Robotic Versus Open Extrahepatic Biliary Reconstruction for Iatrogenic Bile Duct Injury

The majority of biliary surgery for benign disease is undertaken via the traditional “open” approach. Two early reports regarding laparoscopic and robotic repair for biliary injury after laparoscopic cholecystectomy showed acceptable results.^1,2 Overall, however, clinical data are lacking in regard to the minimally invasive approach, partly due to technical challenges associated with the operation itself. The robotic system facilitates fine suturing with improved dexterity under a magnified 3-dimensional visualization. These advantages overcome some of the technical limitations associated with conventional laparoscopy. Since 2016, we have adopted robotic techniques in our program almost exclusively for all hepatopancreatobiliary resections. We aim to compare our short-term outcomes of patients undergoing a robotic or “open” biliary reconstruction for the treatment of iatrogenic common bile duct injuries. We hypothesize that the robotic approach to biliary reconstruction is safe, feasible, and effective compared to the traditional “open” approach.

After an Institutional Review Board (IRB) approval, we prospectively followed 14 patients who underwent robotic and “open” bile duct reconstruction with Roux-en-Y (RY) hepaticojejunostomy or, in few instances, choledochoduodenostomy (n = 1) or primary repair (n = 1) without RY reconstruction. Demographic and perioperative variables were collected and compared utilizing Mann-Whitney U test/Student’s t-test and Fisher’s exact test. Postoperative complications are defined by Clavien-Dindo classification. For illustrative purposes, data are presented as median (mean ± SD).

Fourteen consecutive patients undergoing robotic (n = 8) vs “open” (n = 6) bile duct reconstruction for iatrogenic bile duct injury were included. Patients undergoing the robotic vs “open” approach were 50 (52 ± 16.3) vs 56 (57 ± 20.5) years of age with a BMI of 28 (29 ± 7.4) vs 31 (31 ± 5.6) kg/m². American Society of Anesthesiology (ASA) class of 2 (2 ± .5) vs 3 (3 ± .5) (P=.003). Operative duration was shorter 259 (297 ± 72.6) minutes for the robotic cohort vs 331 (343 ± 81.0) minutes for the ‘open” cohort. Estimated blood loss (EBL) was lower for patients undergoing robotic reconstruction 50 (70 ± 28.3) vs 150 (260 ± 253.5) mL (P = .05). There were no intraoperative complications for either approach. There was only one postoperative complication with a Clavien-Dindo score greater than III after an “open” bile duct reconstruction, which is related to respiratory distress requiring reintubation and intensive care unit (ICU) admission. This patient had a significant history of chronic obstructive pulmonary disease (COPD) originating from 50 years of heavy tobacco smoking. We did not see any postoperative bile leaks. There were no in-hospital or 30-day mortalities. For patients undergoing robotic vs “open” approach, 0 vs 1 patient was readmitted within 30 days of hospital discharge due to abdominal pain. This readmission was resolved with conservative management utilizing intravenous fluid and oral narcotic pain medication. With a median follow-up of 22 (27 ± 19.5) months, an anastomotic stricture occurred at 13 and 10 months in one patient undergoing robotic and one patient undergoing “open” approach, respectively. Both patients underwent successful percutaneous biliary dilations by interventional radiology service.

Bile duct injury is a known complication of laparoscopic cholecystectomy (LC). More than 750 000 laparoscopic cholecystectomies are performed every year in the United States. ² Several recent studies have assessed the risk of bile duct injury (BDI) from LC and the incidence of BDI varies between .09% and .3%.^3,4 Studies have also shown that patients who required bile duct reconstruction were ten times more likely to die within a year when compared to those who did not experience complications after LC. ³ Evidence-based recommendations for safe cholecystectomy and prevention of BDI have been put forward after a consensus conference of multiple surgical societies (SAGES, AHPBA, SSAT) held in Boston, MA in 2018. ⁴ Due to such initiative and improvement in the cognitive skill of general surgeons, the incidence of BDI continues to decrease over the last two decades. BDI also leads to grave consequences for the operating surgeon as it remains one of the most common causes of legal litigation against general surgeons. ⁴

Although most patients with BDI are still reconstructed by hepatobiliary/transplant surgeons via traditional “open” technique, minimally invasive surgery has been reported in a few small case reports.^3,4 Laparoscopic RY hepaticojejunostomy is feasible in expert hands but is technically challenging due to the complex nature of the dissection and the requirement of fine suturing. Valezquez et al described his experience of 24 patients who underwent laparoscopic RY hepaticojejunostomy successfully with no mortality; however, the incidence of bile leak was high (17% of patients). It is widely known that anastomotic leak is a predisposing factor for later anastomotic stricture requiring reintervention.

