Gastroenterology
Volume 136, Issue 1 , Pages 32-34, January 2009

Celiac Disease and Its Complications: A Time Traveller's Perspective

  • Joe West

      Affiliations

    • Corresponding Author InformationAddress requests for reprints to: Joe West, Division of Epidemiology and Public Health, University of Nottingham Medical School, Queen's Medical Centre, Nottingham, NG7 2UH, UK

Division of Epidemiology and Public Health, University of Nottingham Medical School, Queen's Medical Centre, Nottingham, United Kingdom

published online 01 December 2008.

Article Outline

 

See “Presentation and long-term follow-up of refractory celiac disease: comparison of type I with type II” by Malamut G, Afchain P, Verkarre V, et al, on page 81; and “Increased risk for non-Hodgkin lymphoma in individuals with celiac disease and a potential familial association,” by Gao Y, Kristinsson SY, Goldin, et al, on page 91.

In 1888, a short story “The Chronic Argonauts” by H.G. Wells1 was published by the Royal College of Science in which his concept of a time machine was first introduced. Writing in the preface to the 1931 edition of his now famous novel “The Time Machine” Wells says that “it is obviously a work of an inexperienced writer, but certain originalities in it saved it from extinction.” Samuel Gee, although far from inexperienced, in the same year wrote about another original concept that has stood the test of time.2 If Gee had had at his disposal Wells's Time Machine and decided to travel a mere 120 years into the future rather than the 802,000 the Time Traveller did, what would be his view of the current understanding of the “Celiac affection” he so memorably described?

Not unlike the experience of H.G. Wells's character, when arriving in 2008 Gee would see a landscape in one sense markedly different from his own observations and yet in other ways palpably familiar. Celiac disease as we now know it is, rather than being considered a rare and incurable disease, both common (at least in terms of its serologic markers in the general population3) and readily treatable with, as predicted by Gee himself, a dietary intervention. Yet, as in the 19th century, serious and life-threatening complications of celiac disease persist today. Although for the first 50 years or so of the history of celiac disease enteropathy-associated (or type) T-cell lymphoma (EATL) and refractory celiac disease remained unclassified, during the latter part of the 20th century our recognition and to some extent understanding of these conditions improved considerably. However, given the length of time that we have had since Gee's original description, there remains a surprising amount that is unknown.

Back to Article Outline

In Gastroenterology This Month 

From previous, work we know that there is an increased risk of lymphoproliferative malignancy and in particular EATL among people with celiac disease; however, the estimates of the magnitude of the associations and how specific they are to the different types of lymphoma remain imprecise.4 Refractory celiac disease by contrast remains a recondite clinical entity. In addition, how important the risks are of these conditions to people with celiac disease and the general population at large are challenging to discern from previous work. In this area, our understanding can be improved through epidemiology and 2 contrasting studies, in terms of their methodology, are reported in Gastroenterology this month.

Back to Article Outline

Celiac Disease and Lymphoproliferative Malignancy 

Gao et al5 describe a large, case-control study of non-Hodgkin lymphoma (NHL), Hodgkin lymphoma (HL), and chronic lymphocytic leukemia, in which they used a series of linked, population-based databases to identify cases of cancer matched to controls and determine which of these had celiac disease diagnosed >1 year before the cancer diagnosis. Essentially, the findings mirror those of similar case-control studies in terms of the overall risk estimates with respect to NHL6, 7 show an association midway between those observed before in the Swedish cohort studies for HL8, 9 and confirm no association with chronic lymphocytic leukemia7 (most of the relevant studies are listed in Table 1 of their paper). These findings are important as because this study is approximately 10 times larger than the nearest of its kind we can be confident that the findings are not due to chance. This is the great advantage of the combination of a database study with a case-control design for an exposure such as celiac disease. Because of its size, the authors of this study had the ability to explore the variation in risk over time. They noticed that the increased risk for NHL associated with celiac disease is much lower the later in the study NHL was diagnosed. They also report a curious observation that people with a sibling who had been diagnosed with celiac disease were at a 2-fold increased risk of NHL.

