A nationwide survey of uniportal thoracoscopic anatomical pulmonary resections in Japan.

Introduction
Uniportal video-assisted thoracoscopic surgery (uVATS) was initially reported by Rocco et al. for wedge resection and has subsequently been applied to anatomical pulmonary resection since González-Rivas et al. reported their early experience (1,2). Since then, uVATS has rapidly spread worldwide and several single-center or multicenter studies have demonstrated its safety and feasibility (3-5). Compared to the multiportal approach, the uniportal approach is less invasive, as it requires fewer skin incisions. Al-Ameri et al. demonstrated that thoracoscopic lobectomy performed using the uniportal approach provides similar feasibility and safety to the multiportal approach, while additionally enabling more rapid postoperative recovery (3). Likewise, Chen et al. showed that, in patients undergoing thoracoscopic complex segmentectomy, the uniportal approach resulted in shorter duration of chest drainage and a reduced length of hospital stay, without any compromise in perioperative outcomes when compared with the multiportal approach (4). In our country as well, Matsuura et al. demonstrated that uVATS was associated with a shorter operative time, reduced duration of postoperative drainage, and a shorter length of hospital stay compared with conventional multiport VATS. Moreover, the significant reduction in perioperative analgesic use further supports the notion that uVATS is not only safe but also a less invasive surgical approach (5). However, despite the increasing number of reports, no nationwide data regarding the prevalence and clinical outcomes of anatomical pulmonary resection via uVATS have been identified in the literature. Thus, it remains unclear how widely this technique has been adopted in clinical practice.
To identify it, in the present study, we surveyed nationwide data on anatomical pulmonary resections including lobectomy and segmentectomy performed by uVATS in Japan. This multicenter, retrospective survey was conducted by the Japanese Uniportal Video-assisted Thoracoscopic Surgery Interest Group (JUVIG). We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1-2535/rc).

Methods
The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of the Japanese Red Cross Maebashi Hospital (approval No. 2025-22) and individual consent for this retrospective analysis was waived. The other participating hospitals were informed and agreed to the study.

Design, setting, and data collection
We performed a multi-institutional retrospective cohort study to evaluate the patient characteristics and perioperative outcomes of lobectomy and segmentectomy performed using a uVATS approach for primary lung cancer. In total, 43 institutions including 17 academic and 26 non-academic across the country participated in this study. The process of patient inclusion is summarized in Figure 1. Patients <18 years of age, those who underwent simultaneous bilateral surgery, previous ipsilateral anatomical lung resection, concurrent resection of ≥2 anatomical segments or lobes, or pneumonectomy were excluded. After excluding missing data, pulmonary metastasis, benign tumors, and other diseases, all cases with primary lung cancer were classified into the lobectomy and segmentectomy groups. This restriction was applied to ensure a relatively homogeneous clinical context for evaluating procedure-related safety and technical outcomes of uniportal anatomical lung resection, particularly with respect to the extent of lymphadenectomy and perioperative management. Trends in the annual proportion of lobectomies and segmentectomies were first assessed. Next, distributions of the institutional case volume were calculated. Finally, any patient characteristics and perioperative outcomes were evaluated in each group. The primary outcome was the incidence of procedure-related complications, and the secondary outcomes included other perioperative parameters.

