RECONSTRUCTIVE Indocyanine Green SPY Elite–Assisted Sentinel Lymph Node Biopsy in Cutaneous Melanoma Jason M. Korn, Alejandra Tellez-Diaz, Marisa Bartz-Kurycki, Brian Gastman,
M.D. M.D. M.D. M.D.
Cleveland, Ohio; and Buffalo, N.Y.
Background: Sentinel lymph node biopsy is the standard of care for intermediate-depth and high-risk thin melanomas. Recently, indocyanine green and near-infrared imaging have been used to aid in sentinel node biopsy. The present study aimed to determine the feasibility of sentinel lymph node biopsy with indocyanine green SPY Elite navigation and to critically evaluate the technique compared with the standard modalities. Methods: A retrospective review of 90 consecutive cutaneous melanoma patients who underwent sentinel lymph node biopsy was performed. Two cohorts were formed: group A, which had sentinel lymph node biopsy performed with blue dye and radioisotope; and group B, which had sentinel lymph node biopsy performed with radioisotope and indocyanine green SPY Elite navigation. The cohorts were compared to assess for differences in localization rates, sensitivity and specificity of sentinel node identification, and length of surgery. Results: The sentinel lymph node localization rate was 79.4 percent using the blue dye method, 98.0 percent using the indocyanine green fluorescence method, and 97.8 percent using the radioisotope/handheld gamma probe method. Indocyanine green fluorescence detected more sentinel lymph nodes than the vital dye method alone (p = 0.020). A trend toward a reduction in length of surgery was noted in the SPY Elite cohort. Conclusions: Sentinel lymph node mapping and localization in cutaneous melanoma with the indocyanine green SPY Elite navigation system is technically feasible and may offer several advantages over current modalities, including higher sensitivity and specificity, decreased number of lymph nodes sampled, decreased operative time, and potentially lower false-negative rates. (Plast. Reconstr. Surg. 133: 914, 2014.) CLINICAL QUESTION/LEVEL OF EVIDENCE: Diagnostic, II.
M
elanoma accounts for 10 percent of cutaneous malignancies; however, at least 65 percent of skin cancer–related deaths are attributed to melanoma.1 Lymph node status is the most important prognostic factor for patients who present with early-stage cutaneous melanoma.2 Sentinel lymph node biopsy in melanoma was pioneered by Morton et al. in 1991 with the use of vital dyes, and later improved on by Krag From the Department of Plastic Surgery, Cleveland Clinic Foundation; and the School of Medicine and Biomedical Sciences, University at Buffalo. Received for publication September 11, 2013; accepted October 7, 2013. Presented at the 58th Annual Meeting of the Plastic Surgery Research Council, in Santa Monica, California, May 2 through 4, 2013; and the 56th Annual Meeting of the Ohio Valley Society of Plastic Surgeons, in Indianapolis, Indiana, May 17 through 19, 2013. Copyright © 2014 by the American Society of Plastic Surgeons DOI: 10.1097/PRS.0000000000000006
914
et al. with the addition of radioisotopes such as technetium-99m sulfur colloid.3–5 When used in conjunction, vital dye and radioisotope can accurately identify the sentinel node in 96 percent of cases of cutaneous melanoma.6 Sentinel lymph node biopsy has since become the standard of Disclosure: Brian Gastman, M.D., has a consulting agreement with LifeCell Corp. None of the other authors has a financial interest in any of the products or devices mentioned in this article.
Supplemental digital content is available for this article. Direct URL citations appear in the text; simply type the URL address into any Web browser to access this content. Clickable links to the material are provided in the HTML text of this article on the Journal’s Web site (www. PRSJournal.com).
