Mortality and timing of surgery in the left-sided infective endocarditis: an Italian multicentre study

OBJECTIVES: Observational studies on early surgery in infective endocarditis have conflicting results. This study aims to compare the treatment strategies for early surgery (within 2 weeks of diagnosis) and late surgery/medical therapy in terms of survival among patients with the left-sided infective endocarditis. METHODS: This study included patients with the left-sided infective endocarditis registered between 2006 and 2010 in the Italian Registry of Infective Endocarditis (RIEI). A Cox proportional hazards model was used to estimate the effect of these treatment strategies on overall survival and included sociodemographic and clinical characteristics associated with treatment, risk factors for mortality and early surgery as a time-dependent covariate to avoid indication and immortal time biases. RESULTS: Among the 502 patients included, 184 (36.7%) underwent early surgery. Of the remaining 318 patients, 138 underwent late surgery. The early surgery group had fewer patients with comorbidities and with enterococcus as the causative microorganism, but this group had more complicated cardiac conditions. No difference in mortality risk was estimated between the treatment groups including early surgery as time-dependent variables (adjusted hazard ratio  = 0.95, 95% confidence interval 0.55-1.63), while a distorted and overestimated beneficial effect of surgery was estimated considering surgery as a non-time-dependent variable (adjusted hazard ratio 0.41, 95% confidence interval 0.25-0.70). CONCLUSIONS: Our study did not confirm a better overall survival in patients undergoing early surgery. However, even with the use of statistical techniques to control biases, we could not draw definitive conclusions that early surgery is not beneficial. Our results need to be assessed by randomized trials before any changes in clinical practice can be recommended.