Feasibility, Safety, and Early Outcomes of Image‐Guided Segmentectomy Using Near‐Infrared Fluorescence Dye for Tumor Visualization and Margin Identification: A Collaborative Effort by the Surgical and Radiological Teams

ABSTRACT

Introduction

Despite advances in lung cancer management, it remains the leading cause of cancer‐related deaths. Low‐dose computed tomography (LDCT) screening has increased detection of small, difficult‐to‐palpate lung lesions.

Materials and Methods

This retrospective study at Chang Gung Memorial Hospital (2014–2022) evaluated the feasibility of image‐guided segmentectomy (I‐segmentectomy) using indocyanine green (ICG) for lesion localization and intersegmental plane navigation.

Results

A total of 260 patients with 266 pulmonary lesions were enrolled in the study cohort, with 122 lesions undergoing image‐guided segmentectomy (I‐segmentectomy). After propensity score matching, lesions resected using the I‐segmentectomy method provided appropriate resection margins and margin‐to‐tumor ratios, particularly for lesions larger than 1 cm. Additionally, operation times were shorter with I‐segmentectomy. Survival analysis showed no significant differences in disease‐free and overall survival; although I‐segmentectomy maintained a 100% survival rate.

Conclusion

Overall, I‐segmentectomy with dual ICG fluorescence imaging is a feasible, safe, and effective method for ensuring adequate resection margins in difficult‐to‐discern lung lesions. Further prospective studies are necessary to validate these findings and assess long‐term outcomes.

Image‐guided segmentectomy using near‐infrared fluorescence dye improves tumor localization and intersegmental plane identification. Compared with traditional segmentectomy, it achieves wider resection margins and higher margin‐to‐tumor ratios, especially for tumors > 1 cm, without increasing operative time. This technique may reduce local recurrence risk; further long‐term validation is needed.

1. Introduction

Despite improvements in the diagnosis, staging, and treatment of lung cancer, it remains the leading cause of cancer‐related deaths worldwide [1]. Low‐dose computed tomography screening (LDCT) is indeed effective in reducing lung cancer mortality [2]. However, through LDCT screening of high‐risk populations, an increasing number of smaller lung lesions are being detected [3]. Some of these lesions are located in the deeper lung tissues, while others primarily present as ground‐glass opacities. The emergence of lesions that are difficult to palpate or access with the naked eye and endoscope instruments has driven the evolution of localization methods and more precise resection methods in thoracic surgery. Various methods for tumor localization before or during thoracoscopic surgery have been proposed, [4, 5, 6] such as CT‐guided percutaneous assisted localization, transbronchial‐guided assisted localization, hybrid OR image‐guided assisted localization, and so on. Similarly, some surgeons adopt comprehensive preoperative preparation for these non‐palpable nodules, collaborating with specialized radiologists to confirm the lesion's location within a specific lung segment or CT reconstructions when there is a lack of reliable and accessible lesion localization support [7]. Segmentectomy is employed as one of the surgical approaches for lesion removal. Based on the data from the European Society of Thoracic Surgeons database, segmentectomy is more suitable for patients who have impaired respiratory function, higher ASA scores, and relevant comorbidities [8, 9]. Segmentectomy, as a fundamental anatomical resection of the lung, has been proven to be no less effective than lobectomy for malignant lung tumors smaller than two centimeters in size recently [10, 11]. However, it is worth noting that research on segmentectomy has shown a higher probability of postoperative local [10] recurrence. Given the challenges in uncertainty in marking and adequate resection margins in segmentectomy, there is an urgent need to develop a simple and minimally invasive procedure that can be performed at any healthcare facility. The purpose of this article is to discuss the feasibility of image‐guided segmentectomy (I‐segmentectomy), which involves the simultaneous application of indocyanine green dye for lesion localization and confirming the segmental plane in uniportal VATS segmentectomy, as well as its ability to provide appropriate resection margins in segmentectomy.