In general, the risks of NHL reported in the literature4 and in this study are likely to remain overestimates owing to either bias or confounding. In this instance, as in most other studies, bias may play its part in the ascertainment of celiac disease among cases of NHL (or vice versa), through visiting a physician and the attendant investigations that are carried out. Although a 1-year lag period in this particular study will have tempered the effect, it will not have removed it. With respect to confounding, surprisingly few variables have evaluated systematically either in this study or others primarily owing to the few exposed cases in each study.

The observation of an increased risk for NHL among siblings of people with celiac disease should be viewed cautiously before being considered causal. No more sophisticated attempt other than simple logistic regression was made in the analysis, and this will have overestimated the precision of their estimate by having not taken into account any effect of “clustering.”10 The finding therefore could well have transpired by chance; in addition some ascertainment bias, as described, may have occurred. However, if it is a true association, then it is at least equally likely that shared environmental factors are the explanation than any based on genetics, as the authors speculate.11, 12

Back to Article Outline

Refractory Celiac Disease 

Malamut et al's report13 of 60 cases of refractory celiac disease differs markedly with the Swedish database study just discussed. It describes in detail the various characteristics at diagnosis and the consequences that occur over a reasonable period of follow-up in these patients, observed in 6 centers in France. Essentially, the findings are similar to that of Al-Toma et al14 with respect to the characteristics at the time of diagnosis, including age, gender, and human leukocyte antigen status. The consequences of acquiring a diagnosis of refractory celiac disease I or II are also not greatly different between those patients diagnosed in a single center in Holland14 and those reported here from a wider referral population in France. Although refractory celiac disease 1 (by the author's definition) follows a relatively benign course with survival characteristics not vastly different from the majority of people with celiac disease,15, 16 refractory celiac disease II confers a substantial shortening of life expectancy. It is also clear from this report that the treatment of these individuals is far from satisfactory. Although corticosteroids may improve the symptoms, few patients are able then to stop taking them without relapse. Steroid-sparing immunosuppressant therapy does not seem to work and there was no evidence of preventing the onset of lymphoma, unsurprisingly.

However, before accepting of the accuracy of all of these conclusions, one must take into account the small numbers of people identified and the manner in which the case series was assembled. The authors are reliant upon 2 conditions. First, that the precision of their definition of refractory celiac disease is excellent, and second that the people in their study represent an unselected group. The latter point seems unlikely to be true because they are in general physicians with an interest in celiac disease and a certain amount of referral bias is therefore unavoidable. The authors focus a fair amount on the 2 cases of lymphoma that developed in the 14 people labeled refractory celiac disease I and on this basis propose twice yearly screening with either computed tomography or magnetic resonance imaging. Although they are only suggesting this policy for a minority of celiac disease patients, caution should be exercised as when dealing with numbers as small as this; any error or misclassification in the original diagnosis or definition (of refractory celiac disease) could lead to unwarranted x-ray exposure with its attendant risks of malignancy.17 In addition, there is no guarantee of implementing a successful treatment should an EATL be identified.

Unfortunately, the study by Malamut et al13 does not give us a clear estimate of the occurrence of refractory celiac disease. The available data from Derby, United Kingdom,18 comprising an unselected, population-based cohort allows a useful comparison with the current study in terms of the occurrence of refractory celiac disease among people with celiac disease. From 1978 to 2005, there were 713 people identified in Derby with celiac disease who reported strictly taking a gluten-free diet, the vast majority of whom were incident (personal communication, Geoff Holmes). Of these only 5 (0.7%) did not respond to the diet and remained clinically markedly unwell with evidence of ulcerative jejunoileitis. These 5 individuals represent the only people with celiac disease in the Derby cohort who could be considered as having refractory celiac disease II. Of these 5, only 1 developed lymphoma—an EATL.

