Two-channel gastric pacing prototype device pulse generator (black box) is attached to external wires in a patient with refractory GP. These external wires stimulate electrodes that were surgically attached to the serosa of the stomach when the device is activated after meals
High-Frequency/Low-Energy GES (Neurostimulation)
This approach is also referred to as “short-pulse stimulation” and applies pulses with duration of 300 ms, at a frequency of approx. 12 cpm which is four times the physiologic rate of stomach . This higher frequency does not entrain the slow wave of stomach and has no effect on gastric dysrhythmias. The main purpose of neurostimulation is to treat intractable nausea and vomiting . Gastroparetic patients who are the best candidates for gastric neurostimulation have daily nausea and vomiting refractory to aggressive antiemetic and prokinetic drug therapy for at least 1 year in duration.
Different programming parameters were studied in humans and animal models to evaluate response of symptoms and gastric emptying. Studies done by Familoni et al. with low-energy/high-frequency device initially reported increased gastric contractility in canines and improvement in symptoms and liquid gastric in a diabetic patient with severe GP [13, 14]. The first implantable device named Enterra Therapy System (Medtronic, Inc. Minneapolis, MN) was developed to incorporate those high-frequency low-energy parameters and was approved by the Food and Drug Administration (FDA) under a Humanitarian Use Device status based on the result of a multicenter double-blinded crossover study called WAVESS (World Anti-Vomiting Electrical Stimulation Study ). Thirty-three patients with chronic gastroparesis (17 diabetic and 16 idiopathic) underwent implantation of this GES device which provided continuous high-frequency/low-energy gastric electrical stimulation parameters after surgery. They were randomized in a double-blind crossover design to stimulation ON or OFF for 1-month periods. The second phase was unblinded where all patients were programmed to stimulation ON and followed a further 6 and 12 months. The weekly vomiting frequency (WVF ) was a primary objective. In the double-blinded phase , there was a significant reduction in self-reported vomiting frequency in the ON vs. OFF period (P < 0.05). In the unblinded portion of the study, vomiting frequency decreased significantly (P < 0.05) at 6 and 12 months. Scores for symptom severity and quality of life significantly improved (P < 0.05) at 6 and 12 months; however, gastric emptying was only modestly accelerated . Overall 80% of the patients reported more than 50% improvement in symptoms after the total of 12 months of follow-up.
Subsequently, Enterra therapy has been studied in multiple open labeled clinical trials which have shown sustained and significant improvement in symptoms such as nausea and vomiting in patients who have failed aggressive medical therapies. When a controlled study was performed again in 2010, 55 patients with refectory diabetic GP patients were implanted with the Enterra system. All patients had the system turned ON for 6 weeks after surgery. Patients were then randomly assigned to groups that had consecutive 3-month crossover periods with the device either ON or OFF. The devices then were turned ON in all patients and they were followed for up to 12 months. There was a significant reduction in nausea and vomiting in the initial 6 weeks but there was no significant differences shown between the ON and OFF treatments during the 3 months crossover period. However, there was a significant decline in WVF from baseline values (median reduction, 68%; P < 0.001) by 12 months when all patients had devices tuned ON. The study participants also had improvements in total symptom score, gastric emptying, quality of life, and median days in the hospital .
Previous studies have shown higher treatment failure of GES in patient with idiopathic compared to diabetic GP. In multicenter randomized crossover study evaluated the efficacy of GES in 33 idiopathic GP patients, the stimulator was turned ON for 6 weeks after the surgery followed by double-blind randomization to consecutive 3 month crossover periods with the device either ON or OFF followed by an unblinded treatment period of 4.5 months. A total of 25 patients completed the crossover period and 21 patients continued 12 months follow-up with device ON . During the unblinded first 6 weeks ON period, there was a significant reduction in WVF from baseline (61.2%, P < 0.001). During the subsequent crossover period the median reduction of WVF approx. 17% (P > 0.10) between ON and OFF phase of the study was not significantly different. At 1 year, the mean WVF remained decreased by 87% (P < 0.001), and it was accompanied by improvements in GP symptoms, gastric emptying, and days of hospitalization (P < 0.05) .
Both studies [15, 16] did not show significant improvement in WVF during the cross over periods. This observation may be related to the presence of other variables that contributed to the symptom reporting. During the first 6 weeks following implantation, all patients had the device activated. One plausible theory is that initial activation of the system may lead to a “memory” or “imprinting” effect on the CNS pathway which was activated and which may have led to a sustained response during the crossover period even when the device was turned off. Differences in symptoms may potentially be confounded by effects of the initial surgery related to the use of pain medications, alterations in glucose control, and the placebo effect of the surgery itself . However, placebo effects tend to last for not more than few weeks, hence the sustained response observed for more than 1 year would seem to be related to the effects of the GES therapy.
