Optimal Parameters for Gastric Electrical Stimulation Therapy for Long-Term Symptom Control in Patients with Gastroparesis

: Background : Gastric electrical stimulation (GES) therapy is indicated for the treatment of drug-refractory gastroparesis (GP). However, the long-term effects of GES therapy on GP symptom control and identification of the optimal parameters to activate this long-term efficacy have not been investigated. Methods : We conducted a retrospective cohort analysis of 57 GP patients who received GES and pyloroplasty (PP). The interrogation of the GES system and assessment of GP symptoms were conducted at the initiation of GES therapy and during follow-up visits. We determined the changes in GES parameters including voltage (V), impedance (I), and current (C). The outcome was total symptom score (TSS), which was measured by self-reported GP symptoms. Results : The mean age of patients was 44 ( ± 14) years, and 72% were females. The etiology for GP was diabetes mellitus in 72% and idiopathic in 28%. The median duration of GES follow-up was 47 months (range 5–73) A significant decrease was found in individual symptom scores and the TSS ( − 10.8; 95%CI: − 12.6, − 9.08) compared to baseline scores ( p < 0.0001). During follow-up, readings for I (515 vs. 598 Ω ), V (3.3 vs. 4.8 V), and C (6.5 vs. 8.4 mA) significantly increased ( p ≤ 0.0001 for all parameters). Higher GES settings were associated with lower TSS in the adjusted analysis (RC, − 1.97; 95%CI: − 3.81, − 0.12, p = 0.037). Conclusions: these findings suggest that adjusting GES parameters over time based on optimizing symptom improvement should be incorporated into the long-term care of patients receiving gastric neurostimulation therapy.

The management of GP consists of the improvement in gastric emptying when possible, the reduction in symptoms, and dietary modifications [1,2].Pharmacological therapy is aimed at symptomatic control through the use of prokinetic and antiemetic agents [2].Depending on the etiology, measures such as optimizing glycemic control in diabetics and the discontinuation of gastroparesis-inducing medications are also important [1].Dietary modifications to preserve oral nutrition and correct fluid, electrolyte, and nutritional deficiencies are recommended as part of the initial management.However, enteral or parenteral nutrition may be needed for severe GP patients who are losing weight regardless of concurrent dietary modifications and pharmacotherapy [1].In cases of GP, where patients have intractable nausea and vomiting despite medical management and/or severe medication-induced adverse effects limiting pharmacology options, gastric electrical stimulation (GES) therapy is a valuable option.
The use of GES is indicated for the treatment of refractory gastroparesis (GP) symptoms and has been available for the last two decades.The Enterra Therapy System was approved by the U.S. Food and Drug Administration as a humanitarian device exemption for patients with refractory diabetic or idiopathic GP in March 2000 [2].This surgically implanted device delivers high-frequency, low-energy stimulation to the stomach.Since its introduction, the beneficial outcomes from GES therapy have been documented in the literature, specifically significant improvements in total symptom score [5][6][7][8][9][10], quality of life [9,11], and a reduced need for enteral or parenteral nutrition [5,12].Additionally, the use of GES therapy in refractory GP patients has been associated with a reduction in the utilization of health care services and associated costs [6,12].The efficacy of GES therapy relies on the effects generated by the GES parameters.It is still unclear which of the GES parameters is more critical for achieving sustainable GP symptom control.
Despite the established efficacy and benefits of GES therapy, there is limited literature on the recommended GES parameters for the device during long-term treatment.Therefore, we designed this study with a null hypothesis to show no changes in the total symptom scores after administration of the GES therapy, and the pre-defined hypothesis was to establish changes in primary and secondary outcomes before and after GES intervention.Our primary objective was to evaluate the long-term changes in GP total symptom score (TSS) in two etiological groups of GP patients.The secondary objective/aim was to evaluate the changes in the GES parameters associated with changes in primary and secondary clinical outcomes.Primary endpoints (outcome measures) were the improvement in the total and individual GP symptoms scores (vomiting, nausea, early satiety, bloating, postprandial fullness, and pain) and their association with the stimulation parameters delivered by the GES system.

