Neurology · Medical Literature
Endoscopic versus catheter-based minimally invasive evacuation of spontaneous supratentorial intracerebral hemorrhage: a systematic review and meta-analysis.
Abstract
Minimally invasive surgery (MIS) for spontaneous supratentorial intracerebral hemorrhage (sICH) includes endoscopic evacuation and catheter-based evacuation (aspiration or drainage ± thrombolysis), but comparative performance remains uncertain. We performed a systematic review and meta-analysis comparing MIS approaches. PubMed, Embase, and CENTRAL were searched (December 2025) for studies in adults undergoing endoscopic or catheter-based MIS for supratentorial sICH. Random-effects meta-analysis was conducted. Prespecified subgroup analyses stratified catheter-based techniques by thrombolysis use. Risk of bias and certainty of evidence was assessed. Twenty-nine studies (8221 patients; 5041 endoscopic; 3180 catheter-based) were included. Endoscopic evacuation was associated with lower mortality (RR 0.64; 95% CI: 0.47-0.87; P < 0.01; I2 = 44.8%), rebleeding (RR 0.58; 95% CI: 0.39-0.87; P < 0.01; I2 = 22.2%), postoperative hematoma volume (MD -5.6 mL; 95% CI: -10.30-0.90; P < 0.05; I2 = 96.9%), and higher day-one evacuation rate (MD 24.5%; 95% CI: 14.60-34.40; P < 0.001; I2 = 99.7%). Overall complications (RR 0.98; 95% CI: 0.78-1.25; P = 0.90; I2 = 48.5%) and infection (RR 1.03; 95% CI: 0.64-1.66; P = 0.05; I2 = 43.6%) were similar. Catheter-based approaches had lower intraoperative blood loss (MD 69.0 mL; 95% CI: 27.7-110.4; P < 0.001; I2 = 96.9%). Subgroup analyses suggested that mortality and evacuation rate differences favored catheter-based cohorts receiving thrombolysis. Functional outcomes diverged: GCS/GOS favored endoscopy (SMD 0.27) while mRS marginally favored catheter-based approaches (MD 0.28), a between-group difference below the threshold generally considered clinically meaningful. Endoscopic evacuation was associated with lower mortality, rebleeding, and residual hematoma burden, while catheter-based approaches retained perioperative efficiency advantages. Functional findings were mixed and of uncertain clinical significance. Given predominantly non-randomized evidence and heterogeneity, these findings should be interpreted as strategy-specific associations.
Clinical relevance
In adults with spontaneous supratentorial intracerebral hemorrhage undergoing minimally invasive surgery, does endoscopic evacuation produce superior clinical and radiographic outcomes compared with catheter-based evacuation (aspiration or drainage ± thrombolysis)? __SNAPSHOT__ {"studyDesign":"Systematic review and meta-analysis (random-effects model) with prespecified subgroup analyses stratified by thrombolysis use within the catheter-based arm; risk of bias and certainty of evidence formally assessed","setting":"Multi-database search (PubMed, Embase, CENTRAL) through December 2025; constituent studies represent multiple international centers and settings — individual study settings not reported in abstract","sampleSize":"n = 8,221 across 29 studies (endoscopic: n = 5,041; catheter-based: n = 3,180)","population":"Adults undergoing minimally invasive surgery for spontaneous supratentorial intracerebral hemorrhage; specific inclusion criteria beyond this (hematoma volume thresholds, GCS cutoffs, hemorrhage location breakdown, baseline characteristics) not reported in abstract","intervention":"Endoscopic hematoma evacuation","comparator":"Catheter-based evacuation: aspiration or drainage with or without thrombolysis (prespecified subgroup analysis stratified by thrombolysis use)","primaryOutcome":"Not explicitly designated as a single primary endpoint in the abstract; reported outcomes include mortality, rebleeding, postoperative residual hematoma volume, and day-one evacuation rate — the primary outcome as pre-registered is not reported in the abstract","followUp":"Not reported in abstract","fundingSource":"Not reported in abstract"}
Key findings
Across 29 studies (n = 8,221), endoscopic evacuation was associated with statistically significant reductions in mortality compared with catheter-based approaches (RR 0.64; 95% CI: 0.47–0.87; P < 0.01; I² = 44.8%) and in rebleeding (RR 0.58; 95% CI: 0.39–0.87; P < 0.01; I² = 22.2%). The absolute mortality risk reduction and number needed to treat (NNT) cannot be calculated without the baseline event rates in the control arm, which were not provided in the abstract. Endoscopic evacuation was also associated with lower postoperative hematoma volume (mean difference [MD] −5.6 mL; 95% CI: −10.30 to −0.90; P < 0.05), though this outcome demonstrated very high heterogeneity (I² = 96.9%), severely limiting interpretability. Day-one evacuation rate was meaningfully higher with endoscopic surgery (MD +24.5%; 95% CI: +14.60 to +34.40; P < 0.01). The abstract was truncated and does not report functional outcome data (e.g., modified Rankin Scale), quality of life, duration of hospitalization, infectious complications, or other safety endpoints; these are not reported. Subgroup analyses stratified by thrombolysis use within catheter-based techniques were prespecified but results were not included in the provided abstract. Certainty of evidence ratings (GRADE or equivalent) were performed but not reported in the abstract.
