The reliability of fMRI in brain tumor patients: a critical review and meta-analysis.

The reliability of fMRI in brain tumor patients: a critical review and meta-analysis.

World Neurosurg. 2019 Feb 08;:

Authors: Metwali H, Raemaekers M, Kniese K, Kardavani B, Fahlbusch R, Samii A

Abstract
INTRODUCTION: Formal studies that validated fMRI against direct cortical stimulation (DCS) show inconsistencies.
METHODS: We performed a review of literature and meta-analysis, including the studies that validated the fMRI using DCS according to PRISMA guidelines. The meta-analysis was performed using the studies that met our inclusion criteria. The positive predictive values (PPV), negative predictive values (NPV), false omission rates (FOR), false discovery rates (FDR), sensitivity and specificity have been calculated. We were able to identify the factors that could influence the fMRI analysis.
RESULTS: Accordingly, six studies concerning with language activation, and two studies concerning with motor activation were eligible for this meta-analysis. The PPV of fMRI for detection of eloquent motor cortex was 59.5% and the FDR was 40.5%. The NPV was 96% and the FOR 3.9%. The sensitivity of fMRI for detection of eloquent motor cortex was 91%, and specificity 76%. As for language activations, the PPV was 71.7%, NPV 80%, FDR 28.2%, and FOR 19.5%. The sensitivity was 80% and specificity 71.5%. Factors that could have caused mismatch or that affect the reliability have been analyzed.
CONCLUSION: Our current analysis of the available evidence in literature suggests that fMRI itself (due to neurovascular uncoupling), or its analysis suffer from limitations regarding reliability when validated against DCS. Based on the existing studies, our meta-analysis indicates that using the fMRI alone for surgical planning could lead to undesirable outcome.

PMID: 30743033 [PubMed - as supplied by publisher]

from A via a.sfakia on Inoreader http://bit.ly/2SJaMht