A Propensity Score-Weighted Analysis of Short-Term Corticosteroid Therapy for Refractory Pain Following Spontaneous Subarachnoid Hemorrhage.

Abstract

Background: Corticosteroids are prescribed for refractory headache in patients with spontaneous subarachnoid hemorrhage (SAH) despite limited supporting evidence. We hypothesized that a short course of corticosteroids would reduce pain.

Methods: We reviewed all patients who received corticosteroids for refractory headache following spontaneous SAH within our institutional database. Pain was measured by a numeric rating scale (NRS) every 2 h. The primary outcome was maximum daily NRS score; secondary outcomes were the mean daily NRS score and daily opioid consumption. Propensity scores were developed using potential predictors of corticosteroid use, including age, sex, pretreatment 24-h pain burden, and the number of analgesics being used to control pain. Inverse probability treatment weighting (IPTW) was used to balance baseline covariate distributions between patients receiving corticosteroids and control patients. Generalized estimating equations were used to analyze longitudinal NRS scores and oral morphine equivalents based on the weighted cohort.

Results: A total of 213 patients were included. The mean age was 55 (SD 13) years, and 141 of 213 (66%) were female. Of 213 patients, 195 (92%) had a low clinical grade (i.e., Hunt-Hess grades 1-3). Seventy patients were prescribed corticosteroids on postbleed day 5 (SD 3.3) on average, with an average of 26 (SD 10) mg of dexamethasone over 48 h. Patients receiving corticosteroids and controls were well balanced on baseline predictors of treatment status. After IPTW, we found that corticosteroid therapy reduced the daily maximum pain NRS score by 0.59 (SE = 0.39, p = 0.12), 0.96 (SE = 0.42, p = 0.02), and 0.91 (SE = 0.46, p = 0.048) on days 1-3, respectively, after adjusting for control effects. The mean daily pain NRS score and daily opioid use were nonsignificantly reduced in the 3 days following corticosteroid initiation after adjusting for control effects.

Conclusions: Short-term corticosteroids only slightly reduced maximum pain severity after spontaneous SAH. Other analgesic strategies are required to manage refractory pain in this population.