Justin D. Roberts

Substance abuse at the time of left ventricular assist device implantation is associated with increased mortality

BACKGROUND Advanced heart failure teams are often faced with the decision of whether or not to offer a left ventricular assist device (LVAD) to patients who have end-stage heart failure and recent or ongoing substance abuse. The outcomes of these patients after LVAD implantation are unknown. METHODS Baseline predictors and outcomes were collected and analyzed from patients with active substance abuse and a cohort of patients without active substance abuse matched for age, INTERMACS profile and year of implantation. The primary outcome was all-cause mortality. Secondary outcomes included rates of listing for cardiac transplantation, transplantation and chronic drive-line infection. RESULTS The cohort consisted of 20 consecutive LVAD recipients with active substance abuse and 40 recipients without active substance abuse. During a median follow-up period of 2.3 years (IQR 1.4 to 3.6), the substance abuse group had 3.2 times the rate (hazard) of death compared with a matched cohort (HR 3.2, 95% CI 1.2 to 8.0, p < 0.05). Furthermore, the rate of listing for transplant was 69% lower (rate ratio 0.31, p < 0.0005), rate of cardiac transplant was 89% lower (rate ratio 0.11, p < 0.0005), and risk of chronic drive-line infection was 5.4 times higher (rate ratio 5.4, p < 0.0005) in the substance abuse group. CONCLUSIONS Active substance abuse in patients who received an LVAD was associated with increased mortality and overall poor outcomes. Larger scale data will be needed to confirm these findings and to inform decision-making in this population.

Improving the Efficiency of Heart Failure Care

Background Ambulatory pulmonary artery (PA) pressure-directed clinical management of Heart Failure (HF) patients has been shown to reduce HF hospitalizations; however the work flow associated with remote hemodynamic monitoring in such patients has not been studied. We performed a time and motion study in a group of patients with heart failure. Methods A non-interventional, single site “time and motion” study of usual care processes was conducted in the Heart Failure clinic of a 630-bed community hospital between July - October 2017. All enrolled patients were NYHA class III. Patients previously implanted with an ambulatory PA pressure sensor (CardioMEMSTM, Abbott; CMEM group), as well as sensor-eligible patients who had not previously received the implant (non-CMEM group), were recruited at a routine HF clinic visit. The usual care visit, for both CMEM and non-CMEM group, was observed from the time the patient arrived at the clinic to the time they left. The in-clinic observation was quantified based on the time spent in the prep area, exam room area, dictation, and scheduling desk. Primary reason for telephone calls made to CMEM group was captured. Results The HF clinic workflow was observed for 53 patients (n = 24 CMEM, n  =  29 non-CMEM). The mean clinic visit time were 48:55 ± 15:34 minutes for CMEM and 55:57 ± 21:42 minutes non-CMEM (p  =  0.07). 75% of the visit time was spent in the exam room with the provider. Telephone call duration was 5:47 ± 15:09 minutes (N = 92) with a median of 2:13 minutes of which, 52% were related to review of pulmonary artery pressures, 29% to HF monitoring, 12% to labs, and 3% to medication change. Conclusion This is the first characterization of the practical implications of utilizing remote hemodynamic monitoring. The additional time spent during follow-up calls was partially offset by shorter office visits. In addition, since most of the office time involved the provider (exam room) as opposed to nurse time (phone calls), the utilization of CardioMEMS™ may improve provider efficiency. The economic implications of remote hemodynamic management and office visits for HF patients can be studied based on these data.