Surgical Endoscopy

, Volume 25, Issue 5, pp 1672–1676

Laparoscopic detection of sentinel node in gastric cancer surgery by indocyanine green fluorescence imaging

Authors

    • Department of SurgeryOsaka Medical Center for Cancer and Cardiovascular Diseases
  • Kentaro Kishi
    • Department of SurgeryOsaka Medical Center for Cancer and Cardiovascular Diseases
  • Masahiko Yano
    • Department of SurgeryOsaka Medical Center for Cancer and Cardiovascular Diseases
  • Koji Tanaka
    • Department of SurgeryOsaka Medical Center for Cancer and Cardiovascular Diseases
  • Masaaki Motoori
    • Department of SurgeryOsaka Medical Center for Cancer and Cardiovascular Diseases
  • Masayuki Ohue
    • Department of SurgeryOsaka Medical Center for Cancer and Cardiovascular Diseases
  • Hiroaki Ohigashi
    • Department of SurgeryOsaka Medical Center for Cancer and Cardiovascular Diseases
  • Akemi Takenaka
    • Department of PathologyOsaka Medical Center for Cancer and Cardiovascular Diseases
  • Yasuhiko Tomita
    • Department of PathologyOsaka Medical Center for Cancer and Cardiovascular Diseases
  • Osamu Ishikawa
    • Department of SurgeryOsaka Medical Center for Cancer and Cardiovascular Diseases
New Technology

DOI: 10.1007/s00464-010-1405-3

Cite this article as:
Miyashiro, I., Kishi, K., Yano, M. et al. Surg Endosc (2011) 25: 1672. doi:10.1007/s00464-010-1405-3

Abstract

Background

Indocyanine green (ICG) fluorescence imaging is a promising technique for detection of sentinel node (SN) as it avoids unnecessary resection. However, the ICG fluorescence imaging system cannot be used in laparoscopic surgery because of technological difficulties.

Methods

A prototype laparoscopic detection system comprising an electron multiplier charge-coupled device (EM-CCD) as the detector and a xenon lamp as the light source was developed. The CCD camera head was attached to the end of a specially designed laparoscope that could transmit ICG fluorescence. The system allows visualization of both color and fluorescence images. Laparoscopic surgery in ten patients with gastric cancer included SN biopsy using ICG dye, ICG fluorescence images using our system, and laparoscopy-assisted gastrectomy (LAG) with lymphadenectomy. SNs were sliced into 2-mm sections for histological examination and imprint cytology.

Results

Immediately after intraoperative ICG injection by endoscopy, the laparoscopic ICG fluorescence imaging system allowed easy visualization of the lymphatic vessels draining from the primary gastric tumor toward the lymph nodes and traced the moving injected dye, whereas lymph vessels and nodes were hardly recognized by ICG green color through a standard laparoscope. Surgeons could confirm the removed lymph nodes stained with ICG by fluorescent imaging. SNs were successfully detected in all patients. Three patients had suspicious metastases in the SNs; two patients had lymph node metastases only in the SNs.

Conclusions

Our newly developed laparoscopic ICG fluorescence imaging system is promising in the detection of SNs in laparoscopic gastric cancer surgery. The preliminary results suggest an easier and shorter learning curve of dye-guided SN biopsy in laparoscopic gastric surgery.

Keywords

Sentinel nodeGastric cancerIndocyanine green fluorescence imagingLaparoscopic surgeryNear-infrared light

The sentinel node (SN) technique allows direct and accurate assessment of lymph node status with minimal morbidity [1]. This technique is based on the concept that the tumor-bearing status of the SN reflects the tumor status of the remaining nodes. Our group reported the application of SN biopsy using indocyanine green (ICG) in open gastric cancer surgery, with a high success rate and a high degree of accuracy [2]. The dye-guided method is safe, convenient, and cost-effective, whereas the use of the radioactive probe-guided method in general hospitals is limited due to legal considerations and costs of radioactive substances [3, 4]. However, the dye-guided method has certain limitations, such as loss of visibility in dense fat and the rapid transit of the dye, and thus adequate training is required [57]. These limitations are more critical in laparoscopic surgery. Therefore, improvement of the detection method is important for clinical application of this technique, especially in laparoscopic surgery.

The SN biopsy guided by ICG fluorescence imaging was originally reported on in a preliminary study of breast cancer by Kitai et al. [8]. They concluded that this imaging technique is promising for further clinical exploration. More recently, we reported a clinical exploration/feasibility study of the novel technique for SN biopsy using ICG fluorescence imaging in gastric cancer surgery [9]. The imaging system was, however, for open surgery, not laparoscopic surgery. Due to technical difficulties, a laparoscopic system was not available at the time of that study. We have completed a preliminary study with a newly developed prototype system for laparoscopic surgery.

