Control of Systemic and Local Inflammation with Transforming Growth Factor β Containing FormulasFrom the Department of Pediatric Gastroenterology, Chelsea and Westminster Hospital, London, United Kingdom Correspondence: Dr. J.M.E. Fell, Paediatric Gastroenterology, Chelsea and Westminster Hospital, 369 Fulham Rd., London SW10 9NH, UK. Electronic mail may be sent to j.fell{at}imperial.ac.uk.
Enteral nutrition therapy with liquid diet has been shown to be effective
in achieving clinical remission in intestinal Crohn's disease. The mechanism
of action of this therapy, however, is still poorly understood. As part of our
assessment of the action of 3 related polymeric enteral therapies, we have
used a variety of techniques to document the histological and cytokine
responses, in the mucosa and, systemically, to these treatments. The feeds
studied (AL110, Modulen IBD and ACD004 [Nestle, Vevey, Switzerland]) all have
casein as the protein source, are lactose free and are rich in transforming
growth factor β (TGF-β). They have all been shown to induce clinical
remission associated with mucosal healing. In the case of Modulen IBD, as well
as mucosal macroscopic and histological healing there was a fall in mucosal
proinflammatory cytokines: interleukin-1 mRNA in colonic and ileal,
interleukin-8 mRNA in the colon and interferon The primary objective of therapy in patients with Crohn's disease is to achieve clinical remission with minimal side effects. Because the inflammatory process is focused on the intestinal mucosa, clinical remission should ideally be associated with a reduction in mucosal inflammation and mucosal healing. Several studies have shown that enteral nutrition therapy, using a variety of different formulas (elemental, semi-elemental, and polymeric) as the sole source of nutrition, is effective in achieving remission in Crohn's disease.1,2 From meta-analysis, the efficacy appears to be less than that of corticosteroids, although pediatric trials seem to produce higher remission rates for enteral therapy than adult studies.3 It is in this pediatric group of patients that enteral therapy has been most widely used (and studied) in recent years, due in particular to the potentially deleterious effects on linear growth of the standard alternative therapy of corticosteroids.4-6 The effect of therapeutic interventions on the intestinal mucosa in Crohn's disease has only been studied in a few instances. Most notably, infliximab has been shown to result in mucosal healing and inducing clinical remission in both adults and children.7,8 The importance of mucosal response to therapy has been emphasized and indeed proposed as a more significant end point than simple clinical remission.9 As part of our assessment of the action of 3 related enteral therapies, we have used a variety of techniques to document the histologic and cytokine responses, in the mucosa and systemically, to these treatments.
The studies described below on the effects of enteral nutrition on the intestinal mucosa have all been with polymeric diets, administered for 6–8 weeks before return to normal diet. The initial formula used was AL110 (Nestle, Vevey, Switzerland). This is a casein-based, lactose-free formula that is rich in transforming growth factor β (TGF-β).10 Results from experiments with mouse models of intestinal inflammation have suggested that TGF-β plays a central role in the suppression of the mucosal Th1 inflammatory response.11 This raised the possibility that exogenous TGF-β could have a beneficial effect on the T helper 1 (Th1) dominant inflammatory process of Crohn's disease.12-14 The subsequent formulas studied, Modulen IBD and ACD004 (Nestle, Vevey, Switzerland), have been adapted from AL110.15,16 They are both still casein based and rich in TGF-β (>24 ppm in Modulen IBD). Instead of the relatively low calorie density of 0.7 kcal/mL for AL110, they both have a density of 1.0 kcal/mL, thus requiring smaller volumes to be taken. In the case of ACD004, the fat composition was also modified (increased n-3 polyunsaturated fatty acids).16
