Ultrasound-Guided, Indocyanine Green-Directed Robot-Assisted Breast Surgery for Benign Tumours

By Nanqiu Liu, Guochun Zhang, Chongyang Ren, Jiyuan Cao, Ning Liao

Affiliations

1. Department of Breast Surgery, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Southern Medical University, Guangzhou, China

doi: 10.29271/jcpsp.2026.08.996

ABSTRACT
Objective: To compare the feasibility, safety, and cosmetic satisfaction of ultrasound-guided, indocyanine green (ICG)-assisted robotic lumpectomy with open lumpectomy for benign breast tumours.
Study Design: A retrospective cohort study.
Place and Duration of the Study: Department of Breast Surgery, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Southern Medical University, Guangzhou, China, from June to September 2023.
Methodology: A step-by-step protocol was developed for surgeons to guide ultrasound-guided ICG injections for robotic lumpectomy. Clinicopathological characteristics of the patients, intraoperative data, postoperative complications, and aesthetic outcomes were analysed. Continuous variables were compared using an independent-samples t-test if they were normally distributed; otherwise, the Mann–Whitney U test was used. Categorical variables were compared using a chi-square test or a corrected chi-square test.
Results: A total of 68 patients treated concurrently were included in the study; of whom, 30 underwent a robot-assisted local tumour excision through an axillary incision, while 38 had their tumour excision through a standard open incision over the breast. Operation time was significantly longer in the robotic group (96.0 vs. 42.5 minutes; p <0.001). No significant intergroup difference was found in postoperative complications. Patients who underwent robotic surgery achieved significantly higher rates of the highest level of satisfaction with the wound site (96.7% vs. 52.6%; p <0.001), scar appearance (93.3% vs. 71.1%; p = 0.020), and overall breast look (86.7% vs. 60.6%, p = 0.017).
Conclusion: This surgical approach, using ICG injection and ultrasound guidance, provides a potentially feasible and safe method for tumour localisation during robotic lumpectomy.

Key Words: Aesthetic outcome, Breast surgery, Indocyanine green, Robot-assisted surgery, Ultrasound-guidance.

INTRODUCTION

Robot-assisted surgery is being employed for nipple-sparing mastectomy (R-NSM) and breast reconstruction procedures, particularly in Asia and Europe.^1,2 However, some experts have questioned the justification for using this technology, parti- cularly regarding the increased operative time and cost. Its use in the United States is largely limited to clinical trials.³ Nevertheless, there are definite advantages to using robotic instruments compared with endoscopic instruments for minimally  invasive  breast  surgery.
 

Robot-assisted surgery provides a ten-fold magnification imaging system in a 3D visual field using a flexible arm with unique joint movements and a comfortable ergonomic design. In addition, robot-assisted breast surgery through an axillary incision yields greater patient satisfaction, greater skin and nipple  sensation,  and  a  superior  aesthetic  outcome.⁴

Multiple studies have verified the feasibility and safety of robot-assisted total mastectomy, even suggesting that the compli- cation rate is the same or even lower than conventional (open) mastectomy at the expense of longer operating time.⁵ R-NSM was associated with a lower rate of postoperative complications (e.g. nipple-areolar complex necrosis and poor wound healing).⁶ Other surgical approaches, such as robot-assisted axillary lymph node dissection and latissimus dorsi flap procedure, have also been reported, further enlarging the application  of  robotic  technology  in  breast  surgery.⁷

However, there are no reports using robotic surgery for local excisions of benign breast tumours through an axillary incision. This surgical limitation is likely because robot-assisted breast surgery cannot reliably identify the margins of a breast tumour without  tactile  sensation.

Thus, there is an unmet clinical need for robot-assisted surgery for the local excision of larger tumours (e.g., those larger than 2 cm), particularly lesions located deep within the breast or in the medial quadrants, as well as for reliable intraoperative tumour identification and complete excision.

