Gilbert Habib

Modification of the diagnostic criteria proposed by the Duke Endocarditis Service to permit improved diagnosis of Q fever endocarditis.

BACKGROUNDnQ fever endocarditis is a life-threatening disease for which the diagnosis is usually based on serology. The major microbiologic criterion for the diagnosis of infectious endocarditis (two separate positive blood cultures) cannot be achieved in most routine laboratories because of the biohazard associated with the culture of Coxiella burnetii, the etiological agent of Q fever.nnnPURPOSEnRecently, new criteria for the diagnosis of infectious endocarditis have been proposed, and in this study we attempted to assess the suitability of these criteria specifically for the diagnosis of Q fever endocarditis.nnnPATIENTS AND METHODSnTo achieve this aim, we first selected from our series 20 recent cases in whom endocarditis had been confirmed following valvular pathological examination, and for whom microbiological evidence for the involvement of C burnetii was available. Then, we applied the criteria proposed by the Duke Endocarditis Service (ie, C burnetii positive serology being considered a minor criterion) to this cohort of patients but excluding pathological findings. Although the Duke Endocarditis Service criteria confirmed diagnosis in 16 of the patients, 4 were misclassified as possible cases (20%). However, when the Q fever serological results (using an 1/800 antiphase I immunoglobulin G cut off) and single blood culture results were changed from minor to major diagnostic criteria, endocarditis was confirmed in them all. A second time, prospectively, we applied the Duke Endocarditis Service criteria to a further 5 patients affected with Q fever endocarditis. Strict application of these criteria resulted in 1 of the 5 being misdiagnosed. Applying the suggested modification for C burnetii results, all 5 were confirmed as having infectious endocarditis.nnnCONCLUSIONnWe propose that the modifications discussed in this study be applied to the Duke Endocarditis Service criteria in order that the diagnosis of C burnetii induced endocarditis is improved.

Prognostic Factors in 61 Cases of Staphylococcus aureus Prosthetic Valve Infective Endocarditis from the International Collaboration on Endocarditis Merged Database

Staphylococcus aureus prosthetic valve infective endocarditis (SA-PVIE) is associated with a high mortality rate, but prognostic factors have not been clearly elucidated. The International Collaboration on Endocarditis merged database (ICE-MD) contained 2212 cases of definite infective endocarditis (as defined using the Duke criteria), 61 of which were SA-PVIE. Overall mortality rate was 47.5%, stroke was associated with an increased risk of death, and early valve replacement was not associated with a significant survival benefit in the whole population; however, patients who developed cardiac complications and underwent early valve replacement had the lowest mortality rate (28.6%).

Evaluation of the LightCycler SeptiFast test in the rapid etiologic diagnostic of infectious endocarditis.

The SeptiFast test (Roche Diagnostics) is a new commercial molecular technique that has emerged for the detection of bacteria in blood. We compared in this study the sensitivity of blood culture to a commercially available broad-range real-time polymerase chain reaction (PCR) assay for the detection in blood of 19 bacterial species and six fungal species (SeptiFast test, Roche Diagnostics) in 63 patients with infectious endocarditis (IE). The SeptiFast test is not more sensitive for organisms such as Streptococci, Enterococci, and Staphylococcus aureus (11/29 versus 12/29 for blood culture). It has detected less commonly coagulase-negative Staphylococci (0/15 versus 3/15, Pu2009=u20090.2) and significantly fewer other microorganisms (0/6 versus 4/6, Pu2009=u20090.03). However, bacteria were detected from three IE treated by antibiotics, with blood culture negative on admission. The SeptiFast test may be useful in cases of IE in patients treated with antibiotics before admission.

Emergence of endocarditis due to group D streptococci: findings derived from the merged database of the International Collaboration on Endocarditis

The aim of the present study was to compare the epidemiological and clinical characteristics of Streptococcus bovis endocarditis with those of endocarditis caused by oral streptococci, using data obtained from a large international database of uniformly defined cases of infective endocarditis. S. bovis, a well-known cause of infective endocarditis, remains the common name used to designate group D nonenterococcal streptococci. In some countries, the frequency of S. bovis endocarditis has increased significantly in recent years. Data from the International Collaboration on Endocarditis merged database was used to identify the main characteristics of S. bovis endocarditis and compared them with those of infective endocarditis (IE) due to oral streptococci. The database contained 136 cases of S. bovis IE and 511 cases of IE due to oral streptococci. Patients with S. bovis IE were significantly older those with IE due to oral streptococci (63±16 vs. 55±18 years, P<0.00001). The proportion of streptococcal IE due to S. bovis increased from 10.9% before 1989 to 23.3% after 1989 (P=0.0007) and was 56.7% in France as compared with 9.4% in the rest of Europe and 6.0% in the USA (P<0.00001). Patients with S. bovis IE had more comorbidity and never used intravenous drugs. Complication rates, rates of valve replacement, and mortality rates were similar in the two groups. In conclusion, this study confirmed that S. bovis IE has unique characteristics when compared to endocarditis due to oral streptococci and that it emerged in the 1990s, mainly in France, a finding that is yet unexplained.

