Comparison of segmentectomy and lobectomy for non-small cell lung cancer with visceral pleural invasion Authors: Masaya Tamura, Takashi Sakai, Naoki Furukawa, Yujiro Bunno, Marino Yamamoto, Ryohei Miyazaki, Hironobu Okada Published: 2025 ABSTRACT BackgroundVisceral pleural invasion (VPI) is known to have a significant impact on staging and prognosis in NSCLC and is an important factor in determining surgical strategy. The aim of this study was to compare the outcomes of surgical procedure (segmentectomy vs. lobectomy) with VPI positive patients.MethodsA retrospective study was conducted on 218 VPI positive NSCLC patients were studied, with the segmentetomy group (n = 42) designated as cohort.1, cases with tumor diameter less than 2 cm (n = 22) designated as cohort.2, and cases with greater than 2 cm (n = 20) designated as cohort.3. Each group and 150 cases of lung lobectomy were analyzed. Cause of death and recurrence mode were investigated.ResultsThere was no difference in OS, DSS and RFS between segmentectomy and lobectomy after propensity matched analysis. There was no difference in OS and RFS in the group of cases < 2 cm. In patients with tumors > 2 cm, recurrence-free survival was numerically higher in the lobectomy group than in the segmentectomy group, although this difference was not statistically significant (p = 0.08). There was no difference in lung cancer deaths, but there was a trend toward more deaths from multiple disease in the lobectomy group (p = 0.07), and pleural dissemination recurrence was significantly more common in the segmentectomy group than in the lobectomy group in the > 2 cm group (p = 0.03).ConclusionIn patients with VPI-positive lung cancer, segmentectomy may offer a better prognosis for those with tumors measuring < 2 cm. Nevertheless, lobectomy should be considered first for patients with tumors measuring > 2 cm. Our study provides novel insights by stratifying outcomes based on tumor size, highlighting pleural dissemination risk and death causes, which may support individualized surgical decision-making. Although no statistically significant difference was found, lobectomy may be more appropriate for patients with tumors > 2 cm based on the observed recurrence patterns and potential for better local control. Background Surgical resection is currently the mainstay of curative treatment for early-stage non-small cell lung cancer (NSCLC). Visceral pleural invasion (VPI) has a significant impact on staging and prognosis and is an important factor in determining surgical strategy [1, 2]. Previous studies demonstrated that NSCLC patients with VPI have more progressive tumors and are at an increased risk of mediastinal lymph node metastasis [3, 4] and recurrence [5]. Therefore, lobectomy, which allows more adequate resection, has been considered the standard technique [6]. Nevertheless, segmentectomy is increasingly indicated for the preservation of pulmonary function, particularly in elderly patients and those with comorbidities. It was recently suggested that segmentectomy provides a comparable prognosis to that associated with lobectomy in patients with small tumors [7], and its appropriateness is also being examined in patients with pleural involvement. However, there is limited evidence on the prognostic comparison of segmentectomy versus lobectomy in NSCLC with VPI. Moreover, there is no clear consensus on the impact of tumor size on surgical selection and prognosis. The impact of resection technique on the recurrence pattern and cause of death has also not been fully investigated. In the present study, we retrospectively analyzed the effect of tumor size and surgical technique on prognosis, recurrence pattern, and cause of death in patients with NSCLC with pleural invasion. In particular, we aimed to gain insight into future treatment strategies by comparing the outcomes of segmentectomy and lobectomy in two groups of patients with tumor diameters measuring < 2 cm and > 2 cm. Unlike previous studies, which often did not stratify VPI-positive cases by tumor size or assess recurrence patterns in detail, our study uniquely analyzes the impact of tumor size on surgical outcomes in VPI-positive NSCLC using propensity score-matched cohorts. We also explore causes of death and recurrence modes, providing clinically relevant insights into personalized surgical decision-making. This study offers additional clinical value by