The extracranial versus intracranial approach In frontoethmoidal encephalocele corrective surgery: a meta-analysis.

The debate between the extracranial and the intracranial approach for frontoethmoidal (FEE) encephalocele corrective surgery was not summarized yet. The extracranial approach is traditionally believed being inferior to the intracranial approach, but convincing evidence was missing. To provide robust evidence, we conducted a meta-analysis on the incidence of cerebrospinal fluid (CSF) leakage, its progression to infection, the reoperation to treat the leakage, and the recurrence rate between the two techniques. We performed a meta-proportion pooled analysis and meta-analysis on eligible literature following the recommendation of PRISMA guidelines. The outcome of interest was the incidence of CSF leakage, the CSF leakage that progressed into an infection, the reoperation rate to treat the leakage, and the recurrence rate. We included 28 studies comprising 1793 patients in the pooled prevalence calculations. Of the 28 studies, nine studies describing 730 patients were eligible for meta-analysis. The prevalence of CSF leakage was 8% (95% CI, 0.04-0.12) in the intracranial approach and 10% (95% CI, 0.01-0.23) in the extracranial approach The subgroup analysis of the intracranial approach showed higher CSF leakage prevalence in the frontal craniotomy approach (9%; 95% CI, 0.03-0.16) than the subfrontal osteotomy (6%; 95% CI, 0.03-0.12). Meta-analysis study revealed a significantly higher risk of CSF leakage (OR 2.82; 95% CI, 1.03-7.72), a higher reoperation rate (OR 5.38; 95% CI: 1.13 - 25.76), and the recurrence rate (RR 4.63; 95% CI, 1.51-14.20) for the extracranial approach. The event of infected CSF leakage (OR 3.69; 95% CI, 0.52-26.37) was higher in the extracranial than intracranial approach without any statistical significance. The extracranial approach was associated with a higher risk of CSF leakage, reoperation rate to treat the CSF leakage, and the recurrence rates. The infected CSF leakage between the extracranial and intracranial approaches showed no significant difference.