Study variables
The following variables were extracted from the case report forms: study period, board certification status of the operating surgeon, age, sex, body mass index (BMI), smoking history expressed in pack-years, preoperative % forced expiratory volume in one second (%FEV1), clinical stage, resected lobe, extent of lymphadenectomy, operative time, intraoperative blood loss, duration of postoperative chest drainage, postoperative length of stay, conversion to a multiport approach, conversion to thoracotomy, postoperative morbidity (defined as Clavien-Dindo grade ≥ III) (6), hospital readmission within postoperative 30 days, 30-day postoperative mortality, 90-day postoperative mortality, and surgical procedure-related complications (7). The extent of lymphadenectomy was classified into two categories: ND0–1 and ND2. ND2 was defined as either a systemic or lobe-specific mediastinal lymph node dissection, whereas ND0–1 included no lymphadenectomy or hilar lymphadenectomy only, without mediastinal lymphadenectomy. In Japan, board certification in general thoracic surgery is granted by the Japanese Association for Chest Surgery after completion of a structured training program, which includes several years of society membership, a minimum required number of thoracic surgical cases, and successful completion of a written examination. Board-certified surgeons are therefore considered fully trained specialists rather than trainees.
In the present study, surgical procedure-related complications were defined as perioperative adverse events directly attributable to the technical aspects of anatomical lung resection. For analytical clarity, these complications were classified in advance into three categories.
The first category consisted of intraoperative complications, including critical events occurring during surgery, such as significant vascular injury, unintended division of the pulmonary vein, and bronchial injury. Significant vessels were defined as segmental or more proximal branches of the pulmonary artery, pulmonary veins proximal to the segmental branches, as well as major vessels such as the aorta, subclavian artery, superior vena cava, brachiocephalic vein, subclavian vein, and azygos vein. These events resulted in interruption of the procedure and typically required intraoperative modification or urgent intervention.
The second category comprised technical postoperative complications, which were postoperative events considered to be directly related to intraoperative technique. These included prolonged air leak (PAL) defined as the persistence of air leakage for ≥5 postoperative days, delayed pulmonary pneumothorax requiring re-drainage, excessive pleural effusion defined as pleural effusion requiring chest drainage for more than 5 days postoperatively or re-drainage after initial chest tube removal, recurrent laryngeal or phrenic nerve palsy, delayed hemothorax, empyema, chylothorax, bronchial fistula, wound infection, and other similar complications. These events reflected technical challenges in procedural execution or tissue handling.
The third category consisted of reintervention within 24 hours after surgery, defined as emergent re-thoracotomy required during the immediate postoperative period. These included reoperations performed for severe complications such as intrathoracic bleeding, uncontrolled air leakage, or wound bleeding.
This categorization enabled an objective evaluation of technical performance and safety associated with uVATS. In cases where ≥4 surgical procedure-related complications were observed, the three most relevant complications were recorded for analysis, based on the operating surgeon’s judgment, with priority given to complications considered to have the greatest clinical impact, such as those requiring additional intervention or associated with major morbidity.

Definitions
The uVATS approach was defined as a procedure in which all surgical instruments and the thoracoscope were introduced concurrently through a single skin incision of ≤4 cm in length (8). In all cases, no metal rib-spreader was used, and the uniportal thoracoscopic approach was performed without rib spreading, using soft tissue retraction only. All procedures were performed exclusively under monitor vision without direct visualization of the operative field. The choice of intercostal space for the incision, as well as the specific surgical techniques employed—including the methods used to expose and divide the pulmonary vessels, bronchi, and lung parenchyma—was left to the discretion of the operating surgeon and was not restricted by protocol.

Statistical analysis
Continuous variables were expressed as medians with interquartile ranges (IQR), whereas categorical variables were expressed as absolute numbers and corresponding percentages. All statistical analyses were conducted using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R.

