To Örebro University

Objectives: Postoperative pulmonary complications (PPC) are common following lung resections, but there is no consensus in the literature on the definition of a clinically relevant PPC. This study aimed to use the Melbourne Scoring Scale (MGS) to determine the frequency and predictors of PPC in patients scheduled for lung resection on suspicion of or due to cancer.

Methods: In a prospective observational design, we assessed 87 consecutive patients following lung resections in Aalborg University Hospital, Denmark. Patients were preoperatively classified as being at high PPC-risk (n= 68) or low PPC-risk (n=19), based on the presence of one or more of the items: FEV₁ or carbon monoxide diffusion capacity (DLCO) ≤70%, age ≥70 years or scheduled pneumonectomy. Data on PPC was collected daily and re-evaluated two weeks postoperatively. Multivariate regression analysis was used to evaluate variables associated with PPC.

Results: The actual frequency of PPC according to the MGS was 11% (n=10), all cases within the predefined high-risk group, with pneumonia accounting for 10% of the cases. We found that preoperative FEV₁ and DLCO ≤60% were significantly associated with a higher PPC risk (area under the ROC curve 0.851), 95% CI 2.2-56.6 and 1.1-36.8 for FEV1 and DLCO, respectively.

Conclusions: The MGS can be used to identify patients at high risk of postoperative clinically relevant PPC after lung resections, in particular in patients with preoperative values of FEV1 ≤ 60% or DLCO ≤ 60%. More research is needed to evaluate the effect of preventable interventions targeting patients at high-risk of developing PPC.