الفهرس 
ContentsOnly 14 pages are availabe for public view
 |  | from | 142 |  |  |
| | | from | 142 | | |
Abstract
Roux-en-Y hepaticojejunostomy is considered the procedure of choice for biliary reconstruction. However, hepaticojejunostomy anastomotic stricture (HJAS) remains a frequent long-term complication even in hepatobiliary centers [7, 8, 100, 139, 179]. HJAS is a serious complication and, if untreated, will result in cirrhosis, portal hypertension and eventually hepatic failure from repeated cholangitis and intrahepatic stones Revision of hepaticojejunostomy, performed to manage HJAS, is a technically challenging procedure that usually necessitates a high anastomosis in hilar and intrahepatic board. This carries a morbidity of approximately 25% and a mortality of about 2%. In addition, success rate of surgical repair decreases with each attempt for surgical intervention Percutaneous transhepatic balloon dilation for HJAS is an alternative option to redo-hepaticojejunostomy, but has a high failure rate. Reasons for failure include presence of coagulopathy, failure to gain access to a non-dilated biliary system, the need of serial dilation and absence of high expertise Endoscopic management (via either balloon dilatation or stenting of the stricture) with use of double balloon or single balloon enteroscope is a safe, highly effective option in patients with HJAS [14, 16, 184-186]. However, the endoscopic access to the anastomotic site is difficult due to the lengthy pathway in Roux-en-Y reconstruction. Reports of successful endoscopic intervention are still few with limited number of cases. The need for high expertise of the endoscopist, and unavailability of the enteroscopes in all endoscopy units are the major limitations. Many modifications of hepaticojejunostomy with provision of an ”access loop” were developed to shorten the distance traveled by the endoscope in the jejunal loop to reach the hepaticojejunostomy anastomotic site. The earliest description of the access loop formed adjunct to hepaticojejunostomy involved the fashioning of cutaneous stoma in the Roux limb of hepaticojejunostomy (hepaticocutaneousjejunostomy). This technique was abandoned due to excessive bile loss and unpleasant side-effects of stoma with associated cutaneous irritation and excoriation [187-189]. Routine early closure of the stoma has been recommended to avoid these side-effects after removal of retained stones following surgery [190]. Stoma closure is associated with significant complications of wound infection, fistula formation and parastomal hernia. Furthermore, many patients required reopening of the stoma for removal of recurrent stones after only 2 years of follow-up In 1977, Kan and Tsung-chih describe a modified hepaticojejunostomy with subcutaneous access loop for treatment of intrahepatic stones. The technique was designed as a good alternative to the hepaticocutaneousjejunostomy to provide an easy and quick access to the biliary tree to remove stones repeatedly without the need for surgery. In his case-series of six patients, only 2 cases needed incision of the subcutaneous blind loop for removal of residual calculi [21]. In 1978, Chao et al. reported a study in which modified hepaticojejunostomy with subcutaneous access loop was used in 116 patients for treatment of intrahepatic stones. Cho¬langiography and/or choledochoscopy can be done through the subcutaneous conduit of the choledocho-jejunostomy to follow the course of the disease post¬operatively. Retained stones, worms, ductal stenosis, or suppurative cholangitis were found 41 times in 53 cases over a period of 2 months to 6 years postoperatively. Except for 4 failures, the rest were treated successfully by Dormia basket or other instruments through this route [22]. In series reported by Kassem et al. forty-two patients with intrahepatic stones underwent surgery with construction of a subcutaneous hepaticojejunal access loop. There were no specific complications attributable to the construction of the access loop. The subcutaneous access loop was used to gain access to the biliary tree in 28 patients with residual or recurrent stones. A total of 55 procedures (range 1-5) were attempted with successful access achieved in all cases and successful stone clearance in 21 of the 28 patients, and all of them were symptom free for at least 12 months after the last procedure. Partial stone clearance (failed group) was achieved in the remaining seven patients. Six of them had different degrees of biliary strictures Until recently, the modified hepaticojejunostomy with subcutaneous access loop is still in the zone of case series. The technique is not standardized in all studies Variations present among the studies regarding the exact position of the access loop at the anterior abdominal wall (epigastric or right subcostal, and subparietal, subfacial or subcutaneous) and method of intervention (endoscopic, radiological or both) Also, the advantage of modified hepaticojejunostomy with subcutaneous access loop to provide a percutaneous