Why the topic matters: limitations of previous research

Vitamin D is traditionally viewed as a regulator of inflammation and repair processes. Clinical literature frequently reports that patients with keloids, acne, or hypertrophic scars have lower serum 25(OH)D levels than healthy individuals.

However, these data have a fundamental limitation: they do not demonstrate causality. Low 25(OH)D could result from lifestyle, comorbidities, skin characteristics, phototype, or numerous unmeasured factors linked to both vitamin D and scarring. Consequently, a practical doubt remains: does serum vitamin D itself influence the risk of pathological scarring, or are we merely observing an association?

Methodology: Mendelian randomization

To address the question of causality, the authors of the study published in the Journal of Cosmetic Dermatology in 2025 [1] performed a two-sample Mendelian randomization (MR) using summary data from Genome-Wide Association Studies (GWAS). A GWAS is a type of genetic study that scans the genomes of large populations (thousands to hundreds of thousands) to find small genetic variations — single-nucleotide polymorphisms (SNPs) — that occur more frequently in people with a particular trait or disease. This enables the identification of genetic markers associated with vitamin D levels or predisposition to scar formation.
The logic of Mendelian randomization rests on the fact that each individual receives their set of genetic variants (SNPs) at conception, independent of lifestyle, diet, or comorbidities. This resembles a lottery: some "draw" variants are associated with higher vitamin D levels, while others are associated with lower levels. This randomness is used as a tool: instead of comparing people with different vitamin D blood levels (which are influenced by many external factors), the authors compare groups with different genetic predispositions toward those levels. This approach reduces the influence of confounding factors and avoids "reverse causality" where it is unclear which came first — low vitamin D or the condition itself.

Key methodological points:

• The primary factor studied (exposure) was genetically predicted serum 25(OH)D levels, derived from a large-scale GWAS of hundreds of thousands of participants of European ancestry.
• Outcomes (keloid, acne, hypertrophic scar) were also sourced from independent GWAS datasets to minimize sample overlap.
• Strict quality criteria were used to select genetic markers: only variants with an indisputable statistical association with vitamin D levels were chosen. To avoid data duplication, "clumping" was performed to exclude genetically linked markers located too close to each other. The authors also ensured that the chosen genetic instruments had sufficient "strength" (F-statistic significantly above the standard threshold of 10), guaranteeing the reliability of the findings.
• The primary method for calculating the causal effect was Inverse-Variance Weighted (IVW). Results were further validated using several sensitivity analyses (MR-Egger, weighted median, weighted mode, and MR-PRESSO) to assess whether findings were biased by the diversity of genetic instruments (heterogeneity) or by their influence on the outcome through pathways unrelated to vitamin D (horizontal pleiotropy). Additionally, a "leave-one-out" analysis was conducted by removing each SNP in turn to ensure that no single marker disproportionately affected the overall conclusion.

Study results: keloids and hypertrophic scars

For keloids, genetically predicted 25(OH)D levels showed no significant causal effect on keloid formation risk [1]. All analytical methods yielded a consistent result: no link was found. The genetic instruments performed correctly, showing no significant heterogeneity or signs of influencing the outcome through unintended pathways. The leave-one-out analysis confirmed that the conclusion remained unchanged regardless of any single marker.

For hypertrophic scars (HS), the results were identical: neither the primary method nor the additional analyses revealed a statistically significant causal relationship with 25(OH)D levels. The findings remained stable across all reliability checks.

Practical interpretation and limitations

The practical interpretation is subtle but vital. This study does not suggest that vitamin D is "unnecessary" for the skin or healing; the authors reiterate its role in repair and inflammatory regulation. However, with respect to the risk of developing keloids and hypertrophic scars, genetically predicted serum 25(OH)D levels in the European population showed no causal effect. This supports the view that low 25(OH)D found in patients with pathological scars in observational studies may be a marker of other factors rather than a direct cause of scarring.

Several limitations affecting the translation of these results into practice were highlighted: the study was conducted only in Europeans (limiting generalizability to other populations), and the mathematical model assumed a linear relationship (without accounting for potential complex threshold effects). The role of vitamin D-binding protein could not be fully assessed due to a lack of strong genetic markers (SNPs). For more precise confirmation, the authors suggest using multi-omics integration in future research.

The discussion also notes that the association between vitamin D deficiency and the risk of keloids or HS may vary across different ethnic groups, further emphasizing the need for population-specific studies.

Conclusion

Correcting vitamin D deficiency remains a general medical goal. Still, this study does not support using serum 25(OH)D alone as a "target" for preventing keloids or hypertrophic scars, given the lack of causality.

The clinical value of this research lies in reducing unrealistic expectations of vitamin D as a standalone preventive strategy for pathological scarring and highlighting the need for further research across diverse ethnic groups and other components of vitamin D metabolism.

Source:

1. Ji J., Shi X., Tang E. et al. Serum vitamin D levels and its relationship with keloid, acne, or hypertrophic scar: a two-sample Mendelian randomization study. J Cosmet Dermatol 2025: 24870398.