Prophylactic epidural blood patch for cerebrospinal fluid leakage after intrathecal drug delivery system implantation in patients with refractory cancer pain: a multi-center retrospective cohort study.

INTRODUCTION

Current guidelines for postdural puncture cerebrospinal fluid (CSF) leakage advocate conservative management, extrapolated from general populations. Patients with refractory cancer pain (RCP) often have advanced disease, malnutrition, and hypoalbuminemia, which may compromise tissue repair and reduce the reliability of spontaneous dural closure. The role of prophylactic epidural blood patch (PEBP) in this high-risk subgroup is undefined.

METHODS

This multicenter retrospective cohort study enrolled RCP patients undergoing intrathecal drug delivery system (IDDS) implantation (Jan 2023-Aug 2025). Patients received either intraoperative PEBP (10-15 mL autologous blood) or conservative management (strict bed rest for 24-48 h, intravenous hydration 1,500-2,000 mL/day, oral caffeine). Propensity score matching (1:1) balanced recorded baseline and available procedural covariates. The primary outcome was moderate-to-severe postural headache syndrome (PHS).

RESULTS

After matching (78 patient pairs, N = 156), PEBP was associated with lower rates of moderate-to-severe PHS (6.41% vs. 25.64%; p < 0.001), exploratory imaging-confirmed CSF leakage (7.69% vs. 24.36%; p = 0.003), and remedial EBP requirement (2.56% vs. 15.38%; p = 0.006). The absolute risk reduction for moderate-to-severe PHS was 19.23%, corresponding to a number needed to treat (NNT) of approximately 5. Headache duration and hospital stay were also shorter in the PEBP group. Subgroup findings were exploratory. Minor adverse events occurred in 11.54% of PEBP patients. No clinically evident neuraxial seeding was observed during short-term follow-up (median 3 months, range 2-5 months).

CONCLUSION

In this retrospective cohort, intraoperative PEBP was associated with fewer early CSF leakage-related complications and shorter recovery after adjustment for recorded covariates. These findings support further prospective evaluation of PEBP in carefully selected patients. Prospective randomized controlled trials are warranted to validate these results. Causal inference, long-term safety, and patient-centered benefit require confirmation.