Article Text

Colonic wall thickness and pancreatic enzymes in cystic fibrosis
  1. JOHN A DODGE, Honorary Professor in Child Health
  1. Singleton Hospital, Sketty
  2. Swansea SA2 8QA, UK

    Statistics from Altmetric.com

    Editor,—The paper by Ramsden and colleagues1 showed no evidence that colonic wall thickness differs in children with cystic fibrosis who take high or low strength preparations of pancreatic enzymes, or the product they used previously. They found that, on average, colonic wall thickness was greater in children with cystic fibrosis than in controls, which confirms the observations made by Oppenheimer and Esterley2 more than 20 years ago (many years before any of the present enzyme formulations existed).

    Despite these negative results, the authors seem to have accepted the hypothesis (which has little or no published evidence to support it) that there is a connection between colonic thickening, fibrosing colonopathy, and methacrylic acid copolymer Eudragit L30 D55. They state that this copolymer “in high doses has a toxic effect on the gut of animals”, but give no reference for this statement. They also say that no association has been found between fibrosing colonopathy and Creon 25000 (which does not contain Eudragit) in the United Kingdom epidemiological study of fibrosing colonopathy.3 I was the chairman of that study, but, for the following reasons, was unhappy with the investigation and its published presentation. Firstly, the selection and inclusion of cases of fibrosing colonopathy rested on the opinion of a single pathologist who was not blinded to the treatment, and who has been the main proponent of the copolymer hypothesis. Although there were two other pathologists available for consultation, they were never allowed to see all the material submitted to the study, which included verified and possible cases. Therefore, there may be other cases of fibrosing colonopathy which were excluded from epidemiological analysis. Secondly, the lack of association between Creon 25000 and fibrosing colonopathy in cases that were included was true only for the months leading up to surgery. The pattern of enzyme use up to 12 months before operation showed that 11 of the 14 affected children were on Creon. No information was collected on why they changed from Creon to other preparations, but in the single published case which studied this, it was found that while on Creon there was a lack of response to abdominal symptoms, which persisted when the brand of enzyme was changed.4 It was impossible in that case, as in most others, to identify exactly when the lesion of fibrosing colonopathy developed; nearly all cases were initially diagnosed with distal intestinal obstruction syndrome (DIOS) before the diagnosis of fibrosing colonopathy.

    A review of the diagnostic criteria for fibrosing colonopathy and treatment for the disease is clearly needed. Ramsdenet al have found no supporting evidence for the co-polymer hypothesis, but their discussion suggests that they continue to believe it.1 It is surprising that they made no reference to the authoritative American study which reports 29 patients with fibrosing colonopathy. This study confirms a strong correlation between fibrosing colonopathy and high daily doses of pancreatic enzymes, but found no link with the strength, coating, or manufacturer of the products.5 In my view, the epidemiological and pathological evidence should be reviewed by an independent team; unfortunately, the pathological material upon which the UK report was based cannot be found. Until an impartial investigation takes place, it is inappropriate to accept hypotheses which cannot be verified.

    Acknowledgments

    Conflict of interest: During the past few years, I have accepted consultation fees and hospitality from all of the companies marketing pancreatic enzymes in the United Kingdom—that is, Solvay, Cilag, and E Merck. At present, I am a consultant to Scandipharm (USA).

    References

    Dr Ramsden and colleagues comment:

    We are pleased that Professor Dodge accepts our finding that the children in our study, who had cystic fibrosis and who had received high strength pancreatic enzymes containing methacrylic acid copolymer Eudragit L30 D55 (MAC), had a similar colonic wall thickness to controls.1-1 However, this finding does not exonerate MAC as an important causative factor in the rare cystic fibrosis patient who develops fibrosing colonopathy. However, our results are reassuring because they do not support the suggestion that there are now many cystic fibrosis patients who have subclinical colonic damage, as a result of taking high strength pancreatic enzymes.1-2

    We do not accept Professor Dodge’s statement that there is “little or no published evidence” to support the link between fibrosing colonopathy and MAC. Published, individual case reports in which enzyme brands are identified, show that all patients with pathologically confirmed fibrosing colonopathy have received MAC coated products at some stage.1-3-1-11 There are no reports of patients with pathologically confirmed fibrosing colonopathy taking solely non-MAC coated products. A detailed analysis of the patterns of use of pancreatic enzymes in fibrosing colonopathy shows that the condition has occurred only in patients taking MAC coated enzymes for at least six months.1-10 The UK epidemiological study, of which Professor Dodge was a coauthor, found a significant dose related association with only MAC coated brands.1-11 We consider that the design of the North American study could not show brand difference, as the controls came from the same cystic fibrosis centre as the index case and, thus, risk factors that varied systematically between centres were missed.1-12 Experimental data show submucosal fibrosis in pigs after seven days of exposure to MAC,1-13 and fibrosing colonopathy-like changes after chronic exposure to the coating.1-14 Therefore, we cannot accept the contention that there is little evidence to support the role of MAC in fibrosing colonopathy.

    In the single published case mentioned by Professor Dodge, it was clearly shown that growth failure occurred when the child was switched to Nutrizym 22, but not while taking Creon 25 000.1-15

    There are obvious differences of opinion about the role of MAC in causing fibrosing colonopathy, and, despite the wealth of supporting evidence, we support the proposed review of the existing evidence by independent experts. This is particularly important for cystic fibrosis patients who continue to take MAC containing pancreatic enzyme preparations that are still available in the UK (Pancrease HL and Nutrizym GR), and which are widely used in the United States.

    References

    1. 1-1.
    2. 1-2.
    3. 1-3.
    4. 1-4.
    5. 1-5.
    6. 1-6.
    7. 1-7.
    8. 1-8.
    9. 1-9.
    10. 1-10.
    11. 1-11.
    12. 1-12.
    13. 1-13.
    14. 1-14.
    15. 1-15.
    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.