Perioperative surgical outcomes and learning curves of thoracoscopic basal segmentectomy.

Introduction
Low-dose computed tomography (CT) has been widely adopted in clinical practice, significantly improving the detection rate of early-stage non-small cell lung cancer (NSCLC) (1-3). Lung cancer remains one of the most prevalent and deadly diseases worldwide and is the leading cause of cancer-related death (3,4). While lobectomy has long been the standard treatment for early-stage lung cancer, anatomic segmentectomy is increasingly being performed with comparable long-term outcomes, particularly for nonsolid tumors with ground-glass opacity (GGO) (5-7). Recent research from the Japan Clinical Oncology Group (JCOG) 1211 trial evaluated the safety of segmentectomy for NSCLC tumors up to 2 cm in size, demonstrating that segmentectomy should be considered a standard treatment option for patients with predominantly GGO NSCLC, of which the tumor size is 2 cm or less (8).
Compared with lobectomy, pulmonary segmentectomy is more technically challenging because of the need for precise dissection of anatomical structures (9). Basal segmentectomy is particularly difficult because of the deep location of the basal segments and the frequent anatomical variations (10). The key to successful segmentectomy for NSCLC lies in the ability to intraoperatively identify both the tumor and the intersegmental plane (11). Segmentectomies involving S9, S10, or both are particularly challenging due to the complexity of the anatomy in these areas (12). These types of pulmonary resections, which require the creation of multiple intersegmental planes, are generally referred to as complex pulmonary resections. They are associated with higher rates of morbidity and mortality, particularly due to prolonged air leaks (PALs) resulting from the construction of several intersegmental planes. Other concerns include longer hospital stays and extended periods of drainage. Additionally, postoperative recovery of pulmonary function may be impaired, as altered lung structure can hinder the expansion of the remaining lung tissue (13).
In this study, we assess the feasibility of basal segment resection in the surgical treatment of early-stage NSCLC by analyzing the perioperative outcomes of both complex and simple basal segment resections. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1652/rc).

Methods

Patients selection
We retrospectively reviewed the medical records of 190 patients who underwent anatomical basal segmentectomy at China-Japan Friendship Hospital between January 2020 and July 2024 (Figure 1). Among these 190 patients, 66 underwent single basal segmentectomy, 22 underwent combined basal segmentectomy, 10 underwent single basal subsegmentectomy, and 5 underwent combined basal subsegmentectomy. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of China-Japan Friendship Hospital (No. 2023-KY-151) and individual consent for this analysis was waived due to the retrospective nature.
The surgical margin was set to exceed the maximum diameter of the tumor. The specimens were sent for intraoperative frozen section pathology to verify whether the resection margins were adequate. All cases of adenocarcinoma were staged and histologically classified via the 8th edition of the tumor-node-metastasis (TNM) staging system (14).
The inclusion criteria were as follows: (I) chest high-resolution CT suggesting the possibility of early-stage lung cancer; (II) adequate pulmonary function to tolerate segmentectomy; (III) no evidence of distant metastasis on preoperative examination; (IV) thoracoscopic basal segment resection or basal subsegmentectomy performed in our department; and (V) pulmonary nodules with a diameter of ≤2 cm.
The exclusion criteria were as follows: (I) small cell lung cancer; (II) pulmonary segment resection of other areas, including full basal segmentectomy; and (III) patients who underwent neoadjuvant therapy.
The objective of this study was to identify factors influencing the duration of basal segmentectomy, surgical costs, postoperative drainage duration, and length of hospital stay, providing practical insights and reference value for clinical practice.

Data collection
There are differences between males and females in terms of pulmonary anatomical hormone levels (15) and comorbidities, which may lead to variations in surgical difficulty and postoperative recovery rates. Elderly patients exhibit reduced pulmonary functional reserve and impaired tissue healing capacity, potentially resulting in prolonged surgical duration and other factors. Thus, we summarized the clinical and perioperative data of 190 patients, including sex, age, nodule location, total hospitalization costs, nodule size on CT, nodule composition, actual nodule size, surgical approach (single-port vs. multi-port), surgical route (venous, arterial, or mixed), surgical procedure (based on actual excision), postoperative pathology, the number and sites of lymph node metastases, intraoperative blood loss, duration of postoperative drainage, length of hospital stay, 30- and 90-day perioperative mortality, survival data, tumor staging and grading, and other relevant factors. Intraoperative blood loss was assessed by the volume in the collecting bottle connected to the negative pressure suction device. The criteria for removing thoracic drainage tubes were based on ensuring that no air leakage occurred, which was confirmed by instructing the patient to cough. If no bubbles appeared in the drainage bottle, it was considered free from air leakage. Factors influencing surgical time, postoperative drainage duration, length of hospital stay, and total hospitalization costs were analyzed.

