Transillumination of Calot’s Triangle on Laparoscopic Cholecystectomy: A Feasible Approach to Achieve a Critical View of Safety

Background Laparoscopic cholecystectomy is currently one of the most commonly performed procedures globally. Morbidity of laparoscopic cholecystectomy is low; however, bile duct injury is still a feared complication. Despite worldwide efforts, the global incidence of bile duct injury remains higher for laparoscopic cholecystectomy compared with open cholecystectomy. Despite the general belief that the learning curve and lack of laparoscopic skills represent the most common causes of bile duct injuries, the principal cause is the misidentification of biliary anatomy. The aim of our study is to determine if laparoscopic transillumination is a feasible approach to bile and vascular structures visualization during laparoscopic cholecystectomy because the only other method for real-time visualization is fluorescent cholangiography, which can be cost-prohibitive and requires specialized equipment and training. Materials and methods We performed a retrospective comparison of outcomes between the transillumination approach in 10 patients receiving laparoscopic cholecystectomy (group A) and a control group of 50 conventional laparoscopic cholecystectomy patients (group B). We compared demographic data, type of surgery, operative time, bleeding, intraoperative and postoperative complications, and hospital stay. We used a conventional four-port positioning for laparoscopic cholecystectomy, and a 5-mm/30° scope was used as a light source and placed behind the area identified as Calot’s triangle. Results Group A consisted of 10 patients (9 women, 1 man), with a mean age of 50.7 (± 17.4) years. The mean body mass index (BMI) in group A was 26.8 (± 0.65) kg/m2. In group A, three of the cholecystectomies were conducted as emergency procedures. Group B consisted of 50 patients (40 women, 10 men), with a mean age of 49.7 (±15.2) years. The mean BMI in group B was 27.5 (±4.5) kg/m2, and two cholecystectomies were emergency procedures. In comparing the transillumination approach with conventional cholecystectomy, we found no statistical differences in operative time, bleeding, complications, or mean hospital stay. Conclusions Laparoscopic transillumination is a feasible method for real-time visualization of Calot’s triangle structures. Our initial experience with transillumination did not provide better outcomes than conventional cholecystectomy.


Materials and methods
We performed a retrospective comparison of outcomes between the transillumination approach in 10 patients receiving laparoscopic cholecystectomy (group A) and a control group of 50 conventional laparoscopic cholecystectomy patients (group B). We compared demographic data, type of surgery, operative time, bleeding, intraoperative and postoperative complications, and hospital stay. We used a conventional four-port positioning for laparoscopic cholecystectomy, and a 5-mm/30° scope was used as a light source and placed behind the area identified as Calot's triangle.

Results
Group A consisted of 10 patients (9 women, 1 man), with a mean age of 50.7 (± 17.4) years. The mean body mass index (BMI) in group A was 26.8 (± 0.65) kg/m 2 . In group A, three of the cholecystectomies were conducted as emergency procedures. Group B consisted of 50 patients (40 women, 10 men), with a mean age of 49.7 (±15.2) years. The mean BMI in group B was 27.5 (±4.5) kg/m 2 , and two cholecystectomies were emergency procedures. In comparing the transillumination approach with conventional cholecystectomy, we found no statistical 1 2, 3

Introduction
Laparoscopic cholecystectomy (LC) is currently one of the more common surgical procedures worldwide. In 2006, in the United States, approximately 917,000 LCs were performed [1]. LC is a safe procedure with low complication rates, especially when it is performed electively. Nevertheless, bile duct injury (BDI) remains a serious complication that concerns both the patient and the surgeon due to its impact on quality of life, survival, and medicolegal issues. The incidence of BDI is in LC 0.32% to 0.52%, which is high compared to the 0.1% to 0.2% rate reported for open cholecystectomy [1]. However, the incidence of BDI has declined; from 1994 to 1999, the pooled incidence was 0.52% to 0.84%, and from 2010 to 2014, the incidence was 0.02% to 0.4% [1]. The incidence is higher in developing countries [1,2].
Many techniques have been proposed to obtain intraoperative imaging of the bile ducts and vessels during LC; however, only fluorescent cholangiography (FC) offers a real-time visualization, however, it can be cost-prohibitive and requires specialized equipment and training. We conducted this study to describe the feasibility of an alternative technique to visualize the structures of Calot's triangle in real-time, guiding the dissection to achieve a successful critical view of safety (CVS).