Back to Article Outline

Summary 

From the perspective of people with celiac disease, the case-control study confirms that there is an association between clinically diagnosed celiac disease and an increased risk of lymphoproliferative malignancy and gives a precise estimate of the contemporary risk. Taken with the other available estimates the relative risk of NHL is around 3- to 4-fold compared with the general population. If we instead focus on absolute risks as measured by incidence rates the available estimates from Derby, United Kingdom, showed that for people with celiac disease the absolute risk of NHL was 1 in 1421 person-years and for small bowel lymphoma 1 in 5684 person-years.18 From a wider, population-based study also from the United Kingdom, the risk beyond 1 year after a diagnosis of celiac disease of any lymphoproliferative malignancy was approximately 1 in 1200 person-years.19

From the perspective of the general population and the possibility of preventing cases of NHL we can use the figures from the current case-control study to calculate the population attributable fraction. To do this we must assume that the relationship between NHL and celiac disease is causal. From the overall estimate from the current study the population attributable fraction is about 0.1%; that is, if we were to remove celiac disease as a risk factor for NHL, then we would prevent 0.1% of NHL occurrence. One could argue that “undetected celiac disease” if diagnosed and treated might prevent a greater proportion of NHL. However, evidence from the 2 case-control studies that included screening for undetected celiac disease suggest that the association with this condition is not as strong as in clinically diagnosed disease.6, 20 In addition, there is no evidence that imposing a gluten-free diet on such “screen-detected” individuals prevents malignant complications.

From a general population perspective, the incidence of EATL in the Netherlands is reported to be approximately 1 per million person-years21 and is of a similar order of magnitude in Scotland.22 Of the 43 people with refractory celiac disease II in the current study,13 16 developed lymphoma. Thus, we can crudely estimate that refractory celiac disease II is about 3 times more common than EATL itself—perhaps around 0.3 per 100,000 population. From the perspective of people with celiac disease, the current study highlights the severity of the clinical course in refractory celiac disease while not giving us an absolute risk for the majority of people with celiac disease. Nevertheless, taking the data from Derby presented into account, we can be fairly confident that this serious complication of celiac disease is unusual.

In summary, Samuel Gee's likely understanding of the state of celiac disease were he to visit us in 2008 is as follows. Celiac disease is now commonly found, but severe and fatal outcomes from it remain rare. Lymphoproliferative malignancy is such an infrequent complication of celiac disease that the majority of gastroenterologists may never see it among the population of celiacs they diagnose and follow-up. Although dietary treatment, which works for almost all, is now available for those unlucky few who do not respond, aside from long-term corticosteroids and the associated side effects thereof, no good alternative yet exists. The precise size of this group of patients, and the magnitude of the risk to them, is still being elucidated, and we can only hope that a visit a further 120 years in the future will reveal progress in these respects.

Back to Article Outline

 

I thank Tim Card and Geoff Holmes for their comments on the manuscript. I am particularly grateful to Geoff Holmes for the use of his data presented in the text on his experience of refractory celiac disease, which are otherwise unpublished.

