Simple Summary
Guidelines recommend using the segment count method (SCM) for prediction of post-surgical lung function (DL,CO and FEV1) after lung cancer resection. Earlier studies indicated that a 10% prediction error of SCM concerning FEV1 is common. We explored the prevalence of prediction error in DL,CO in a retrospective cohort. A DL,CO prediction error >10% was observed in 74%, most often due to underestimation (63%, overestimation: 11%). Prediction error increased inversely with pre-surgical DL,CO, indicating that recovery of DL,Co is larger than expected. Future studies should investigate underlying mechanisms.
Abstract
Background: The segment count method (SCM) is standard for predicting post-operative pulmonary function tests (ppo-PFTs) in patients undergoing surgical resection for non-small cell lung cancer. Accurate prediction is pivotal for providing optimal treatment; however, earlier studies indicate that SCM underestimates ppo-PFTs. This study aimed to investigate whether this underestimation correlates with the specific lobe resected or pre-operative PFT values. Methods: This retrospective single-centre study evaluated patients who underwent surgical resection (wedge, segmentectomy, or lobectomy) for lung cancer. Clinical data were extracted from electronic medical files. PFTs (forced expiratory volume in the first second, FEV1, and diffusing capacity of the lungs for carbon monoxide, DL,CO) were performed before and within 1 year after surgery and reported as % predicted. The SCMs were calculated as the pre-operative PFTs minus the proportion related to the number of resected segments. Pre-operative DL,CO was categorised as 40–60%, 61–80%, and >80%. Results: In total, 117 patients were included (lobectomy n = 102, wedge resection n = 15). PFTs were preserved with wedge resection. In patients undergoing lobectomy, a >10% prediction error of SCM on DL,CO was observed in 74% (FEV1: 63%) and most often due to higher-than-predicted (DL,CO: 63%; FEV1: 52%) than lower-than-predicted PFT (DL,CO: 11%; FEV1: 11%). A higher BMI was significantly associated with lower-than-predicted PFTs. No association was found regarding the resected lobe. Notably, the prediction error of SCM increased with decreasing DL,CO: post-operative DL,CO was preserved in patients with pre-operative DL,CO 40–60%. Conclusions: The segment count method exhibits systematic imprecision that correlates inversely with baseline DL,CO levels. Further studies are needed to validate this finding and to investigate how the SCM can be refined to reduce prediction error, especially in patients with the lowest preoperative PFTs.