2. Materials and Methods

2.1. Study Populations

This was a retrospective study from January 2014 to January 2022 using data from the Chang Gung Memorial Hospital Linkou Branch. We included patients who underwent uniportal thoracoscopic lung surgery and selected those who received segmentectomy with and without ICG lesion localization and segmental plane navigation. Therefore, a nodule difficult to be easily located during naked‐eye surgery is defined as a nodule on preoperative CT scan, exhibiting (a) a nodule smaller than 1.5 cm with semi‐solid or solid characteristics, (b) a pure ground glass nodule smaller than 2 cm, or (c) a nodule smaller than 2 cm having a subpleural distance‐to‐diameter ratio equal to or greater than one. In the early stages, we used to discuss with radiologists the location of nodules within lung segments as part of the preoperative planning for segmentectomy. In addition, during this period, we adopted a judgment method for determining the intersegmental plane using the inflation and deflation technique. However, with the maturation and increased accessibility of localization techniques, the surgical team collaborates with specialized radiologists not only for preoperative planning but also for the localization of these small nodules since 2019. In addition to utilizing the inflation–deflation method, the determination of the intersegmental plane also involves the systematic injection of ICG to delineate the resection boundaries. During the study period, among 945 cases of lung anatomic resections, which included 318 cases of lung segmentectomy, excluding 33 cases where segmentectomy was performed due to tumors larger than 2 cm but with poorer physical condition, 285 lesions smaller than 2 cm on preoperative CT that were defined to be difficult to locate intraoperatively by naked eye underwent lung segmentectomy. Among the total of 285 lung segmentectomies, 144 were performed without preoperative localization, 19 underwent preoperative localization using methylene blue or hook wire, and 122 were performed using the I‐segmentectomy method (Figure 1).

FIGURE 1.

Schema for the study population. UVATS, uniportal video‐assisted thoracoscopic surgery.

2.2. Pre Operation Surgical Planning and Localization Techniques

Before surgery, each patient undergoes a comprehensive preoperative examination, including a chest CT scan. Surgeons and radiologists collaborate to identify the correct lung segment for the targeted nodules and assess variations in secondary bronchioles and feeding vessels. Indocyanine green (ICG) is used for CT‐guided localization (Diagnogreen; Daiichi‐Sankyo Co. Ltd., Tokyo, Japan). The patient's position—supine, prone, or semi‐decubitus—depends on the nodule's location on the CT image. Radiologists prefer a vertical needle path perpendicular to the pleural surface but will consider a tangential path if vital or bony structures block the vertical route. The localization process starts with an initial CT scan, followed by skin sterilization and local anesthesia using 5–10 mL of 2% lidocaine. A fine needle (21‐gauge, 15‐cm or 22‐gauge, 8.9‐cm) is inserted near the intercostal region. A second CT scan confirms the needle path. The needle is then advanced to the targeted lung parenchyma with additional lidocaine for nerve block until it reaches the nodule. ICG mixed with iohexol (Omnipaque 300; GE Healthcare, Cork, Ireland) is deployed, followed by another CT scan to verify the tattoo's location. A 0.3 mL ICG is injected at the deepest puncture site. If the intrapulmonary needle path is longer than 1 cm, additional ICG is injected during needle withdrawal at 1 cm intervals. Another 0.1 mL ICG is injected at the subpleural region for detection during surgery. A final CT scan confirms the staining location and checks for complications. After localization, radiologists communicate results with the surgical team. Depending on operating room availability, patients are either transferred directly to the surgical suite or return to the ward. Some patients opt for intraoperative localization using a Hybrid OR to avoid awake needle puncture. The localization approach involves a vertical entry perpendicular to the pleura, unless significant blood vessels obstruct the path. This method follows previously published protocols [6, 12], using the same ICG formula and tattoo technique (Figure 2A).

FIGURE 2.

Detailed process of image‐guided segmentectomy (I‐seg). (A) Difficult‐to‐palpate lung lesions treated with pre‐operative or intraoperative lesion localization with indocyanine green (ICG). (B) Illustration of Image‐guided segmentectomy (I‐seg) in the surgical field of view.

