Abstract

Background: This study aim to evaluate the feasibility of threedimensional (3D) modeling using open-source free software for preoperative planning and intraoperative guidance in uniportal video-assisted thoracoscopic segmentectomies.

Methods: Between October 2020 and November 2023, 50 patients (27 males, 23 females; mean age: 60±11.1 years; range, 34 to 78 years) who underwent uniportal video-assisted thoracoscopic pulmonary segmentectomy with preoperative 3D modeling were retrospectively analyzed. Preoperative 3D modeling was performed using computed tomography with an open-source 3D software program. The virtual models exported to the mobile device were compared with the anatomical structures of the patient intraoperatively. The patients were divided into two groups as simple and complex segmentectomy according to the characteristics of the surgical procedures.

Results: The overall matching success rate of the virtual 3D models with intraoperatively identified bronchovascular structures was 99.27%. The overall variation rate was 36% (n=18) among all patients. There was a significant difference between the two groups in terms of the bronchovascular variation. The bronchovascular variation rate was 11.1% (n=2) in Group 1 and 50% (n=16) in Group 2 (p=0.006).

Conclusion: Three-dimensional modeling using open-source software for preoperative planning and intraoperative guidance is a reliable method for detecting bronchovascular structures of the target segment with high accuracy in uniportal video-assisted thoracoscopic surgery segmentectomy.

Lung-sparing surgical techniques have gained increasing importance with the rise in early-stage lung cancer diagnoses and improvements in radiological detection. In this evolving landscape, segmentectomy is becoming more than just an alternative; it's emerging as a key strategy in selected patients.

Recent studies by the Japanese Clinical Oncology Group (JCOG0802)[1] a nd A ltorki e t a l.[2] (Alliance/CALGB140503) demonstrated the benefits of sublobar resections compared to lobectomy, suggesting that the segmentectomy procedure particularly may become the standard surgical procedure in terms of overall survival rate in patients with clinical Stage 1A lung cancer. Therefore, thoracic surgeons are increasingly interested in minimally invasive segmentectomy procedures. However, segmentectomy is a more complex surgical procedure than standard lobectomy due to its anatomical complexity and variations at segmental levels, including the bronchovascular structures.[3] G ossot et a l.[4] reported that to overcome the complexity of video-assisted thoracoscopic surgery (VATS) segmentectomy, the surgeon would not only need to master the techniques, but would also need to rely on various technologies such as preoperative three-dimensional (3D) modeling and/or 3D printing.[4]

Preoperative 3D modeling and 3D printing are also widely used in other surgical specialties (interventional neuroradiology, otorhinolaryngology, orthopedics, etc.).[5, 6] Preoperative 3D modeling technology plays a critical role in lung resection, particularly in segmentectomy. As there are many variations in the pulmonary arteries and veins, a detailed preoperative understanding of the patient's specific anatomy can contribute to safe pulmonary surgery and prevent complications.[3] M any s oftware programs have been developed for 3D modeling. The goal of all software is to preoperatively visualize the anatomical structures of the lung for surgeons using two-dimensional (2D) computed tomography (CT) images to improve the safety and accuracy of VATS segmentectomy.[7, 8] Nevertheless, professional skill requirements and high time consumption hinder the widespread use of these systems in clinical practice.[9] AI-based software automatically learns from raw data and rapidly completes the 3D model to improve the clinical efficiency.[10] Unfortunately, owing to the high cost of AI-based software in Türkiye, open-source free software continues to be used to perform segmentectomy as intended. In the present study, we aimed to evaluate the feasibility of 3D modeling using open-source free software for preoperative planning and intraoperative guidance in uniportal VATS segmentectomy, a complex procedure.