Cumulative sum (CUSUM) analysis
The CUSUM method was used for the analysis of the learning curve (16,17). The learning curve was assessed via both the operating time (OT) and the cumulative sum of the operating time (CUSUMOT) for all the cases. We assessed the best fit for the plot and detected the change in the slope of the CUSUM learning curve. The number of required cases was calculated from the inflection point of the curve that represented the best fit for the plot.

Statistical analyses
Continuous variables with a normal distribution are presented as the means ± standard deviations. Variables with a nonnormal distribution are expressed as medians [interquartile ranges (IQRs)]. Categorical and count data are presented as frequencies and percentages. Continuous variables use Spearman analyzation to analyze correlation, and classified variables use non-parametric Mann-Whitney U test. A P value of <0.05 was considered to indicate statistical significance. CUSUM analysis was employed to analyze the learning curves of anatomical basal segmentectomy.

Operative procedure
Preoperative evaluations typically include a physical examination; cardiopulmonary function tests; blood tests; and imaging studies of the brain, upper abdomen, and bones. The locations of the target nodules, as well as the adjacent structures, anatomical variations, and positional relationships of the bronchi and blood vessels in the basal segment, were carefully assessed via high-resolution computed tomography (HRCT) to plan the appropriate surgical resection.
We collapsed the lung on the operative side and anesthetized the patient via differential ventilation. Either a uniport or multiport approach was used for port access. The surgical techniques employed included the following:
❖ Arterial approach: dissection was carried out through the pulmonary fissure, with segmental arteries, bronchi, and veins dissected sequentially from top to bottom before being individually severed.

❖ Venous approach: dissection starts through the lower pulmonary vein, first severing the veins supplying the segmental area, followed by freeing the bronchus and artery from the deep aspect of the vein before being cut separately.

❖ Mixed approach: dissection begins through the pulmonary fissure to expose and sever the artery, followed by dissection and division of segmental veins through the lower pulmonary vein and, finally, severing the segmental bronchi. Alternatively, dissection may start through the lower pulmonary vein to expose and sever the segmental veins first, followed by sequential dissection and division of the arteries and bronchi within the pulmonary fissure.

The resection margin was required to exceed the diameter of the nodule. If achieving an adequate margin with single segment resection was not feasible, extended segmentectomy or combined segmentectomy with adjacent subsegments was performed. Lymph node sampling was performed instead of dissection when lung cancer patients presented with pure ground-glass nodules (GGNs). Mediastinal lymph node dissection was carried out for solid tumors and tumors with partially solid GGNs.

Results

Clinical and perioperative characteristics
We included records of 190 patients who underwent thoracoscopic anatomical basilar segmentectomy. The mean age of the patients was 54.65 years, with 30.0% men and 70.0% women. Among these cases, 106 (55.79%) involved single basal segmentectomy, 44 (23.16%) involved combined basal segmentectomy, 18 involved segmentectomy combined with subsegmentectomy, 15 (7.89%) involved single basal subsegmentectomy, and 7 (3.68%) involved combined basal subsegmentectomy. The average duration of chest drainage was 3.24 days, and the mean postoperative hospitalization time was 4.24 days. Pathological examination confirmed the following diagnoses: 17 cases of adenocarcinoma (AIS), 83 cases of minimally invasive adenocarcinoma (MIA), 73 cases of invasive adenocarcinoma (IAC), 1 case of pulmonary metastasis from thyroid cancer, and 16 cases of benign lesions. Conversion to thoracotomy occurred in one patient (0.53%, 1/190). We defined postoperative tube retention for 5 days or more as PAL (18). There are a total of 38 cases (2%) of postoperative complications, including 31 cases of PAL (16.3%), 2 cases of atrial fibrillation (0.01%), one case each of lung infection, pulmonary embolism, cerebral infarction, subcutaneous emphysema and coeliac chest. There were no cases of perioperative mortality, and no patients experienced local recurrence. The postoperative pathological margins were negative. No 30-day or 90-day perioperative mortality was observed in either group. Details of the segmentectomy procedures are provided in Table 1.

