Factors associated with gastrointestinal obstruction in phytobezoar patients: an emergency ultrasound study.

INTRODUCTION

Phytobezoars are a common cause of acute abdominal pain in northern China and may lead to serious complications including gastrointestinal obstruction (GIO). Early identification of factors associated with GIO is essential for timely intervention. This study aimed to investigate the ultrasound characteristics of phytobezoars and identify factors associated with GIO.

METHODS

A retrospective analysis was conducted on 162 patients with ultrasound-diagnosed phytobezoars in the emergency department, divided into GIO (n = 36) and non-GIO (n = 126) groups. Demographics, bezoar characteristics, and ultrasound findings were compared. Univariate analyses (Chi-square test, independent t-test, and Mann-Whitney U-test) and binary logistic regression were used to identify factors associated with GIO. Firth penalized logistic regression including location and a subgroup analysis restricted to pyloric/small-intestinal bezoars were performed as sensitivity analyses.

RESULTS

Phytobezoars occurred predominantly in elderly females, with peak incidence from October to April. Compared with the non-GIO group, the GIO group had a higher proportion of males (p < 0.05), higher median age (68 vs. 63 years, p < 0.05), and smaller bezoar diameter (41.11 ± 8.57 vs. 48.20 ± 13.75 mm, p = 0.004). Notably, all GIO cases had bezoars located in the pylorus or small intestine, whereas none of the 101 gastric body bezoars were associated with obstruction. The primary logistic regression model (excluding location due to complete separation) identified male sex (OR = 3.761), age (OR = 1.054), and bezoar diameter (OR = 0.961) as associated factors (all p < 0.05). However, after accounting for location-the dominant predictor of GIO (OR = 220.30, p < 0.001)-these associations were no longer statistically significant.

DISCUSSION: The incidence of phytobezoars follows distinct seasonal patterns. Bezoar location is the strongest determinant of GIO, and the associations of sex, age, and diameter observed in the primary model appear to be partly confounded by location. Emergency ultrasound enables rapid identification of bezoar location at the bedside, which may assist in early clinical decision-making. Prospective studies with independent diagnostic confirmation are needed to validate these findings.