Wedge resection for pulmonary sequestration guided by indocyanine green fluorescence and 3-dimensional reconstruction Authors: Qing Liang, Shaomin Lin, Zihao Chen, Zhifei Guo, Qichan Gao, Xuening Yang, Hongxia Tian, Yilong Wu, Wenzhao Zhong, Qiang Nie Published: 2026 ABSTRACT ObjectivesPulmonary sequestration (PS) is a rare congenital lung malformation with an often-indistinct boundary between sequestered and normal lung tissue, leading to lobectomy and potential loss of lung function. This study evaluated whether watershed analysis combined with intraoperative indocyanine green fluorescence can accurately delineate the PS boundary and enable wedge resection.MethodsWe analyzed consecutive patients who underwent surgical resection for PS at our center between January 2023 and September 2025. Preoperative contrast-enhanced computed tomography with 3-dimensional reconstruction was used to simulate the PS boundaries of normal lung tissue and to assess whether wedge resection was feasible. During video-assisted thoracoscopic surgery, the PS-feeding systemic artery was divided, followed by rapid intravenous indocyanine green injection and fluorescence imaging to visualize the interface between sequestered and normal lung tissue, allowing completion of the wedge resection.ResultsForty-seven patients were included: 35 underwent watershed-guided wedge resection and 12 underwent lobectomy. In all cases, intraoperative fluorescence accurately matched the preoperative 3-dimensional watershed simulation. By preoperative simulation of the expected boundary, wedge resection was feasible in 74.5% (35/47) of patients in this cohort. Compared with lobectomy, watershed-guided wedge resection was associated with shorter operative time, less blood loss, shorter chest tube duration, and shorter length of stay.ConclusionsThe watershed topographic mapping technique we developed can accurately predict the boundaries of PS preoperatively, perform fluorescence verification intraoperatively, and allow for wedge resection of PS to avoid lobectomy, while ensuring that the surgical procedure is safe and reliable. 3D watershed with ICG fluorescence guide parenchyma-sparing wedge resection for PS. Central Message3D watershed analysis plus ICG fluorescence enables safe, parenchyma-sparing wedge resection for pulmonary sequestration with better perioperative outcomes and lung function than lobectomy. PerspectiveThis study shows that CT-based 3D watershed planning, validated by intraoperative ICG fluorescence, enables precise wedge resection for pulmonary sequestration, helping avoid incomplete excision, preserve lung function, and inform when lobectomy is necessary. 3D watershed with ICG fluorescence guide parenchyma-sparing wedge resection for PS. Central Message 3D watershed analysis plus ICG fluorescence enables safe, parenchyma-sparing wedge resection for pulmonary sequestration with better perioperative outcomes and lung function than lobectomy. Perspective This study shows that CT-based 3D watershed planning, validated by intraoperative ICG fluorescence, enables precise wedge resection for pulmonary sequestration, helping avoid incomplete excision, preserve lung function, and inform when lobectomy is necessary. Pulmonary sequestration (PS) is a rare congenital lung malformation in which a portion of lung parenchyma is isolated from the normal tracheobronchial tree and receives its arterial supply from the systemic circulation rather than the pulmonary artery.1 Venous drainage usually occurs via the pulmonary veins but may also involve systemic veins.2 The aberrant systemic artery most commonly arises from the thoracic aorta, followed by the abdominal aorta, the intercostal arteries, and the subclavian arteries; more unusual sources, including coronary and internal thoracic arteries, have also been described.2, 3, 4, 5, 6, 7 Although most PS lesions are supplied by a single aberrant vessel, multiple feeding arteries may be present in a minority of cases.8,9 PS accounts for approximately 0.15% to 6.4% of congenital lung malformations.10 PS is classified as intralobar or extralobar, depending on whether it shares the visceral pleura with adjacent lung; intralobar PS is more common.11 Rarely, both types may coexist in the same patient.12,13 PS predominantly affects the lower lobes, especially the left lower lobe, and the patient may remain asymptomatic for years. When complicated by infection, patients can present with recurrent fever, cough, expectoration, and occasionally hemoptysis.14,15 Historically, PS has been treated by open thoracotomy with resection of the affected lung and ligation of the aberrant systemic artery. With the advancement of video-assisted thoracoscopic surgery, thoracoscopic resection has become the preferred approach in many centers.16 However, several technical challenges remain. On conventional computed tomography (CT) scans, the interface between sequestered and normally perfused lung parenchyma is often difficult