Search in the Abstract Database

Abstracts Search 2014

* = Presenting author

P168. Predicting outcome in acute ulcerative colitis: Comparison of the Travis and Ho scores using UK IBD Audit data

A.M. Churchhouse1, R.W. Lynch1, A. Protheroe2, I.D. Arnott1, 1NHS Lothian, Gastroenterology, Edinburgh, United Kingdom, 2RCP, CEEU, London, United Kingdom


Patients with severe Ulcerative Colitis (UC) are commonly identified using the Truelove and Witts [1] criteria on presentation. The Travis [2] and Ho [3] scores are subsequently used to identify patients with severe UC who are at high risk of failing medical therapy and needing second line therapy or colectomy. To date there has been no direct comparison between Travis and Ho scores to determine which is superior.


We analysed data from 3049 patients with ulcerative colitis collected during the 2010 round of the UK IBD audit. 984 patients had acute severe UC according to the Truelove and Witts criteria. Those that failed steroid therapy were scored using both Travis and Ho criteria and allocated into either a “Travis high risk” (n = 190) or “Travis low risk” (n = 201) group and either a “Ho high risk” (n = 165), “Ho medium risk” (n = 150) or “Ho low risk” (n = 65) group. We assessed whether further medical therapy or surgical intervention varied between these groups.


Age and sex did not differ between the groups. Patients requiring surgery did not differ between the high risk groups (Travis 49%, n = 93 and Ho 51%, n = 84, respectively). However, only 35% (n = 53) in the medium risk Ho group, 26% (n = 17) in the low risk Ho group and 32% (n = 65) in the low risk Travis group underwent surgery. Similarly 41% (n = 78) and 38% (n = 63) of patients in the high risk Travis and Ho groups respectively were treated with ciclosporin, whereas only 34% (n = 51) were treated in the medium risk Ho group, 25% (n = 16) in the low risk Ho group and 27% (n = 55) in the low risk Travis group. Resistance to ciclosporin correlated with increasing risk stratification, although this failed to reach statistical significance (Figure 1). However, the use of anti-TNF and trial drugs was the same across all three groups, although like ciclosporin, the tendency to anti-TNF resistance also increased with increasing risk group.

Figure 1. Colectomy by era.


The Travis and Ho scores are equally able to identify patients who are at high risk of failing medical therapy and needing colectomy or second line medical therapy. The Ho score identifies an intermediate risk group which also has an intermediate response to second line therapy. This may be of value clinically in defining treatment decisions in some patients. Both scores are useful tools to aid clinical decision making but do not replace timely multidisciplinary care for these patients.

1. S C Truelove and L J Witts, (1955), Cortisone in Ulcerative Colitis, Br Med J, 1041–1048, 2(4947).

2. S P L Travis, J M Farrant, C Ricketts, D J Nolan, N M Mortensen, M G W Kettlewell, D P Jewell, (1996), Predicting outcome in severe ulcerative colitis, Gut, 905–910, 38(6).

3. G T Ho, C Mowat, C J R Goddard, J M Fennell, N B Shah, R J Prescott, J Satsangi, (2004), Predicting the outcome of severe ulcerative colitis: development of a novel risk score to aid early selection of patients for second-line medical therapy or surgery., Aliment Pharmacol Ther, 1079–87, 19(10).