Methods

This single-center, retrospective study was conducted at Ondokuz Mayıs University Faculty of Medicine, Department of Thoracic Surgery between October 2020 and November 2023. A total of 196 consecutive patients who underwent anatomical lung resection using uniportal VATS were screened. Among them, 50 patients (27 males, 23 females; mean age: 60±11.1 years; range, 34 to 78 years) who underwent uniportal VATS segmentectomy were included in the study. All segmentectomies were performed by a surgeon with experience in more than 300 cases of pulmonary anatomical resection. Inclusion criteria were as follows: patients with peripheral nodules less than 2 cm in size, no calcific lymph nodes, respiratory function unsuitable for lobectomy, benign pathologies (aspergilloma, bronchiectasis, etc.), and centrally located secondary lung malignancies who underwent uniportal VATS segmentectomy. Patients who could not obtain CT of the thorax with an intravenous contrast agent due to impaired renal function were excluded from the study. A written informed consent was obtained from each patient. The study protocol was approved by the Ondokuz Mayis University Clinical Research Ethics Committee (date: 25.04.2024, no: KAEK 2024/185). The study was conducted in accordance with the principles of the Declaration of Helsinki.

The patients were divided into two groups: simple and complex segmentectomy, according to the characteristics of the surgical procedures. Group 1 (n=18): Simple segmentectomy includes right/left S6, left S1+2+3, and left S4+5 and is completed by creating only one linear intersegmental plane. Group 2 (n=32): Resections other than simple segmentectomy, which are performed by creating more than one intersegmental plane and are called complex segmentectomies.[11] The flowchart of the study is shown in Figure 1

Figure 1: Study flowchart. VATS: Video-assisted thoracoscopic surgery.

3D modelling and preoperative planning

All patients underwent preoperative CT imaging using a 64-detector MDCT scanner (GE Healthcare Discovery CT750 HD scanner, WI, USA), and iohexol (Opaxol, MDS Sağlık Ürünleri Tic. A.Ş., İstanbul, Türkiye) as a contrast agent (70 mL of non-ionic contrast agent (350 mg I/ML) was injected into the antecubital vein at 2 mL/s via a power injector). Digital imaging and communications in medicine (DiCOM) data were transferred to a dedicated workstation. Based on the contrast-enhanced CT images, 3D modelling was performed with a software-based evaluation using the software tools "Segment Editor" and "Segmentation" of the open-source 3D Slicer software (v. 5.1.1, https://www.slicer.org). For optimal results, contrast-enhanced CT images with a slice thickness of ?1.25 mm were used for 3D modelling. Lung reconstruction was performed by identifying and segmenting the pulmonary bronchovascular structures and placing the tumor. The use of contrast medium combined with accurate contrast timing allows the system to automatically perform 3D modelling of pulmonary arteries and veins. In addition, segment extraction was performed using virtual simulation to visualize the resection site and extent of the surgical margin. The 3D reconstruction and virtual simulation of the resection for each patient were recorded in an interactive video file in which all pulmonary structures could be individually selected. The files were transferred to a mobile device. All reconstructions were performed by an operating surgeon. The patients were evaluated and approved by two independent radiologists.

Surgical technique

The patient was placed in the lateral decubitus position with the intact side facing down. A doublelumen intubation tube was placed, and general anesthesia with one-lung ventilation was performed. For uniportal surgery, a single 3 to 4-cm long incision was made in the fifth or sixth intercostal space between the anterior and mid-axillary line, as described in our previous study.[12] The incision was closed by using a silicone wound protector. A 10-mm 30° thoracoscope (HOPKINS® Forward-Oblique 30° Telescope, Karl Storz, Tuttlingen, Germany) was used as the camera, and an endoscopic sealer/separator (LigaSure? Maryland, MN, USA) was used as an energy device for tissue dissection. An endovision system integrated with an infrared light source (IMAGE1 STM Rubina® NIR/ICG, Karl Storz, Tuttlingen, Germany) was used to determine the intersegmental planes in cases with intravenous indocyanine green (ICG). The preferred surgical instruments for vascular and bronchial dissection were the node grasper (snake), dissector clamp, endovascular clamp, right clamp, and aspirator.

In all the patients, the pulmonary hilum was released to provide more space for lung manipulation. The 3D model of the patient on the mobile device was compared intraoperatively with the patient?s anatomical structures. Vascular structures of the targeted segment were dissected, ligated with an endostapler or hemo- lock, and divided. Segmental bronchial division was performed using an endostapler. After bronchial separation, segmentectomy was completed by insufflation-deflation of the lung or the intravenous ICG method, and the fissure and intersegmental margins were separated with an endostapler.

