A new study explores the possibility of increased mortality rates due to persistent villous atrophy in patients diagnosed with and treated for celiac disease.
Past research has shown an association between villous atrophy in the small intestine (damage to the fingerlike projections in the intestine that absorb nutrients from food) in celiac disease patients and increased mortality rates. In a recent study published in the Alimentary Pharmacology and Therapeutics journal, researchers set out to determine if healing of the small intestine decreases this risk of mortality.
Benjamin Lebwohl, MD, MS, Assistant Professor of Clinical Medicine and Epidemiology at the Celiac Disease Center at Columbia University in New York, is the lead author of this recent study. To help explain the study results and what they mean for people with celiac disease, Dr. Lebwohl participated in a Q&A with the National Foundation for Celiac Awareness (NFCA).
Q. Past research has shown that villous atrophy leads to increased mortality. In contrast, your study concluded that persistent villous atrophy does not increase mortality risk. You discuss a variety of reasons for this finding; can you highlight one of these rationales here?
A. One potential reason that persistent villous atrophy was not associated with mortality was that patients may have altered their behavior after learning the results of the follow-up biopsy. A patient learning that she has persistent villous atrophy may then meet with a dietician, making an effort to improve adherence to the gluten-free diet. If subsequent strict adherence to the diet is protective, then perhaps that would explain our finding of no mortality difference after the follow-up biopsy.
Q. On unadjusted Cox proportional hazard modeling, persistent villous atrophy was associated with an increased risk of death. After adjustment, this increased risk was no longer significant. Can you briefly explain “unadjusted Cox proportional hazard modeling?” If the results prior to adjustment showed an increased risk in mortality, why do the final results show no increased risk?
A. Cox proportional hazard modeling is a method to estimate the risk of death over time. In an observational study, individuals are followed by varying durations until either death, loss to follow-up (which is rare in Sweden), or the end of the study period. This modeling provides estimates of the risk of death in one group (in this case persistent villous atrophy) compared to another (mucosal healing). It also allows us to adjust for variables that may be different between these groups, which is important because these variables may themselves affect survival.
In this case, patient age was an important variable. As it turned out, the group with persistent villous atrophy was significantly older, on the whole, than those with mucosal healing. As a result, an unadjusted comparison would find that the group with persistent villous atrophy would have a higher rate of death; this is unsurprising given that older age is a very important predictor of death! But once we adjust for age (and several other variables, including gender, duration of celiac disease, and socioeconomic status) we found that there was no difference in mortality risk between those with persistent villous atrophy and those with mucosal healing.
Q. In light of your findings showing that persistent villous atrophy does not cause an increased risk of mortality, can you speak to the importance of strict adherence to the gluten-free diet? For example, some people may be now tempted to occasionally cheat on their diet, but you actually discuss the possibility of those who undergo follow-up biopsy as having an improved diet adherence.
A. This study does not justify cheating on the diet! For one, it could very well be the case that those with persistent villous atrophy subsequently increased their adherence after receiving their biopsy result. Second, mortality is an important (and the ultimate) outcome but this study did not investigate other important outcomes such as cancer risk and other consequences of celiac disease.
Q. This study mainly focused on “persistent” villous atrophy. Can these results be expanded to draw conclusions about intestinal damage done occasionally due to accidental gluten ingestion, such as cross-contamination when dining out? Would occasional accidental ingestion be enough to cause persistent villous atrophy?
A. While we believe that the group with persistent villous atrophy probably had lower dietary adherence than those with mucosal healing (at least before their biopsy), it is difficult to be certain about an individual patient’s threshold regarding gluten ingestion and villous atrophy. While some patients have persistent villous atrophy that is due to occasional accidental ingestion of small amounts of gluten, other patients may admit to more frequent dietary lapses and yet still have mucosal healing on their biopsy. While it is often useful, the follow-up biopsy is an imperfect way of measuring for gluten exposure in the individual. Follow-up biopsy results should be interpreted in the context of the whole clinical picture, which includes a session with a dietician with expertise in celiac disease and the gluten-free diet.
Q. In looking at future studies, is it possible to investigate if there is a difference in the mortality risk in people who were diagnosed earlier than others? For example, if one person was diagnosed within one year of symptom presentation and another did not receive a diagnosis for ten years, would you expect there to be a difference in mortality risk for these two people?
A. It would be interesting to know whether the length of illness prior to diagnosis of celiac disease predicts mortality, or if this modifies the impact of the follow-up biopsy on mortality risk. The difficulty with this approach is that, in the absence of frozen serum that could be tested for celiac disease antibodies, it is often difficult to be sure how long a patient has had celiac disease prior to diagnosis. Symptoms may come on slowly and gradually, and symptom onset does not necessarily correlate with disease onset. So there is much more to learn about the effect of undiagnosed celiac disease on clinical outcomes.
Learn more about the Celiac Disease Center at Columbia University by following them on Facebook.