- Email: [email protected]
Advancing robotic sentinel lymph node detection with indocyanine green (ICG) fluorescence: Optimal concentration for discrimination of the sentinel node
e156
Abstracts / Gynecologic Oncology 130 (2013) e1–e169
376 Feasibility of robotic extraperitoneal para-aortic lymphadenectomy for gynecologic cancer A. Bats, C. Bensaïd, A. Achouri, L. Makke, C. Nos, F. Lecuru. Hôpital Européen Georges-Pompidou, Paris, France. Objective: As the most recent evolution of minimally invasive techniques, robotic technology has increased in surgical procedures for gynecologic oncology. Robotic extraperitoneal para-aortic lymphadenectomy has never been reported. The aims of our study were to describe the technique, the feasibility, and results of robotic extraperitoneal paraaortic lymphadenectomy. Methods: Eight patients undergoing robotic extraperitoneal para-aortic lymphadenectomy using the Da Vinci system were evaluated retrospectively: 4 with cervical cancer and 4 with endometrial cancer. Extraperitoneal para- aortic lymphadenectomy was performed using a similar surgical technique as previously described by laparoscopy. The procedure was carried out using 4 port sites: 1 for the camera, 1 each for the no. 1 and no. 3 arms of the Da Vinci robot system, and 1 for the assistant. Results: The operation was completed in all but 1 patient. Beyond para-aortic lymphadenectomy, 3 patients with endometrial cancer had transperitoneal bilateral pelvic lymphadenectomy and hysterectomy, and 1 patient with advanced cervical cancer had anterior pelvectomy. The median age of patients was 61 years (range, 5266 years) and body mass index was 25.7 (range, 21.6-28.5). The median operating time was 270 min (range, 195-302 min). The median number of para-aortic lymph nodes removed reached 17 (range, 13-20), the median decrease in hemoglobin was 1.1 g/dL (range, 0.7-1.9 g/dL), and median hospital stay was 5 days (range, 47). There was 1 intraoperative complication (bilateral pneumothorax) and 1 postoperative complication (desaturation). Conclusions: Robotic-assisted extraperitoneal para-aortic lymphadenectomy via the Da Vinci system appears feasible and safe. doi:10.1016/j.ygyno.2013.04.435
377 Rate of bowel herniation in patients undergoing robotic surgery N. Goldman1, J. Wright2, S. Lewin2, T. Herzog2, W. Burke1. 1The Valley Hospital, Paramus, NJ, 2Columbia University Presbyterian Hospital, New York, NY. Objective: There is debate within the surgical literature over whether to close the fascia after operating laparoscopically. It is widely accepted that trocar wounds N10 mm should be closed, while smaller trocar wounds of 5 mm need not be closed. With the advent of robotic surgery, a new “middle ground” has presented itself with the use of 8-mm trocars for the robotic instruments. We reviewed our robotic surgical cases in an effort to determine if there is a need, based on an increased herniation rate, to routinely close the 8-mm robotic trocar fascia. Methods: All robotic cases performed from August 2006 to June 2012 were included for review. Minimum follow-up time was 4 months. Our configuration was the same for all cases and included a 10-mm camera trocar in the midline, approximately 3-4 cm above the umbilicus; 2 8- mm lateral trocars; 1 10-mm left lower quadrant trocar; and a 5-mm assistant trocar at Palmer’s point in the left upper quadrant. Cases were examined for bowel herniation through each trocar fascia. Multivariate analysis was performed to determine if herniation was related to age, race, body mass index (BMI), medical comorbidity, or type of procedure performed. Results: There were 845 robotic cases performed in the study time period, which translated to 845 5-mm trocar wounds, 1,690 8-mm trocar wounds, and 1,690 10-mm trocar wounds. None of the 5-mm trocar sites, none of the 8-mm trocar sites, and 3/1,690 (0.18%) 10-mm trocar sites experienced bowel herniation. Two hernias occurred at the
site of the camera trocar and 1 hernia was at the left lower quadrant site. There was no predilection to bowel herniation based on age, race, BMI, medical comorbidity, or type of procedure performed. Conclusions: Bowel herniation after laparoscopy creates an emergent situation, causes increased medical cost, and results in decreased patient satisfaction. Our experience showed a low rate of herniation when larger (10-mm) fascial defects were closed and no herniation with smaller unclosed fascial defects (8-mm and 5-mm). We conclude that trocar sites smaller than 10-mm may not need to have the fascia closed to prevent bowel herniation. doi:10.1016/j.ygyno.2013.04.436
