Risk factor and prediction model construction for oral mucosal pressure injuries in patients with endotracheal intubation in emergency intensive care unit

Objective To explore the risk factors for oral mucosal pressure injuries (OMPI) in patients with endotracheal intubation in the emergency intensive care unit (EICU) and to construct a nomogram prediction model based on these factors.

Methods A case-control study design was adopted to retrospectively collect clinical data from 209 adult patients with endotracheal intubation admitted to EICU. The patients were divided into OMPI group(53 patients)and non-OMPI group (156 patients) based on whether OMPI occurred during the observation period. The clinical data of the two groups were analyzed, and multivariate Logistic regression analysis was used to screen risk factors for OMPI in patients with endotracheal intubation in the EICU. R software was used to draw a nomogram prediction model, and the predictive performance of the model was evaluated through the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis.

Results Statistically significant differences were observed between the two groups in prone position ventilation, vasoconstrictor use, consciousness at the time of intubation, Acute Physiology and Chronic Health Evaluation Ⅱ (APACHE Ⅱ) score at the time of intubation, and duration of endotracheal intubation (P < 0.05). The results of multivariate Logistic regression analysis showed that prone position ventilation (OR=2.545, 95%CI, 1.261 to 5.135), vasoconstrictor use (OR=1.984, 95%CI, 1.162 to 3.387), inability to express complaints at time of intubation (OR=3.618, 95%CI, 1.891 to 6.924), high APACHE Ⅱ score (OR=2.394, 95%CI, 1.322 to 4.336), and long duration of endotracheal intubation (OR=3.995, 95%CI, 1.857 to 8.593) were all risk factors for OMPI in patients with endotracheal intubation in the EICU (P < 0.05). ROC curve analysis showed that the area under the curve of the nomogram prediction model was 0.881; calibration curve analysis showed that the mean absolute error between the predicted probability and the actual probability of the model was 0.016; and decision curve analysis showed that the prediction model had practical value in clinical practice.

Conclusion Prone position ventilation, vasoconstrictor use, inability to express complaints at the time of intubation, high APACHE Ⅱ score, and long duration of endotracheal intubation are all risk factors for OMPI in patients with endotracheal intubation in the EICU. The nomogram model constructed based on these factors has good predictive performance for OMPI risk.