Data Collection
Our original research project, titled The Long-Term Subjective and Objective Outcomes of Combining Pyloroplasty with Gastric Electrical Stimulation in Drug-Refractory Gastroparetic Patients, was designed to investigate clinical efficacy, safety, and unbiased outcomes of concurrent GES and pyloroplasty (PP) interventions.This submitted manuscript is a sub-study predicated on the question of the relationship between changes in the GES parameters and their association with GP symptom improvement.We did not generate any sub-study-specific objectives at the beginning of this project, but rather we decided to learn about the GES parameters from our data obtained during this observational study.
Therefore, we conducted a retrospective cohort study to evaluate how changes in GES parameters correlate with the efficacy of the stimulation in reducing GP symptoms, after receiving IRB (Institutional Review Board) approval.The IRB, which is our University's equivalent of an Ethics Committee, is dedicated to the protection of human rights and well-being and guarantees the safety of human subjects involved in research.
We retrospectively reviewed the medical records of 57 patients with refractory GP.All patients were implanted with the Enterra device for GES therapy and simultaneously underwent a Heineke-Mikulicz pyloroplasty (PP) procedure by laparotomy or robotassisted approaches.

1.
Our primary endpoint of this project was the total change in symptom scores measured using a validated instrument TSS, while secondary endpoints were alterations in individual symptom scores including vomiting, nausea, early satiety, bloating, postprandial fullness, and abdominal pain.2.
Our null hypothesis was no changes in the total symptom scores after administration of the GES device.Our secondary hypotheses were to establish if specific GES parameters are associated with the total symptom scores and/or their individual components regardless of underlying gastroparesis etiologies.Hence, the secondary objective was to assess the impact of gastric stimulation parameters on clinical outcomes (GP symptoms), through a comparative analysis of the two groups.This research aimed to determine if any of the patient etiology populations derive the greatest benefit from the intervention.

Inclusion/Exclusion Criteria
There were many inclusionary and exclusionary criteria utilized to qualify GP patients for the surgical GES and PP therapies, under our master original protocol and clinical investigation.For example, patients had to have documented delayed gastric emptying, drug-refractory symptoms, frequent episodes of nausea and vomiting, being hospitalized due to the complication of their diabetic gastroparesis, losing weight, and possibly being dependent on the jejunostomy as a main source of nutritional support.This current article is a sub-study, which is part of the original project but asks a separate research question (GES parameters in our case), and it contributes to the parent protocol objectives with study participants.
All qualified GP patients have been under the care of the same team of clinicians since the initiation of the stimulation therapy.An interrogation of the GES system and assessment of GP symptoms were performed at the initiation of GES therapy and during follow-up visits.These patients returned to the clinic for 1-2 follow-up visits per year after receiving the GES system.

Introduction of the GES System
The Gastric Electrical Stimulation (GES) system, also called bioelectric Enterra Therapy, was originally developed by Medtronic and approved in March of 2000 by the FDA through the humanitarian device exemption process, as a Humanitarian Use Device (HUD).The Enterra ® Therapy System is a minimally invasive, customizable, reversible and implantable device that provides gastric stimulation to treat chronic, intractable, nausea and vomiting in drug-refractory patients, aged 18 to 70 years old, whose symptoms of gastroparesis are caused by diabetes mellitus (DM) or an unknown (idiopathic ID) origin.Therefore, by the FDA definition, the utilization of GES in clinical settings is limited to those 2 etiological groups of qualified patients, included in our research study.The batteries of the device could last up to 8-10 years, depending on the settings of the energy use.

The GES Implantation Procedure
The Enterra GES System consists of 3 major components: (1) a battery-powered implantable pulse-generator (Medtronic Enterra Therapy Model 3116 or Model 37800), (2) two 35-cm-long leads with 1 cm intramuscular electrodes on the ends (Model 4351; Medtronic), and (3) an external programmer for adjusting output parameters.
The surgical implantation of the pulse generator can be accomplished through an open laparotomy or robotic-assisted laparoscopic procedure.Parallel to one another, two permanent electrodes separated by 1 cm are inserted 9 and 10 cm from the pylorus along the greater curvature in the mascularis propria layer of the stomach.Intraoperative upper endoscopy is performed to verify that the electrodes do not penetrate into the lumen of the stomach.Subsequently, the pulse generator is positioned and secured with sutures in a subcutaneous pocket located above the abdominal wall fascia in the upper mid-quadrants of the abdomen, either on the right or left side, with the opposite ends of the leads connected permanently to the stimulator.Antibiotic-containing solution is used to thoroughly irrigate the pocket, and the patient is administered intravenous antibiotics two days before and following the operation.It is preferable to initiate GES therapy in the operating room by utilizing the "default" settings.GES can also be activated before the patient's discharge from the hospital when they are able to tolerate a soft food diet.