to define. Consequently, lobectomy or sometimes segmentectomy is frequently performed, particularly for intralobar PS, where a clear anatomic boundary is lacking.17, 18, 19, 20 Extralobar PS is typically well encapsulated and can usually be managed with simple excision or wedge resection; lobectomy is rarely required unless there is atypical intraparenchymal extension, severe inflammation, or an indistinct interface with adjacent lung. These strategies have important limitations. Extensive anatomic resection may unnecessarily sacrifice normal lung parenchyma and lead to long-term loss of pulmonary function, especially in younger patients. Conversely, attempts at parenchyma-sparing resection without clear boundary definition risk incomplete excision of sequestered tissue, with subsequent hemoptysis, recurrent infection, and persistence of preoperative symptoms.9,12,14 Multiple previous reports have described the use of intraoperative indocyanine green (ICG) fluorescence to demarcate intralobar PS and guide parenchyma-sparing resection.21, 22, 23, 24 These studies support the utility of fluorescence-based perfusion demarcation, but questions remain regarding preoperative planning, anticipation of complex aberrant arterial anatomy, and reproducible definition of the expected demarcation pattern before vessel division. Watershed analysis is a recently developed technique for real-time localization of nonpalpable pulmonary nodules on the lung surface.25 The method uses preoperative 3-dimensional (3D) reconstruction of pulmonary vessels to identify the arterial branches supplying the target nodule. Intraoperatively, temporary occlusion of these vessels followed by intravenous ICG injection and near-infrared thoracoscopy reveals a fluorescence “watershed” on the lung surface, delineating the perfusion territory of the occluded artery and guiding wedge resection. A key advantage of this approach is the ability to simulate the postocclusion fluorescence pattern on preoperative 3D images, enabling precise surgical planning. We hypothesized that this principle could be adapted to PS. In this study, we applied arterial watershed mapping to preoperatively delineate the perfusion territory of PS-feeding arteries and used intraoperative ICG fluorescence to visualize the interface between sequestered and normal lung parenchyma. We evaluated the feasibility and safety of this technique and compared perioperative outcomes and pulmonary function at 6 months between patients undergoing watershed-guided wedge resection and those undergoing lobectomy. Methods Patients This was a real-world cohort study. We included consecutive patients with PS who underwent surgical resection at Guangdong Provincial People's Hospital between January 2023 and September 2025. The study flow is shown in Figure 1. The diagnosis of PS was established preoperatively on the basis of contrast-enhanced CT demonstrating sequestered lung parenchyma with aberrant systemic arterial supply, in combination with clinical and radiologic findings. Patients were excluded if imaging was insufficient to confirm PS or if they had concomitant major thoracic procedures during the same operation. Patients were grouped according to the procedure ultimately performed. The watershed group included patients in whom watershed analysis was used preoperatively and ICG fluorescence was used intraoperatively to guide wedge resection. The lobectomy group included patients who received watershed topography analysis but ultimately underwent lobectomy. 3D reconstruction of lung imaging data and ICG allergy assessment were performed preoperatively, and written informed consent for surgery was obtained from each patient after explanation of the risks and benefits of the procedure.Figure 1The flowchart of patients included in this study. 3D, 3-Dimensional; ICG, indocyanine green. The flowchart of patients included in this study. 3D, 3-Dimensional; ICG, indocyanine green. Preoperative Evaluation All patients underwent 64-slice multidetector CT scanning using our institutional protocol. Thin-section CT data (1.0-mm slice thickness) were imported into 3D reconstruction software (Optomedic SAT, OPTO-FM2000; Guangdong OptoMedic Technologies), the software was used as an auxiliary tool for surgical planning and navigation, and it has received regulatory approval from the China National Medical Products Administration (approval no. 20242210434). This retrospective study was approved by the Institutional Ethics Committee of Guangdong Provincial People's Hospital (KY2024-296-01, April 7, 2024). The requirement for written informed consent for study participation was waived because of the retrospective nature of the study. The data underlying this article are available in the article and in its online supplementary material. Further requests for data can be made to the corresponding author. All patients provided routine written informed consent for the surgical procedure. Automated 3D reconstructions