The most common scenario that the esophageal surgeon faces with respect to gastroparesis is the patient with severe, refractory GERD who is referred for antireflux surgery but who has unrecognized gastroparesis. Alternatively, patients may be referred for surgical treatment of gastroparesis but have significant GERD, possible due to years of gastric distension and vomiting. Either way, a formal gastric emptying scintigraphy (GES) is mandatory for patients suspected to have gastroparesis as part of their standard comprehensive GERD evaluation, including upper endoscopy to rule out mechanical obstruction of the stomach. Adherence to published recommendations for standardized meals and a 4-h emptying measurement is important to maximize the accuracy of the test. The quality of radionucleide GES is known to differ amongst radiology centers due to a wide variability in testing protocols and normal values. Tougas et al. [3] and Ziessman et al. [23] have refined the test to create standardization of the test meal and imaging protocols at 0, 1, 2, and 4 h. Using this protocol, diagnosis of GP is based on retention of the tracer at each interval. Specifically, 90 % of persistent tracer remaining at one hour, 60 % at 2 h, or 10 % at 4 h are diagnostic of gastroparesis. This has shown to be more sensitive to detecting gastroparesis than the previously thought for only 2 h, and other interval protocols.

Laparoscopic Heineke–Mikulicz pyloroplasty can be performed using the same trocar configuration as for the concurrent antireflux surgery (Fig. 21.3). The liver retractor is inserted through the far right trocar and used to expose the pylorus. With the surgeon standing at the patient’s left side, a gentle Kocher maneuver is performed to mobilize the pylorus to decrease tension on the eventual suture line. The attachments on the cephalad and caudad sides of the pylorus are divided taking care to avoid the underlying portal structures and maintain adequate blood supply to the pylorus. Precise identification of the pylorus, usually aided by the location of the vein of Mayo, is mandatory and can be facilitated by intra-operative upper endoscopy. After mobilization, the pylorus is grasped by the surgeon’s left hand and the assistant using the subxiphoid trocar. Our preference is to use the ultrasonic shears in the surgeon’s right hand, which enables tissue division, sealing, and dissection. Using the ultrasonic shears, a gastrotomy is created about 2–3 cm proximal to the pylorus. The gastrotomy is then extended across the pylorus for a 5 cm full thickness pyloromyotomy (Fig. 21.4). The assistant grasper is then repositioned with one jaw inside the lumen at the level of the pylorus on the cephalad edge to keep the suture line elevated and maximize the sewing angle for the surgeon. The complete myotomy is confirmed by viewing the muscular ridge in the center of the enterotomy. The defect is then closed transversely using a running 2-0 absorbable monofilament suture. This suture choice avoids chronic inflammation of the repair which can be seen with permanent alternatives. The first suture is placed to reapproximate the linear (proximal and distal) ends of the myotomy into the center of the future suture line and run cephalad, away from the surgeon, toward the assistant grasper (Fig. 21.5). Care should be taken to avoid “back-walling” the closure in the center but deeper bites can be taken as the end of the suture line is approached to decrease the dog-ear. When the end is reached, the assistant removes the grasper and the suture is tied to itself just past the end of the myotomy. The second suture is then placed, starting at the caudad end of the myotomy and running toward the start of the previous suture in the center of the suture line. The assistant retracts the original suture in the center using a needle driver. The suture is then secured to the tail of the first suture to complete the closure (Fig. 21.6). Intra-operative submersion air-leak and/or methylene blue tests are performed using the upper endoscope. No omental patch is placed. A 15f closed suction drain is then placed in the gastrohepatic ligament along the suture line. Contrast study is performed POD #1 to ensure patent pyloroplasty and confirm the absence of a suture line leak. Nausea is treated aggressively with IV antiemetics and the patients are kept on a PPI for 6 weeks. Routine 4 h GES is performed at 3 months and the patient’s symptoms are reassessed.

Feeding jejunostomy tubes and decompressive gastrostomy tubes can be helpful in the care of the gastroparetic patient. The paretic stomach is particularly sensitive to post-fundoplication gas-bloat. Decompressive gastrostomy tubes have been advocated for gastroparetics undergoing fundoplication [17, 30]. We also advocate this approach but reserve it for patients undergoing fundoplication and pyloroplasty who also report severe bloating and/or vomiting to assist with these symptoms during recovery. A majority of the tubes are successfully removed by 3 months when the swelling from the pylorus has resolved and the ability to belch, or naturally vent, recovers. Feeding jejunostomy tubes are also helpful in severe gastroparetics. In a selected group of gastroparetic patients presenting frequent admissions for dehydration and severe malnourishment, a laparoscopic jejunostomy tube can serve as an effective “life-line” [1, 31]. Usually, the fundoplication-pyloroplasty can be performed at the time of the feeding jejunostomy but occasionally, the feeding tube is placed alone if the patient is simply too debilitated to proceed with more surgery up front.

Special mention should be made regarding the obese patient with gastroparesis and GERD. The same considerations exist regarding recommendations for RY gastric bypass and are covered elsewhere in this book. The only difference is that we recommend removing the remnant stomach as it is diseased and if left behind it risks continuing to cause symptoms. This does increase the morbidity of the operation overall and should be discussed with the patient ahead of time.

Iatrogenic gastroparesis is most often secondary to surgical procedures susceptible of causing a vagal nerve injury such as fundoplication, bariatric surgery, or procedures in which vagotomy is needed such as anti-ulcer operations [30]. As expected, the incidence of iatrogenic post-operative gastroparesis seems to increase with the number of re-operations in a given patient. It is important to be aware of gastroparetic symptoms in a post-fundoplication patient as the gastric dysfunction may eventually lead to a hiatal hernia recurrence or failed fundoplication.

Perhaps the most challenging group of patients are those who are refractory to dietary modification, medical therapy, and pyloroplasty. Unfortunately, debilitating symptoms, particularly pain and nausea, may be persistent regardless of the impact of pyloroplasty on objective gastric emptying [32–34]. It is appropriate to consider gastrectomy for this group of patients [35–37]. Our group has recently presented our experience with 35 patients who underwent a laparoscopic subtotal or total gastrectomy for refractory gastroparesis. Of these, 46 % had previous pyloroplasties, 54 % had previous fundoplications, and 11 % had previous feeding or venting tubes that failed to result in clinical improvement [38]. At a median of 6 months follow-up, there was complete resolution in the pre-operative symptoms in 72 % of patients with nausea, 79 % of those with belching, 89 % of those with bloating, and in 50 % of those with chronic abdominal pain. While the results are promising, the operative morbidity of gastrectomy in gastroparesis is not trivial ranging from 20 to 40 % [35, 37, 38]. These patients often have severe gastritis, retained food in the stomach and in general are not healthy. Malnutrition or poor nutrition is not uncommon among patients with gastroparesis, making for tenuous anastomotic healing [39]. Because of these factors, the procedure should be reserved for refractory patients.

Another option for refractory gastroparesis is gastric nerve stimulation as the implant has been shown to reduce nausea and vomiting, and ameliorate chronic pain [40]. This is a particularly appealing option for high-risk patients with primary nausea. However, a recent study has demonstrated that gastrectomy may be superior to gastric stimulation in the primary treatment of gastroparesis [41]. The decision to proceed with gastrectomy or gastric stimulation after failed pyloroplasty is a challenge and one that needs to be individualized to the unique needs of the patient. There is little if any published literature to help guide the clinician. We have found that patients who suffer from predominantly bloating and retained food do well with a gastrectomy while primary nausea often responds to gastric stimulation. While more work needs to be done in this field, our current clinical algorithm for refractory gastroparesis is shown in Fig. 21.7.