stratifying patients with VPI based on tumor size and evaluating not only survival outcomes but also recurrence patterns and causes of death. These aspects have been underexplored in previous studies. Our findings may help refine surgical strategy, particularly in elderly or high-risk populations where lung function preservation is a critical consideration. Materials and methods Patient population The internal review board (ERB-111369) at our hospital approved this study. Overall, 922 patients with NSCLC who underwent curative lung resection from 2012 to 2022 were included in the present investigation. Of those, 218 patients with VPI were evaluated (Fig. 1). Patients who underwent segmentectomy (n = 42) were classified into cohort 1; those with tumors measuring < 1.0 cm (n = 8), 1.0–2.0 cm (n = 14), 2.1–3.0 cm (n = 12) and > 3.0 cm were classified into cohort 2 A, 2B, 3 A and 3B, respectively. This stratification was performed to align with the 8th edition of the TNM classification, which recognizes 1 cm and 2 cm cutoffs as clinically relevant for staging. These patients as well as 150 patients who underwent lobectomy were analyzed; subsequently, cohort 1 (n = 38), cohort 2 (n = 22), and cohort 3 (n = 20) were background-matched by the grade of VPI, tumor size, age, sex, histology, and pathological nodal stage. Table 1 shows the background of the 42 and 150 patients who underwent segmentectomy and lobectomy, respectively. Table 2 shows the patient background after propensity score matching (PSM). Fig. 1Flow chart of patient selection. NSCLC: non-small cell lung cancer Jan: January Dec: December VPI: visceral pleural invasion Flow chart of patient selection. NSCLC: non-small cell lung cancer Jan: January Dec: December VPI: visceral pleural invasion Table 1Characteristics of patients undergoing segmentectomy or lobectomyFactorsSegment resection(n = 42)Lobar resection(n = 150)p valueGender0.59Male30103Female1247Age0.24< 72 years2077≥ 72 years2273Lesion size< 0.0001< 3.1 cm3160≥ 3.1 cm1190Histology0.46Non adenoca1653Adenoca2697Pathological stage0.01IB2357IIA29IIB633IIIA738IIB413Pathological Nodal Stage0.11N03098N1617N2635Pleural invasion0.09pl12283pl21229pl3838 Table 1Characteristics of patients undergoing segmentectomy or lobectomyFactorsSegment resection(n = 42)Lobar resection(n = 150)p valueGender0.59Male30103Female1247Age0.24< 72 years2077≥ 72 years2273Lesion size< 0.0001< 3.1 cm3160≥ 3.1 cm1190Histology0.46Non adenoca1653Adenoca2697Pathological stage0.01IB2357IIA29IIB633IIIA738IIB413Pathological Nodal Stage0.11N03098N1617N2635Pleural invasion0.09pl12283pl21229pl3838 Characteristics of patients undergoing segmentectomy or lobectomy Table 2Patient characteristics of propensity score matched cohortFactorsSublobar resection(n = 38)Lobar resection(n = 38)p valueGender0.32Male2824Female1014Age0.91< 72 years1819≥ 72 years2019Lesion size0.79< 3.1 cm2829≥ 3.1 cm109Histology0.41Non adenoca1411Adenoca2427Pathological stage0.72IB2425IIA11IIB52IIIA79IIB11Pathological Nodal Stage0.52N02826N154N258Pleural invasion0.86pl12224pl2119pl355 Table 2Patient characteristics of propensity score matched cohortFactorsSublobar resection(n = 38)Lobar resection(n = 38)p valueGender0.32Male2824Female1014Age0.91< 72 years1819≥ 72 years2019Lesion size0.79< 3.1 cm2829≥ 3.1 cm109Histology0.41Non adenoca1411Adenoca2427Pathological stage0.72IB2425IIA11IIB52IIIA79IIB11Pathological Nodal Stage0.52N02826N154N258Pleural invasion0.86pl12224pl2119pl355 Patient characteristics of propensity score matched cohort Surgical procedure (segmentectomy or lobectomy) was selected based on a comprehensive evaluation of the patient’s general condition, pulmonary function, comorbidities, and tumor location. Segmentectomy was preferentially performed in patients with poor pulmonary reserve, higher surgical risk (e.g., advanced age, cardiovascular or respiratory comorbidities), or when the tumor was located in an anatomically suitable segmental region. Conversely, lobectomy was the preferred procedure when patients were deemed fit for major resection. To reduce potential selection bias inherent in this retrospective study design, propensity score matching (PSM) was employed using variables such as age, tumor size, histology, pathological nodal stage, and degree of pleural invasion (PL1–PL3), as detailed in Table 2. Patient outcomes and follow-up Standard follow-up for patients included chest computed tomography scans, physical examinations, and blood tests that were conducted every 3–6 months for the first 3 years following the procedure, and every 6–12 months thereafter. 