Assessing the benefits of lung surgery: 3D reconstruction vs. conventional methods - a pilot study

Video-assisted thoracoscopic surgery (VATS) is increasingly performed as sublobar lung resection, which is planned with 3D reconstructions of HRCT-scans. To assess the coherency between lung tissue quality as calculated with Visible Patient 3D-reconstructions software (VP) and pulmonary function (PF). This single-centre pilot study included 20 patients (mean age 68 y ± 10, 10 male) who underwent VATS resection of the right upper lobe between October 2020 and October 2022. VP simulations of lung volume (healthy to total) were related to pre- and 6 months postoperative PF and compared with predictions based on conventional methods (5% per segment rule, formula proposed by Brunelli et al., 2005). Patients (8/20) with higher postoperative than preoperative FEV1 showed a lower proportion of healthy tissue in preoperative VP simulations (mean (SD): 76 (18) % vs. 89 (10) %, p=0.045), measured postoperative FEV1 [l] (2.4 (0.56) to 2.3 (0.52) vs. 2.3 (0.55) to 2.6 (0.63)). The measured postoperative DLCO [mmol/(min*kPa)] remained constant (6.3 (2.3) to 6.3 (2.5) vs. 5.7 (1.5) to 6.2 (1.7)). Predictions using conventional methods underestimate the actual postoperative FEV1. Calculations using the 5% rule (2.0 (0.47) vs. 2.3 (0.58)) correlated better than the Brunelli formula (1.7 (0.32) vs. 1.8 (0.40)), which required a correction factor. DLCO calculated with the Brunelli formula predicted the measured DLCO (6.3 (2.7) vs. 6.2 (1.7)), while the 5% rule failed (5.6 (2.4) vs. 5.4 (1.6)). This pilot study shows that VP appears to be better at identifying patients who will benefit from surgery than conventional methods. However, further multicentric studies are needed to confirm our initial results.