Indocyanine green (ICG) has been used in other anatomical areas for localising tumours or specific anatomical structures during robot-assisted surgery for liver resections, gastric cancer lymph node dissection, and intraoperative vessel and lymphatic imaging.^8,9 For example, real-time ICG fluorescence imaging during a robotic-assisted right hepatectomy reduced postoperative bile leakage without prolonging operative time.¹⁰ A prospective non-randomised cohort study demonstrated that the mean number of harvested lymph nodes was significantly higher using the Firefly system- assisted ICG tracer during a radical robotic distal gastrec- tomy.¹¹ ICG is a safe near-infrared imaging agent that facilitates visual localisation of breast tumours during retromammary robot-assisted breast surgery, thereby enabling safe and complete local excision.

It was hypothesised that a robotic surgery excision approach through a small axillary incision could be successful when guided to the tumour by ICG mapping, correctly identify the tumour and its border, and yield good aesthetic results. This study aimed to investigate robot-assisted local excision of benign breast tumours using ultrasound-guided ICG injections and to evaluate the feasibility and safety of this sur- gical approach.

METHODOLOGY

This was a descriptive study in which data were collected concurrently from all patients undergoing local excision for benign breast tumours, including both robot-assisted and conventional open surgery at the Department of Breast Surgery, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Southern Medical University, Guangzhou, China, from June to September 2023. Informed consent for the use of clinical records was obtained from the included participants. The study was approved by the Ethics Committee of Guangdong Provincial People’s Hospital (Permission No.KY2023-731-02).

Patients aged ≥18 years with benign breast lesions (fibroadenoma or intraductal papilloma) confirmed by core needle biopsy, who underwent preoperative tumour measurement by breast ultrasound or MRI, were included. All patients provided informed consent for robotic or conventional open lumpectomy alongside regular postoperative follow-up. Patients allergic to iodine or ICG were excluded.

The data were collected according to a pre-designed protocol, including age, body mass index (BMI), menstrual status, tumour size, circumferences of bilateral upper arms, intraoperative blood loss, operation time, complications (wound infection, skin or nipple necrosis) and patient-reported aesthetic outcomes. Operating time was collected as: total operation time, pre-docking time, docking time, and console time. The pre-docking time was defined as the time from the start of skin incision to the completion of the establishment of operating space in the axilla. Docking time was defined as the interval from the start of glove-port creation to completion of robotic system setup. Console time was defined as the time from the start of the dissection of the skin flap to the removal of the tumour specimen. A standardised questionnaire was conducted for patients’ cosmetic satisfaction.¹²

The patients undergoing robotic surgery were placed in a semi-lateral position so that the breast would fall away from the chest wall, thus increasing the retromammary space between the breast tissue and the pectoralis muscle. The ipsilateral arm was wrapped in a sterile drape and attached to a support stand (Figure 1A). The da Vinci Xi platform (Intuitive Surgical, Sunnyvale, CA, USA), equipped with the Firefly imaging system, was used for all robot-assisted breast surgeries.

The first assistant surgeon stood on the operative side of the patient, the second assistant surgeon on the contralateral side, and the primary surgeon at the console. An oblique 4 cm incision was made with an electrotome along the midaxillary line of the axilla. The skin and subcutaneous tissue were dissected medially to the lateral edge of the pectoralis major muscle. The dissection continued until an operating space was created in the retromammary space between the breast tissue and the pectoralis fascia. A 60 mm flap protector (Nuo De Medical, China) was inserted into the incision, and a sterile rubber glove was used to make the glove port. The periphery of the flap protector was covered by the wrist of the sterile rubber glove and secured with a suture. The thumb, middle finger, and little finger of the glove were selected, and a 1 cm longitudinal incision was made at the fingertip with scissors. Three 8 mm trocars (Intuitive Surgical, Sunnyvale, CA, USA) were placed and fixed through the fingertip incision, respectively (Figure 1B).

During docking, the patient cart was placed posterior to the patient. The first assistant operated the patient cart and rotated the robotic arms to the opposite side of the patient by the overhead boom, freeing up operating space inside the surgical field. Docking was facilitated by a laser positioning system to aid alignment with the surgical site. A 30° camera (Intuitive Surgical, Sunnyvale, CA, USA) was used through an 8 mm trocar, and a monopolar curved scissors and bipolar forceps (Intuitive Surgical, Sunnyvale, CA, USA) were inserted through an 8 mm trocar, respectively (Figure 1C, D).

