Home

The National Health and Nutrition Examination Survey (NHANES) has monitored total homocysteine (tHcy) concentrations in a nationally-representative sample of the US population since 1991. Until recently, however, data could not be compared across survey periods because of changes in analytical methods and specimen matrices. Such an analysis of these data could supplement current knowledge regarding whether the US folic acid fortification program has modified national plasma tHcy concentrations.

We examined tHcy data in the prefortification NHANES III survey (phase II, 1991-1994) and in 3 postfortification survey periods (1999-2000, 2001-2002, and 2003-2004). We applied method adjustment equations to the survey data based on method comparison studies of separate samples. Persons with chronic kidney disease were excluded from the analyses. Mean plasma tHcy concentrations decreased by 8%, 9%, and 10% for adolescent, adult, and older men and by 6%, 3%, and 13% for women, respectively, from before to after fortification.

Concentrations remained unchanged between the first and third postfortification survey periods. Prevalence estimates of increased plasma tHcy concentrations (>13 µmol/L) for older men and women decreased from prefortification (32% and 20%, respectively) to postfortification (14% and 5%, respectively) but remained unchanged thereafter (16% and 14%, respectively [males] and 5% and 9%, respectively [females]).

After adjusting for method changes, we quantified a prefortification to postfortification decrease in circulating tHcy concentrations of about 10% in a national sample of the US population. This change is similar to effects seen in intervention trials with folic acid and in smaller observational studies.

Discussion

Circulating tHcy concentrations have been measured as part of the NHANES survey for 3 years before and 6 years after the introduction of folate fortification. Until now, however, tHcy data could not be compared across survey periods because of changes in analytical methods and specimen matrices. We derived method adjustment equations to allow for time trend analysis.

This process had several strengths: (a) sample sizes for method comparisons were of sufficient size, except the EDTA plasma-to-serum comparison; (b) all 3 method comparisons spanned the reference interval of tHcy concentrations; (c) each comparison was conducted over a period of several days to capture the interrun variability; (d) the small adjustment between the Abbott IMx and AxSYM platforms was consistent with a previous report (5, 28); (e) all measurements were adjusted to the AxSYM method, which has shown accurate agreement with international standard reference material (36); and (f) the processing time of whole blood was standardized separately in NHANES 1991-1994 and NHANES 1999-2004, minimizing imprecision in tHcy values due to inconsistent sample handling.

We have presented a detailed analysis of circulating tHcy concentrations in the US population. Although the complex method adjustment performed to allow for time trend analysis generated some uncertainty, this uncertainty was small enough to allow the quantification of a modest decrease in plasma tHcy concentrations from before to after fortification. The decrease of approximately 10% corresponds well with effects seen in intervention trials with folic acid and in smaller observational studies. We saw no significant changes in tHcy concentrations during the 6 postfortification years. Stabilization of tHcy concentrations may largely depend on the stabilization of blood folate concentrations as well as on other possible causes of increased concentrations, including but not limited to vitamin B^sub 12^ status. Therefore, monitoring of all related biomarkers should continue, and further investigations of other possible causes of high tHcy concentrations should be undertaken.