Indocyanine green fluorescence navigation in laparoscopic hepatectomy: a retrospective single-center study of 120 cases

Laparoscopic hepatectomy is now performed widely to cure benign and malignant liver diseases [1]. To mark the demarcation line, portal staining or inflow clamping of the target area is recommended in conventional open anatomical liver resection [2‐4]. Conversely, in laparoscopic hepatectomy, this requires advanced skills. The lack of tactile perception of laparoscopic forceps and the complexity of intraoperative ultrasound examination makes it challenging to localize the tumor and confirm the demarcation line, limiting the application of this technology [5].

Indocyanine green (ICG), once bound to protein, can emit fluorescence (peaking at 840 nm) under the illumination of near-infrared light [6]. Because it can be absorbed exclusively by hepatocytes and excreted through bile without enterohepatic recirculation, ICG has gain the attention of hepatobiliary surgeons over the last two decades [7]. Intraoperative ICG fluorescence navigation might be a complementary solution to overcome the limitation of laparoscopic hepatectomy [4, 8]. Its application can also improve the detection rate of liver tumor focus, especially small lesions in an early stage [9‐11]. Moreover, laparoscopic hepatectomy using ICG fluorescence navigation is associated with less intraoperative blood loss, a lower transfusion rate, less postoperative complications, and reduced hospital stay, as well as a higher R0 resection rate to avoid a positive surgical margin [12‐15].

Despite all these advantages, ICG fluorescence navigation has some inherent drawbacks. First, ICG is usually injected intravenously a few days prior to the operation and bile metabolism malfunctions in diseased (especially cirrhotic) liver, and residual ICG will affect the observation [13, 16]. Second, regenerated nodules might show as false-positive if the bile duct is compressed [9]. Third, ICG fluorescence can only penetrate about 5–10 mm of tissue, so deep tumors cannot be detected superficially and positive- and negative-staining must be combined. More studies are needed to demonstrate the importance of ICG fluorescence navigation in laparoscopic hepatectomy.

We conducted this retrospective study, based on our 120-patient, single-center experience of laparoscopic hepatectomy, to investigate the role of ICG fluorescence navigation in laparoscopic hepatectomy and to assess the influence of the timing of ICG administration on intraoperative observation.

Recent studies have demonstrated the safety and efficiency of ICG fluorescence navigation in laparoscopic hepatectomy [12‐15]. This study also showed that the intraoperative and postoperative indexes of the ICG-FN group were comparable to or even better than those of the non-ICG-FN group. For patients with malignant tumors, the residual ICG inside the lesion may emit fluorescence under near-infrared light illumination, which helps the surgeon find the lesion quickly, especially if the tumor is in an “inconvenient” location such as the right posterior lobe [10, 11, 17]. The implementation of a negative staining technique is also helpful to identify the liver segment/lobe boundary and the resection plane. Energy devices such as the ultracision harmonic scalpel (Johnson & Johnson) are capable of breaking the liver parenchyma in the non-vascular area between the liver segments, reducing the need for vascular clips, reducing the operative time, preventing large vessel injury in the liver segments, and minimizing intraoperative blood loss. This was confirmed in our study. As a result of advances in surgical techniques, the overall intraoperative transfusion rate was low, with no difference between the groups. Although studies have shown that fluorescence navigation can reduce the incidence of bile leakage and liver abscess after hepatic segmentectomy and lobectomy [12], and minimize the postoperative hospital stay [13, 14], this study did not find that ICG fluorescence navigation reduced postoperative complications or the postoperative hospital stay. However, Spearman’s rank correlation and stepwise Logistic regression analysis indicated that ICG fluorescence navigation was associated with shorter operative time and less intraoperative blood loss. It was also a protective factor for less intraoperative blood loss, proving its safety and effectiveness. In this study, the liver resection type (anatomical vs. non-anatomical) did not affect operation time, hilar occlusion time, intraoperative blood loss, or postoperative hospital stay.

