Reply to Mendall

We thank Dr. Mendall for his comments about our study describing the increasing average weight of patients with Crohn’s disease in clinical trials over the years and the significant correlation between this novel finding and disease activity. This was a hypothesis-generating study of the data published in connection with previous clinical trials. The author states that fecal calprotectin is a useful surrogate marker for Crohn’s disease risk and could thus be useful in future epidemiological studies. The role of fecal calprotectin in the association between obesity and Crohn’s disease is a valid one. It was not possible to address this in our study of previous clinical trials of therapeutic agents. The associations ascribed to obesity, such as decreased physical activity and a high-fat diet, are ecological and do not impute causality with Crohn’s disease. Adiposity in general may be proinflammatory as a result of multiple adipokine and cytokine effects. We agree that obesity might have a causal effect in Crohn’s disease and could possibly affect its clinical outcome. Even in the absence of adiposity, a high-fat diet might accelerate disease pathogenesis in Crohn’s disease, and alteration of the microbiome and barrier function could underlie this association. Unfortunately, body mass index, which is the most routinely used clinical weight measurement, is a very insensitive marker of adiposity. A large proportion of malnourished patients with Crohn’s disease have a normal body mass index, despite having an abnormal body composition and muscle function. This observation might explain the discrepancy in the findings between the European Prospective Cohort Study and the Nurses’ Health study. A bimodal distribution in the incidence of Crohn’s disease based on obesogenic factors is a contentious issue. The present data do not suggest that the incidence of obesity is higher in pediatric Crohn’s disease, but surgical risk is significantly associated with an obese status, possibly indicating more aggressive disease in this cohort. The observed difference, when compared with the adult data, could be explained by the confounders of reduced disease duration, less life-years exposure to the proinflammatory obese state, and a tendency to a different disease phenotype in the pediatric Crohn’s cohort. Further work needs to be undertaken to investigate the association between adiposity and its effects on the incidence and severity of Crohn’s disease both in adult and pediatric population.