The safety of sodium glucose transporter 2 inhibitors and trends in clinical and hemodynamic parameters in patients with left ventricular assist devices.

BACKGROUND: The safety and impact of sodium glucose transporter 2 inhibitors (SGLT2-I) in patients with left ventricular assist devices (LVAD) are unknown. METHODS: A retrospective analysis of all consecutive patients who underwent LVAD Heart Mate 3 (HM3) implantation at a single medical center and received SGLT2-I therapy following surgery was conducted. LVAD parameters, medical therapy, laboratory tests, echocardiography, and right heart catheterization (RHC) study results were recorded and compared before and after initiation of SGLT2-I. RESULTS: SGLT2-I medications were initiated in 29 (21%) of 138 patients following HM3 implantation (23 (79%) received Empagliflozin and 6 (21%) Dapagliflozin). The mean age at the time of LVAD implantation was 62 ± 6.7 years, 25 (86%) were male, and 23 (79%) had diabetes mellitus. The median time from HM3 implantation to SGLT2-I initiation was 108 days, IQR (26-477). Following SGLT2-I therapy, the daily dose of furosemide decreased from 47 to 23.5 mg/day (mean difference = 23.5 mg/d, 95% CI 8.2-38.7, p = 0.004) and significant weight reduction was observed (mean difference 2.5 kg, 95% CI 0.7-4.3, p = 0.008). Moreover, a significant 5.6 mm Hg reduction in systolic pulmonary artery pressure (sPAP) was measured during RHC (95% CI 0.23-11, p = 0.042) in a subgroup of 11 (38%) patients. LVAD parameters were similar before and after SGLT2-I initiation (p > 0.2 for all). No adverse events were recorded during median follow-up of 354 days, IQR (206-786). CONCLUSION: SGLT2-I treatment is safe in LVAD patients and might contribute to reduction in patients sPAP.

Effect of levosimendan infusion prior to left ventricular assist device implantation on right ventricular failure.

OBJECTIVE: Investigate the safety and efficacy of preoperative levosimendan in patients undergoing left ventricular assist device (LVAD) implantation. METHODS: Consecutive patients who received LVADs (HeartMate-2, 3, HVAD) in a single tertiary medical center (2012-2018). INTERMACS profile 1 patients were excluded. The primary outcome was post-LVAD right ventricular failure (RVF) and inhospital mortality rates. The secondary outcomes included other clinical, echocardiographic and hemodynamic parameters at follow-up. RESULTS: Final cohort consisted of 62 patients (40[65%] in the levosimendan group and 22[35%] in the no-levosimendan group). Post-operative RVF rate and inotrope or ventilation support time were similar in the levosimendan and no-levosimendan groups (7.5% vs. 13.6%; P = 0.43, median of 51 vs. 72 h; P = 0.41 and 24 vs. 27 h; P = 0.19, respectively). Length of hospitalization, both total and in the intensive care unit, was not statistically significant (median days of 13 vs. 16; P = 0.34, and 3 vs. 4; P = 0.44, respectively). Post-operative laboratory and echocardiographic parameters and in-hospital complication rate did not differ between the groups, despite worse baseline clinical parameters in the Levosimendan group. There was no significant difference in the in-hospital and long term mortality rate (2.5% vs. 4.5%; P > 0.999 and 10% vs. 27.3% respectively; P = 0.64). CONCLUSIONS: Levosimendan infusion prior to LVAD implantation was safe and associated with comparable results without significant improved post-operative outcomes, including RVF.

Hemodynamic Changes After Left Ventricular Assist Device Implantation Among Heart Failure Patients With and Without Elevated Pulmonary Vascular Resistance.

Background: Left ventricular assist devices (LVADs) may reverse elevated pulmonary vascular resistance (PVR) which is associated with worse prognosis in heart failure (HF) patients. We aim to describe the temporal changes in hemodynamic parameters before and after LVAD implantation among patients with or without elevated PVR. Methods: HF patients who received continuous-flow LVAD (HeartMate 2&3) at a tertiary medical center and underwent right heart catheterization with PVR reversibility study before and after LVAD surgery. Patients were divided into 3 groups: normal PVR (<4WU); reversible PVR (initial PVR ≥4WU with positive reversibility); and non-reversible (persistent PVR ≥4WU). Results: Overall, 85 LVAD patients with a mean age of 58 years (IQR 49-64), 65 patients (76%) were male; 60 patients had normal PVR, 20 patients with reversible and 5 patients with non-reversible PVR pre-LVAD. All patients with elevated PVR (≥4WU) had higher pulmonary pressures (PP) and increased trans-pulmonary gradient (TPG) compared to patients with normal PVR (p < 0.05). Patients with non-reversible PVR were more likely to have a significantly lower baseline cardiac output (CO) compared to all other groups (p ≤ 0.02). Hemodynamic parameters and PVR post LVAD were similar in all study groups. Patients with baseline elevated PVR (reversible and non-reversible) demonstrated a significant improvement in PP and TPG compared to patients with normal baseline PVR (p ≤ 0.05). The improvement in CO and PVR post-LVAD in the non-reversible PVR group was significantly greater compared to all other groups (p < 0.01). There were no significant differences between study groups in post LVAD and post heart transplantation course. Conclusion: Hemodynamic parameters improved after LVAD implantation, regardless of baseline PVR and reversibility, and enabled heart transplantation in patients who were ineligible due to non-reversible elevated PVR. Our findings suggest that mitigation of elevated non-reversible PVR is related to reduction in PP and increase in CO.