18 F-fluorodeoxyglucose positron emission tomography and magnetic resonance imaging were also performed when necessary. Median duration of follow-up in our study was 54 months (mean: 49.2; range: 24–134 months). The primary outcome was recurrence-free survival (RFS), defined as the period from initial surgical resection to the first evidence of tumor recurrence. Recurrence was classified based on radiological findings, including chest computed tomography (CT) and fluorodeoxyglucose positron emission tomography (FDG-PET), as well as clinical assessment. Specifically, pleural dissemination (pleural seeding) was diagnosed by identification of new pleural nodules or plaques on imaging studies consistent with metastatic spread, accompanied when possible by histopathological confirmation through biopsy or cytology of pleural fluid. In cases where histological confirmation was not feasible, diagnosis was based on characteristic radiological progression and clinical course consistent with pleural dissemination. This comprehensive approach ensured accurate identification and classification of recurrence patterns. Statistics The associations between clinical outcomes and imaging and clinical metrics were evaluated using univariate and multivariate Cox proportional hazards regression models. Variables with p-values < 0.05 in univariate analysis were included in the multivariate analysis to identify independent prognostic factors. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated. A p-value < 0.05 was considered statistically significant. The 95% confidence intervals were calculated, and the percentages were two-sided. For continuous variables, presented as the mean ± standard deviation, the student’s t-test was employed to assess significant differences. For categorical variables, the chi-squared test was utilized to assess statistical significance. The SAS software package (SAS Institute, Inc., Cary, NC, USA) was used for statistical analyses. The p-values < 0.05 indicate statistically significant differences. Results Figure 2 shows the overall survival (OS), disease-specific survival, and RFS associated with each procedure in cohort 1. No statistically significant difference was found between the two groups (p = 0.24, 0.34, and 0.88, respectively). Figure 3 shows the OS, disease-specific survival, and RFS linked to each procedure after PSM in cohort 1. No statistically significant difference was found between the two groups (p = 0.47, 0.46, and 0.69, respectively). Fig. 2Kaplan-Meyer curve for patients who underwent different surgical procedures (Segmentectomy vs. Lobectomy). (A) Overall survival (B) Disease specific survival and (C) Recurrence free survival. Lob: Lobectomy Seg: Segmentectomy Kaplan-Meyer curve for patients who underwent different surgical procedures (Segmentectomy vs. Lobectomy). (A) Overall survival (B) Disease specific survival and (C) Recurrence free survival. Lob: Lobectomy Seg: Segmentectomy Fig. 3Kaplan-Meyer curve for patients who underwent different surgical procedures (Segmentectomy vs. Lobectomy): propensity score-matched analysis (1:1). (A) Overall survival (B) Disease specific survival and (C) Recurrence free survival Lob: Lobectomy Seg: Segmentectomy Kaplan-Meyer curve for patients who underwent different surgical procedures (Segmentectomy vs. Lobectomy): propensity score-matched analysis (1:1). (A) Overall survival (B) Disease specific survival and (C) Recurrence free survival Lob: Lobectomy Seg: Segmentectomy Figure 4 shows the OS and RFS following each procedure after PSM in cohort 2; similarly, no difference was found between the two groups (p = 0.17 and 0.94, respectively). Figure 5 shows the OS and RFS after each surgical procedure in the group of patients with tumors measuring > 2 cm (cohort 3). In patients with tumors > 2 cm (cohort 3), the lobectomy group showed a numerically higher recurrence-free survival compared to the segmentectomy group; however, the difference did not reach statistical significance (p = 0.08) (Fig. 5). Fig. 4Kaplan-Meyer curve for patients with tumor ≦ 2 cm who underwent different surgical procedures (Segmentectomy vs. Lobectomy): propensity score-matched analysis (1:1). (Cohort2) (A) Overall survival (B) Disease specific survival