Statistical analysis

Statistical analysis was performed using the IBM SPSS version 26.0 software (IBM Corp., Armonk, NY, USA). Continuous data were presented in mean ± standard deviation (SD) or median (min-max), while categorical data were presented in number and frequency. Compliance with normal distribution was analyzed using the Shapiro-Wilk test. An independent two-sample t-test was used to compare normally distributed surgical times between groups. The Pearson correlation coefficient was used to examine the relationship between normally distributed data and Spearman rho correlation coefficient was used to examine the relationship between non-normally distributed data. The Wilcoxon test was used to compare the success rates of the non-normally distributed 3D models. A p v alue o f < 0.05 w as considered statistically significant.

Results

Uniportal VATS segmentectomy was successfully performed in 50 patients. Segmentectomies were performed in 34% (n=17) of the right upper lobe, 24% (n=12) of the right lower lobe, 26% (n=13) of the left upper lobe, and 16% (n=8) of the left lower lobe. The resected segments and their distribution according to the lobes are presented in Table 1. The overall matching success rate of the 3D model with the bronchovascular structures identified intraoperatively was 99.27%. The arterial, venous, and bronchial matching success rates were similarly high (99.33%, 99% and 100%, respectively). Anatomical variations in the bronchovascular structures of the resected segments were evaluated. The overall variation rate was 36% (n=18) among all patients. Examples of the detected variations are shown in Figure 2.

Table 1: Distribution of segmentectomies

Figure 2. (a) In the patient who was scheduled for left upper lobe tri-segmentectomy (simple), the A4-5 artery is presented to proceed from the main pulmonary artery to the lingula segment together with the A3 artery from the in front of the upper lobe bronchus. (b) Ascending A2 artery was shown to originate from A6 in a patient with a right upper lobe S2 segmentectomy (complex) plan. (c) For right upper lobe S2 segmentectomy (complex), the V2 venous branch was shown to drain into the left atrium from behind the right main bronchus. (d) For right S1 segmentectomy (complex), B1a and B1b bronchial branches were shown to originate from B2 and B3 branches separately.

The patients were divided into two groups according to the type of segmentectomy performed: simple (Group 1) and complex segmentectomy (Group 2). There were no statistically significant differences between the two groups in terms of age and sex. The most common reason for resection in both the groups was primary lung cancer. Simple segmentectomies were performed in the left lung and lower lobes and complex segmentectomies were performed in the right lung and upper lobes (p=0.001; p=0.022). There was a significant difference between the two groups in terms of bronchovascular variation. The rate of bronchovascular variation was 11.1% (n=2) in Group 1 and 50% (n=16) in Group 2 (p=0.006). The success of preoperative 3D model matching with intraoperative bronchovascular structures did not differ between groups (Group 1/Group 2; 99.07%, 99.38%). Intraoperative complications (Group 1/Group 2; 11.1%, 3.1%) and postoperative complications (Group 1/Group 2; 5.6%/12.5%) did not significantly differ between the groups. There was no statistically significant difference in the median duration of chest tube removal (Group 1/Group 2; 1.5 /1 day) or median duration of hospital stay (Group 1/Group 2; 3 /3 days). The distribution of the groups is shown in Table 2.

Table 2. Distribution of variables according to groups

Discussion

In this study, we report the effects of 3D modeling on preoperative planning and intraoperative guidance in uniportal VATS segmentectomy. The 3D modeling of all patients was performed using conventional multi-slice CT images and open-source software. The 3D models were uploaded to mobile devices for intraoperative visualization. The videos uploaded to the mobile device allowed the surgeon to examine the vascularization and bronchial tree in the operating room. Intraoperative dissection and visualization matched the 3D model 99.27% of the time. There was no significant difference between simple and complex segmentectomies in terms of the 3D model matching characteristics. However, there was a significant difference between the two groups in terms of the anatomical bronchovascular structures. Variations were detected in the bronchovascular structures of 50% of the patients undergoing complex segmentectomy. There was no mortality in our study, with an intraoperative complication rate of 6% and postoperative complication rate of 10%.

