Accuracy of Faecal Immunochemical Test to Predict Endoscopic and Histological Healing in Ulcerative Colitis: A Prospective Study Based on Validated Histological Scores.
J Crohns Colitis. 2017 Sep 01;11(9):1071-1077
Authors: Shi HY, Chan FKL, Chan AWH, Higashimori A, Kyaw M, Ching JYL, Luk AKC, Wong SH, Wu JCY, Sung JJY, Ng SC
Background and Aims: Endoscopic and histological healing are associated with improved clinical outcomes in ulcerative colitis [UC]. We aimed to investigate the predictive value of faecal immunochemical test [FIT] for endoscopic and histological healing in UC.
Methods: We measured quantitative FIT and faecal calprotectin [FC] in 140 consecutive UC patients who underwent colonoscopy. We assessed the diagnostic accuracy of FIT for predicting endoscopic healing using the Mayo endoscopic subscore [MES 0/1] and for histological healing using the Geboes score [< 2.0] and Nancy index [grade ≤ 1]. The predictive abilities of FIT were compared with those of FC.
Results: FIT had an area under the curve [AUC] of 0.77 (95% confidence interval [CI] 0.67-0.86, p < 0.001) for endoscopic healing, an AUC of 0.77 [95% CI 0.67-0.86, p < 0.001] using the Geboes score, and 0.77 [95% CI 0.66-0.85, p < 0.001] using the Nancy Index for histological healing. The AUC of FIT was comparable to that of FC for endoscopic healing [p = 0.773] and histological healing [p = 0.767-0.960], and was comparable to colonoscopy for histological healing [p = 0.384-0.673]. FIT < 50 ng/ml predicted endoscopic healing with a sensitivity, specificity, and positive predictive value [PPV] of 72%, 68%, and 82%, respectively, and for histological healing with a sensitivity, specificity, and PPV of 73-75%, 67%, and 78-80%, respectively. Combining FIT with FC led to a higher specificity [90%] for histological healing. Over 85% of patients with FIT < 50 ng/ml and FC < 50 μg/g achieved histological healing.
Conclusions: FIT is highly sensitive and accurate to predict endoscopic and histological healing in UC. It represents a promising non-invasive tool for monitoring mucosal healing in UC.
PMID: 28881876 [PubMed – in process]