2.3. Surgical Techniques

After lesion localization, the surgery began with the patient positioned in a lateral decubitus posture. A single skin incision of 3 cm was made along the anterior axillary line at the 5th intercostal space, and a wound protector was applied at the incision site. A 10 mm, 30° Video S1 scope was used during surgery ((Karl Storz, Tuttlingen, Germany and Olympus, Tokyo, Japan)). At the beginning, the surgeon would use the near‐infrared fluorescence mode to observe the location of the target lesion. If dye dispersion occurred, the surgeon would use N/S 2000 mL for chest cavity irrigation to reduce the artifact effect of targeting dye dispersion. An endo‐stapler (Johnson & Johnson Institute, Cincinnati, Ohio and Medtronic, Minneapolis, MN, USA) was used to separate the bronchus, lung parenchyma, and associated pulmonary vessels; these were discussed among surgical team members and collaborating radiologists. When the targeted vessels and bronchus were resected sequentially, the surgeon turned on the near‐infrared fluorescence mode to observe the target lesion and its planned resected pulmonary segment. The anesthetic nurse injected indocyanine green (ICG), reconstituted in distilled water to a final concentration of 2.5 mg/mL, through the systemic circulation at a dose of 0.25 mg/kg, followed by a 10 mL flush of sterile normal saline. The surgeon then marked the segmental boundary with an electric hook and resected the marked area using staplers (Figure 2B). The resected specimen was retrieved with a plastic bag through the previously created incision wound. Post‐operation, an air leak test was practiced for all patients. If an air leak was found, primary suture repair or a PGE patch would be applied for the lung parenchyma injury area. Finally, one 16 Fr chest tube was left for drainage.

3. Data Collection and Statistical Analysis

Clinical charts were retrospectively reviewed. Descriptive statistics for continuous variables were expressed as mean ± standard deviation (SD) and categorical variables as numbers (percentages). Categorical variables were summarized as numbers and percentages and compared using the χ test or Fisher's exact test. Continuous variables were summarized as means and standard deviations and compared using Student's t or Wilcoxon rank‐sum tests when appropriate. To ensure more accurate comparisons, we used age, gender, tumor size, FEV1, and diagnosis as parameters for 1:1 matching of patients who underwent I‐segmentectomy with those who did not, using SPSS version 27 (IBM, USA). Survival plots were done in Python 3.12 with matplotlib and the lifelines package.

4. Results

In this study, we excluded patients who underwent lesion localization using methylene blue and hook wire, and only compared cases with or without undergoing the I‐segmentectomy method. Among 260 patients, there were 266 pulmonary lesions undergoing segmentectomy, with females comprising the majority at 54.62% of the overall study population (Table 1). The average age was 61.05 ± 11.63. Out of the 266 lung segmentectomy procedures, 122 lesions were treated using image‐guided segmentectomy.

TABLE 1.

Clinicopathologic characteristics of difficult‐to‐palpate lesions patients receiving segmentectomy with/without image‐guided segmentectomy (I‐seg).

Patient baseline (n = 260)
Variables		Variables
Age	61.05 ± 11.63	Body mass index	24.50 ± 3.37
Gender		Smoking
Male	118 (45.38%)	Quit	28 (10.77%)
Female	142 (54.62%)	Active	35 (13.46%)
		No	197 (75.77%)
ECOG		Previous malignancy history
0	180 (69.23%)	Yes	103 (39.62%)
1	80 (30.77%)	No	157 (60.38%)
ACS history		Family history of lung cancer
Yes	29 (11.15%)	Yes	19 (7.31%)
No	231 (88.85%)	No	241 (92.69%)
COPD		FEV1 (l)	2.18 ± 0.68
Yes	30 (11.54%)
No	230 (88.46%)
Nodule characteristic (n = 266)
CT nodule size (cm)	1.30 ± 0.52	CT pleural to nodule distance (cm)	1.13 ± 0.85
Nodule location		Nodule diagnosis
RUL	56 (21.10%)	Primary lung cancer	189 (71.05%)
RML	21 (7.90%)	Metastatic malignancy	31 (11.65%)
RLL	71 (26.70%)	Benign lesion	46 (17.30%)
LUL	73 (27.40%)
LLL	45 (16.90%)