There were more variations in the anatomy of the lung, particularly in the segmental anatomy, than in the lobar anatomy. These variations greatly increase the risk of vascular injury during segmentectomy, and high blood flow and velocity in the pulmonary circulation can cause high volumes of hemorrhage within a short time after injury.[13] Furthermore, in segmentectomy, which is a parenchyma-sparing surgery, only the bronchovascular structures of the target segment should be divided to minimize the functional capacity of the remaining lung segments. This requires that patient-specific segmental anatomy be precisely delineated with 3D modeling in the preoperative period. Whether segmentectomy performed with preoperative 3D modeling is superior to non-3D procedures has not yet been included in the guidelines.[14] Large-series studies on this subject are required. Fan et al.[13] argued that 3D modeling was an excellent technology for the assessment of pulmonary bronchovascular anatomy in their study. The opensource 3D Slicer software that we used in our study does not depend on a platform and can be installed on a portable laptop computer. As the software does not require expertise in computer technologies, it can be used by thoracic surgeons with radiologic competence. The model, created from thin-section CT images with a maximum thickness of 1.25 mm, allows detailed examination of the bronchovascular structures. The model is converted to a video format for review on a mobile device, and can be compared with structures in living tissues during the operation. It can also provide an idea of the amount of tissue to be removed by identifying target segment boundaries with anatomical landmarks. The widespread use of this technology would have a positive impact on the learning curve for segmentectomy surgery. In the literature, studies suggest that preoperative 3D modeling for the evaluation of bronchovascular variations may improve surgical accuracy and safety.[15, 16] H owever, W u e t a l.[17] suggested that preoperative 3D lung simulation could reduce operation time and ensure proper segmental margins and lymph node dissection. In our study, in which we used preoperative 3D modeling for anatomical segmentectomy, we compared the preoperative 3D lung model with intraoperative anatomical structures and found a high matching rate.

Segmentectomy is divided into simple and complex segmentectomies according to the surgical procedures and intersegmental planes.[11] Simple segmentectomy includes the right S6, left S6, left S1+2+3, and left S4+5 and is completed by creating only one linear intersegmental plane. Other segmentectomies, which are completed by creating more than one intersegmental plane, are classified as complex segmentectomies.[18] W ang e t a l.[19] reported that routine preoperative high-resolution CT (HRCT) images would be sufficient for simple segmentectomy, but emphasized the importance of preoperative 3D simulation technology for complex segmentectomy. In our study, the overall rate of bronchovascular variations was 36%. We observed a bronchovascular variation rate of 11.1% in simple segmentectomies and 50% in complex segmentectomies. Similar to the literature, a significant difference was found between simple and complex segmentectomies in terms of bronchovascular variation. This result emphasizes the importance of 3D modeling, particularly for patients scheduled for complex segmentectomy.

The complexity of the segmental anatomy of the lung and narrowing of the distal diameter of the bronchovascular structures are factors that facilitate damage to these structures during dissection. However, preoperative 3D modeling of the lung virtually reveals the direction of extension of the bronchovascular structures and their relationship with the surrounding tissues. Studies have shown that preoperative 3D modeling of the lung reduces intra- and postoperative complication rates.[20, 21] In a comprehensive systematic review, Xiang et al.[22] showed that segmentectomy performed with preoperative 3D modeling could achieve better intraoperative and postoperative outcomes in terms of operative time, conversion rate, postoperative hospital stay and complications. Hu et al.[23] a rgued t hat p reoperative 3 D m odeling h ad n o effect on postoperative outcomes, although it had a positive effect on intraoperative complications. By contrast, Xu et al.[24] in their study comparing groups with and without 3D modeling, postoperative drainage times were significantly better in the group using 3D modeling. In this study, minor intraoperative bleeding occurred in two cases of simple segmentectomy, and bronchial injury occurred in one case of complex segmentectomy. None of the patients were converted to open thoracotomy or extended resection due to major complications. We believe that this result may be directly related to the reduced risk of intraoperative complications with 3D modeling.

Nonetheless, the present study has certain limitations. First, the sample size was not sufficiently large, and the study was designed retrospectively. A prospective study with a large case series is required for more efficient analysis. Second, the generalizability is limited due to the single-center design of the study. Third, it shows the intraoperative effects of 3D modeling and the early postoperative results. A large case series is needed to compare long-term results. Fourth, we considered the fact that intersegmental planes were not determined by the same method in all patients as a limitation. Although the main vascular structures of the target segment were compared with the 3D model, the distal borders, which play an important role in determining the segment boundaries, could not be evaluated using the standard method in all patients.