www.PRSJournal.com
Volume 133, Number 4 • SPY Elite for Sentinel Lymph Node Biopsy care for intermediate depth melanomas and thin melanomas with high-risk features (i.e., ulceration or mitotic index >1).7 Despite advances in medical technology, pathologic analysis, and surgical techniques, false-negative sentinel lymph node biopsy is still reported to occur in 13 percent (range, 0 to 34 percent) of cases.2,6,8–10 A false-negative sentinel lymph node is defined as a regional nodal recurrence in a previously sampled negative sentinel lymph node basin. False-negative sentinel lymph node biopsy has been attributed to multiple causes, including errors in pathologic analysis, technical errors in nuclear medicine, surgical excisional errors, and anatomical causes secondary to variable lymphatic drainage, most notably in the head and neck region.6,9,10 In addition, the biological nature of cutaneous melanoma is characterized by slow dissemination of metastatic cells in lymphatic channels or large tumor deposits blocking or diverting lymphatic drainage, both of which can contribute to false-negative sentinel lymph node biopsy.11,12 Furthermore, nonlocalization of sentinel lymph nodes is noted in up to 5 percent of cases. Thus, to optimize lymph node staging and regional control of melanoma, new techniques have been used to better identify sentinel lymph nodes, including single-photon emission computed tomographic/computed tomographic scanning, immunohistochemistry, polymerase chain reaction, mobile gamma cameras, and, recently, the fluorescein dye indocyanine green.13,14 Indocyanine green is a small substance molecule (774.96 Da) tricarbocyanine dye and is the only near-infrared fluorophore that is approved by the U.S. Food and Drug Administration for human use.15 Indocyanine green offers several inherent advantages over other modalities of sentinel lymph node biopsy. Specifically, indocyanine green associates with albumin, making it an excellent agent for evaluating vascular and lymphatic systems. Indocyanine green can be excited in the near-infrared spectrum between 760 and 785 nm and the fluorescence can be imaged between 820 and 840 nm.16 Near-infrared imaging has high penetration into living tissue, and indocyanine green can be visualized transcutaneously in real time up to a depth of 0.5 to 1.0 cm.15–17 Indocyanine green does not use radiation, can be seen for up to 10 hours after injection, and has a lower risk of severe anaphylactoid reactions (0.05 percent) compared with patent blue (0.3 percent) and isosulfan blue (1.1 percent). Finally, in contrast to the vital dyes, indocyanine green does not stain
or obscure tissue characteristics or risk tattooing.15 Fujiwara et al. were the first to describe sentinel lymph node detection in 10 patients with skin cancer (seven melanoma patients) using indocyanine green and the Photodynamic Eye near-infrared imaging system (Hamamatsu Photonics K.K., Hamamatsu, Japan), but no comparisons were performed between other sentinel node detection modalities.18 In the United States, the SPY Elite near-infrared imaging system (LifeCell Corp., Branchburg, N.J.) is used in plastic and reconstructive surgery to assess tissue perfusion with the aid of intravascular injected indocyanine green. Limited studies have been conducted on the ability of indocyanine green–assisted sentinel lymph node biopsy in melanoma. To our knowledge, none has been performed with the use of the SPY Elite System. The primary objective of the current study was to evaluate the feasibility of using the indocyanine green SPY Elite System intraoperatively for navigation of sentinel lymph node biopsy in primary cutaneous melanoma cases. Secondarily, we aimed to compare the sensitivity of indocyanine green SPY Elite navigation with vital dyes [Lymphazurin (Covidien, Mansfield, Mass.) or methylene blue] and/or radioisotopes (technetium-99) in sentinel lymph node identification. In addition, we hope to determine whether indocyanine green SPY Elite can localize unique sentinel lymph nodes not identified by other modalities and result in a more selective and time-efficient sentinel lymph node biopsy.
PATIENTS AND METHODS After institutional review board approval was obtained, cases of primary cutaneous malignant melanoma without clinical or radiologic evidence of metastasis, which underwent sentinel lymph node biopsy at the Cleveland Clinic between November of 2010 and March of 2013, were identified. Patients were excluded if they did not meet the National Comprehensive Cancer Network criteria for sentinel lymph node biopsy or did not have a diagnostic biopsy report available. Data collected included patient demographics, melanoma clinical and histologic data (including clinical staging, location, Breslow thickness, ulceration, and regression), operative time, length of followup, and lymphatic recurrence. Cutaneous melanoma staging was assigned based on the seventh edition of the American Joint Committee on Cancer staging system.19 Intraoperative lymph node examination data included ability to localize the sentinel lymph node, number of sentinel lymph
915