of the tracheobronchial tree, pulmonary arteries and veins, and aberrant systemic arteries were generated and manually reviewed by the surgical team. Using the built-in watershed analysis function, the software simulated the perfusion territory of the aberrant PS-feeding artery. Using the built-in watershed function, the software estimated the perfusion territory of the systemic feeding artery and projected the expected pleural demarcation after feeding-artery division and intravenous ICG injection. The predicted boundary was used to plan the intended resection margin. The presence of pulmonary arterial inflow to the sequestered lung and Pryce classification were assessed on contrast-enhanced CT/CT angiography and 3D reconstruction by 2 thoracic surgeons and 1 radiologist, with disagreements resolved by consensus. Surgical Procedure All procedures were performed with the patient receiving general anesthesia and single-lung ventilation. A 3- to 4-cm muscle-sparing anterolateral utility incision was used as the primary access for video-assisted thoracoscopic surgery. Thoracoscopic images and 3D reconstructions were displayed side by side on dual monitors to facilitate intraoperative navigation. In the watershed group, the aberrant systemic artery supplying the PS was identified by correlating thoracoscopic findings with the preoperative 3D vascular map. The target artery was carefully dissected, encircled, and temporarily controlled with a colored slipknot for traction and confirmation of the target vessel. After the anatomy was confirmed, the systemic feeding artery was divided using an endoscopic stapler. Immediately thereafter, 5 mL of ICG at a concentration of 2.5 mg/mL (total dose, 12.5 mg) was rapidly injected through a peripheral vein. Near-infrared fluorescence imaging was performed using Optomedic SAT (OPTO-FM2000). Two regions became visible: a fluorescent area representing normal lung tissue and a nonfluorescent area corresponding to the perfusion territory of the divided systemic feeding artery. The fluorescence demarcation line was marked on the visceral pleura using electrocautery. Agreement between simulated and observed boundaries was assessed qualitatively by the operating surgeons using intraoperative fluorescence images and corresponding 3D reconstructions. Guided by the preoperative 3D watershed simulation and the intraoperative fluorescence boundary, wedge resection of the PS was performed using an endoscopic stapler, with the aim of completely remove the sequestered tissue while preserving as much healthy lung as possible. The extent of resection was tailored according to lesion size, anatomic location, and surrounding parenchymal condition (Figure 2). In cases with suspected dual supply or an equivocal fluorescence pattern, additional intraoperative perfusion testing was performed by temporarily occluding relevant pulmonary arterial branches followed by repeat ICG injection to confirm the demarcation (Video 1).Figure 2Watershed analysis in PS surgery intraoperatively. A, CT scan indicating that the artery supplying the PS originates from the thoracic aorta. B and C, Preoperative 3D reconstruction and simulation of the target territory on the basis of contrast-enhanced CT. D and E, Intraoperative identification and division of the PS artery using a stapler. F and G, After the injection of ICG, the boundary between the target territory and adjacent normally perfused lung is visualized using a near-infrared thoracoscopic system. H and I, Wedge resection of the PS tissue revealed an abscess cavity upon dissection. PS, Pulmonary sequestration; CT, computed tomography; 3D, 3-dimensional; ICG, indocyanine green. Watershed analysis in PS surgery intraoperatively. A, CT scan indicating that the artery supplying the PS originates from the thoracic aorta. B and C, Preoperative 3D reconstruction and simulation of the target territory on the basis of contrast-enhanced CT. D and E, Intraoperative identification and division of the PS artery using a stapler. F and G, After the injection of ICG, the boundary between the target territory and adjacent normally perfused lung is visualized using a near-infrared thoracoscopic system. H and I, Wedge resection of the PS tissue revealed an abscess cavity upon dissection. PS, Pulmonary sequestration; CT, computed tomography; 3D, 3-dimensional; ICG, indocyanine green. In the lobectomy group, patients underwent a preoperative assessment using watershed analysis to evaluate the feasibility of wedge resection. However, during intraoperative exploration, if the lung showed no fluorescence after dividing the PS artery and injecting ICG, or if the PS artery supply area was extensive, causing loss of function in the lower lung tissue, and wedge resection might result in incomplete or nonexpansion of the remaining lung tissue, the surgical plan was altered intraoperatively to perform a lobectomy. Standard thoracoscopic lobectomy techniques were used, with dissection