To investigate the efficacy of arterial partial pressure of oxygen (PaO2), procalcitonin (PCT) combined with ROX index in predicting the timing of tracheal intubation in patients with acute severe pancreatitis (SAP). A case-control study was conducted. A total of 148 patients with SAP admitted to Hunan Provincial People's Hospital from January 2019 to December 2022 were selected as the research objects. According to whether endotracheal intubation was used after admission during hospitalization, the patients were divided into the intubation group (102 cases) and non-intubation group (46 cases). Gender, age, white blood cell count (WBC), lymphocyte count (LYM), platelet count (PLT), C-reactive protein (CRP), hemoglobin (Hb), PCT, PaO2, arterial partial pressure of carbon dioxide (PaCO2), arterial bicarbonate ion (HCO3-) 1 day after admission, arterial lactic acid (Lac), lactate dehydrogenase (LDH), heart rate (HR), respiratory rate (RR), pulse oxygen saturation (SpO2), oxygenation index (PaO2/FiO2), blood pressure, worst ROX index (ROX index = SpO2/FiO2/RR) within 30 minutes of admission and 30 minutes before intubation of the two groups were measured. Multivariate Logistic regression was used to analyze the independent risk factors for the timing of endotracheal intubation in patients with SAP. The receiver operator characteristic curve (ROC curve) was used to determine the optimal predictive cut-off value for endotracheal intubation. There were no significant differences in age, gender, WBC, LYM, CRP, Hb, LDH, HR and blood pressure at admission between the two groups. The PLT, Lac, PCT and RR in the intubation group were significantly higher than those in the un-intubation group, and HCO3-, PaO2, SpO2, PaO2/FiO2, the worst ROX index within 30 minutes after admission and 30 minutes before intubation were significantly lower than those in the non-intubation group (all P < 0.05). Logistic regression analysis showed that the worst ROX index within 30 minutes before intubation was the largest negative influencing factor for the timing of tracheal intubation in SAP patients [odds ratio (OR) = 0.723, 95% confidence interval (95%CI) was 0.568-0.896, P = 0.000], followed by PaO2 (OR = 0.872, 95%CI was 0.677-1.105, P < 0.001). PCT was the positive influencing factor (OR = 1.605, 95%CI was 1.240-2.089, P < 0.001). ROC curve analysis showed that the area under the ROC curve (AUC) of PaO2, PCT, the worst ROX index within 30 minutes before intubation and the combination to evaluate the tracheal intubation time of patients with SAP were 0.715, 0.702, 0.722 and 0.808, the sensitivity was 78.1%, 75.0%, 81.5% and 89.3%, the specificity was 66.7%, 59.0%, 73.2% and 86.4%, and the best cut-off value was 60.23 mmHg (1 mmHg ≈ 0.133 kPa), 2.72 μg/L, 4.85, and 0.58, respectively. The AUC of the combination of PaO2, PCT and the worst ROX index within 30 minutes before intubation predicted the timing of tracheal intubation in patients with SAP was significantly greater than using each index alone (all P < 0.01). The worst ROX index within 30 minutes before intubation combined with PaO2 and PCT is helpful for clinicians to make a decision for tracheal intubation in patients with SAP.
Read full abstract