Plastic and Reconstructive Surgery • April 2014 nodes identified, and unique sentinel lymph nodes identified by each one of the three sentinel lymph node biopsy methods evaluated in this study (i.e., preoperative lymphoscintigraphy, radioisotope/handheld gamma probe, and blue dye and indocyanine green fluorescence). Localization was defined as successful identification of at least one sentinel lymph node. The localization rate for each sentinel lymph node method was calculated as the ratio of localization and nonlocalization. Localization of unique sentinel lymph nodes was defined as any additional sentinel lymph node on which biopsy was performed exclusively on the basis of a certain sentinel lymph node biopsy method not identified by other modalities. Results of post–sentinel lymph node biopsy histologic nodal evaluation for metastases were documented. A false-negative sentinel lymph node biopsy was defined as regional recurrence within a lymph node basin that was negative for sentinel lymph node biopsy. Length of follow-up was calculated from the date of surgery to the date of last follow-up or death. All cases were performed a single surgeon (B.G.). Two cohorts were developed based on the use of indocyanine green SPY Elite navigation: group A consisted of patients who had sentinel lymph node biopsy performed with a combination of vital dye and radioisotope/handheld gamma probe, and group B had sentinel lymph node biopsy performed using a combination of radioisotope/handheld gamma probe and indocyanine green SPY Elite navigation (Fig. 1). All patients included in this study underwent cutaneous preoperative lymphoscintigraphy with filtered technetium-99m sulfur colloid on the day of the sentinel lymph node biopsy procedure (Fig. 2, left). Filtered technietium-99m radiocolloid (582 to 325 μCi) was injected subcutaneously in two to four aliquots around the primary tumor site 2 to 4 hours before the sentinel lymph node biopsy procedure, and static and flow images were obtained. Furthermore, in patients with head and neck melanoma, single-photon emission computed tomographic/computed tomographic scans of these regions were also obtained. Foci of radiopharmaceutical accumulation identified by preoperative lymphoscintigraphy and, in head and neck cases, visualized on single-photon emission computed tomographic/computed tomographic images, were marked. Intraoperatively, either 0.5 to 3.0 ml of blue dye (isosulfan blue or methylene blue) (group A) or 0.5 to 1.5 ml of fluorescent indocyanine green (group B) was injected intradermally at the primary tumor site.
916
Fig. 1. Flowchart demonstrating study design. Patients were excluded who did not fulfill National Comprehensive Cancer Network guidelines for sentinel lymph node biopsy. All consecutive patients treated after April of 2011 had indocyanine green SPY Elite navigation and were considered group B. SLNB, sentinel lymph node biopsy; RI, radioisotope; VD, vital dye; ICG, indocyanine green SPY Elite navigation.
At the time of surgery, the skin incision was made at the most radioactive site, denoted by the handheld gamma probe. All lymph node basins labeled by lymphoscintigraphy were examined. All lymph nodes identified by radioactivity and either blue dye or fluorescence were sampled as sentinel lymph nodes. Fluorescence imaging was performed with the real-time indocyanine green fluorescence infrared camera SPY Elite System (Fig. 2, right). (See Video, Supplemental Digital Content 1, which demonstrates injection of indocyanine green and lymph node dissection by means of the SPY Elite System. The technique of injection of indocyanine green peritumorally in the same patient as shown in Figure 2. The video demonstrates the lymphatic uptake and trans cutaneous identification of the sentinel lymph node and the dissection. This video does not represent real time and has been condensed and edited, http://links.lww.com/PRS/A962.) Ex vivo gamma probe counts were recorded and back-table imaging was performed of the sentinel nodes by means of SPY Elite (Fig. 3). Surgical time was collected from the anesthesia record and defined as the time from first incision to complete closure. The data were analyzed using R software (version 2.15; Vienna, Austria). All values of p < 0.05 were considered statistically significant. Categorical factors were summarized using frequencies and percentages, whereas continuous measures
Volume 133, Number 4 • SPY Elite for Sentinel Lymph Node Biopsy
Fig. 2. (Left) Preoperative lymphoscintigraphic image of a patient from group B demonstrating three sentinel lymph nodes in the left inguinal region (same patient as shown in Supplemental Digital Content 1). (Right) Intraoperative SPY Elite view of the same patient demonstrating transcutaneous identification of sentinel lymph nodes and lymphatics. This location corresponded to the “hottest region” with the gamma probe. (See Video, Supplemental Digital Content 1, http://links.lww.com/PRS/A962.)
were described using medians and percentiles because the distributions of these measures were largely nonparametric. Comparisons of cohorts on unordered factors were performed using Pearson chi-square tests or Fisher’s exact tests in variables with low frequencies. Ordered factors were compared by means of Mantel-Haenszel chi-square tests. Continuous measures were compared using Wilcoxon rank sum tests between groups and Wilcoxon rank sum tests within groups.
