Addition of pyloroplasty may improve glycemic control and refractory early satiety in gastroparesis at rates similar to gastric neurostimulation alone: a retrospective analysis

Abstract

Objectives

Gastroparesis that is refractory to standard dietary and medical management may benefit from surgical treatment with gastric electrical neurostimulation, which has shown promise in reducing symptoms of the disease. Pyloroplasty may serve an adjunctive role to a gastric stimulator, but the precise benefit remains unclear. The present study compares reported rates of symptom improvement following gastric neurostimulator implantation with and without pyloroplasty.

Materials and methods

A single center retrospective analysis of consecutive patients who received operative management for symptom refractory gastroparesis from 1 January 2020 to 31 December 2021 was performed. Subjects were assigned to cohorts based on treatment with gastric electrical stimulation alone (GES-only) or combined with pyloroplasty (GES + PP). A survey-based assessment was administered post-operatively that evaluated cardinal symptoms of gastroparesis (nausea, vomiting, early satiety) before and after treatment.

Results

In total, 42 patients (15 GES-only, 27 GES + PP) were included in the study. Both groups reported a high degree of improvement in global symptom control following surgery (93% vs 81%) with no differences between treatment cohorts (p = 0.09). Early satiety demonstrated better improvement in patients who received gastric stimulation alone (p = 0.012). Subgroup analysis of diabetic gastroparesis patients showed a 2.2% decrease in hemoglobin A1c levels in the GES + PP group (p-0.034).

Conclusions

Symptom reduction in refractory gastroparesis appears to improve after placement of a gastric neurostimulator with or without the addition of a pyloroplasty procedure.

Keywords:

• Gastroparesis
• implantable neurostimulator
• gastric emptying
• pyloroplasty

Introduction

Gastroparesis (GP) is a motility disorder of the stomach whose defining clinical presentation is delayed or absent emptying of the stomach into the duodenum [1]. The disorder often presents with accompanying nausea, vomiting, early satiety (ES), and anorexia [2]. The gold standard for managing GP includes dietary modifications and prokinetic agents [1,2]. Patients with refractory symptoms may benefit from operative treatment, but the standard-of-care for refractory GP remains controversial. Gastric electrical stimulation (GES) device implantation and surgical management with pyloroplasty (PP) have been shown to improve symptoms [3–6]. There is also a notable trend touting PP combined with GES as best practice [7–9]. Petrov et al. advocate for a stepwise approach to treatment, while Zoll and colleagues specifically identified the need for additional research into patients’ post-treatment symptoms between interventions [10].

This study aimed to further bridge the literature gap on treatment effectiveness for refractory GP. We hypothesized that patients who receive combined treatment with GES and PP will report a greater reduction in symptoms than if undergoing either modality in isolation. Specifically, this study compared the reported rates of symptoms among two treatment groups: those who underwent GES implantation and those who received a combination of GES and PP.

Materials and methods

Following approval by the Kirk Kerkorian School of Medicine Institutional Review Board (IRB 1579811-2), a retrospective review of a single institution’s electronic medical registry was performed. Included were patients greater than 18 years of age who underwent treatment of medically refractory GP with placement of GES with or without PP between 1 January 2020 and 31 December 2021 at University Medical Center of Southern Nevada. PP was performed via Heineke-Mikulicz stricturoplasty technique and not by other methods (transpyloric balloon dilatation, gastric peroral endoscopic myotomy). Diagnosis of GP was established via two repeat 4-hour radioactive scintigraphy tests with a radiotracer admixed in a solid meal. Exclusion criteria included patients who were pregnant, had an active bacterial or viral infection, or were diagnosed with a malignancy. Patients who declined study participation or were unable to provide informed consent for the study were also excluded. A final total of 41 patients were included for analysis and assigned to cohorts based on the surgical management used: GES-only or GES + PP.

A survey-based questionnaire was administered to all subjects, which queried patients on: diagnosis of GP (age at diagnosis, onset of GP-related symptoms); postoperative outcomes [5-point modified Likert scale assessing specific and overall symptoms, alterations to diabetes regimen including insulin requirements and serum hemoglobin A1c (HbA1c) levels]. Likert items were coded as: 1 = Significantly worse, 2 = Somewhat worse, 3 = Unchanged, 4 = Somewhat improved, 5 = Significantly improved. The survey design was selected given the disruptive effect of the COVID-19 pandemic on clinical scheduling and in-person evaluation.

The primary outcome was clinical improvement between treatment cohorts, defined in this study as an overall reduction in symptoms between the patient’s initial baseline survey evaluation and following surgical management. Secondary outcomes included pre- and postoperative changes in specific symptoms (nausea, vomiting, ES). Subgroup analysis of the diabetic GP cohort was performed to assess for changes in serum HbA1c levels, frequency of diabetic ketoacidosis (DKA) episodes, and changes in baseline insulin requirements.

