15% to 20% of ulcerative colitis (UC) patients will have a flare warranting hospitalization, with approximatey 20% progressing to colectomy. Data on predictors of rapid progression to colectomy (<30 d) are limited. Infliximab (IFX) is now common practice in the management of hospitalized UC patients but standard dosing regimens are currently based on the ACT1 and 2 trials performed in outpatient moderate-severe UC. Hospitalized UC patients have a higher inflammatory burden resulting is more rapid IFX clearance, suggesting higher up-front dosing of IFX may be needed to avoid colectomy. Our aim was to use propensity-scoring methodology to examine predictors of 30-days outcomes and determine whether initial IFX dosing strategy is associated with 30-days colectomy rates.Methods:
This was a retrospective study of UC patients admitted or transferred to a tertiary care IBD service between January 2011 and April 2016. Inclusion criteria were: UC diagnosis confirmed prior to admission based on standard criteria; hospitalized for acute UC; IFX-naïve; and at least one induction dose of IFX administered during hospitalization. The primary outcome was 30-days colectomy rate; additional outcomes included 90 days colectomy rate, need for additional IFX within 7 days of initial infusion, length of stay (LOS), and medical and surgical complication rate. Inverse probability-weighted propensity-scores (IWPS) were constructed to account for potential selection bias between patients receiving IFX 5 mg/kg (standard group, SD) and 10 mg/kg (high-dose group, HD). Univariate and multivariate testing were performed.Results:
146 patients were analyzed, 120 SD and 26 HD. HD patients were younger (35 [16–82] versus 22 [8–86] yr), otherwise the groups were similar in terms of demographics, baseline disease characteristics, and former/current non-IFX anti-TNF use. Mean admission CRP was higher in the SD versus HD group (68 versus 37.8 mg/L), but other parameters including albumin and hemoglobin (hb) nadir were similar. Twenty-five patients underwent colectomy by 30 days (17.1%) and 33 patients by 90 days (22.6%). Fewer patients in the HD group (15.4%) underwent colectomy versus the SD group (17.5%) at 30 days (P < 0.001). Twenty-two patients in the SD group compared to 0 patients in the HD group needed an additional IFX dose within 7 days of the first IFX infusion. The HD group had shorter LOS and fewer complications. Hb nadir (aOR 1.43, 1.23–1.58, P < 0.001), admission CRP (aOR 1.01, 1.00–1.02, P = 0.02), pediatric provider type (aOR 20.6, 5.4–78.7, P < 0.001), and need for additional IFX within 7 days (aOR 4.4, 1.49–13.2, P = 0.007) were independently associated with 30-days colectomy rate; receiving 10 mg/kg as the initial dose was associated with 80% lower odds of 30-days colectomy (aOR 0.20, 0.06–0.67, P = 0.009). Albumin <3 g/dL, hb nadir <8 g/dL, and CRP >10 g/dL, predicted 30-days colectomy with 91% sensitivity and 40% specificity (AUC 0.72, 0.60–0.84, P = 0.001).Conclusions:
Up-front dosing of IFX 10 versus 5 mg/kg is associated with an 80% reduction in 30-days colectomy, fewer complications, and shorter LOS. There should be a low threshold for early colectomy instead of additional IFX if the response to the initial infusion is inadequate. Prospective studies validating our findings are needed.