Enterococcal prosthetic valve infective endocarditis: report of 45 episodes from the International Collaboration on Endocarditis-merged database

Enterococcal prosthetic valve infective endocarditis (PVE) is an incompletely understood disease. In the present study, patients with enterococcal PVE were compared to patients with enterococcal native valve endocarditis (NVE) and other types of PVE to determine differences in basic clinical characteristics and outcomes using a large multicenter, international database of patients with definite endocarditis. Forty-five of 159 (29%) cases of definite enterococcal endocarditis were PVE. Patients with enterococcal PVE were demographically similar to patients with enterococcal NVE but had more intracardiac abscesses (20% vs. 6%; p=0.009), fewer valve vegetations (51% vs. 79%; p<0.001), and fewer cases of new valvular regurgitation (12% vs. 45%; p=0.01). Patients with either enterococcal PVE or NVE were elderly (median age, 73 vs. 69; p=0.06). Rates of in-hospital mortality, surgical intervention, heart failure, peripheral embolization, and stroke were similar in both groups. Patients with enterococcal PVE were also demographically similar to patients with other types of PVE, but mortality may be lower (14% vs. 26%; p=0.08). Notably, 93% of patients with enterococcal PVE came from European centers, as compared with only 79% of patients with enterococcal NVE (p=0.03). Thus, patients with enterococcal PVE have higher rates of myocardial abscess formation and lower rates of new regurgitation compared to patients with enterococcal NVE, but there are no differences between the groups with regard to surgical or mortality rates. In contrast, though patients with enterococcal PVE and patients with other types of PVE share similar characteristics, mortality is higher in the latter group. Importantly, the prevalence of enterococcal PVE was higher in the European centers in this study.

Escherichia coli endocarditis: seven new cases in adults and review of the literature.

Described here are seven new cases of infective endocarditis due to Escherichia coli, including four involving prosthetic valves, followed by a review of similar cases in the literature. The review identified cases according to the modified Duke’s criteria and revealed 16 cases reported before 1960, 5 between 1960 and 1980, and 11 after 1980. Currently, patients diagnosed with E.xa0coli endocarditis are older than the patients diagnosed before 1960 (p<0.05), and they are often diabetic with underlying heart disease. Prosthetic valves are frequently involved (p<0.05), and the principal source of infection is the urinary tract. Surgery is often necessary. The mortality rate associated with this type of infection has decreased since 1960, but it remains high, with 17% calculated for the present series of seven new cases. The data presented here suggest that elderly patients with prior valve disease or prosthetic valve and E.xa0coli urinary tract infection should be examined for endocarditis.

Candida Endocarditis: Contemporary Cases from the International Collaboration of Infectious Endocarditis Merged Database (ICE-MD)

Candida infective endocarditis (IE) is increasingly common, yet most reports have been single-center reviews. We evaluated 16 patients with Candida IE nested within a cohort of 2,022 patients with IE. Prosthetic valve IE was more common in patients with Candida (50% vs 17%); mortality was 37% for patients with Candida.

Lactobacillus reuteri and Escherichia coli in the human gut microbiota may predict weight gain associated with vancomycin treatment

