New UC histology index 'reproducible, responsive and valid'

Last Updated: 2015-10-29

By Reuters Staff

NEW YORK (Reuters Health) - An international team of pathology and biostatistics experts has developed a new instrument for measuring ulcerative colitis (UC) disease activity.

While the Robarts Histology Index (RHI) still requires external validation, "the development of RHI is an important advance," Dr. Barrett Levesque of the University of California, San Diego and colleagues write.

There are existing histological activity indices for UC, but "the limited data available indicate their operating properties may be suboptimal," Dr. Levesque and his team wrote in their report online October 16 in Gut.

Neither the Geboes score (GS) nor the modified Riley score (MRS) have been adequately validated, they add, and while studies have found intra-rater reliability is good for both measures, inter-rater reliability is not.

The investigators had previously used the RAND consensus process to generate histological scoring conventions for the GS and MRS. In the current study, they sought to reassess the reliability of the GS and MRS using the new scoring methods, and develop a new index using the most reliable items.

Four pathologists scored 48 UC colon biopsies using the GS, MRS, and a visual analog scale (VAS) global rating on three separate occasions two weeks apart. For each index, the researchers measured intra-rater and inter-rater reliability using interclass correlation coefficients. The most reliable items were incorporated into the RHI, on which the total score can range from 0 (no disease activity) to 33 (severe disease activity).

The RHI can be calculated as 1 x chronic inflammatory infiltrate (4 levels) plus 2 x lamina propria neutrophils (4 levels) plus 3 x neutrophils in epithelium (4 levels) plus 5 x erosion or ulceration (4 levels after combining Geboes 5:1 and 5:2).

In the validity testing phase, the researchers compared predicted versus observed correlations between the RHI and clinical and endoscopic outcome measures using linear regression.

"RHI was shown to be reproducible, responsive and valid," they wrote.

Multiple new therapies are being developed for UC, Dr. Levesque and his team write, which may lead to improved patient care, but "has created a substantial challenge."

They conclude: "Recruitment of patients into clinical trials has become increasingly difficult, and therapies are being developed for UC with evolving definitions of treatment response that specify deeper levels of remission than endoscopic healing. As such, histological evaluation of disease activity is likely to become an integral part of the management of UC. We believe that RHI will help enable future clinical research in UC."

Dr. Levesque did not respond to an interview request by press time.

SOURCE: http://bit.ly/1jVnXXP

Gut 2015.