Programming of the GES System
The programming of the GES system was originally performed in the operating room.The N'Vision Clinician Programmer assisted in confirming that the optimal impedance (resistance between 2 electrodes) was in the normal range of <800 Ω.Initially, GES was programmed to standardized/default parameters: pulse width, (PW) 330 µs (microseconds); (current) (I) amplitude, 5 mA; rate, 14 Hz; cycle ON: 0.1 s; cycle OFF: 5.0 s.During follow-up visits, parameters such as (pulse width) PW and rate with ON and OFF cycles remained unchanged for all of our patients.
At each subsequent visit, the system was interrogated and parameters were recorded.Voltage depends on the measurements of the impedance (R; a typical range of 200-800 Ω), using a basic Ohm's Law equation, which states that voltage equals resistance multiplied by current (V = IR).We also assessed the overall status of the GP symptoms.If symptoms, particularly nausea and vomiting, were persistent and there seemed to be no other explanation for their presence, we generally decided to increase voltage by a small increment, such as 0.5 to 1 volt.
Similarly, we lowered the voltage in order to preserve battery life and prolong the time that Enterra's device delivers stimulation if N and V symptoms were improving.In a case when GES impedance was >800 Ω, abdominal X-ray and/or upper GI endoscopy were performed to investigate potential technical complications caused by electrode dislodgement or electrode penetration through the muscle into the lumen of the stomach.

Scheduling of the Follow-Up Visits
We have a predetermined "GES Calendar" for the follow-up appointments occurring 3 and 6 months post surgery, and subsequently at intervals of no less than 6 months.Based on our extensive experience of monitoring GES patients for a maximum of 7 years, including patients who travel varying distances to visit us, we strongly recommend following this "6-month blueprint" for modeling GP patients' clinical visits.

Statistical Analysis
Categorical data were summarized using frequency and percentages.However, quantitative data are summarized either with the mean and standard deviation (SD) or with median and interquartile range (IQR).The baseline characteristics of patients including the GES parameters are summarized for the total cohort and by the GP etiology.All the baseline characteristics, GES parameters, and TSS were compared between GP etiology using an unpaired t-test or a Fisher's exact test.The changes in the GES parameters between baseline and follow-up were compared using a two-sided paired t-test.The effect size is summarized with mean changes and a 95% confidence interval (CI).In addition, the relative percent change at follow-up in the GES parameters was also computed and is reported.The TSS and its individual components were compared between baseline and follow-up using either a Wilcoxon signed rank test or a paired t-test.The unadjusted association of each covariate including GES parameters and changes in GES parameters were associated with overall TSS at follow-up using linear regression analysis.We further evaluated the effect of changes in each GES parameter after adjusting for age, GI etiology, and corresponding baseline GES parameters.We further validated associations by additionally adjusting for the duration of follow-up and sex in the linear regression analyses.These analyses were also replicated in the short-term (<12 months) and long-term follow-up (≥12 months) of patients separately.In the regression analysis, we used z-standardized values of GES parameters so that the effect of GES parameters can be compared directly.The results of regression analyses are summarized with the regression coefficient (RC), its 95%CI, and the p-value.
The p-values were considered statistically significant at 0.05.All data analyses were conducted using Stata V17.