Outcome variables
To identify risk factors for prolonged surgical time during basal segmentectomy, we treated surgical time as dependent variables, the independent variables included sex, age, nodule size, CT location (external/central/internal), nodule laterality (left/right), nodule composition (GGN/non-GGN), lung fissure development (complete/incomplete), surgical approach (via fissure/nonfissure approach), surgical entry (single-port/multiport), intraoperative blood loss, and postoperative pathology (AIS/MIA/IAC/benign lesions and others). Given the complexity of surgical procedure, studies have shown that segmentectomy of the S9/S10 segments is relatively more challenging than other lung segment resections are (19). Therefore, we categorized the surgical procedure into two groups: those including S9/S10 resections and those excluding them. The dependent variable was analyzed via Spearman analysis and Mann-Whitney U test (Table 2), which revealed that the surgical approach (P=0.001), surgical procedure (P=0.01), and intraoperative blood loss (P<0.001) significantly influenced surgical time. Variables with P<0.05 in the univariate analysis were included in the multiple linear analysis (Table 3). This revealed that surgical approach (<0.001), surgical procedure (P=0.04) and intraoperative blood loss (P<0.001) were significant predictors of surgical time.
We use the same method to analyze the total cost of hospitalization (Table 4). The results revealed that age (P=0.003), surgical procedure (P<0.001) and postoperative pathology (P=0.02) significantly affected total hospitalization costs. Variables with a P value <0.5 in the multiple linear analysis were included in the multivariate analysis (Table 5), which confirmed that the surgical procedure (P=0.01) significantly influenced the total hospitalization costs. Specifically, when segment resection included S9 or S10, the total hospitalization costs were significantly greater.
In terms of surgical method, the hospitalization cost for segmentectomy that included S9/S10 resection was 10,946±1,771 dollars, whereas that for segmentectomy excluding S9/S10 resection was 10,134±2,751 dollars (P=0.03).
To identify factors contributing to PAL and prolonged extubation time following basal segmentectomy (Table 6). We found that lung fissure development (P=0.002) significantly affected the postoperative tube retention time when extubation occurred after 5 days or more. These variables were then included in multiple linear analysis, which revealed that incomplete lung fissure development (P=0.002) significantly increased the likelihood of postoperative pulmonary air leakage. According to the classification of lung fissure development, the postoperative tube retention time for well-developed lung fissures was 3.09±1.18 days, whereas for incomplete lung fissure development, it was 3.94±1.56 days (P<0.001).
As the number of cases increased, a downward trend in OT (to make the chart more intuitive, the OT is expressed in minutes.) was observed (Figure 2). The CUSUMOT learning curve was modeled. The CUSUM analysis revealed that the curve entered a stable phase after the case 119. The average operation time before and after this case is respectively 138.1±47.7 and 119.1±39.4 min (P=0.005), and there exists significant difference.