The initial study into the effect of enteral feeding on the intestinal mucosal was with AL110. Beattie et al10 studied seven children (12–17 years), predominantly with small bowel Crohn's disease. They were treated with AL110 as the sole source of nutrition for 8 weeks. At the end of that time, not only had all cases improved clinically (decrease in Lloyd Still index) and biochemically (decrease in sedimentation rate [ESR] and C-reactive protein [CRP]), but there had also been mucosal healing. Blinded analysis of mucosal biopsies from the terminal ileum showed improvement in 6 of 7 cases, with complete healing in 2 children. After these initial results, we proceeded with a more detailed study using Modulen IBD (8 weeks) in 29 children (8–17 years) with active Crohn's disease affecting both small and large bowel (7 small intestine only, 17 small and large intestine, 5 large intestine only), as defined by barium meal and follow through and the macroscopic appearances at ileocolonoscopy.15 Once again, there was mucosal healing in association with clinical improvement (remission in 79%). In the terminal ileum, there was histologic improvement in 13 of 20 children, and in the colon there was improvement in 12 of 25 children; complete healing in the terminal ileum and colon occurred in 8 and 2 cases, respectively. A similar result was also found after nutrition therapy with ACD004.16 In this latter study, the clinical improvement after enteral nutrition was also shown for the first time to be associated with an improvement in quality of life. On regression analysis, improvements in quality of life did not correlate with the degree of mucosal healing. At first sight this was a somewhat surprising observation, but on reflection it is an indication that the quality of life of children affected by Crohn's disease is influenced by numerous factors, not all directly related to the severity of mucosal inflammation at any one time point. The studies reported thus far have all lacked a comparative control group. However, there has been a recent report of a randomized controlled trial, comparing the effects of either enteral nutrition (Modulen IBD) or corticosteroids on 32 children with active Crohn's disease, which included an endoscopic assessment before and after treatment.17 The results showed that although the rate of clinical remission was similar between the 2 groups, the improvement in the mucosa was significantly greater in children receiving enteral nutrition. These results were consistent with previous observations in adult patients that reported endoscopic mucosal healing of just 13% in patients in remission after corticosteroids therapy, and poor correlation between clinical and endoscopic features.18,19
Several studies have shown that the clinical response to enteral nutrition is associated with a fall in markers of inflammation in the blood, with a fall in ESR and CRP.5,10,15,16 Indeed, the ESR has been incorporated into one of the scoring systems of disease activity, the pediatric Crohn's disease activity index (PCDAI).20 With Modulen IBD treatment, we have also demonstrated a fall in serum tumor necrosis factor  (TNF)
levels,15,21,
whereas with AL110 a fall in interleukin-6 (IL-6) has been
reported.22 The
fall in levels of cytokines which are otherwise significantly elevated in
active Crohn's
disease23 is
probably secondary to the mucosal response, but may well account for some of
the changes in clinical parameters after enteral therapy.
The clinical response to enteral nutrition, the fall in systemic inflammatory mediators and mucosal healing in response to nutrition therapy are associated with a fall in mucosal proinflammatory cytokines. The initial evidence for this was provided by Breese et al.24 They showed that treating 6 children with Crohn's with enteral nutrition using an elemental feed, Flexical (Maed Johnson, Uxbridge, UK) in 3 and a polymeric feed (AL110) in the others, resulted in a decrease in the number of mucosal IL-2 and interferon  (IFN) producing cells. These changes were
similar to those observed after treatment with corticosteroids or cyclosporin.
