Use of palliative sedation for intractable symptoms in the palliative care unit of a comprehensive cancer center

Background There is wide variation in the frequency of reported use of palliative sedation ( PS) to control intractable and refractory symptoms in terminally ill patients. The aim of this study was to determine the frequency and outcomes of PS use and examine patterns of practice after establishment of a policy for the administration of midazolam for PS in our palliative care unit (PCU). Materials and methods This retrospective study reviewed PCU admissions for 2004 and 2005 and pharmacy records to identify patients who received chlorpromazine, lorazepam, or midazolam for PS in the PCU. Data on indication for PS, drug used, and discharge outcome were assessed for each patient. Results During the period studied, there were 1,207 PCU admissions. Of these patients, 186 (15%) received PS; and 143 (41%) of the 352 patients who died in the PCU received PS. The median age of PS patients was 58 (range, 20-84) years, and 106 (57%) were male. The most common indications for PS were delirium, 153 cases (82%); dyspnea, 11 (6%); and multiple indications, 12 ( 6%). Midazolam was used in 18 PS cases (10%). Six (55%) of 11 patients with dyspnea received midazolam for PS, compared with 12 (7%) of 175 patients with other indications for PS (p<0.001). Forty-three (23%) of 186 PS patients were discharged alive, compared with 812 (80%) of 1,021 patients who did not receive PS (p<0.001). Conclusions PS was required in 15% of PCU admissions, and 23% of PS patients were discharged alive. Our findings suggest a potential for significant underreporting of overall PS. If our institution's policy on midazolam use for PS were less restrictive, midazolam use might increase. More research is needed to define the optimal agent for inducing rapid, effective, and easily reversible PS.