Abstract
Objective: 1) To assess the association between tongue position, classified by both Mallampati (MP) and Friedman Tongue Position (FTP) scales, and severity of sleep apnea (OSA) as determined by the apnea-hypopnea index (AHI). 2) To determine which of these assessment methods is most closely correlated with OSA severity.
Method: PubMed, MEDLINE, and Cochrane Trial Registry through December 2011 were searched, combined with review of relevant article bibliographies. All studies were assessed by 3 reviewers. Systematic review and random-effects meta-analysis of studies assessing the association of tongue position and OSA severity were performed. Outcomes are reported as correlations.
Results: Ten studies met inclusion criteria and had data suitable for pooling (2515 patients). Overall, increasing tongue position (categorical variable) was significantly associated with increasing OSA severity (continuous variable of AHI) with a correlation of .351 (95% CI .094, .564; P = .008). Sensitivity analysis revealed correlations of .184 (95%CI .052, .310; P = .006) and .388 (95% CI .049, .646; P = .026) for MP and FTP, respectively. Analysis of publication bias revealed a nonsignificant Egger’s regression intercept; however, 4 imputed values to the right of the mean were found using Duval and Tweedie’s trim and fill method. Adding these values yielded an overall correlation of .498 (95% CI .474, .521).
Conclusion: Both the MP and FTP assessment techniques are significantly correlated with OSA severity. Publication bias does not affect or skew our results. The strength of the correlation between AHI and tongue position is greater for FTP than for MP, though the 95% confidence intervals for their respective correlation coefficients overlap.
Get full access to this article
View all access options for this article.