Statistical analysis was performed using IBM SPSS Version 25 (Chicago, IL). The Student’s t-test was used for normally distributed independent samples. Categorical variables were evaluated with the Chi square test. Non-parametric ordinal data was measured using a Mann-Whitney U Test. P values <.05 were considered statistically significant.

Results

A total of 42 patients (18 with idiopathic GP, 24 with diabetic GP) met inclusion criteria and responded to the survey (Table 1). Most respondents were female (85%) with a mean age of 49 years. All study participants reported refractory GP to standard treatments including dietary modifications, prokinetic agents, and symptom-focused adjuncts. 15 underwent GES placement only while 27 had a combination therapy of GES and PP. Of the latter, 20 received simultaneous GES placement and PP while the remaining underwent GES placement first, were found to be inadequately treated, and underwent subsequent PP. All patients presented with nausea and vomiting as the primary GP symptom. Follow-up period was variable: the patient followed the longest underwent their index procedure in 2004, a total maximum follow-up time of 17 years.

Table 1. Characteristics of the study population (N = 42).

			Surgical management
Characteristics		Overall sample, N (%)	GES-only, N = 15	GES + PP, N = 27	P-value
Gastroparesis etiology	Idiopathic	18 (42.9)	6 (14.3)	12 (28.6)	0.780
	Diabetic	24 (57.1)	9 (21.4)	15 (35.7)
Gender	Male	7 (16.7)	4 (9.5)	3 (7.1)	0.195
	Female	35 (83.3)	11 (26.2)	24 (57.1)
Race/Ethnicity	Caucasian	35 (88.1)	13 (31.0)	22 (52.4)	0.666
	Non-Caucasian	7 (14.3)	2 (4.8)	5 (11.9)
Body mass index, kg/m^2, mean (range)		29.9 (17.9 to 55.8)	33.5 (17.9 to 55.8)	27.7 (20.2 to 41.5)	*0.037
Age at surgical evaluation, years, mean (range)		49 (21 to 72)	49 (23 to 72)	48 (21 to 69)	0.842
Age at diagnosis of gastroparesis, years, mean (range)		41 (18 to 68)	42 (21 to 57)	39 (18 to 68)	0.535
Diabetes mellitus subtype	Type 1	11 (26.8)	3 (7.3)	8 (19.5)	0.341
	Type 2	13 (31.7)	6 (14.6)	7 (17.1)

Abbreviations: GES, gastric-electrical stimulation; PP, pyloroplasty/pyloromyotomy.

* Statistically significant.

Reported outcomes are summarized in Table 2. Study participants in both cohorts had better overall symptom control at three months post-op, with 93% of and 81% of GES-only and GES + PP reporting at least some degree of improvement, respectively. There was no difference in reported symptomatic relief among treatment groups post-procedurally (median 5 vs 4, p = 0.09). Individual symptom subgroups (nausea, vomiting, ES) showed improvement following both treatment modalities. ES showed more improvement in patients who underwent the GES procedure only (median 5 vs 4, p = 0.012).

Table 2. Reported changes in symptoms and rates of hospitalization following surgical treatment for refractory gastroparesis.

		Patient response, N
Category	Treatment group	Significantly improved	Somewhat improved	No change	Somewhat worse	Significantly worse	P Value
Overall symptoms	GES-only	12	2	1	0	0	0.091
	GES + PP	13	9	5	0	0
Nausea	GES-only	10	4	1	0	0	0.080
	GES + PP	10	11	5	0	1
Vomiting	GES-only	11	3	1	0	0	0.107
	GES + PP	12	10	3	1	1
Early satiety	GES-only	9	5	1	0	0	*0.012
	GES + PP	7	8	10	2	0

Abbreviations: GES, gastric electrical stimulation; PP, pyloroplasty/pyloromyotomy.

* Statistically significant.

Among subjects with type 1 DM, insulin requirements were mostly unchanged while the reported frequency in DKA decreased in 40% of all treated patients (Table 3). Serum HbA1c levels dropped 2.2% in patients who underwent combination GES and PP procedures (p = .034).

Table 3. Outcomes of patients with diabetic gastroparesis (N = 24).

Category		Frequency	P-value
Insulin requirements	Increased	2
	Decreased	2
	Unchanged	6
DKA episode frequency	Increased	0
	Decreased	4
	Unchanged	6
HbA1c, Mean ± SD	GES-only, before surgery	13.3 ± 1.1	0.603
	GES-only, after surgery	10.7 ± 6.9
	GES + PP, before surgery	10.1 ± 2.7	*0.034
	GES + PP, after surgery	7.9 ± 1.3

Abbreviations: DKA, diabetic ketoacidosis; GES, gastric electrical stimulation; HbA1c, hemoglobin A1c; PP, pyloroplasty/pyloromyotomy.

