A New Way of Reconstruction in Pancreaticoduodenectomy. “The Anatomical Sequence”. Preliminary Report

The number of previously reported cholangitis episodes in choleco-duodenostomy (CD)is according to some authors around 10% [6]. This technique (CD) has been the probable origin of the theory that the anatomical reconstruction favors the reflux of food into the biliary tract, the presence of the blind sac between the anastomosis and the papilla would justify stagnation and subsequent ascending infection. Some authors in 2146 patients undergoing choledoco-duodenostomy report an incidence of 0.73% [18]. Patil et al. [11] used hepatic-duodenostomy as a biliarydigestive anastomosis in the treatment of choledochal cysts in 56 patients and report only one cholangitis secondary to stenosis of the anastomosis 18 years after its creation (0.56%), which would cast doubt on reflux as the cause of cholangitis. These authors performed the anastomosis with the duodenum, 2 cm from the pylorus and always at the level of the biliary crossroads, although they do not report the diameter of the same. On the other hand, the experience with Billroth 1 type reconstructions in gastrectomies has been abandoned because the biliary reflux from the duodenum to the gastric remnant could become disabling and this would also justify the conduct of placing the biliary anastomosis before the gastric one. There is little literature referring to biliary complications of PD. Cholangitis is reported by different authors between 1 and 18.6%. (2-6, 8, 9, 19, 20), most of the episodes appear within the first 30 days [4,9].

Ueda et al. (19) among 18 patients with refractory cholangitis, report that in 17 of them the same appeared in the first year after surgery. End-to-side hepatic-duodenostomy is the procedure that most closely resembles the anatomic sequence reconstruction we have used. In a comparison between hepatic-jejunal anastomosis and hepatic-duodenostomy in the treatment of choledochal cysts, the authors report that patients with hepatic-jejunal anastomosis had more cholangitis than those who underwent hepaticduodenostomy (15% vs 3%) [21]. In the same sense, but comparing both anastomoses in liver transplantation, the percentage of cholangitis was the same between both anastomoses (HD 14% vs HJ 12%) [22]. It is likely that pyloric preservation in our patients plays a fundamental role both in avoiding cholangitis as well as in avoiding bile reflux into the stomach and the consequent alkaline gastritis or even esophagitis. Among the causes of cholangitis in PD, anastomotic stenosis, stones, intestinal obstruction, afferent loop syndrome and jejunal peristalsis disorders are the most frequently reported [6,9]. The cause of early episodes of cholangitis, which are the most frequent, [4] could be attributed to minimal biliary stenosis due to acute inflammation, ileus, peristalsis disorders and contamination by resistant germs, [9] while late episodes would be associated to a greater extent with anastomotic stenosis. Some authors [8] have found that the main predictive factor of stenosis is a thin bile duct. Other authors [20] reaffirm this concept by stating that a bile duct less than 15 mm in diameter is a risk factor for stenosis and based on this they propose performing a hepaticoplasty to increase the diameter of the anastomotic mouth and promote a greater flow of bile into the intestine. Our patients had a mean diameter of 12 mm and there were no episodes of cholangitis in the mean follow-up of 13 months.

Other factors associated with postoperative cholangitis are resection for benign pathology, prolonged surgery time and persistent elevation of alkaline phosphatase [19], the latter with a value greater than 440 should lead to expect a stenosis [6]. The use of preoperative percutaneous biliary drainage or stents would favor the appearance of cholangitis by causing micro-traumas in the biliary tract [6,10]. Because we agree with this last observation, we leave the percutaneous drains preferably before the biliary crossroads functioning as external biliary drainage, we have noticed that in those patients with drains that run through the common bile duct they produce traumatic choledocytitis that, depending on the time it is left in place, makes its dissection difficult in surgery and forces an anastomosis to be made over an inflamed biliary edge. We have used the anatomical sequence in 16 patients with no episodes of cholangitis or bile reflux into the stomach to date, and this new arrangement seemed more harmonious to us. It facilitates endoscopic access to the biliary tract in a simple manner, a very important factor since, as far as we know, endoscopic access in other types of reconstructions is extremely cumbersome, [23] expanding therapeutic procedures. In the imaging studies we were able to observe a small amount of reflux with complete filling of the stomach and a significant portion of the jejunum in the patients in whom we performed serial esophageal-gastro-duodenal radiographs.

In conclusion, the proposed technique is novel. [24] There are still a few cases for a definitive assertion, this should be increased and will probably require multi-center participation for this objective. In future communications we will try to make up for the lack of more in-depth complementary methods to confirm the absence of cholangitis and gastric reflux, as well as the performance of an ERCP to determine one of the strong points of this reconstruction, which is biliary access by endoscopy.