378 Advancing robotic sentinel lymph node detection with indocyanine green (ICG) fluorescence: Optimal concentration for discrimination of the sentinel node K. Levinson1, H. Mahdi2, L. David2, P. Escobar2. 1Cleveland Clinic, Cleveland, OH, 2The Cleveland Clinic Foundation, Cleveland, OH. Objective: To determine the optimal dose of ICG to accurately distinguish the sentinel node from surrounding tissue. Methods: The study was performed on healthy female pigs weighing 40-60 kg. After induction of anesthesia, all pigs underwent exploratory laparotomy, dissection of the bladder, and a colpotomy to reveal the cervical os. Using a 21-gauge needle, 0.5 mL of normal saline was injected at the 3 o’clock and 9 o’clock positions as a control. Four concentrations of ICG were constituted for doses of 1,000 mcg, 500 mcg, 250 mcg, and 175 mcg per 0.5 mL. ICG was then injected at the 3 o’clock and 9 o’clock positions on the cervix. The SPY camera (Novadaq Technologies Inc.) was used to track ICG into the sentinel node and to quantify the intensity of light emitted. SPY technology uses an intensity scale from 1 to 256, which was used to determine the difference in intensity between the sentinel node and surrounding tissues. This difference was calculated at the time the sentinel node was first identified and when the sentinel node reached maximum intensity. The average intensity in the surrounding tissues was calculated by taking the average intensity in 4 surrounding quadrants adjacent to the node. Results: A sentinel node was identified at all doses except for the 175 mcg dose, at which ICG stayed in the cervix and vasculature only (Table and Figure). For the 1,000 mcg dose, the sentinel node was only delineated once maximum intensity was reached. At that time, the difference in intensity between the surrounding tissue and the node was 119 (251 vs. 132). For both the 500-mcg and 250-mcg doses, the sentinel node was identified before reaching maximum intensity. At that time, the difference was 172 (191 vs. 19) for the 500-mcg dose and 84 (124 vs. 40) for the 250-mcg dose. At maximum intensity, the difference between the surrounding tissue and the node was 207 (251 vs. 44) for the 500mcg dose and 159 (251 vs. 92) for the 250-mcg dose. Conclusions: For sentinel lymph node detection, the dose of ICG is related to the ability to discriminate the sentinel node from the surrounding tissue. ICG dose of 250-500 mcg successfully identify a sentinel lymph node with more distinction from the surrounding tissues.
Table. Discrimination of the Sentinel Node
Abstracts / Gynecologic Oncology 130 (2013) e1–e169
e157
doi:10.1016/j.ygyno.2013.04.438 doi:10.1016/j.ygyno.2013.04.437
379 Is robotic hysterectomy really less painful than traditional laparoscopy? S. Lange, N. Rozario, J. Hall, R. Higgins, D. Tait, R. Naumann. Carolinas Medical Center, Charlotte, NC. Objective: To compare narcotic use (as a surrogate for postoperative pain) in patients who had a total robotic hysterectomy (TRH) compared to those with total laparoscopic hysterectomy (TLH) for endometrial cancer. Methods: All cases of TRH for the treatment of endometrial cancer (EC) from 1/08 until 7/12 were reviewed with institutional review board approval. These were matched with TLHs for age, body mass index (BMI), type of procedure, and the use of intravenous nonsteroidal anti-inflammatory drugs (NSAIDs). Exclusion criteria included cases that were converted to open hysterectomy and patients with pre-existing chronic pain. The cumulative dose of narcotics, including intraoperative doses, were calculated for each patient during the first 6, 12, 18, and 24 hours and converted to fentanyl equivalents for comparison. Results: There were 100 TLHs and 63 TRHs, with no difference in age, BMI, or use of NSAIDs. Statistically significant differences were noted between the estimated blood loss and operative time of the 2 groups (P = 0.007 and P = 0.01, respectively). The mean fentanyl use at 24 hours for TLH was 484 mcg compared to 511 mcg (P = 0.38). The mean values at the first 6, 12, 18, and 24 hours were all lower for TLH than TRH, but the values did not reach statistical significance (Figure). There was no significant difference in narcotic use with respect to BMI, operative time, number of ports, and type of cuff closure. Conclusions: Patients undergoing TRH for EC used on average more narcotic analgesics at 6, 12, 18, and 24 hours from the time of incision. However, the difference was not statistically significant. There was no evidence that the TRH is any less painful than TLH. Although claims by the robotic manufacturer have suggested that this device decreases pain, there is no objective evidence to support this conclusion.