of the pulmonary artery, pulmonary vein, and bronchus on the basis of preoperative 3D reconstruction, to remove the affected lobe. Lymph node sampling was not performed. Data Collection and Outcomes Demographic data (age and sex), clinical presentation, PS location and type (intralobar vs extralobar), and aberrant artery origin were recorded. Perioperative data included operative time, intraoperative blood loss, type and extent of resection, resected lung volume (measured from the specimen), duration of chest tube drainage, postoperative hospital stay, and complications within 30 days and up to 3 months. Pulmonary function testing, including forced expiratory volume in 1 second (FEV1), was performed preoperatively and at 6 months postoperatively in all patients. The primary functional outcome was the absolute change in FEV1 (6-month minus preoperative value); postoperative FEV1 at 6 months was also analyzed. Statistical Analysis Continuous variables are presented as mean ± standard deviation or median (interquartile range) depending on distribution and compared between groups using the unpaired Student t test for normally distributed data or the Mann-Whitney U test for non-normally distributed data, as appropriate. Categorical variables are expressed as counts (percentages) and compared using the χ2 test or Fisher exact test. All statistical analyses were performed with GraphPad Prism 9 (GraphPad Software). Results Between January 2023 and September 2025, 47 patients underwent surgical resection for PS at our center. All patients underwent preoperative watershed analysis and were planned for wedge resection; ultimately, 35 patients (74.5%) underwent watershed-guided wedge resection, whereas 12 patients (25.5%) underwent lobectomy. Baseline demographic and clinical characteristics, including age, sex distribution, symptom profile, and PS location, were similar between groups (Table 1). All patients completed preoperative 3D reconstruction and watershed analysis. Aberrant systemic arteries originated most commonly from the thoracic aorta, with additional origins from the abdominal aorta, musculophrenic artery, and, in rare cases, the hepatic artery. In terms of Pryce classification, no Pryce type I lesions were identified in either group. The watershed-guided wedge resection group included 2 patients with Pryce type II and 33 with Pryce type III, whereas the lobectomy group included 1 patient with Pryce type II and 11 with Pryce type III; the distribution did not differ between groups (P > .9999), suggesting comparable anatomic patterns of sequestration across cohorts.Table 1Patient characteristics in both groupsCharacteristicsWatershed groupLobectomy groupP valueTotal3512Sex Male17 (48.57%)4 (33.33%).5047 Female18 (51.43%)8 (66.67%)Median age (range), y36 (13-66)30 (9-52).1842Type of PS Intralobar31 (88.57%)12 (100.00%).5597 Extralobar4 (11.43%)0 (0.00%)PS location LLL24 (68.57%)8 (66.67%)>.9999 RLL11 (31.43%)4 (33.33%)Origin of systemic feeding artery Thoracic aorta32 (91.43%)9 (75.00%).1857 Abdominal aorta2 (5.71%)1 (8.33%) Musculophrenic artery1 (2.86%)1 (8.33%) Hepatic artery0 (0.00%)1 (8.33%)Number of PS arteries One artery32 (91.43%)8 (66.67%).1019 Two arteries2 (5.71%)3 (25.00%) Three arteries1 (2.86%)1 (8.33%)Pryce classification>.9999 Pryce I00 Pryce II21 Pryce III3311Operative approach VATS34 (97.14%)9 (75.00%).0459 Thoracotomy1 (2.86%)3 (25.00%)PS, Pulmonary sequestration; LLL, left lower lobe; RLL, right lower lobe; VATS, video-assisted thoracoscopic surgery. Table 1Patient characteristics in both groupsCharacteristicsWatershed groupLobectomy groupP valueTotal3512Sex Male17 (48.57%)4 (33.33%).5047 Female18 (51.43%)8 (66.67%)Median age (range), y36 (13-66)30 (9-52).1842Type of PS Intralobar31 (88.57%)12 (100.00%).5597 Extralobar4 (11.43%)0 (0.00%)PS location LLL24 (68.57%)8 (66.67%)>.9999 RLL11 (31.43%)4 (33.33%)Origin of systemic feeding artery Thoracic aorta32 (91.43%)9 (75.00%).1857 Abdominal aorta2 (5.71%)1 (8.33%) Musculophrenic artery1 (2.86%)1 (8.33%) Hepatic artery0 (0.00%)1 (8.33%)Number of PS arteries One artery32 (91.43%)8 (66.67%).1019 Two arteries2 (5.71%)3 (25.00%) Three arteries1 (2.86%)1 (8.33%)Pryce classification>.9999 Pryce I00 Pryce II21 Pryce III3311Operative approach VATS34 (97.14%)9 (75.00%).0459 Thoracotomy1 (2.86%)3 (25.00%)PS, Pulmonary sequestration; LLL, left lower lobe; RLL, right lower lobe; VATS, video-assisted thoracoscopic surgery. Patient characteristics in both groups PS, Pulmonary sequestration; LLL, left lower lobe; RLL, right lower lobe; VATS, video-assisted thoracoscopic surgery. All patients in the watershed group tolerated the ICG injection without allergic reactions or other adverse events. After division of the PS-feeding systemic artery, ICG injection followed by near-infrared thoracoscopy (Optomedic SAT, OPTO-FM2000) revealed a clear fluorescence boundary on the lung surface in every case. The shape and distribution of the fluorescent and nonfluorescent