Results
Overall, 4,026 case report forms for patients who underwent uVATS major pulmonary resection between April 2018 and March 2023 were obtained from the participating centers. After excluding 57 patients with missing data, 259 cases of pulmonary metastasis, 149 benign tumors, and 15 other cases, 3,546 patients with primary lung cancer were included in the final analysis. Of these, 2,780 comprised the lobectomy group and 766 comprised the segmentectomy group.
Over the study period, both the total number of uVATS anatomical lung resections and the proportion of segmentectomies progressively increased, as illustrated in Figure 2. This temporal change likely reflects a growing preference for segmentectomy, possibly driven by accumulating evidence demonstrating its oncologic validity and by refinements in surgical techniques. In addition, the distribution of institutional case volume is shown in Figure 3.
Patient characteristics and perioperative outcomes in the lobectomy group are summarized in Table 1. In the lobectomy group, 89.7% of procedures were performed by board-certified surgeons. The median age was 72 years, with 64.6% younger than 75 years. Females accounted for 43.7% of patients. The median BMI was 22.8 kg/m2 (IQR, 20.7 to 25.2 kg/m2). Regarding smoking history, the median smoking index was 21 pack-years (IQR, 0 and 47 pack-years). Preoperative pulmonary function was well preserved in most patients, with a median %FEV1 of 91.6% (range, 78.0 to 107.1%). Systematic nodal dissection (ND2) was performed in 86.2% of cases. The median operative time was 170 min, and median blood loss was 20 g. Median postoperative drainage lasted 2 days, and the median hospital stay was 7 days. Conversion to multiport occurred in 2.6% and to thoracotomy in 3.6%. Among lobectomy cases requiring conversion to thoracotomy (n=99), the most common cause was bleeding (n=38), followed by inflammatory lymph node invasion (n=23), severe pleural adhesion (n=16), bronchial injury (n=6), and other reasons (n=16). Postoperative morbidity of Clavien-Dindo grade ≥3 was observed in 9.8%. Readmission within 30 days occurred in 2.6%, and 30- and 90-day mortality rates were 0.3% and 0.5%, respectively. Surgical procedure-related complications occurred in 15.1% of patients in the lobectomy group. A detailed breakdown of these complications is presented in Table 2. Intraoperative complications were documented in 4.0%, mainly significant vessel injury (3.4%). Among patients who experienced significant vascular injury, conversion to thoracotomy was required in 38 of 94 cases (40.4%) in the lobectomy group, while the remaining 56 cases (59.6%) were managed thoracoscopically. Technical postoperative complications were recorded in 11.7%, with PAL being the most common (7.6%). Reintervention within 24 hours occurred in 0.4%, primarily due to thoracic bleeding or persistent air leaks.
Patient characteristics and perioperative outcomes in the segmentectomy group are summarized in Table 3. In the segmentectomy group, 90.3% of procedures were performed by board-certified surgeons. The median age was 72 years, with 61.9% younger than 75 years. Females represented 52.2% of patients. The median BMI was 22.6 kg/m2 (IQR, 20.4 to 25.0 kg/m2). The median smoking index was 12 pack-years (IQR, 0 to 43 pack-years), and pulmonary function was preserved, with a median preoperative %FEV1 of 91.5% (IQR, 76.2–107.2%). The most frequently treated site was the left upper lobe (43.3%), followed by the right lower lobe (22.5%). Limited nodal dissection (ND0–1) was performed in 79.8% of cases. The median operative time was 154 min, and median blood loss was 5 g. Postoperative drainage lasted a median of 1 day, and the median hospital stay was 6 days. Conversion to multiport occurred in 1.8% and to thoracotomy in 2.4%. Among segmentectomy cases requiring conversion (n=18), bleeding was also the most frequent cause (n=8), followed by inflammatory lymph node invasion (n=3), severe pleural adhesion (n=3), bronchial injury (n=1), and other reasons (n=3). Postoperative morbidity of Clavien-Dindo grade ≥3 was reported in 5.6%. Readmission within 30 days occurred in 1.2%. Thirty- and 90-day mortality rates were 0.3% and 0.4%, respectively. Surgical procedure-related complications occurred in 10.4% of patients in the segmentectomy group. A detailed breakdown of these complications is presented in Table 4. Intraoperative complications were observed in 3.8%, mainly vessel injury (3.5%). In the segmentectomy group, conversion to thoracotomy was required in 8 of 27 cases (29.6%), whereas 19 cases (70.4%) were managed without conversion. Technical postoperative complications were seen in 6.9%, with PAL being the most frequent (4.6%). Only 1 case (0.1%) required reintervention within 24 hours.
To further explore potential differences between lobectomy and segmentectomy, an exploratory propensity score-matched analysis was conducted. After matching, baseline patient characteristics and perioperative variables were well balanced between the two groups (Table S1). Comparisons of perioperative outcomes after matching are presented in Table S1, while surgical procedure-related complications are summarized in Table S2. Given the exploratory nature of this analysis and the inherent differences in indications and anatomical complexity between the two procedures, the results should be interpreted cautiously and are not intended to establish comparative effectiveness. In addition, causes of readmission within 30 days after the operation are summarized in Table S3. The most common cause of readmission in the lobectomy group was PAL, whereas empyema was the most frequent reason for readmission in the segmentectomy group.