Baseline Clinical Characteristics of Gastroparesis Cohort
Overall, we had 57 patients diagnosed with gastroparesis.Among them, 79% were of white Hispanic ethnicity, with a female/male ratio of 41/16 (72/28%).The mean age of patients was 44 (±14) years, and 72% were female (Table 1).The etiology for GP was diabetes mellitus in 72% and idiopathic in 28%.The mean duration of GP before the introduction of GES was 6 (±4) years (range: 1-20 years).The median time to the most recent follow-up visit was 47 months (range: 5-73 months).The mean initial and follow-up TSS scores were 18.8 (±3.7) and 7.7 (±6.2) points, respectively.The mean baseline stimulation parameters were as follows: an impedance of 515.1 (±92.8)Ω, voltage of 3.3 (±1.1)V, and current of 6.5 (±2.1) mA.The mean follow-up stimulation parameters were an impedance of 598.4 (±116.8)Ω, voltage of 4.8 (±1.0)V, and current of 8.4 (±2.3) mA.These characteristics were not found to be different in diabetes or idiopathic patients except for gender differences (Table 1 and Supplemental Table S1).In terms of co-morbidities, a majority (61%) of our diabetic GP patients were diagnosed with peripheral neuropathy.Additionally, 47% of them suffered from hyperthyroidism, 30% experienced headaches/migraines, 14% had a history of some cardiovascular conditions, and 4 (7%) patients were on hemodialysis.Furthermore, 51% of these patients were diagnosed with anxiety and depression.By the GES protocol we are utilizing, we made sure our patients were not on any narcotics during the gastric emptying test.After the initiation of GES therapy, there were very few individuals who had to be on opioids for back pain utilizing a patch of lidocaine, Tramadol, or hydrocodone to control their discomfort.We did not request that patients discontinue any prokinetics or antiemetics when applying GES stimulation, so they were able to stop, taper off, or continue using metoclopramide, domperidone, Phenergan, ondansetron, or other medications they were taking before the surgery or as a backup when needed.Additionally, diabetic patients were receiving insulin, metformin, or using GLP-1 medications.Gabapentin for peripheral neuropathy was often prescribed for DM patients.The most useful drugs for depression and anxiety were TCAs with amitriptyline, mirtazapine, and benzodiazepines (Lorazepam).

Effect of GES Intervention on Primary and Secondary Outcomes
All parameters of the GES system such as impedance (p < 0.0001), voltage (p < 0.0001), and current (p = 0.0001) were found to be significantly higher at the most recent follow-up compared to the initiation of GES therapy (Table 2).The highest increase was noticed in voltage (42%), followed by current (26%) and impedance (15%).It was imperative to ascertain whether both GP etiological groups had a similar favorable response to gastric stimulation as a control for their symptoms and whether the GES parameters were adequate and equivalent in both the DM and ID groups.According to our data, we can infer that no significant variations were seen in the utilization of various GES metrics and patient TSS assessments.The information is presented in Table 3.The combination of the GES changes in stimulation parameters and the presence of pyloroplasty resulted in a significant improvement in TSS at follow-up compared to the score at the origination of GES therapy (7.7 ± 6.2 vs. 18.8 ± 3.7, p < 0.0001, Table 4) with a 60% decrease in mean TSS.Additionally, a significant decrease was found in each individual patient's symptom score at the most recent follow-up compared to the baseline (p < 0.0001 for all, Table 4).

Association of Baseline, Follow-Up, and Changes in the GES Parameters with Total Symptoms Score
In the unadjusted analysis, older age (RC= −0.12, p = 0.03), follow-up impedance (RC = −2.86,p = 0.001), and the change in impedance (RC = −2.38,p = 0.009) were associated with decreases in TSS.In contrast, a higher follow-up current value (RC = 2.78, p = 0.002) was associated with increases in TSS (Table 5).In the adjusted analysis, the change in impedance remained statistically significant (RC = −1.97;95%CI: −3.84, 0.03, p = 0.038) for decreases in TSS (Table S2).Although not significant, a change in the current parameter was positively associated with TSS at follow-up (RC = 1.66; 95%CI: −0.18, 3.51, p = 0.077).These associations remained unchanged even after additionally adjusting for covariates in the model (Supplementary Table S2).In stratified analysis by follow-up, we observed higher changes in impedance and voltage associated with lower TSSs at followup, while higher changes in current were associated with increases in TSS among patients with short-term follow-up (Supplementary Table S3).

