| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
1 School of Pharmacy, Texas Tech University, Health Sciences Center, 1300 Cowter Dr. Amarillo, TX 79106
To the Editor:
In their report, Infante-Rivard et al. (1) presented evidence that homocysteine (Hcy) values were largely <15 µmol/L in the participants in their case–control study. The cases were born at their institution over a 2-year period and had birthweights below the 10th percentiles for gestational age and sex. The control group included babies born at or above the 10th percentiles. The mothers were also included in the study. These investigators initiated their research based on an assumption that higher maternal and newborn Hcy concentrations in plasma would increase the risk of intrauterine growth restriction through placental thrombosis. In contrast to their proposed hypothesis, however, they concluded that mothers with small babies had lower Hcy concentrations than those giving birth to larger infants.
There are several issues with the conclusion that the authors made in their article that I would like to address. The first is that the maternal total homocysteine (t-Hcy) was measured within 48 h postpartum. This measurement does not reflect the actual t-Hcy during pregnancy, because t-Hcy may have been decreased or increased during the 48 h after labor. A proper investigation should include specimen collection at various points during pregnancy (first, second, and third trimesters), immediately after labor, at least 1 week postpartum, and ideally even before pregnancy to determine whether the difference between the obtained t-Hcy values is meaningful and statistically significant. Without these data, conclusions and arguments such as those made by Infante-Rivard et al. (1) in their report are not valid. These investigators previously published a reference range study (2) for t-Hcy for maternal blood; however; that study suffers from the same problem of inappropriate sampling time.
Infante-Rivard et al. (1) indicated that, contrary to their hypothesis, the probability of a mother giving birth to a baby with growth restriction decreased with increasing t-Hcy; i.e., birthweight increased with t-Hcy concentrations. There were no conclusive statistical data included in their communication showing that the values for t-Hcy were significantly different between control infants and cases or between their mothers. Although the mean t-Hcy of 5.11 µmol/L [confidence interval (CI), 4.95–5.26 µmol/L; range, 1.76–14.03 µmol/L] for case mothers in Table 2 of their report (1) seems different from the mean t-Hcy of 5.59 µmol/L (CI, 5.41–5.76 µmol/L; range, 1.92–15.98 µmol/L) for control mothers, it does not say much about the individuals at risk. It would have been more interesting to see the distribution of the results and their quartiles in boxplots so that the reader could see whether there is a significant overlap between the data in the two groups. The confidence interval for the mean is calculated based on "mean ± 2 SE", and although this conveys that the two populations are different, it does not provide additional information on the individuals at risk. In the first paragraph of their discussion, Infante-Rivard et al. (1) claimed that the results for Hcy in the newborns were in the same direction as in the mothers, but it is hard to believe that the mean of 4.99 µmol/L (CI, 4.84–5.15 µmol/L; range, 1.03–17.94 µmol/L) reported in Table 2 for newborn cases is statistically different from the mean of 5.06 µmol/L (CI, 4.92–5.21 µmol/L; range, 0.73–15.62 µmol/L) for the newborn controls, particularly when the analytical CV was 10%. Again, presentation of the data in a boxplot would have been of great help in illustrating the trends.
The authors raised the question of what factors could explain their unexpected findings. They indicated that because the range of t-Hcy values fell below the cutoff for mild hyperhomocysteinemia, the postulated atherothrombotic effects (3) in that range (<15 µmol/L) may be either altered or nonexistent. They provided an alternative theory by referencing the report of Zappacosta et al. (4), in which the authors state that low in vitro concentrations of Hcy do not act as a prooxidant but show an antioxidant effect. One should note that there may be a third mechanism worth mentioning. Because all of the t-Hcy values were below the cutoff for mild hyperhomocysteinemia, those values should be considered as within the reference interval. However, individuals with a t-Hcy value closer to the upper reference limit may have had a better diet during pregnancy. Hcy is derived from methionine, an essential amino acid (5), and this could qualify it as a novel nutritional biomarker for risk assessment of intrauterine growth restriction during pregnancy if their conclusion is valid. Obviously this needs to be substantiated further and requires appropriate statistical analysis of the t-Hcy concentrations for the cases and controls. To do this, an appropriate reference range study for t-Hcy is needed in pregnant and nonpregnant women in conjunction with a study of weight gain during pregnancy. This may further support the conclusions of Zappacosta et al. (4) because better nutrition may provide sufficient t-Hcy to work as an antioxidant. Therefore, a higher t-Hcy concentration may mean that the mother of an infant with a higher birthweight had a better diet, which in turn means that there were more nutrients available in the maternal blood to be transferred to the fetus. On the other hand, Haulrik et al. (6) showed that a high-protein, high-methionine diet did not lead to increased Hcy concentrations compared with a low-protein, low-methionine diet in overweight adults. They also showed that Hcy concentrations after a 3-month intervention were inversely associated with vitamin B12 intake and with weight change.
Infante-Rivard et al. (1) claim that their results were comparable with other North American results (7)(8)(9). I did not see any comparability between the results obtained by these authors and the cited references (7)(8)(9). Walker et al. (7) measured Hcy in nonpregnant control women and in healthy pregnant women during the first, second, and third semesters. However, they did not measure postpartum Hcy. Walker et al. (7) also found that Hcy decreased during pregnancy. This is evident from Fig. 2 of their publication (7). Therefore, the results reported by Infante-Rivard et al. (1) are not comparable to those reported by Walker et al. (7), especially because the latter group did not measure Hcy in women after delivery. Malinow et al. (8) hypothesized in their publication that Hcy, an amino acid that is not a constituent of proteins, crosses the maternal-placental-fetal interphases and is taken up by the fetus. To test this hypothesis, Malinow et al. (8) determined the concentration of plasma Hcy in the maternal vein and neonatal umbilical vessels at the time of delivery. Their findings demonstrated a progressive decrease in the concentration of plasma Hcy going from the maternal vein to the umbilical vein and to the umbilical artery. Their data support the hypothesis that Hcy is sequestered by the fetus. In my view this study had a proper design for the time of specimen collection. They sampled the blood at the time of delivery but not 48 h postpartum. Pagan et al. (9) studied serum Hcy in smoking and nonsmoking pregnant women (18–30 weeks of gestation only). They found that the mean (SD) Hcy in smokers [5.7 (3.4) µmol/L] was not different from that in the nonsmokers [4.9 (1.6) µmol/L]. Interestingly, the difference in Hcy concentrations between the two groups was almost the same as the one reported by Infante-Rivard et al. (1); however, Pagan et al. (9) did not report the difference as significant.
In conclusion, additional data are required to support the unexpected relationship observed by Infante-Rivard et al. (1) between plasma Hcy and intrauterine growth restriction. In addition, the roles of vitamin B12 and folic acid should also be investigated.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
