Introduction With the increasing number of gastric cancer cases in young patients in Zambia, we set out to investigate the possibility of reversing gastric atrophy, a precancerous lesion, using micronutrient-antioxidant supplementation in a retrospective analysis of a published randomised controlled trial. We recently reported that low antioxidant levels were associated with gastric cancer in Zambia. In an earlier study, we had demonstrated that people with the HIV infection tend to have higher gastric pH than those without HIV.

Methods Archival samples from a randomised controlled trial, (Kelly et al Trans R Soc Trop Med Hyg. 2008; 102:194–9) were used in this study. These were collected from healthy volunteers in Misisi, a densely populated and impoverished township in Lusaka, and carefully stored at –80oC in a secure laboratory at the University Teaching Hospital, the largest referral hospital in Zambia. In this controlled trial, 500 volunteers were randomly allocated to either a micronutrient-antioxidant supplementation or placebo. The supplements contained vitamins, A, B1, B2, B6, B12, C, D3, E, niacin, folic acid, iron, zinc, copper and selenium. We analysed 215 samples collected in 2005, from subjects who had taken either supplementation or placebo for a median of 19 (range 14 to 27) months. Gastric atrophy was determined using pepsinogen 1 to 2 ratio of less than 3.0 using BIOHIT ELISA kits, Helsinki, Finland, according to the manufacturer’s instructions. Fasting gastric pH was available on 121 participants. The presence of atrophy was compared between the intervention and the placebo groups. Other factors analysed included the effect of HIV infection, age, body mass index (BMI), smoking, alcohol intake and gastric pH.

Results Gastric atrophy was found in 8 (7.8%) of 103 subjects on supplementation, and 7 (6.3%) of 112 on placebo (RR 1.24; 95%CI 0.47–3.3; P = 0.22). HIV infection was diagnosed in 5 participants with atrophy and 61 without (RR 1.07; 95%CI 0.37–3.2; P = 1.0). The lack of effect of supplementation on atrophy was not changed after stratification for HIV status (M-H OR 2.0; P = 09.20). Gastric atrophy was found to be more prevalent in those above the age of 40 years. In univariate and multivariate analysis, BMI, smoking, alcohol intake showed no impact on gastric atrophy. Gastric pH and pepsinogen 1:2 ratio were inversely correlated (Spearman’s r = –0.34; P = 0.0001).

Conclusion An average of nine months of micronutrient-antioxidant supplementation has no impact on gastric atrophy in Zambian healthy adults. The high gastric pH seen in HIV patients can not be attributed to gastric atrophy.

Disclosure of Interest None Declared.