical data. SETTING: Patients studied in a tertiary care Hospital with cardiac surgery available. PATIENTS: Between 1984 and 1994 there were ten cases of candida endocarditis in nine patients, four male and five female, mean age--45 +/- 12 years (31-65). INTERVENTIONS: The following parameters were analysed: clinical (predisposing factors, clinical evolution, complications, therapy and mortality), echocardiographic (presence of vegetations, abscesses, valvular regurgitations). Patients studied in other Centres and referred to our Department only for examination (echocardiograms) were excluded from this analysis. RESULTS: Eight cases in seven patients were prosthetic valve endocarditis and two native valve endocarditis. No patient was drug addicted. Seven cases of prosthetic valve endocarditis developed less than one year after surgery and another had a gynecological fungal infection as the cause of the endocarditis. Four patients had had previous endocarditis. There were four embolic events and three developed heart failure. There were three perivalvular infections, six valvular regurgitations and only one case with huge vegetations on echocardiography. Nine patients were treated with amphotericin B, in five fluocytosin was added and in four ketoconazol, which was replaced by flukonazol in one patient. Therapy was continued for at least eight weeks. Six patients were operated during the acute stage and one died. One patient was operated on late after the infection. Three patients died during the active stage. In a follow up of 5.2 +/- 4.8 years (8 months to 8 years) there was one fatal candida endocarditis relapse, one fatal candida sepsis, one non cardiac death, one patient developed a periprosthetic leak and one had recurrent systemic embolization. Abscesses/pseudoaneurysms were found in five out of seven patients submitted to surgery. CONCLUSION: Candida infective endocarditis has a bad prognosis, specially in those patients not operated early; it develops in patients with predisposing factors, which in our series were a previous infective endocarditis (four patients) and/or a prosthetic valve implantation less than one year before; it has important morbidity with multiple embolic events, perivalvular involvement, valvular regurgitation and heart failure.     

Intramuscular clindamycin for therapy of infective endocarditis. Report of 23 cases and review of the literature

Twenty-three patients with infective endocarditis received intramuscular clindamycin (Cleocin) for treatment. Thirteen had acute Staphylococcus (S.) aureus endocarditis but none had involvement of the aortic valve. Eleven of these 13 infections were heroin-related and involved the tricuspid valve.Twenty-one patients were successfully treated. Two patients with heroin-related S. aureus infection failed to respond to intramuscularly administered clindamycin, but responded to retreatment with methicillin. There have been 34 reported cases of endocarditis treated with clindamycin. Although 80 percent of all cases due to staphylococci responded favorably, almost all were heroin-related tricuspid valve infections. In addition 91 percent of cases due to aerobic streptococci responded but, surpisingly, treatment failed in three of four cases of anaerobic endocarditis. Although clindamycin can be useful in streptococcal endocarditis and in some cases of heroin-related S. aureus tricuspid endocarditis, caution should be exercised in its use. It is "less" bactericidal than the penicillins or cephalosporins, and organisms have become resistant during treatment. Furthermore, patients with anaerobic endocarditis have not responded well, and data are not available to recommend administration of clindamycin for acute S. aureus infections engrafted on the aortic or mitral valve.     

Prosthetic valve endocarditis: clinicopathologic analysis of 22 necropsy patients with comparison observations in 74 necropsy patients with active infective endocarditis involving natural left-sided cardiac valves

Clinical and morphologic features are described in 22 necropsy patients with endocarditis involving rigid-framed prosthetic valves: aortic in 15 patients and mitral in 7. The interval from valve replacement to onset of symptoms of prosthetic valve endocarditis was less than 2 months in 8 patients and longer than 2 months in 14 patients. The most frequent infecting organism was the Staphylococcus (13 patients). In each of the 22 patients the infection was located behind the site of attachment of the prosthesis to the valve ring, and the infection spread to adjacent structures in 13 patients, 11 of whom had aortic prostheses. Prosthetic detachment causing severe regurgitation occurred in 12 of the 15 patients with an infected aortic valve prosthesis, and in 2 of the 7 with an infected mitral valve prosthesis. Prosthetic obstruction by vegetative material occurred in 5 of 7 patients with prosthetic mitral infection and in only 1 of 15 with prosthetic aortic infection. High degrees of conduction defects developed in seven patients with aortic prosthetic valve endocarditis: complete heart block in five, and complete left bundle branch block in two. Comparison of observations in the 22 patients with prosthetic valve endocarditis with those in 74 patients with active infective endocarditis involving natural left-sided cardiac valves revealed significant (P less than 0.05) differences in the percent with ring abscess, hemodynamic consequences of the endocarditis (valve stenosis), frequency of Staphylococcus as the causative organism and percent with complete heart block or left bundle branch block. No significant differences were observed between the two groups when comparing age, sex, type of underlying valve disease or frequency of organ infarcts of splenomegaly.     

