Fig. 8.1
a Wide open pylorus after surgical pyloroplasty. b Appearance of a surgical pyloroplasty done more than 60 years previously, in a 100-year old patient
After antrectomy, the remaining mid-stomach needs to be anastomosed to the intestine. Billroth described 2 ways to do this: either the body of the stomach is sewn to the duodenal bulb directly (Billroth I), or to a loop of jejunum just past the ligament of Treitz (Billroth II) (Fig. 8.2). The mnemonic for remembering this is that a Billroth I has one opening from the stomach, whereas a Billroth II has two openings. In reality, one of the openings is actually the afferent limb, carrying pancreatic juice and bile from the duodenum. The other is the efferent limb (the mnemonic here is the E in Efferent stands for Exit). A Billroth II anastomosis can be created in an antecolic (in front of the colon) or retrocolic fashion; however, the difference cannot usually be detected endoscopically (Fig. 8.3).
Fig. 8.2
Diagrams showing the difference between a Billroth I and b Billroth II anastomosis (A afferent limb, E efferent limb)
Fig. 8.3
Endoscopic appearance of a Billroth I anastomosis (gastroduodenal anastomosis) and b Billroth II anastomosis
It is fortunate that in most Billroth II cases the surgeon chooses a site to anastomose the stomach to the intestine that is not too far from the ligament of Treitz. For this reason, the ampulla of Vater is typically reachable, even with a duodenoscope. The afferent limb (to be cannulated to reach the ampulla) is typically smaller and goes “down.” The opening that is easier to get into is usually the efferent limb. The presence of bile is usually another clue that you are in the afferent limb, although on occasion this is not reliable; not infrequently both limbs need to be checked in order to identify the afferent limb.
The appearance of a Billroth I anastomosis is somewhat “nondescript.” Paradoxically, ERCP can be more difficult with a BI compared to a BII because the ampulla is displaced proximally, making proper orientation of the papilla sometimes quite difficult (Fig. 8.4). The papilla is easier to orient properly in patients with Billroth II anatomy, but the orientation of the CBD and PD is turned 180°/inverted (Fig. 8.5), with the bile duct being below and the pancreatic duct above.
Fig. 8.4
Fluoroscopic appearance of ERCP scope in a Billroth I or b Billroth II anastomosis
Fig. 8.5
Billroth II anatomy, view of ampulla with impacted stone, as seen with a forward-viewing endoscope and b side-viewing endoscope
Rarely a “double pylorus” can be encountered (Fig. 8.6); this is not a postsurgical appearance, but basically a gastroduodenal fistula that has formed from a pre-pyloric ulcer penetrating to the duodenal bulb.
Fig. 8.6
Double pylorus; one opening is the native pyloric channel and the other is a gastroduodenal fistula caused by a peptic ulcer
In cases of pyloric or duodenal obstruction from post-ulcer scarring, the gastric outlet can be very stenotic. Rather than resection of the stenosis, a gastrojejunostomy can be created instead. Some endoscopists have described creating a gastrojejunal anastomosis using the lumen opposing Axios stent (Fig. 8.7a–c). It is possible that with the refinement of this technique, it might be done more commonly than placement of a duodenal stent in cases of duodenal obstruction.
Fig. 8.7
Endoscopic gastrojejunostomy with an Axios stent. a Forward-viewing EUS scope used to identify jejunal loop, punctured with needle and contrast injected. b Immediately after Axios placement between the stomach and the jejunum. c View of jejunum through lumen of Axios stent
On occasion, a repeat operation is necessary after peptic ulcer surgery, particularly with the development of one of the common post-gastrectomy syndromes; this is typically conversion of a Billroth I or Billroth II to Roux-en-Y with gastrojejunostomy. The length of the Roux limb is variable, but a limb length of at least 50 cm is necessary for adequate diversion of bile away from the gastric pouch. Roux limb lengths of 50–150 cm are typically utilized, and a biliopancreatic limb of another 50–100 cm can make reaching the ampulla even more difficult than following a Billroth II, often requiring devices to assist in deep intubation or, in some cases, making endoscopic access to the ampulla impossible.
