Increasingly earlier detection of small lung cancers may permit the possibility of anatomical sublobar resection. Robotic assisted thoracic surgery (RATS) facilitates complex segmentectomy but identification of the intersegmental boundary remains a drawback to minimally invasive techniques. We have employed two contrasting methods using either intravenous (IV) or intrabronchial (IB) indocyanine green (ICG) and compared the efficacy. The location of the lesion was identified on CT reconstruction. Patients were allocated to either method by surgeon preference. IV ICG was injected during RATS after division of the target segmental artery. A novel aerosolized IB technique (by fibreoptic bronchoscopy) was used to inject the target segmental bronchus before RATS. All operations were performed using the Da Vinci Xi robot (Intuitive Surgical, Sunnyvale, CA, USA) and near infrared thoracoscopy (Firefly®). IV ICG was administered in 18 patients and IB ICG was given in 13 patients. The intersegmental plane was identified by the fluorescence demarcation line, which was then divided using the robotic stapler. On univariate analysis, there were no intergroup differences between IV and IB in the patient demographics (median age 70.5 years vs 71 years), number of staple firings (8 vs 10, p>0.05), success in identifying the intersegmental plane (94 vs 100% respectively, p>0.05), postoperative air leak duration (2 vs 3 days, p >0.05), or TNM staging of tumours (T1 64% vs 86% respectively, p>0.05). The total anaesthetic time (230 vs 290 mins, p<0.05) and the operative duration (171.5 vs 193.5 mins, p<0.05) were significantly less in the IV ICG group. In IB ICG, more than half of the segmentectomies involved the apical segment of either lower lobe (5 R S6, 1 L S6; 3 L S1-3, 2L S4-5), whereas in IV ICG more complex segmentectomies were performed (2 R S1-2, 2 R S6, 2 R S7-10; 2 L S1-2, 5 L S1-3, 2 L S4-5, 1 L S7-10, 1 L S7-8). There was no mortality nor major complications in either group. One case was converted to thoracotomy due to extensive adhesions in the IV ICG group. In the same group, there was one R1 resection due to parietal pleura involvement.TableComparison between IV and IB ICG in RATS segmentectomiesTotal number of patientsIV ICG (n=18)IB IVG (n=13)Excluded patients1 case converted to thoracotomy1 wedge resection (frozen section negative)1 lobectomy (abnormal anatomy)Patient age (median, range)70.5 (35-84)71 (50-89)Sex (M/F)11/77/6Target segments (R/L)R S1-2 2R S6 2R S7-10 2L S1-2 2L S1-3 5L S4-5 2L S7-10 1L S7-8 1R S6 5L S1-3 3L S4-5 2L S6 1Operative duration (median, range)171.5 mins (111-228)193.5 mins (178-275)No of staple firings (median, range)8 (2-14)10 (7-15)Success in identifying the intersegmental plane (%)16/17 (94%)11/11 (100%)Major complications00Conversion to VATS or open1(adhesions)0Tumour clearance (%)R0 16/17 (94%)R0 11/11 (100%)NSCLC (%)NSCLC 14/17 (82%)NSCLC 7/11 (63%)T-status in NSCLC (%)T1 9/14 (64%)T1 6/7 (86%) Open table in a new tab In RATS segmentectomy, identification of the intersegmental plane is facilitated by the intraoperative use of ICG. Both IV and IB routes are feasible with comparable results. As the IV route appears to reduce the use of theatre time, a future randomized comparison is suggested.