Search tips
Search criteria 


Logo of gutGutView this articleSubmit a manuscriptReceive email alertsContact usBMJ
Gut. 2007 October; 56(10): 1452.
PMCID: PMC2000266


Living with Lille

[filled triangle] Louvet A, Naveau S, Abdelnour M, et al. The Lille model: a new tool for therapeutic strategy in patients with severe alcoholic hepatitis treated with steroids. Hepatology 2007;45:1348–54.

Debate as to the efficacy of corticosteroid treatment in alcoholic hepatitis continues to polarise opinions and generate more heat than light. One of the real issues for steroid treatment in alcoholic hepatitis is that in an individual patient, 40 mg of prednisolone for 28 days may do more harm than good. Therefore, early identification of those who are unlikely to respond is obviously desirable.

Louvet et al aimed to develop a model to identify steroid non‐responders. All the patients included in the study had histological confirmation of alcoholic hepatitis and clinically severe disease with a Maddrey discriminant function of [gt-or-equal, slanted]32 at entry. The Lille model was developed using data from 320 historical patients treated with steroids, then the performance of the model was validated using a prospective cohort of 118 patients. The model combining both pre‐treatment and on‐treatment variables (age, renal insufficiency, albumin, prothrombin time, bilirubin and evolution of bilirubin on day 7 of steroid treatment) was highly predictive of death at 6 months (p<0.001) with an area under the receiver operating characteristic curve of 0.89. The Lille model was superior to Child‐Pugh, MELD, Glasgow and Maddrey scores. Those who were above a cut‐off value of 0.45 on this model (40% of patients) had a markedly lower 6‐month survival (25%) compared with the rest (85%) (p<0.0001).

Lille is another addition to an ever growing list. Nevertheless, the model's credibility derives from the highly characterised patient population that is used and the track record of the Lille group. Using their model, the authors argue, will allow steroid non‐responders to be identified early and considered for other (novel) treatments.

MAP: not the answer for Crohn's disease

[filled triangle] Selby W, Pavli P, Crotty B, et al. Two‐year combination antibiotic therapy with clarithromycin, rifabutin, and clofazimine for Crohn's disease. Gastroenterol 2007;132:2313–9.

Since the isolation of Mycobacterium avium subspecies paratuberculosis (MAP) from Crohn's tissues in 1984, there has been ongoing speculation that MAP may be a specific bacterial cause of Crohn's disease. Conventional anti‐tuberculosis treatments probably do not eradicate MAP and have been ineffective in prolonged treatment for Crohn's. If prolonged treatment with an antibiotic combination that does eradicate MAP resulted in long‐term remission of Crohn's disease, then this would provide the best evidence that MAP is causative.

This multicentre Australian trial seemed best placed to answer this question and results have been eagerly awaited. A group of 213 patients with active Crohn's disease were randomised to receive clarithromycin 750 mg, rifabutin 450 mg and clofazimine 50 mg per day plus a tapering course of prednisolone (40 mg reducing to 0 over 16 weeks) or matching placebo plus prednisolone. The antibiotics were increased gradually up to the full doses over 4 weeks to minimise adverse events. Patients in remission at week 16 continued into the maintenance phase of the study. Significantly more patients in the antibiotic arm entered the maintenance phase (67/102 (66%) vs 55/111 (50%), p = 0.02). Of the 122 patients entering the maintenance phase, 39% on antibiotics had at least one relapse between week 16 and 52 (vs 56% on placebo, p = 0.054) and at week 104, the figures were 26% and 43%, respectively (p = 0.14). During the third year, 59% on antibiotics had a relapse, vs 50% on placebo (p = 0.54). Only 32 patients completed this study. The benefits in the antibiotics group during induction of remission may be a non‐specific antibacterial effect.

The antibiotics used would have good intracellular penetration and concomitant corticosteroids would further optimise antibiotic effect. This study, with its 3‐year follow‐up, shows that antibiotics that eliminate MAP have no effect on the course of Crohn's disease and thus this bacteria is unlikely to be causative.

When more does seem to be better … !

[filled triangle] Chowdhury MM, Dagash H, Pierro A. A systematic review of the impact of volume of surgery and specialization on patient outcome. Brit J Surgery 2007;94:145–161.

Does specialisation and/or high volume surgery improve outcome? For nearly a century, this most contentious issue has been plagued by rhetoric and opinion rather than evidence‐based practice. The implications of a positive answer could and should radically change operations such as ileoanal pouch, stopping the performance of low volume in multiple non‐specialist centres and encouraging regional, high volume specialisation.

The authors of this study systematically reviewed articles examining the effects of one or more of three variables (hospital volume of surgery, surgeon volume and specialisation) on outcome measured by length of hospital stay, mortality and complication rate. Their search identified 1075 articles published between 1957 and 2002 relevant to the study, of which 163 (9 904 850 patients) fulfilled their entry criteria. These 163 articles examined 42 different surgical procedures, spanning 13 surgical specialties. High‐volume hospitals had significantly better outcomes in 74.2% of studies but this effect was limited in prospective studies. High‐volume surgeons had significantly better outcomes in 74% of studies and specialist surgeons in 91% of studies.

Whether current clinical practice should be changed on the basis of this article is a difficult question. However, the results of the recent UK national inflammatory bowel diease audit suggest that in the UK, many centres are performing only a few ileoanal pouches per year, with variable outcomes. Perhaps it is time, therefore, to consider centralising these sorts of complex procedures in specialist high‐volume centres.

Doing it with your eyes closed …

[filled triangle] Chen SAC, Rex DK. Endoscopist can be more powerful than age and male gender in predicting adenoma detection at colonoscopy. Am J Gastroenterol 2007;102:862–3.

It's all very “new” Labour and a bit depressing for the average gastroenterologist. Everything is monitored and scrutinised … Tony Blair makes me worry about my emissions (out goes the BMW M5), the wife's talking about buying an electric car, next door have erected a 20 foot wind turbine from B&Q due to a community carbon footprint drive and Charles Kennedy is now arrested for having a crafty one on the train. Well, there's more to come!

Doug Rex has again published a corker. In this study he assessed the adenoma detection rates of nine endoscopists who did >10 000 procedures and the effects of endoscopist on adenoma detection rate compared with the well‐established predictors of adenoma, namely age and male gender. The results are dramatic. Among patients >50 years, the range of detection of at least one adenoma per colonoscopy by the nine endoscopists was 15.5–41.1%, at least two adenomas, 4.9–20%, and three adenomas, 0.8–10%. Clearly, the differences between endoscopists were significant and it now appears that who performs the colonoscopy is a more important predictor than age and gender in predicting adenoma rates!

Articles from Gut are provided here courtesy of BMJ Publishing Group