The robotic surgical platform enhances precise hand movements while neutralizing tremors, thus facilitating very fine dissection in small spaces. A key aspect of any biliary reconstruction is a tension-free anastomosis between a well-perfused hepatic duct and small bowel. It is very important to identify all bile duct sectors when planning for biliary reconstruction. All major bile ducts larger than 3-4 mm need to be drained and this sometimes requires more than one anastomosis. This kind of reconstruction, with multiple fine biliary anastomoses, is greatly facilitated by the robotic system with acceptable outcomes. Our comparative data showed that the robotic approach leads to significantly lower estimated blood loss when compared to the “open” approach while eliminating the need for ICU admission (Table 1).

OPEN IN VIEWER

Table 1.

Demographic and Perioperative Variables.

Robotic versus open biliary reconstruction
Variables	Robotic	Open	Total/P-values
# Of patients	n = 8	n=6	n = 14
Age (years)	50(52 ± 16.3)	56(57 ± 20.5)	P = .62
Sex (W/M)	5W/3M	4W/2M	P = 1.00
BMI (kg/m2)	28(29 ± 7.4)	31(31 ± 5.6)	P = .59
ASA class	2(2 ± .5)	3(3 ± .5)	P = .003*
Intraoperative variables
Roux-en-Y hepaticojejunostomy (n)	7	5	P = 1.00
Primary bile duct repair without reconstruction (n)	0	1
Choledochoduodenostomy (n)	1	0	P = 1.00
Operative duration (min)	259(297 ± 72.6)	331(343 ± 81.0)	P = .29
Estimated blood loss (mL)	50(70 ± 28.3)	150(260 ± 253.5)	P = .05 *
Intraoperative complications (n)	0	0	P = 1.00
Postoperative Variables
ICU Length of stay (Days)	0	0(1 ± 1.3)	P = .05 *
Clavien-Dindo complications grade ≥ III (n)	0	IVa (1)	P = .43
Biliary anastomotic leak (n)	0	0	P = 1.00
Length of stay (days)	4(4 ± 1.4)	5(5 ± 1.2)	P = .18
30-day readmission (n)	0	1	P = .43
Anastomotic stricture (22-month follow-up)	1	1	P = 1.00

In our hepatobiliary program, a tertiary referral center, we frequently accept postcholecystectomy biliary complications from surrounding community hospitals. We apply the robotic approach as our preferred method for biliary resection and reconstruction, including reconstruction for BDI, choledochal cyst excision, perihilar/extrahepatic cholangiocarcinoma resection, and pancreaticoduodenectomy. For BDI repair, we routinely use both robotic ultrasound and intraoperative cholangiogram to confirm biliary anatomy and the type of injury (with or without a concomitant vascular injury). In our center, reconstruction is undertaken in a RY hepaticojejunostomy fashion. We utilize all four robotic arms and a Gelpoint™ (Applied Medical, Rancho Santa Margarita, CA, USA) in the right lower quadrant for a bedside surgeon to assist with suctioning and exposure. Most of the porta hepatic dissection is undertaken using fenestrated bipolar forceps and hook electrocautery/scissors electrocautery. Once a 60-cm Roux limb has been created and brought up antecolically, we use two 9-inch 3-0 running V-Loc™ (Medtronic, Minneapolis, MN, USA) stitches to create the posterior wall followed by the anterior wall of the hepaticojejunostomy anastomosis. The sutures are started at nine o’clock position and ended at three o’clock position, where eventually both of them are tied together with the knot placed on the outside. A closed suction drain is placed routinely near the anastomosis. We find this technique technically simple, safe, and reproducible with minimal postoperative complications such as bile leak and anastomotic stricture. Most of the patients are discharged home on postoperative day four without an abdominal drain.

In conclusion, the minimally invasive technique utilizing a robotic system for biliary reconstruction is safe and feasible with excellent short-term outcomes. Robotic BDI repair leads to significantly lower blood loss and eliminates the need for ICU admission. The robot is our “tool of choice” for minimally invasive hepatobiliary surgery. We advocate that this modern approach should be considered an alternative to the traditional “open” method when local expertise is available.