The effect of steroids as an adjunct to surgical treatment in patients with chronic subdural hematoma

Roger Lodewijkx, Dana Catharina Holl, Kari-Anne M Slot, Victor Volovici, Clemens M F Dirven, Ruben Dammers, William Peter Vandertop, Dagmar Verbaan

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

The role of steroids as an adjunct to surgery for chronic subdural hematoma (cSDH) remains un-clear. We evaluated the effect of steroids as an adjunct to surgery on recurrence rates, complica-tions, and mortality. We retrospectively collected data of 525 patients operated for cSDH between January 2010 and April 2015 at the Amsterdam University Medical Centers (AMC) and Erasmus Medical Center Rotterdam. Data from patients with and without steroid use as an adjunct to sur-gery were obtained from medical records and compared using the Chi-square test, independent samples t-test, Mann-Whitney U test, where applicable. Associations between adjuvant steroid use and complications were analyzed with univariable (penalized likelihood) logistic regression analysis. Multivariate logistic regression was performed to analyze the influence of adjuvant steroid use on recurrence. Propensity score matching was used to assemble a cohort of patients with simi-lar baseline characteristics. 278 of the 525 patients (53%) were treated with adjuvant steroids. Sur-gery for recurrences occurred less in patients of the steroid group (9% vs 14%, odds ratio (OR) 0.57; 95%CI 0.33 - 0.99), but the effect was not significant after correction for confounders (adjusted odds ratio (aOR) 0.59; 95% CI 0.33 - 1.05). In the steroid group, delirium (10% vs 3%, OR 3.99; 95%CI 1.72 - 9.29) and dysregulated glucose levels occurred more frequently (2% vs 0%, OR 11.81; 95% CI 1.38 - 1542.79) but multivariate analysis was not possible. After propensity score matching, McNemar's chi-square test showed that adjuvant steroid use was not significantly associated with recurrence rate (p = 0.10). Steroids as an adjunct to surgery in patients with cSDH did not have a favorable effect on the recurrence rate in our data after controlling for confounders.

Original languageEnglish
JournalJournal of Neurotrauma
DOIs
Publication statusE-pub ahead of print - 31 Mar 2021

Cite this