Discussion
Over the past decade, increasing research has demonstrated that there is no significant difference in overall survival between sublobar resection and lobectomy in patients with a tumor diameter ≤2 cm. For example, JCOG0802, JCOG0804, JCOG1211, and Cancer and Leukemia Group B (CALGB)140503 reported that sublobar resection or segmentectomy in patients with T1N0 stage lung cancer yields similar oncological outcomes to lobectomy for tumors ≤2 cm in size (8,20-22). While segmentectomy has been widely studied, only a few studies with small sample sizes have specifically focused on thoracoscopic basal segmentectomy.
Dai et al. (19) compared the perioperative outcomes and oncologic prognoses of combined versus simple basal segment resection performed via single-port thoracoscopy versus multiport thoracoscopy. The results indicated that the surgical time for combined basal segment resection was significantly longer than that for simple basal segment resection. Previous studies have highlighted that S9/S10 segment resection is one of the most challenging thoracoscopic lung segmentectomy procedures (23), with significantly longer surgical times than simple segment resection (13,24-26). This study considered various factors, and the results revealed that surgical procedure (i.e., whether S9/S10 was involved) significantly influenced surgical time (P=0.038). Specifically, segmentectomy, including S9 or S10, required significantly more time than did procedures excluding these segments. The reason may be that the S9–10 segments are typically more variable in vascular and bronchial anatomy, deeper in location, and more challenging to identify and handle (12). Resection of these segments often involves multiple intersegmental planes (13), increasing the risk of complications and contributing to longer surgical times.
We found that the operation time of single-hole thoracoscopy is shorter than that of multi-hole thoracoscopy (P<0.001), and calculated that the average operation time of single-hole and multi-hole is 1.92 and 2.29 h respectively. We analyzed that this may be due to the fact that in single-hole thoracoscopy, the lens is directly aimed at the surgical target area, and the instrument and the lens are at the same mouth. Enter the surgical field in parallel and operate in the vertical plane. The surgical field and the surgical operation are similar to conventional thoracotomy, which may be one of the reasons for the short time of single-hole thoracoscopic surgery, and more experienced surgeons were assigned to execute the procedures in the single-port video-assisted thoracic surgery (VATS) group, whereas less experienced surgeons were assigned to perform the procedures in the 2 or multi-port VATS (27). In addition, when there is more bleeding during the operation, the operation time will also be significantly prolonged (P<0.001), bleeding leads to repeated steps such as suction, hemostasis and exposure reconstruction, thus blocking the pace of promoting the operation in the conventional anatomical order, and if necessary, incision increase or open conversion will also occur. This view is consistent with the overview of thoracic bleeding management and its impact on surgery time (28).
This study reported 31 cases of PAL, accounting for 16.3% of the total study population. Since PAL was the main postoperative complication observed, we used the number of postoperative days under drainage as a proxy for postoperative complications. However, the absence of formal Clavien-Dindo grading is a limitation of our retrospective analysis, and future studies should apply standardized classification systems for better comparability. Previous studies have reported no significant difference in the occurrence of postoperative complications between single-port and multiport lung segment or lobectomy procedures (29). Takamori et al. reported that there was no significant difference in the duration of chest tube retention (30), and other studies have suggested that surgeons are increasingly favoring noninterlobar fissure approaches, which have led to a significant reduction in the median postoperative hospitalization time compared with the interlobar fissure approach (31). On this basis, we also included an analysis of the factors affecting the postoperative drainage time to determine whether there was preoperative emphysema, although the results were not significant. Our findings indicate that the development of pulmonary fissures significantly impacts the postoperative duration of drainage tube placement and overall hospitalization. In cases of incomplete pulmonary fissure development, sharp dissection may damage the surface of the lung tissue, which can prolong the duration of drainage tube placement.
This study revealed that as the number of cases increased, the OT tended to decrease, which is consistent with findings from previous studies (31). As the number of cases increased, the learning curve tended to gradually stabilize, indicating that with the accumulation of experience, the variability in OT decreased. Specifically, the CUSUMOT values in the early stages were relatively high, suggesting that in the initial phase of learning, the OT was longer, possibly due to lower technical proficiency and familiarity with the procedure. With the accumulation of surgical experience, the CUSUMOT values progressively decreased, indicating improved technical skills and a reduction in OT.
There are several limitations in this study. First, it is a retrospective analysis, which is subject to selection bias that cannot be fully eliminated. Additionally, since the surgeries were performed by different surgical teams, there may be variability in surgical practices or preferences. Second, the follow-up period in this study was relatively short, with the longest duration not exceeding 1 year. As a result, long-term prognostic outcomes, including recurrence, metastasis, and mortality, were not addressed in this study. We plan to conduct further research with extended follow-up periods to assess survival outcomes over the long term. The CUSUM of the operation time is for the overall team of surgeons, not the individual, so it can only reflect the overall proficiency rather than the individual.

Conclusions
In summary, our research suggests that the perioperative outcomes of thoracoscopic basal segmentectomy are influenced by various factors, with significant increases in operative time and total costs when segments containing S9/S10 segments are resected. However, regardless of the surgical approach used, no significant differences in short-term prognosis were observed. The primary postoperative complication was pulmonary air leakage, which was associated mainly with incomplete lung fissure development.

Supplementary
The article’s supplementary files as