Practice implications
This meta-analysis suggests that endoscopic evacuation is associated with lower mortality and rebleeding rates, as well as faster and more complete initial clot removal, compared with catheter-based MIS in spontaneous supratentorial ICH. The mortality risk ratio of 0.64 represents a potentially clinically important signal; however, absolute risk reduction and NNT cannot be derived without baseline event rates from the comparator arm. The extremely high heterogeneity for postoperative hematoma volume (I² = 96.9%) indicates that the mean difference of −5.6 mL should be interpreted with great caution, as it likely reflects substantial variation in surgical technique, hematoma location, imaging timing, and patient selection across constituent studies rather than a homogeneous treatment effect. Functional outcomes — the most patient-relevant endpoints — are not reported in the abstract, making it impossible to determine whether the imaging and mortality advantages translate into meaningful differences in neurological recovery or independence. Whether results apply to specific subpopulations (deep vs. lobar hemorrhage, large vs. small hematoma, hemorrhage with or without intraventricular extension, or patients treated at different time windows from onset) cannot be assessed. The finding that catheter-based aspiration without thrombolysis may differ from thrombolysis-assisted drainage is mechanistically important but subgroup results are not available in the abstract. The body of included literature likely encompasses predominantly observational and non-randomized data given the known paucity of head-to-head RCTs comparing these two MIS modalities directly, which substantially limits causal inference. IMPORTANT LIMITATION OF THIS SUMMARY: Only the abstract was available; full results, GRADE certainty ratings, risk of bias assessments, and complete outcome tables were not accessible for review. __LIMITATIONS__ 1. Only the abstract was available for this summary; full methodology, GRADE certainty ratings, sensitivity analyses, and complete subgroup results could not be evaluated. 2. Very high heterogeneity for the hematoma volume outcome (I² = 96.9%) renders that pooled estimate unreliable; the sources of heterogeneity are not described in the abstract. 3. The included literature likely contains a mixture of study designs (RCTs and observational studies); the proportion and weighting of non-randomized studies are not reported, raising concerns about confounding and selection bias. 4. Functional neurological outcomes (mRS, Barthel Index, GCS) are not reported in the abstract, leaving the most clinically meaningful endpoints unaddressed. 5. Baseline patient characteristics (age, hematoma volume, GCS, hemorrhage location, time to surgery) across the two treatment arms are not reported, precluding assessment of comparability. 6. Subgroup results stratifying catheter-based MIS by thrombolysis use — a mechanistically critical distinction — are not available in the abstract. 7. Duration of follow-up and funding sources are not reported. 8. Generalizability may be limited if included studies are predominantly from East Asian centers, where operator experience and patient selection criteria may differ from Western practice. 9. Statistical significance across all outcomes does not confirm clinical importance in the absence of absolute effect sizes and patient-centered endpoints. __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ This meta-analysis of 29 studies (n = 8,221) found that endoscopic evacuation of spontaneous supratentorial ICH was associated with lower mortality, less rebleeding, and higher day-one clot clearance compared with catheter-based MIS, but very high heterogeneity for hematoma volume and the absence of reported functional outcomes limit the strength of these conclusions. Causal inference is further constrained by likely inclusion of non-randomized studies and unknown baseline comparability between treatment groups. These results are hypothesis-generating and should inform prospective RCT design rather than prompt immediate changes in technique selection outside of experienced endoscopic centers.