RESULTS We identified 147 patients with a new primary cutaneous malignant melanoma without clinical or radiologic evidence of metastasis who underwent sentinel lymph node biopsy during the study
period. Fifty-seven patients were excluded because they did not meet National Comprehensive Cancer Network criteria for sentinel lymph node biopsy. Thus, a total of 90 consecutive patients with a new diagnosis of cutaneous malignant melanoma constituted the study population analyzed (Fig. 1). The mean age of the patients in the cohort was 57.8 ± 15.8 years, and there was a slight female (52.2 percent) predominance. The most common primary location of the melanoma was head and neck in 31.0 percent (n = 28), followed by trunk in 30 percent (n = 27), lower extremity in 19 percent (n = 17), and upper extremity in 20 percent (n = 18). Mean Breslow level at the time of diagnosis was 1.93 ± 2.14; ulceration data were available for only 88 patients, and ulceration was observed in 16.7 percent (15 of 88) of the patients (Table 1).
Video. Supplemental Digital Content 1 demonstrates the lymphatic uptake and transcutaneous identification of the sentinel lymph node and the dissection. This video does not represent real time and has been condensed and edited, http://links.lww.com/PRS/A962.
917
Plastic and Reconstructive Surgery • April 2014
Fig. 3. (Left) Gross photograph of two sentinel lymph nodes from same patient shown in Figure 2, left, after removal; note the relative gamma counts for the two nodes. (Right) Near-infrared indocyanine green SPY Elite image of the same nodes; note the second node appears to fluoresce more.
Of the 90 sentinel lymph node biopsy procedures performed, 39 were performed using a combination of radioisotope/handheld gamma probe and blue dye methods (group A), and 51 were performed using a combination of radioisotope/ handheld gamma probe and indocyanine green fluorescence methods (group B) (Fig. 1). Patient age, sex, anatomical location, melanoma clinical T stage, mitotic index, and ulceration were similar between groups (p > 0.05). Median follow-up for the study population was 6.7 months (range, 0 to 25.0 months). Group A had a median follow-up of 15.2 months (range, 0 to 25.0 months), whereas group B was followed for a median of 3.5 months (range, 0 to 11.2 months). The localization rate for patients who underwent sentinel lymph node biopsy with a combination of radioisotope/handheld gamma probe and blue dye (group A) was 100 percent (39 of 39 cases), whereas the localization rate for patients who underwent sentinel lymph node biopsy with a combination of radioisotope/handheld gamma probe and indocyanine green fluorescence (group B) was 98 percent (50 of 51 cases). The total number of lymph nodes (sentinel and nonsentinel) submitted to biopsy was 4.36 ± 5.07 for group A and 3.9 ± 3.99 for group B; this difference was nonsignificant (p = 0.97). The sentinel lymph node localization rate was 79.4 percent using the blue dye method, 98.0 percent using the fluorescence method, 97.8 percent using the radioisotope/handheld gamma probe method, and 96.7 percent with the cutaneous lymphoscintigraphy method (Table 2). Within group B, there were three cases without localization by lymphoscintigraphy or the radioisotope/handheld gamma probe method. In one of these cases, we were able to localize the sentinel
918
lymph nodes with the fluorescence method; this unique sentinel lymph node, however, was negative for metastasis. Within group A, there were seven cases without sentinel lymph node localization by the blue dye method, all of which localized using the radiocolloid with hand-held gamma probe method (Fig. 4). The indocyanine green SPY Elite method detected more sentinel lymph nodes than the vital dye method alone. The mean number of sentinel lymph nodes localized was 1.71 ± 2.43 using the blue dye method and 1.86 ± 1.18 using the fluorescence method (p = 0.020) (Table 3). The mean number of sentinel lymph nodes detected with the fluorescence method was 1.86 ± 1.18 and the mean number of sentinel lymph nodes detected with the radioisotope/handheld gamma probe method was 1.86 ± 1.22 (p = 0.99) (Table 4). When comparing the mean number of sentinel lymph nodes identified by each sentinel lymph node biopsy method by location of the primary melanoma, a significant difference was detected in the lower extremity melanomas, with fluorescence identifying a mean sentinel lymph node number of 0.8 ± 0.45. In contrast, the blue dye method identified a mean number of 1.64 ± 0.5 sentinel lymph nodes (p = 0.016). There were no other statistically significant differences noted when comparing the mean number of sentinel lymph nodes identified by each sentinel lymph node biopsy method by primary site location. Sentinel lymph node metastases were identified in 12 of 90 patients (13.3 percent) (12.8 percent from group A and 13.7 percent from group B), all of whom went on to have completion lymph node dissections (Fig. 4). Recurrences over the period of follow-up were noted in five of the
Volume 133, Number 4 • SPY Elite for Sentinel Lymph Node Biopsy Table 1. Patient and Tumor Characteristics Total No. Sex Female Male Anatomical location Head and neck Lower extremity Trunk Upper extremity Mitotic index Mitotic index
M.D. M.D. M.D. M.D.