Long-term observations have also been published. Brody and colleagues reported sustained symptoms response and pain reduction during their follow-up period of 8 years . In a prospective non-randomized study 79 patients had GES implanted for refractory GP between 2003 and 2013 and were analyzed for pre- and postoperative pain and function scores over time at a single institution. Symptom scores were available for 60 participants at baseline, 52 participants at 1 year, 14 participants during years 2–3, and 18 participants for 4–8 years. Overall, symptom reductions were maintained for 8 years for both functional and pain symptoms. At 1 year follow-up, 44 and 31% of the participants experienced at least a 25% reduction in symptom distress for functional and pain symptoms, respectively. At 4–8 year follow-up, 67 and 33% of the participants experienced at least a 25% reduction in symptom distress for functional and pain symptoms, respectively . This study has limitations as it was a non-randomized and also data was available for less than 40% of the patients beyond 1 year follow-up. However, the finding of patients reporting substantial improvement beyond 12 months would be further evidence against a “placebo” effect . These patients had hospitalizations in the time preceding the GES surgery and had been clinically unstable. These aspects were now changed.
The largest study on long-term safety and efficacy of GES therapy was reported in a case series study where 221 patients with refractory GP (n = 142 diabetic, n = 48 idiopathic, and n = 31 postsurgical) who were treated with the Enterra device were followed for up to 10 years. At 1 year, 188 (85%) of the initial 221 subjects enrolled were available for follow-up. Total symptom score was reported to have decreased by 53% ± 32% (P < 0.001). Participants with diabetic GP had greater symptom reduction than those with postsurgical and idiopathic GP (55% vs. 48% vs. 47%, respectively). Of 119 subjects with gastric emptying data, 26% normalized their results after GES therapy (P < 0.05). In addition, 89% of patients were able to stop jejunostomy tube feeding within 12 months and had a significant improvement in their weight. There was a reduction of hospitalization days by 87% (P < 0.001) in the last year of follow-up for all patients with diabetic GP. Overall, the use of GP medications in all subject groups was reduced after 1 year of GES (74% at baseline vs. 56% for prokinetics, P = 0.05; and 65% at baseline vs. 58% for antiemetics, P = 0.025) . Patients with idiopathic GP had less response compared to diabetic GP subjects. Idiopathic GP patients represent a heterogeneous mixture of patients etiologically compared to other groups and they report more abdominal pain which is the symptom least likely to respond to neurostimulation .
Meta-analysis of 10 studies (n = 601) by Chu et al. showed significant improvement of total symptom severity score (P < 0.00001) and gastric retention at 2 h (P = 0.003) and 4 h (P < 0.0001) in patients with diabetic GP with GES. However, gastric retention at 2 h (P = 0.18) in idiopathic GP patients and gastric retention at 4 h (P = 0.23) in postsurgical GP patients receiving GES were not significant .
Predictors of Response
There are some factors that have been identified which could predict a suboptimal outcome to GES . Patients with concomitant chronic opioid use will do worse due to inhibition of gastric emptying by opioid use as well as induction of nausea and vomiting by central mechanisms. GES improves nausea and vomiting which may lead to less abdominal pain; however, when chronic abdominal pain is the predominant preoperative symptom, use of GES should be carefully considered as pain control is not a primary goal of the therapy. Idiopathic GP patient tend to have higher abdominal pain levels than diabetic GP which could also explain the higher treatment failure of GES in idiopathic GP patients.
Patients with different disease processes such as rumination syndrome, dumping syndrome, cyclic vomiting syndrome, and eating disorders can have nausea and vomiting, which will not improve with GES. Hence, making a right diagnosis is crucial. 50% of patients with GP are lacking the normal ICC populations in their antral smooth muscle based on full-thickness biopsies done during surgery, and overall this ICC deficiency has been found to have higher association with a suboptimal treatment outcome with GES [20, 21].
How Is Surgery Accomplished of the Enterra System?
The Enterra gastric stimulation system consists of three main elements: a pair of leads, a pulse generator, and a programming system (Fig. 25.2). Two electrodes are implanted surgically, by either laparotomy or laparoscopy, depending on the expertise and training of the surgeon, in the muscular layer of the body of the stomach, along the greater curvature, approx. 1 cm apart 9 and 10 cm from the pylorus generally on the greater curvature. Leads from the electrodes are connected to a pulse generator which is placed in a subcutaneous pocket in the abdominal wall (Fig. 25.3) in the left or right upper quadrant. The pulse generator was adapted from existing devices in clinical use, which could sustain long-term requirements of a low energy type of stimulation. It is programmed by an external interrogator which both monitors and determines parameters, e.g., 5 mA as a standard current, 14 Hz micro-second, cycle on and cycle off 0.1 and 5 s, respectively. There are no controlled trials regarding the best programming parameters . The usual initial setting is the “default” setting where current and voltage are based on the resistance in ohms between the two gastric electrodes. Further adjustments may or may not be warranted. Clinicians can increase current and voltage in increments of 20–30% during follow-up if a patient reports poorly controlled symptoms in the hope that energy may be helpful in reducing symptoms [18, 22]. However this is very subjective. Battery life of the pulse generator is estimated to be at least 5–10 years, depending on the pulse parameters used . When the battery is depleted, the pulse generator is replaced by local intervention. Hospital stay is short, approximately 2–3 days, following laparoscopic insertion, and is shorter when compared to placement via laparotomy (6.4 days) .
Enterra system components (From Reddymasu SC: Severe Gastroparesis: Medical Therapy or Gastric Electrical Stimulation, Clin Gastroenterol Hepatol. 2010; 8(2):117–24) (Reprint permission obtained from Elsevier publication)