Using the flexibility of the glove port, the surgeon could adjust the position of the robotic arms within the port and create a larger working space by increasing the distance between the arms, reducing the risk of instrument collisions. The camera trocar was placed slightly higher than the trocars used for the scissors and forceps to minimise collisions. During console operation, the assistant at the table could adjust the position of the robotic arms when necessary.

Figure 1: Operative images for surgical techniques in robotic-assisted lumpectomy. (A) A patient was placed in the lateral position with the arm on the operating side attached to the support stand. (B) Glove port with trocars. Trocars 1, 2, and 3 were used for insertion of monopolar curved scissors, camera, and bipolar forceps, respectively. (C) Completion of gas insufflation. (D) Completion of the insertion of robotic arms. (E, F) Injection of the ICG guided by ultrasound examination was performed by a well-trained surgeon. (G) Intraoperative tumour imaging in the white light model in robotic-assisted lumpectomy. (H) Intraoperative tumour imaging in Firefly mode during robot-assisted lumpectomy demonstrated the tumour margin (red dotted line), thereby facilitating precise surgical dissection. (I) Tumour specimen in white light model. (J) Tumour specimen in the Firefly model.

The port with the trocar was positioned in the retromammary space before the injection to minimise the time between ICG injection and the surgical excision. Carbon dioxide (CO₂) insufflation was used to maintain an insufflation pressure of 10 mmHg.

The exact location of the tumour was confirmed by intraoperative breast ultrasound. ICG (25 mg vials) was prepared as a 0.625 mg/mL solution in sterile water. A 1 mL percutaneous intratumoral injection of the solution was administered using a 1 mL syringe under ultrasound guidance by a well-trained surgeon. The ICG was carefully injected into the posterior and central portions of the tumour, allowing diffusion through-out the lesion (Figure 1E, F).

The surgery was switched to the da Vinci Xi platform when docking was completed. Bipolar forceps were used for traction to maintain exposure. A retromammary dissection was carried out using monopolar curved scissors, beginning at the lateral edge of the pectoralis major muscle and dissecting medially along the pectoralis fascia to establish an operating space. The camera was inserted in a plane behind the breast through an axillary incision.

The tumour location and posterior margins were visible in the Firefly model within 3 to 5 minutes of the injection, guiding the surgeon to accurately identify the tumour margins and precisely resect the tumour (Figure 1G-J). The specimen was then extracted through the axillary incision. The wound was sutured after proper haemostasis. A drainage tube was placed through a 1cm incision 2cm below the inframammary fold along the lateral edge of the pectoralis major muscle.

For conventional breast lumpectomy, patients were placed in a supine position with the ipsilateral upper arm in an abducted position. A radial incision was made overlying the tumour or through a peri-areolar incision. The incision length was determined according to tumour size. The breast tumour and peritumoral tissue were excised using an electrotome. A drainage tube was placed 5 cm below the breast incision, haemostasis was completed, and the incision was sutured closed.

The patients' reported aesthetic outcome was surveyed using a validated questionnaire¹² with 9 standardised questions, using 4 scales for each answer, graded as 1 = unsatisfied, 2 = fair, 3 = satisfied, and 4 = very satisfied.

The Kolmogorov–Smirnov test was used to assess normality. Continuous variables were presented as mean ± standard deviation (SD) when normally distributed; otherwise, they were presented as median (interquartile range [IQR]). Diffe- rences between groups were assessed using an independent- sample t-test; if variables were not normally distributed, the Mann-Whitney test was applied. Categorical variables were presented as frequency (percentage), and differences between groups were assessed using the chi-square test.

Figure 2: Comparison of the operating time for the first ten patients with that of the last ten patients. (A) Operation time with case accumulation. (B-E) Comparison of operation time in two robot-assisted surgery groups. (B) Total operation time of first 10 patients vs. last 10 patients (125.0 min vs. 88.0 min; p <0.001); (C) Pre-docking time (27.9 min vs. 13.5 min; p = 0.002); (D) Docking time (13.3 min vs. 10.0 min; p = 0.123); (E) Console time (47.7 min vs. 22.2 min; p <0.001).

A p-value <0.05 was considered statistically significant, and all tests were two-tailed. All statistical analyses were performed with the use of the SPSS statistical package (Version 27.0).

