Multicenter retrospective cohort links two chronic subdural haematoma procedures to lower recurrence than burr hole drainage

In a multicenter retrospective cohort study of 265 adults with chronic subdural haematoma, middle meningeal artery embolization and neuroendoscopic-assisted evacuation with middle meningeal artery disconnection were associated with lower six-month recurrence and reoperation rates than burr hole drainage. Neuroendoscopic-assisted evacuation with artery disconnection was also associated with the highest neurological improvement rate, although the procedures differed in operation time, follow-up imaging, additional surgery and complications.

The study included patients treated at Nanchong Central Hospital and Guangyuan Central Hospital in Sichuan, China, between July 2020 and February 2026. Eligible patients were 18 to 90 years old, had symptomatic chronic subdural haematoma of at least three weeks’ duration, and had either maximum haematoma thickness above 10 mm or midline shift above 5 mm. Of the 265 participants, 46 received middle meningeal artery embolization, 152 underwent burr hole drainage and 67 received neuroendoscopic-assisted evacuation with middle meningeal artery disconnection.

The primary outcome was radiological recurrence within six months, defined as increased haematoma thickness on follow-up CT, with or without worsening symptoms, regardless of whether further surgery was required. An independent radiologist blinded to treatment allocation re-measured haematoma thickness and midline shift. Secondary outcomes included neurological improvement, defined as a decrease of at least one point on the modified Rankin Scale; reoperation; any additional surgery; complications; operation time; hospital stay; and imaging follow-ups.

Six-month recurrence occurred in 30 of 152 patients treated with burr hole drainage, a rate of 19.74% (95% CI 13.73–26.96%). Recurrence occurred in two of 46 patients treated with embolization, a rate of 4.35% (95% CI 0.53–14.84%), and three of 67 patients treated with neuroendoscopic evacuation and artery disconnection, a rate of 4.48% (95% CI 0.93–12.53%). The overall difference among the three groups was statistically significant (p = 0.001), while the difference between embolization and neuroendoscopic evacuation was not. Recurrence odds were higher with burr hole drainage than with embolization or neuroendoscopic evacuation, and the blinded radiological measurements produced the same group estimates.

Neurological improvement was documented in 59 of 67 patients in the neuroendoscopic group, or 88.06% (95% CI 77.82–94.70%), compared with 71.71% in the burr hole drainage group and 60.87% in the embolization group; the overall difference was significant (p = 0.003). Reoperation for recurrence was required in 19.08% of the burr hole drainage group, compared with 4.35% after embolization and 4.48% after neuroendoscopic evacuation, also a significant overall difference (p = 0.002).

Embolization did not produce the lowest rate under every measure. Five patients, or 10.87% of the embolization group, underwent salvage burr hole drainage during the same hospitalization because of persistent or worsening symptoms without radiological progression. When recurrence-related reoperations and this salvage procedure were combined, any additional surgery occurred in 15.22% of the embolization group, 19.08% of the burr hole drainage group and 4.48% of the neuroendoscopic group. After Bonferroni-corrected pairwise comparisons, the neuroendoscopic group had a significantly lower rate than the burr hole drainage group, but the other pairwise differences were not significant.

The embolization procedure had the shortest median operation time, 31 minutes, compared with 45 minutes for burr hole drainage and 70 minutes for neuroendoscopic evacuation. Median hospital stay was seven days after embolization, nine days after burr hole drainage and eight days after neuroendoscopic evacuation. The embolization and neuroendoscopic groups each had a median of four imaging follow-ups, compared with six in the burr hole drainage group. Complication rates did not differ significantly: 8.70% after embolization, 9.21% after burr hole drainage and 5.97% after neuroendoscopic evacuation.

One patient in the embolization group experienced permanent blindness after inadvertent embolization through a dangerous meningo-ophthalmic anastomosis. No comparable severe event was reported in the other groups. The authors described embolization as offering rapid procedures and low recurrence, but cautioned that it does not immediately remove an established haematoma and that five patients required salvage drainage for delayed resolution of mass effect. Neuroendoscopic evacuation with artery disconnection allowed direct removal of liquid and solid haematoma, coagulation of bleeding points and disruption of fibrous septations, but required the longest operation.

The observational design limits causal interpretation. Treatment was not randomized, and selection may have reflected operator preference, patient willingness and local resources. Although measured baseline characteristics were balanced, unmeasured factors such as frailty, disease severity and operator experience could have influenced the results. The embolization and neuroendoscopic groups were small and had few recurrence events, preventing multivariable adjustment, propensity-score matching and reliable time-to-event analysis. Secondary outcomes were exploratory without correction for multiple testing, and follow-up beyond one year was not established.

Based on the findings and their clinical interpretation, the authors proposed individualized selection that accounts for haematoma characteristics, patient condition and available resources. They presented burr hole drainage as a reasonable option for first-onset, non-septated haematoma; embolization as potentially suitable for older or medically frail patients and those with coagulopathy, subject to caution because of blindness risk; and neuroendoscopic evacuation with artery disconnection as potentially appropriate for septated, organized, recurrent or markedly space-occupying haematomas. The study itself was observational, so these positions require confirmation in larger prospective comparative research.

Subject of Research: Comparative outcomes of three treatments for chronic subdural haematoma in a 265-patient retrospective cohort

Article Title: Comparative outcomes of middle meningeal artery embolization, burr hole drainage, and neuroendoscopic-assisted evacuation with MMA disconnection for chronic subdural haematoma: a multicenter retrospective cohort study

Article References: Song, X., Zhang, Y., Luo, B., & Tang, X. (2026). Comparative outcomes of middle meningeal artery embolization, burr hole drainage, and neuroendoscopic-assisted evacuation with MMA disconnection for chronic subdural haematoma: a multicenter retrospective cohort study. Frontiers in Neurology, 17, Article 1975364. https://doi.org/10.3389/fneur.2026.1975364

Image Credits: AI Generated

DOI: 10.3389/fneur.2026.1975364

Keywords: chronic subdural haematoma, middle meningeal artery embolization, burr hole drainage, middle meningeal artery disconnection, neuroendoscopy, haematoma recurrence, retrospective cohort study, neurological outcome, neurosurgery, Comparative, outcomes, middle