Symptom Associations with Delayed Gastric Emptying Vary by Disease Phenotype

Background & Aims: The association between upper gastrointestinal (GI) symptoms and delayed gastric emptying time (GET) is debated. This study utilized Body Surface Gastric Mapping (BSGM) and the 'Auckland Classification' BSGM phenotyping scheme to investigate whether symptom associations with GET vary across different mechanistic phenotypes. Methods: A pooled analysis was performed on two prospective datasets of 194 patients with chronic upper GI symptoms. Participants underwent simultaneous BSGM (Gastric Alimetry) and gastric emptying testing (breath test or scintigraphy). Validated time-of-test symptom profiling was recorded at 15-minute intervals using 0-10 Likert-type scales. Patients were classified: 'Spectral Abnormal' (abnormal electrophysiology), 'Sensorimotor' (symptoms correlate with gastric electrical amplitude), 'Continuous' (symptoms uncorrelated), and 'NA' (no phenotype). Results: In the analyzed cohort, 24% had delayed GET. Overall, delayed GET was associated with a higher total symptom burden (15.6 delayed vs. 8.8 normal, p=0.006; r=0.30, p<0.001), specifically postprandial fullness (p=0.004) and early satiation (p=0.015). However, associations varied significantly by phenotype. In the 'Sensorimotor' phenotype, nausea was associated with delayed GET (4.9 vs. 1.0, p=0.027; r=0.59, p=0.009), while total symptom burden was 23.3 vs. 8.4 (p=0.135). Conversely, patients with 'Continuous' or 'Spectral Abnormal' phenotypes, or who were unclassified, showed no symptom associations. Conclusions: Delayed GET is weakly associated with an increased burden of upper GI symptoms. However, gastroduodenal disorders are heterogeneous, such that phenotyping reveals this association to be exclusive to patients with a 'Sensorimotor' phenotype. These results could improve targeting of therapies that address gastric emptying.