In this meta-analysis, we provide insights into the efficacy and safety of flow diverter treatment with or without coiling of large and giant intracranial aneurysms. The aim was to address the question of which treatment is safer and more effective, given the complexity of such aneurysms and the varying conclusions that should be found in the literature. We found that the adjunctive coiling of the aneurysm did not statistically increase the adequate occlusion rate (OKM C-D). However, the treatment with FD + C was associated with lower complication rates, especially device-related and ischemic complication rates.
Several meta-analyses have attempted to evaluate outcomes and complications with the use of FD, such as the study presented in 2013 by Brinjikji and colleagues [18]. They included 1451 patients with 1654 intracranial aneurysms treated with FD, reporting a complete occlusion rate of 76% (95% CI [70%,81%]) at 6 months, a procedure-related morbidity of 5% (95% CI [4%, 7%]), and a mortality of 4% (95% CI [3%, 6%]). A more recent meta-analysis in 2021 by Jia-Lin Xia and colleagues compared the treatment of unruptured aneurysms with FD versus coil embolization, including 8 studies with 839 patients treated with FD and 2734 with coils. They found a higher rate of complete occlusion at 6 months with the use of FD (OR 0.28 [0.09, 0.85]; I2:82%; p:0.02) [19]. In subgroup analysis, FD treatment was associated with significantly higher complete occlusion rates at 6 months for large or giant aneurysms (OR 0.12 [0.07, 0.21]; I2:0%; p:<0.01), while no differences were observed in non-large/giant aneurysm groups (OR 1.10 [0.46, 2.60]; I2:18%; p:0.83). In 2022, Li and colleagues compared FD treatment with conventional coil treatment, including 18 studies with 1001 patients treated with FD and 1133 with coils [20]. They found that the FD group had larger aneurysms (standardized mean difference [SMD] 0.22 [0.03, 0.41]; p: 0.02). Angiographic results for FD treatment indicated a higher rate of complete occlusion (OR 2.55 [1.70, 3.83]; I2:68%; p:<0.01) and lower rates of recurrence (OR 0.24 [0.12, 0.46]; I2:36%; p:<0.01) and retreatment (OR 0.31 [0.21, 0.47]; I2:30%; p:<0.01). There was a higher risk of complications in the FD group compared to the endovascular coil group (OR 1.4 [1.01, 1.96]; p:0.04).
In our study, we found that 83.9% (453/541; 95% CI 80.8%-87%; I2:0%; p:0.21) had a favorable occlusion rate OKM C-D, consistent with findings presented by Shehata, like the PEDESTRIAN study, and better than those presented by Brinjikji [18, 21, 22]. When dichotomized by therapeutic modality, the FD + C had a favorable occlusion of 85.6% (249/291; 95% CI 81.5%-89.6%), while the FD-only group was 81.6% (204/250; 95% CI 76.8%-86.4%), and the comparison was inconclusive (RR 1.06 [0.96, 1.17]; I2:0 %; p:0.28).This aligns with Sweid and colleagues' findings, although their final recommendation is the combined use for large and giant aneurysms [23]. When analyzing subgroups, we note a statistically significant trend for OKM D in favor of FD without coadjuvant coils (RR 1.09 [0.99, 1.20]; I2:0 %; p:0.07). Als, we highlight that few articles discriminated aneurysm size, limiting statistical power (33.3%).
The PEDESTRIAN study, presented in 2021, reported long-term follow-up on 835 patients with 1000 aneurysms treated with the Pipeline flow diverter [22]. They obtained a low rate of overall complications, including 3.6% (30/835) for stroke (mostly caused by stent thrombosis) and 1.8% (23/835) for hemorrhagic complications. The overall morbidity rate, including subarachnoid hemorrhage, was 2.7% (23/835). The all-cause mortality rate was 4.6% (38/835), with a neurological mortality rate of 3.1% (26/835). Cox logistic regression identified aneurysm size as the only significant predictive factor for neurological death (HR 6.37 [95% CI: 2.41–16.81]; p = 0.01). These findings do not align with the results of this study.
On the contrary, our results coincide with those presented by Sweid, who reported similar hemorrhagic complications between groups (5.5% for coadjuvant coils and 4.8% for FD-only, without statistical significance, p:0.81) as well as ischemic complications (12% for coadjuvant coils and 6.3% for FD-only, without statistical significance, p: 0.21) [23].
Regarding complications, we found that the coadjuvant use of coils resulted in fewer general complications (RR 0.56 [0.33, 0.95]; I2:0%; p:0.03). However, subgroup analysis (related to FD, hemorrhagic, and ischemic complications) was inconclusive, although complications related to FD showed a trend favoring the use of coils (RR 0.46 [0.20, 1.05]; I2:0%; p:0.06).
LIMITATIONS
This study has some limitations. There are no RCT studies, and among the observational studies, most are retrospective. Most of the articles present treatments up to the ophthalmic segment, potentially introducing a bias in the distal and posterior circulation. Some articles used different dimension to define large aneurysms, and it could have an effect in the occlusion rate. Additionally, a specific type of FD was not analyzed, despite known differences. Some articles did not use the OKM classification, requiring adaptation to the mentioned classification, and there was no standardized follow-up time, so the last follow-up was mentioned. The articles that used coils as coadjuvant do not specify the quantity used and do not reference light or dense packing. Despite these limitations, we can draw important conclusions.