This observed fall in Th1 cytokines in response to enteral nutrition was
investigated in greater detail in our study into the effects of Modulen
IBD15 (clinical
response and mucosal healing described above). Mucosal IL-1β, IL-8,
IL-10, IFN, and TGF-β mRNA, was measured by a quantitative reverse
transciptase-polymerase chain reaction method in biopsies from terminal ileum
and colon, before and after treatment. The most striking observation was a
fall in IL-1β mRNA in both the colon and terminal ileum. There were also
changes in response to treatment in 3 of the other cytokines measured,
although these responses differed between the 2 disease sites assessed. In the
terminal ileum, there was a fall in IFN mRNA and a concurrent rise in
TGF-β mRNA. In the colon, on the other hand, there were no significant
changes in IFN or TGF-β mRNA, but there was a decrease in IL-8
mRNA. These data provided strong confirmatory evidence for the
anti-inflammatory effects of enteral nutrition on the inflamed mucosa of
children with Crohn's disease, and added further weight to the significance of
the observed mucosal healing in response to this therapy. The question as to whether these changes are influenced by the specific formulas used, and in particular whether the exogenous TGF-β present in the feedings used was influencing the clinical and/or mucosal outcomes, cannot be answered without comparative clinical studies. In general, the feeding formulation has not seemed to have a major influence on outcome. A meta-analysis of elemental vs polymeric feeding was unable to demonstrate a difference in efficacy.2 The only other study that has focused on intestinal mucosal inflammation and its response to enteral nutrition has shown results similar to our own. This study, which used an elemental formula (E028, Scientific Hospital supplies, Liverpool, UK) as the therapy, demonstrated a fall in luminal IL-1β in response to treatment, using a whole gut lavage method to retrieve luminal cytokines.25
Early trials with enteral nutrition frequently excluded cases with colonic disease in the belief that nutrition therapy was less effective for disease at that site. We have shown that treatment with polymeric diets can induce colonic mucosal healing and a fall in proinflammatory cytokines. Not only is there a fall in IL-1β, but also in IL-8, both of which are implicated in the inflammatory process of colonic Crohn's.26 Reviewing the clinical response to therapy of 62 children with disease at different sites, we have observed that remission rates after nutrition therapy were lower for disease in the colon alone (47%), in contrast with small bowel (90%) or ileocolonic disease (85%).27 Thus, although randomized trials have yet to make a comparison of the remission rates for enteral nutrition and corticosteroids by disease site, there may well be some validity in the reticence of some clinicians in using enteral nutrition for colonic disease. This raises the more general prospect of tailoring treatment approaches depending on the phenotype of the Crohn's disease in individual cases.
The fall in mucosal proinflammatory cytokines, and the documented mucosal healing in response to the polymeric formulas described here, indicate that these formulas are influencing the disease process. Thus, the clinical remission achieved is a result of a reduction in the inflammatory process, rather than a consequence of some other nutrition effect. The mechanism by which the mucosal inflammation is reduced in response to this treatment has been the cause of much speculation. A likely mechanism is the potential effect of enteral nutrition on the intestinal microflora.28 Other potential mechanisms have been explored. The lipid content of enteral therapies, for example, appears to influence its efficacy,29,30 but whether other components such as TGF-β have an impact on the relative efficacy of different formulas is at present not answerable. Studies into the clinical application of the enteral formulas AL110, Modulen IBD, and ACD004 reported here were funded in part by Nestec, Nestlé, Switzerland. Received for publication December 6, 2004. Accepted for publication February 1, 2005.
DiscussionDr Walker-Smith: If I could just comment on the 8 weeks, this was on the advice of a pediatric endocrinologist, Martin Savage, because either he would use 6 weeks and he thought that 8 weeks was the shortest time possible which could give auxological significant measurement in growth, so it was really on auxological endocrine grounds, we went for 8 weeks.Dr Powell-Tuck: I am going to bang on about this business of colonic because I think it is important. Thomas told us that he knows that it is triggered by the bacteria in the colon and the ileum and yet you are saying that the treatments seem to work less in the colon, and I have a question. Your data that shows it doesn't work is based on very small numbers and I was also very interested, very pleased that you showed that data of severity. The severe cases were responding but they weren't going into remission, and they are simply the same cases that the colonic disease is, the severe cases, they are responding but they are not quite going into remission. This is very important