Body site distribution of cutaneous malignant melanoma in relationship to patterns of sun exposure

AIM: To compare the clinical and morphological characteristics of patients with Streptococcus bovis endocarditis with those of patients with endocarditis caused by other microorganisms. METHODS: 177 consecutive patients (Streptococcus bovis, 22; other streptococci, 94; staphylococci, 44; other, 17) with definite infective endocarditis according to the Duke criteria were included. All patients underwent transthoracic and transoesophageal echocardiography. In 88 patients, findings from surgery/necropsy were obtained. RESULTS: S bovis endocarditis was associated with older patients, with a higher mortality (p = 0.04), and with a higher rate of cardiac surgery (p < 0.001) than other microorganisms, although embolic events were observed less often (p = 0.02). Pathological gastrointestinal lesions were detected in 45% of the patients. Multiple valves were affected in 68% of the patients with S bovis endocarditis and in 20% of those with other organisms (p < 0.001). Moderate or severe regurgitation occurred more often in S bovis endocarditis than with other microorganisms (p = 0.05). When surgery or necropsy was performed, infectious myocardial infiltration of the left ventricle was confirmed histopathologically in 36% of the patients with S bovis endocarditis and in 10% of those with other organisms (p = 0.002). CONCLUSIONS: S bovis endocarditis is a severe illness because of the more common involvement of multiple valves, and of the frequent occurrence of haemodynamically relevant valvar regurgitation and infectious myocardial infiltration.     

N-acetylcysteine prevents development of the hyperdynamic circulation in the portal hypertensive rat

Infective endocarditis, defined as pathologically or clinically definite by the Duke criteria, was observed in 14 transplant recipients at our institutions. In addition, we reviewed 32 previously reported cases in solid organ transplant recipients. The spectrum of organisms causing infective endocarditis was clearly different in transplant recipients than in the general population; 50% of the infections were due to Aspergillus fumigatus or Staphylococcus aureus, but only 4% were due to viridans streptococci. Fungal infections predominated early (accounting for six of 10 cases of endocarditis within 30 days of transplantation), while bacterial infections caused most cases (80%) after this time. In 80% (37) of the 46 cases in transplant recipients, there was no underlying valvular disease. Seventy-four percent (34) of the 46 cases were associated with previous hospital-acquired infection, notably venous access device and wound infections. Three patients with S. aureus endocarditis had had an episode of S. aureus bacteremia > 3 weeks prior to the diagnosis of endocarditis and had received treatment for the initial bacteremia of < 14 days' duration. The overall mortality rate was 57% (26 of 46 patients died), with 58% (15) of the 26 fatal cases not being suspected during life. Endocarditis is an underappreciated sequela of hospital-acquired infection in transplant recipients.     

Propionibacterium acnes infectious endocarditis

BACKGROUND: Propionibacterium acnes, a Gram positive microaerophilic bacteria is considered to have a low level of virulence. Nevertheless, it can be associated with infective endocarditis. We report 2 cases and a review of the literature. CASE REPORTS: The first man developed an acute prosthetic valve infective endocarditis and died. The second case was also a prosthetic valve infective endocarditis in a patient who developed infectious spondylitis. DISCUSSION: P. acnes is an uncommon causal agent in infective endocarditis and appears to have a predilectio for prosthetic valves. A prior history of skin infection is rare. P. acnes grow quite slowly, often requiring 7 to 14 days for identification. All isolates of P. acnes from blood or valve do not necessarily mean contamination.     