Abbreviations: ACS, acute coronary syndrome; COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume in 1 s; LLL, left lower lobe; LUL, left upper lobe; RLL, right lower lobe; RML, right middle lobe; RUL, right upper lobe.

4.1. Outcome of I‐Segmentectomy

In 122 attempted I‐segmentectomies, there were three instances of dye dispersion. Although these did not significantly affect the surgical procedure and were managed by extensive irrigation of the thoracic cavity with normal saline, the image quality was relatively less clear compared to lesions without dye dispersion. In the remaining 119 I‐segmentectomies (97.5%), which consisted of two cases, we discovered intraoperatively that due to arterial variations, the resected segment area needed to be extended beyond what was initially planned preoperatively. Otherwise, other lesions with I‐segmentectomy were visible and believed to be consistent with the preoperative assessment. Apart from this, one pre‐operation localization patient suffered from subclinical pneumothorax without any further clinical management.

4.2. Comparative Results of Lesion With/Without I‐Segmentectomy

All surgical procedures were completed using uniportal VATS. In the initial analysis of the entire study cohort, lesions that did not undergo I‐segmentectomy group had longer operation times, larger lesion sizes on CT, and on pathological examination. Lesions that underwent I‐segmentectomy group had a larger ratio of the nearest resection margin to tumor size. Detailed characteristics were described in Table 2. To further analyze the impact of I‐segmentectomy on resection margins, the ratio of resection margin to tumor size, the entire study cohort was divided into two groups with 1:1 propensity score matching. Eighty‐six paired lesions were enrolled in the final analysis (Table 2). In subsequent analysis, we found that patients who underwent I‐segmentectomy had shorter operation times and larger resection margins. Although the ratio of resection margin to tumor diameter was larger, it did not reach statistical significance. Other factors such as hospital stay, duration of chest tube placement, and surgery‐related complications showed no significant differences. Given that different tumor sizes directly impact the difficulty of achieving sufficient resection margins and the resection margin to tumor diameter ratio, we further subdivided the matched lesions into two subgroups: those with tumors smaller than or equal to 1 cm and those with tumors larger than 1 cm (Table 3). In the subgroup with tumors smaller than 1 cm, there were no significant differences in resection margin distance or the ratio of resection margin to tumor diameter, regardless of whether I‐segmentectomy was used. Conversely, in the subgroup with tumors larger than 1 cm, there were significant differences in operation time and resection margin.

TABLE 2.

Comparative clinical parameters before and after propensity score matching for lesions resection with and without image‐ guided segmentectomy (I‐seg).