because it is a recurrent clinical message that keeps getting thrown back which is that we don't use this in colonic disease and if we want to encourage nutrition treatment in adult practice for example, if every time we try to do it and it can't be used in colonic disease we get a bit stuck. So that is why I am banging on about it, I think it is very important clinically. Do you really believe that colonic disease does not respond to Modulen or any other enteral treatment? Dr Fell: To answer the last question first, we have got good evidence that colonic disease responds to enteral nutrition. We can't say whether it is better than anything else, but it does respond and there is mucosal healing and cytokines go down. Unfortunately, the cytokine data numbers were too small to do a proper sub-group analysis in picking out those limited cases that had pure colonic disease, and what do we mean by colonic disease? There is isolated colonic disease where there is no small bowel involvement, and maybe that is truly a different phenotype (the truly UC type Crohn's) from the phenotype where there is some small bowel involvement somewhere, and that would appear to come out from the small numbers looking at different disease sites. So maybe colonic disease is different from Crohn's disease that also affects the colon. And unfortunately with the cytokine data, I can't tell you whether those cytokines respond in the colon. All colonic patients responded to enteral nutrition, so that would have included purely isolated colonic disease and ileo-colonic disease patients. And also we have included patients that have no macroscopic colonic disease, so for them the colon is getting better as well because we included all cases responding with regards to cytokine data, so even if there is no evidence of inflammation, logically there is still a response. Dr Marteau: Two short questions, the first one is what do you know about the pharmacokinetics of TGF-β, when we ingest it, is it absorbed, does it reach the end of the small bowel, does it reach the colon? So how does it work? And the second question, I will present tomorrow a retrospective study in adults using Modulen and it seems at least in this French study that a high CRP level was predictive of a higher chance of efficacy of Modulen. Was it the same in your study in children? Dr Fell: The high CRPs do all come down so it is not that the CRP responds and responds quickly within 2 weeks, and I never compared that with the clinical responders, and certainly with infliximab of course the high CRP patients are the ones who respond best, the ones with more inflammation rather than presumably the ones with more fibrosis. The bioavailability of TGF-β, it gets through the stomach, I think we know that, where then it gets through I don't think we know. Dr Schiffrin: This question is quite relevant and we are now finishing a study to see whether we can detect TGF-β in ileostomised patients after an administration of one-half a liter. So we are doing this, we are trying to understand if it is absorbed or it gets to the end of the ileum, it goes into the colon. These are questions that are truly relevant and we are trying to answer to them. Dr Triantafyllidis: You mentioned before that Modulen is rich in TGF. What actually means rich? And my second question is, when dealing with clinical response, is it good to measure anthropometric parameters to estimate the clinical response like skinfold thickness, BMI and so on? I didn't hear anything about that. And also you mentioned that there are alterations in the serum levels of some cytokines. What about the measurement of cytokines in the tissue? I suppose that it is better and clinically more relevant to measure cytokines in the tissue. Would you like to comment on these 3 questions? What means rich? Dr Fell: Rich in TGF-β. My understanding whether TGF-β gets into the feeding is if it is present anyway and the issue with AL110, and also with Modulen, is that in the manufacturing process the TGF-β was not destroyed. I have some data that looks at the TGF-β levels in different products and different infant formulas that show that if you process the constituents of milk differently, the naturally occurring TGF-β may or may not be destroyed. So it is by chance really that in AL110 the processing didn't kill off the TGF-β, it was still in active form. Thus, they could move directly to Modulen to make absolutely sure that it is not degraded by the manufacturing process. But it is not added, it is there naturally and it has not been degraded by the processing. Dr Huggett: Just to add to what John says which is perfectly correct, TGF-β2 is a natural component of human breast milk and cow's milk. The AL110, which was the predecessor to Modulen IBD, was the product where we found the greatest amounts of TGF-β2 even compared with other milk derived products. So as John says during the manufacture of Modulen, the TGF-β is protected through other proteins in the product, which results in a much greater amount of TGF-β in Modulen than in other products. It is resistant to digestive enzymes in that form and stomach acid as well. Dr Fell: The question 2, which was the anthropometric response, we looked at