In our study, supplementation with vitamins C and E did not reduce the frequency of the primary outcome or any of its components. We chose the primary outcome of new-onset pregnancy-associated hypertension with evidence of maternal, fetal, or neonatal complications, rather than the diagnosis of preeclampsia, so that we could assess whether therapy would prevent serious complications rather than merely modify diagnostic findings. We did not require the presence of proteinuria as part of the primary outcome, since severe hypertension without proteinuria can be associated with adverse maternal and fetal outcomes.20
Furthermore, a diagnosis of proteinuria that is based on the qualitative assessment of random urine samples or even on protein-to-creatinine ratios cannot be compared with a diagnosis that is based on 24-hour urine collections.21–24
Since our primary outcome was not based on a conventional diagnosis of preeclampsia, we used preeclampsia as a major secondary outcome. The rates of mild preeclampsia, severe preeclampsia, the HELLP syndrome, and eclampsia were not significantly affected by vitamin treatment. There was also no evidence of a benefit with vitamin therapy with respect to any of the other prespecified secondary outcomes.
Several other trials have examined the effectiveness of vitamins C and E in preventing preeclampsia. 11–14
The doses of vitamins that were used in those trials were the same as those used in our trial. Each of the previous trials had a smaller sample than that in our study; only one of the previous trials included low-risk subjects.13
In no other study was therapy initiated as early as it was in our trial. The Vitamins in Pre-eclampsia study (VIP; Current Controlled Trials number, ISRCTN62368611) of antioxidant supplementation to prevent preeclampsia in high-risk women showed that there was an increased rate of complications among women and infants when women received antioxidant vitamins during pregnancy. 14
In that study, there were significantly more babies with low birth weight (a prespecified outcome) in the group that received vitamin supplementation than in the control group; gestational hypertension and the use of antihypertensive therapy were also more common in the vitamin group. Although the rates of perinatal death were similar in the two groups, post hoc analyses showed that the stillbirth rate was higher (and the neonatal death rate lower) in the vitamin group than in the control group. In contrast, we did not find significant between-group differences in the rates of low birth weight and stillbirth. We did find an increase in the frequency of gestational hypertension in the vitamin group, as compared with the placebo group. We did not collect data on the use of antihypertensive medications, but a significant increase in the use of antihypertensive therapy was found with vitamin supplementation in the previous trial involving low-risk women.13
Why was therapy with these antioxidant vitamins not successful in altering the outcome of hypertension in pregnancy in our study or in previous studies? It is, of course, possible that although oxidative stress is present in preeclampsia, it is not important in the pathophysiology of the condition. Alternatively, oxidative stress may be relevant to pathogenesis in only a subgroup of women, with no appreciable benefit of antioxidants for the overall population. It has been suggested, in relation to other trials, that the women may already have had adequate concentrations of vitamins C and E before therapy.25
In our study, almost 80% of the women were taking prenatal vitamins that contained an average of 100 mg of vitamin C and 22 IU of alpha-tocopherol when they entered the study. Doses of ascorbate of 150 mg per day result in nearly maximal plasma and tissue concentrations. Administration of 1000 mg per day increases the plasma concentration by only 25% above that achieved with 150 mg per day.26
The doses of alpha-tocopherol used in this study, however, would be expected to increase plasma concentrations substantially above those achieved with prenatal vitamins.26
The possibility that therapy might be effective specifically in women who are deficient in these vitamins was not supported by the World Health Organization study (ISRCTN86677348) of vitamin C and E supplementation, which showed that supplementation with these vitamins, as compared with placebo, did not reduce the risk of preeclampsia in a high-risk and nutritionally deficient population (relative risk with vitamins, 1.0; 95% CI, 0.9 to 1.3).11
The doses of vitamins C and E used in our trial were determined on the basis of the preliminary study, in which these doses not only appeared to reduce the frequency of preeclampsia but also decreased objective evidence of oxidative stress.10
The timing of the administration of antioxidants is also important. The antioxidants need to be present at the time of a relevant pro-oxidant challenge. Burton and Jauniaux found that the initiation of intervillous blood flow at 8 to 10 weeks of gestation was associated with a burst of oxidative stress.27
Antioxidant therapy in our study was initiated in the 9th to 16th week of pregnancy, with 44% of women beginning treatment before the 13th week of pregnancy. In a post hoc subgroup analysis limited to women treated before the 13th week of pregnancy, there was no apparent benefit of vitamin supplementation. We also cannot be certain that other antioxidants would not have been effective.
In summary, supplementation with vitamin C (at a dose of 1000 mg daily) and vitamin E (at a dose of 400 IU daily) did not reduce the rates of either serious adverse outcomes of pregnancy-associated hypertension or preeclampsia among low-risk, nulliparous women. Previous studies have shown a similar lack of efficacy among high-risk women and among women who were likely to have had a deficiency of vitamins C and E. The findings of these several studies provide no support for the use of vitamin C and E supplementation in pregnancy to reduce the risk of preeclampsia or its complications.