| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
LETTERS TO THE EDITOR |
Department of Nutrition
Bussie 594
North-West University, Potchefstroom Campus
Private Bag X6001
Potchefstroom 2520
South Africa
E-mail: machteld{at}oomvanlieshout.net
Helen Keller International
Asia Pacific
Singapore
Department of Human Nutrition
Wageningen University
Wageningen, Netherlands
Dear Sir:
Haskell et al (1) recently reported estimates of the vitamin A equivalency factors of ß-carotene in oil, spinach, and orange-fleshed sweet potato that differ from those found by our laboratories (2-6). They reported factors of 6:1, 10:1, and 13:1, respectively, whereas we reported estimates of 
2.7:1, 26:1, and 12:1, respectively (2-6). The estimates for yellow, orange, and red fruit and vegetables (YORFV) are almost the same, but those for ß-carotene in oil and dark-green leafy vegetables (DGLV) are considerably different. In this letter, we discuss the possible reasons for and consequences of these differences.
Haskell et al used the paired deuterated-retinol-dilution (DRD) technique, which is based on the dilution of isotopically labeled [2H4]retinyl acetate before and after unlabeled retinol or ß-carotene is provided, as a supplement or through the diet. Three of our studies used the plateau isotopic enrichment (PIE) technique, which is based on reaching a plateau of isotopic enrichment during prolonged intake of multiple low doses of ß-carotene and retinol, each specifically labeled with 10 13C atoms. For the PIE technique, the level of the plateau depends on the change in the body pool of unlabeled retinol and ß-carotene as a result of concurrent intake of dietary ß-carotene.
First, we address the estimates of ß-carotene in oil. With use of the PIE technique, we found that amounts of 2.4 (95% CI: 2.1, 2.7) (4), 2.7 (95% CI: 2.5, 2.8) (5), and 3.2 (95% CI: 2.7, 3.7) (6) µg ß-carotene in oil have the same vitamin A activity as 1 µg retinol in oil, both in healthy Dutch adults (6) and in Indonesian schoolchildren with marginal vitamin A status and of whom 62–74% were infested with intestinal parasites (4, 5). Haskell et al hypothesize that their higher estimate (6.3 µg) is due to high dietary fiber intake from the meals provided (58 g/d). In our studies, dietary fiber content was indeed considerably lower, 20–30 g/d
(values for references 4 and 5 were extrapolated from reference 2, and for reference 6, the content was analyzed at 24 g/d), and it has been reported that some dietary fibers lower the bioavailability of some carotenoids (7, 8). However, it is also possible that the corn oil capsule used in Haskell et al's study for administering the ß-carotene, the size of which was unspecified, was too small for the 2.25 mg ß-carotene to be completely solubilized, which was previously found to result in unrealistic estimates (5).
Second, we address the DGLV estimates. As also mentioned by Haskell et al, there were several differences between their study and ours. In their study, Bangladeshi men with adequate vitamin A status (average serum retinol concentration: 1.27 µmol/L) were dewormed and received pureed sautéed spinach, whereas our studies were conducted among Indonesian children (2, 5) and Vietnamese breastfeeding women (3), all of whom had low vitamin A status and most of whom were infested with intestinal parasites. In our studies, the foods were provided as they are normally consumed, almost uncut and stir-fried. Disruption of the matrix has previously been found to almost double the bioavailability of ß-carotene (8). Extrapolation of Haskell et al's estimate to uncut spinach would then give a conversion factor of 19:1, which is much closer to our results. In addition, we previously hypothesized that the negative effect of intestinal parasites on vitamin A equivalency is greater if ß-carotene is entrapped in more complex matrices, such as in DGLV, than when it is in YORFV (2). The comparable estimates for ß-carotene in oil in Dutch (6) and Indonesian (4, 5) volunteers support this theory, as do the comparable estimates found for YORFV by Haskell et al and by our laboratories.
We found that YORFV are a better source of retinol than are DGLV, whereas Haskell et al found the opposite. Haskell et al argue that some of the difference may be due to the limitations, with which we agree, of using the serum retinol response technique. Two of our studies used this technique after 9–10-wk interventions with diets rich in vitamin A from either YORFV, DGLV, or dietary retinol compared with a low–vitamin A diet (2, 3). However, because our findings were confirmed in a study using the PIE technique (5), we do not think this explains the differences between our and Haskell et al's findings (2, 3, 5).
The CV of the initial and final pool sizes in Haskell et al's study was considerable (44–78%). Proper comparison of their and our vitamin A equivalency estimates requires data about the spread of their estimates. The spread around the estimates obtained from de Pee et al's study (2) was also large, but the findings were confirmed in a subsequent similar study in Vietnam (3). The PIE technique does have a high precision, has been proven to be reproducible (4-6), and is being validated against liver stores of retinol. Although the DRD technique has been validated for assessing vitamin A status, for which it is probably the most accurate technique available, the paired DRD technique has not yet been proven to have adequate precision for quantifying vitamin A equivalency. An additional advantage of the PIE technique is that it requires only 21–42 d for assessing the vitamin A equivalency of ß-carotene in oil or in a dietary source, respectively (4-6), instead of the 113 d required by the paired DRD technique.
In conclusion, we strongly advocate the quest for reliable techniques for estimating the vitamin A equivalency of provitamin A carotenoids. Moreover, reliable estimates of the vitamin A equivalency of dietary ß-carotene are essential for guiding public health policies developed for those who depend on provitamin A carotenoids for maintaining their vitamin A status. Therefore, we stress that such estimates should be obtained in populations at risk of vitamin A deficiency, such as children and pregnant or lactating women, under circumstances realistic for their living conditions and representing their food choice and preparation methods. This is required to confirm our findings (9) and others' statements (10) that the vitamin A activity of dietary ß-carotene (12:1 for healthy populations in developed countries; 10) appears to be lower for malnourished populations. Therefore, we meanwhile advocate the use of the more conservative 21:1 conversion factors for dietary ß-carotene for persons at risk of vitamin A deficiency (9).
ACKNOWLEDGMENTS
The authors had no conflicts of interest to report.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  C. E. Harjes, T. R. Rocheford, L. Bai, T. P. Brutnell, C. B. Kandianis, S. G. Sowinski, A. E. Stapleton, R. Vallabhaneni, M. Williams, E. T. Wurtzel, et al. Natural Genetic Variation in Lycopene Epsilon Cyclase Tapped for Maize Biofortification Science, January 18, 2008; 319(5861): 330 - 333. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. C. Khan, C. E West, S. de Pee, D. Bosch, H. D. Phuong, P. J. Hulshof, H. H. Khoi, H. Verhoef, and J. G. Hautvast The contribution of plant foods to the vitamin A supply of lactating women in Vietnam: a randomized controlled trial Am. J. Clinical Nutrition, April 1, 2007; 85(4): 1112 - 1120. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