weight gain and there was significant weight gain not surprisingly with the nutrition treatment. The weight gain was at the order of 2–3 kilos in the 2 months period, so not enormous. Since in clinical practice it has become clear that the patients actually gain weight when they come off the feeding. During the food reintroduction phase when patients are feeling well and being told that they can eat anything they like rather than having food prescribed, that is when there is the anthropometric response. It is, once the inflammatory process has gone away and you are allowed to eat normally, an artifact of the way we are treating these patients which is allowing them to eat ad libitum. They don't respond nutritionally as much as one would hope and it is quite difficult to give them their extra 30% of calories to achieve catch up growth. I would prescribe in the expected way: +10%–20%, which is quite hard to get in if you are anorexic from a systemic inflammation. Then towards the end of treatment, patients start becoming hungry and that is where the weight gain kicks off. The last question was the cytokine response. We have looked at the mucosal biopsies; we have got a lot of data on the cytokine response there. What is happening elsewhere in the tissues we don't know. The only data we had was on the TNF and unfortunately there is no data on IL-6. Dr Walker-Smith: A young student went to university and he kept on supplementing himself with enteral feeding. He ended up quite obese because he was eating as well. He really felt that it was a good thing to continue with his Modulen and made himself obese. Dr Paerregaard: You have beautifully demonstrated that Modulen is efficacious in treating active Crohn's disease and I think we should be grateful to the Nestlé company for developing a specific Crohn's disease treatment formula, but Modulen has not been compared in a randomized fashion to other formulas that may be used to treat active Crohn's disease and a key issue is the addition of TGF-β. There seems to be data coming up to verify that TGF-β stays intact in the gut, when you put it in the mouth and eat it. But in order to convince me to use Modulen, I need to know whether TGF-β is also able to modify intestinal inflammation. Do you have any data on that issue and could some of the animal models, the mouse colitis models, could they be used to verify this putative action of TGF-β? Dr Fell: I think Eduardo is going to be talking a bit later on about animal models and the putative effect of TGF-β. You are right, there has been no comparative study. I think that the difficulty we have with enteral nutrition is almost that the benefit of further studies, I am not sure, is actually there; to do a study that would show a small benefit between the different feeds you would need large numbers firstly, and secondly the answers that you get from the questions is relatively unimportant. The feeding works and if it doesn't work you move on to something else is the practical way of dealing with this treatment. Whether we really want to know whether one feed is largely better than another, I don't know whether that is worth knowing if the remission rates that come out of these studies are all of the order of 80%. And if you do a study you are never going to illustrate properly, you cannot very easily do studies of nutrition vs something else blind. You may be able to blind different feeds but if you are going to have a different feed it is very hard just to modify 1 component. As soon as you are taking one thing out, other things are going to happen, if you just try to take out TGF-β, that is 1 question. Maybe it is worth knowing that but it is an awful lot of work just to answer 1 very small question on nutrition when maybe you actually should be modifying fats or maybe you should modify the amino acid profile. The difficulty we have had is when they have done comparative studies, the data that comes out are often not very interesting. To be remotivated to look at whether one feed is really better than another, what we should do is to actually compare taste. That is more important, and/or a set of studies that look at the quality of life as the major importance. Maybe that is a better way forward. I think the treatment approach is more important than the subtle differences in different feeds. We have to get something that works. That is probably all we need to know and then the more important question is where does it fit in since it is harmless and safe and making sure that it is then used in the right way before escalating too quickly to toxic treatments. Dr Walker-Smith: The n-3 formula in fact didn't taste as well. That formula came to existence for a sort of academic reason of putting more n-3 following the meeting in Boston. It was felt necessary because of the change of formula you should go through the same protocol in changing mucosal healing. But one of the problems with the n-3 formula in fact related to taste. Do you have any data on that, taste problems in the children? Dr Fell: When we designed that, the initial studies, its efficacy was just shown, it worked like all other enteral nutrition and we didn't specifically focus on what the patients felt about