ICG dissolved in sterilized water was usually administrated preoperatively via a peripheral vein. ICG distributed in liver tissues with blood flow will be excreted through bile by normal hepatocytes, leading to tumor-specific staining, which enables the tumor to be visualized easily. As fluorescence penetrates the tissue to a depth of only 5–10 mm [6], superficial tumors near the liver capsular can be detected (Fig. 1a). To retain maximal liver parenchyma and reduce the postoperative complications caused by lack of liver volume, parenchymal-sparing non-anatomic resection could be performed, maintaining integrity around the tumors in the liver parenchyma and achieving a curative effect [18]. Intraoperative real-time navigated resection along the fluorescence boundaries was well accepted (Fig. 1b). Anatomic hepatectomy is a better option for non-neoplastic lesions such as hepatolithiasis and tumors located deep in the liver parenchyma not able to be well observed by tumor specific staining. Intraoperatively, the Glissonean pedicle of the lesion-involved segment/lobe was dissected and clipped (Fig. 2a, b), and then ICG was injected intravenously. ICG cannot be delivered into the area to be removed and the remnant liver was stained negatively (Fig. 2c). In the process of liver parenchyma resection, real-time fluorescence navigation was also used to obtain a satisfactory resection section (Fig. 2d). The combination of positive and negative staining can improve the success rate of ICG fluorescence navigation, although extensive clinical training and practical experience is required.

Previous studies have suggested that the timing of preoperative ICG injection may affect the fluorescence imaging of tumor-specific staining. It is thought that the injection should not be given fewer than 4 days preoperatively; and that administration 0–3 days preoperatively may decrease the detection rate of tumors [19, 20]. However, the flexibility of medical resource allocation might be poorer and the waiting time before surgery would be prolonged in this setting. A small retrospective study indicated that a shorter injection time can also result in a satisfactory imaging effect [16]. However, as the ICG administration timing remains inconclusive, another cohort study was begun last year [21]. In our experience, administering ICG 0–3 days preoperatively results in comparable staining and imaging satisfaction to 4–7 days preoperatively (p = 0.547). Moreover, there were no significant differences in objective indexes, such as operative time, hilar occlusion time, intraoperative blood loss and postoperative hospital stay (p > 0.05 for all), when ICG was administered 0–3 days preoperatively. ICG injection timing closer to the operation day meant shorter preoperative waiting time and better medical resources allocation.

The surgical margin around a malignant tumor is another concern of hepatobiliary surgeons, and related to tumor recurrence and patient survival. R0 resection means that the surgical margin is at least 1 mm away from the tumor boundary and no tumor cells are visible under the microscope. R0 resection was reported to be associated with better long-term survival [22]. ICG fluorescence navigation facilitates the comprehensive application of positive and negative staining and real-time navigation technology to achieve R0 resection [15]. Furthermore, it was demonstrated that narrow resection margin (width ≤ 10 mm) was a predictor for overall survival [23]. A meta-analysis of 34 studies involving 11,147 hepatectomy patients concluded that patients with a wide margin (> 10 mm) had a better prognosis, suggesting that this should be one of the goals of hepatectomy [24]. Recent studies also identified that a wide surgical margin is predictive of a better long-term prognosis [25‐27]. During laparoscopic surgery, the lack of tactile perception of laparoscopic forceps and the 2-dimensional image on the monitor in most centers might affect surgeons’ evaluation of the resection range, which could be insufficient to preserve liver parenchyma, resulting in a narrow margin or a positive margin. Our study found no difference in R0 resection rates between the groups, regardless of whether ICG fluorescence navigation was used, but a wide margin was easier to attain in the ICG-FN group. Spearman’s rank correlation and stepwise logistic regression analysis suggested that ICG fluorescence navigation was an important correlation factor and predictor of a wide margin. As a wide surgical margin should be a target of resection of malignant liver tumors [24‐27], these results indicated that the patients in the ICG-FN group might have a lower postoperative recurrence rate and longer postoperative survival, although this still needs to be verified by long-term follow-up data.

In conclusion, laparoscopic hepatectomy is now performed widely because of its minimal invasion and enhanced recovery, although it requires skill, training, and practice by surgeons. As an emerging technology, ICG fluorescence navigation is safe and efficient in laparoscopic hepatectomy. It also helps to achieve a wide surgical margin, which often results in a better prognosis. The findings of this study suggest that the administration of ICG closer to the operation, 0–3-days preoperatively, is acceptable.