regions closely matched the preoperative 3D watershed simulations (Figure 3), supporting the accuracy of the technique for delineating the perfusion territory of the target systemic artery. In 35 patients, the fluorescence demarcation boundary allowed tailored wedge resection that encompassed the intended target territory while preserving adjacent normal lung. In the remaining 12 patients, intraoperative assessment demonstrated extensive parenchymal destruction and markedly compromised perfusion of the residual lower-lobe parenchyma after systemic feeding-artery division, as evidenced by limited fluorescence after ICG injection. Because wedge resection would likely leave a poorly perfused remnant with inadequate expansion and function, the procedure was converted to lobectomy. Compared with the lobectomy group, watershed-guided resection was associated with a significantly shorter operative time (98.80 ± 35.82 minutes vs 179.9 ± 34.00 minutes; P < .0001) and lower intraoperative blood loss (15.46 ± 7.22 mL vs 33.58 ± 32.03 mL; P = .0027). Resection volume measured from the resected PS pathologic tissue was smaller in the watershed group than in the lobectomy group (72.14 ± 30.83 cm3 vs 228.30 ± 111.6 cm3; P < .0001). Duration of postoperative chest tube drainage was also shorter in the watershed group (41.94±13.28 hours vs 56.25±17.56 hours; P = .0049). Meanwhile, patients undergoing wedge resection guided by watershed topography had a hospital stay of 2.71 ± 0.67 days, significantly shorter than that of patients undergoing lobectomy (5.33 ± 1.37 days, P < .0001).Figure 3The white line indicates the boundary between the target territory and adjacent normally perfused lung. The white line indicates the boundary between the target territory and adjacent normally perfused lung. Postoperative complications differed between groups. In the lobectomy group, 2 patients developed massive hemothorax within 2 days after surgery and required reoperation. Intraoperative exploration identified rupture and bleeding of a residual PS-feeding artery in 1 patient and bleeding from the pulmonary artery stump in the other. In addition, 1 patient in the lobectomy group developed a pulmonary infection 1 month postoperatively, accompanied by right middle lobe atelectasis; further evaluation led to the diagnosis of a bronchopleural fistula, and this patient also required reoperation. In contrast, no severe complications occurred in the watershed group. One patient had increased postoperative pleural drainage with a mild decline in hemoglobin level, raising concern for intrathoracic bleeding; this was managed conservatively with intravenous etamsylate and batroxobin. Drainage subsequently decreased, and follow-up imaging of the chest showed no progression of pleural effusion, allowing safe chest tube removal without further intervention. Pulmonary function was evaluated in all patients preoperatively and at 6 months postoperatively. Preoperative FEV1 was similar between groups (3.62 ± 0.54 L vs 3.28 ± 0.64 L; P = .0789). At 6 months after surgery, patients in the watershed group had greater FEV1 than those in the lobectomy group (3.35 ± 0.59 L vs 2.70 ± 0.64 L; P = .0024). The mean decrease in FEV1 was 0.27 ± 0.25 L in the watershed group compared with 0.58 ± 0.17 L in the lobectomy group (P < .0001), indicating less postoperative functional loss (Table 2).Table 2Differences between the 2 surgical methodsCharacteristicsWatershed groupLobectomy groupP valueOperative time, min98.80 ± 35.82179.9 ± 34.00<.0001Intraoperative blood loss, mL15.46 ± 7.2233.58 ± 32.03.0027Resection volume, cm372.14 ± 30.83228.30 ± 111.6<.0001Postoperative tube drainage, h41.94 ± 13.2856.25 ± 17.56.0049Postoperative length of stay, d2.71 ± 0.675.33 ± 1.37<.0001Complications Pneumonia01 Hemothorax12 DVT/PE00 Bronchopleural fistula01 Reoperation03 Readmission02 Prolonged air leak00 Arrhythmia00 Empyema00 Atelectasis01FEV1/L Before surgery3.62 ± 0.543.28 ± 0.64.0789 6 mo after surgery3.35 ± 0.592.70 ± 0.64.0024Mean decrease in FEV10.27 ± 0.250.58 ± 0.17<.0001DVT, Deep vein thrombosis; PE, pulmonary embolism; FEV1, forced expiratory volume in 1 second. Table 2Differences between the 2 surgical methodsCharacteristicsWatershed groupLobectomy groupP valueOperative time, min98.80 ± 35.82179.9 ± 34.00<.0001Intraoperative blood loss, mL15.46 ± 7.2233.58 ± 32.03.0027Resection volume, cm372.14 ± 30.83228.30 ± 111.6<.0001Postoperative tube drainage, h41.94 ± 13.2856.25 ± 17.56.0049Postoperative length of stay, d2.71 ± 0.675.33 ± 1.37<.0001Complications Pneumonia01 Hemothorax12 DVT/PE00 Bronchopleural fistula01 Reoperation03 Readmission02 Prolonged air leak00 Arrhythmia00 Empyema00 Atelectasis01FEV1/L Before surgery3.62 ± 0.543.28 ± 0.64.0789 6 mo after surgery3.35 ± 0.592.70 ± 0.64.0024Mean decrease in FEV10.27 ± 0.250.58 ± 0.17<.0001DVT, Deep vein thrombosis; PE, pulmonary embolism; FEV1, forced expiratory volume in 1 second. Differences between the 2 surgical methods