Cleveland, Ohio; and Buffalo, N.Y.
Background: Sentinel lymph node biopsy is the standard of care for intermediate-depth and high-risk thin melanomas. Recently, indocyanine green and near-infrared imaging have been used to aid in sentinel node biopsy. The present study aimed to determine the feasibility of sentinel lymph node biopsy with indocyanine green SPY Elite navigation and to critically evaluate the technique compared with the standard modalities. Methods: A retrospective review of 90 consecutive cutaneous melanoma patients who underwent sentinel lymph node biopsy was performed. Two cohorts were formed: group A, which had sentinel lymph node biopsy performed with blue dye and radioisotope; and group B, which had sentinel lymph node biopsy performed with radioisotope and indocyanine green SPY Elite navigation. The cohorts were compared to assess for differences in localization rates, sensitivity and specificity of sentinel node identification, and length of surgery. Results: The sentinel lymph node localization rate was 79.4 percent using the blue dye method, 98.0 percent using the indocyanine green fluorescence method, and 97.8 percent using the radioisotope/handheld gamma probe method. Indocyanine green fluorescence detected more sentinel lymph nodes than the vital dye method alone (p = 0.020). A trend toward a reduction in length of surgery was noted in the SPY Elite cohort. Conclusions: Sentinel lymph node mapping and localization in cutaneous melanoma with the indocyanine green SPY Elite navigation system is technically feasible and may offer several advantages over current modalities, including higher sensitivity and specificity, decreased number of lymph nodes sampled, decreased operative time, and potentially lower false-negative rates. (Plast. Reconstr. Surg. 133: 914, 2014.) CLINICAL QUESTION/LEVEL OF EVIDENCE: Diagnostic, II.
M
elanoma accounts for 10 percent of cutaneous malignancies; however, at least 65 percent of skin cancer–related deaths are attributed to melanoma.1 Lymph node status is the most important prognostic factor for patients who present with early-stage cutaneous melanoma.2 Sentinel lymph node biopsy in melanoma was pioneered by Morton et al. in 1991 with the use of vital dyes, and later improved on by Krag From the Department of Plastic Surgery, Cleveland Clinic Foundation; and the School of Medicine and Biomedical Sciences, University at Buffalo. Received for publication September 11, 2013; accepted October 7, 2013. Presented at the 58th Annual Meeting of the Plastic Surgery Research Council, in Santa Monica, California, May 2 through 4, 2013; and the 56th Annual Meeting of the Ohio Valley Society of Plastic Surgeons, in Indianapolis, Indiana, May 17 through 19, 2013. Copyright © 2014 by the American Society of Plastic Surgeons DOI: 10.1097/PRS.0000000000000006
914
et al. with the addition of radioisotopes such as technetium-99m sulfur colloid.3–5 When used in conjunction, vital dye and radioisotope can accurately identify the sentinel node in 96 percent of cases of cutaneous melanoma.6 Sentinel lymph node biopsy has since become the standard of Disclosure: Brian Gastman, M.D., has a consulting agreement with LifeCell Corp. None of the other authors has a financial interest in any of the products or devices mentioned in this article.
Supplemental digital content is available for this article. Direct URL citations appear in the text; simply type the URL address into any Web browser to access this content. Clickable links to the material are provided in the HTML text of this article on the Journal’s Web site (www. PRSJournal.com).
