Abstract 215: Cardiac Surgeons’ Treatment Preferences for Infective Endocarditis Based on Patients’ Drug Use Status

Purpose: With the current opioid epidemic, U.S. cardiac surgeons are increasingly facing complex medical decision-making for patients with injection drug use-associated infective endocarditis (IDU-IE). Guidelines for IE recommend that normal indications for surgery be applied to IDU-IE. We evaluated cardiac surgeons’ treatment preferences for IDU and non-IDU-IE.

Methods: We sent an anonymous survey to 2398 cardiac surgeons in the U.S., of whom 254 (10.6%) responded. Of those respondents, 208 completed the questions of interest. Treatment options were presented as hypothetical clinical scenarios of IE with categorical responses. Scenarios varied based on patient drug use status, patient adherence to addiction treatment, and nature of the valve (recurrent vs. native). Treatment preferences were classified as operative vs non-operative. McNemar and Wilcoxon signed-rank tests were used to determine differences in response to scenarios. Statistical significance was defined as P< 0.05. Respondents’ demographic data were collected.

Results: In patients with native valve non-IDU-IE, 131 (63%) would operate, and 77 (37%) would not. For patients with native valve IDU-IE and adherent to addiction treatment, 76 (36.5%) would not operate, while 132 (64.5%) would. In cases of recurrent IE, 175 (93.1%) surgeons would operate on non-IDU-IE vs 55 (26.4%) on IDU-IE. The difference in treatment preference for recurrent IDU-IE vs recurrent non-IDU-IE was statistically significant by McNemar’s test (OR 0.16, CI [0.08, 0.27] P <0.001). Nearly three quarters (152, 73.1%) stated no limits on operations for non-IDU-IE in case of recurrence, vs 34 (16.5%) for IDU-IE. Others limited to 1 (83, 40.3%), 2 (79, 38.4%),or 3 (10, 4.9%) operations for IDU-IE, vs 1 (13, 6.3%), 2 (30, 14.5%) or 3 (13, 6.3%) for non-IDU-IE. Wilcoxon signed-rank test showed a statistically significant difference in the limit number of operations for recurrent non-IDU-IE vs recurrent IDU-IE (z = -4.9, P <0.001). Most respondents report having denied operating for IDU-IE (132, 63.5%), unlike 49 (23.6%) who have not, while 24 (11.5%) have delayed operating. Most respondents who have denied operating did not refer patients to a different surgeon or hospital. Active drug use and recurrent endocarditis were the leading reasons cited for denying operating for IDU-IE.

Conclusion: Cardiac surgeons have different treatment preference for IE based on patients’ underlying substance use disorder diagnosis. Surgeons are less likely to offer operation for primary and recurrent valve infection to patients with IDU-IE, and patient adherence to addiction treatment may be protective against denial of operation. This indicates a need for increased education of cardiac surgeons on addiction, and a multidisciplinary approach to care in order to address the disparity in treatment and management between IDU-IE and non-IDU-IE.