The use of palliative sedation for intractable symptoms in a palliative care unit (PCU) in a comprehensive cancer center

Abstract

8577 Background: There is wide variation in the frequency of reported use of palliative sedation (PS) to control intractable and refractory symptoms. Institutions have established policies for midazolam infusion in cases of PS. The indications and outcomes of this procedure have not been well characterized Methods: Our midazolam policy for PS requires 1:1 nursing for the first 24 hours and documentation of discussions regarding sedation. We reviewed our PCU database for all admissions for the first 11 months of 2005. We used pharmacy records for all patients who received medications used for sedation (chlorpromazine, lorazepam, midazolam). We reviewed all charts of pts who received any of these drugs to establish if the indication had been PS. Results: 148/484 admissions died in the PCU [31%]. 65/484 admissions (13%), and 47/ 148 patients who died (32%) received PS. Median age of patients (pts) was 58, 42 pts were male [65%], and the most frequent primaries observed were lung 24 [37%], hematologic 12 [18%], head and neck 7 [11%], and gastrointestinal 7 [11%]. Results are indicated in the table. * 2 patients had more than one indication for sedation The main causes for PS in our patients were delirium 57 [88%], dyspnea 6 [9%], and bleeding 4 [6%]. 18/65 patients who received PS [35%] were discharged alive, versus 318/419 [76%] who did not receive PS [p< 0.001]. Midazolam was used in 11/65 episodes [17%]. 4/6 pts with PS for dyspnea received midazolam [66%], versus 8/57 with PS for delirium or bleeding [14%], p=0.01]. 18/54 pts who received PS using other drug were discharged alive [33%], versus 0/11 pts who received midazolam [p=0.02]. Conclusions: Palliative sedation was required in 32% of pts who died in the hospital. Reporting midazolam utilization rates for monitoring overall PS outcomes, results in significant under reporting. Midazolam was used more frequently in cases of progressive dyspnea and poor prognosis. Less restrictive policies in the use of midazolam may result in more use for PS. Data accrual continues. [Table: see text] No significant financial relationships to disclose.