Sentinel lymph node (SLN) biopsy is now widely used for several gynecologic malignancies, as it has been shown to be both effective and safe for selecting patients who would benefit from lymphadenectomy, for example those women diagnosed with in vulvar and breast cancers [1]. However, due to the confounding bilateral multifarious drainage pattern of the uterus and cervix, the SLN procedure is more challenging in cervical and endometrial cancer patients than in breast or vulvar cancer patients [2].

Use of the SLN procedure in endometrial and cervical cancers have recently been investigated using traditional blue dyes or radiocolloids, or a combination of both, with variable results [3-5]. However, exposure to ionizing radiation, and the need for a specialized nuclear medicine unit, have limited the use of the radioactive tracer technique. More recently, the intraoperative near-infrared (NIR) fluorescence imaging method, using indocyanine green (ICG), has emerged as an effective technique for SLN detection during laparotomy in cervical cancer patients [6]. Due to the limited number of reports of NIR imaging being used for laparoscopic cervical surgery prior to 2012, we began our own prospective comparative assessment of the laparoscopic SLN detection rates for the ICG and blue colorimetric methods.

The use of patient data in this study was reviewed and pre-approved by Osaka University’s Ethical Committee for Human Subjects (Ethics Ref: 11077) on December 15, 2011. Written informed consent was obtained from all patients. All patients (n=16) who had planned SLN mapping using fluorescence imaging between May 2012 and August 2014 were identified and prospectively entered into the study database.

Indocyanine green (ICG, 25-mg vial) was purchased from Daiichi Sankyo (Tokyo, Japan). A 2.5 mg/ml ICG stock solution was made by adding 10 ml of sterile water to the vial. About 500 μM stand-alone working solution for ICG injection was made by mixing 8 ml of the stock with 42 ml of sterile water. The dose of 500 μM ICG was chosen based on previous optimal dose studies [7,8]. Usually, prior to initiating laparoscopic entry, the cervix is divided submucosally into 0-, 3-, 6-, and 9-o’clock positions and a total of from 2 to 4 ml of ICG working solution is injected vaginally. Patent blue dye was injected concurrently in 14 of the 16 cases, using the same method as described for ICG.

During the study period, beginning May 2012, we performed 16 laparoscopic surgeries using NIR-FI-SLNB, visualizing lymph node ICG uptake via an InfraRed Imaging System (Olympus corporation, Tokyo, Japan). The cervix was injected with tracer after the patient was prepped and draped, but prior to the insertion of any uterine manipulator. After creating the pneumoperitoneum, we opened the retroperitoneum and identified the SLN (Figures 1 and 2). For safety and validation, in all 16 cases a systemic pelvic lymph node dissection was performed after the SLN biopsy.

Figure 1.

— Panoramic view of right side pelvis after opening the paravesical space and pararectal space.

Figure 2.

— NIR-FI in the same patients shows the flow of lymph vessels and sentinel lymph nodes.

Patient demographics, surgical data, and final pathology results were collected and examined. After conducting definitive histology, all 16 cases were staged according to the 2009 International Federation of Gynecology and Obstetrics Classification system. All lymph nodes were handled in a standardized manner, without ultrastaging. The patient’s median age, BMI, estimated blood loss (EBL), time to find the SLN, number of SLN identified, and anatomic location of the SLN were recorded. The time to perform the SLN mapping was defined as the time from the injection of tracer until the surgeon felt that an SLN was, or was not, identified. Overall detection and bilateral detection rates to find the SLN were calculated. Fisher’s exact test was used to compare the SLN detection and mapping rates for the dyes used.

Sixteen patients with endometrial or cervical cancer were prospectively included into this study. Their mean age was 47.1 (range 30-67) years and mean BMI was 21.1 (range 16.9-24.6) kg/m2. The majority of the patients were diagnosed as having cervical cancer (Table 1).

Operating time spent specifically on SLN mapping was available from patient records for 12/16 (75%) patients. The mean surgical time to complete the SLN mapping was 26.8 (range 17-40) minutes. The mean number of SLNs removed per patient was 4.4 (range 1-12). During the 16 procedures there were no adverse intraoperative complications.

NIR-FI-SLNB mapping was used in all 16 cases, although blue dye was used concurrently in only 14/16 cases. SLNs were most frequently identified in the obturator region (62.3%), followed by the internal iliac (19.7%) and external iliac (18%) regions.

Using ICG, we were able to identify the SLN in 93.7% (15/16) of the cases, with bilateral pelvic mapping in 75% (12/16) (Table 2). With blue dye, they were able to identify the SLN in only 64.3% of 14 cases, with bilateral pelvic mapping in 14.3% (Table 3). When comparing mapping rates between the two methods, the mapping rate was higher with ICG than with the blue dye (64.3 vs. 93.7%, p = 0.1337). In bilateral hemi-pelvic mapping, ICG was significantly superior to blue dye (14.3 vs. 75%, p = 0.0009; Table 3).