DVT, Deep vein thrombosis; PE, pulmonary embolism; FEV1, forced expiratory volume in 1 second. Patients underwent CT examinations of the chest at 3 months, 6 months, and 1 year postoperatively. At 3 months, 3 of 35 patients who underwent watershed-guided wedge resection had minimal pleural effusion and mild inflammatory changes at the resection margin. By 1 year, CT scans of the chest of all patients who underwent watershed-guided wedge resection showed no atelectasis of the residual lung, pulmonary infection, or pleural effusion. The PS artery stump was gradually encapsulated by connective tissue, with no bleeding, rupture, or bronchopleural fistula (Figure E1). None of the patients required further related treatment. Currently, 7 patients who underwent wedge resection on the basis of watershed topography map have been followed up for more than 3 years. During the follow-up period, CT scans of the chest did not indicate postoperative recurrence, pleural infection, or other complications, and all patients have continued to receive regular dynamic follow-up. Discussion In this single-center cohort study, we performed refined 3D reconstructions of patients with PS using enhanced CT scans preoperatively and applied watershed topographic mapping technology to simulate the boundary between PS and normal tissue. This simulation was validated using intraoperative ICG fluorescence, demonstrating that the predicted range matched the actual range and guided wedge resection of PS. This strategy may facilitate preoperative planning by anticipating the expected demarcation pattern, whereas intraoperative ICG fluorescence provides real-time confirmation of the perfusion boundary. Compared with lobectomy, watershed-guided wedge resection was associated with shorter operative time, reduced blood loss, smaller resection volume, fewer major complications, and better preservation of FEV1 at 6 months. Surgical management of PS is challenging because of its heterogeneous anatomy and the often-indistinct interface between sequestered and normal lung. Reported complications of PS resection include life-threatening intraoperative hemorrhage from aberrant systemic arteries, postoperative hemothorax, prolonged air leak, empyema, and persistent infection.8,11,15 For intralobar PS, which often lacks a clear anatomic boundary, many centers perform division of the aberrant systemic artery followed by lobectomy. Extralobar PS is typically well encapsulated and is usually managed with simple excision or wedge resection; lobectomy is rarely required except in atypical cases with intraparenchymal extension or severe inflammation.17, 18, 19, 20 Both approaches have inherent limitations. Extensive anatomic resection may unnecessarily sacrifice normal lung tissue, resulting in long-term loss of pulmonary function. Conversely, attempts at parenchyma-sparing resection without clear boundary definition risk leaving residual sequestered tissue, leading to hemoptysis, recurrent infection, and persistence of symptoms. Our findings align with previous reports highlighting these trade-offs and underscore the need for techniques that can reliably delineate the PS perfusion territory. The watershed-guided technique addresses several of these limitations. First, automated 3D CT reconstruction allows detailed visualization of pulmonary arterial and venous anatomy, which is difficult to achieve consistently with 2-dimensional imaging alone.25 This approach enables preoperative visualization of the relationship between aberrant systemic arteries, pulmonary vessels, and surrounding lung parenchyma. Second, the software we developed can estimate the tissue of the PS-feeding arteries and simulate the expected fluorescence area on the lung surface after arterial occlusion and ICG injection. This simulation supports preoperative planning by predicting the boundary between the target territory and adjacent normally perfused lung and informing the intended resection margin. Third, intraoperative navigation by matching thoracoscopic findings with 3D reconstructions facilitates identification, encirclement, and division of target vessels, potentially reducing the risk of vascular injury. After division of the aberrant artery, ICG fluorescence provides real-time visualization of the boundary between the target territory and adjacent normally perfused parenchyma. This allows tailored wedge resection while preserving as much uninvolved lung as possible. The clinical relevance of these advantages is reflected in our outcomes. Watershed-guided resection was associated with shorter operative time and lower blood loss than lobectomy and resulted in smaller resection volumes and shorter chest drainage duration. In Pryce II intralobar sequestration, a dual blood supply may be present anatomically; however, we hypothesize that the long-standing high-pressure systemic feeding artery progressively compromises the low-pressure pulmonary arterial microcirculation within the