Background: Endoscopic scores, e.g. Mayo Endoscopic Subscore (MES), are validated scoring systems used to grade endoscopic severity in ulcerative colitis (UC). However these scores don't incorporate any measure of disease extent. The aim of this study was to determine the feasibility of using a composite score to quantify UC inflammatory burden.
Methods: A retrospective analysis of faecal calprotectin (FC) results was performed (Jan 2015-June 2016). UC patients with contemporaneous endoscopic evaluations and FC measurements were included. The DUBLIN score was calculated as a product of the MES and disease extent, i.e. E1 = limited to sigmoid/rectum E2 = limited to left colon (distal to splenic flexure) E3 = proximal to splenic flexure. Analysis of correlation between DUBLIN score and its individual components with FC, an objective marker of inflammation was performed. ROC curves were compared.
Results: 279 FC results (n=181 patients, confirmed UC) were included. n=74 had contemporaneous endoscopy; it was possible to calculate the DUBLIN score in 70 cases. Patient characteristics and endoscopic activity as per Table 1.
The median FC result was 186 ug/L (IQR 23–1108ug/L), 77.5ug/L (IQR15–833 ug/L) and 32ug/L (IQR 15–432ug/L) in those with pancolitis, left sided colitis and proctosigmoiditis respectively.
There was a significant correlation between the MES and FC (r=0.359; p<0.01). A significant correlation was also identified between C-reactive protein (r=0.270, p<0.01) and Albumin (r=−0.129, p=0.04) and endoscopic activity.
Extent of disease also correlated significantly with FC (r=0.286; p=0.016). However, a stronger correlation was observed between DUBLIN score and FC (r=0.394; p<0.01). ROC curves demonstrated an AUC =0.79 for the MES in predicting FC; AUC=0.75 when calculated for disease extent. A greater area under the curve was observed when ROC was constructed for the Dublin score. (AUC=0.87) (Figure 1)
Conclusions: The DUBLIN score is a simple composite score that can be calculated at the time of endoscopy or retrospectively. These results show that it has potential as a measure of inflammatory burden in UC. A user-friendly scoring system that better reflects overall luminal inflammation may be particularly useful in personalising therapy.