RESULTS

Sixty-eight patients were included in this study, thirty of whom underwent a robot-assisted tumour excision, and thirty- eight had their tumour excision through a standard open incision (Table I). The average tumour size in the robot-assisted lumpectomy group and the conventional open lumpectomy group was comparable (p >0.05). All but 3 patients were premenopausal. Of the 30 patients in the robot-assisted breast surgery group, 28 had breast fibroadenoma, and 1 had a benign phyllodes tumour and intraductal papilloma, respectively. Of the 38 patients in the open lumpectomy group, 24 had breast fibroadenoma, and 8 had intraductal papilloma. All benign tumours were completely excised, and the surrounding capsule remained intact in all procedures. No patients experienced the ICG-related adverse reactions.

There was no significant difference in intraoperative blood loss between the two groups. There was no significant diffe- rence in the incidence of postoperative complications. Two patients in the robot-assisted surgery group and 3 in the open surgery group developed a mild wound infection that healed spontaneously without antibiotics. None of the patients who underwent robot-assisted surgery showed motor or sensory abnormalities of the upper extremity after surgery. The length of incision was significantly shorter in the robot-assisted surgery group compared to the open surgery group (4.0 cm vs. 5.0 cm; p <0.001).

Table I: Clinicopathologic characteristics and clinical outcomes of included patients.
 

Variables	Robot-assisted surgery group	Conventionally open surgery group	p-values
Number of patients	30	38
Age (years)	31.8 ± 10.8	34.6 ± 9.3	0.267
BMI (kg/m2)	20.9 ± 2.7	21.1 ± 3.3	0.728
Tumour laterality (%)			0.695
Left	18 (60.0)	21 (55.3)
Right	12 (40.0)	17 (44.7)
Tumour size (cm)	2.7 ± 1.1	2.2 ± 1.1	0.076
Menstrual status (%)			>0.99
Pre-menopause	29 (96.7)	36 (94.7)
Post-menopause	1 (3.3)	2 (5.3)
Pathology (%)
Breast fibroadenoma	28 (93.4)	24 (63.2)
Intraductal papilloma	1 (3.3)	8 (21.0)
Breast adenosis	0 (0.0)	2 (5.3)
Phyllodes tumour	1 (3.3)	4 (10.5)
Intraoperative blood loss (mL)	5.0 (5.0~5.0)	5.0 (2.0~5.0)	0.281
Total operation time (min)	96.0 (89.7~120.0)	42.5 (35.0~50.0)	<.001
Length of wound (cm)	4.0 (3.5~4.0)	5.0 (4.0~5.0)	<0.001
Complications	2	3	>0.99
Infection	2	3
Skin necrosis	0	0
Nipple or areola necrosis	0	0
Seroma	0	0
Movement disorder of the upper extremities	0	0
Continuous variables are presented as mean ± standard deviation (SD) or median (interquartile range [IQR]), while categorical variables are presented as n (%). The independent-samples t-test, Mann–Whitney U test, and chi-square test were applied.

Table II: Patient-reported aesthetic outcomes in the two groups.