permanent biliary access is obvious, this technique in not free of major disadvantages. Opening and closing the skin and jejunal loop were needed after each endoscopic procedure with the risk of wound infection or jejunal fistula. Also, subparietal or subcutaneous placement of the closed end of the afferent jejunal loop resulted in difficulty of transjejunal endoscopy with the need of fluoroscopic assistance and a risk of failure to enter the access loop. Some patients may need more than single intervention in a short period to gain satisfactory results which is disappointing to the patients due to re¬peated opening and closure of the loop [22, 191]. In our study, we fixed the access loop in right subcostal region as it is the most popular position in the recent studies, technically feasible and seems comfortable to the patient during intervention. We supposed that the subcutaneous position of the access loop to be the site of choice of endoscopic intervention. We did it in the first six cases of our series and the subcutaneous access loop could be felt by palpation if the patient is thin. Then, we modified the position of the closed access loop, in the remaining cases, to be fixed beneath the aponeurosis in the same region (subfacial position). This is due to patient dissatisfaction from distention of the subcutaneous loop and bloating in the right subcostal region occurring after meals. This characteristic drawback is not reported in previous studies. We are convinced that the subcutaneous position of the access loop is not the best to provide a good quality of life for these patients. In addition, the subcutaneous loop has the risk of parastomal hernia on long-term follow-up, which we found in two cases in our study. On the other, the subfascial position avoid these drawbacks but on the expense of feasibility of endoscopic access. This appeared on the two cases that needed an endoscopic intervention for HJAS. We found that the metallic clips marking are not largely beneficial in this situation due to the two-dimensional image of fluoroscopy. Al-Ghnaniem and Benjamin used a circular steel wire as radiological marker for the access loop in their study[192]. Modified hepaticojejunostomy with gastric access loop was first innovated by Sitaram et al in 1998 [18]. Ten patients had undergone gastric access loop. Access loop was entered easily with the gastroscope in five patients in whom it was attempted. The follow-up was between 3 and 24 months. Jayasundara et al. reported 27 patients who had hepaticojejunostomy with gastric access loop. Three patients (11%) developed strictured HJ that were successfully managed endoscopically through the access loop. During the follow up period of about two and half years, they did not discover gastric access related morbidity using the dyspepsia disability score [19]. In a series with 16 cases, Hamad MA and El-Amin H assessed different construction of gastric access loop in the form of bilioenterogastrostomy the overall success rate of endoscopic access to the HJ through the three types of BEG was 87.5%, while it was 100% for BEG type III, which is a construction similar to the previous series. The mean follow up period was 23 months. They had one strictured HJ (6.25%) which was successfully treated endoscopically by stenting [20]. Hepaticojejunostomy with gastric access loop did not gain popularity by most of hepatobiliary centers and until recently, only few publications studied the technique. This is probably due to four reasons: 1) the fear of biliary gastritis from the presence of bile in stomach, 2) the risk of cholangitis from food particles entering the access loop, 3) Adding another anastomosis to the procedure with the risk of gastric fistula, and finally 4) the therapeutic benefit of endoscopic access to HJ anastomosis is questionable. In our study, only two patients in RYHJ-GA reported dyspeptic symptoms shortly after the procedure which were managed effectively with PPI. Neither of them nor other patients who underwent endoscopic evaluation were found to have endoscopic evidence of gastritis. These results are in agreement with previous studies [18-20, 197]. Also, there is no case of GJ stomal ulcer, a complication reported in one patient of sitaram et al series [18]. On the other hand, most of the patients show presence of bile in the stomach. We believe that the nature of pure bile content in the stomach of our patients is totally different from that of the duodenogastric reflux which contains a mixture of bile and pancreatic enzymes. The presence of activated pancreatic enzymes is responsible for the injurious effects in reflux biliary gastritis and esophagitis [198, 199]. Moreover, the amount of bile is in the stomach is significantly small in comparison to the copious amount of bile in the access loop (which is shown once the endoscope entering the GJ). This may be explained by the strong gastric muscular layer which contracts during digestion closing the opening of GJ. Also, the bile flow to the stomach is upstream against