Poor prognostic value of the modified Mallampati score: a meta-analysis involving 177 088 patients

L H Lundstrøm1, M Vester-Andersen, A M Møller

  • 1Department of Anaesthesia and Intensive Care, Herlev Hospital, Copenhagen University Hospital, Herlev, Denmark.

Insights

The modified Mallampati score is not a reliable standalone predictor for difficult tracheal intubation. This meta-analysis indicates its prognostic value is insufficient, suggesting it should be part of a larger predictive model.

Area of Science:

  • Anesthesiology
  • Critical Care Medicine
  • Otolaryngology

Background:

  • The modified Mallampati score is a widely used clinical tool to predict difficult tracheal intubation.
  • However, its accuracy and prognostic value as a standalone test remain debated.

Purpose of the Study:

  • To conduct a comprehensive meta-analysis evaluating the modified Mallampati score's effectiveness in predicting difficult tracheal intubation.
  • To assess its diagnostic accuracy using pooled estimates of sensitivity, specificity, and likelihood ratios.

Main Methods:

  • A systematic review and meta-analysis of published studies were performed.
  • Data from 55 studies, including 177,088 patients, were analyzed using random-effects models and summary receiver operating characteristic (SROC) curves.
  • Meta-regression was employed to investigate heterogeneity.

Main Results:

  • The SROC curve showed an area under the curve (AUC) of 0.75.
  • A modified Mallampati score of III or IV yielded a pooled odds ratio of 5.89 for difficult intubation.
  • Pooled sensitivity was 0.35 and specificity was 0.91, with positive and negative likelihood ratios of 4.13 and 0.70, respectively. Significant heterogeneity was observed across studies.

Conclusions:

  • The modified Mallampati score demonstrates inadequate prognostic value as a standalone predictor for difficult laryngoscopy or tracheal intubation.
  • Its performance appears worse than previously estimated, highlighting the need for caution when relying solely on this score.
  • The score may be more useful as a component within a multivariate model for predicting difficult tracheal intubation.