Materials and methods

A prototype laparoscopic detection system comprising an electron multiplier charge-coupled device (EM-CCD) as the detector and a xenon lamp as the light source was developed. The principle of the newly developed laparoscope system is based on the photodynamic eye (PDE) system (Hamamatsu Photonics, Hamamatsu, Japan) [10]. The light source was a xenon light that included a near-infrared wavelength so that it could generate the fluorescence of ICG. The detector was a color EM-CCD with high sensitivity not only in the visible wavelength but also in near-infrared region, instead of the ordinary CCD, which has less-than-ideal sensitivity. The CCD camera head was attached at the end of a custom-designed simple-to-use laparoscope that could transmit the ICG fluorescence. The system could view color images and also fluorescence images alternately by pressing a footswitch. The video signals were sent to a digital video processor and displayed on a monitor.

SN biopsy was conducted as described previously [2]. For this study, in principle 2-10 times dilution ICG (Diagnogreen, Dai-Ichi Sankyo Pharm. Co., Tokyo, Japan) in a volume of 2-4 ml (0.25-1.25 mg/0.5 ml, 4-8 injection sites) was injected just around the primary tumor using an intraoperative endoscopic puncture needle. All lymph nodes that stained green or were bright with fluorescence by the ICG fluorescence imaging system were excised before gastrectomy and were sliced into 2-mm sections for intraoperative histological examination with hematoxylin and eosin (H&E) staining. The sliced nodes were also examined simultaneously by imprint cytology.

Seven male and three female patients with a mean age of 68 ± 8 years (range = 54-79 years; median = 68 years) were enrolled and treated at the Osaka Medical Center for Cancer and Cardiovascular Diseases from February 2008 to February 2010 (Table 1). All patients had clinically T1 gastric cancer and underwent laparoscopy-assisted gastrectomy (LAG) with lymphadenectomy according to the gastric cancer treatment guidelines in Japan [11]. Nine patients received laparoscopy-assisted distal gastrectomy (LADG) and one patient received laparoscopy-assisted total gastrectomy (LATG). The preoperative diagnosis was based on gastric endoscopy, abdominal ultrasonography, and computed tomography. Informed consent was obtained from each patient preoperatively and the study was approved by the Human Ethics Review Committee of Osaka Medical Center for Cancer and Cardiovascular Diseases.
Table 1

Profile of enrolled patients with clinically T1 gastric cancer

 

Tumor location

Operative procedure

Pathological depth (TNM, 7th ed.)

No. of nodes

Nodal metastases

 

Portion

Cross-sectional

SN

Non-SN

1

Lower third of stomach

Anterior wall

LADG

pT1a

3

42

2

Lower third of stomach

Posterior wall

LADG

pT1b

3

46

3

Lower third of stomach

Greater curvature

LADG

pT1b

2

52

4

Lower third of stomach

Lesser curvature

LADG

pT1a

2

55

5

Lower third of stomach

Anterior wall

LADG

pT1b

3

29

Sentinel node

6

Middle third of stomach

Lesser curvature

LADG

pT1b

2

47

7

Lower third of stomach

Lesser curvature

LADG

pT2

6

86

Sentinel node

8

Lower third of stomach

Lesser curvature

LADG

pT1a

1

60

9

Middle third of stomach

Greater curvature

LADG

pT1a

4

20

10

Upper/middle third of stomach

Lesser curvature

LATG

pT1b

5

79

Both

LADG laparoscopy-assisted distal gastrectomy, LATG laparoscopy-assisted total gastrectomy

Results

None of the patients had an adverse event after ICG injection. Immediately after intraoperative endoscopic injection of ICG in the peritumor area, the laparoscopic ICG fluorescence imaging system visualized the lymphatic vessels draining the primary tumor toward the lymph nodes with gradual flow of the injected dye seen as a bright fluorescent image. Individual nodes could be visualized even through adipose tissue by the laparoscopic fluorescence imaging system within approximately 5 min after the dye injection (Fig. 1A, B). The laparoscopic fluorescence imaging system could visualize the fluorescent lymphatic channels and nodes that would otherwise hardly be recognized through dense fat when using the green color only with a standard laparoscopic view (Fig. 1C, D). The system was easy to use because not only fluorescence images but also color images could be obtained by stepping on the footswitch (Fig. 1E, F). Fluorescently labeled nodes were dissected out as SNs and sliced for intraoperative histological and cytological examination. Surgeons could confirm easily the removed lymph nodes stained with ICG by fluorescent imaging (Fig. 2). SN biopsy was completed within 30 min after dye injection and did not prolong surgery.
https://static-content.springer.com/image/art%3A10.1007%2Fs00464-010-1405-3/MediaObjects/464_2010_1405_Fig1_HTML.jpg
Fig. 1

