Endoscopic and histologic remission are therapeutic targets in ulcerative colitis (UC). Endocytoscopy (EC), provides optical magnification up to 520x, this real-time imaging modality offers the potential to assess histologic healing without tissue sampling. The objective of this study was to evaluate the diagnostic performance of EC to assess endoscopic and histologic healing in UC.

We conducted a prospective observational study including consecutive patients with UC in clinical remission undergoing colonoscopy. EC and biopsies were performed and in the segment with the highest previously documented inflammatory activity. Endoscopic activity was evaluated using the Mayo endoscopic subscore and UCEIS. Clinical activity (PRO2) and biochemical markers (fecal calprotectin and C-reactive protein) were recorded. Endocytoscopy (ECI-H290ECI, Olympus) was evaluated using three different scores (ELECT, ICC and ESS) using NBI and 1% methylene blue staining. Three independent endoscopist reviewed EC videos for interobserver correlation. Histologic activity using Nancy index served as the gold standard. 

Thirty-seven patients were included. Mean age was 52,3 (SD16,6) and 56,8% were female. Baseline characteristics are shown in Table 1. Only three patients (8,6%) had demonstrated mucosal healing on the prior endoscopy. Clinically, patients were in remission, with normal bowel movements number (73% reporting 1 bowel movement/day) and no rectal bleeding in 100%.  Median fecal calprotectin was 39.5 µg/g (IQR 19.1-91.3), CRP 1.2mg/L (IQR 0.5-5.4), hemoglobin 14.8 g/dL (IQR 13.6-15.4), and albumin 4.5 g/dL (IQR 4.4-4.7).

Most patients were in endoscopic remission, with 90% with Mayo 0 and UCEIS 0. Histologic remission (Nancy 0-1) was present in 78,4% of patients.  

EC assessment showed 81% patients scored as ELECT 0 or 1, 83% as ICC 0 or 1 and 57% as ESS 3. 

Concordance between Nancy and EC indexes was good. 93,1% of patients in histologic remission were correctly classified as ELECT 0 or 1, ICC 0 or 1 while 62% as ESS 3.

Among 6 patients with significant histologic activity (Nancy 2-3), 3 were clasified as ELECT 2-3, 2 as ICC 2 and 4 as ESS 4.

Interobserver agreement was good for ELECT score (kappa 0.67), and moderate for ICC and ESS (kappa 0.4). Concordance was 83.3% for ELECT and 66,7% for ICC and ESS scores. Of individual variables vascular architecture, distance between crypts, and abscence of crypt abscesses had good agreement, while capilar morphology and presence of inflammatory infiltrate only had 33,3% concordance.  

Endocytoscopy allows accurate in vivo assessment of histologic remission in patients with ulcerative colitis, with an excellent negative predictive value and could be a feasible alternative to biopsies in patients in endoscopic remission.