and (C) Recurrence free survival Lob: Lobectomy Seg: Segmentectomy PMA: propensity matched analysis Kaplan-Meyer curve for patients with tumor ≦ 2 cm who underwent different surgical procedures (Segmentectomy vs. Lobectomy): propensity score-matched analysis (1:1). (Cohort2) (A) Overall survival (B) Disease specific survival and (C) Recurrence free survival Lob: Lobectomy Seg: Segmentectomy PMA: propensity matched analysis Fig. 5Kaplan-Meyer curve for patients with tumor > 2 cm who underwent different surgical procedures (Segmentectomy vs. Lobectomy): propensity score-matched analysis (1:3). (Cohort3) (A) Overall survival (B) Disease specific survival and (C) Recurrence free survival Lob: Lobectomy Seg: Segmentectomy PMA: propensity matched analysis Kaplan-Meyer curve for patients with tumor > 2 cm who underwent different surgical procedures (Segmentectomy vs. Lobectomy): propensity score-matched analysis (1:3). (Cohort3) (A) Overall survival (B) Disease specific survival and (C) Recurrence free survival Lob: Lobectomy Seg: Segmentectomy PMA: propensity matched analysis Table 3 shows the causes of death according to each procedure after PSM. There was no difference between the groups in the rate of deaths related to lung cancer; however, there was a trend toward more deaths from other diseases in the lobectomy group versus the segmentectomy group. Table 4 shows the recurrence patterns after each procedure in the patients with tumors measuring > 2 cm. The segmentectomy group had a significantly higher rate of pleural dissemination recurrence than the lobectomy group. Table 3Summary of causes of death during follow-upSegmentectomy(n = 38)Lobectomy(n = 38)p value Alive 28 (73.7%)24 (63.1%) Cause of death lung cancer 6 (15.8%)5 (13.2%) Other 4 (10.5%)9 (23.7%)0.07 other cancer 2 pneumonia 32 IP 11 AMI 1 Renal failure 1 pulmonary embolism 1 hypoglycemic attack 1IP: interstitial pneumonia AMI: acute myocardial infarction Table 3Summary of causes of death during follow-upSegmentectomy(n = 38)Lobectomy(n = 38)p value Alive 28 (73.7%)24 (63.1%) Cause of death lung cancer 6 (15.8%)5 (13.2%) Other 4 (10.5%)9 (23.7%)0.07 other cancer 2 pneumonia 32 IP 11 AMI 1 Renal failure 1 pulmonary embolism 1 hypoglycemic attack 1IP: interstitial pneumonia AMI: acute myocardial infarction Summary of causes of death during follow-up IP: interstitial pneumonia AMI: acute myocardial infarction Table 4Recurrence pattern by surgical procedure after propensity matched analysisSegmentectomy (n = 20)Lobectomy (n = 60)p value None 11 (55.0%)43 (71.7%) Mode of recurrence Loco-regional 1 (5.0%)5 (8.3%) Dissemination 4 (20.0%)3 (5.0%)0.03 Distant 4 (20.0%)9 (15.0%) Table 4Recurrence pattern by surgical procedure after propensity matched analysisSegmentectomy (n = 20)Lobectomy (n = 60)p value None 11 (55.0%)43 (71.7%) Mode of recurrence Loco-regional 1 (5.0%)5 (8.3%) Dissemination 4 (20.0%)3 (5.0%)0.03 Distant 4 (20.0%)9 (15.0%) Recurrence pattern by surgical procedure after propensity matched analysis Univariate and multivariate Cox proportional hazards regression analyses were performed to identify independent prognostic factors for overall survival (OS), recurrence-free survival (RFS), and disease-free survival (DFS). In the multivariate analysis, gender, whole tumor size were identified as independent prognostic factors significantly associated with worse OS and pathological stage with worse RFS (p < 0.05). Visceral pleural invasion grade (PL2, PL3 vs. PL1) showed a trend toward worse prognosis but did not reach statistical significance. The details of the analysis are presented in Table 5. Table 5Multiple logistic regression analysis to predict overall survival (OS), disease specific survival (DSS) and recurrence free survival (RFS)VariablesOSDSSRFSUnivariate HRp valueMultivariate HRp valueUnivariate HRp valueMultivariate HRp valueUnivariate HRp valueMultivariate HRp value(95%CI)(95%CI)(95%CI)(95%CI)(95%CI)(95%CI)Age1.46 (0.75–2.84)0.262.01 (1.22–3.29)0.0061.86 (1.07–3.26)0.029༊1.33 (0.84–2.12)0.22Gender (male)2.39 (1.31–4.35)0.0052.47 (1.30–4.69)0.006༊1.89 (1.15–3.49)0.0142.41 (1.38–4.22)0.002༊1.17 (1.03–1.64)0.18Surgical procedure (≧ Lob)1.21 (0.58–2.67)0.581.19 (0.66–2.12)0.561.27 (0.47–1.32)0.37Whole tumor size8.11 (2.02–18.5)0.0021.98 (1.02–3.76)0.02༊2.46 (1.45–12.6)0.0071.89 (0.99–2.12)0.081.21 (1.09–1.36)0.0081.29 (0.79–2.11)0.31CEA (≧ 5.0)1.09 (0.96–1.32)0.421.12 (0.98–1.12)0.341.24 (1.03–1.65)0.21Histology (non-adenocarcinoma)1.16 (0.63–2.14)0.621.21 (0.76–1.93)0.411.57 (0.89–1.45)0.151.19 (0.73–1.95)0.47Pathological stage (≧ II)2.26 (1.16–4.39)0.0161.73 (0.81–3.68)0.152.49 (1.58–3.96)< 0.0012.01 (1.23–4.32)0.03༊2.84 (1.67–4.83)0.0011.89 (1.12–2.98)0.02༊Pleural invasion (pl 2 or 3)1.51 (0.96–2.37)0.071.65 (0.91–3.01)0.091.80 (1.13–2.86)0.0141.14 (0.68–1.93)0.60*indicates a statistically significant difference Table 5Multiple logistic regression analysis to predict overall survival (OS), disease specific survival (DSS) and recurrence free survival (RFS)VariablesOSDSSRFSUnivariate HRp valueMultivariate HRp valueUnivariate HRp valueMultivariate HRp valueUnivariate HRp valueMultivariate HRp value(95%CI)(95%CI)(95%CI)(95%CI)(95%CI)(95%CI)Age1.46 (0.75–2.84)0.262.01 (1.22–3.29)0.0061.86 (1.07–3.26)0.029༊1.33 (0.84–2.12)0.22Gender (male)2.39 (1.31–4.35)0.0052.47 (1.30–4.69)0.006༊1.89 (1.15–3.49)0.0142.41 (1.38–4.22)0.002༊1.17 (1.03–1.64)0.18Surgical procedure (≧ Lob)1.21 (0.58–2.67)0.581.19 (0.66–2.12)0.561.27 (0.47–1.32)0.37Whole tumor size8.11 (2.02–18.5)0.0021.98 (1.02–3.76)0.02༊2.46 (1.45–12.6)0.0071.89 (0.99–2.12)0.081.21 (1.09–1.36)0.0081.29 (0.79–2.11)0.31CEA (≧ 5.0)1.09 (0.96–1.32)0.421.12 (0.98–1.12)0.341.24 (1.03–1.65)0.21Histology (non-adenocarcinoma)1.16 (0.63–2.14)0.621.21 (0.76–1.93)0.411.57 (0.89–1.45)0.151.19 (0.73–1.95)0.47Pathological stage (≧ II)2.26 (1.16–4.39)0.0161.73 (0.81–3.68)0.152.49 (1.58–3.96)< 0.0012.01 (1.23–4.32)0.03༊2.84 (1.67–4.83)0.0011.89 (1.12–2.98)0.02༊Pleural invasion (pl 2 or 3)1.51 (0.96–2.37)0.071.65 (0.91–3.01)0.091.80 (1.13–2.86)0.0141.14 (0.68–1.93)0.60*indicates a statistically significant difference Multiple logistic regression analysis to predict overall survival (OS), disease specific survival (DSS) and recurrence free survival (RFS) *indicates a statistically significant difference Additional subgroup analysis based on tumor diameter (< 1.0 cm, 1.0–2.0 cm, and 2.1–3.0 cm) was performed. There were no statistically significant differences in OS or RFS between segmentectomy and lobectomy in the < 1.0 cm (Cohort 2 A) and 1.0–2.0 cm groups (Cohort 2B). However, in the 2.1–3.0 cm group (Cohort 3 A), lobectomy was associated with a trend toward improved RFS compared to segmentectomy (p = 0.11), consistent with the original analysis. To further investigate the impact of pleural invasion depth on prognosis, we performed a subgroup analysis stratified by the degree of pleural invasion (PL1, PL2, and PL3). Although the sample size in each subgroup was limited, the analysis revealed statistically significant worse recurrence-free survival with increasing pleural invasion (PL3 and PL2 > PL1) (Fig. 6). Notably, in patients with PL3, lobectomy showed a more favorable recurrence pattern compared to segmentectomy, including a lower incidence of pleural dissemination. These findings are summarized in Fig. 7. Fig. 6Kaplan-Meyer curve for recurrence free survival. Subgroup analysis stratified by the degree of pleural invasion (PL1, PL2, and PL3) Kaplan-Meyer curve for recurrence free survival. Subgroup analysis stratified by the degree of pleural invasion (PL1, PL2, and PL3) Fig. 7Kaplan-Meyer curve for patients who underwent different surgical procedures (Segmentectomy vs. Lobectomy). (A) Pl 1 (B) Pl 2 and (C) Pl3 Kaplan-Meyer curve for patients who underwent different surgical procedures (Segmentectomy vs. Lobectomy). (A) Pl 1 (B) Pl 2 and (C) Pl3 Discussion In the present study, we investigated the relationship between tumor size and surgical technique and its impact on prognosis in the surgical treatment of NSCLC with VPI. The results showed that in patients with tumors measuring < 2 cm, segmentectomy was as good as or better than lobectomy, while lobectomy was associated with numerically better outcomes in patients with tumors > 2 cm, without reaching statistical significance. These findings may be useful in reconsidering the indications for reduction surgery in patients with pleural invasion. Although VPI is a pathological finding typically confirmed postoperatively, our analysis focused on evaluating the oncologic safety of segmentectomy in patients found to be VPI-positive. While this limits its utility in preoperative surgical decision-making, the results provide useful insights for postoperative risk stratification and future treatment strategies, particularly in the era of improved