Endovascular coiling vs. surgical clipping for unruptured intracranial aneurysm: A meta-analysis

With increasing use of high-resolution imaging of brain, unruptured aneurysms are more and more frequently detected. With the advances in treatment techniques, an increasing number of aneurysms are now occluded using endovascular coiling instead of conventional surgical clipping. However, the better modality for unruptured intracranial aneurysm has been poorly understood.

The objective of this meta-analysis was to compare the outcomes between endovascular coiling and surgical clipping among patients with unruptured intracranial aneurysms.

PubMed, Embase, Web of Science, CENTRAL, and SIGLE were electronically searched from January 1, 1990 to March 13, 2012 with no language restriction for randomized or nonrandomized clinical controlled trials. Article screening and data extraction were conducted in duplicate. Results were statistically pooled through Review Manager 5 and StatsDirect 2.7.9.

Seven studies met our inclusion criteria. The pooled risk ratios (coiling vs. clipping) were 0.59 (95% CI = 0.23-1.54) for death; 0.37 (95% CI = 0.10-1.41) for bleeding; 0.78 (95% CI = 0.38-1.58) for cerebral ischemia; 0.87 (95% CI = 0.70-1.08) for occlusion of aneurysm; 0.53 (95% CI = 0.18-1.52) for independence in daily activities. The pooled rates of death, bleeding, ischemia, occlusion of aneurysm, and mRS no less than 3 were 1% (95% CI = 0-2%), 2% (95% CI = 0-5%), 8% (95% CI = 4-13%), 82% (95% CI = 64-95%), and 5% (95% CI = 1-10%) for endovascular coiling, respectively, and 1% (95% CI = 0-2%), 6% (95% CI = 3-10%), 9% (95% CI = 5-15%), 95% (95% CI = 90-98%), and 8% (95% CI = 3-14%) for surgical clipping, respectively. We failed to evaluate quality of life and cognitive outcome due to insufficient data. Both meta-regression and sensitivity analysis showed consistent results. Furthermore, Begg's test and Egger's test failed to detect publication bias.

We suggest that endovascular coiling and surgical clipping bear similar risk ratios of death, bleeding, cerebral ischemia, occlusion of aneurysm, and independence in daily activities and encourage further studies on quality of life and cognitive outcome. However, albeit the results in this meta-analysis are robust, due to great clinical heterogeneity and low quality of studies, the results in this meta-analysis should be interpreted with caution.