Background:Antibiotics, used for 60 years to promote weight gain in animals, have been linked to obesity in adults and in children when administered during early infancy. Lactobacillus reuteri has been linked to obesity and weight gain in children affected with Kwashiorkor using ready-to-use therapeutic food. In contrast, Escherichia coli has been linked with the absence of obesity. Both of these bacteria are resistant to vancomycin.Objectives and methods:We assessed vancomycin-associated weight and gut microbiota changes, and tested whether bacterial species previously linked with body mass index (BMI) predict weight gain at 1 year. All endocarditis patients treated with vancomycin or amoxicillin in our center were included from January 2008 to December 2010. Bacteroidetes, Firmicutes, Lactobacillus and Methanobrevibacter smithii were quantified using real-time PCR on samples obtained during the 4–6 weeks antibiotic regimen. L. reuteri, L. plantarum, L. rhamnosus, Bifidobacterium animalis and E. coli were quantified on stool samples obtained during the first week of antibiotics.Results:Of the193 patients included in the study, 102 were treated with vancomycin and 91 with amoxicillin. Vancomycin was associated with a 10% BMI increase (odds ratio (OR) 14.1; 95% confidence interval (CI; 1.03–194); P=0.047) and acquired obesity (4/41 versus 0/56, P=0.01). In patients treated with vancomycin, Firmicutes, Bacteroidetes and Lactobacillus increased, whereas M. smithii decreased (P<0.05). The absence of E. coli was an independent predictor of weight gain (OR=10.7; 95% CI (1.4–82.0); P=0.02). Strikingly, a patient with an 18% BMI increase showed a dramatic increase of L. reuteri but no increase of E. coli.Conclusion:The acquired obesity observed in patients treated with vancomycin may be related to a modulation of the gut microbiota rather than a direct antibiotic effect. L. reuteri, which is resistant to vancomycin and produces broad bacteriocins, may have an instrumental role in this effect.

Diagnosis of Coxiella burnetii pericarditis by using a systematic prescription kit in cases of pericardial effusion: an 8‐year experience

The detection and treatment of pericarditis remains a challenging problem, and the aetiology is unknown in 40–85% of cases [1,2]. As a result, a large proportion of cases are labelled idiopathic pericarditis. The advent of echocardiography has clarified the definition from pericarditis to pericardial effusion, which is a standardized entity. Pericardial effusion may be caused by a wide variety of infections, including those caused by Coxiella burnetii [3]. In an effort to reduce this ratio, we have previously developed [4], on 204 patients hospitalized in Marseilles with pericardial effusion, a diagnostic strategy that mandated the systematic use of a battery of non-invasive tests for the diagnosis of benign pericardial effusion. This allowed reduction of the number of pericarditis cases classified as idiopathic when compared to an intuitive prescription of tests [5]. Q fever was the main reported aetiology in our experience [4], which is particularly interesting, because it is a treatable disease. Herein, we present a 8 years of experience of Q fever pericarditis diagnosed with a systematic prescription kit in cases of pericardial effusion.

Images in cardiovascular medicine. Massive biventricular thrombosis as a consequence of myocarditis: findings from 2-dimensional and real-time 3-dimensional echocardiography.

A43-year-old man with medical history of gastroenteritis 2 weeks previously was referred to our intensive care unit for acute chest pain. At admission, the ECG showed negative T waves in V1, V2, V3, and V4 leads, and his troponin serum level was 0.6 ng/mL. His C-reactive protein level was elevated at 70 mg/L, and the serum blood count showed hyperleukocytosis with hyperlymphocytosis and thrombocytosis. Two-dimensional transthoracic echocardiogram revealed a dilated and hypokinetic left ventricle (LV) and a biventricular thrombosis (Figure 1). A dramatic and mobile apical thrombus appeared in the LV cavity; a smaller one near to the septo-basal wall was better assessed by real-time, 3-dimensional transthoracic echocardiogram (Figure 2, Movie I, and Movie II). Another thrombus was observed in the apex of the right ventricle (Figure 2, Movie III). Abdominal computed tomography scan revealed a massive splenic infarction that explained the thrombocytosis. The coronary angiogram was normal. Facing a large and mobile LV apical thrombus and a high risk of new embolization, we performed a total thrombectomy in an urgent setting through an aortotomy and both a left and right atriotomy. Despite an early postoperative recurrence of a small LV thrombus, the outcome was favorable, with disappearance of this thrombus after anticoagulation therapy, spontaneous resolution of the inflammatory syndrome, and an improvement in LV function. The presumptive final diagnosis was myocarditis complicated by biventricular thrombosis.A 43-year-old man with medical history of gastroenteritis 2 weeks previously was referred to our intensive care unit for acute chest pain. At admission, the ECG showed negative T waves in V1, V2, V3, and V4 leads, and his troponin serum level was 0.6 ng/mL. His C-reactive protein level was elevated at 70 mg/L, and the serum blood count showed hyperleukocytosis with hyperlymphocytosis and thrombocytosis. Two-dimensional transthoracic echocardiogram revealed a dilated and …