Document resource
Background To analyze the influencing factors for acute gastrointestinal injury (AGI) in critically ill patients with chronic obstructive pulmonary disease (COPD) who undergo mechanical ventilation, and to develop a predictive model.Methods A total of 908 critically ill COPD patients who received mechanical ventilation between 2008 and 2022 were selected from the MIMIC database as study subjects. Based on the occurrence of AGI, they were divided into two groups: the complication group (n=158, with AGI) and the non-complication group (n=750, without AGI). Univariate analysis was performed on the clinical data of both groups. Factors with statistically significant differences were then incorporated into a multivariate stepwise logistic regression model. A predictive model was developed based on the identified influencing factors, and its predictive performance was evaluated.Results The duration of intubation was longer in the complication group compared to the non-complication group (P < 0.05), the APACHE II score was higher (P < 0.05), the PaO 2/FiO2 ratio (oxygenation index) was lower (P < 0.05), and the incidence of ventilator-associated pneumonia (VAP) was higher (P < 0.05). Multivariate stepwise logistic regression analysis showed that prolonged intubation [OR = 1.083 (95% CI: 1.014, 1.157)], concurrent VAP [OR = 2.226 (95% CI: 1.280, 3.172)], and a high APACHE II score [OR = 1.354 (95% CI: 1.101, 1.607)] were independent risk factors for AGI in mechanically ventilated, critically ill COPD patients (all P < 0.05). A predictive model was constructed based on the partial regression coefficients of the independent variables, and a nomogram was developed. The area under the curve (AUC) for this model was 0.880, with a sensitivity of 87.6% (95% CI: 0.771, 0.982) and a specificity of 87.8% (95% CI: 0.856, 0.901). The model demonstrated a good fit on the calibration curve.Conclusions Prolonged intubation, concurrent VAP, and a high APACHE II score are independent risk factors for AGI in mechanically ventilated, critically ill COPD patients. The predictive model based on these factors can accurately estimate the risk of developing AGI in this patient population.