Effects of restricted caffeine intake by mother on fetal, neonatal and pregnancy outcomes

Caffeine is a stimulant found in tea, coffee, cola, chocolate and some over-the-counter medicines. Conflicting results found in the literature make it difficult for health professionals to advise pregnant women about avoiding caffeine during pregnancy. Clearance of caffeine from the mother's blood slows down during pregnancy. Some authors of observational studies have concluded that caffeine intake is harmful to the fetus, causing growth restriction, reduced birthweight, preterm birth or stillbirth. The newborn could also have withdrawal symptoms if the mother has a high intake of caffeine (more than eight cups of coffee per day).

Two studies met the inclusion criteria but only one contributed data to the outcomes of interest. The study was based in Denmark. Women less than 20 weeks pregnant were randomly assigned to drinking caffeinated instant coffee (568 women after exclusions) or decaffeinated instant coffee (629 women). Drinking three cups of coffee a day in early pregnancy had no effect on birthweight, preterm births or growth restriction.

Both included studies were randomised controlled trials. One randomly allocated pregnant women to either caffeinated or decaffeinated groups. It was unclear from the other whether allocation concealment was undertaken. Blinding of personnel and study participants was satisfactory in both studies while blinding of outcome assessor was not clearly stated. Attrition bias was also not clearly explained in one study. The results from the one trial that provided data for analysis showed that there was no evidence of an effect of caffeine avoidance on the outcomes birthweight, preterm birth or small-for-gestational age.

Two outcomes were assessed and assigned a quality rating using the GRADE methods. Evidence for these two outcomes, namely birthweight and frequency of preterm birth, was assessed as of low quality, with downgrading decisions due in part to the relatively small sample sizes and the wide confidence interval of the one included trial that contributed data.

There is insufficient evidence to confirm or refute the effectiveness of caffeine avoidance on birthweight or other pregnancy outcomes.

Authors' conclusions: 

There is insufficient evidence to confirm or refute the effectiveness of caffeine avoidance on birthweight or other pregnancy outcomes. There is a need to conduct high-quality, double-blinded RCTs to determine whether caffeine has any effect on pregnancy outcome.

Read the full abstract...
Background: 

Maternal caffeine consumption during pregnancy may have adverse effects on fetal, neonatal and maternal outcomes.

Objectives: 

This review investigates the effects of restricting caffeine intake by mothers on fetal, neonatal and pregnancy outcomes.

Search strategy: 

We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (16 January 2015), scanned bibliographies of published studies and corresponded with investigators.

Selection criteria: 

Randomised controlled trials (RCTs) including quasi-RCTs investigating the effect of caffeine and/or supplementary caffeine versus restricted caffeine intake or placebo on pregnancy outcomes.

Data collection and analysis: 

Two review authors independently assessed trials for inclusion and risk of bias, extracted data and checked them for accuracy.

Main results: 

Two studies met the inclusion criteria but only one contributed data for the prespecified outcomes. Caffeinated instant coffee (568 women) was compared with decaffeinated instant coffee (629 women) and it was found that reducing the caffeine intake of regular coffee drinkers (3+ cups/day) during the second and third trimester by an average of 182 mg/day did not affect birthweight (g) (mean difference (MD) 20.00, 95% confidence interval (CI) -48.68 to 88.68; one study, 1197 participants; low quality evidence), preterm birth (risk ratio (RR) 0.81, 95% CI 0.48 to 1.37; one study, 1153 participants; low quality evidence) or small-for-gestational age (RR 0.97, 95% 0.57 to 1.64; one study, 1150 participants). Risk of bias was moderate in both studies.

Two outcomes were assessed and assigned a quality rating using the GRADE methods. Evidence for these two outcomes (birthweight and preterm birth) was assessed as of low quality, with downgrading decisions due to the relatively small sample sizes and the wide confidence interval of the one included trial that contributed data. Neither of the studies reported on any of the other primary outcomes (low birthweight; first trimester fetal loss; perinatal mortality; fetal hypoxia; fetal tachycardia) or on any of the reviews neonatal or maternal outcomes.