Thrombocytopenia, a new marker of bad prognosis in patients with infective endocarditis

Abstract

Endocarditis 863 had chronic renal failure (23.9% vs 6.9%; p<0.001) and chronic obstructive pulmonary disease (14.6% vs 7.8%; p=0.019) more commonly. Enterococcus (14.9% vs 7.4%; p=0.011) and Streptococcus bovis (8.8% vs 3.8%; p=0.024) were isolated more frequently in diabetics. Septic shock (29.2% vs 16.4%; p=0.005) was more common among diabetics, and mortality (43.5% vs 30%; p=0.008) was higher in this group. Considering the treatment of DM, septic shock (33.3%; p=0.011) and death (50.8%; p=0.012) were more frequent in patients with oral antidiabetics. Multivariate analysis showed that DM had an independent association with development of septic shock (OR 2.282; 95% CI 1.186-4.393), but it was not a predictor of in-hospital mortality. Conclusions: In left-sided IE, DM is independently associated with the development of septic shock, especially in patients treated with oral antidiabetics. Aim: Thrombocytopenia in patients with sepsis has been described as a marker of bad prognosis. Our purpose was to analyze if the presence of thrombocytope-nia has a prognostic impact in patients with left-sided infective endocarditis (IE). Methods: We analyzed 698 consecutive episodes of IE prospectively recruited in three referral hospitals between 1996 and 2011. They were classified in two groups: Group I (n=213) episodes of IE who had thrombocytopenia in blood analysis at admission; and Group II (n=485) those who did not have it. Thrombocy-topenia was defined as a platelet count below 150000/μl. Results: The age and gender distribution were similar in both groups. Regarding to comorbidities, diabetes (25.4% vs. 18.3%, p=0.03), chronic anemia (24.5% vs. 17.6%, p= 0.03) and immunosuppression (11.8% vs. 4.5%, p<0.001) were more frequent in Group I. We found a greater percentage of prosthetic valves in Group II, while patients without any previous cardiopathy were more prevalent in Group I (29.6% vs. 21.6%, p=0.02). S. aureus (28.6% vs. 10.9%, p<0.001) and Gram negative bacilli (8% vs. 3.1%, p=0.004) were more frequently isolated in Group I, were S. viridans (8% vs. 15.1%, p=0.01) and culture negative IE (10.3 vs. 16.3, p=0.04) were more common in Group II. Clinical presentation with neurological (p<0.001), renal (p=0.05), cutaneous (p<0.001) manifestations and septic shock (12.2% vs. 3.5%, p<0.001) at admission were more frequent in Group I. These patients showed also more frequency of hemorrhagic skin lesions (11% vs. 4%, p=0.002), splenomegaly (14.2% vs 7.4%, p=0.005) and coma (5.6% vs. 0.8%, p<0.001). Presence of pseudoaneurysms were more frequently found in Group II (12.2% vs. 18.9%, p=0.03). There were no differences in other echocardiographic findings (valvular insufficiency, presence and size of vegetations). During hospitalization , persistent signs of infection (44.5% vs. 31.8%, p=0.001) and septic shock were more frequent in Group I (27% vs. 13.5%, p<0.001). The need of surgery was similar in both groups (53.5% vs. 56.1%, p=0.535), but much higher mortality was observed in patients with thrombocytopenia (40.4% vs. 25.3%, p<0.001). Conclusions: Thrombocytopenia at admission of patients with IE identifies a highest risk group. Causative microorganisms are more virulent in this group. These patients develop septic shock more frequently and showed higher mortality. Purpose: Renal failure (RF) has been proven as a risk factor of death in left-sided infective endocarditis (LSIE). The objective of this work is to analyze the incidence of de novo renal failure and the factors involved in the development of renal failure in a cohort of patients with LSIE. Methods: De novo renal failure was defined as the absence of creatinine levels over 2 mg/dl at admission, due to either chronic renal failure or de novo condition , and rise of the creatinine levels over 2 mg/dl during the first month after admission. Among 705 patients with LSIE diagnosed consecutively in three ter-tiary centers from 1996 to 2012, 292 had normal renal function at admission and did not underwent surgery or die within one month from the diagnosis. Of them, 94 developed de novo RF (32%), whereas 192 did not (68%). We performed a univariate analysis of 84 variables and then a multivariate analysis to determine the independent risk factors for the development of RF. Results: In the univariate analysis, the variables associated with the development of RF were: early onset prosthetic valve IE (53% vs 27%, p=0.009), heart failure at admission (43 vs 25%, p=0.002), abscence of previous fever (64 vs 76% p=0.04), abdominal pain (19 vs 7%, p= 0,002), S. aureus (12 vs 21%, p=0.03), coagulase-negative Staphylococci (10 vs 19% p=0.03), treatment with vancomycin (51 vs 36%, p= 0.02) and gentamicin (67 vs 79%, p=0.04). The development of RF was also associated with new onset heart failure symptoms (38 vs 60%, p< 0.001). In hospital mortality was more frequent in the group with de novo RF (35% vs 11% p<0.001). In the multivariate analysis, the presence of heart failure at admission (OR 2.2; 95%IC 1.3-3.9), infection by S. aureus (OR 2.3; 95% CI 1.1-4.7), coagulase-negative Staphylococci (OR 2.5; 95% CI 1.05-4.9) and the use of gen-tamicin (OR 2; 95% CI 1.07-3.7) were associated with higher risk of de novo renal failure. Conclusions: The development of de novo renal failure in patients with LSIE is related to the presence of heart failure at admission, the infection by S. aureus or coagulase-negative Staphylococci and the treatment with gentamicin. Neurologic complications (NC) are a major cause of morbidity and mortality in pts with infective endocarditis (IE) and may prevent pts from receiving valve surgery. Methods: Among 399 cases of Duke-Li definite left-sided IE collected during a one-year prospective population-based survey, 100 had at least one symptomatic neurological complication (SNC). NC included transient ischemic attack, ischemic stroke, cerebral hemorrhage, meningitis, brain abscess, intracranial my-cotic aneurysm and unprecised cerebral event. A neuroimaging procedure was performed in 183 of the remaining pts which diagnosed 35 additional asymp-tomatic (AsNC) and was considered normal in 148 pts (NoNC). Characteristics of pts with any NC, SNC, AsNC and NoNC were compared. Results: Among pts with SNC, 40 episodes occurred after the beginning of IE antibiotic therapy (21 (52%) during the first 48 hours). Pts with NC (n=135) had more frequently a prosthetic valve (28% vs 16%), extra cerebral embolic events, mitral IE (especially calcified mitral annulus), septic shock, elevated CRP, larger vegetation , lower LVEF and lower rate of significant valvular regurgitation than the 148 NoNC patients. SNC pts were older, had higher levels of CRP, lower frequency of significant valvular regurgitation and had more often IE due to Staphylococcus aureus (38% vs 14%) than AsNC. AsNC pts were younger, more often had community acquired IE, other embolic events and intracardiac abscess than NoNC pts. Mechanical prosthetic IE (OR 3.4 [1.6-7.5], p=0.007), S.aureus IE (OR 1.8 [1.0-3.2], p=0.03) and mitral IE (OR 2.0 [1.2-3.4], p=0.008) were predictive of the occurrence of NC. Among the 135 NC pts, age, renal failure, septic shock and S.aureus as responsible microorganism were predictive of in-hospital (33%) and 1-year mortality (38%). Furthermore, a symptomatic NC was also predictive of 1-year mortality. Rate of surgery was not different between pts with NC and without NC (52% vs 57%) but was significantly higher in AsNC than in SNC (77% vs 43%). In hospital mortality was higher in pts with NC than in noNC pts (33% vs 17%, p=0.001) and was higher in SNC than in AsNC (42% vs 9%, p=0.0003). Among the 135 pts with NC, 95 had a surgical indication (71%) which was performed in 70 of them (mortality 20%) and not performed in 25 (mortality 68%). Conclusion: The presence of NC is associated with a poor prognosis when it is symptomatic. However, when NC is discovered on systematic neuroimaging, it is associated with a very high rate of surgery and a better prognosis than that of pts with SNC and even without NC, suggesting a protective role of surgery. Purpose: Injection drug users (IDUs) account for a disproportionally large number of the hospitalisations for infective endocarditis (IE). However the frequency of previous IE and cardiac abnormalities in a community-based cohort of active IDUs is unknown. The aim of the present study was to assess cardiac abnormalities with echocardiography in a community-based cohort of active IDUs focusing on signs of previous IE. Methods: A community-based cohort of non-hospitalised, active IDUs was screened with transthoracic echocardiography at five different supervised injection facilities in Copenhagen, Denmark. Participation was anonymous and data on substance abuse and heart disease were self-reported. Valves were described as normal, with regurgitation, thickening of leaflets and/or with vegetations. Echocar-diographic findings were summarized as "probable previous IE" if two out of three criteria (moderate to severe regurgitation/thickening of leaflets/vegetations) were present for at least one valve.