Severe experimental folate deficiency in a human subject - a longitudinal investigation of red-cell folate immunoassay errors as megaloblastic anaemia develops

Red Blood Cell Folate Likely Overestimated in Australian National Survey: Implications for Neural Tube Defect Risk

In 2009, the Australian government mandated the addition of folic acid to bread flour to reduce the incidence of neural tube defects (NTD)-affected pregnancies. In 2011-2012, the Australian Health Measures Survey (AHMS) reported a mean red blood cell (RBC) folate in women of reproductive age (16-44 y) of 1647 nmol/L. Over 99% of women had an RBC folate ≥ 906 nmol/L, a concentration consistent with a very low risk of NTDs if a woman became pregnant. However, RBC folate was measured using an immunoassay, which is not a recommended method due to questionable accuracy. The microbiological assay is the preferred method for RBC folate measurement. To determine whether the immunoassay method may have led to spurious conclusions about the folate status of Australian women, we collected fasting blood samples from 74 healthy non-pregnant, non-lactating women (18-44 y) and measured RBC folate using both the immunoassay and microbiological methods. Mean RBC folate (95% confidence interval) concentration measured with the immunoassay method was 1735 (1666, 1804) nmol/L compared with 942 (887, 1012) nmol/L using the microbiological method. No woman had an RBC folate < 906 nmol/L using the immunoassay method, whereas 46% of women had an RBC folate < 906 nmol/L using the microbiological method. The NTD risk was estimated to be 0.06% using the immunoassay method and 0.14% using the microbiological method. RBC folate using AHMS survey may have underestimated NTD risk in Australian women.

Severe experimental folate deficiency in a human subject - a longitudinal study of biochemical and haematological responses as megaloblastic anaemia develops

BACKGROUND

The currently accepted theory, that the human liver store of folate is limited to about four months, is based on the findings of Victor Herbert and others of the era before folate fortification of food. A recent model, developed by Lin et al., predicts far greater liver folate storage capacity than reported by Herbert. The conflict between Herbert's and Lin's models needs to be resolved experimentally, however current research is restricted because ethical considerations prevent such risky experimentation on patients or healthy human volunteers. The objective was to provide a detailed record of the biochemical and haematological responses to the development of severe experimental folate deficiency in an initially replete human subject.

METHODS

This 58 year old male severely depleted himself of folate, using a folate-deficient diet, until overt megaloblastic anaemia developed. The biochemical and haematological responses were monitored by routine blood tests. Daily intake of dietary supplements prevented deficiencies of other relevant nutrients.

RESULTS

The rate of change of all analytes was significantly slower, and the delay before any change for several analytes was significantly longer, than reported for previous experiments. The time before reporting of abnormal biochemical and haematological results was therefore very significantly longer than reported by Herbert, but was consistent with the recent model of Lin et al. Serum folate and red-cell folate became abnormally low after 219 and 413 days respectively. Macrocytic anaemia was produced after 469 days, and megaloblastic anaemia was confirmed by bone marrow biopsy on day 575. Folate starvation ceased on day 586, and recovery was complete on day 772.

CONCLUSIONS

The currently accepted four month time scale for development of megaloblastic anaemia from folate deficiency, based on the early work of Herbert and others, is not consistent with the results from this study. The > 300 day liver folate storage time, predicted by the model of Lin et al., is supported by this experiment. Self-experimentation has produced a detailed record of the biochemical and haematological responses to severe experimental folate deficiency, whereas using patients or healthy volunteers as subjects would be unethical.