www.PRSJournal.com
Volume 133, Number 4 • SPY Elite for Sentinel Lymph Node Biopsy care for intermediate depth melanomas and thin melanomas with high-risk features (i.e., ulceration or mitotic index >1).7 Despite advances in medical technology, pathologic analysis, and surgical techniques, false-negative sentinel lymph node biopsy is still reported to occur in 13 percent (range, 0 to 34 percent) of cases.2,6,8–10 A false-negative sentinel lymph node is defined as a regional nodal recurrence in a previously sampled negative sentinel lymph node basin. False-negative sentinel lymph node biopsy has been attributed to multiple causes, including errors in pathologic analysis, technical errors in nuclear medicine, surgical excisional errors, and anatomical causes secondary to variable lymphatic drainage, most notably in the head and neck region.6,9,10 In addition, the biological nature of cutaneous melanoma is characterized by slow dissemination of metastatic cells in lymphatic channels or large tumor deposits blocking or diverting lymphatic drainage, both of which can contribute to false-negative sentinel lymph node biopsy.11,12 Furthermore, nonlocalization of sentinel lymph nodes is noted in up to 5 percent of cases. Thus, to optimize lymph node staging and regional control of melanoma, new techniques have been used to better identify sentinel lymph nodes, including single-photon emission computed tomographic/computed tomographic scanning, immunohistochemistry, polymerase chain reaction, mobile gamma cameras, and, recently, the fluorescein dye indocyanine green.13,14 Indocyanine green is a small substance molecule (774.96 Da) tricarbocyanine dye and is the only near-infrared fluorophore that is approved by the U.S. Food and Drug Administration for human use.15 Indocyanine green offers several inherent advantages over other modalities of sentinel lymph node biopsy. Specifically, indocyanine green associates with albumin, making it an excellent agent for evaluating vascular and lymphatic systems. Indocyanine green can be excited in the near-infrared spectrum between 760 and 785 nm and the fluorescence can be imaged between 820 and 840 nm.16 Near-infrared imaging has high penetration into living tissue, and indocyanine green can be visualized transcutaneously in real time up to a depth of 0.5 to 1.0 cm.15–17 Indocyanine green does not use radiation, can be seen for up to 10 hours after injection, and has a lower risk of severe anaphylactoid reactions (0.05 percent) compared with patent blue (0.3 percent) and isosulfan blue (1.1 percent). Finally, in contrast to the vital dyes, indocyanine green does not stain
or obscure tissue characteristics or risk tattooing.15 Fujiwara et al. were the first to describe sentinel lymph node detection in 10 patients with skin cancer (seven melanoma patients) using indocyanine green and the Photodynamic Eye near-infrared imaging system (Hamamatsu Photonics K.K., Hamamatsu, Japan), but no comparisons were performed between other sentinel node detection modalities.18 In the United States, the SPY Elite near-infrared imaging system (LifeCell Corp., Branchburg, N.J.) is used in plastic and reconstructive surgery to assess tissue perfusion with the aid of intravascular injected indocyanine green. Limited studies have been conducted on the ability of indocyanine green–assisted sentinel lymph node biopsy in melanoma. To our knowledge, none has been performed with the use of the SPY Elite System. The primary objective of the current study was to evaluate the feasibility of using the indocyanine green SPY Elite System intraoperatively for navigation of sentinel lymph node biopsy in primary cutaneous melanoma cases. Secondarily, we aimed to compare the sensitivity of indocyanine green SPY Elite navigation with vital dyes [Lymphazurin (Covidien, Mansfield, Mass.) or methylene blue] and/or radioisotopes (technetium-99) in sentinel lymph node identification. In addition, we hope to determine whether indocyanine green SPY Elite can localize unique sentinel lymph nodes not identified by other modalities and result in a more selective and time-efficient sentinel lymph node biopsy.
PATIENTS AND METHODS After institutional review board approval was obtained, cases of primary cutaneous malignant melanoma without clinical or radiologic evidence of metastasis, which underwent sentinel lymph node biopsy at the Cleveland Clinic between November of 2010 and March of 2013, were identified. Patients were excluded if they did not meet the National Comprehensive Cancer Network criteria for sentinel lymph node biopsy or did not have a diagnostic biopsy report available. Data collected included patient demographics, melanoma clinical and histologic data (including clinical staging, location, Breslow thickness, ulceration, and regression), operative time, length of followup, and lymphatic recurrence. Cutaneous melanoma staging was assigned based on the seventh edition of the American Joint Committee on Cancer staging system.19 Intraoperative lymph node examination data included ability to localize the sentinel lymph node, number of sentinel lymph
915