it. We looked at quality of life and the quality of life improved as the disease improved. Unfortunately we didn't really focus on the patients' experience. With the impact of quality of life scores publications, 2 now and perhaps 3 available, at least we can now start looking more holistically at what the patients experience with treatment. With Modulen this is more of a subjectivity of taste because we have mentioned the word fish oil. Dr Walker-Smith: I remember individual children who had been in a trial and later they went back onto Modulen and they said it was much better, Modulen is much better tasting, that is very subjective and anecdotal. Dr Fell: You can't even compare taste of healthy people because when you are ill the taste is more disturbed anyway. Dr Damiao: In your study, did you use Modulen in refractory patients or corticoid dependent patients, or did you use the diet as a primary measure? Dr Fell: All patients had active Crohn's, so they were all on no treatment when they went into the study apart from 1 on a very low dose of steroids. Less than one-third were not new cases, so most were patients were presenting with their very first disease activity episode of Crohn's, so completely newly diagnosed. We are dealing with the patients that are most likely to have an inflammatory process going on, rather than a chronic fibrotic process that often go to pediatric studies. We are usually dealing with patients at the beginning of their story and some of the difference in remission rates may well reflected this. We are dealing more with inflammatory rather than fibrotic chronic cases, and infliximab probably would work better in children and anecdotally it does work better in children because they are more inflammatory rather than chronically inflamed. So most of these cases were new cases. Dr Barabino: I didn't understand the percentage of patients in your study that got endoscopic remission. Is it 85%? Dr Fell: The endoscopic remission was, it is different in the small bowel and in the large bowel, it is of the order of 50%–60% in the 2 disease sites, whereas histologic remission was actually strongly determined in the colon. Dr Lionetti: With regard to the colon involvement, I think personal experiences are important also. We can have a confrontation here about our personal experience even if data is not published sometimes. So my experience, in children who do not respond to these kind of treatments, I usually treat them, those with extensive colonic involvement (left colon involvement) and we check the efficacy in 2 weeks time. Usually we have a partial response, we don't induce remission. What I am doing now, I add steroids to that therapy. So I keep them on exclusive liquid diet and I add steroids, and usually I can induce remission this way. And I have some data also about the mucosal healing in 1 or 2 of these patients. I would like to have your opinion about that, if you do the same or if you stop enteral nutrition and you change way and you give steroids to them. And then another observation: I agree with you that sometimes we see the weight gain after the end of exclusive course. We have to think about that because the mechanism is independent of the nutrition status. And when they go back to the normal diet they gain weight because they are in remission, but we sometimes keep them drinking something like 40% of the daily caloric intake as Modulen or a liquid diet, but they can have a normal diet. Dr Fell: The first question which is the decision time at 2 weeks, our current clinical practice is that nearly all patients with active Crohn's go to enteral nutrition certainly the first time around, and by 2 weeks 1 or 2 things will happen, their life is better and therefore they are complying well with the treatment and therefore can continue with the 8 week course, or they haven't responded and they are not complying. The 2 go together because we very rarely use nasogastric tubes, so the poor compliance is in itself probably a marker of lack of response. Now what do you then do at the 2-week decision time? By that time the patient may well have completely become averse to that treatment and it is actually quite difficult to continue with a treatment that hasn't worked for 2 weeks. So we would add corticosteroids as well. Whether they stay on Modulen or not is up to them but most of the children actually will not and would require escalating to a nasogastric tube feeding. So that is individual; some stay on Modulen enteral nutrition, some don't. What do I do at the end of the treatment? At 8 weeks we go through a laborious instruction process, which is trying to be systematic in adding foods sequentially and very often these patients react to individual foods. Anecdotally these children respond to individual feeds and see what they are reacting to, and then whether they stay on supplementation or not is the patient's choice, and when the children don't want to and the parents do, then it is up to the family to decide what they are going to do. Dr Heuschkel: Our practice is very similar to John's. My own view is that if you are not achieving a good remission within 2 weeks then to put a child through another 6 weeks of exclusive nutrition