Surgery of infectious endocarditis

Thirty to fifty percent of patients with infective endocarditis are operated on during the active phase of the disease; this percentage is higher in case of some valvular localizations (aortic), in case of early prosthetic valve endocarditis, in case of some microorganisms (Staphylococcus aureus, gram-negative, fungus, intracellular microorganism). Operative death (at 30 days) is below 10% in native valve endocarditis, close to 50% in early prosthetic valve endocarditis, and below 20% in late prosthetic valve endocarditis. When active infective disease has been healed by medical treatment alone, half the patients need surgery in the first 2 years of follow-up; the indications for surgery are the functional status, the degree of valvular leaks and other lesions, the degree of ventricular dilatation.     

Prosthetic valve endocarditis leading to valve replacement: a case study

Infective endocarditis (IE) is a pathologic condition of native or prosthetic heart valves or endocardium, which may result in valve destruction and congestive heart failure. It occurs more frequently in men than in women, and there is an increased trend in the elderly. The following conditions predispose patients to IE: congenital and rheumatic heart disease, calcification or stenosis of a valve, prosthetic valve surgery, a previous episode of endocarditis, poor dentition, parenteral drug abuse, and placement of intravascular lines or devices. Effective treatment frequently involves a combination of intense antibiotic therapy and surgical repair. Risk of death from IE is related to age over 60, diagnosis of staphylococcal infection, involvement of an aortic or prosthetic valve, and the presence of any of the following sequelae of endocarditis: congestive heart failure, embolic phenomenon, and neurologic deficit. Clinicians should suspect endocarditis in patients presenting with fever of unknown origin and who are at risk for endocarditis. Timely evaluation with transthoracic or transesophageal echocardiography may identify patients in the early stages of endocarditis and direct the patient to definitive therapy. Early treatment of native and prosthetic valve endocarditis may decrease its overall morbidity and mortality. This case study illustrates some of the challenges in effectively managing prosthetic valve endocarditis.     

Hospital practices influence the pattern of infective endocarditis

OBJECTIVE: To identify factors contributing to infective endocarditis at a major teaching hospital. METHODS: Retrospective review of clinical records of patients diagnosed with endocarditis by standard case definitions with respect to causative organisms, clinical features and outcome. RESULTS: One hundred and ninety-three episodes of endocarditis seen between 1979 and 1992 at Westmead Hospital, Sydney, were reviewed. In the 174 cases where the causative organism was isolated, 75 (43%) were Staphylococcus aureus and 50 (29%) were viridans streptococci. Nosocomial acquisition and/or inter-hospital transfer accounted for 83 episodes; 48 (58%) S. aureus (P < 0.001) and nine (11%) viridans streptococci (P < 0.001). In cases from the local community, viridans streptococci were more common than S. aureus (37% versus 25%); these included 18 episodes (14 S. aureus) in intravenous drug users. CONCLUSION: We conclude that, compared with community-acquired infections, the aetiology of endocarditis in a large teaching hospital is influenced strongly by the prevalence of nosocomial endocarditis and the need for interhospital transfer of complicated cases.     

Calcaneal osteotomy

Echocardiographic observations are described in 25 opiate addicts with active infective endocarditis involving apparently previously normal valves. Infective endocarditis was isolated to the tricuspid valve in 11 patients, involved both right- (tricuspid valve) and left-sided valves in 7 and was isolated to the left-sided valves in 7 (mitral valve in 6). Twenty patients (80%) had tricuspid valve regurgitation, 12 had mitral regurgitation, 3 had aortic regurgitation and none had pulmonary valve regurgitation. Considering the 75 cardiac valves (excluding the pulmonary) in the 25 patients, echocardiographic abnormalities consistent with active infective endocarditis were detected in 26 (74%) of the 35 clinically incompetent valves but in none of the 40 competent valves. Comparison of the 20 incompetent tricuspid valves with the 12 incompetent mitral valves indicated that (1) the echocardiogram was less sensitive in detecting tricuspid valve lesions, (2) rupture of tricuspid valve chordae tendineae was absent or not detectable, and (3) tricuspid valve vegetations tended to be larger.     

