Characteristics and risk factors for poor outcome in patients with systemic vasculitis involving the gastrointestinal tract

Gastrointestinal (GI) involvement was described to be a poor prognostic factor in systemic necrotizing vasculitis. Its prognostic significance may vary according to clinical presentation and vasculitis subtype. This study investigated risk-factors associated to poor outcome in GI-involvement of vasculitis. Patients with systemic vasculitis as defined by the 2012 Chapel Hill Consensus Conference and presenting with GI involvement were retrospectively included. Baseline characteristics, treatments and outcome were recorded. Primary endpoint was a composite of admission to intensive care unit (ICU), emergency surgical procedure, or death. Two hundred and thirteen patients were included. Vasculitis were distributed as follows: 41% IgA vasculitis, 27% ANCA-associated vasculitis, 17% polyarteritis nodosa (PAN), and 15% other vasculitis. Eighty-three (39%) patients fulfilled the composite primary endpoint within 6 months. Predictive factors associated with the primary endpoint included PAN subtype (OR 3.08, 95% CI 1.29–7.34), performance status (OR 1.40, 1.05–1.87), use of morphine (OR 2.51, 0.87–7.24), abdominal guarding (OR 3.08, 1.01–9.37), ileus (OR 2.29, 0.98–5.32), melena (OR 2.74, 1.17–6.42), increased leukocytes (per G/L, OR 1.05, 1.00–1.10), low hemoglobin (per g/dL, OR 0.80, 0.71–0.91) and increased CRP (log mg/L, OR 1.21, 0.94–1.56). A risk prediction model for the achievement of primary endpoint had a very good performance [C-statistics 0.853 (0.810 to 0.895], and for overall survival as well. Vasculitis presenting with GI involvement have a poor outcome in more than one third of cases. An easy-to-use risk prediction model had a very good performance to predict the admission to ICU, emergency surgical procedure, or death. [Display omitted] •Gastrointestinal involvement in systemic vasculitis is a poor prognostic factor.•Risk factors associated to poor outcome were included in a scoring system.•These features quantified the predicted risk for clinical deterioration.