EndoFLIP Guided Assessment of Pyloric Distensibility Identifies Associations With Delayed Gastric Emptying and Symptoms of Gastroparesis.

BACKGROUND

Pyloric dysfunction assessed by EndoFLIP has been implicated in the pathogenesis of gastroparesis. However, the relationship between pyloric dysfunction, gastric emptying, and severity of gastroparesis symptoms remains unclear.

AIM: Investigate the association of pyloric distensibility index (DI) and area under the curve (AUC) for a DI to volume curve with percent of gastric retention and severity of gastroparesis symptoms.

METHODS

We retrospectively reviewed the electronic charts of 321 adult patients who underwent graded pyloric sphincter distension over a range of 30-70 mL balloon volume using an EF-322 EndoFLIP catheter during a sedated upper endoscopy. To be included in the analysis, participants should have also completed questionnaires about severity of gastroparesis symptoms before EndoFLIP (n = 229). Gastric emptying study (GES) test results completed within the last 10 years were abstracted from chart review (n = 132). AUC for adjacent balloon volumes (40-50, 50-60 and 60-70 mL) were calculated using the trapezoid formula. Associations between DI, AUC, gastric emptying and GI symptoms were assessed using non-parametric independent sample tests, and Spearman's correlation. Receiver operating characteristic (ROC) curve was used to calculate the ability of DI and AUC to distinguish between those with normal and abnormal gastric emptying.

RESULTS

Median age of the study population was 50 years (IQR 20.0). Majority of patients were female (79.9%), and White (89.1%). More than half of those with prior GES had gastroparesis (N = 68, 51.5%). Pyloric DIs at 60 and 70 mL and AUCs at 50-60 and 60-70 mL were smaller in those with delayed gastric emptying compared to those with normal gastric emptying. Pyloric DI at 50 mL (rho = -0.16, p = 0.014) and AUCs at 40-50 mL (rho -0.16, p = -0.018) and 50-60 mL (rho = -0.16, p = 0.019) showed significant associations with severity of stomach fullness, decreasing as severity of the symptom worsened. Pyloric DI of 8.9 mm2/mmHg at 60 mL (94.7% sensitive, 26.2% specific, AUC 0.61, p = 0.029) and 6.25 mm2/mmHg at 70 mL (92.5% sensitive, 31% specific, AUC 0.62, p = 0.038) distinguished between normal and delayed gastric emptying. In linear mixed-effects modeling, anesthesia type demonstrated a volume-dependent effect on pyloric DI, with monitored anesthesia care (MAC) associated with higher DI than general anesthesia (GA) at increasing balloon volumes (anesthesia × volume p < 0.001). Further, female sex was independently associated with lower DI (β -3.55, p = 0.002), but this was offset by a significant anesthesia × gender interaction, whereby females exhibited relatively higher DI under monitored anesthesia care compared to general anesthesia (β +3.49, p = 0.005).

CONCLUSION

Pyloric sphincter DI and AUC at higher balloon volumes differentiated those with normal from those with delayed gastric emptying. The relationship with severity of gastroparesis symptoms was limited to stomach fullness. The effect of type of anesthesia on pyloric distensibility depends on balloon volume and sex of the patient. Pyloric DI and AUC at 60 and 70 mL could be helpful to phenotype the pyloric sphincter in patients with gastroparesis symptoms.