Diastolic Plateau - Invasive Hemodynamic Marker of Adverse Outcome Among Left Ventricular Assist Device Patients.

Background: Diastolic plateau is an invasive hemodynamic marker of impaired right ventricular (RV) diastolic filling. The purpose of the current analysis was to evaluate the prognostic importance of this sign in left ventricular assist device (LVAD) patients. Methods: The analysis included all LVAD patients who received continuous-flow LVAD (HeartMate 3) at the Sheba medical center and underwent right heart catheterization (RHC) during follow up post-LVAD surgery. Patients were dichotomized into 2 mutually exclusive groups based on a plateau duration cutoff of 55% of diastole. The primary end point of the current analysis was the composite of death, heart transplantation, or increase in diuretic dosage in a 12-month follow-up period post-RHC. Results: Study cohort included 59 LVAD patients with a mean age of 57 (IQR 54-66) of whom 48 (81%) were males. RHC was performed at 303 ± 36 days after LVAD surgery. Patients with and without diastolic plateau had similar clinical, echocardiographic, and hemodynamic parameters. Kaplan-Meier survival analysis showed that the cumulative probability of event at 1 year was 65 ± 49% vs. 21 ± 42% for primary outcomes among patients with and without diastolic plateau (p Log rank < 0.05 for both). A multivariate model with adjustment for age, INTERMACS score and ischemic cardiomyopathy consistently showed that patients with diastolic plateau were 4 times more likely to meet the study composite end point (HR = 4.35, 95% CI 1.75-10.83, p = 0.002). Conclusion: Diastolic plateau during RHC is a marker of adverse outcome among LVAD patients.

The impact of diabetes on short-, intermediate- and long-term mortality following left ventricular assist device implantation.

OBJECTIVES: Type 2 diabetes mellitus (DM) is a frequent comorbidity among patients suffering from advanced heart failure necessitating a left ventricular assist device (LVAD) implant. The goal of this study was to evaluate the impact of type 2 DM on early and long-term outcomes of patients following an LVAD implant. METHODS: We performed an observational cohort study in a large tertiary care centre in Israel. All data of patients who underwent a continuous flow LVAD implant between 2006 and 2020 were extracted from our departmental database. Patients were divided into 2 groups: group I (patients without diabetes) and group II (patients with diabetes). We compared short-term (30-day and 3-month) mortality, intermediate-term (1- and 3-year) mortality and long-term (5 year) mortality between the 2 groups. RESULTS: The study population included 154 patients. Group I (patients without diabetes) comprised 88 patients and group II (patients with diabetes) comprised 66 patients. The mean follow-up duration was 38.2 ± 30.3 months. Short- and intermediate-term mortality (30 days, 1 year and 3 years) was higher in the group with DM compared with the group without DM but did not reach any statistically significant difference: 16.1% vs 9.8% (P = 0.312), 24.2% vs 17.3% (P = 0.399) and 30.6% vs 21.9% (P = 0.127) respectively. Long-term 5-year mortality was significantly higher in the group with DM compared to the group without: 38.7% vs 24.4% (P = 0.038). Furthermore, predictors of long-term mortality included diabetes (hazard ratio 2.09, confidence interval 1.34-2.84, P = 0.004), as demonstrated by regression analysis. CONCLUSIONS: Patients with diabetes and those without diabetes have similar 30-day and short- and intermediate-term mortality rates. The mortality risk of diabetic patients begins to increase 3 years after an LVAD implant. Diabetes is an independent predictor of long-term, 5-year mortality after an LVAD implant. CLINICAL TRIAL REGISTRATION: Ethical Committee of Sheba Medical Centre, Israel, on 2 December 2014, Protocol 4257.