Materials And Methods
We conducted a single-center retrospective analysis of our prospectively constructed database, evaluating perioperative results from 10 non-consecutive patients in whom LC was performed using transillumination of Calot's triangle (group A) by two experienced surgeons, with 50 patients (group B) who received conventional LC. Patients were matched for age, sex, and body mass index (BMI). Demographic data, type of surgery, operative time, bleeding, intraoperative and postoperative complications, and hospital stay were analyzed. When data were not available, we reviewed the patients' medical records.
Data are presented as mean and standard deviation for continuous variables, and categorical variables are expressed in percentages. Student's t-test and chi-square tests were employed to evaluate group differences. We considered a p-value lower than 0.05 as statistically significant.

Transillumination technique
We used an open (Hasson) technique to enter the abdominal cavity with a transverse supraumbilical incision of 10 to 12 mm, then CO 2 pneumoperitoneum was initiated at 12 mmHg, a 10-mm/30° scope was introduced, and three additional ports were placed under direct visualization: a 10-mm subxiphoid port and two 5-mm subcostal ports.
After an initial laparoscopy, the camera directly faced the gallbladder, and early exposure was achieved using a 5-mm grasper to retract the infundibulum in a caudolateral direction. After that, a 5-mm/30° scope was introduced through the most lateral 5-mm port, placing it behind Calot's triangle. To achieve transillumination of the area, the light of the 5-mm scope had to be rotated to face the triangle, and the 10-mm scope's light has to be gradually decreased until visualization of the aimed structures is achieved ( Figure 1). Depending on the anatomy of the area, the cystic duct (CD) and cystic artery (CA) may be identified. Another elective maneuver to clarify the anatomy in an attempt to identify the common bile duct (CBD) was performed by carefully advancing the 5-mm scope to locate the CBD's 3 o'clock and 9 o'clock arteries. Once this procedure is concluded, the 10-mm scope's light returns to optimal intensity, and the 5mm scope is removed from the abdominal cavity to continue dissection (Figures 2, 3). However, when the visualization of the structures cannot be achieved, the posterior peritoneal layer could be opened followed by another attempt of transillumination ( Figure 4). Dissection during transillumination is not advisable because the visualization of adjacent structures is reduced.  Thin arrow: cystic artery; thick arrow: cystic duct; star: structure free tissue.

Results
Group A consisted of nine women and one man, with a mean age of 50.   Data expressed as mean ± SD or as percentage (%).
The one complication noted in group A was Clavien-Dindo grade I. Specifically, it was a contained abdominal wall hematoma secondary to local anesthetic infiltration for the subxiphoid port, with no need for additional intervention. Of the four complications noted in group B, three were Clavien-Dindo grade I, including a Strasberg A BDI, treated conservatively, a hepatic subcapsular hematoma with no need of further interventions, and one liver bed bleeding with no need for transfusions or interventions. The other patient had a Clavien-Dindo grade III, developing a biloma that needed percutaneous management.
Group A had less bleeding than group B (28.5 ± 17.4 mL vs. 33.2 ± 32.5 mL) and longer operative time (90.1 ± 29 minutes vs. 82.9 ± 27 minutes). The difference was not statistically significant.
Visualization of structures was accomplished in all the procedures in group A. For the seven elective procedures, initial visualization of structures was achieved in two cases; visualization was achieved in the other five cases after the removal of the peritoneal layer. For the three emergency procedures, initial dissection was needed to achieve structure visualization with transillumination ( Table 3).

Discussion
The leading cause of a BDI was initially attributed to the learning curve and lack of technical skills, knowledge, or judgment in cholecystectomies. However, this cause accounts only for 3% of the cases [3]. The principal cause of BDI (97%) is the misinterpretation of biliary anatomy [3]. Injuries often arise from the misidentification of the CBD for the CD or by dissecting too close to the CBD or common hepatic duct (CHD). In such cases, the biliary anatomy was not fully recognized given the presence of connective tissue or inflammation; therefore, the wrong duct is separated, and the surgeon proceeds to remove the gallbladder from the liver. Any duct identified on this final step is often reported as an accessory structure, and the right hepatic artery (RHA) ends up being mistaken for the CA [3,4].
In 1995, Strasberg et al. proposed the CVS for conclusive anatomical identification to prevent BDI. CVS has since been validated worldwide and endorsed by many surgical societies as a feasible approach to a safe cholecystectomy [5,6]. Since the introduction of the CVS, many studies have validated its feasibility and utility in diminishing BDI, with some series even reporting zero incidences of BDI [7]. Eighty percent of BDIs occur while attempting a CVS [8], and Strasberg et al. published a three-step roadmap for avoiding BDI that consists of the following sequence [9]. First, surgeons should try to get a secure, anatomical identification of key structures. When this is not possible, the surgeon should not attempt a total cholecystectomy ("inflection point"). For finishing the operation safely, Strasberg's last point includes conversion or bail out procedures such as cholecystostomy, subtotal cholecystectomy, either fenestrated or reconstituting, conversion, or even concluding the procedure and referring the patient appropriately [9].
Mascagni et al. emphasized a maneuver described by Strasberg et al. to achieve a complete CVS when cystic structures are short or the CA enters high into the gallbladder [10]. The maneuver consists of identifying the CA and dividing it to achieve a better exposition of the CD and, subsequently, the correct CVS. However, misidentification of the structures could lead to RHA injury or the CD itself, and, in fewer cases, a BDI could present [10]. Therefore, knowledge of the arterial and biliary anatomical variations is needed, and physicians should consider that in only 81.5% of cases the CA is found in the hepatocystic triangle [11,12].

