John L. Klein

Clinical Presentation, Etiology, and Outcome of Infective Endocarditis in the 21st Century: The International Collaboration on Endocarditis–Prospective Cohort Study

BACKGROUNDnWe sought to provide a contemporary picture of the presentation, etiology, and outcome of infective endocarditis (IE) in a large patient cohort from multiple locations worldwide.nnnMETHODSnProspective cohort study of 2781 adults with definite IE who were admitted to 58 hospitals in 25 countries from June 1, 2000, through September 1, 2005.nnnRESULTSnThe median age of the cohort was 57.9 (interquartile range, 43.2-71.8) years, and 72.1% had native valve IE. Most patients (77.0%) presented early in the disease (<30 days) with few of the classic clinical hallmarks of IE. Recent health care exposure was found in one-quarter of patients. Staphylococcus aureus was the most common pathogen (31.2%). The mitral (41.1%) and aortic (37.6%) valves were infected most commonly. The following complications were common: stroke (16.9%), embolization other than stroke (22.6%), heart failure (32.3%), and intracardiac abscess (14.4%). Surgical therapy was common (48.2%), and in-hospital mortality remained high (17.7%). Prosthetic valve involvement (odds ratio, 1.47; 95% confidence interval, 1.13-1.90), increasing age (1.30; 1.17-1.46 per 10-year interval), pulmonary edema (1.79; 1.39-2.30), S aureus infection (1.54; 1.14-2.08), coagulase-negative staphylococcal infection (1.50; 1.07-2.10), mitral valve vegetation (1.34; 1.06-1.68), and paravalvular complications (2.25; 1.64-3.09) were associated with an increased risk of in-hospital death, whereas viridans streptococcal infection (0.52; 0.33-0.81) and surgery (0.61; 0.44-0.83) were associated with a decreased risk.nnnCONCLUSIONSnIn the early 21st century, IE is more often an acute disease, characterized by a high rate of S aureus infection. Mortality remains relatively high.

Candida infective endocarditis

Candida infective endocarditis (IE) is uncommon but often fatal. Most epidemiologic data are derived from small case series or case reports. This study was conducted to explore the epidemiology, treatment patterns, and outcomes of patients with Candida IE. We compared 33 Candida IE cases to 2,716 patients with non-fungal IE in the International Collaboration on Endocarditis—Prospective Cohort Study (ICE-PCS). Patients were enrolled and the data collected from June 2000 until August 2005. We noted that patients with Candida IE were more likely to have prosthetic valves (pu2009<u20090.001), short-term indwelling catheters (pu2009<u20090.0001), and have healthcare-associated infections (pu2009<u20090.001). The reasons for surgery differed between the two groups: myocardial abscess (46.7% vs. 22.2%, pu2009=u20090.026) and persistent positive blood cultures (33.3% vs. 9.9%, pu2009=u20090.003) were more common among those with Candida IE. Mortality at discharge was higher in patients with Candida IE (30.3%) when compared to non-fungal cases (17%, pu2009=u20090.046). Among Candida patients, mortality was similar in patients who received combination surgical and antifungal therapy versus antifungal therapy alone (33.3% vs. 27.8%, pu2009=u20090.26). New antifungal drugs, particularly echinocandins, were used frequently. These multi-center data suggest distinct epidemiologic features of Candida IE when compared to non-fungal cases. Indications for surgical intervention are different and mortality is increased. Newer antifungal treatment options are increasingly used. Large, multi-center studies are needed to help better define Candida IE.

