Development and validation of a nomogram model to predict primary graft dysfunction in patients after lung transplantation based on the clinical factors.

Abstract

Primary graft dysfunction (PGD), a significant complication that can affect patients' prognosis and quality of life, develops within 72 h post lung transplantation (LTx). Early detection and prevention of PGD should be given special consideration. The purpose of this study was to create a clinical prediction model to forecast the occurrence of PGD. We collected information on 622 LTx patients from Wuxi People's Hospital from 2016 to 2020 and used the data to construct the prediction model. Information on 224 patients from 2021 to June 2022 was used for external validation. We used LASSO regression for variable screening. A nomogram was developed for model presentation. Distinctness, fit, and calibration were used to evaluate the performance of the model. Subjects with respiratory failure, who received fresh frozen plasma, donor age, donor gender, donor mechanism of death, donor smoking, donor ventilator use time, and donor PaO 2/FiO 2 ratio were independent predictor variables for the occurrence of PGD. The area under the curve of the nomogram was .779. The Hosmer-Lemeshow test showed a good model fit (P= .158). The calibration curve of the nomogram is fairly close to the ideal diagonal. Moreover, the decision curve analysis revealed a positive net benefit of the model. External validation also confirmed the reliability of the model. The nomogram of PGD based on clinical risk factors in postoperative LTx patients was established with high reliability. It provides clinicians and nurses with a new and effective tool for early prediction of PGD and early intervention.