Evidence indicates hypervitaminosis A may be attributed to overconsumption of natural preformed vitamin A (VA) and overlapping VA intervention strategies. Hypervitaminosis A can disrupt metabolic processes; however, the extent and mechanisms of these impacts are not well understood. This study aims to assess metabolic differences related to hypervitaminosis A and VA supplementation by performing metabolomics analysis. A subsample of South African preschoolers participating in the country’s VA supplementation programme was selected. Participants were divided into two groups: adequate VA (n 15; 0·59–0·99 µmol/g total liver reserve and high VA (n 15; ≥ 1·0 µmol/g total liver reserve). Serum samples were collected at baseline and 28 d after consuming a 200 000 IU VA supplement. Lipidomics and oxylipins assays were conducted using ultraperformance LC-MS. At baseline, unsaturated lysophosphatidylcholines and unsaturated phosphatidylcholines were significantly lower in the high VA group (P < 0·05). A group-by-time interaction with VA supplementation was observed for polyunsaturated lysophosphatidylcholines and polyunsaturated phosphatidylcholines (P < 0·05). Additionally, a group effect was noted for oxylipins, and a time effect in response to VA supplementation was seen with decreased arachidonic acid and lipoxygenase- and non-enzymatically derived oxylipins (P < 0·05). Hypervitaminosis A is associated with modifications in lipids involved in cell structure and signalling, particularly unsaturated lysophosphatidylcholines and phosphatidylcholines. Further research is needed to identify the mechanisms behind these modifications, their physiological effects and their potential as biomarkers of elevated vitamin A status.