Non-HACEK Gram-Negative Bacillus Endocarditis

Context Infective endocarditis due to non-HACEK organisms has been considered to be associated with injection drug use. Contribution Analysis of 2761 cases of patients with infective endocarditis from an international collaborative of 61 hospitals found that non-HACEK organisms account for fewer than 2% of the cases, and that most patients with non-HACEK endocarditis had infections associated with health care. Of patients with non-HACEK infections, 59% had implanted endovascular devices or prosthetic valves, but only 4% had injection drug use. More than one half of patients with non-HACEK infections required cardiac surgery and 24% died. Implication Infective endocarditis due to non-HACEK organisms is a rare but frequently fatal condition. It is much more frequently associated with implanted endovascular devices than with injection drug use. The Editors Infective endocarditis caused by non-HACEK (species other than Haemophilus species, Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella corrodens, or Kingella species) gram-negative bacilli is a rare and poorly characterized disease. The literature describing non-HACEK gram-negative bacillus endocarditis primarily consists of several small case series from the 1970s and 1980s of outbreaks in injection drug users in large urban areas, such as Detroit (1, 2), Cleveland (3), and San Francisco (4, 5). As a result, endocarditis due to non-HACEK gram-negative bacilli has been considered to be almost exclusively associated with injection drug use (6, 7). In contrast to this reporting bias, however, non-HACEK gram-negative bacillus endocarditis has been occasionally reported to be a nosocomial problem, particularly in patients with early endocarditis after cardiac surgery (811). The International Collaboration on Endocarditis Prospective Cohort Study (ICE-PCS) database was created in 1999. From 1 January 2000 to 31 August 2005, 2761 patients with definite endocarditis from 61 centers in 28 countries were prospectively enrolled. This resource offers a unique opportunity to evaluate the epidemiology, characteristics, and outcome of endocarditis due to non-HACEK gram-negative bacilli in a large, contemporary, and international cohort of well-characterized patients with endocarditis. Methods The International Collaboration on Endocarditis Prospective Cohort Study Hospitalized patients with endocarditis (12) were identified prospectively by using site-specific procedures to ensure consecutive enrollment. Informed consent (oral or written) was obtained from all patients according to local institutional review board or ethics committee instructions. A standard case report form containing 275 variables was completed for each patient on enrollment at the participating site. The ICE-PCS database is maintained at the Duke Clinical Research Institute, Durham, North Carolina, which serves as the coordinating center for the ICE studies, with approval from the institutional review board. We included all patients with endocarditis from sites that met performance criteria for participation. These site criteria included 1) minimum enrollment of 12 cases per year in a center with access to cardiac surgery, 2) the presence of patient identification procedures to ensure consecutive enrollment and to minimize ascertainment bias (as described elsewhere) (13, 14), 3) high-quality data with query resolution, and 4) institutional review board or ethics committee approval or waiver based on local standards. All patients from sites that did not meet these criteria (totaling 494 case-patients from 14 sites) were excluded. Sample We included patients who had both definite endocarditis according to the modified Duke criteria (12) and isolation of a pure culture of an aerobic gram-negative bacillus from the bloodstream or valve. To ensure that the diagnosis of gram-negative endocarditis was accurate, the following additional criteria were applied when interpreting the blood culture results: 1) the patients bacteremia had to meet the definition for persistently positive blood cultures when applying the modified Duke criteria; 2) a single blood culture positive for a gram-negative organism was not considered to constitute a minor microbiological criterion when applying the modified Duke criteria; and 3) patients with endocarditis due to anaerobes, Brucella species, HACEK organisms, or other fastidious gram-negative pathogens (for example, Pasteurella species) or polymicrobial infections were excluded. Definitions We used published definitions of health carerelated variables (15, 16). Nonnosocomial health careassociated infection was defined as a health careassociated infection that was not acquired as a hospital inpatient (for example, hemodialysis, outpatient cancer chemotherapy, or receipt of intravenous antibiotics at home) (16). A nosocomial infection was defined as a health careassociated infection that was acquired after at least 48 hours as a hospital