Anti-reflux Surgery
There remains an important role for anti-reflux surgery despite the widespread use and effectiveness of proton pump inhibitors. Very large hiatal hernias can be very symptomatic and are not uncommonly found in patients with dysplastic Barrett’s esophagus (BE). Correction of the hiatal hernia can improve symptoms, and many of these patients may be able to stop taking PPIs. Patients with regurgitation particularly benefit from a fundoplication. There is only limited long-term data of the benefit of repairing a hiatal hernia in dysplastic BE in terms of preventing recurrence or progression of the dysplasia [3]. However, it is known that after BE ablation, recurrence of dysplasia is not a rare event [4]. So theoretically, at least, HH repair may be of benefit in patients with dysplastic BE. In fact, Velanovich reported in his series of patients significant improvements in recurrence of BE and persistence of BE in patients who had a Nissen fundoplication before, at the same time as, or after ablation [5].
There are several different hiatal hernia repairs that have been described, and each may have benefit in specific cases. In the present day, these are almost always done laparoscopically [6]. The standard repair is a Nissen fundoplication, which is a 360° “wrap” of the fundus around the distal esophagus in association with reducing the stomach down from the mediastinum into the abdomen. A Dor fundoplication is an anterior wrap generally between 90 and 180°. This is a less robust procedure regarding control of reflux; but, it is advantageous in situations where a tighter wrap is not desired (i.e., patients with motility disorders). A Toupet fundoplication creates a 270° posterior wrap. The Hill repair recreates the angle of His and does not wrap the fundus around at all and is largely of historical significance. A Collis fundoplasty, or esophageal lengthening procedure, may be required in patients with foreshortened esophagus, where the stomach cannot be easily pulled down into the abdominal cavity.
There are subtle differences in endoscopic appearance of fundoplication between the different surgeries [7]. However, the Nissen is the most common done fundoplication. It has an appearance of “stacked coins” with a deep posterior groove and a shallower anterior groove (Fig. 8.8). A Dor or Toupet is similar, but the wrap does not go completely around. It may not be obvious that a patient had a Collis fundoplasty, as the surgical changes are typically covered by the wrap. As mentioned above, the proper position of a fundoplication is around the distal esophagus. Thus, endoscopically, the GEJ should be located beneath the pinch of the diaphragm/top of the fundoplication (often very close and indistinguishable from one another). If the GEJ can be seen above this level, it likely represents some degree of malposition of the fundoplication which can be associated with dysphagia.
Fig. 8.8
Normal view of Nissen fundoplication with the so-called stack of coins appearance
Resective Surgery for Benign and Malignant Disease
Esophagus
Esophagectomy is pursued in resectable cases of esophageal cancer (usually after neoadjuvant chemoradiation), and on occasion for benign conditions (unusual cases of dysplastic Barrett’s, or end stage achalasia). In all of these cases, the stomach is “tubularized,” then anastomosed to the proximal esophagus. Stricturing of the surgical anastomosis is, unfortunately, common, particularly if there has been a leak postoperatively (Fig. 8.9), and usually requires multiple frequent dilation procedures and/or esophageal stenting. Occasionally, there is Barrett’s epithelium remaining after esophagectomy.
Fig. 8.9
a Anastomotic stricture after esophagectomy. b After incisional treatment with insulated tip knife; suture material is visible at 6 o’clock position
The appearance of the foregut anatomy after esophagectomy can be surprisingly similar to the normal situation, with the exception of the esophagogastric anastomosis in the chest. The entire stomach remains, although narrower, and there is an antrum and pylorus. Pyloroplasty is usually done after an esophagectomy because of concern for pylorospasm related to the obvious need for a truncal vagotomy in the process of resection. Some surgeons will choose to do Botox injection into the pylorus rather than pyloroplasty; current practices in the field are in evolution [8].