Can We Make The In-Hospital LVAD Deactivation Process At End-of-life Comfortable?

Abstract

Background Despite growth in the use of durable left ventricular assist devices (LVAD), best practices about comfort medications prior to and at the time of deactivation process at end-of-life (EOL) have yet to be elucidated. We aim to describe details of comfort medications administered prior to in-hospital LVAD deactivation. Methods We retrospectively reviewed patients with durable LVAD who died at our institution following device deactivation between January 2017 and March 2020. Patient data included demographics, baseline characteristics and details regarding the comfort medications, their dosing and methods of administration used 24 hours prior and at the time of the deactivation process. Results 58 patients were included in the quantitative analysis; 70% were males, 70% African American, median age 62 years (range 31 to 80). Most deactivations occurred in the cardiac surgical ICU (N=48, 82%); N=9 (16%) occurred in the advanced heart failure unit and one in the emergency department. Prior to deactivation, comfort focused medications used in the ICU included opioids (N=48, 100%), most commonly fentanyl patient-controlled analgesic (average continuous dose of 90 mcg/hour, average bolus dose of 208 mcg). Benzodiazepines were used frequently (N=30, 68%), most commonly lorazepam (average dose of 3.2mg) and antisecretory agents (N=24, 50%) used was glycopyrrolate (average dose of 0.4mg). Sedatives such as dexmedetomidine, propofol and ketamine drips were used as adjuncts in the ICU (N=22, 46%). Of patients who underwent device deactivation in the intermediate care unit, hydromorphone PCA (N=4, 44%) was most commonly used (average continuous dose 0.5mg/hour, average bolus dose of 3.5 mg) followed by fentanyl PCA in 2 patients (average continuous dose 22.5 mcg/hour, average bolus dose of 170 mcg). Other patients in the intermediate care unit received either fentanyl or hydromorphone bolus doses prior to device deactivation. Lorazepam was used in almost all patients (N=8, 89%) prior to device deactivations in the intermediate care unit (average dose 2.3 mg) along with glycopyrrolate (N=5, 56%; average dose 0.36 mg). The patient who underwent device deactivation in the ER received bolus doses of hydromorphone (total dose of 4mg) along with lorazepam prior to deactivation. Conclusion In this analysis of comfort medications used for in-hospital device deactivations, most patients were placed on fentanyl through PCA along with lorazepam and glycopyrrolate prior to device deactivation. Dexmedetomidine drip was sometimes used as an adjunct in patients who underwent device deactivations in the ICU. We plan to draft institutional comfort medication guidelines for in-hospital device deactivations. Despite growth in the use of durable left ventricular assist devices (LVAD), best practices about comfort medications prior to and at the time of deactivation process at end-of-life (EOL) have yet to be elucidated. We aim to describe details of comfort medications administered prior to in-hospital LVAD deactivation. We retrospectively reviewed patients with durable LVAD who died at our institution following device deactivation between January 2017 and March 2020. Patient data included demographics, baseline characteristics and details regarding the comfort medications, their dosing and methods of administration used 24 hours prior and at the time of the deactivation process. 58 patients were included in the quantitative analysis; 70% were males, 70% African American, median age 62 years (range 31 to 80). Most deactivations occurred in the cardiac surgical ICU (N=48, 82%); N=9 (16%) occurred in the advanced heart failure unit and one in the emergency department. Prior to deactivation, comfort focused medications used in the ICU included opioids (N=48, 100%), most commonly fentanyl patient-controlled analgesic (average continuous dose of 90 mcg/hour, average bolus dose of 208 mcg). Benzodiazepines were used frequently (N=30, 68%), most commonly lorazepam (average dose of 3.2mg) and antisecretory agents (N=24, 50%) used was glycopyrrolate (average dose of 0.4mg). Sedatives such as dexmedetomidine, propofol and ketamine drips were used as adjuncts in the ICU (N=22, 46%). Of patients who underwent device deactivation in the intermediate care unit, hydromorphone PCA (N=4, 44%) was most commonly used (average continuous dose 0.5mg/hour, average bolus dose of 3.5 mg) followed by fentanyl PCA in 2 patients (average continuous dose 22.5 mcg/hour, average bolus dose of 170 mcg). Other patients in the intermediate care unit received either fentanyl or hydromorphone bolus doses prior to device deactivation. Lorazepam was used in almost all patients (N=8, 89%) prior to device deactivations in the intermediate care unit (average dose 2.3 mg) along with glycopyrrolate (N=5, 56%; average dose 0.36 mg). The patient who underwent device deactivation in the ER received bolus doses of hydromorphone (total dose of 4mg) along with lorazepam prior to deactivation. In this analysis of comfort medications used for in-hospital device deactivations, most patients were placed on fentanyl through PCA along with lorazepam and glycopyrrolate prior to device deactivation. Dexmedetomidine drip was sometimes used as an adjunct in patients who underwent device deactivations in the ICU. We plan to draft institutional comfort medication guidelines for in-hospital device deactivations.