Discussion
Although the Enterra device has been used for GES therapy for more than 20 years, there is limited research on the change in GES settings and how it affects clinical outcomes.A few previous studies that evaluated the stimulation parameters for the GES system concluded that altering the GES parameters can improve GP symptoms and that these alterations may need to be tailored based on the patient's response [13][14][15][16][17]. Abidi et al. recommended using an algorithmic method to change stimulation parameters in response to patients' symptoms.According to this study's findings, patients who did not respond to the first stimulation settings were able to have at least a 50% reduction in symptoms when using their algorithm as opposed to a 36% improvement with GES therapy alone [16].According to McCallum et al., an increase of 1 mA in the stimulation current led to an increase in voltage value across all GP etiologies, as well as an improvement in their GP symptoms [13].Both of these studies showed that increasing some parameters, and consequently the energy provided during utilization of GES, led to an improvement in symptoms, despite using different approaches for altering device stimulation parameters.
Our approach to this research focused on modifying the stimulation parameters and adjusting the voltage to the right level for each patient to achieve optimal symptomatic improvements.Any change in the voltage only occurred after the careful assessment of symptom status, particularly the degree of nausea/vomiting, was completed.
When comparing readings from the beginning of GES therapy to the most recent follow-up, our findings showed a 42% rise in mean impedance over time, which led to a 15% increase in mean voltage and a 26% increase in mean current (Table 2).When comparing TSS between the start of GES therapy and the most recent follow-up visit, our study revealed significant improvements in TSS and each individual symptom score, with nausea and vomiting showing the largest numerical improvement (Table 4).In addition, our study demonstrated that adopting this rational strategy to update the GES settings over an extended period of time reduced the mean numerical improvement in TSS by approximately 60%.Also, based on our clinical experiences, we have been evaluating GP patients, who were followed up at our center for a second opinion related to the lack of response to GES therapy initiated by the referring physicians.Those interesting cases of patients presenting with non-traditional device stimulation settings and uncontrolled GP symptoms highlighted the potentially detrimental effects of non-traditional, more "adventurous" stimulation configurations, and the need for further guidance to optimize GES stimulation parameters.Our adjustments of the GES parameters (Case #1: voltage of 10 V, current of 22 mA changed to 5 V, and 8.1 mA.Case #2: frequency changed from 28 Hz to 14 Hz; time ON/OFF from 2 s/3 s to 0.1 s/5 s) reported symptom alleviation after returning the parameters to their original, almost default settings.This re-introduction of the appropriate, conservative settings was able to restore the long-term control of GP symptoms and patients' QoL.Note: Those two patients were not included in our cohort analysis and were not part of the current investigation.
Our research suggests, when considered collectively, that a rational approach to GES parameter modification results in excellent and long-lasting symptomatic control in GP patients, independent of the underlying etiological cause.The longer use of GES therapy was also made possible in our study by the system's conservative programming, which prevents early battery depletion.Generally speaking, drug-refractory GP patients undergoing GES therapy appear to symptomatically benefit from a gradual increase in current/voltage parameters, but the ideal stimulation parameters appear to differ amongst patients and must be appropriately titrated based on symptoms of nausea and vomiting.
It is also important to mention that the FDA has specified two GP etiological categories under the approved status of the GES, which allowed us to enroll only individuals with diabetic and idiopathic gastroparesis.While performing very sophisticated research studies, many of which are funded by NIH/NIDDK grants, we are constantly looking for potential distinctions and similarities between those groups of gastroparetics.We believe that understanding their neurobiological and pathophysiological bases, disease progression, clinical presentations, systemic complications, and complex responses to treatment alternatives is critical to our findings and the development of new diagnostic tests and therapy modules.As a result, we believe it was necessary to learn whether both GP groups responded to gastric stimulation with the same positive control of their symptoms, and if the extent of GES parameters is sufficient and comparable in the DM and ID groups (Table 3).It might have been possible to observe significant effects of GES intervention only in one of the gastroparesis etiologies.Hence, we decided to look at two groups of patients with gastroparesis symptoms attributed to specific etiologies with different pathophysiologies and analyzed their research data separately.We believe our findings will allow clinicians to personalize the GES therapy based on our recommendations.
Previous research has demonstrated that diabetic and postsurgical GP patients have more symptom alleviation with GES therapy with pyloroplasty than idiopathic GP patients [2,9,13].When examining stimulation parameters stratified by GP etiology, past studies have revealed varying results.Abidi et al. observed that postsurgical GP patients needed the most energy to maintain symptom improvement.In contrast, McCallum et al. discovered that idiopathic GP patients needed the most energy but also had the worst symptom outcomes when comparing diabetic, idiopathic, and postsurgical GP patients [13,16].The authors noted that idiopathic GP is made up of a diverse range of patients who frequently experience abdominal discomfort/pain, which is less likely to respond to GES therapy.Our study did not find any differences in GES parameters or TSS between diabetic and idiopathic GP patients.Therefore, the optimum settings in the GES parameters may improve TSS, regardless of their etiology.
Although we did make cautious adjustments to the GES stimulation parameters as necessary, the retrospective design of our study had the drawback that we did not have a predefined stepwise method for altering parameters at follow-up visits depending on patients' complaints.We acknowledge the limitations inherent in our research, namely the limited sample size of patients counted in the analysis, and the frequency of visits including the interrogation of the system.However, it is worth acknowledging that as the sole institution overseeing this project, we were able to implement identical programming principles executed by a single individual, a member of our team.As investigators, we actively participated in the clinical long-term follow-up care of GP patients and continued to focus on the energy depletion of GES batteries.We perform pulse generator replacements if necessary, while on average, we can administer the delivery of the GES therapy for up to 10 years.
In conclusion, our study evaluated the changes in patient-reported symptom scores and GES setting characteristics during the course of long-term treatment for refractory diabetic and idiopathic GP patients.We conclude that maintaining parameters close to the default settings and making adjustment decisions for voltage and current based on specific symptoms of nausea and vomiting are very optimal approaches to programming in GES patients.However, the frequency, pulse width, and on/off were never altered from the default settings in our study.Our patients showed considerable improvement in TSS during the extended follow-up lasting up to 73 months, utilizing conservative and symptom-based methods of adjusting GES parameters.
Through this article, we aim to assist fellow investigators in integrating our guidelines into their interrogation of the GES system.Additionally, we provide the rationale to support the use of our recommended parameters, which not only contribute to symptom improvement but also extend the lifespan of GES batteries.