	Robot-assisted group				Open surgery group				p-values
	Fair (%)	Satisfied (%)	Very satisfied (%)	Mean score	Fair (%)	Satisfied (%)	Very satisfied (%)	Mean score
Q1.  Surgical wound position satisfaction	0 (0.0)	1 (3.3)	29 (96.7)	3.97 ± 0.18	3 (7.9)	15 (39.5)	20 (52.6)	3.45 ± 0.64	<0.001
Q2.  Postoperative breast appearance        satisfaction – no dressing	0 (0.0)	4 (13.3)	26 (86.7)	3.87 ± 0.34	0 (0.0)	15 (39.4)	23 (60.6)	3.61 ± 0.49	0.017
Q3.  Scar appearance satisfaction	0 (0.0)	2 (6.7)	28 (93.3)	3.93 ± 0.25	0 (0.0)	11 (28.9)	27 (71.1)	3.71 ± 0.46	0.020
Q4.  Scar length satisfaction	0 (0.0)	3 (10.0)	27 (90.0)	3.90 ± 0.30	3 (7.9)	7 (18.4)	28 (73.7)	3.66 ± 0.62	0.087
Q5.  Postoperative bilateral breast         symmetry satisfaction	0 (0.0)	4 (13.3)	26 (86.7)	3.87 ± 0.34	0 (0.0)	11 (28.9)	27 (71.1)	3.71 ± 0.46	0.123
Q6.  Postoperative breast appearance        satisfaction – with dressing	0 (0.0)	5 (16.7)	25 (83.3)	3.83 ± 0.37	0 (0.0)	12 (31.6)	26 (68.4)	3.68 ± 0.47	0.159
Q7.  Postoperative nipple-areola position s        atisfaction	0 (0.0)	6 (20.0)	24 (80.0)	3.80 ± 0.40	0 (0.0)	6 (15.8)	32 (84.2)	3.84 ± 0.37	0.651
Q8.  Postoperative bilateral breast size        satisfaction	0 (0.0)	5 (16.7)	25 (83.3)	3.83 ± 0.37	0 (0.0)	11 (28.9)	27 (71.1)	3.71 ± 0.46	0.236
Q9.  Preoperative breast appearance        satisfaction	0 (0.0)	6 (20.0)	24 (80.0)	3.80 ± 0.40	0 (0.0)	8 (21.1)	30 (78.9)	3.79 ± 0.41	0.915
Categorical variables are presented as n (%), and chi-square tests were applied.

Patients who underwent robot-assisted surgery were significantly more satisfied with the aesthetic outcome than those who underwent open surgery. Thus, when the postoperative results of a concurrent group of 30 patients having robot- assisted surgery were compared to 38 patients having an open surgical excision of benign breast tumours, significantly more patients in the robot-assisted group were very satisfied with their surgical wound position (96.7% vs. 52.6%; p <0.001), with their scar appearance (93.3% vs. 71.1%; p = 0.020), and with their overall breast appearance (86.7% vs. 60.6%; p = 0.017, Table II).

One disadvantage of robot-assisted surgery is the total operating room time, with the cost implication. Overall, the total operation time of the robot-assisted group was significantly longer than that of the conventionally open group (96.0 min vs. 42.5 min, p <0.001; Table I). However, the total operation time decreased as case experience accumulated over the 4 months of the study, especially console time, which was generally <25 min (Figure 2A). With increasing surgical experience, total operative time decreased by 37 minutes when the first ten patients were compared with the last ten patients. Thus, there was a statistically significant improvement in total operating time (125.0 min vs. 88.0 min; p <0.001), pre-docking time (27.9 min vs. 13.5 min; p = 0.002) and console time (47.7 min vs. 22.2 min; p <0.001, Figure 2B-E), while there was no statistically significant difference in docking time (13.3 min vs. 10.0 min; p = 0.123).

DISCUSSION

Fluorescence imaging has received considerable attention as a means of intraoperative navigation. Optical fluorescence imaging can be used for tumour localisation in real-time during surgery with greater sensitivity than direct visual inspection and palpation.¹³ ICG, as a tracer, is approved and usually used for tumour localisation, lymph node tracing, and blood supply assessment through preoperative intravenous injection.¹⁴ The feasibility and safety of ICG guidance have been verified in previous studies.

It has been reported that intravenous injection of ICG can be used for breast tumour localisation and resection margin evaluation; however, the specificity and sensitivity of this technique have yet to be verified, and the exact physiological mechanism of ICG preferential uptake by tumour tissue after intravenous injection remains unclear. Local ICG injection is currently used for tumour lymph node dissection. Chen et al. localised the tumour and marked the margins using preoperative local ICG injection under gastroscopic guidance, demonstrating that local ICG injection may facilitate tumour localisation and assessment of resection margins.¹⁵ Meanwhile, the study found that local ICG injection had no obvious complications or adverse reactions, suggesting that local ICG injection is a safe method. For breast lesions, Aydogan et al. injected ICG into local lesions directly for tumour localisation under ultrasound guidance for the first time, and under the guidance of a near-infrared sensitive camera for the determination of the skin incision and surgical margins by observing the area of fluorescence.¹⁶ Later, a study using local injection of ICG to detect non-palpable breast cancer found that this method had an acceptable oncological safety after 19 months of follow-up.¹⁷ Additionally, there are some studies using local ICG injection to detect the breast lesion,^18,19 suggesting the feasibility and safety of local intratumoral ICG injection for tumour localisation and margin delineation.