Plastic and Reconstructive Surgery • April 2014 nodes identified, and unique sentinel lymph nodes identified by each one of the three sentinel lymph node biopsy methods evaluated in this study (i.e., preoperative lymphoscintigraphy, radioisotope/handheld gamma probe, and blue dye and indocyanine green fluorescence). Localization was defined as successful identification of at least one sentinel lymph node. The localization rate for each sentinel lymph node method was calculated as the ratio of localization and nonlocalization. Localization of unique sentinel lymph nodes was defined as any additional sentinel lymph node on which biopsy was performed exclusively on the basis of a certain sentinel lymph node biopsy method not identified by other modalities. Results of post–sentinel lymph node biopsy histologic nodal evaluation for metastases were documented. A false-negative sentinel lymph node biopsy was defined as regional recurrence within a lymph node basin that was negative for sentinel lymph node biopsy. Length of follow-up was calculated from the date of surgery to the date of last follow-up or death. All cases were performed a single surgeon (B.G.). Two cohorts were developed based on the use of indocyanine green SPY Elite navigation: group A consisted of patients who had sentinel lymph node biopsy performed with a combination of vital dye and radioisotope/handheld gamma probe, and group B had sentinel lymph node biopsy performed using a combination of radioisotope/handheld gamma probe and indocyanine green SPY Elite navigation (Fig. 1). All patients included in this study underwent cutaneous preoperative lymphoscintigraphy with filtered technetium-99m sulfur colloid on the day of the sentinel lymph node biopsy procedure (Fig. 2, left). Filtered technietium-99m radiocolloid (582 to 325 μCi) was injected subcutaneously in two to four aliquots around the primary tumor site 2 to 4 hours before the sentinel lymph node biopsy procedure, and static and flow images were obtained. Furthermore, in patients with head and neck melanoma, single-photon emission computed tomographic/computed tomographic scans of these regions were also obtained. Foci of radiopharmaceutical accumulation identified by preoperative lymphoscintigraphy and, in head and neck cases, visualized on single-photon emission computed tomographic/computed tomographic images, were marked. Intraoperatively, either 0.5 to 3.0 ml of blue dye (isosulfan blue or methylene blue) (group A) or 0.5 to 1.5 ml of fluorescent indocyanine green (group B) was injected intradermally at the primary tumor site.
916
Fig. 1. Flowchart demonstrating study design. Patients were excluded who did not fulfill National Comprehensive Cancer Network guidelines for sentinel lymph node biopsy. All consecutive patients treated after April of 2011 had indocyanine green SPY Elite navigation and were considered group B. SLNB, sentinel lymph node biopsy; RI, radioisotope; VD, vital dye; ICG, indocyanine green SPY Elite navigation.
At the time of surgery, the skin incision was made at the most radioactive site, denoted by the handheld gamma probe. All lymph node basins labeled by lymphoscintigraphy were examined. All lymph nodes identified by radioactivity and either blue dye or fluorescence were sampled as sentinel lymph nodes. Fluorescence imaging was performed with the real-time indocyanine green fluorescence infrared camera SPY Elite System (Fig. 2, right). (See Video, Supplemental Digital Content 1, which demonstrates injection of indocyanine green and lymph node dissection by means of the SPY Elite System. The technique of injection of indocyanine green peritumorally in the same patient as shown in Figure 2. The video demonstrates the lymphatic uptake and trans cutaneous identification of the sentinel lymph node and the dissection. This video does not represent real time and has been condensed and edited, http://links.lww.com/PRS/A962.) Ex vivo gamma probe counts were recorded and back-table imaging was performed of the sentinel nodes by means of SPY Elite (Fig. 3). Surgical time was collected from the anesthesia record and defined as the time from first incision to complete closure. The data were analyzed using R software (version 2.15; Vienna, Austria). All values of p < 0.05 were considered statistically significant. Categorical factors were summarized using frequencies and percentages, whereas continuous measures
Volume 133, Number 4 • SPY Elite for Sentinel Lymph Node Biopsy
Fig. 2. (Left) Preoperative lymphoscintigraphic image of a patient from group B demonstrating three sentinel lymph nodes in the left inguinal region (same patient as shown in Supplemental Digital Content 1). (Right) Intraoperative SPY Elite view of the same patient demonstrating transcutaneous identification of sentinel lymph nodes and lymphatics. This location corresponded to the “hottest region” with the gamma probe. (See Video, Supplemental Digital Content 1, http://links.lww.com/PRS/A962.)
were described using medians and percentiles because the distributions of these measures were largely nonparametric. Comparisons of cohorts on unordered factors were performed using Pearson chi-square tests or Fisher’s exact tests in variables with low frequencies. Ordered factors were compared by means of Mantel-Haenszel chi-square tests. Continuous measures were compared using Wilcoxon rank sum tests between groups and Wilcoxon rank sum tests within groups.
