High Ultra-Processed Food Consumption Is Associated with Clinical Exacerbation in Patients with Crohn's Disease in Remission: A Prospective Cohort Study

Abstract

Introduction: Diet can impact the clinical course of Crohn's disease (CD). However, the importance of food processing level is uncertain. We aimed to evaluate whether ultra-processed foods (UPFs) are associated with an increased risk of relapse among patients with CD in clinical remission.

Methods: In this prospective cohort study, adult patients with CD in clinical remission were followed for 1 year or until clinical relapse, as assessed by the Harvey Bradshaw index. Dietary intake was assessed using a food frequency questionnaire and a dedicated validated processed food questionnaire to evaluate UPF consumption. Multivariable Cox proportional hazard models were used to estimate the adjusted hazard ratio of disease relapse according to high or low UPF consumption.

Results: A total of 111 patients with CD (mean age 37.9 ± 14.0 years) in clinical remission were followed. Over a median period of 12 months [interquartile range, 8-12], 24 patients (21.6%) experienced clinical relapse. Fecal calprotectin levels were increased at the relapse visit compared to baseline (193.0 μg/g [114.0-807.0] and 79.0 μg/g [17.0-274.5], respectively, p = 0.002). A higher intake of UPFs was associated with an increased risk of CD relapse (HR = 3.86; 95% CI 1.30-11.47). Ultra-processed breads and pastries, as well as oils and spreads, were the UPF subgroups with the strongest positive associations (HR = 3.27; 95% CI 1.26-8.45 and HR = 2.76; 95% CI 1.02-7.45, respectively).

Conclusion: UPF consumption is associated with an increased risk of relapse in patients with CD in remission. This study, along with future research, can contribute to establishing nutritional guidelines aimed at reducing UPF consumption in this population.

Keywords: Diet; Inflammatory bowel diseases; Processed food questionnaire; Ultra-processed foods.

Fig. 1.

Flow chart of the study. *Twenty-two patients (19.8%) did not complete the 1-year follow-up: 1 patient withdrew due to a change in medical treatment prompted by a significant increase in fecal calprotectin, one due to an imbalance related to type 2 diabetes, another due to a new cancer diagnosis, two due to personal reasons, two became pregnant during the follow-up, and fifteen were lost to follow-up. IBD, inflammatory bowel disease; PFQ, processed food questionnaire; FFQ, food frequency questionnaire; CD, Crohn’s disease; FC, fecal calprotectin.

Fig. 2.

Comparison of time to clinical relapse in patients with CD stratified by ultra-processed foods consumption. The cut-off for low and high ultra-processed food consumption was determined by the median number: 3.6 servings/day.

Fig. 3.

Forest plot of the adjusted hazard ratios for clinical relapse in patients with CD based on the consumption of various UPF food groups. High intake of each UPF group was defined by the median number of servings consumed per day within that group. The Cox proportional hazards model with time-dependent covariates was used for analysis. HRs are adjusted for sex, age (years), smoking status, energy intake (kcal/day), saturated fat (% of total kcal), and monounsaturated fatty acids (% of total kcal). Ultra-processed food groups are as follows: (1) Bread pastries and starch: different types of industrial bread, crackers, ready-to-eat/heat pastries or pizza, salty purchased snacks, ready-to-heat French fries, or gnocchi. (2) Oil and spreads: mayonnaise, margarine, ready-made sauces for cooking or seasoning, coconut cream, different types of spreads, and different types of industrial ready-to-eat salads (excluding unprocessed oils such as olive oil and other vegetable oils). (3) Ultra-processed meat: processed red meat, processed poultry, hot dogs, sausages, and pastrami. (4) Sweet products and desserts: ready-made cake, sweetened yogurt, breakfast cereals, ice cream, energy bars, and ready-made sweet/chocolate bars. (5) Ultra-processed beverages: sweetened soft drinks, ready fruit juices, sweetened ready milk beverages, diet sweetened soft drinks or juices, energy drinks, and alcoholic liqueurs. HR, hazard ratio; CI, confidence interval; UPF, ultra-processed food.