All SLN detected by blue dye were also detected by NIR- FI-SLNB, but the converse was not true. In addition, a single confirmed metastatic-positive lymph node was identified in one (6.3%) of the 16 patients using ICG however this was not detectable by the blue colorimetric method.

Despite many published positive results [9], SLN biopsy for cervical and endometrial cancer is not yet widely practiced. This may be due to the challenging methodology and sometimes false-negatives obtained with traditional SLN identification techniques, which use injections of radiolabeled isotopes, such as Technetium, in conjunction with blue dyes, typically methylene blue or Lymphozurin). Although the isotopes need to be injected several hours before the surgery, to enable tracer uptake and transit-time to the SLN, the use of radiolabeled isotopes requires additional coordination with the Nuclear Medicine Department, and includes painful cervical injections. Hand-held or laparoscopic Geiger counters are used intraoperatively to detect isotope uptake by the SLNs, with the blue-dye acting as a visible guide to the SLN’s exact location.

Considerable experience and skill are required in the identification of SLN’s using this technology. Some suggest that a learning curve of 30 patients is necessary to achieve a level of competency associated with an acceptable false negative rate [10]. We hypothesized that these challenges could be overcome using the higher sensitivity of NIR imaging of fluorescent dyes. In addition, we used the novel optical filtering arrangement of the Olympus InfraRed Imaging System for visualizing lymph node drainage field. This would allow simultaneous real-time detection of bright-field color images and fluorescence images from multiple dyes (ICG and blue dye) using a single camera, with potential additive detection sensitivity. Thus, we introduced the ICG method into our surgical practice as potentially both a more effective and more convenient method than traditional isotope/blue dye techniques. We then compared their SLN detection rate using the ICG method with that of the blue colorimetric method.

The present results showed that NIR-FI-SLNB was significantly superior to the blue colorimetric method in bilateral SLN detection rate (75% vs. 14.3%, p = 0.0009). The bilateral detection rate of NIR-FI-SLNB (75%) was directly comparable to reports using radioisotope(RI) [9].

In this study, two findings were of major importance: first, all SLNs detected by blue dye were also detected by NIR-FI-SLNB, but the converse was not true; second, a single metastatic-positive lymph node was identified in one of the 16 patients, and critically, this positive SLN was detected only by the NIR-FI-SLNB method.

There are two reasons to explain the difference in results between the blue dye and NIR-FI-SLNB methods. Firstly, we could recognize the lymph vessels and nodes more clearly using the NIR optics system, as compared with visible light detection of the blue dye, even when the camera’s surgical field was smudged by blood. Secondly, ICG fluorescent imaging was more prominently visualized through the visceral and retroperitoneal fat, as compared with the blue dye. Fluorescence imaging was able to identify lymphatics through up to approximately 3 mm of undissected tissue, assisting the surgeon to identify the lymphatic channel before it was interrupted, facilitating the ease of dissection [11].

From 2010 to 2017, use of ICG for SLN biopsy for gynecologic cancers has now been reported in 28 case series. Among these reports, there were four series using only laparotomy, five using only laparoscopy, six included cases of both laparoscopy and laparotomy, and 13 of the series used a robotic surgeon [12-17]. Six series compared the SLN detection rate of blue dye vs. ICG. Among them, four series used robots and two series reported both laparotomy and laparoscopy cases. To our knowledge, our study is unique in that it compares ICG and blue dye results using only laparoscopy.

In all 28 reports, ICG was found to be superior to blue dye for detection of SLN (Table 4). Tanaka et al. compared ICG’s detection rate to three different tracers (RI, indocyanine green, indigo carmine). They also reported that patients who underwent a laparoscopic procedure had a higher SLN detection rate (100% vs. 77%, p < 0.01) and lower false negative (FN) rate (0% vs. 7.4%, p < 0.01) than those who underwent laparotomy [16].

In order to better directly compare the actual detection rates of various SLN tracers, we believe it is mandatory to determine the surgical procedure being used for the comparison (laparotomy or laparoscopy).

A key limitation of this study is the small number of patients that could be safely used, as at the time this was still largely an experimental procedure for this type of uterine gynecological cancer. During the pre- sent study we gained such confidence that ICG was significantly superior to use of blue dye. We abandoned the blue colorimetric method altogether and currently use only the ICG-NIR method.

In conclusion, fluorescence imaging of SLN’s with endoscopic ICG NIR imagers is both a more effective and more convenient method than traditional blue colorimetric method. During SLN biopsy, the concomitant use of a blue dye might be no longer necessary.

The authors have no conflicts of interest.