Individual nodes could be visualized, even through adipose tissue, by the laparoscopic fluorescence imaging system (B) but not by standard laparoscopy (A). Thin arrows: a green-colored and fluorescently labeled infrapyloric lymph node; thick arrows: a fluorescently labeled but not green-colored suprapyloric lymph node. A fine lymph vessel toward a lymph node that is hardly recognized by the green color only with standard laparoscopy (C) was recognized by ICG fluorescent imaging (D). Our technique could provide fluorescence images (F) as well as color images (E) (Color figure online)

https://static-content.springer.com/image/art%3A10.1007%2Fs00464-010-1405-3/MediaObjects/464_2010_1405_Fig2_HTML.jpg
Fig. 2

Surgeons were able to confirm by fluorescent imaging that the excised lymph nodes stained ICG green. Arrows: a negative control lymph node recognized as a nonsentinel node in the surgical field by green color and the ICG fluorescence imaging system (Color figure online)

The fluorescent bright nodes were successfully detected as SNs in all patients (range = 1-6 nodes; mean = 3.1 ± 1.5; median = 3.0) with the laparoscopic ICG fluorescence imaging system (Table 1). Histopathological examination of H&E-stained sections indicated that seven patients had no lymph node metastases in both SNs and non-SNs. Imprint cytology also indicated that seven patients had no metastases in SNs and that three patients had suspicious metastases. Two patients had lymph node metastases in SNs but not in non-SNs.

Discussion

Compared with the radio-guided method, the dye-guided method is potentially safer, more convenient, and more cost-effective but requires adequate training. The significance of the technical learning curve for SN biopsy has already been discussed [5]. False-negative SN biopsy may lead to local control failure. The removal of lymph nodes in gastric cancer is indicated for now.

ICG is a popular diagnostic reagent approved clinically [12, 13], and allergic reactions to ICG are fewer than those to blue dyes such as isosulfan blue [14]. ICG has an absorption peak of 800 nm in vivo and is detected as green in color. Although detection by absorption spectroscopy is more sensitive than color perception, Kitai et al. [8] noted that the sensitivity of fluorescence spectroscopy is greater than that of absorption spectroscopy. They also reported that an ICG solution could be detected embedded 10 mm deep in a material that has optical properties similar to that of human tissue. In comparison, ICG detection by near-infrared absorption spectroscopy is limited to a depth of only 3 mm [15]. The difference could be critical in SN detection. In this context, Ishikawa et al. [15] reported on an obese patient with a false-negative SN when the infrared ray electronic endoscopy (IREE) system [16] was used. In contrast, a study using a larger population reported that the ICG fluorescence imaging allowed highly sensitive image-guided intraoperative SN mapping in open gastric cancer surgery [17] following our pioneering study of the novel detection technique for SN biopsy [9].

The SN biopsy is useful not only for improved tumor staging by detecting metastasis, but also for helping to decide to change the therapeutic approach for patients with gastric cancer to less invasive treatment, including a laparoscopic approach as a minimal-access procedure. Nevertheless, laparoscopic SN biopsy has several problems that need to be overcome [18, 19]. The shine-through effect from the injection site seriously restricts the use of the radio-guided method. Tonouchi et al. [20] reported a false-negative case resulting from a detection error attributed to the shine-through effect of the radio-guided method. Several limitations of the dye-guided method, such as loss of visibility in dense fat and rapid transit of the dye, are more critical in laparoscopic surgery. Thus, a newly developed laparoscopic ICG fluorescence imaging system, unavailable commercially at present, would overcome such limitations of the dye-guided SN biopsy. The wavelength of light emitted from the Astral lamp, but not the ordinary light lamp, influences ICG fluorescence imaging. Fortunately, since the light emitted by the laparoscope into the peritoneal cavity does not influence the ICG fluorescence imaging system, a dark environment is not necessary and laparoscopic exploration or surgery can be conducted in such environment. Further technological developments are expected for the laparoscopic system because the commercially available system for open surgery provides sharper images, based on our own experience.

In the present study, all patients had clinically T1 gastric cancer and underwent LAG with lymphadenectomy. Follow-up studies showed that a small number of these patients developed lymph node metastases. Like the system for open surgery [17], further studies using larger population samples are required; however, our report indicates that the ICG fluorescence imaging system is a promising technique, because it simplifies the dye-guided SN biopsy not only in open [9, 21] but also in laparoscopic gastric cancer surgery. Based on our limited experience, participating surgeons felt that the procedure was simple and easy to master and it can shorten the learning curve, thus overcoming the current problem related to the dependency of accuracy of SN biopsy on the individual surgeon.

Acknowledgments

The authors thank Mr. Mitsuharu Miwa for the excellent technical support. This study was supported in part by the Otsuka Grant.

Disclosures

Drs. Miyashiro, Kishi, Yano, Tanaka, Motoori, Ohue, Ohigashi, Tomita, and Ishikawa, and Ms. Takenaka have no conflicts of interest or financial ties to disclose.

Copyright information

© Springer Science+Business Media, LLC 2010