gms | German Medical Science

Aim: Some orally taken drugs (e.g. NSAIDs) are well-known to cause gastrointestinal adverse effects but differ in their respective potency. In order to assess these properties early in clinical development, esophagogastroduodenoscopy (EGD) is often used to monitor effects on the upper gastrointestinal (GI) mucosa. However, even in clinically healthy individuals mucosal lesions are observed, casting doubt on the utility of this biomarker.The objectives of the present analysis were to describe and evaluate the incidence of findings in the upper GI tract in healthy volunteers and to assess a possible association with demographics and social habits.

Method: A baseline EGD was performed to detect any pre-existing lesions in three clinical trials. The esophageal, gastric and duodenal (until pars descendens) mucosa was assessed by visual examination. Findings were described by type and number of observation (reddening, erosion, ulcer or other), size and location within the organ and by overall clinical relevance as assessed by the endoscopist. Clinical relevant findings excluded volunteers from trial participation. As potential covariates, gender, age, height, body weight, body mass index (BMI), consumption of alcohol, caffeine containing beverages and tobacco were tested. Data were compared descriptively between groups with and without clinically relevant finding re above mentioned covariates (one-way ANOVA, Pearson’s Chi Squared Test, p≤0.05).

Results: A total of 317 EGDs were included in the present analysis (278 (87.7%) in males, 39 (12.3%) in females; aged 32.1±8.2years, body weight 75.8±10.5kg, BMI 24.09±2.5kg/m², 214 consuming alcohol, 266 (of 312) consuming caffeine containing beverages and 39 (of 155) smokers). Of these, 137 EGDs (43.2%) showed at least one clinical relevant finding. The most frequent finding was erosion, 36 in esophagus, 92 in stomach, 20 in duodenum, and 10 ulcers in the stomach were observed). In contrast to all social habits tested, age, body weight and BMI showed statistically significant effects on the occurrence of pathological findings (age p=0.013, body weight p=0.002, BMI p<0.001).

Conclusion: Upper GI tract mucosal lesions are frequent in clinically healthy individuals. This suggests a poor validity of low-grade lesions found in EGD as surrogate parameters predicting the gastrointestinal risk of drugs suspected to cause GI lesions.