The dosage of ICG administered in this work was determined with reference to relevant published literature. The concentration of the ICG solution is 2.5 mg/mL or 1.25 mg/mL.¹⁴ Since the rapid diffusion of 1.25 mg/mL ICG affected the assessment of tumour location and boundaries, the ICG solution was diluted (0.625 mg/mL), and no patients had ICG- related adverse reactions in the present group. No remarkable time-dependent enlargement of fluorescent staining was observed around lesions, likely because intact capsules of benign tumours hinder rapid interstitial diffusion of injected ICG. However, the study has also reported that different ICG concentrations have different effects on the fluorescence imaging.²⁰ The optimal dosage and concentration of ICG require further investigation.

Meanwhile, robot-assisted breast surgery could be used to locally excise a breast tumour through an axillary incision, leaving no scars over the breast. However, this surgical approach could not be performed reliably without a method for tumour localisation and boundary delineation. Therefore,  a technique combining ultrasound guidance with local injections of ICG has been developed to localise the benign breast tumour and map margins with the Firefly model of the da Vinci robot instrument. Recently, the study using this technique to delineate tumour boundary in breast cancer has further preliminarily verified the feasibility of this approach.²¹

Robot-assisted surgery is clearly an advantageous procedure without any scars on the breast, compared with open surgery, which results in a relatively visible scar over the breast, especially in patients prone to scarring. A covert incision or scar in the axilla may present better patient satisfaction. The differences in patient satisfaction were notable, with significant differences observed between robot-assisted and conventional open surgery in postoperative breast appearance, surgical wound position, and scar appearance. Using ultrasound-guided injections of ICG to map the tumour location and its boundaries enabled the console surgeon to locate the tumour more accurately. Since this technique describes a retromammary dissection, patients with lesions located posteriorly and larger size may be ideal candidate for this technique.

Patient-reported surveys demonstrated that robot-assisted surgery yielded superior aesthetic outcomes. However, due to the limited number of included patients, whether this result is superior to that of open surgery requires confirmation in future studies with larger sample sizes. Regarding operating time, the use of more complicated robotic equipment clearly has a learning curve. The total operation time improved to within 100 minutes, with console time within 25 minutes. The pre-docking time was significantly shortened through improved perioperative preparation and optimisation of operating room team procedures.

In terms of cost, robotic lumpectomy inevitably results in higher total hospital expenditure than conventional open surgery. The additional cost is primarily attributable to fixed robotic platform usage fees and disposable robotic instruments. Prospective cohort analysis verified that robotic breast surgery incurs approximately $4000 greater costs per case compared to conventional open procedures in the United States.² The average cost of robotic breast surgery is about 31,338,659.0 Korean Won (approximately USD 20000) in Korea.²² As reflected in some studies, the clinical penetration of robotic breast surgery varies considerably across different national healthcare frameworks.²³ Restricted by varied healthcare funding allocation patterns and distinct medical insurance developmental stages, robotic surgical resources are mainly centralised in a small number of large urban medical institutions, which is a common developmental feature for emerging economies during the popularisation of high-end surgical equipment.²⁴ The total hospitalisation cost of robotic breast surgery is around USD 2500 higher per patient compared with open lumpectomy at the study centre. With revised medical insurance policies in several Chinese cities and the increasing adoption of domestically manufactured robotic platforms, pilot insurance reimbursement programmes for robotic breast surgery are being implemented in China, thereby improving the economic affordability of this procedure for patients. Further health-economic studies are warranted to evaluate the future cost-effectiveness of robotic breast surgery.

This study has some limitations. First, given the retrospective cohort design and limited sample size for comparative analysis, potential selection bias is inevitable. Further large-scale studies with optimised research designs, alongside longer-term post- operative follow-up, are required to verify the efficacy and cosmetic outcomes of this surgical approach. Second, the operative time for robotic lumpectomy was longer than that for open surgery. As surgical experience accumulates, the total operative time is expected to decrease further, and the learning curve requires additional investigation. Third, a detailed long-term cost-effectiveness analysis was not performed in the present study, and relevant economic evaluations require further investigation.