sequestered parenchyma, rendering any pulmonary arterial inflow functionally insignificant for effective alveolar gas exchange and normal arteriovenous transit. This concept was supported by an intraoperative perfusion test. We first temporarily occluded the posterior segmental and basal pulmonary arterial branches to the lower lobe and then injected ICG, which produced a sharp pleural fluorescence demarcation with no staining in the posterior and basal segments. After subsequent division of the systemic feeding artery and repeat ICG injection, an equally distinct boundary appeared, and importantly, it was identical to the boundary observed during basal pulmonary arterial occlusion. This concordance indicates that the perfusion territory governed by the systemic feeding artery is not materially altered by the pulmonary arterial circulation in this setting; therefore, systemic feeding-artery control alone is sufficient to generate a reliable ICG-based demarcation between sequestered and normal lung (Video 1). Severe postoperative complications occurred only in the lobectomy group in our series. Two patients developed massive hemothorax requiring early reoperation because of bleeding from a residual PS-feeding artery or the pulmonary artery stump, and 1 patient developed a bronchopleural fistula requiring reoperation for recurrent infection and atelectasis. Such events have been reported after resection for PS, particularly in anatomically complex cases.2,17, 18, 19, 20 Patients undergoing lobectomy more often had extensive parenchymal destruction and poor residual perfusion intraoperatively, which may partly explain the greater complication rates observed. In the watershed-guided group, no major complications occurred, only 1 patient had suspected intrathoracic bleeding that was managed conservatively. Although our study was not powered to detect differences in rare events, these findings support the technical feasibility of watershed-guided wedge resection and suggest potential perioperative safety advantages. Given that many patients with PS are young and otherwise healthy, preserving lung function is a critical consideration. Our data suggest a functional advantage of a parenchyma-sparing strategy. FEV1 at 6 months was greater in the watershed group than in the lobectomy group, and the mean decline in FEV1 was smaller. In some patients who underwent wedge resection, postoperative lung function improved compared with preoperative levels. We speculate that a possible reason is that after the removal of nonfunctional lung tissue, the normal tissue within the same lobe is relieved of pressure, the ventilation-perfusion ratio is altered, pulmonary ventilation and gas exchange function are enhanced, and both lung capacity and FEV1 are increased. Despite the advantages of watershed guidance, wedge resection is not suitable for all PS cases. Watershed analysis was attempted in 12 patients who were expected for wedge resection; however, these patients ultimately underwent lobectomy because of lesion extent and parenchymal destruction. One patient had a large intralobar PS located in the posterior basal segment of the left lower lobe, supplied by 3 aberrant arteries arising at the level of the 10th rib arch and coursing around the spine and ribs in close association with the intercostal arteries and veins. Given the bulky lesion and complex vascular anatomy, we considered lobectomy to be a safer and more reliable procedure than wedge resection. The other 11 patients had long-standing symptoms and had previously been misdiagnosed with bronchiectasis. Both demonstrated extensive inflammatory destruction of the lower lobe on imaging. Preoperative watershed analysis was performed, and assessment included ICG injection after division of the PS-feeding systemic artery. In both cases, the lower lobe appeared stiff and fibrotic; after arterial division, the affected lung changed from bright red to dark reddish-black, and histology later confirmed multiple cystic cavities filled with purulent fluid. Notably, fluorescence was minimal in the lower lobe after ICG injection, suggesting severely compromised perfusion of the residual parenchyma. Under these circumstances, lobectomy was considered the most appropriate option to ensure complete removal of nonfunctional tissue and to avoid leaving a poorly perfused, dysfunctional remnant. These cases illustrate that watershed-guided wedge resection should be viewed as part of a spectrum of surgical options rather than a universal solution. Patient selection should account for lesion size, chronicity of infection, degree of parenchymal destruction, and vascular anatomy. In advanced or complex cases, particularly those with severe fibrosis or loss of function in adjacent lung, lobectomy may remain the optimal treatment despite accurate perfusion mapping. Safe division of the aberrant systemic artery remains a key concern in PS surgery. Previous reports have highlighted the risk of catastrophic