A clinical risk model for the evaluation of bronchopleural fistula in non-small cell lung cancer after pneumonectomy.

BACKGROUND: There are no reliable risk factors to predict bronchopleural fistula (BPF) formation in patients undergoing pneumonectomy for non-small cell lung cancer (NSCLC). This study aims to create a validated clinical model based on the risk factors for BPF after pneumonectomy. The model to estimate the risk of BPF may help select patients for intervention therapy to reduce the rate of BPF after pneumonectomy.
METHODS: This retrospective analysis included 684 patients with NSCLC who underwent pneumonectomy at our institution from 1995 to 2012. The rates of BPF were estimated by the Kaplan-Meier method. Univariate and multivariate analyses were performed to identify the independent risk factors for the BPF and based on which a clinical model for the prediction of the incidence of BPF was formed.
RESULTS: The incidence of BPF was 4.4% (30 of 684 patients). Three factors were independently associated with BPF after pneumonectomy for NSCLC: neoadjuvant therapy (hazard ratio, 2.479), diabetes mellitus (hazard ratio, 1.061), and age 70 years or older (hazard ratio, 1.175). A scoring system for BPF was developed by assigning 2 points for a major risk factor (neoadjuvant therapy) and 1 point for each minor risk factor (diabetes mellitus and age ≥ 70 years). The 684 patients were divided into a low-risk group (score, 0 to 1), moderate-risk group (score, 2), and high-risk group (score, ≥ 3), with respective incidences of early BPF after pneumonectomy of 2.4%, 18.2%, and 58.3%
CONCLUSIONS: This model, based on readily available clinical characteristics, can estimate the risk of BPF after pneumonectomy in the NSCLC patients, independent of early BPF and late BPF classifications. This model could be used to select patients for intervention therapy (parenteral alimentation, control of blood glucose level, oxygen therapy, and strengthening the antibiotic treatment) if validated in independent data sets.