A Prediction Model for Postoperative Pulmonary Complication in Pulmonary Function-Impaired Patients Following Lung Resection

Abstract

Xiaowei Mao,1,2,* Wei Zhang,1,3,* Yi-Qian Ni,1 Yanjie Niu,1 Li-Yan Jiang1 1Pulmonary and Critical Care Medicine, Shanghai Jiao Tong University, Shanghai Chest Hospital, Shanghai, People’s Republic of China; 2Pulmonary and Critical Care Medicine, Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University, Hangzhou, People’s Republic of China; 3Department of Internal Medicine, American-Sino Women’s & Children’s Hospital, Shanghai, People’s Republic of China*These authors contributed equally to this workCorrespondence: Li-Yan JiangPulmonary and Critical Care Medicine, Shanghai Chest Hospital, Shanghai Jiao Tong University, No. 241 West Huaihai Road, Shanghai, People’s Republic of ChinaTel/Fax [phone]-3801Email [email protected]: Most patients with lung cancer have impaired pulmonary function. Single pulmonary function parameters have been suggested as good indices for predicting postoperative pulmonary complications (PPC). The purpose of this retrospective study was to construct a prediction model, including more than one pulmonary function parameter, for better prediction of PPC in patients with lung cancer and impaired pulmonary function.Patients and Methods: Our database of patients who underwent lung resection for non-small cell lung cancer was reviewed and those with impaired pulmonary function were enrolled. Clinical data, including PPC, were recorded. Univariate and logistic regression analyses were applied to explore potential predictors and a prediction model constructed based on the results of logistic regression.Results: Patients with impaired pulmonary function (n = 124) were enrolled. Most patients were male, current smokers, > 60 years old, and had adenocarcinoma and mild ventilatory dysfunction or diffusion dysfunction. In univariate analysis, we identified six pulmonary function parameters that differed significantly between the PPC and non-PPC groups. Receiver operating characteristic curves were used to determine the best cutoff values. In logistic regression, only forced expiratory volume in 1 second/forced vital capacity (FEV1/FVC%), peak expiratory flow (PEF%), and post predictive operation (ppo)-FEV1% remained significant. Based on these results, we constructed a prediction model for PPC including FEV1/FVC%, PEF%, and ppo-FEV1%, which had an good diagnostic performance of, with 76.7% sensitivity and 67.6% specificity.Conclusion: Our prediction model, including the pulmonary function parameters, FEV1/FVC%, PEF%, and ppo-FEV1%, shows excellent performance for predicting PPC in patients with lung cancer and impaired pulmonary function following resection, and has potential for wide application in clinical practice.Keywords: non-small-cell lung, retrospective studies, forced expiratory volume, logistic models, respiratory function tests

Keywords

• non-small-cell lung
• retrospective studies
• forced expiratory volume
• logistic models
• respiratory function tests;