inpatient. Prosthetic endocarditis was defined as endocarditis involving a prosthetic heart valve or implanted endovascular device, such as a permanent cardiac pacemaker, cardioverter defibrillator, or aortic stent. Statistical Analysis Patients with definite non-HACEK gram-negative bacillus endocarditis were compared with all other patients with definite endocarditis in the ICE-PCS database. Continuous variables are presented as medians and 25th and 75th percentiles. Categorical variables are presented as frequencies and percentages of the specified group. Univariable comparisons were made by using the Wilcoxon rank-sum test or the chi-square test as appropriate. For all tests, a P value of 0.05 or less was considered statistically significant. Missing data for each variable were excluded from the denominator as indicated in Table 1. All statistical analyses were performed by using SAS software (version 8.2, SAS Institute, Cary, North Carolina). Table 1. Frequency of Individual Duke Criteria among 49 Patients with Non-HACEK Gram-Negative Bacillus Endocarditis* Role of the Funding Source The study did not receive funding. Results Of the 2761 patients with definite endocarditis, 49 (1.8%) had endocarditis due to non-HACEK gram-negative bacilli. Twenty-six of these patients (53%) were enrolled from Europe; 11 (22%) from North America; and the remainder from South America, New Zealand, Australia, the Middle East, and Asia. Patient enrollment was constant throughout the study period. Characteristics of Non-HACEK Gram-Negative Bacillus Endocarditis Patients with non-HACEK gram-negative bacillus endocarditis were more likely to have had symptoms for more than 1 month than were patients infected with other pathogens (90% [95% CI, 82% to 98%] vs. 77% [CI, 75% to 79%], respectively; P= 0.035) (Table 2). Injection drug use was uncommon in patients with non-HACEK gram-negative bacillus endocarditis and in patients with endocarditis due to other organisms (4% [CI, 0% to 9%] vs. 10% [CI, 9% to 11%]; P= 0.20). In contrast, health care contact was a statistically significant risk factor for non-HACEK gram-negative bacillus endocarditis (57% [CI, 43% to 71%] vs. 30% [CI, 28% to 32%]; P< 0.001), largely because the proportion of nosocomial infections was higher in the non-HACEK gram-negative bacillus endocarditis group (39% [CI, 25% to 53%] vs. 14% [CI, 13% to 15%]; P< 0.001). The Figure shows the routes of acquisition of non-HACEK gram-negative bacillus endocarditis compared with Staphylococcus aureus endocarditis (15) and all other causes of endocarditis in the ICE-PCS database. Table 2. Characteristics of Patients with Non-HACEK Gram-Negative Bacillus Infective Endocarditis and Those with Other Causes of Endocarditis* Figure. Routes of acquisition among patients with definite endocarditis due to non-HACEK gram-negative bacilli, Staphylococcus aureus , and other pathogens. HACEK = Haemophilus species, Actinobacillus actinomycetemcomitans, Cardiobacterium hominis, Eikenella corrodens, or Kingella species. Implanted endovascular devices were more common in patients with non-HACEK gram-negative bacillus endocarditis than in patients with other pathogens (29% [CI, 16% to 42%] vs. 11% [CI, 10% to 12%]; P< 0.001). Patients with non-HACEK gram-negative bacillus endocarditis were also statistically significantly more likely than patients with other causes of endocarditis to have a presumed source of infection involving the genitourinary or nonoral gastrointestinal tract (35% [CI, 22% to 48%] vs. 12% [CI, 11% to 13%]; P< 0.001). A nondental invasive procedure within 60 days before symptom onset was more likely in patients with non-HACEK gram-negative bacillus endocarditis than in patients with other causes of endocarditis (38% [CI, 24% to 52%] vs. 19% [CI, 18% to 20%]; P= 0.002). Intracardiac abscesses were statistically significantly more common in patients with non-HACEK gram-negative bacillus endocarditis than in patients with endocarditis due to other organisms (25% [CI, 13% to 37%] vs. 14% [CI, 13% to 15%]; P= 0.034). The in-hospital mortality rate was 24% (CI, 12% to 36%) for patients with non-HACEK gram-negative bacillus endocarditis and 17% (CI, 16% to 18%) for patients with other causes of endocarditis (P= 0.190). Of the 49 patients with non-HACEK gram-negative bacillus endocarditis, 20 (41%) had native-valve endocarditis and 29 (59%) had prosthetic endocarditis. All 49 cases were confirmed as definite endocarditis by the modified Duke criteria: 22 (45%) were histopathologically (16 patients [33%]) or macroscopically (at surgery in 6 patients [12%]) confirmed (Table 1). Of the 16 patients with pathologic confirmation, 8 had valve cultures, 2 had device cultures, and 1 had an aortic aneurysm culture. Microbiology of Non-HACEK Gram-Negative Bacillus Endocarditis The most common pathogens in patients with non-HACEK gram-negative bacillus endocarditis were Escherichia coli (14 patients [29%]) and Pseudomonas aeruginosa (11 patients [22%]). Othe

