Efficacy of Primary Surgical Versus Medical Intervention for Treatment of Left-Sided Infective Endocarditis

Abstract

Left-sided staphylococcal, streptococcal, and enterococcal infective endocarditis (IE) is associated with poor clinical outcomes. Our primary aim is to compare clinical outcomes of staphylococcal, streptococcal, and enterococcal IE patients who undergo valve replacement surgery and outcomes of patients who are treated solely with antibiotics. All patients were treated medically or surgically for left-sided staphylococcal, streptococcal, or enterococcal IE at our institution from 1998 to 2014 and were retrospectively studied. The primary outcome of interest was 30-day and 1-year mortality, and secondary outcomes included posttreatment septic shock, embolic events, stroke, and end-stage renal disease at 30 days. Inverse probability treatment weights, derived from propensity scores, were used to balance the medical and surgical cohorts across clinical risk factors, The Society of Thoracic Surgeon scores, and pathogens. Outcomes were compared comprehensively and in a staphylococcal-only subanalysis. Study population consisted of 245 surgical patients and 164 medical patients. Mortality at 30 days was higher in the medical cohort, both in aggregate and for staphylococcal only (all, 7% vs 16%, P < .001; staphylococcal only, 7% vs 22%, P < .001). Surgical patients had a higher incidence of septic shock and renal dysfunction; however, stroke and embolic events at 30 days were not different between cohorts. Cox survival analysis demonstrated that surgical treatment provided a 1-year survival benefit, with a hazard ratio of 0.48 (95% confidence interval, 0.36 to 0.64) that was robust regardless of pathogen. Compared with medical management, valve replacement surgery in patients with left-sided staphylococcal, streptococcal, or enterococcal IE appears to confer a survival advantage at 30 days and 1 year.