Y-ECCO Literature Review: Viktor Domislović

Viktor Domislović

13 3 ViktorViktor Domislović 
Viktor Domislović

Cumulative burden of inflammation predicts colorectal neoplasia risk in Ulcerative Colitis: A large single-centre study

Choi CR, Al Bakir I, Ding NJ, Lee GH, Askari A, Warusavitarne J, Moorghen M, Humphries A, Ignjatovic-Wilson A, Thomas-Gibson S, Saunders BP, Rutter MD, Graham TA, Hart AL
Gut. 2017 Nov 17. pii: gutjnl-2017-314190. doi: 10.1136/gutjnl-2017-314190.



Currently, the total duration of Ulcerative Colitis (UC) is regarded as a proxy measure for the cumulative inflammatory insult; this approach has disadvantages since it is assumed that the degree of inflammatory activity remains constant over time. In fact, patients with persistently active disease and those with quiescent disease have a small probability of having the same accumulated inflammatory damage. Moreover, assessment of the severity of inflammation on the basis of just a single or the most recent colonoscopy provides only a cross-section of the individual's disease history without taking into consideration the time component and significant inflammatory burden, which may have an important effect on carcinogenesis. This study examined the association between the cumulative inflammatory burden (CIB) and the risk of developing colorectal neoplasia (CRN) in patients with UC, and compared the CIB with known risk factors like duration of colitis and severity of inflammation based on a single recent colonoscopy. A pragmatic risk stratification strategy was suggested, which could be used to accurately predict the risk of development of CRN in patients with UC in daily clinical practice.

This retrospective single-centre study enrolled 987 patients with extensive UC who were under colonoscopy surveillance, of whom 97 (9.8%) developed CRN over a median of 13 years of follow-up. At each surveillance, the severity of microscopic and macroscopic inflammation was scored from 0 (no activity) to 3 (severe activity). For each surveillance interval, the CIB was calculated as the average microscopic severity between the two surveillance episodes multiplied by the length of the surveillance interval. For example, if one surveillance interval was between 2004 and 2006, with corresponding microscopic severity of 3 (severe) and 2 (moderate), the CIB for this surveillance interval would be 3+2/2=2.5 multiplied by the length of the surveillance interval (2 years), which is 5. The overall CIB was then obtained by summing the CIB scores from all surveillance intervals.

Key findings

The results of the study strongly suggest that the risk of developing CRN is associated with the CIB, indicating that the risk of neoplasia in patients with UC is positively correlated to the total amount of inflammatory damage accumulated over time.

Accumulative inflammatory burden was strongly associated with CRN risk, with a twofold increase in risk for approximately 10, 5 or 3.3 years of continuously mild, moderate or severe active microscopic inflammation, respectively. Inflammation-based risk stratification was only sufficiently predictive when information from multiple colonoscopies was used, and it is also important to note that severity of inflammation in the most recent colonoscopy alone was not significantly associated with CRN risk. The importance of the CIB was further emphasised by the fact that the predictive accuracy of inflammation severity markers (i.e. mean severity, maximum severity and persistency) improved in a linear fashion when more surveillance procedures were included in the calculation of these scores.

Although it is clear that the most accurate risk estimation may be achieved only by assessing a patient’s entire surveillance history, this is often impractical in daily clinical practice, especially when dealing with complex patients with a long disease history. The mean severity of microscopic inflammation derived from all surveillance procedures (based on the segment worst affected by colitis only) performed in the last 5 or 10 years (consisting of approximately three to four colonoscopies) proved an accurate marker of the CRN risk and can be rapidly calculated in a busy clinical practice.

The persistency of inflammation was significantly associated with the risk of developing CRN. This association remained significant even after adjusting for the mean severity. This result suggests that patients with more persistent, chronically active colitis are at risk of developing CRN independently of inflammation severity. Thus, cancer risk management should involve early identification of patients with persistently active disease.


Accurate estimation of CRN risk is unlikely to be achieved using only a single or the most recent colonoscopy; rather, assessment of multiple colonoscopies performed over many preceding years is required to accurately estimate cumulative effect and influence of inflammation. The CRN risk stratification should involve assessment of multiple surveillance episodes to take into account the effect of the CIB. Calculation of the mean severity of microscopic inflammation over the preceding 5 or 10 years may offer an accurate and pragmatic risk stratification strategy. Moreover, patients with severe and persistent inflammation should be medically optimised with the aim of achieving mucosal healing and prioritised to undergo more intensive surveillance protocols to allow timely and early detection and intervention for CRN.

Limitations of this study are its retrospective design, incomplete assessment of the possible effects of interobserver variability in grading inflammation, and lack of separate validation of endoscopic and histological inflammation scores. Furthermore, given the fluctuating nature of inflammation severity over time, any change in the severity that occurred during the period between examinations would not have been captured. This is of particular significance for the CIB, since it assumes a constant degree of inflammation during each surveillance interval, which could lead to misinterpretation of results. To obtain a more complete view of the patient’s inflammatory activity during the interval, one should consider using other clinical indices such as the clinical Mayo score or the Clinical Activity Index (CAI) as well as biochemical indices including calprotectin and lactoferrin, but this should be verified in future research.


Viktor Domislović is a Gastroenterology and Hepatology resident at the Clinical Hospital Centre, Zagreb and is currently performing postgraduate study at the University of Zagreb. Viktor is interested in both the clinical and the scientific aspects of IBD: Epidemiology, clinical nutrition, biomarkers and microbiota. 13 3 Viktor
Viktor Domislović 
Viktor Domislović  

Posted in ECCO News, Y-ECCO Literature Reviews, Committee News, Y-ECCO, Volume 13, Issue 2