CONCLUSION

The current study described a novel surgical approach using robot-assisted breast surgery, guided by ultrasound injections of ICG to localise benign breast tumours. It was found to be a safe approach that yielded a good aesthetic outcome.

FUNDING:
The study was supported by the National Natural Science Foundation of China (82273256 to Ning Liao).

ETHICAL APPROVAL:
The study was approved by the Ethics Committee of Guangdong Provincial People’s Hospital, Guangzhou, China (Permission No. KY2023-731-02).

PATIENTS’ CONSENT:
Informed consent for the use of clinical records was obtained from the included participants.

COMPETING INTEREST:
The authors declared no conflict of interest.

AUTHORS’ CONTRIBUTIONS:
NL, GZ: Methodology, data analysis, data collection, supervision and writing the original draft.
CR, JC: Data collection, data, analysis, and review.
NL: Conceptualisation, methodology, supervision, and reviewing.
All authors approved the final version of the manuscript to be published.
 

REFERENCES

1. Toesca A, Sangalli C, Maisonneuve P, Massari G, Girardi A, Baker JL, et al. A randomised trial of robotic mastectomy versus open surgery in women with breast cancer or BrCA mutation. Ann Surg 2022; 276(1):11-9. doi:10.1097/SLA. 0000000000004969.
2. Lai HW, Chen DR, Liu LC, Chen ST, Kuo YL, Lin SL, et al. Robotic versus conventional or endoscopic-assisted nipple- sparing mastectomy and immediate prosthesis breast reconstruction in the management of breast cancer: A prospectively designed multicenter trial comparing clinical out- comes, medical cost, and patient-reported outcomes (RCENSM-P). Ann Surg 2024; 279(1):138-46. doi: 10.1097/ SLA.0000000000005924.
3. Morrow M. Robotic mastectomy: the next major advance in breast cancer surgery? Br J Surg 2021; 108(3):233-4. doi: 10.1093/bjs/znab010.
4. Kim JH, Toesca A, Pozzi G, Gazzetta G, Marrazzo E, Park HS. Controversies and strengths of robot-assisted mastectomy. Eur J Cancer Prev 2023; 32(4):388-90. doi: 10.1097/CEJ.00 00000000000812.
5. Egan KG, Selber JC. Modern innovations in breast surgery: Robotic breast surgery and robotic breast reconstruction. Clin Plast Surg 2023; 50(2):357-66. doi: 10.1016/j.cps. 2022.11.004.
6. Nessa A, Shaikh S, Fuller M, Masannat YA, Kastora SL. Postoperative complications and surgical outcomes of robotic versus conventional nipple-sparing mastectomy in breast cancer: Meta-analysis. Br J Surg 2024; 111(1):znad336. doi: 10.1093/bjs/znad336.
7. Ahn JH, Park JM, Choi SB, Go J, Lee J, Kim JY, et al. Early experience of robotic axillary lymph node dissection in patients with node-positive breast cancer. Breast Cancer Res Treat 2023; 198(3):405-12. doi: 10.1007/s10549-022- 06760-8.
8. Tanaka T, Nishie R, Ueda S, Hashida S, Miyamoto S, Terada S, et al. Robot-assisted modified radical hysterectomy with removal of lymphatic vessel using indocyanine green: A new method. Int J Med Robot 2022; 18(6):e2451. doi: 10. 1002/rcs.2451.
9. Rocha A, Domínguez AM, Lecuru F, Bourdel N. Indocyanine green and infrared fluorescence in detection of sentinel lymph nodes in endometrial and cervical cancer staging - a systematic review. Eur J Obstet Gynecol Reprod Biol 2016; 206:213-9. doi: 10.1016/j.ejogrb.2016.09.027.
10. Potharazu AV, Gangemi A. Indocyanine green (ICG) fluorescence in robotic hepatobiliary surgery: A systematic review. Int J Med Robot 2023; 19(1):e2485. doi: 10.1002/rcs.2485.
11. Fujimoto D, Taniguchi K, Takashima J, Kobayashi H. Indocyanine green tracer-guided radical robotic distal gastrectomy using the firefly^™ system improves the quality of lymph node dissection in patients with gastric cancer. J Gastrointest Surg 2023; 27(9):1804-11. doi: 10.1007/s11605-023-05740-7.
12. Lai HW, Chen ST, Lin SL, Chen CJ, Lin YL, Pai SH, et al. Robotic nipple-sparing mastectomy and immediate breast reconstruction with gel implant: Technique, preliminary results and patient-reported cosmetic outcome. Ann Surg Oncol 2019; 26(1):42-52. doi: 10.1245/s10434-018-67 04-2.
13. Lauwerends LJ, van Driel P, Baatenburg de Jong RJ, Hardillo JAU, Koljenovic S, Puppels G, et al. Real-time fluorescence imaging in intraoperative decision making for cancer sur-gery. Lancet Oncol 2021; 22(5):e186-95. doi: 10.1016/ s1470-2045(20)30600-8.
14. Wang X, Teh CSC, Ishizawa T, Aoki T, Cavallucci D, Lee SY, et al. Consensus guidelines for the use of fluorescence imaging in hepatobiliary surgery. Ann Surg 2021; 274(1): 97-106 doi: 10.1097/sla.0000000000004718.
15. Chen QY, Xie JW, Zhong Q, Wang JB, Lin JX, Lu J, et al. Safety and efficacy of indocyanine green tracer-guided lymph node dissection during laparoscopic radical gastrectomy in patients with gastric cancer: A randomised clinical trial. JAMA Surg 2020; 155(4):300-11 doi: 10.1001/jamasurg. 2019.6033.
    
