The Utility Of Digoxin In Reducing Recurrent Gastrointestinal Bleeding And Significant Epistaxis Post-Left Ventricular Assist Device

Background To assess the impact of Digoxin (Dig) utilization on recurrent gastrointestinal bleeding (rGIB) and severe epistaxis, which are both common complications in patients implanted with left ventricular assist devices (LVAD). Literature suggests that Dig utilization may reduce the risk and incidence of initial GIB, however scare literature is present regarding the of effect of Dig on rGIB and epistaxis. Methods Our initial cohort consisted of 346 patients implanted with LVADs on Dig use who survived index admission for LVAD implantation. Subsequent analyses included two separate sub-cohorts; rGIB (n=77) and epistaxis (n=84). All episodes of epistaxis and rGIB were captured up to 2 years of follow-up. Severe epistaxis was defined as one requiring admission with ENT consultation and at least a 1-point drop in hemoglobin, while moderate was defined as an encounter in the ED/clinic to address epistaxis with intervention. Other degrees of epistaxis were classified as mild and excluded from the analysis. Dig use was defined as dig prescribed at the time of rGIB or from the time of LVAD implantation (for epistaxis) with at least 30 days of continuous use. Clinical events as a function of time were investigated using Kaplan-Meier curve analysis. To assess influence of digoxin utilization on clinical outcomes, univariate and multivariate Cox regression analysis was utilized. Results Mean age of the rGIB cohort was 62 years (±10) and epistaxis 57 years (±14). Dig was prescribed in 41% of the rGIB and 42% of the epistaxis cohorts. A total of 77 patients experienced at least one GIB (22 % of study cohort) with 39 patients (11 % of study cohort) experiencing at least one episode of recurrent GIB. The median time from LVAD implant to first GIB was 62 days (IQR: 26, 231 days) and from first to second GIB was 57 days (IQR: 20, 163 days). In a multivariate analysis after controlling for gender, white blood cell count and beta blocker use at 6 months, Dig use was associated with a reduced risk of rGIB (Fig: HR: 0.40; 95% CI: 0.17-0.93, p=0.03). A total of 84 patients experienced epistaxis, 18 patients had severe epistaxis, while 12 patients had moderate epistaxis and 54 patients had mild epistaxis. The median time from LVAD implantation to first episode of severe epistaxis was 90 days (IQR: 55, 146 days) and moderate epistaxis was 168 days (IQR: 102, 322 days). In a univariate analysis Dig use was not associated with a reduced risk of severe epistaxis (HR: 0.54; 95% CI: 0.18-1.66, p=0.28) Conclusions : Dig utilization is potentially associated with a reduced risk of rGIB but has no effect on severe epistaxis in patients implanted with LVAD. Larger cohorts and prospective data will be required confirm these findings. To assess the impact of Digoxin (Dig) utilization on recurrent gastrointestinal bleeding (rGIB) and severe epistaxis, which are both common complications in patients implanted with left ventricular assist devices (LVAD). Literature suggests that Dig utilization may reduce the risk and incidence of initial GIB, however scare literature is present regarding the of effect of Dig on rGIB and epistaxis. Our initial cohort consisted of 346 patients implanted with LVADs on Dig use who survived index admission for LVAD implantation. Subsequent analyses included two separate sub-cohorts; rGIB (n=77) and epistaxis (n=84). All episodes of epistaxis and rGIB were captured up to 2 years of follow-up. Severe epistaxis was defined as one requiring admission with ENT consultation and at least a 1-point drop in hemoglobin, while moderate was defined as an encounter in the ED/clinic to address epistaxis with intervention. Other degrees of epistaxis were classified as mild and excluded from the analysis. Dig use was defined as dig prescribed at the time of rGIB or from the time of LVAD implantation (for epistaxis) with at least 30 days of continuous use. Clinical events as a function of time were investigated using Kaplan-Meier curve analysis. To assess influence of digoxin utilization on clinical outcomes, univariate and multivariate Cox regression analysis was utilized. Mean age of the rGIB cohort was 62 years (±10) and epistaxis 57 years (±14). Dig was prescribed in 41% of the rGIB and 42% of the epistaxis cohorts. A total of 77 patients experienced at least one GIB (22 % of study cohort) with 39 patients (11 % of study cohort) experiencing at least one episode of recurrent GIB. The median time from LVAD implant to first GIB was 62 days (IQR: 26, 231 days) and from first to second GIB was 57 days (IQR: 20, 163 days). In a multivariate analysis after controlling for gender, white blood cell count and beta blocker use at 6 months, Dig use was associated with a reduced risk of rGIB (Fig: HR: 0.40; 95% CI: 0.17-0.93, p=0.03). A total of 84 patients experienced epistaxis, 18 patients had severe epistaxis, while 12 patients had moderate epistaxis and 54 patients had mild epistaxis. The median time from LVAD implantation to first episode of severe epistaxis was 90 days (IQR: 55, 146 days) and moderate epistaxis was 168 days (IQR: 102, 322 days). In a univariate analysis Dig use was not associated with a reduced risk of severe epistaxis (HR: 0.54; 95% CI: 0.18-1.66, p=0.28) : Dig utilization is potentially associated with a reduced risk of rGIB but has no effect on severe epistaxis in patients implanted with LVAD. Larger cohorts and prospective data will be required confirm these findings.