380 Learning curve associated with becoming proficient in completing a robotic-assisted total laparoscopic hysterectomy: A single-institution gynecologic oncology fellowship experience S. Sandadi, J. Gadzinski, S. Lee, D. Chi, E. Jewell, C. Brown, G. Gardner, R. Barakat, M. Leitao. Memorial Sloan-Kettering Cancer Center, New York, NY. Objective: To describe the learning curve associated with training fellows in completing robotic-assisted total laparoscopic hysterectomies. Methods: All patients scheduled to undergo a robotic procedure from 5/15/07–5/22/12 were identified. Various intraoperative time points were captured per our ongoing quality assessment. Fellow participation per procedure was also documented. For the current analysis, we focused on the learning curve of fellows for the time to complete a hysterectomy (from initiation of developing the retroperitoneal space to completion of the colpotomy). All cases were performed jointly and under the direct supervision of attending faculty. The dual-console platform was used for all cases starting in 2009. Appropriate statistical tests were used. Results: Of the 1,754 planned robotic cases, 1,626 were completed robotically and 128 were converted to laparotomy. Fifty-seven fellows, including gynecologic, urologic, and surgical fellows, participated in 99.5% of the cases. A total of 1,035 hysterectomies (including 56 radical hysterectomies) were performed. During this time period, 11 gynecologic oncology fellows completed at least 1 robotic-assisted total laparoscopic hysterectomy. Fifty-four hysterectomies were performed from 5/15/07– 7/6/08; none were completed by a fellow. From 7/7/08–5/21/12, 981 hysterectomies were completed robotically, 256 of these (26.1%) by the 11 fellows. The rate of hysterectomy completion by a fellow increased from 2.1% in 2008 to 42.4% by the end of 2011. Prior to completing a hysterectomy, the median number of hysterectomies in which a fellow participated to some degree on the console was 16 (range, 11-40). Median amount of time for a fellow to complete a hysterectomy decreased from 60 minutes in 2009 (n = 27 cases) to 31 minutes in 2011 (n = 147 cases). Based on the recorded completion times in which the 11 fellows completed a hysterectomy, it required ~33 cases per fellow to be able to perform the hysterectomy in 30 minutes or less. Conclusions: These data suggest that the learning curve associated with hysterectomy requires completion of ~ 30 cases by the fellow after an initial median experience of 16 cases. Our data suggest that a minimum of 50 total cases is required during fellowship to achieve proficiency performing a robotic hysterectomy. The dual-console platform greatly enhanced fellow training. doi:10.1016/j.ygyno.2013.04.439
Abstracts / Gynecologic Oncology 130 (2013) e1–e169
376 Feasibility of robotic extraperitoneal para-aortic lymphadenectomy for gynecologic cancer A. Bats, C. Bensaïd, A. Achouri, L. Makke, C. Nos, F. Lecuru. Hôpital Européen Georges-Pompidou, Paris, France. Objective: As the most recent evolution of minimally invasive techniques, robotic technology has increased in surgical procedures for gynecologic oncology. Robotic extraperitoneal para-aortic lymphadenectomy has never been reported. The aims of our study were to describe the technique, the feasibility, and results of robotic extraperitoneal paraaortic lymphadenectomy. Methods: Eight patients undergoing robotic extraperitoneal para-aortic lymphadenectomy using the Da Vinci system were evaluated retrospectively: 4 with cervical cancer and 4 with endometrial cancer. Extraperitoneal para- aortic lymphadenectomy was performed using a similar surgical technique as previously described by laparoscopy. The procedure was carried out using 4 port sites: 1 for the camera, 1 each for the no. 1 and no. 3 arms of the Da Vinci robot system, and 1 for the assistant. Results: The operation was completed in all but 1 patient. Beyond para-aortic lymphadenectomy, 3 patients with endometrial cancer had transperitoneal bilateral pelvic lymphadenectomy and hysterectomy, and 1 patient with advanced cervical cancer had anterior pelvectomy. The median age of patients was 61 years (range, 5266 years) and body mass index was 25.7 (range, 21.6-28.5). The median operating time was 270 min (range, 195-302 min). The median number of para-aortic lymph nodes removed reached 17 (range, 13-20), the median decrease in hemoglobin was 1.1 g/dL (range, 0.7-1.9 g/dL), and median hospital stay was 5 days (range, 47). There was 1 intraoperative complication (bilateral pneumothorax) and 1 postoperative complication (desaturation). Conclusions: Robotic-assisted extraperitoneal para-aortic lymphadenectomy via the Da Vinci system appears feasible and safe. doi:10.1016/j.ygyno.2013.04.435