Back to Article Outline

References 

  1. Wells HG. The chronic Argonauts. The Science Schools Journal. 1888;Apr–Jun
  2. Gee S. On the coeliac affection. St Bartholomew's Hospital Reports. 1888;24:17–20
  3. Lohi S, Mustalahti K, Kaukinen K, et al. Increasing prevalence of coeliac disease over time. Aliment Pharmacol Ther. 2007;26:1217–1225
  4. Catassi C, Bearzi I, Holmes GK. Association of celiac disease and intestinal lymphomas and other cancers. Gastroenterology. 2005;128:S79–S86
  5. Gao Y, Kristinsson SY, Goldin , et al. Increased risk for non-Hodgkin lymphoma in individuals with celiac disease and a potential familial association. Gastroenterology. 2008;136:91–98
  6. Catassi C, Fabiani E, Corrao G, et al. Risk of non-Hodgkin lymphoma in celiac disease. JAMA. 2002;287:1413–1419
  7. Smedby KE, Hjalgrim H, Askling J, et al. Autoimmune and chronic inflammatory disorders and risk of non-Hodgkin lymphoma by subtype. J Natl Cancer Inst. 2006;98:51–60
  8. Askling J, Linet M, Gridley G, et al. Cancer incidence in a population-based cohort of individuals hospitalized with celiac disease or dermatitis herpetiformis. Gastroenterology. 2002;123:1428–1435
  9. Smedby KE, Akerman M, Hildebrand H, et al. Malignant lymphomas in coeliac disease: evidence of increased risks for lymphoma types other than enteropathy-type T cell lymphoma. Gut. 2005;54:54–59
  10. Williams RL. A note on robust variance estimation for cluster-correlated data. Biometrics. 2000;56:645–646
  11. Kaprio J. Science, medicine, and the future (Genetic epidemiology). BMJ. 2000;320:1257–1259
  12. Rothman KJ, Greenland S. Causation and causal inference in epidemiology. Am J Public Health. 2005;95(suppl 1):S144–S150
  13. Malamut G, Afchain P, Verkarre V, et al. Presentation and long-term follow-up of refractory celiac disease: comparison of type I and type II. Gastroenterology. 2009;136:81–90
  14. Al-Toma A, Verbeek WH, Hadithi M, et al. Survival in refractory coeliac disease and enteropathy-associated T-cell lymphoma: retrospective evaluation of single-centre experience. Gut. 2007;56:1373–1378
  15. Peters U, Askling J, Gridley G, et al. Causes of death in patients with celiac disease in a population-based Swedish cohort. Arch Intern Med. 2003;163:1566–1572
  16. Solaymani-Dodaran M, West J, Logan RF. Long-term mortality in people with celiac disease diagnosed in childhood compared with adulthood: a population-based cohort study. Am J Gastroenterol. 2007;102:864–870
  17. Wall BF, Kendall GM, Edwards AA, et al. What are the risks from medical X-rays and other low dose radiation?. Br J Radiol. 2006;79:285–294
  18. Card TR, West J, Holmes GK. Risk of malignancy in diagnosed coeliac disease: a 24-year prospective, population-based, cohort study. Aliment Pharmacol Ther. 2004;20:769–775
  19. West J, Logan RF, Smith CJ, et al. Malignancy and mortality in people with coeliac disease: population based cohort study. BMJ. 2004;329:716–719
  20. Mearin ML, Catassi C, Brousse N, et al. European multi-centre study on coeliac disease and non-Hodgkin lymphoma. Eur J Gastroenterol Hepatol. 2006;18:187–194
  21. Verbeek WH, Van De Water JM, Al-Toma A, et al. Incidence of enteropathy-associated T-cell lymphoma: a nation-wide study of a population-based registry in The Netherlands. Scand J Gastroenterol. 2008;43:1322–1328
  22. Shack LG, Wood HE, Kang JY, et al. Small intestinal cancer in England & Wales and Scotland: time trends in incidence, mortality and survival. Aliment Pharmacol Ther. 2006;23:1297–1306

 The author discloses no conflicts.

PII: S0016-5085(08)02059-3

doi:10.1053/j.gastro.2008.11.026

Refers to article:

  • Editorial Accompanies Article Presentation and Long-Term Follow-up of Refractory Celiac Disease: Comparison of Type I With Type II , 06 October 2008

    Georgia Malamut, Pauline Afchain, Virginie Verkarre, Thierry Lecomte, Aurélien Amiot, Diane Damotte, Yoram Bouhnik, Jean–Frédéric Colombel, Jean–Charles Delchier, Matthieu Allez, Jacques Cosnes, Anne Lavergne–Slove, Bertrand Meresse, Ludovic Trinquart, Elizabeth Macintyre, Isabelle Radford–Weiss, Olivier Hermine, Nicole Brousse, Nadine Cerf–Bensussan, Christophe Cellier
    Gastroenterology January 2009 (Vol. 136, Issue 1, Pages 81-90)

  • Editorial Accompanies Article Increased Risk for Non-Hodgkin Lymphoma in Individuals With Celiac Disease and a Potential Familial Association , 26 September 2008

    Ying Gao, Sigurdur Y. Kristinsson, Lynn R. Goldin, Magnus Björkholm, Neil E. Caporaso, Ola Landgren
    Gastroenterology January 2009 (Vol. 136, Issue 1, Pages 91-98)

Gastroenterology
Volume 136, Issue 1 , Pages 32-34, January 2009

Article Tools