Discussion
This study was primarily conducted by JUVIG, with participation from high-volume centers in Japan that had performed a large number of uVATS procedures, thereby reflecting the current status of its implementation. Although several reports have described patient characteristics and perioperative outcomes of uVATS anatomical pulmonary resections, to the best of our knowledge, this is the first study to present nationwide data.
Although uniportal anatomical lung resection can also be applied to pulmonary metastases or benign lesions, these entities often involve different surgical indications, extents of lymphadenectomy, and perioperative risk profiles compared with primary lung cancer. Therefore, such cases were intentionally excluded to avoid clinical heterogeneity that could confound the assessment of procedure-related complications and technical safety. By restricting the study population to patients with primary lung cancer, we aimed to evaluate the safety and feasibility of uVATS anatomical resection under relatively uniform surgical and perioperative conditions.
According to the 2023 annual report of the Japanese Association for Thoracic Surgery, 693 institutions are accredited for thoracic surgery in Japan (9). In 2023, a total of 28,371 lobectomies were performed nationwide, including 18,403 thoracoscopic (64.9%) and 5,256 robotic (18.5%) procedures. Similarly, among 9,427 segmentectomies performed during the same year, 7,190 (76.3%) were thoracoscopic and 1,596 (16.9%) were robotic. In the present study, data were collected from 43 institutions (6.2%). The total numbers of thoracoscopic lobectomies and segmentectomies performed by these participating centers in fiscal year 2022 were 785 and 288, respectively, corresponding to approximately 2.8% and 3.1% of all national cases, and 4.3% and 4.0% of the national thoracoscopic cases, respectively. This study was conducted mainly among institutions belonging to JUVIG. As most high-volume centers performing uniportal thoracoscopic surgery in Japan are JUVIG members, this cohort is considered to include nearly all uniportal anatomical pulmonary resections performed nationwide, except for a small number of cases.
In this study, the rates of significant vessel injury in uVATS were 3.4% for lobectomy and 3.5% for segmentectomy, which were comparable to those reported in previous studies involving cases performed exclusively with multiport approaches (10-12). In general thoracic surgery, avoiding intraoperative major vessel injury is critical to ensuring operative safety. Although the overall incidence of vascular injury was not negligible, a substantial proportion of such events could be managed without conversion to thoracotomy, underscoring the importance of surgical experience and intraoperative judgment in real-world practice. Taken together, these findings suggest that uVATS anatomical lung resection can be performed with acceptable perioperative safety when applied by experienced teams, while recognizing the inherent technical challenges of this approach. In this context, the causes of conversion to thoracotomy—predominantly bleeding, inflammatory lymph node invasion, and severe adhesions—highlight the practical limitations and technical challenges encountered during uVATS in real-world practice. These observations further emphasize the importance of careful patient selection and intraoperative decision-making when performing complex anatomical lung resections via a uniportal approach.
Operative time and the incidence of PAL are commonly used surrogate markers of surgical performance, particularly in the context of learning curve assessment. Previous studies have often used reductions in operative time as a surrogate marker for the learning curve (13). PAL, in contrast, is one of the most common complications following pulmonary resection, and its prevention not only facilitates early mobilization but also reduces the risk of secondary complications such as pneumonia (14). According to earlier reports, operative times for uVATS lobectomy have been described as 130–160 min, with PAL rates of 2.9–13%, whereas uVATS segmentectomy has been reported with operative times of 106–149 min and PAL rates of 2.2–12.7% (13,15-19). Although the operative times in our study were slightly longer than those reported previously, this may reflect the inclusion of multiple institutions with varying case volumes and levels of surgical experience, thereby representing real-world practice in Japan. These results are within the range of those documented in previous studies. Although this was not a direct comparative analysis, our findings suggest that uVATS anatomical lung resections are being performed in Japan with acceptable perioperative outcomes in real-world practice.
The observed discrepancy between the relatively short duration of postoperative chest drainage and the longer length of hospital stay may be explained by several factors. First, criteria for chest tube removal and hospital discharge were not standardized across participating institutions, reflecting real-world practice in a nationwide multicenter setting. In addition, the healthcare system in Japan, where prolonged hospitalization does not necessarily impose substantial additional financial burden on patients, may influence postoperative management. As a result, patients often prefer to remain hospitalized until full recovery, rather than being discharged immediately after chest tube removal. This practice may contribute to a longer postoperative hospital stay compared with reports from other countries, despite early removal of chest drains.
Importantly, following the JCOG0802 and CALGB140503 trials, which demonstrated that segmentectomy is non-inferior to lobectomy, the proportion of segmentectomies has steadily increased across various surgical approaches, including uniportal and multiportal thoracoscopy, mini-thoracotomy, and open thoracotomy (20,21). A similar trend was observed in our nationwide cohort, where the rate of segmentectomy gradually increased over the study period, even though the cohort included not only early-stage non-small cell lung cancer but also advanced tumors. This trend is most likely attributable to the growing influence of recent clinical trial evidence. Although this temporal change was not the central focus of our study, it nevertheless represents an important evolution in surgical practice and warrants further investigation in future research. In addition to randomized trials demonstrating the oncological non-inferiority of segmentectomy, the final results of the JCOG1710A prospective cohort study further supported the feasibility and safety of anatomical segmentectomy, including technically demanding procedures, when performed under standardized surgical protocols in experienced centers (22). These findings may have further accelerated the adoption of segmentectomy in Japan by alleviating concerns regarding technical complexity and perioperative safety.
Although a direct comparison between lobectomy and segmentectomy was not performed in the present study, the rate of PAL in the segmentectomy group (4.6%) was lower than that in the lobectomy group (7.6%). This observation should be interpreted with caution. Although residual lung parenchyma covering the divided surface after segmentectomy may partly contribute to postoperative air leak control, the present analysis did not adjust for patient-, tumor-, or anatomy-related factors (23,24). In real-world practice, segmentectomy may be preferentially selected for peripheral or technically less complex tumors, whereas lobectomy may more frequently be performed for central or anatomically complex cases, including those with fused fissures. Such selection bias could have influenced the observed difference in PAL rates between the two procedures. In addition, the proportion of segmentectomy gradually increased over the study period, suggesting that segmentectomy was often performed after surgeons had gained experience with the uniportal approach through lobectomy, which may also have affected perioperative outcomes. In contrast, the JCOG0802 trial reported a higher incidence of PAL in the segmentectomy group (25). This discrepancy may be attributed to differences in patient selection, study periods, the proportion of complex segmentectomies, and variations in techniques for intersegmental plane division, although these factors were not clearly described.
Finally, an exploratory propensity score-matched analysis was performed as a supplementary assessment; however, given the inherent differences in indications and anatomical complexity between lobectomy and segmentectomy, these results should be interpreted with caution and were not intended to establish comparative effectiveness.

