Background: Dietary therapy is an effective treatment for eosinophilic esophagitis (EoE), however, the influence of body mass index (BMI) on treatment response is unclear.
Aim: To determine whether outcomes to dietary therapy in adults with EoE vary by BMI or change in weight during treatment.
Methods: In this retrospective cohort study conducted at a large academic medical center, subjects were adults with EoE treated with dietary therapy. Dietary therapy for EoE included elemental, empiric, and targeted elimination diets. Clinical characteristics and treatment responses to diet therapy were extracted from electronic health records. BMI was calculated at EoE diagnosis and post-treatment endoscopy. Standard BMI categories were used (overweight/obesity: BMI>25.0, healthy weight: BMI 18.5-24.9). Treatment response was defined by global symptom, endoscopic, and histologic responses (<15 eos/hpf, with additional assessment of ≤6 and <1 eos/hpf). We assessed the relationship of BMI and weight gain on treatment response using bivariate and multivariable analyses. In addition, BMI vs weight gain class interaction was assessed to determine whether the influence of weight gain on outcomes differed across BMI classes.
Results: We identified 169 EoE patients treated with dietary therapy, over half of whom were overweight/obese (57%). Baseline characteristics and dietary treatment type were similar between adults with overweight/obesity and healthy weight (Table 1). Dietary therapy included empiric elimination (81%), targeted elimination (18%), and elemental (1%) diets. Diet type did not statistically differ over the study period (p=0.94) or by BMI (p=0.88). Weight gain with dietary therapy occurred in 38% of adults with overweight/obesity and 37% with healthy weight (p=0.88) and did not differ by diet type (p=0.22). Treatment response was not different between adults with overweight/obesity and healthy weight (Fig 1A). However, compared to those with no weight gain during dietary therapy, weight gain was associated with lower histologic response at <15 (19% vs 47%, p<0.001), <6 (13% vs 38%, p<0.001), and <1 (5% vs 16%, p=0.03), lower endoscopic response (47% vs 61%, p=0.07), and lower symptomatic response (56% vs 72%, p=0.03) (Fig 1B). Multivariable logistic regression revealed a 1-unit increase in BMI with dietary therapy was associated with a decrease in the odds of histologic response at <15 (aOR 0.65; 95% CI 0.44-0.95), ≤6 (aOR 0.66; 95% CI 0.45-0.98), and <1 (aOR 0.61; 95% CI 0.39-0.96) and symptom response (aOR 0.50; 95% CI 0.30-0.84). There was no significant BMI/weight gain interaction.
Conclusions: In adults with EoE, weight gain during diet therapy is associated with decreased treatment response, regardless of baseline weight status. Whether this is a sign of treatment non-adherence or related to underlying EoE pathophysiology remains to be determined.

Table 1. Comparison of adults with EoE treated with dietary therapy by baseline weight status (overweight/obesity (BMI>25.0) vs healthy weight (BMI 18.5-24.9)) and by with and without weight gain
Figure 1. Treatment response to dietary therapy in adults with eosinophilic esophagitis by (A) weight status (overweight/obesity vs normal weight) and (B) by change in weight (weight gain vs no weight gain)