imaging and intraoperative diagnostic tools. Furthermore, considering the potential prognostic impact of VPI, investigation into the role of adjuvant therapies in these patients is warranted and should be addressed in future studies. Pleural involvement is an important factor in the staging of early-stage NSCLC. The International Association for the Study of Lung Cancer has recognized the importance of VPI for up-staging in the 7th edition of the NSCLC staging system and incorporated it into the tumor-node-metastasis (TNM) staging system [8]. Subsequent studies have revealed that VPI is linked to poor prognosis in patients with early-stage NSCLC, and the PL classification has been reconsidered for inclusion in the 8th edition of the TNM staging system [9]. Several investigations, including the Japan Clinical Oncology Group 0802 study, have demonstrated the safety and effectiveness of segmental resection with hilar and mediastinal lymph node dissection in patients with early-stage NSCLC and tumors measuring ≤ 2 cm [7, 10]. Regarding the prognosis of early-stage lung cancer with VPI, the 5-year survival rates of patients with stage IB NSCLC who underwent surgical reduction or lobectomy were comparable [11]. Nonetheless, Yu et al. [12] showed that patients with tumors measuring < 2 cm and VPI had an improved long-term prognosis following resection. Furthermore, it has been reported that partial resection is sufficient in older patients (aged >80 years) with clinical stage I lung cancer even if the tumor is VPI-positive [13]. Several recent studies have reported reduced resection with VPI as the primary focus and lobectomy [13–15]; however, the appropriate surgical treatment remains controversial. In patients with tumors measuring < 2 cm, the local spread of the tumor is relatively low, and even with pleural invasion, the tumor is likely to be confined to the area; thus, zonal resection may provide adequate local control. In addition, zonal resection may be a useful option, especially for elderly patients or those with impaired respiratory function, because it is superior in preserving lung function, improves postoperative quality of life, and reduces the risk of other diseases. Nevertheless, if the tumor measures > 2 cm, it is likely to have more advanced pleural involvement, and there is a risk that a regional resection may be inadequate in terms of microinvasion and spread to adjacent areas. Lobectomy allows for more extensive lymph node dissection and local control, which may improve prognosis. In the present study, the frequency of pleural dissemination recurrence was significantly higher in the segmentectomy group than in the lobectomy group among patients with tumors > 2 cm. While we initially speculated that this might be related to the influence of resection margins or peritumoral pathological characteristics, we acknowledge that our study did not include detailed pathological margin analysis or assessment of the extent of microscopic invasion. Therefore, such conclusions remain hypothetical. It is also possible that pleural recurrence is more strongly influenced by the inherent biological aggressiveness of the tumor, rather than the surgical procedure itself. Future studies involving molecular profiling, immunohistochemical analyses, and detailed pathological evaluation are warranted to clarify the mechanisms underlying pleural recurrence. Furthermore, it should be noted that segmentectomy was selected in some patients with tumors > 2 cm due to advanced age, impaired pulmonary function, or other significant comorbidities that made lobectomy infeasible. Therefore, the choice of segmentectomy in these patients may have reflected a salvage strategy rather than a standard treatment plan. This introduces a potential selection bias, which could impact the interpretation of the outcomes in this subgroup. Although we attempted to mitigate this bias through propensity score matching (PSM), the possibility of residual confounding cannot be excluded. Therefore, our findings regarding tumors > 2 cm should not be overgeneralized, and further prospective studies are warranted to validate these results. The higher incidence of death from other causes in the lobectomy group may be due to the more invasive nature of the procedure, decreased pulmonary function after the procedure, and systemic risks derived from patient background. Therefore, it is important to