This was a single-center, retrospective observational cohort study of 132 patients with IgA vasculitis and gastrointestinal involvement who had complete endoscopic data. The study examined endoscopic features, such as ulceration and multi-segment GI involvement, as exposures related to hospitalization length.
The primary outcome was prolonged length of hospitalization (LOS >13 days). Endoscopic ulceration was independently associated with prolonged LOS (OR 2.40, 95% CI 1.09–5.27, p=0.029). Multi-segment GI involvement was also independently associated with prolonged LOS (OR 2.58, 95% CI 1.20–5.58, p=0.016).
For the secondary continuous outcome, ulceration was associated with a 27.7% longer LOS (RR 1.277, 95% CI 1.047–1.564, p=0.018). Multi-segment involvement was associated with a 34.7% longer LOS (RR 1.347, 95% CI 1.035–1.748, p=0.025).
Safety and tolerability data were not reported. Key limitations include the single-center retrospective design, potential confounding, and the fact that associations do not imply causation. Practice relevance is that endoscopic phenotyping may inform clinical assessment in IgA vasculitis with GI involvement, but findings are associations, not prognostic predictions.
View Original Abstract ↓
To evaluate associations between endoscopic features and hospitalization outcomes in IgA vasculitis (IgAV) with gastrointestinal (GI) involvement, focusing on endoscopic subtypes associated with prolonged hospitalization.
We analyzed 132 IgAV patients with GI involvement and complete endoscopic data at a large-volume center (January 2019 to December 2024). Clinical manifestations, endoscopic characteristics, laboratory data, and abdominal imaging findings at admission were reviewed. Prolonged length of stay (LOS) was defined as LOS exceeding the 75th percentile (>13 days). Multivariable logistic regression and gamma regression (log link) were used to examine factors associated with prolonged LOS and LOS as a continuous outcome. Sensitivity analyses additionally adjusted for treatment variables and renal involvement. Laboratory parameters were compared across endoscopic subgroups in exploratory analyses.
The median age of the patients was 18 years (IQR: 15.0–45.5), with a male-to-female ratio of 2:1. Abdominal pain was the most common presenting symptom. Bowel wall thickening was the most frequent finding on abdominal imaging. The most prevalent endoscopic finding was congestion/edema, followed by erosions, ulcers, and petechiae/ecchymosis. Endoscopic ulceration (OR 2.40, 95% CI 1.09–5.27, p=0.029) and multi-segment GI involvement (OR 2.58, 95% CI 1.20–5.58, p=0.016) were independently associated with prolonged LOS. Gamma regression showed that ulceration was associated with a 27.7% longer LOS (RR 1.277, 95% CI 1.047–1.564, p=0.018) and multi-segment involvement with a 34.7% longer LOS (RR 1.347, 95% CI 1.035–1.748, p=0.025). The association with ulceration remained significant after adjusting for treatment factors, whereas the association with multi-segment involvement was attenuated. The association of ulceration with LOS was more pronounced in patients aged ≤18 years. In sensitivity analyses, additional adjustment for renal involvement did not materially change the associations between endoscopic features and LOS.
In IgAV with GI involvement, endoscopic ulceration is the endoscopic feature most consistently associated with longer hospitalization, and the association appears stronger in patients aged ≤18 years. Multi-segment involvement shows a weaker association that attenuates after treatment adjustment. Overall, these findings support the clinical value of endoscopic phenotyping to inform clinical assessment and should be interpreted as associations rather than prognostic predictions.