Antioxidants and preterm prelabour rupture of the membranes Fiona Mathews, a Andrew Neil b Objective To investigate the relationships of dietary intakes and serum levels of antioxidant nutrients to the risk of prelabour rupture of the membranes preterm (PPROM). Design A prospective observational cohort study. Setting A district general hospital in the south of England. Population Eight hundred and twelve white nulliparous women with singleton pregnancies. Methods Serum antioxidant concentrations and dietary intakes were measured in the second trimester. Associations with the risk of PPROM were investigated. Main outcome measure Preterm prelabour rupture of the membranes. Result Serum concentrations of the antioxidant lutein, but no other nutrient, was associated with the risk of PPROM. Maternal smoke exposure (serum cotinine) also predicted PPROM. Women with high serum lutein concentrations (top third) in early pregnancy had four times greater risk than women in the lowest third (95% CI 1.3 – 11.9; P ¼ 0.009) after adjusting for smoke exposure. Similar results were obtained using lutein levels from later pregnancy. Conclusion We found no evidence that antioxidant nutrients are protective against PPROM. Rather, high levels of lutein were associated with an increased risk. INTRODUCTION Preterm birth is a prime cause of perinatal morbidity and mortality. Rupture of the membranes (ROM) prelabour is the most common antecedent, occurring in 30–40% of cases. 1,2 Prelabour rupture of the membranes preterm (PPROM) affects between 2.0% and 3.5% of pregnancies, 2 yet it has been the subject of few epidemiological studies, and there is debate about the identified risk factors. A higher incidence of PPROM in smokers has been reported by some studies, 3,4 while others have found the elevated risk limited to heavy smokers, 5,6 and some have been able to demonstrate no association at all. 7,8 Similarly, previous termination of pregnancy has sometimes, 9 but not always, 3,4 been found to be an independent risk factor. Previous preterm delivery 3 and consumption of more than three cups of coffee per day 4 have also been associated with PPROM. The aetiology of PPROM remains the subject of specu- lation. Intrauterine infection has a role in some cases, 10 but ascension of infection from the vagina after membrane rupture is more usual. 5 Damage may be caused to the membranes by substances produced by bacteria in the genital tract, or by the inflammatory response of the mother to these bacteria. 11 Because the amnion is an avascular structure, which depends on the amniotic fluid for its maintenance, nutritional status may also be involved in the pathogenesis of PPROM. 4,12 The hypothesis that in- creased intake of antioxidant nutrients may help to prevent PPROM is currently the subject of a NIH-funded random- ised controlled trial in the United States. In vitro, vitamins C and E have been shown to protect the chorion and amnion from damage due to reactive oxygen species. 13 Inadequate dietary intake of these antioxidant nutrients has therefore been proposed as a cause of membrane damage in vivo, hence, of PPROM. 12 The proposed mechanism is that vitamin C is required to scavenge oxygen species in the amniotic fluid and vitamin E prevents lipid peroxidation within the membranes. In addition, vitamin C is involved in collagen synthesis, and perturbed collagen composition has been noted in the amnions of patients with PPROM. 14 One study has previously associated low plasma levels of vitamin C with prelabour ROM, but the results of dietary assessments made during this study were not reported. 15 There are no published data on the relationships of diet- ary intake, or plasma levels of lipid-soluble antioxidants, on the risk of PPROM. We have therefore investigated the relationships of nutritional factors to PPROM as part of a wider study of maternal nutrition, smoking and preg- nancy outcome, for which birthweight was the primary outcome variable. 16 – 19 The study of PPROM was spec- ified in our protocol at the outset of the study as being BJOG: an International Journal of Obstetrics and Gynaecology May 2005, Vol. 112, pp. 588–594 D RCOG 2005 BJOG: an International Journal of Obstetrics and Gynaecology www.blackwellpublishing.com/bjog a Department of Zoology, University of Oxford, UK b Division of Public Health and Primary Health Care, University of Oxford, Institute of Health Sciences, UK Correspondence: Dr F. Mathews, Department of Zoology, University of Oxford, South Parks Road, Oxford, OX1 3PS, UK. DOI:10.1111/j.1471-0528.2005.00500.x