construct an individualized treatment strategy that takes into account tumor factors and the general condition of the patient, as well as the presence or absence of complications when selecting a surgical approach; furthermore, a detailed analysis of risk factors is necessary. According to the evidence, in NSCLC patients with pleural invasion, regional resection may be a safe and effective option for those with tumors measuring < 2 cm, while lobectomy may be more appropriate for those with tumors measuring > 2 cm. Moreover, the degree of pleural involvement (PL1, PL2, and PL3) has been reported to affect prognosis [9]. In response to this, we performed an additional subgroup analysis stratified by PL status. The results suggested that patients with PL3 had worse recurrence-free and overall survival than those with PL1 or PL2, particularly in the segmentectomy group. This may be attributed to the increased risk of microscopic pleural dissemination in cases of deeper invasion. In patients with PL3, lobectomy may provide superior local control due to more extensive resection. Therefore, when considering segmentectomy in VPI-positive patients, PL classification should be taken into account. One of the novel contributions of this study is the identification of a significantly higher risk of pleural dissemination in patients with tumors > 2 cm undergoing segmentectomy. This finding suggests that, despite the increasing trend of segmentectomy in small NSCLC, careful attention should be paid when VPI is present. Furthermore, the observation that lobectomy patients had more deaths from non-cancer causes highlights the importance of tailoring the surgical approach to each patient’s comorbidities and general condition, which has not been extensively discussed in prior literature. This study had several limitations. Firstly, this is a retrospective, single-center investigation with a limited number of cases. Secondly, regarding the type of resection, limited resection was performed for high-risk patients, whereas lobectomy was more desirable. Thirdly, although propensity score matching was applied to reduce selection bias, the number of patients in each subgroup after matching was relatively small, especially in cohort 3 (patients with tumors > 2 cm undergoing segmentectomy, n = 20). This small sample size may have reduced the statistical power of the analysis, increasing the risk of type II error. Consequently, some observed differences may not have reached statistical significance despite potential clinical relevance. These findings should therefore be interpreted with caution, and further large-scale, prospective studies are warranted to validate our results. Moreover, this study did not include subgroup analyses based on driver gene mutations such as EGFR and ALK alterations, which are known to influence prognosis and treatment strategies in NSCLC. Due to the retrospective nature and limited availability of genetic data for all patients, these factors were not incorporated into the current analysis. Future prospective studies including comprehensive molecular profiling are warranted to elucidate the impact of genetic mutations on surgical outcomes in NSCLC patients with visceral pleural invasion. Additionally, further pathological reassessment incorporating genetic and molecular markers may provide deeper insights into tumor biology and prognosis. Conclusions In patients with VPI-positive lung cancer, segmentectomy may offer a favorable prognosis for those with tumors measuring ≤ 2.0 cm, while lobectomy may be more appropriate for tumors > 2.0 cm. Nevertheless, lobectomy should generally be considered the first-line approach for patients with tumors measuring > 2 cm, except in cases where segmentectomy is selected based on patient-specific risk factors. Caution should be taken when interpreting the segmentectomy outcomes in this subgroup due to potential selection bias. A balance between local tumor control and preservation of pulmonary function should be considered when selecting a surgical approach. Our results, particularly the stratified analysis by tumor size and recurrence pattern, may contribute to a more individualized and evidence-based surgical strategy for NSCLC patients with VPI. Therefore, future prospective, multicenter studies are warranted to validate our findings and to establish more definitive treatment strategies for VPI-positive NSCLC patients.