the peristalsis of the jejunal loop. The hypothesis that RYHJ-GA has a risk of cholangitis due to entering of stomach contents into the access loop contradicts the results of our study and previous ones [18-20, 197]. No patient without stricture was reported to have cholangitic attacks. It is thought that during gastric emptying, the contracting gastric musculosa directs food to the patent dependent pyloric orifice rather than the relatively closed GJ. In our study, there is no case of gastric fistula. This is in consistent with the previous studies [18-20, 197]. We suppose that the risk of gastric fistula is similar to the risk of intestinal fistula from enteroenterostomy, which is extremely rare in elective settings. Regarding perioperative data, patients who underwent RYHJ-SA had comparable operative time (219.9 ± 34.2 vs 235.2 ± 21.4 (minutes); P= 0.059) and blood loss (312.5 ± 36.1vs 297.3 ± 18.9; P= 0.06) to those underwent RYHJ-GA. The variation in operative time largely depends on the difficulty in hilar dissection rather than adding a GJ anastomosis or subcutaneous access loop to the standard procedure. Also, Ileus, bile leakage, wound infection, pulmonary complications and length of hospital stay were comparable between RYHJ-SA group and RYHJ-GA group. No reported mortalities in the first 30 postoperative days. Post-operative biliary leakage in the present study was 17.3% which is in the upper range reported in literature [179, 200, 201], but insignificantly different in both groups (P=0.71). Probably, the demanding anastomoses with narrow ducts or at a high level of hilar board performed in most cases of our series can justify these results. All cases of bile leak stopped spontaneously without intervention. The endoscopic management via the access loop is greatly different between both groups. In RYHJ-SA, endoscopy into the access loop was successful in only one case of three patients with HJAS in whom therapeutic intervention was attempted. Unfortunately, the endoscopy failed to dilate the stricture. Two cases underwent redo-hepaticojejunostomy with uneventful follow-up. Due to the medicolegal issue, the endoscopic intervention, in our and previous studies, was of therapeutic intent and not for prophylactic evaluation of HJAS. This has a disadvantage of postponing the endoscopic dilatation until HJAS being very tight and non-dilatable. In Kassem series, 6 cases failed to benefit from subcutaneous access loop to remove intrahepatic stones due to HJAS [191]. As the majority of previous reports were using RYHJ with subcutaneous access loop to provide a biliary access to treat recurrent intrahepatic stones where cholangiohepatitis is endemic in their countries, this put a question mark on the actual benefit of using the subcutaneous access loop when HJAS is the major concern on long-term follow-up. In RYHJ- GA, endoscopy into the access loop via GJ was successful in all patients in whom it was attempted. Endoscopy into the access loop via GJ was successful in all patients in whom it was attempted. The endoscopic maneuver was easy with end-view gastroscope and we did not face any obstacle from GJ stricture that was reported in three out of eleven patients in Selvakumar et al series [197]. Three cases of RYHJ- GA needed more than 1 endoscopy due to evidence of HJAS. It should be noted not all patients underwent the first endoscopy at the planned time (3 months postoperative) due to the COVID-19 pandemic waves. Interestingly, the two cases of HJ stricture underwent their first endoscopy at 5 and 7 months. Fortunately, all 3 cases had successful endoscopic dilation and balloon sweeping of biliary mud (one case) or stone (2 cases), without stenting. Although stenting after dilatation of biliary stricture remains a controversial issue [202, 203], we do not prefer stenting as it acts as foreign body in the biliary system, leads to repeated episodes of cholangitis and encourages formation of intrahepatic stones. The follow-up liver functions of all 3 cases returned to normal after successful endoscopic management. The incidence of HJAS was comparable between the two groups (P=0.68). However, the follow-up period in this study is relatively short and we need a longer follow-up period for accurate comparison of HJAS incidence between both techniques, as many patients in RYHJ-GA group underwent prophylactic endoscopic dilatation. Many studies reported that most of HJAS develop within 5 years, and 90% within 7-10 years [139, 181]. Longer follow up of our patients in the RYHJ-GA group who underwent endoscopic dilatation would clarify the benefit, if any, of this dilatation on the avoidance of future stricture formation. Moreover, those who may develop HJAS have the potential benefit of endoscopic management of their condition without the need for revisional surgery. Although our experience in this study is limited to 26 cases of RYHJ-GA and RYHJ-SA, the safety of the gastric access and the easy ability to access the anastomotic site for balloon dilatation has convinced us that RYHJ-GA