Risk of reoperative valve replacement for failed mitral and aortic bioprostheses

BACKGROUND: One factor influencing the choice of mechanical versus bioprosthetic valves is reoperation for bioprosthetic valve failure. To define its operative risk, we reviewed our results with valve reoperation for bioprosthetic valve failure. METHODS: Records of 400 consecutive patients having reoperative mitral, aortic, or mitral and aortic bioprosthetic valve replacement from January 1985 to March 1997 were reviewed. RESULTS: Reoperations were for failed bioprosthetic mitral valves in 219 patients, failed aortic valves in 153 patients, and failed aortic and mitral valves in 28 patients. Including 26 operations (6%) for acute endocarditis, 153 operations (38%) were nonelective. One hundred nine patients (27%) had other valves repaired or replaced, and 72 (18%) had coronary bypass grafting. The incidence of death in the mitral, aortic, and double-valve groups was respectively, 15 (6.8%), 12 (7.8%), and 4 (14.3%); and the incidence of prolonged postoperative hospital stay (>14 days) was, respectively, 57 (26.0%), 41 (26.8%), and 8 (28.6%). Only 7 of 147 patients (4.8%) having elective, isolated, first-time valve reoperation died. Multivariable predictors (p < 0.05) of hospital death were age greater than 65 years, male sex, renal insufficiency, and nonelective operation; and predictors of prolonged stay were acute endocarditis, renal insufficiency, any concurrent cardiac operation, and elevated pulmonary artery systolic pressure. CONCLUSIONS: Reoperative bioprosthetic valve replacement can be performed with acceptable mortality and hospital stay. The best results are achieved with elective valve replacement, without concurrent cardiac procedures.     

Infectious endocarditis and congenital heart defects in adults

BACKGROUND: Infective endocarditis still remains a cardiological menace. However, the type of predisposing diseases has changed: the incidence of rheumatic heart disease in advanced countries has declined, advances made in the surgical and medicamentous treatment of inborn heart disease are the reasons why we are encountering, with increasing frequency, infective endocarditis which develops on their background. METHODS AND RESULTS: The objective of the investigation was to assess the frequency of infective endocarditis and predisposing diseases. During the time interval between 1987 and 1991 16 patients (50% younger than 50 years) were hospitalized with the diagnosis of infective endocarditis. Rheumatic valvular damage and inborn heart disease were the predisposing factor in 25%. All patients were younger than 30 years (mean 24, range 18-30), and half the patients suffered from defects of the ventricular septum. Other defects were tetralogy of Fallot and inborn aortic stenosis. In 10% of the patients infective endocarditis developed on the background of a mitral valve prolapse with regurgitation. Echocardiographic examination confirmed the diagnosis in almost 70% by revealing vegetation. Bacteriological examination revealed the agent in 60%, most frequently it was Streptococcus viridans. The mortality rate in the group was 13%. CONCLUSIONS: The recorded incidence of infective endocarditis, 1.5 pro mille, is consistent with data in the literature. Corrected and not corrected heart disease plays an important role as predisposing disease. Despite the opportunity of intensive antibiotic treatment, the mortality remains high--13%.     

Prosthetic valve endocarditis: superiority of surgical valve replacement versus medical therapy only

The objective of our study was to assess the long-term outcome of patients with prosthetic valve endocarditis. We used a multicenter, prospective, observational study design. Six university teaching hospitals with high volume cardiothoracic surgery participated. Seventy-four patients with prosthetic valve endocarditis as defined by explicit, objective criteria were selected for participation. All patients were followed up prospectively for 1 year. Thirty-one percent and 69% had development of endocarditis within 60 days of valve insertion ("early") and after 60 days ("late"), respectively. The most common causes were Staphylococcus epidermidis (40%), Staphylococcus aureus (20%), streptococcal species (18%), and aerobic gram-negative bacilli (11%). Physical signs of endocarditis (new or changing murmur, stigmata, emboli) were seen in 58%. At 6 months and 12 months, mortality was 46% and 47%, respectively. Surgical replacement of the infected valve led to significantly lower mortality (23%) as compared with medical therapy alone (56%), as assessed by both univariate and multivariate analyses (p < 0.05). Improved outcome was seen for the surgical group even when controlling for severity of illness at time of diagnosis. From these findings we conclude that accurate assessment of outcome in prosthetic valve endocarditis requires long-term follow-up of at least 6 months following diagnosis. Surgical therapy warrants greater scrutiny; evaluation in controlled clinical trials is appropriate.     

