Comparison of Various Measures for Anticipating Difficult Laryngoscopy in Turkish Population: An Observational Study
Derya Özden Omaygenç1, Oylum Selçuk21Department of Anesthesiology, University of Health Sciences Türkiye, Istanbul Haseki Training and Research Hospital, Istanbul, Türkiye2Department of Anesthesiology, University of Health Sciences Türkiye, Istanbul Training and Research Hospital, Istanbul, Türkiye
INTRODUCTION: Difficult laryngoscopy (DL) was established as the most accurate determinant of difficult intubation. Here, we sought to assess the diagnostic value of various bedside tests for predicting DL. We also aimed to create a regression model in this context.
METHODS: 137 patients were included in the study. Demographic features and eight diagnostic variables were evaluated for DL predictivity. These were retrognathia, presence of buck teeth, modified Mallampati test (MMT), upper lip bite test (ULBT), sternomental distance, interincisor distance (IID), thyromental distance, and neck circumference. DL was identified by Grade III-IV view during laryngoscopy according to the Cormack-Lehane classification.
RESULTS: The frequency of DL was 27% (n=37) in our sample population. Among predictive tests, IID was lower in patients with DL, while mean MMT score, frequency of retrognathia, and that of Grade 3 in ULBT were significantly higher. According to the results of logistic regression analyses IID (Odds Ratio [OR]: 0.504, Confidence intervals [CI] 95% [0.260–0.978], p=0.043), MMT score (OR: 2.001, CI 95% [1.159–3.454], p=0.013), and presence of retrognathia (OR: 0.108, CI 95% [0.019–0.613], p=0.012) were determined as independent predictors of DL. Our predictive model (two out of three factors: IID ≤4 cm, MMT score≥3, and retrognathia) anticipated DL with a sensitivity of 35.1%, a specificity of 91%, a negative predictive value of 79.1%, and an accuracy of 75.9%.
DISCUSSION AND CONCLUSION: None of the bedside tests or their dual combinations had considerable success for predicting DL in our study. Considering this fact and variation of the predictive performance of bedside tests for DL among different ethnicities, we composed a distinctive regression model. This model anticipated DL with reasonable specificity and accuracy rates.
Makale Dili: İngilizce