377 Rate of bowel herniation in patients undergoing robotic surgery N. Goldman1, J. Wright2, S. Lewin2, T. Herzog2, W. Burke1. 1The Valley Hospital, Paramus, NJ, 2Columbia University Presbyterian Hospital, New York, NY. Objective: There is debate within the surgical literature over whether to close the fascia after operating laparoscopically. It is widely accepted that trocar wounds N10 mm should be closed, while smaller trocar wounds of 5 mm need not be closed. With the advent of robotic surgery, a new “middle ground” has presented itself with the use of 8-mm trocars for the robotic instruments. We reviewed our robotic surgical cases in an effort to determine if there is a need, based on an increased herniation rate, to routinely close the 8-mm robotic trocar fascia. Methods: All robotic cases performed from August 2006 to June 2012 were included for review. Minimum follow-up time was 4 months. Our configuration was the same for all cases and included a 10-mm camera trocar in the midline, approximately 3-4 cm above the umbilicus; 2 8- mm lateral trocars; 1 10-mm left lower quadrant trocar; and a 5-mm assistant trocar at Palmer’s point in the left upper quadrant. Cases were examined for bowel herniation through each trocar fascia. Multivariate analysis was performed to determine if herniation was related to age, race, body mass index (BMI), medical comorbidity, or type of procedure performed. Results: There were 845 robotic cases performed in the study time period, which translated to 845 5-mm trocar wounds, 1,690 8-mm trocar wounds, and 1,690 10-mm trocar wounds. None of the 5-mm trocar sites, none of the 8-mm trocar sites, and 3/1,690 (0.18%) 10-mm trocar sites experienced bowel herniation. Two hernias occurred at the
site of the camera trocar and 1 hernia was at the left lower quadrant site. There was no predilection to bowel herniation based on age, race, BMI, medical comorbidity, or type of procedure performed. Conclusions: Bowel herniation after laparoscopy creates an emergent situation, causes increased medical cost, and results in decreased patient satisfaction. Our experience showed a low rate of herniation when larger (10-mm) fascial defects were closed and no herniation with smaller unclosed fascial defects (8-mm and 5-mm). We conclude that trocar sites smaller than 10-mm may not need to have the fascia closed to prevent bowel herniation. doi:10.1016/j.ygyno.2013.04.436
378 Advancing robotic sentinel lymph node detection with indocyanine green (ICG) fluorescence: Optimal concentration for discrimination of the sentinel node K. Levinson1, H. Mahdi2, L. David2, P. Escobar2. 1Cleveland Clinic, Cleveland, OH, 2The Cleveland Clinic Foundation, Cleveland, OH. Objective: To determine the optimal dose of ICG to accurately distinguish the sentinel node from surrounding tissue. Methods: The study was performed on healthy female pigs weighing 40-60 kg. After induction of anesthesia, all pigs underwent exploratory laparotomy, dissection of the bladder, and a colpotomy to reveal the cervical os. Using a 21-gauge needle, 0.5 mL of normal saline was injected at the 3 o’clock and 9 o’clock positions as a control. Four concentrations of ICG were constituted for doses of 1,000 mcg, 500 mcg, 250 mcg, and 175 mcg per 0.5 mL. ICG was then injected at the 3 o’clock and 9 o’clock positions on the cervix. The SPY camera (Novadaq Technologies Inc.) was used to track ICG into the sentinel node and to quantify the intensity of light emitted. SPY technology uses an intensity scale from 1 to 256, which was used to determine the difference in intensity between the sentinel node and surrounding tissues. This difference was calculated at the time the sentinel node was first identified and when the sentinel node reached maximum intensity. The average intensity in the surrounding tissues was calculated by taking the average intensity in 4 surrounding quadrants adjacent to the node. Results: A sentinel node was identified at all doses except for the 175 mcg dose, at which ICG stayed in the cervix and vasculature only (Table and Figure). For the 1,000 mcg dose, the sentinel node was only delineated once maximum intensity was reached. At that time, the difference in intensity between the surrounding tissue and the node was 119 (251 vs. 132). For both the 500-mcg and 250-mcg doses, the sentinel node was identified before reaching maximum intensity. At that time, the difference was 172 (191 vs. 19) for the 500-mcg dose and 84 (124 vs. 40) for the 250-mcg dose. At maximum intensity, the difference between the surrounding tissue and the node was 207 (251 vs. 44) for the 500mcg dose and 159 (251 vs. 92) for the 250-mcg dose. Conclusions: For sentinel lymph node detection, the dose of ICG is related to the ability to discriminate the sentinel node from the surrounding tissue. ICG dose of 250-500 mcg successfully identify a sentinel lymph node with more distinction from the surrounding tissues.