Incidence and prognosis of embolic events and metastatic infections in infective endocarditis

AIMS: In infective endocarditis, the true incidence of embolic events and metastatic infections remains unknown probably because a large number of events are asymptomatic. The consequences of the prognosis of such events have never been evaluated by a prospective follow-up. This study aimed to assess the incidence of symptomatic or asymptomatic embolic events and metastatic infections in definite infective endocarditis and to determine whether these events carry a risk of mortality. METHODS AND RESULTS: From January 1991 to December 1993, 102 patients with suspected or known infective endocarditis were referred to our institution. Among them, we selected 68 patients (50 males, 18 females, mean age = 52.7 years) exhibiting definite infective endocarditis according to the Duke University criteria. Blood cultures were positive in 49 cases (72%). Echocardiography revealed valvular vegetations in 55 cases (81%). Irrespective of the clinical presentation, patients were examined radiologically by cerebral computed tomography scanning (n = 60), magnetic resonance imaging (n = 3), abdominal computed tomography scanning (n = 32) or abdominal echocardiography (n = 20). Depending on the symptoms, thoracic computed tomography scanning (n = 22), pulmonary angiography (n = 2), ventilation-perfusion scintigraphy (n = 4), or gallium citrate radionuclide scanning (n = 7) were also performed. All patients were prospectively followed-up for a mean period of 21.4 +/- 17.5 months. In 35 patients (51%), 51 embolic or metastatic events were revealed, involving the central nervous system (n = 23), spleen (n = 7), kidney (n = 5), lung (n = 5), liver (n = 4), bone and joint (n = 4), iliac (n = 2) or mesenteric (n = 1) arteries. During the hospital stay, the mortality rate was higher in patients exhibiting embolic or metastatic events (20 vs 12%), but the difference did not reach statistical significance. Kaplan-Meier analysis demonstrated no difference in long-term follow-up. CONCLUSION: Our data suggest that embolic or metastatic events had a high incidence (51%) during infective endocarditis, but were not associated with significant attributable mortality.     

Infective endocarditis in Jerusalem. A comparative analysis of native and prosthetic valve endocarditis

Thirty patients with prosthetic valve endocarditis (PVE) and 62 patients with native valve endocarditis (NVE) observed during a concurrent period of time (1970-80) were analyzed. Patients with PVE were also compared with 697 patients who underwent cardiac valve replacement in the same period, in an attempt to uncover risk factors for PVE. In the operated patients the overall incidence of endocarditis was 4.3%. There were several significant differences between the NVE and PVE groups. Atrial fibrillation was more prevalent in the PVE group (33.4 vs. 11.5%, P less than 0.05), while infection with gram-positive organisms was more prevalent in the NVE patients (79 vs. 47%, P less than 0.01). The outcome of combined medical and surgical treatment showed a much higher mortality rate (50 vs. 6.4%, P less than 0.01) in the PVE group. Staphylococcal endocarditis resulted in 16% mortality in NVE vs. 100% in the PVE group (P less than 0.01). Careful surveillance may hopefully lead to early detection, treatment and better outcome of PVE.     

Development of animal models for the chronobiotics of melatonin analogs

We reviewed the records of patients admitted to our centre with the diagnosis of isolated tricuspid valve infective endocarditis and analysed the clinical presentation, etiopathogenic agent, echocardiographic features and therapeutic approach, namely the indication for cardiac surgery. Between 1988 and 1996, 11 cases of confirmed tricuspid valve endocarditis were identified, corresponding to 5% of the cases of endocarditis admitted to our centre in the same period. A predisposing factor was found in ten of the patients, half of them intravenous drug addicts and Staphylococcus aureus was the most frequent agent isolated. Fever and pleuro-pulmonary manifestations were predominant clinical features. Transthoracic echocardiography had a crucial role in the diagnosis and transesophageal echocardiography was important to characterize vegetations. Four patients underwent cardiac surgery, for persistent infection. In two cases, excision of the vegetations and ring annuloplasty was performed. In two patients not addicted to drugs, the tricuspid valve was replaced with a bioprosthesis, since the extension of the damage to the valve did not allow repair. One patient, with early endocarditis of a tricuspid bioprosthesis died before surgery was attempted.