Variable	Entire cohort			Propensity score matching
	Seg group (n = 144)	I‐seg group (n = 122)	p	Seg group (n = 86)	I‐seg group (n = 86)	p
Age	61.24 ± 12.25	60.83 ± 10.93	0.77	61.48 ± 12.34	61.39 ± 10.70	0.96
Body mass index	25.03 ± 4.31	23.88 ± 3.46	0.02	25.01 ± 4.26	24.01 ± 3.47	0.09
Pre‐operative parameters
Diagnosis			0.076			1
Malignancy	112	108		11	11
Benign	32	14		75	75
Lesion localization
RUL	27	29		14	19
RML	6	15		2	9
RLL	41	30		25	23
LUL	45	28		26	19
LLL	25	20		19	16
CT tumor size (cm)	1.47 ± 0.51	1.14 ± 0.52	< 0.01	1.32 ± 0.42	1.28 ± 0.54	0.57
CT pleural to tumor distance (cm)	1.05 ± 0.85	1.22 ± 0.84	0.11	1.09 ± 0.82	1.19 ± 0.78	0.39
FEV1 (l)	2.16 ± 0.71	2.20 ± 0.64	0.62	2.10 ± 0.72	2.15 ± 0.63	0.60
Peri‐operative and post‐operative parameters
Post OP FEV1 (l)	1.76 ± 0.61	1.91 ± 0.76	0.44	1.65 ± 0.53	1.78 ± 0.61	0.52
Operative time (min)	144.17 ± 47.91	130.29 ± 40.83	0.01	147.72 ± 46.90	130.54 ± 42.82	0.01
Blood loss (mL)	39.60 ± 54.10	32.36 ± 17.41	0.16	42.87 ± 67.59	31.93 ± 17.73	0.15
Drainage duration (days)	2.74 ± 1.58	2.58 ± 1.26	0.37	2.80 ± 1.47	2.62 ± 1.34	0.39
Post OP stay (days)	4.11 ± 3.05	3.74 ± 1.59	0.23	3.83 ± 1.53	3.79 ± 1.72	0.89
Pathological tumor size (cm)	1.54 ± 0.63	1.15 ± 0.55	< 0.01	1.32 ± 0.52	1.31 ± 0.55	0.93
Resection margin (cm)	1.61 ± 1.65	1.90 ± 1.64	0.16	1.41 ± 0.98	1.89 ± 1.59	0.02
Margin to tumor ratio	1.25 ± 1.28	2.06 ± 2.06	< 0.01	1.30 ± 1.13	1.72 ± 1.82	0.07
Post OP complication			0.26			0.62
Yes	15	8		10	8
No	129	114		76	78

Abbreviations: I‐seg, image‐guided segmentectomy; CT, computed tomography; OP, operation; FEV1, forced expiratory volume in 1 s; RUL, right upper lobe; RML, right middle lobe; RLL, right lower lobe; LUL, left upper lobe; LLL, left lower lobe.

TABLE 3.

Subgroup analysis for CT tumor size ≤ 1 cm and > 1 cm after propensity score matching.

Variable	Tumor ≤ 1 cm			Tumor > 1 cm
	Seg group (n = 29)	I‐seg group (n = 33)	p	Seg group (n = 57)	I‐seg group (n = 53)	p
Pre‐operative parameters
Diagnosis			0.60			0.63
Malignancy	25	27		7	5
Benign	4	6		50	48
Lesion localization
RUL	1	6		13	13
RML	1	6		1	3
RLL	8	4		17	19
LUL	10	7		16	12
LLL	9	10		10	6
CT tumor size (cm)	1.06 ± 0.44	0.87 ± 0.32	0.06	1.32 ± 0.42	1.28 ± 0.54	0.34
CT pleural to tumor distance (cm)	1.05 ± 0.68	1.25 ± 0.74	0.28	1.10 ± 0.88	1.15 ± 0.82	0.76
Depth to tumor size (cm)	0.83 ± 0.73	1.29 ± 1.04	< 0.01	0.92 ± 0.77	1.12 ± 0.92	0.14
FEV1 (l)	2.14 ± 0.64	2.14 ± 0.67	0.95	2.07 ± 0.76	2.15 ± 0.61	0.50
Peri‐operative and post‐operative parameters
Post OP FEV1 (l)	1.74 ± 0.63	1.60 ± 0.62	0.71	1.65 ± 0.53	1.78 ± 0.61	0.17
Operative time (min)	136.69 ± 47.91	142.21 ± 45.65	0.61	153.33 ± 50.43	123.35 ± 39.74	< 0.001
Blood loss (mL)	29.26 ± 13.28	32.12 ± 16.64	0.42	49.55 ± 67.59	31.93 ± 17.73	0.13
Drainage duration (days)	2.86 ± 1.51	2.67 ± 1.41	0.60	2.77 ± 1.46	2.58 ± 1.31	0.19
Post OP stay (days)	3.86 ± 1.64	3.91 ± 1.86	0.92	3.81 ± 1.48	3.72 ± 1.65	0.09
Pathological tumor size (cm)	0.78 ± 0.17	0.77 ± 0.15	0.88	1.59 ± 0.41	1.65 ± 0.42	0.49
Resection margin (cm)	1.64 ± 1.21	2.17 ± 2.31	0.27	1.30 ± 0.83	1.71 ± 0.90	0.01
Margin to tumor ratio	2.14 ± 1.42	2.73 ± 2.56	0.27	1.43 ± 1.87	1.93 ± 2.18	0.05