Fluorescent cholangiography
Many studies have reported good results with the use of FC [4,[13][14][15][16]. This method requires administration of indocyanine green (ICG), commonly 2.5 mg or 0.05 mg/kg, 30 minutes before surgery [15]. When near-infrared light is applied to the bile ducts, ICG generates fluorescence, and bile structures can be differentiated from surrounding tissue. An angiography may be performed by administrating a second bolus of ICG during surgery. After 30 seconds to two minutes, the CA might be observed by fluorescence. However, the procedure requires specialized equipment consisting of a small control unit, a charge-coupled device camera, a xenon light source, and a 10-mm laparoscope with specially coated lenses capable of transmitting near-infrared light [4].
The detection of the biliary system with FC differs among structures. In one study, CD was identified in 71.4% to 100% of the cases, CHD was identified in 33.3% to 100%, and the CBD was identified in 50% to 100% [4]. For studies comparing results before and after dissection, the sensitivity increases posterior to dissection for the CD from 72% to 95.1% [14]. In another study, the CA was identified in 85.9% of cases when a second bolus of ICG was administrated [15]. However, two conditions might diminish FC sensitivity: obesity and acute cholecystitis [4,15,16].
When compared with intraoperative cholangiography (IOC), there is moderate-quality evidence that visualization of the CD and CBD improves with FC (relative risk [RR]: 1.16 and 1.00, respectively), and low-quality evidence for CHD visualization (RR: 0.76) [15]. Furthermore, routine use of IOC for LC is debatable due to IOC's limited availability and the requirements of staff trained on the equipment operation. Also, prior dissection and incision of the CD are required. Finally, both the patient and staff members are exposed to radiation [15].
To our knowledge, there are no previous records in the English literature of laparoscopic transillumination; the only precedent in the literature is from 2005 [17], and in 2008 [18], Nychytaĭlo et al. performed verification of biliary ducts and vessels by transillumination on LC and open cholecystectomy with a patented device with promising results. However, no images were provided to clarify the technique applied by the authors, and this method is rarely used [17,18].
In our study, the CA and CD were identified in 10 consecutive LCs by transillumination. In the majority of patients (8/10), the peritoneum covering Calot's triangle needed to be opened prior to successful visualization. There were no biliary injuries in the transilluminated group compared with two in the conventional cholecystectomy group (2/50).
The fact that group A had significantly more emergency procedures might explain the mean longer operative time. Also, despite the higher proportion of emergency procedures in group A, the lower bleeding might indicate a better bleeding control due to the direct visualization of the vascular structures by transillumination and the recognition of a loose safe dissection area.
The main limitations of our study were the small sample size for group A and the retrospective, single-center design. There was also a level of inter-surgeon variability. Future studies could mitigate these limitations by using a prospective case-control study with larger sample sizes and standardizations to reduce inter-surgeon variability.

Conclusions
Calot's triangle transillumination is a promising approach for achieving a CVS during LC and represents an additional option for surgeons to visualize Calot's triangle structures in order to improve dissection safety. This procedure helps to guide Calot's triangle dissection as it highlights loose areas where safe dissection could be achieved; however, it does not replace the general safety principles for LC. The potential benefits of our transillumination method are its reproducibility and feasibility, and it could be attempted with alternative devices rather than a second scope as a light source. More studies on different centers are needed to optimize the procedure and validate this technique.

Additional Information Disclosures
Human subjects: Consent was obtained by all participants in this study. Not aplicable issued approval Not aplicable. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.