     
16. Aydogan F, Ozben V, Aytac E, Yilmaz H, Cercel A, Celik V. Excision of nonpalpable breast cancer with indocyanine green fluorescence-guided occult lesion localisation (IFOLL). Breast care (Basel, Switzerland) 2012; 7(1):48-51. doi: 10.1159/000336497.
17. Liu J, Guo W, Tong M. Intraoperative indocyanine green fluorescence guidance for excision of nonpalpable breast cancer. World J Surg Oncol 2016; 14(1):26. doi: 10.1186/ s12957-016-1014-2.
18. Muraoka A, Kobayashi M. Accurate resection of nonpalpable, ultrasonography undetectable breast cancer tumour by preoperative indocyanine green injection using stereotactic mammography: A case report. Ann Med Surg (Lond) 2022; 79:103965. doi: 10.1016/j.amsu.2022.103965.
19. Kim I, Choi HJ, Ryu JM, Lee SK, Yu JH, Lee JE, et al. The efficacy and safety of indocyanine green-hyaluronic acid mixture (LuminoMark^™) for localisation in patients with non-palpable breast lesions: A multi-centre open-label parallel phase-2 clinical trial. BMC Surg 2021; 21(1):134. doi: 10. 1186/s12893-021-01129-y.
20. Lee YJ, van den Berg NS, Orosco RK, Rosenthal EL, Sorger JM. A narrative review of fluorescence imaging in robotic-assisted surgery. Laparosc Surg 2021; 5:31. doi: 10.21037/ls- 20-98.
21. Liao N, Liu N, Zhang G, Ren C, Zhang W, Li C, et al. Ultrasound-guided, indocyanine green-directed robot-assisted surgery for breast cancer with negative margins: A single- centre study. Eur J Surg Oncol 2026; 52(4):111445. doi: 10.1016/j.ejso.2026.111445.
22. Yu DY, Lee HY, Lee TY, Kim DW, Ku D, Ko SY, et al. Endoscopic mastectomy: A cost-effective and time-efficient alternative to robotic mastectomy with comparable oncologic and safety outcomes. J Breast Cancer 2025; 28(5):347-57. doi: 10.4048/jbc.2025.0090.
23. Khan S, Khan MZ, Saeed R, Ahmed J, Naveed AK, Shah HH. Age of robotic surgery and telemedicine: Is Pakistan lagging behind? Ann Biomed Eng 2024; 52(6):1537-41. doi: 10. 1007/s10439-023-03337-4.
24. Naveed H, Naveed T, Farooqui A. Prevalence of robotic surgery in Pakistan. J Pak Med Assoc 2023; 73(8):1767. doi: 10.47391/JPMA.8131.