Fever in returned travellers presenting in the United Kingdom: recommendations for investigation and initial management.

International travel is increasing. Most physicians and general practitioners will encounter returned travellers with fever and the majority of travel-related infection is associated with travel to the tropics. In those returning from the tropics malaria must always be excluded, and HIV considered, from all settings. Common causes of non-malarial fever include from Africa rickettsial diseases, amoebic liver abscess and Katayama syndrome; from South and South East Asia, enteric fever and arboviral infection; from the Middle East, brucellosis and from the Horn of Africa visceral leishmaniasis. Other rare but important diseases from particular geographical areas include leptospirosis, trypanosomiasis and viral haemorrhagic fever. North and South America, Europe and Australia also have infections which are geographically concentrated. Empirical treatment may have to be started based on epidemiological probability of infection whilst waiting for results to return. The evidence base for much of the management of tropical infections is limited. These recommendations provide a pragmatic approach to the initial diagnosis and management of fever in returned travellers, based on evidence where it is available and on consensus of expert opinion where it is not. With early diagnosis and treatment the majority of patients with a potentially fatal infection related to travel will make a rapid and full recovery.

Mutations in the Plasmodium falciparum cytochrome b gene are associated with delayed parasite recrudescence in malaria patients treated with atovaquone-proguanil.

BackgroundFixed-dose combination antimalarial drugs have played an increasingly important role in the treatment and chemoprophylaxis of falciparum malaria since the worldwide failure of monotherapy with chloroquine. Atovaquone-proguanil is one such combination drug used both for prophylaxis in travellers, and for treatment of acute malaria cases in European hospitals and clinics.MethodsA series of eight atovaquone-proguanil treatment failures and two prophylaxis breakthroughs from four UK hospitals from 2004–2008 were analysed for evidence of mutations in the pfcyt-b gene, previously found to be associated with failure of the atovaquone component.ResultsParasites carrying pfcyt-b mutations were found in five falciparum malaria patients with recrudescent parasitaemia occurring weeks after apparently successful treatment of a primary infection with atovaquone-proguanil. Four of these cases carried parasites with the Tyr268Cys mutation in pfcyt-b, previously reported in two French patients with malaria. In contrast, mutations in pfcyt-b were not found in three patients treated with atovaquone-proguanil who exhibited delayed clearance of the primary infection, nor in two returning travellers with malaria who had used the combination for prophylaxis. Using current and previously published data, mean time to recrudescence of parasites carrying pfcytb codon 268 mutations was estimated as 28.0 days after treatment (95% C.I. 23.0 – 33.0 days), whereas treatment failures without codon 268 mutations received rescue treatment an average of 4.71 days after initial AP treatment (95% C.I. 1.76 – 7.67 days).ConclusionGenetically-determined parasite resistance to atovaquone is associated with delayed recrudescence of resistant parasites three weeks or more after initial clearance of parasitaemia by atovaquone/proguanil therapy. The 268-Cys allele of pfcyt-b may have been overlooked in previous studies of atovaquone-proguanil treatment failure as it is not detected by current RFLP methods.

Non-tuberculous slow-growing mycobacterial pulmonary infections in non-HIV-infected patients in south London

Background: UK data on slow-growing non-tuberculous mycobacterial (NTM) pulmonary infections are sparse and there is little consensus on optimal treatment regimens. Methods: This was a retrospective study of NTM pulmonary infections in a London teaching hospital. Inclusion criteria were culture of slow-growing mycobacteria between 2000 and 2007, age > 18 y, HIV-negative, and meeting American Thoracic Society criteria. Results: Fifty-seven patients were included; 68% were males and the median age was 61 y. Predisposing factors were smoking (70%), alcohol abuse (28%), and chronic obstructive pulmonary disease (37%). Cavitation (56%) and infiltrates (42%) were common radiological findings. The predominant organism was Mycobacterium kansasii (70%). Ninety-three percent of patients with M. kansasii, 63% with Mycobacterium avium intracellulare, 60% with Mycobacterium malmoense, and 25% with Mycobacterium xenopi had clinical disease. Of the 57 patients, 37 were treated and had follow-up data available. Most patients received 3 drugs: rifampicin, ethambutol, and clarithromycin or ciprofloxacin for at least 9 months. Thirty percent experienced drug side effects. M. kansasii treatment had a 100% cure and 10% relapse rate, but 15% died. Conclusions: M. kansasii was the most common NTM and its isolation was predictive of clinical disease. Compared with other studies, treatment with 3 agents had a similar rate of cure and did not appear to reduce the relapse rate of disease, but did increase the risk of side effects.

Melioidosis Acquired by Traveler to Nigeria

We describe melioidosis associated with travel to Nigeria in a woman with diabetes, a major predisposing factor for this infection. With the prevalence of diabetes projected to increase dramatically in many developing countries, the global reach of melioidosis may expand.

Identification of the 'Streptococcus anginosus group' by matrix-assisted laser desorption ionization--time-of-flight mass spectrometry.

Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) provides rapid, accurate and cost-effective identification of a range of bacteria and is rapidly changing the face of routine diagnostic microbiology. However, certain groups of bacteria, for example streptococci (in particular viridans or non-haemolytic streptococci), are less reliably identified by this method. We studied the performance of MALDI-TOF MS for identification of the Streptococcus anginosus group (SAG) to species level. In total, 116 stored bacteraemia isolates identified by conventional methods as belonging to the SAG were analysed by MALDI-TOF MS. Partial 16S rRNA gene sequencing, supplemented with sialidase activity testing, was performed on all isolates to provide gold standard identification against which to compare MALDI-TOF MS performance. Overall, 100u200a% of isolates were correctly identified to the genus level and 93.1u200a% to the species level by MALDI-TOF MS. However, only 77.6u200a% were correctly identified to the genus level and 59.5u200a% to the species level by a MALDI-TOF MS direct transfer method alone. Use of a rapid in situ extraction method significantly improved identification rates when compared with the direct transfer method (P<0.001). We recommend routine use of this method to reduce the number of time-consuming full extractions required for identification of this group of bacteria by MALDI-TOF MS in the routine diagnostic laboratory. Only 22u200a% (1/9) of Streptococcus intermedius isolates were reliably identified by MALDI-TOF MS to the species level, even after full extraction. MALDI-TOF MS reliably identifies S. anginosus and Streptococcus constellatus to the species level but does not reliably identify S. intermedius.

Clonal structure of Streptococcus sanguinis strains isolated from endocarditis cases and the oral cavity

A collection of Streptococcus sanguinis strains from patients with endocarditis (n = 21) and from the oral cavity (n = 34) was subjected to a multi-locus sequence typing analysis using seven housekeeping genes, carbamoyl-phosphate synthetase (carB), Co/Zn/Cd efflux system component (czcD), d-alanyl-d-alanine ligase (ddl), DNA polymerase III (dnaX), glucose-6-phosphate dehydrogenase (gdh), DNA-directed RNA polymerase, beta subunit (rpoB) and superoxide dismutase (sodA). The scheme was expanded by the inclusion of two the putative virulence genes, bacitracin-resistance protein (bacA) and saliva-binding protein (ssaB), to increase strain discrimination. Extensive intra-species recombination was apparent in all genes but inter-species recombination was also apparent with strains apparently harbouring gdh and ddl from unidentified sources and one isolate harboured a sodA allele apparently derived from Streptococcus oralis. The recombination/mutation ratio for the concatenated housekeeping gene sequences was 1.67 (95% confidence limits 1.25-2.72) and for the two virulence genes the r/m ratio was 3.99 (95% confidence limits 1.61-8.72); recombination was the major driver for genetic variation. All isolates were distinct and the endocarditis strains did not form distinct sub-clusters when the data were analysed using ClonalFrame. These data support the widely held opinion that infecting S. sanguinis strains are opportunistic human pathogens.

Gonococcal endocarditis: forgotten but not quite gone

During a routine appointment with her general practitioner in May 2003 a 66 year old lady pointed out a rash on her lower limbs which had been present for several days. She was otherwise well and her only past medical history was of treated hypertension. Blood tests performed at that stage revealed an elevated white cell count (26.2 /10/L) with a marked neutrophilia, a normal platelet count and an erythrocyte sedimentation rate of 92 mm/hr, prompting referral to hospital. At the time of admission nine days later the rash was still her only complaint. On examination she had a temperature of 39.5 degrees Celsius with a normal pulse rate and blood pressure. Auscultation of her heart revealed a soft ejection systolic murmur in the aortic area with no radiation. She was fully orientated and had no neck stiffness or photophobia. Examination of respiratory, gastro-intestinal and neurological systems was normal. There was a raised purpuric rash on her legs and arms together with several large pustules around her ankles (figure). Investigations revealed a continuing leucocytosis and a C-reactive protein (CRP) of 139 mg/L. The chest radiograph was normal. Initially meningococcal infection was suspected and she was commenced on intravenous ceftriaxone. However, three out of five blood culture sets subsequently grew fully sensitive N. gonorrhoeae . An aspirate of one of the pustules, taken after the first dose of antibiotics, was sterile. Assessment by the genito-urinary medicine department confirmed a lack of genito-urinary symptoms and led to empirical chlamydia treatment and contact tracing. After two days of treatment she had defervesced but on day four of her inpatient stay a soft early diastolic murmur, louder in inspiration, was noted. Trans-thoracic echocardiography that day confirmed the diagnosis of endocarditis showing a large mobile mass attached to the pulmonary valve and prolapsing through it. Her antibiotic therapy was switched to intravenous benzylpenicillin 1.2g four hourly. After two weeks of parenteral antibiotics her symptoms had resolved, her white cell count had normalised and her CRP had fallen to 21 mg/L. She discharged herself from hospital against medical advice after 16 days of intravenous therapy and was subsequently lost to follow up.