The objectives of this study were to describe the changing epidemiology of gram-negative infective endocarditis (GNIE) and to identify factors associated with treatment failure and death.

Adult patients with GNIE were included if they met modified Duke criteria for definitive infective endocarditis (IE) between April 2010 and December 2021. Patients were identified using Boolean search terms. Clinical failure was a defined as a composite of all-cause 42-day mortality or microbiologic failure. All analyses were performed using Stata, version 15.1.

One-hundred twenty-three patients were included. The most common pathogens were spp. (43%), (21%), and spp. (14%). Fifty-two percent of cases were among persons who injection drugs (PWID; n = 64), for whom spp. (70%) was the most common cause of GNIE. Overall, patients infected with had higher microbiologic failure rates than other patients (23% vs 6%; = .004). Patients who received combination therapy (n = 53) had comparable median lengths of stay (23 vs 19.5 days; = .412), microbiologic failure rates (11.3% vs 7.1%; = .528), clinical failure rates (18.9% vs 22.9%; = .592), and 90-day mortality rates (13.2% vs 25.7%; = .088) as those treated with monotherapy. After applying stepwise logistic regression, male gender, Pitt Bacteremia Score, and not receiving surgical intervention despite a surgical indication were associated with clinical failure.

This is the first study to identify spp. as the most common etiology of GNIE, which was particularly true among PWID. Microbiologic failures occurred most commonly among , and use of combination antimicrobial therapy did not improve clinical outcomes.