World Neurosurg. 2022 Oct 26:S1878-8750(22)01490-5. doi: 10.1016/j.wneu.2022.10.073. Online ahead of print.
ABSTRACT
OBJECTIVE: Reported rates of failures varies greatly from below 5% up to over 50% and, no meta-analysis to assess the overall prevalence has ever been performed. Herein, the authors estimated the failure rate after internal CSF shunt (ICSFS) insertion and searched for associated factors.
METHODS: Six databases were searched from January 1990 to February 2022. Only original articles reporting the rate of adult shunt failure were included. Random-effects meta-analysis with generalised linear mixed model method and logit transformation was used to compute the overall failure prevalence. Subgroup analysis and meta-regression were implemented to search for associated factors.
RESULTS: Of 1,763 identified articles, 46 were selected, 70,859 ICSFS implantations and for 13,603 shunt failure, suggesting an accumulated incidence of 19.2%. However, the calculated pooled prevalence value and its 95% confidence interval (CI) were 22.7% 95%CI[19.8-5.8]. The CI of the different estimates did not overlap, indicating a strong heterogeneity confirmed by a high I2 of 97.5%, 95%CI[97.1-97.8], p<0.001 with a τ2=0.3. Ninety-five percent prediction interval of shunt failure prevalence ranged from 8.75% up to 47.36%. A meta-regression of prevalence by year of publication found a barely significant decreasing failure rate of about two percent per year (-2.11, 95%CI[-4.02- -0.2], p-value=0.031).
CONCLUSION: Despite being a quite simple neurosurgical procedure, ICSFS insertion has one of the highest risk of complications with failure prevalence involving more than 1 patient out of 5. Nonetheless, all the efforts to lower these frustrating high level of shunt failure seem to be effective.
TRIAL REGISTRATION: This meta-analysis is registered in https://www.crd.york.ac.uk/PROSPERO/, PROSPERO ID: CRD42022339954.
PMID:36309337 | DOI:10.1016/j.wneu.2022.10.073
