OBJECTIVE: Age (in years) of the child has conventionally been used in formulae to estimate the tracheal tube (TT) size. The objective of this retrospective study was to test a weight-based formula (WBF) for uncuffed oral TT in children and compare it with the conventional age-based formula (ABF). METHODS: The patient's age, weight, and size of TT internal diameter (ID) were recorded. For comparative analysis, the actual TT size used was compared with predicted TT size, calculated using both the standard ABF [ID = age (years)/4 + 4 mm] and the WBF [ID = weight (kg)/10 + 3.5 mm]. RESULTS: The Pearson's correlation coefficient for age and actual TT size used was 0.77 (95% CI: 0.74-0.80) and between weight and actual TT used was 0.70 (95% CI: 0.66-0.74). The ABF correctly predicted 51.3% of TT sizes while the WBF correctly predicted 44.8% of TT sizes (P = 0.01). The measures of agreement between the actual and predicted TT size were 0.35 and 0.27 for the ABF and WBF respectively. The difference between the percentages of paired predictions for the ABF and WBF was statistically significant (P < 0.001) suggesting that, when correctly predicting the actual tube size used, the WBF functions for a different subset of the patient cohort than the ABF. CONCLUSIONS: This study suggests that in this patient cohort, the WBF is statistically inferior to the conventional ABF. However, our findings also suggest that the WBF may correctly predict TT sizes in a subset of patients in whom the ABF is inaccurate.
OBJECTIVE: Age (in years) of the child has conventionally been used in formulae to estimate the tracheal tube (TT) size. The objective of this retrospective study was to test a weight-based formula (WBF) for uncuffed oral TT in children and compare it with the conventional age-based formula (ABF). METHODS: The patient's age, weight, and size of TT internal diameter (ID) were recorded. For comparative analysis, the actual TT size used was compared with predicted TT size, calculated using both the standard ABF [ID = age (years)/4 + 4 mm] and the WBF [ID = weight (kg)/10 + 3.5 mm]. RESULTS: The Pearson's correlation coefficient for age and actual TT size used was 0.77 (95% CI: 0.74-0.80) and between weight and actual TT used was 0.70 (95% CI: 0.66-0.74). The ABF correctly predicted 51.3% of TT sizes while the WBF correctly predicted 44.8% of TT sizes (P = 0.01). The measures of agreement between the actual and predicted TT size were 0.35 and 0.27 for the ABF and WBF respectively. The difference between the percentages of paired predictions for the ABF and WBF was statistically significant (P < 0.001) suggesting that, when correctly predicting the actual tube size used, the WBF functions for a different subset of the patient cohort than the ABF. CONCLUSIONS: This study suggests that in this patient cohort, the WBF is statistically inferior to the conventional ABF. However, our findings also suggest that the WBF may correctly predict TT sizes in a subset of patients in whom the ABF is inaccurate.