Limitations
This study has several limitations. First, although it was a multicenter study with a large sample size, its retrospective design may have introduced selection bias. Second, while the study provides a comprehensive overview of the current status of uVATS anatomical lung resections in Japan, it does not include a direct comparison with other approaches, such as multiportal VATS or thoracotomy. Consequently, the relative advantages of uVATS over alternative surgical techniques cannot be assessed. Accordingly, no conclusions regarding the superiority or non-inferiority of uVATS compared with other surgical approaches can be drawn from the present analysis. Although this study focused exclusively on patients with primary lung cancer, detailed oncological data such as the number of dissected lymph node stations, the total number of harvested lymph nodes, the completeness of resection, long-term oncological outcomes, including overall and recurrence-free survival were not collected in the present dataset. This is because the primary aim of this nationwide survey was to evaluate the real-world implementation and procedure-related safety of uniportal thoracoscopic anatomical pulmonary resection under relatively uniform surgical conditions, rather than its oncological efficacy. Therefore, the oncological quality of the procedure could not be fully assessed, and further studies incorporating these parameters are warranted to validate the oncological efficacy of uniportal thoracoscopic anatomical pulmonary resection. In addition, detailed information regarding specific hemostatic techniques used to manage vascular injuries, such as suturing or sealant application, was not available in the present dataset and therefore could not be analyzed. Because this study was retrospective and relied on voluntary data submission from participating centers, potential selection bias related to institutional characteristics and case selection may exist. Therefore, the results should be interpreted with caution regarding generalizability to all institutions nationwide.

Conclusions
Our nationwide analysis indicates that uVATS anatomical pulmonary resection is being performed safely in Japan, with perioperative outcomes that are both acceptable and consistent with international standards.

Supplementary
The article’s supplementary files as