bleeding, especially when the artery arises from the thoracic or abdominal aorta and traverses the posterior mediastinum.2,17, 18, 19, 20 In our experience, 3D reconstruction and watershed mapping improve preoperative understanding of the arterial origin, course, and potential anatomic variants, thereby reducing uncertainty when translating 2-dimensional imaging into operative anatomy. However, watershed analysis does not replace meticulous surgical technique. Secure control and division of PS-feeding systemic arteries continue to depend on careful dissection, appropriate use of staplers or ligatures, and adherence to established principles for managing high-pressure systemic vessels. The surgeon's expertise and intraoperative judgment remain decisive determinants of safety. This study has several limitations. First, watershed-guided resection requires access to contrast-enhanced CT, 3D reconstruction software, ICG, and a fluorescence-capable thoracoscopy system, which may limit applicability in resource-constrained settings. Second, long-standing PS can lead to recurrent infections, progressive fibrosis, and compression-related loss of function in adjacent lung tissue; in such cases, even with perfusion mapping, lobectomy may still be the most appropriate choice. Third, this was a single-center, real-world analysis with a relatively small sample size, and selection bias cannot be excluded because patients with more complex anatomy or extensive tissue destruction were more likely to undergo lobectomy rather than watershed-guided wedge resection. Consequently, observed differences in outcomes may be partly attributable to baseline differences in disease severity. Finally, although we observed better preservation of FEV1 at 6 months in the watershed group, our study was not designed or powered to comprehensively evaluate long-term functional outcomes or quality of life. Larger, preferably multicenter, prospective studies with extended follow-up are needed to validate these findings and to define formal selection criteria for watershed-guided wedge resection. In addition, patient-reported outcomes like exercise tolerance or functional status were not available, which limits interpretation of the clinical significance of the observed FEV1 differences. Despite these limitations, our results support arterial watershed analysis combined with ICG-guided fluorescence imaging as a promising adjunct in the surgical management of PS. By delineating the perfusion territory of the target systemic artery, this approach facilitates parenchyma-sparing wedge resection in appropriately selected patients and may improve perioperative outcomes and functional preservation. Future research should focus on standardizing technical protocols, refining patient selection criteria, and exploring broader applications of watershed-guided techniques to other complex congenital or acquired pulmonary vascular anomalies. Integration of this technology into routine clinical practice will also require cost-effectiveness analyses and assessment of learning curves across institutions. Conclusions Watershed analysis combined with 3D reconstruction and ICG fluorescence imaging provides a feasible strategy for surgical treatment of PS. CT-based 3D planning with watershed simulation may help anticipate perfusion territories and support wedge resection planning for PS, whereas intraoperative ICG fluorescence provides real-time boundary confirmation. Our study provides a preoperative simulation platform that can assist surgeons in predicting the boundaries between PS tissue and normal tissue, as well as complex vascular anatomy. This can help reduce lung function damage caused by excessive resection due to unclear lesion extent or prevent residual tissue infection or other lesions resulting from insufficient resection. In selected patients, watershed-guided wedge resection was associated with shorter operative time, reduced blood loss, fewer major complications, and better preservation of lung function compared with lobectomy. We observed no recurrent symptoms or imaging evidence of residual sequestration during follow-up and no delayed vascular rupture. These findings suggest that watershed-guided wedge resection is a safe and feasible option for selected patients with PS and may facilitate wider adoption of minimally invasive, parenchyma-sparing surgery for this rare congenital malformation. Further multicenter studies with larger cohorts and longer follow-up are warranted to confirm these results and to establish evidence-based guidance for the use of watershed analysis in PS surgery. Webcast You can watch a Webcast of this AATS meeting presentation by going to: Xxx. Conflict of Interest Statement Z. Guo and Q. Gao are affiliated with Guangdong OptoMedic Technologies, Inc, who developed the software and imaging system used in this study. All other authors reported no conflicts of interest. The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.