Introduction The elderly population were underrepresented on the randomized clinical trials (RCTs) for mechanical thrombectomy (MT) in large vessel occlusion (LVO) stroke.

Objective The purpose of this systematic review and meta-analysis was to assess all the available evidence on the efficacy and safety of MT in patients older than 80 years old with LVO stroke.

Methods We used PRISMA guidelines to report our search results. Searches were conducted using the PubMed/Medline, Scopus, Embase, and Cochrane databases up to January 2023, including randomized clinical trials (RCT) and observational studies that included mechanical thrombectomy for patients older than 80 years old with LVO stroke. Our primary outcome was functional independence, defined as modified Rankin Scale score (mRS): 0-2. Secondary outcomes were 90 days mortality, symptomatic intracranial hemorrhage (sICH) and all ICH. We performed a meta-analysis using the random effect model. Pooled proportions were calculated. A sensitivity analysis based on the MT technique was also performed for all outcomes. We conducted the meta-analyses using random-effects models. Statistical heterogeneity and risk of bias across studies were assessed using Baujat plot and funnel plot respectively.

Results Seven studies (4 observational and 3 RCTs) were included (n=999). The overall proportion of patients who achieved a favorable outcome was 0.20 (95%-CI: 0.13 to 0.28). In patients who underwent MT the pooled proportion of functional independence (0.29 [95%-CI: 0.21; 0.38]) was higher than that for the medical management subgroup (0.20 [95%-CI:0.04; 0.20]; p < 0.01). These patients had a higher chance to achieve functional independence: OR: 3.06 [95%-CI: 1.44-6.54], I²=57%, p=0.03. Mortality at 90 days was higher in the elderly group 33.61% vs 20.4%. Symptomatic ICH was 3.8% in the MM, and barely reported on the MT group. Among the seven studies, three subgroups exist with contribution to overall heterogeneity greater than 10 or influence on overall result greater than 0.5. Among these, one study had an influence greater than 4 and heterogeneity contribution greater than 40. The funnel plot appears to be symmetrical with no obvious visual indication of significant publication bias. A linear regression test was performed. The bias was not statistically significant (t-value of 0.79, degrees of freedom of 12, and a p-value of 0.4474). Therefore, results of the funnel plot and linear regression test suggest that the results of the meta-analysis are robust and not influenced by publication bias.

Conclusion Our results suggest that MT in the elderly population is effective to achieve functional independence. Our data was unable to provide information about 90-days mortality, all ICH and sICH. Given our limitations, a RCT comparing MT versus MM in this population would be needed to confirm our results. However, including elderly patients in RCTs is always challenging. So, our results could provide support in the real world to offer this highly effective treatment to this challenging population.

Disclosures J. Sequeiros: None. M. Cornejo: None. F. Terry: None. C. Zackbutton: None. M. Rodriguez: None. H. Zaver: None. O. Saeed: None. N. Goyal: None. V. Inoa: None.