in addition to steroids seems a little over the top. So we generally carry on with some nutritional supplement but let the child go back onto a normal diet, and usually we factor in the early use of immunosuppressants knowing that they already failed to respond to what we know to be a usually successful response to enteral nutrition. Dr Sanderson: There are the patients who are drinking the Modulen but not sufficiently, they are drifting into remission but they are not the ones for which we can say it doesn't work. You put the nasogastric tube down those and then they go into remission very well. It is not always the ones who don't drink that are actually not responding. Dr Fell: The initial interview is to assess what is the status of the patient you are dealing with. There are some patients that you know will not take this feed no matter what you say about it, and for that patient group, there is the occasional patient in which using the nasogastric tube is appropriate. That is actually unusual and I am always conscious that patients that say they are trying hard and really can't take this feed and have a lot of risk factors, they have bad disease. I suspect that they will never respond or you are going to get a partial response. I personally bale out at 2 weeks because what I don't want to do is go through 8 weeks of treatment and then start steroids. This is the part where enteral nutrition gets some sense, because there are truly nonresponders. If you look at the data, there is only a 70% response rate, so there are 30% that are not going to respond, even in the trial circumstance. In trials you try terribly hard to make all your patients take everything and they get a lot of counseling and support, etc, so they comply very well. In clinical trials there is 20%–30% who don't respond. So in clinical practice there are going to be more nonresponders. The compliance is more difficult because they are not part of the trial. Trials are very good for patients, as the ones in the trial do better than someone outside of the trial. But we have to accept that there are some patients that truly don't respond and they may have different diseases or whatever, so I take the personal view that it is appropriate to escalate treatment quickly rather than struggling on for weeks and weeks. Dr Walker-Smith: I would like to comment on the empiric-refeeding regime. We have the data from 1987 from Cambridge from John Hunter who is an adult gastroenterologist and his refeeding protocol was very empirical, but it does seem to actually work in general, doesn't it? When you are introducing food and then when you relapse with one food, that same food, you can actually tolerate it again later on–that is the case isn't it? Dr Fell: It is not a true allergy response, you are reacting to a food and you probably would react to food in general. Where it is helpful is in taking families along with it because families are very often going to be using the websites and they are asking the question as to the importance of food allergy in the course of Crohn's. I found it helpful to use the instruction phase for them to answer that question for themselves to find the answer is negative. They don't find the food is causing it and therefore, it is not a food allergy and it stops children being put on milk-free diet. It stops children going onto wheat-free diets for no good reason. It stops children going onto specific carbohydrate-free diets, and all the potential consequences of being over enthusiastic with interventions. Dr Wenzl: I learned the concept and we are basically following that study protocol. The question now is, what is your correlation between clinical and mucosal healing? That is in your study calculation, what is the incidence of children in clinical remission that did not have mucosal healing, and what do you do with them in real life after the course of treatment? Dr Fell: There will be patients in complete remission that do not have mucosal healing, and the way the study was done, we biopsied the patients at 2 weeks. The true nonresponders were not biopsied, they were discarded. Of the patients that were rebiopsied, nearly all of those are in complete remission and so within that group there is only going to be mucosal healing, in the terminal ileum 50% histologically and in the colon it is going to be in the order of 30%, but there is mucosal improvement in more because of the infliximab. Dr Wenzl: So it does not change your treatment strategy, you do not necessarily want to achieve mucosal healing although that should be the goal? Dr Fell: Yes, it should be the goal and I don't know, and no one has looked yet, whether the healing of mucosa is the one that relapses later. If you have got a nice mucosal sheet there are going to be less bugs getting into the wrong place and therefore your remission should be more established. Dr Walker-Smith: In case there is a mucosal improvement, this is a way of knowing whether you went on longer to complete healing, so that is good news. Really in circumstances where there is no change, mucosal healing is normal. But when there is no change or worsening, and I think one of the advantages of a second endoscopy is when there is mucosal