We read the interesting paper by Lebwohl et al. The authors concluded that there was no association between persistence of villous atrophy and mortality in coeliac disease (CD). However, the study design and the interpretation of results raise some concerns.
In spite of the impressive number of patients evaluated, it is of note that only 26% of the CD-diagnosed patients during the study period had a follow-up biopsy, with the reason why these biopsies were taken unknown and therefore, as the authors mention in their discussion, ascertainment bias cannot be excluded.
It has recently been emphasised that small bowel biopsy remains the gold standard for monitoring mucosal healing in CD.[2, 3] Persistent villous atrophy was associated with a more than threefold increased risk of osteoporosis, refractory CD and malignancies irrespective of symptom resolution on a gluten-free diet (GFD). Thus, the question that emerges is whether the CD patients in this study cohort were adequately followed up.
A recent paper showed that medical follow-up evaluation in patients with CD is highly variable and often inadequate. In fact, follow-up consistent with current American Gastroenterology Association recommendations was observed in only 35% of patients. Currently, clear guidelines on when to repeat small-bowel biopsy are lacking. However, a systematic review of studies reporting the long-term management of CD suggested that follow-up biopsy at 1–2 years after commencement of the GFD is recommended, and that if histological improvement is incomplete, biopsies probably should be performed again in 1–2 years. Thus, the evaluated cohort of patients would not be useful to assess how persistent villous atrophy influenced mortality in CD, as the rate of follow-up small-bowel biopsy was quite low, the reason why these biopsies were taken was unknown, and there were no additional follow-up biopsies performed when histological improvement was incomplete.
Likewise, follow-up biopsy in the study by Lebwohl et al. demonstrated that there was persistent villous atrophy in 43% of the cohort. It is reasonable to think that in these patients, dietary measures to improve GFD adherence were taken, as lack of dietary adherence is considered the main reason for persistence of villous atrophy. Improvement in dietary adherence would have influenced survival, and decreased mortality to the level of patients who have mucosal healing. In this sense, the problem in the long-term management of CD is the nondiagnosed persistent villous atrophy. Only once it is detected, can the opportune dietary measures be established.