Thirty-three studies were identified, of which 32 were located and reviewed. The total number of women included in the studies was not stated.
1. Zinc (28 studies, of which 5 were discarded and only 4 were intervention studies).
The 4 RCTs did not provide good evidence for an increase in birth weight when a zinc supplement was used, or for a reduction in the incidence of low birth weight babies. There was some evidence to suggest that zinc supplements can reduce the incidence of intra-uterine growth retardation in women identified as being at risk of this outcome, but further research is needed before conclusions can be drawn with confidence. No significant differences in mean gestational age at delivery, or in the number of pre-term births in the women taking zinc supplements or placebo, were found in 2 RCTs. However, a third RCT (graded B) found a significant relationship between zinc supplementation and reduced rates of pre-term delivery.
Evidence from cohort and case-control studies indicated that low leucocyte or plasma zinc levels were associated with increased rates of low birth weight and lowered mean birth weight. The balance of evidence currently points to an association between low leucocyte zinc and intra-uterine growth retardation, though this finding has not been supported by work on serum or plasma levels.
2. Vitamin C (5 studies, of which one was discarded).
One study investigated the effect of nutritional counselling on the outcome of pregnancy in rural Greece. However, while biochemical measurements were made of vitamin C and other nutrients, the analysis did not separate out the contribution of individual nutrients to the improved pregnancy outcomes seen with the counselling.
None of the cohort studies showed an association between the amount of vitamin C consumed and the outcome of pregnancy. However, most of the studies had insufficient power to detect small differences.
3. Carotenoids and carotene (2 cohort studies).
No intervention trials were found.
The available data from observational studies were limited, but suggested that in smokers, serum carotene may have an important association with birth weight.
4. Vitamin E (2 cohort studies).
No intervention trials were found.
Of the cohort studies, there was insufficient work of good quality to allow any conclusions to be drawn about the effect of this nutrient.
5. Iron supplements (6 studies).
One intervention study (n=2,912) comparing routine iron supplementation with non-routine treatment found that the mean gestational age was shorter in the infants on the non-routinely supplemented group, although this only reached statistical significance among boys. No differences were seen between the groups in terms of the mean birth weight or placental weight.
One good cohort study indicated that clinically important risks of low birth weight and pre-term delivery were associated with iron deficiency anaemia, as opposed to anaemia from other causes which was associated with no increased risk. In addition, the study indicated that deficiency affects outcome early in pregnancy. The other studies also indicated that, at the other end of the scale, high haematocrit and/or haemoglobin levels were associated with poor outcomes of pregnancy such as pre-term delivery, low birth weight and intra-uterine growth retardation.
6. Selenium studies.
Two studies found no correlation between maternal plasma or serum selenium and the mean birth weight, gestational age, pre-term delivery or intra-uterine growth retardation.