Richard E Frye, Daniel A Rossignol
Background: Mandatory folic acid (FA) fortification prevents neural tube defects but generates unmetabolized folic acid (UMFA) in blood, because hepatic dihydrofolate reductase (DHFR) reduces FA only slowly. Whether UMFA itself, as distinct from high total folate, carries biological consequences is unresolved. Methods: We systematically searched six databases (January 1995-May 2026; PROSPERO CRD420261407616) and included 65 publications describing 61 unique primary studies, yielding 71 analytic contributions (15 mechanistic, 24 observational, 32 interventional). The question of whether equimolar (6S)-5-MTHF substitution alters UMFA, erythrocyte folate, or total homocysteine (tHcy) versus FA was evaluated using random-effects meta-analysis with a DerSimonian-Laird estimator for τ2 and Hartung-Knapp-Sidik-Jonkman small-sample confidence intervals and was rated with GRADE. Mechanistic and observational streams were synthesized narratively and classified by exposure metric (UMFA, total folate, FA intake, fortification). Results: 5-MTHF substitution reduced plasma UMFA (SMD +0.99; 95% CI +0.20 to +1.79; I2 = 30%; k = 3; GRADE low). The two trials reporting erythrocyte folate were directionally discordant (Hedges' g +0.40; +0.08 to +0.73 and -0.38; -0.92 to +0.16; I2 = 83%) and were not pooled. For tHcy, no pooled estimate could be derived; it is reported qualitatively. Observational UMFA-outcome associations were inconsistent, confounded by total folate, and unreplicated. Conclusions: Substitution of (6S)-5-MTHF for FA appears to reduce circulating UMFA. For erythrocyte folate, the two available trials were directionally discordant but insufficient to establish equivalence or non-inferiority. For tHcy, no between-form difference was detected. With two to three trials per outcome, no prespecified margins, and no trial designed for that purpose, the evidence cannot support an equivalence or non-inferiority claim for either outcome and clinical benefit remains unestablished. Current evidence does not establish that UMFA independently causes harm.