In conclusion, 3D modeling with open-source free software for preoperative planning and intraoperative guidance is a safe method for detecting bronchovascular structures in the target segment with high accuracy in uniportal video-assisted thoracoscopic surgery segmentectomy procedures. Complex procedures, such as segmentectomy, can be performed with low complication rates. We also expect that this method may have a positive impact on the learning curve of the surgeon performing the procedure. Therefore, we believe that this would be useful for all simple and complex segmentectomies.

Acknowledgments: We would like to thank radiologists Aslı Tanrivermis Sayit, MD., and Hasan Gundogdu, MD., for checking the reliability of the computed tomography images when they were reconstructed into 3D images.

Data Sharing Statement: The data that support the findings of this study are available from the corresponding author upon reasonable request.

Author Contributions: Idea/concept, design, writing the article, analysis and/or interpretation: S.G., A.Ş.; Control/ supervision, literature review, critical review: A.Ş.; Data collection and/or processing, references and fundings: S.G.;

Conflict of Interest: The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

Funding: The authors received no financial support for the research and/or authorship of this article.

References

1) Saji H, Okada M, Tsuboi M, Nakajima R, Suzuki K, Aokage K, et al. Segmentectomy versus lobectomy in small-sized peripheral non-small-cell lung cancer (JCOG0802/WJOG4607L): A multicentre, open-label, phase 3, randomised, controlled, non-inferiority trial. Lancet 2022;399:1607-17. doi: 10.1016/S0140-6736(21)02333-3.

2) Altorki N, Wang X, Kozono D, Watt C, Landrenau R, Wigle D, et al. Lobar or sublobar resection for peripheral stage IA non-small-cell lung cancer. N Engl J Med 2023;388:489-98. doi: 10.1056/NEJMoa2212083.

3) Ikeda N, Yoshimura A, Hagiwara M, Akata S, Saji H. Three dimensional computed tomography lung modeling is useful in simulation and navigation of lung cancer surgery. Ann Thorac Cardiovasc Surg 2013;19:1-5. doi: 10.5761/atcs. ra.12.02174.

4) Gossot D, Mariolo AV, Seguin-Givelet A. Thoracoscopic anatomical segmentectomies for early-stage lung cancer: The coming challenge. J Thorac Dis 2020;12:4564-7. doi: 10.21037/jtd.2020.04.27.

5) Seguin-Givelet A, Grigoroiu M, Brian E, Gossot D. Planning and marking for thoracoscopic anatomical segmentectomies. J Thorac Dis 2018;10:S1187-94. doi: 10.21037/jtd.2018.02.21.

6) Göçer H, Durukan AB, Tunç O, Naseri E, Ercan E. Evaluation of 3D printed carotid anatomical models in planning carotid artery stenting. Turk Gogus Kalp Dama 2020;28:294-300. doi: 10.5606/tgkdc.dergisi.2020.18939.

7) Yao F, Wang J, Yao J, Hang F, Lei X, Cao Y. Threedimensional image reconstruction with free open-source OsiriX software in video-assisted thoracoscopic lobectomy and segmentectomy. Int J Surg 2017;39:16-22. doi: 10.1016/j. ijsu.2017.01.079.

8) Cui Z, Ding C, Li C, Song X, Chen J, Chen T, et al. Preoperative evaluation of the segmental artery by threedimensional image reconstruction vs. thin-section multidetector computed tomography. J Thorac Dis 2020;12:4196-204. doi: 10.21037/jtd-20-1014.

9) Loftus TJ, Tighe PJ, Filiberto AC, Efron PA, Brakenridge SC, Mohr AM, et al. Artificial intelligence and surgical decision-making. JAMA Surg 2020;155:148-58. doi: 10.1001/ jamasurg.2019.4917.

10) Li X, Zhang S, Luo X, Gao G, Luo X, Wang S, et al. Accuracy and efficiency of an artificial intelligence-based pulmonary broncho-vascular three-dimensional reconstruction system supporting thoracic surgery: Retrospective and prospective validation study. EBioMedicine 2023;87:104422. doi: 10.1016/j.ebiom.2022.104422.

11) Wang X, Wang Q, Zhang X, Yin H, Fu Y, Cao M, et al. Application of three-dimensional (3D) reconstruction in the treatment of video-assisted thoracoscopic complex segmentectomy of the lower lung lobe: A retrospective study. Front Surg 2022;9:968199. doi: 10.3389/fsurg.2022.968199.