should be considered for patients in whom HJAS is anticipated such as patients with intra abdominal abscess or bile collection, external biliary fistula, proximal biliary stricture, non-dilated biliary system, and prior attempts of repair. In conclusion, modified RYHJ with gastric access loop (RYHJ-GA) is comparable to modified RYHJ with subcutaneous access loop (RYHJ-SA) regarding complications. However, the gastric access is more practical for endoscopic management of future HJ anastomotic stricture. This modification should be considered in patients with high risk of HJAS stricture during long-term follow-up to avoid the need for revisional surgery. Roux-en-Y hepaticojejunostomy is considered the procedure of choice for biliary reconstruction. However, hepaticojejunostomy anastomotic stricture (HJAS) remains a frequent serious long-term complication even in hepatobiliary centers. Endoscopic management (via either balloon dilatation or stenting of the stricture) with use of enteroscope is a safe, highly effective option in patients with HJAS and is safer than revisional surgery. However, the endoscopic access to the anastomotic site is difficult due to the lengthy pathway in Roux-en-Y reconstruction. This can be overcome if a short “access loop” to bilio-enteric anastomotic site is available. The aim of our study is to compare modified RYHJ with gastric access loop (RYHJ-GA) with modified RYHJ with subcutaneous access loop (RYHJ-SA) regarding short and long term outcomes and, moreover, to evaluate the feasibility and results of future endoscopic access of the modified bilio-enteric anastomosis. Between September 2017 and December 2019, a total of 70 patients were eligible for our study. Eighteen patients were excluded from the study (11 patients have malignant disease, 5 have dilated CBD >20mm and 2 patients refused the study). Fifty-two patients were randomly assigned to RYHJ-SA (n = 26) or RYHJ-GA (n = 26). There were no significant differences between the two groups regarding mean age, gender distribution, ASA class, indications for biliary shunt, mean CBD diameter, results of preoperative liver function tests, and ERCP performance. Patients who underwent RYHJ-SA had comparable operative time (P= 0.059) and blood loss (P= 0.06) to those underwent RYHJ-GA. Ileus, bile leakage, and wound infection occurred more in patients of RYHJ-SA group than RYHJ-GA group, despite not reaching statistical significance. No reported mortalities in the first 30 postoperative days. Incisional hernia occurred in 2 patients with RYHJ-SA and one patient with RYHJ-GA. Parastomal hernia occurs in two patients of RYHJ-SA group. In RYHJ- SA, endoscopy into the access loop was successful in only one case of three patients with HJAS in whom therapeutic intervention was attempted. Two cases underwent revisional surgery in the form of redo-hepaticojejunostomy with uneventful follow-up. The third patient refused any intervention and was managed with medical treatment. In RYHJ-GA group, three cases of 26 cases did not undergo the planned postoperative endoscopy due to patient refusal. One of them returned with HJAS. Endoscopic access to the HJ was successful but the cannulation failed due to too tight stricture with subsequent failure of anastomotic dilatation. The patient underwent revisional surgery. Of the remaining 23 cases, the planned endoscopic assessment was done with ease. Entry to the access loop through the gastrojejunostomy was successful in all cases. Endoscopic evidence of gastritis was absent. Three cases needed more than one endoscopy due to HJAS. All 3 cases had successful endoscopic dilatation and balloon sweeping of biliary mud (one case) or stone (2 cases), without stenting. The incidence of HJAS was comparable between the RYHJ-SA group (3 cases) and RYHJ-GA group (4 cases) (P=0.68). Longer follow up of our patients in the RYHJ-GA group who underwent endoscopic dilatation would clarify the benefit, if any, of this dilatation on the avoidance of future stricture formation. Moreover, those who may develop HJAS have the potential benefit of endoscopic management of their condition without the need for revisional surgery. Although our experience in this study is limited to 26 cases of RYHJ-GA and RYHJ-SA, the safety of the gastric access and the easy ability to access the anastomotic site for balloon dilatation has convinced us that RYHJ-GA should be considered for patients in whom HJAS is anticipated such as patients with intra abdominal abscess or bile collection, external biliary fistula, proximal biliary stricture, non-dilated biliary system, and prior attempts of repair. In conclusion, modified RYHJ with gastric access loop (RYHJ-GA) is comparable to modified RYHJ with subcutaneous access loop (RYHJ-SA) regarding complications. However, the gastric access is more practical for endoscopic management of future HJ anastomotic stricture. This modification should be considered in patients with high risk of HJAS stricture during long-term follow-up to avoid the need for revisional surgery.