Table. Discrimination of the Sentinel Node
Abstracts / Gynecologic Oncology 130 (2013) e1–e169
e157
doi:10.1016/j.ygyno.2013.04.438 doi:10.1016/j.ygyno.2013.04.437
379 Is robotic hysterectomy really less painful than traditional laparoscopy? S. Lange, N. Rozario, J. Hall, R. Higgins, D. Tait, R. Naumann. Carolinas Medical Center, Charlotte, NC. Objective: To compare narcotic use (as a surrogate for postoperative pain) in patients who had a total robotic hysterectomy (TRH) compared to those with total laparoscopic hysterectomy (TLH) for endometrial cancer. Methods: All cases of TRH for the treatment of endometrial cancer (EC) from 1/08 until 7/12 were reviewed with institutional review board approval. These were matched with TLHs for age, body mass index (BMI), type of procedure, and the use of intravenous nonsteroidal anti-inflammatory drugs (NSAIDs). Exclusion criteria included cases that were converted to open hysterectomy and patients with pre-existing chronic pain. The cumulative dose of narcotics, including intraoperative doses, were calculated for each patient during the first 6, 12, 18, and 24 hours and converted to fentanyl equivalents for comparison. Results: There were 100 TLHs and 63 TRHs, with no difference in age, BMI, or use of NSAIDs. Statistically significant differences were noted between the estimated blood loss and operative time of the 2 groups (P = 0.007 and P = 0.01, respectively). The mean fentanyl use at 24 hours for TLH was 484 mcg compared to 511 mcg (P = 0.38). The mean values at the first 6, 12, 18, and 24 hours were all lower for TLH than TRH, but the values did not reach statistical significance (Figure). There was no significant difference in narcotic use with respect to BMI, operative time, number of ports, and type of cuff closure. Conclusions: Patients undergoing TRH for EC used on average more narcotic analgesics at 6, 12, 18, and 24 hours from the time of incision. However, the difference was not statistically significant. There was no evidence that the TRH is any less painful than TLH. Although claims by the robotic manufacturer have suggested that this device decreases pain, there is no objective evidence to support this conclusion.
380 Learning curve associated with becoming proficient in completing a robotic-assisted total laparoscopic hysterectomy: A single-institution gynecologic oncology fellowship experience S. Sandadi, J. Gadzinski, S. Lee, D. Chi, E. Jewell, C. Brown, G. Gardner, R. Barakat, M. Leitao. Memorial Sloan-Kettering Cancer Center, New York, NY. Objective: To describe the learning curve associated with training fellows in completing robotic-assisted total laparoscopic hysterectomies. Methods: All patients scheduled to undergo a robotic procedure from 5/15/07–5/22/12 were identified. Various intraoperative time points were captured per our ongoing quality assessment. Fellow participation per procedure was also documented. For the current analysis, we focused on the learning curve of fellows for the time to complete a hysterectomy (from initiation of developing the retroperitoneal space to completion of the colpotomy). All cases were performed jointly and under the direct supervision of attending faculty. The dual-console platform was used for all cases starting in 2009. Appropriate statistical tests were used. Results: Of the 1,754 planned robotic cases, 1,626 were completed robotically and 128 were converted to laparotomy. Fifty-seven fellows, including gynecologic, urologic, and surgical fellows, participated in 99.5% of the cases. A total of 1,035 hysterectomies (including 56 radical hysterectomies) were performed. During this time period, 11 gynecologic oncology fellows completed at least 1 robotic-assisted total laparoscopic hysterectomy. Fifty-four hysterectomies were performed from 5/15/07– 7/6/08; none were completed by a fellow. From 7/7/08–5/21/12, 981 hysterectomies were completed robotically, 256 of these (26.1%) by the 11 fellows. The rate of hysterectomy completion by a fellow increased from 2.1% in 2008 to 42.4% by the end of 2011. Prior to completing a hysterectomy, the median number of hysterectomies in which a fellow participated to some degree on the console was 16 (range, 11-40). Median amount of time for a fellow to complete a hysterectomy decreased from 60 minutes in 2009 (n = 27 cases) to 31 minutes in 2011 (n = 147 cases). Based on the recorded completion times in which the 11 fellows completed a hysterectomy, it required ~33 cases per fellow to be able to perform the hysterectomy in 30 minutes or less. Conclusions: These data suggest that the learning curve associated with hysterectomy requires completion of ~ 30 cases by the fellow after an initial median experience of 16 cases. Our data suggest that a minimum of 50 total cases is required during fellowship to achieve proficiency performing a robotic hysterectomy. The dual-console platform greatly enhanced fellow training. doi:10.1016/j.ygyno.2013.04.439