* Statistically significant.

Discussion

GP is a chronic, progressive disease of the stomach with a symptom profile largely derived from delayed gastric emptying [1,11]. Dietary modifications in conjunction with pharmacologic agents serve as first-line treatment in symptomatic GP [3]. Surgical options are utilized in cases where symptoms are refractory [1,6]. GES and PP have been utilized to treat GP both as separate interventions as well as adjunctively. In GES, a gastric neurostimulator is placed, which is postulated to facilitate improved motility via high-frequency, low energy stimulation [5,6]. Pyloroplasty, which can be accomplished by either laparoscopy or endoscopy, assists with gastric emptying through pyloric widening [5,12]. The mechanisms of GES and PP are theoretically complementary and underlie this study’s hypothesis that combination therapy (i.e. GES + PP) would lead to greater improvement in global symptoms compared to GES alone.

In the present study, symptom reduction was reported in a substantial proportion of patients who received either GES or a combination of GES and PP. When comparing GES treatment with and without PP, there was no difference in overall symptoms. The GES-only group was associated with better control of ES compared to combination treatment. Given that all patients underwent at least one invasive procedure, some degree of reported improvement in each cohort was expected due to placebo effects but not measurable in our survey-based study. In patients with insulin-dependent DM, episodic frequency of DKA decreased in about half of patients while those who were treated with combination GES and PP had reduced HbA1c levels.

GES is thought to stimulate motility to potentially improve gastric accommodation, but the precise mechanism remains elusive [1,5,6]. Following the procedure, patients have been shown to have a significant reduction in overall cardinal GP symptoms [8,13–16]. Nausea and vomiting particularly appear to have the most robust response in observational cohort studies [8,13], with some patients being able to discontinue prokinetic medications entirely [17]. Indeed, most participants in the present study endorsed some degree of improvement in all symptom categories, including nausea and vomiting, following gastric electrode placement.

The addition of a pyloroplasty procedure has been posited as an adjunct to GES [18]. In theory, PP’s mechanism is complimentary: surgical pyloric widening reduces the trans-pyloric pressure gradient to facilitate better emptying and possibly alters the vomiting reflex by attenuating pyloric contractions [5,12]. One retrospective study found that gastric emptying improved in concomitant GES and PP beyond what GES accomplishes alone [9]. When evaluating both global and specific symptoms (e.g. nausea, vomiting), combination therapy does shows durable improvement over time but does not appear to confer a significant advantage over GES-only [4,9,18]. Our study’s findings further support the lack of difference between GES and GES + PP, suggesting that GES is effective despite adjunctive measures.

ES symptoms in GP may manifest from impaired gastric accommodation, which via the pacemaking interstitial cells-of-Cajal or smooth muscle-mediated peristalsis can affect the stomach’s ability to properly empty [12]. While most of our patients in both the GES-only and GES + PP cohorts reported an improvement in ES, those who solely underwent GES placement reported a greater effect compared to combination therapy. The temporal separation between accommodation and gastric emptying may in part explain why ES was not as affected in combination treatment, with GES and PP addressing the concurrent processes via different mechanisms [1,18,19]. Why this finding was limited to ES (versus nausea and vomiting) is not entirely clear as the true effect of GES on gastric emptying time is controversial [8].

Diabetic GP appears to confer an improved clinical response to GES compared to those of idiopathic etiology [13,18]. In our patients with insulin dependence, changes in baseline glycemic requirements were mixed: most (60%) had unchanged insulin regimens after surgery. A decrease in serum HbA1c levels was identified in the GES + PP group. This supports previous findings that combination surgical treatment of GP may also have a therapeutic benefit to underlying diabetes, although the precise mechanism is unclear [20–22].

There are several limitations to this study. Its retrospective approach to data abstraction may be prone to selection bias. The descriptive nature and small sample size from a single institution restricts generalizability. The small sample size limits the study’s power, although given the low prevalence of GP the findings do hold merit and support future larger investigations. The study was primarily undertaken during the COVID-19 pandemic, which impacted our decision to formulate a survey-based questionnaire that simplified study participation, but this was not immune to recall bias. Use of a modified Likert scale to assess symptoms was intended to obtain patient-focused changes, but these findings are acknowledged to be subjective in nature. Post-treatment follow-up was also non-standardized. Future studies would benefit from incorporating objective post-procedural measurements of gastric emptying such as follow-up gastric emptying scintigraphy.

Acknowledgements

The authors would like to acknowledge Alex Manning and Christian Borashan for their assistance with data collection.

Disclosure statement

The authors report there are no competing interests to declare.

Additional information

Funding

This research did not receive any specific grant from funding agencies in public, commercial, or not-for-profit sectors.