12) Gurz S, Temel NG, Sengul AT, Buyukkarabacak Y, Pirzirenli MG, Basoglu A. Learning curve for uniportal VATS anatomical pulmonary resections: the activity monitor operating characteristic method. Indian J Surg 2023;85:434-441. doi: 10.1007/s12262-023-03667-6.

13) Fan K, Feng J, Li Y, Liu B, Tao R, Wang Z, et al. Application of three-dimensional reconstruction of left upper lung lobes in anatomical segmental resection. Thorac Cancer 2022;13:1176-83. doi: 10.1111/1759-7714.14379.

14) Postmus PE, Kerr KM, Oudkerk M, Senan S, Waller DA, Vansteenkiste J, et al. Early and locally advanced Non-Small-Cell Lung Cancer (NSCLC): ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2017;28:iv1-21. doi: 10.1093/annonc/mdx222.

15) Chen-Yoshikawa TF, Date H. Update on three-dimensional image reconstruction for preoperative simulation in thoracic surgery. J Thorac Dis 2016;8:S295-301. doi: 10.3978/j. issn.2072-1439.2016.02.39.

16) Sardari Nia P, Olsthoorn JR, Heuts S, Maessen JG. Interactive 3D reconstruction of pulmonary anatomy for preoperative planning, virtual simulation, and intraoperative guiding in video-assisted thoracoscopic lung surgery. Innovations (Phila) 2019;14:17-26. doi: 10.1177/1556984519826321.

17) Wu X, Li T, Zhang C, Wu G, Xiong R, Xu M, et al. Comparison of perioperative outcomes between precise and routine segmentectomy for patients with early-stage lung cancer presenting as ground-glass opacities: A propensity score-matched study. Front Oncol 2021;11:661821. doi: 10.3389/fonc.2021.661821.

18) Handa Y, Tsutani Y, Mimae T, Miyata Y, Okada M. Complex segmentectomy in the treatment of stage IA non-small-cell lung cancer. Eur J Cardiothorac Surg 2020;57:114-21. doi: 10.1093/ejcts/ezz185.

19) Wang R, Zhang Y, Hu Q, Jin K, Huang G, Shen J, et al. Identification of the segmental structures of the right upper lobe of the lung using non-enhanced thin-slice CT. J Thorac Dis 2020;12:1639-44. doi: 10.21037/jtd.2020.03.56.

20) Xue L, Fan H, Shi W, Ge D, Zhang Y, Wang Q, et al. Preoperative 3-dimensional computed tomography lung simulation before video-assisted thoracoscopic anatomic segmentectomy for ground glass opacity in lung. J Thorac Dis 2018;10:6598-605. doi: 10.21037/jtd.2018.10.126.

21) She XW, Gu YB, Xu C, Li C, Ding C, Chen J, et al. Threedimensional (3D)- computed tomography bronchography and angiography combined with 3D-Video-Assisted Thoracic Surgery (VATS) versus conventional 2D-VATS anatomic pulmonary segmentectomy for the treatment of non-small cell lung cancer. Thorac Cancer 2018;9:305-9. doi: 10.1111/1759-7714.12585.

22) Xiang Z, Wu B, Zhang X, Feng N, Wei Y, Xu J, et al. Preoperative three-dimensional lung simulation before thoracoscopic anatomical segmentectomy for lung cancer: A systematic review and meta-analysis. Front Surg 2022;9:856293. doi: 10.3389/fsurg.2022.856293.

23) Hu W, Zhang K, Han X, Zhao J, Wang G, Yuan S, et al. Three-dimensional computed tomography angiography and bronchography combined with three-dimensional printing for thoracoscopic pulmonary segmentectomy in stage IA non-small cell lung cancer. J Thorac Dis 2021;13:1187-95. doi: 10.21037/jtd-21-16.

24) Xu Y, Chen N, Ma A, Wang Z, Zhang Y, Liu C, et al. Threedimensional versus two-dimensional video-assisted thoracic surgery for thoracic disease: A meta-analysis. Interact Cardiovasc Thorac Surg 2017;25:862-71. doi: 10.1093/icvts/ivx219.