RESULTS We identified 147 patients with a new primary cutaneous malignant melanoma without clinical or radiologic evidence of metastasis who underwent sentinel lymph node biopsy during the study
period. Fifty-seven patients were excluded because they did not meet National Comprehensive Cancer Network criteria for sentinel lymph node biopsy. Thus, a total of 90 consecutive patients with a new diagnosis of cutaneous malignant melanoma constituted the study population analyzed (Fig. 1). The mean age of the patients in the cohort was 57.8 ± 15.8 years, and there was a slight female (52.2 percent) predominance. The most common primary location of the melanoma was head and neck in 31.0 percent (n = 28), followed by trunk in 30 percent (n = 27), lower extremity in 19 percent (n = 17), and upper extremity in 20 percent (n = 18). Mean Breslow level at the time of diagnosis was 1.93 ± 2.14; ulceration data were available for only 88 patients, and ulceration was observed in 16.7 percent (15 of 88) of the patients (Table 1).
Video. Supplemental Digital Content 1 demonstrates the lymphatic uptake and transcutaneous identification of the sentinel lymph node and the dissection. This video does not represent real time and has been condensed and edited, http://links.lww.com/PRS/A962.
917
Plastic and Reconstructive Surgery • April 2014
Fig. 3. (Left) Gross photograph of two sentinel lymph nodes from same patient shown in Figure 2, left, after removal; note the relative gamma counts for the two nodes. (Right) Near-infrared indocyanine green SPY Elite image of the same nodes; note the second node appears to fluoresce more.
Of the 90 sentinel lymph node biopsy procedures performed, 39 were performed using a combination of radioisotope/handheld gamma probe and blue dye methods (group A), and 51 were performed using a combination of radioisotope/ handheld gamma probe and indocyanine green fluorescence methods (group B) (Fig. 1). Patient age, sex, anatomical location, melanoma clinical T stage, mitotic index, and ulceration were similar between groups (p > 0.05). Median follow-up for the study population was 6.7 months (range, 0 to 25.0 months). Group A had a median follow-up of 15.2 months (range, 0 to 25.0 months), whereas group B was followed for a median of 3.5 months (range, 0 to 11.2 months). The localization rate for patients who underwent sentinel lymph node biopsy with a combination of radioisotope/handheld gamma probe and blue dye (group A) was 100 percent (39 of 39 cases), whereas the localization rate for patients who underwent sentinel lymph node biopsy with a combination of radioisotope/handheld gamma probe and indocyanine green fluorescence (group B) was 98 percent (50 of 51 cases). The total number of lymph nodes (sentinel and nonsentinel) submitted to biopsy was 4.36 ± 5.07 for group A and 3.9 ± 3.99 for group B; this difference was nonsignificant (p = 0.97). The sentinel lymph node localization rate was 79.4 percent using the blue dye method, 98.0 percent using the fluorescence method, 97.8 percent using the radioisotope/handheld gamma probe method, and 96.7 percent with the cutaneous lymphoscintigraphy method (Table 2). Within group B, there were three cases without localization by lymphoscintigraphy or the radioisotope/handheld gamma probe method. In one of these cases, we were able to localize the sentinel
918
lymph nodes with the fluorescence method; this unique sentinel lymph node, however, was negative for metastasis. Within group A, there were seven cases without sentinel lymph node localization by the blue dye method, all of which localized using the radiocolloid with hand-held gamma probe method (Fig. 4). The indocyanine green SPY Elite method detected more sentinel lymph nodes than the vital dye method alone. The mean number of sentinel lymph nodes localized was 1.71 ± 2.43 using the blue dye method and 1.86 ± 1.18 using the fluorescence method (p = 0.020) (Table 3). The mean number of sentinel lymph nodes detected with the fluorescence method was 1.86 ± 1.18 and the mean number of sentinel lymph nodes detected with the radioisotope/handheld gamma probe method was 1.86 ± 1.22 (p = 0.99) (Table 4). When comparing the mean number of sentinel lymph nodes identified by each sentinel lymph node biopsy method by location of the primary melanoma, a significant difference was detected in the lower extremity melanomas, with fluorescence identifying a mean sentinel lymph node number of 0.8 ± 0.45. In contrast, the blue dye method identified a mean number of 1.64 ± 0.5 sentinel lymph nodes (p = 0.016). There were no other statistically significant differences noted when comparing the mean number of sentinel lymph nodes identified by each sentinel lymph node biopsy method by primary site location. Sentinel lymph node metastases were identified in 12 of 90 patients (13.3 percent) (12.8 percent from group A and 13.7 percent from group B), all of whom went on to have completion lymph node dissections (Fig. 4). Recurrences over the period of follow-up were noted in five of the
Volume 133, Number 4 • SPY Elite for Sentinel Lymph Node Biopsy Table 1. Patient and Tumor Characteristics Total No. Sex Female Male Anatomical location Head and neck Lower extremity Trunk Upper extremity Mitotic index Mitotic index