Conclusions
The inclusion of specific cases illustrates the situation in which using GES parameters that differ from the usual values and are randomly chosen is not advantageous in longterm gastric stimulation.Adopting a rational albeit conservative approach to GES settings which utilize the "default settings" as guidelines while augmenting current and voltage components resulted in a significant relief of GP symptoms.
Our data indicate that adopting a conventional approach when determining GES parameters leads to the best and most sustained management of symptoms in patients with gastroparesis, regardless of the etiology and duration of stimulation therapy.

Supplementary Materials:
The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/gidisord6020037/s1.Table S1: Comparison of device stimulation parameters at follow-up between gastroparesis etiologies, Table S2: Adjusted association of change in each GES parameter with TSS, Table S3: Unadjusted effect of change in each GES parameter on TSS.Informed Consent Statement: Patient consent was waived due to this step is not required by the FDA, as the therapy with a gastric electrical stimulation system is approved.Gastric Electrical Stimulation (GES) is a bioelectric therapy that received the Food and Drug Administration (FDA) approval as a Humanitarian Use Device (HUD) in September 1999 and subsequently, it was permitted to be used as a Humanitarian Device Exemption (HDE) for medication-refractory gastroparesis of diabetic or idiopathic origin in March 2000.HUD label means that the effectiveness of this device has not been established, but the device is safe and could benefit qualified patients diagnosed with GP.The consent form is not mandatory to be executed.

Table 4 .
TSS score comparison between the initiation of GES therapy and the most recent follow-up.

Table 1 .
Baseline characteristics, GES parameters, and TSS of GP patients.

Table 3 .
Comparison of follow-up GES parameters and total symptom score based on gastroparesis etiology.
Device stimulation parameters and total symptom score are reported as mean (standard deviation, or SD).

Table 5 .
Unadjusted association of each cofactor with overall TSS score.