Abbreviations: CT, computed tomography; FEV1, forced expiratory volume in 1 s; I‐seg, image guided segmentectomy; LLL, left lower lobe; LUL, left upper lobe; OP, operation; RLL, right lower lobe; RML, right middle lobe; RUL, right upper lobe.

4.3. Disease Free Survival and Overall Survival of Lesions With/Without I‐Segmentectomy

Finally, we also analyzed the short‐term disease‐free survival and overall survival of patients with primary lung cancer, comparing those who underwent I‐segmentectomy to those who did not. After matching, the median follow‐up time of the matched cohort was 30 months. In terms of disease‐free survival and overall survival, there were no statistically significant differences between the two groups. However, the I‐segmentectomy group maintained 100% disease‐free survival and overall survival throughout the follow‐up period up to December 2023 (p = 0.196, p = 0.091) (Figure 3A,B).

FIGURE 3.

Disease free survival (DFS) and overall survival (OS) for patients receiving segmentectomy with/without image‐guided segmentectomy.

5. Discussion

Segmentectomy has proven effective in treating early‐stage lung cancer patients [10]. However, one potential drawback is a higher rate of local recurrence, prompting a need for more precise segmentectomy techniques. Little discussion has focused on image‐guided segmentectomy (I‐seg) using near‐infrared fluorescent dyes [13, 14], which serve dual purposes of lesion localization and segmental margin confirmation. This method allows surgeons to visualize the surgical margin distance, reducing concerns about inadequate tumor resection margins without significantly increasing surgical equipment or time. Consequently, fluorescent dye guidance aids surgeons in performing appropriate resections, potentially improving outcomes by facilitating the precise removal of lesions. Collaboration with radiologists before surgery further ensures precision, helping avoid insufficient resections that could increase recurrence risk and excessive resections that cause unnecessary lung tissue loss. Given that lung tissue lacks regenerative capacity, precise treatment methods are essential, especially as lung cancer patients become younger. In this retrospective analysis, using dual ICG fluorescence imaging in the I‐segmentectomy group provided appropriate resection margins and margin‐to‐tumor ratios, particularly for lesions larger than 1 cm. Achieving adequate margins is generally easier for smaller tumors, and the shortened operation times observed warrant further discussion. One possibility is that patients who did not undergo I‐segmentectomy were treated earlier in the study period, during the initial development of surgical techniques, potentially influencing results. Another possibility is that I‐segmentectomy enabled surgeons to more accurately identify and resect small lesions intraoperatively, eliminating the need to extend the resection area due to perceived insufficient margins after specimen removal. This finding requires further evaluation and validation.

Patients undergoing I‐segmentectomy demonstrated a higher average margin‐to‐tumor diameter ratio compared to those who did not receive this technique. This difference may reflect more deliberate perioperative planning and execution, as well as reduced intraoperative uncertainty. By combining preoperative localization with intraoperative near‐infrared guidance, I‐segmentectomy enables surgeons to plan resection margins in advance and execute these plans with greater precision. Additionally, the slight spread of fluorescence into the surrounding lung parenchyma over time can extend the final oncologic margin beyond what was initially anticipated at the staple line, further contributing to an increased margin‐to‐tumor ratio.

