Abstract
PURPOSE: Failure to wean from mechanical ventilation can lead to prolonged hospital stays, increased incidence of ventilator-associated pneumonia, and higher mortality rates. This study aimed to explore the effectiveness of the diaphragm contraction pressure index (DCPI) in predicting weaning outcomes in patients undergoing mechanical ventilation, providing a scientific basis for successful weaning in clinical practice. PATIENTS AND METHODS: This prospective observational study included 286 individuals in the derivation cohort and 104 patients in the validation cohort, all of whom completed the spontaneous breathing trial (SBT). During SBT, ultrasound was used to quantify the right hemidiaphragm excursion (DE), diaphragm thickness (DTF) after inspiration and expiration, and DCPI. MIP values were gathered from the mechanical ventilator when the patients breathed peacefully. The derivation cohort determined the cut-off value of DCPI and compared these ultrasound diaphragm parameters. The validation cohort contributes to verifying the accuracy of DCPI. RESULTS: The weaning success group's DCPI in the derivation cohort was significantly higher than that of the weaning failure group (36.67% ± 7.02% vs 24.03% ± 5.78%, P < 0.001). While the area under the receiver operating characteristic curve (ROC) (AUC) of DE, DTF, and MIP was 0.698 (95% CI, 0.615-0.771, P < 0.001), 0.770 (95% CI, 0.693-0.837, P < 0.01), and 0.811 (95% CI, 0.737-0.872, P < 0.001), the ROC of DCPI was 0.954 (95% CI, 0.905-0.982, P < 0.001), indicating good predictive performance for weaning success. The DCPI had a sensitivity of 94.1% and a specificity of 90.8%, with the ideal cut-off value set at ≥30.0%. Similarly, in the validation cohort, the AUC of DCPI for the predicted value is 0.952 (95% CI, 0.854-0.992, P < 0.001). CONCLUSION: Compared with DTF, DE, and MIP, DCPI dramatically improves the accuracy of predicting successful weaning. TRIAL REGISTRATION: No. ChiCTR2100052470, Registered 28 October 2021.