worsening, then you may add in steroids or azathioprine. Would that be the case if you have a child and do repeat endoscopy, if it is worse then you treat with drugs? Dr Fell: Another reason to escalate. Dr Barabino: I would like to comment on partial responder to nutrition feedings. I think that a very important point is whether or not we put azathioprine straight away with enteral feeding. I think that we can have good improvement with azathioprine if we put azathioprine straight away. And we see that when we do endoscopy after enteral feeding sometimes we don't have endoscopic remission but we can have endoscopic remission with azathioprine 6 months or 1 year later. I would like to know your opinion about this. Dr Fell: Clearly if you have more therapy and more aggressive therapy, you have less disease. The question is, I suppose, whether you are an optimist or a pessimist. It would mean that you have overtreated 40% or 50% of your cases that wouldn't have had a relapse within the next year. So you are going to be treating everyone early with azathioprine, it is 1 treatment approach, you are going to be treating cases that would not have relapsed the next year, but also you have a prevented relapse in the other percentage that would have relapsed. It is a philosophical question, should you be treating everyone with high doses, more aggressive treatments? If you are going to do that then you have to have some way of coming off the azathioprine, and you ask yourself the question a year later or 2 years later: is it still necessary because you would be overtreating the same percentage with a potentially toxic drug that they would not otherwise come on to. And as pediatricians we evade the question by patients becoming 16 and therefore we lose them, but at some stage they are going to have to be asked that question, do we need this medicine? If you are starting azathioprine because of significant failures, because you have major relapses and are on unacceptably high-dose steroids or whatever you have in your second course with enteral nutrition and you are relapsing too much, then that is much firmer evidence to stay on a potential toxic drug for a long time. That is my personal pessimistic view of toxicity of azathioprine. Dr Griffiths: We haven't talked very much about the rate of relapse after a successful treatment with enteral nutrition. There is some data in adults, there is a study by Rigaud, which was a comparative enteral nutrition trial and in their paper they have a figure showing a 70% relapse at 12 months. So I was wondering if with your very systematic approach to treating all newly diagnosed children, whether you had similar follow-up data a year later as to rate of relapse, assuming you are not using concomitant azathioprine. I think you alluded to it. Dr Fell: Our relapse rate within 12 months in the Modulen study was <50% but actually there are some that have dropped out already because there have been treatment failures. So the initial of the whole cohort, it is <50% of disease free at a year. Dr Griffiths: So that is lower than in the adult studies which presumably have later disease and is a 1-year longer sort of systematic follow-up? Dr Fell: We have got a few patients for longer than that but we haven't got large enough cohorts to put a percentage on. Dr Sanderson: What about the n-3 study, was it a similar figure for relapse rate? Dr Fell: The n-3 study relapse rate was a bit higher, I think it was 60% relapse rate. Dr Sanderson: We did do the follow-up. We followed them for 5 years and did the relapse rates for both steroid group and the nonsteroid group, but because it was so much later it was never published, and there was no difference in the time in which they had the remission, but both groups did very badly. In the steroid group, most of them relapsed within 5 years. Dr Huggett: As you mentioned before, Modulen was developed from an infant formula and of course we are very limited in what we can put into an infant formula, as it is very regulated. How much more successful do you believe Modulen would be if it were available, for example, in different flavors or in different formats? Dr Fell: Preference and taste. What we can say from the limited data we have is that all trials have been done with the pure product. So if you are going to be quoting the remission rates, the remission rates is 80% with an unflavored product, is it then appropriate to put something in to make it taste different? In practice I personally allow the patients to have a flavor, and if you only allow 1 flavor you are going to get very bored with just having a strawberry flavor for 6 weeks or 8 weeks. We also know that if you don't stick to the feeding regimen enough, from the unpublished data from the n-3 study, it does look as if you add in too many extra things, in other words start getting down so that 30% of your requirements are coming from other foods then maybe enteral nutrition is less effective, so our approach is dogmatic and to allow flavor.
Journal of Parenteral and Enteral Nutrition, Vol. 29, No. 4 suppl,
S126-S133 (2005) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Top
CHARACTERISTICS OF THE FORMULAS...
CLINICAL RESPONSE AND MUCOSAL...