Addressing the issue of fluorescent dye dispersion is crucial. Using normal saline for irrigation in a disrupted environment can reduce the disruption caused by dye dispersion during localization. Furthermore, feedback from the surgical team to the imaging localization team helps maintain high‐quality surgery. With this mechanism, the probability of fluorescent dye overflow is 2.5%; yet, the lesion location can still be identified during surgery, and surgical margins remain tumor‐free. In a survival analysis of early‐stage lung cancer patients, a lower probability of disease recurrence was observed, potentially due to more appropriate resection margins. However, further validation in subsequent follow‐up studies is necessary due to the relatively short follow‐up time. In laparoscopic liver resection for HCC, the combined use of fluorescence images obtained by preoperative and intraoperative ICG administration has proven useful as a real‐time navigation tool [15]. Applying similar concepts in precise lung segmentectomy ensures feasibility and safety.

This study has several limitations. First, in this retrospective analysis, the GGO (ground‐glass opacity) pattern was not included in the overall analysis because it is not yet incorporated into the IASLC guidelines for staging [16]. Additionally, variations in CT slice thickness and the use of different CT machines during the nine‐year enrollment period may have affected the appearance of GGO or the consolidation‐to‐tumor ratio, impacting interpretation [17]. The lack of consensus on the definition and classification of GGO further limits the precision and comparability of this factor across studies. Due to the influence of low‐dose CT screening, the size of detected nodules varied over the study period; therefore, tumor size was used as the primary focus of our analysis. Second, the continuous evolution of surgical techniques and perioperative workflows during the long enrollment period may have influenced outcomes, as this study relies on historical data that may not fully reflect recent advances in surgical approaches or postoperative care. Third, although techniques were aligned between the localization and surgical teams, individual differences in surgeon experience and intraoperative decision‐making are inevitable in non‐single‐operator studies. Moreover, the follow‐up time for patients undergoing I‐segmentectomy was relatively short, which may not be sufficient for comprehensive survival analysis, especially for assessing local recurrence in larger tumors. Notably, our observed local recurrence rate was lower than reported in landmark trials such as JCOG0802 [10], which found a local recurrence rate of approximately 5% at 3 years, highlighting the need for longer follow‐up to verify whether the margin adequacy achieved in our cohort translates into durable oncologic benefit. Despite these limitations, our short‐term findings indicate that patients in the I‐segmentectomy group achieved more appropriate margin distances, which may help reduce the probability of local recurrence. Whether this advantage persists in the long term will require confirmation through future prospective studies with standardized protocols, consistent imaging parameters, and extended follow‐up.

6. Conclusion

Effective communication between the surgical and localization teams has demonstrated that using dual ICG fluorescence imaging in I‐segmentectomy for lesions difficult to discern with the naked eye in the lungs is a feasible, safe, and effective method that provides adequate margin distances.

Author Contributions

Ching Feng Wu: concepturation, data collection, orginal draft writing, formal analysis, revision. Kuei An Chen: concepturation, data collection, revision. Ming Ju Hsieh: data collection, revision. Yu Fu Wu: formal analysis, revision. Tzu Yi Yang: data collection. Ching Yang Wu: concepturation, data collection, orginal draft writing, Revision.

Ethics Statement

The retrospective study was approved by Chang Gung Memorial Hospital Clinical Research Ethical Committee with the decision number (202400105B0).

Consent

Since this study is a retrospective study, informed consent is waived.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Video S1. Segmentectomy was performed using a double ICG navigation technique to ensure adequate margins and precision, achieved through close collaboration between the radiology and surgical teams.

Wu C. F., Chen K. A., Hsieh M. J., Wu Y. F., Yang T. Y., and Wu C. Y., “Feasibility, Safety, and Early Outcomes of Image‐Guided Segmentectomy Using Near‐Infrared Fluorescence Dye for Tumor Visualization and Margin Identification: A Collaborative Effort by the Surgical and Radiological Teams,” Thoracic Cancer 16, no. 15 (2025): e70139, 10.1111/1759-7714.70139.

Data Availability Statement

The data supporting the findings of this study are available upon request from the corresponding author. The data are not publicly available due to privacy and ethical restrictions.