Background

Gestational diabetes (GDM) is glucose intolerance, first recognised in pregnancy and usually resolving after birth. GDM is associated with both short- and long-term adverse effects for the mother and her infant. Lifestyle interventions are the primary therapeutic strategy for many women with GDM. Objectives

To evaluate the effects of combined lifestyle interventions with or without pharmacotherapy in treating women with gestational diabetes. Search methods

We searched the Pregnancy and Childbirth Group's Trials Register (14 May 2016), ClinicalTrials.gov, WHO International Clinical Trials Registry Platform (ICTRP) (14th May 2016) and reference lists of retrieved studies. Selection criteria

We included only randomised controlled trials comparing a lifestyle intervention with usual care or another intervention for the treatment of pregnant women with GDM. Quasi-randomised trials were excluded. Cross-over trials were not eligible for inclusion. Women with pre-existing type 1 or type 2 diabetes were excluded. Data collection and analysis

We used standard methodological procedures expected by the Cochrane Collaboration. All selection of studies, data extraction was conducted independently by two review authors. Main results

Fifteen trials (in 45 reports) are included in this review (4501 women, 3768 infants). None of the trials were funded by a conditional grant from a pharmaceutical company. The lifestyle interventions included a wide variety of components such as education, diet, exercise and self-monitoring of blood glucose. The control group included usual antenatal care or diet alone. Using GRADE methodology, the quality of the evidence ranged from high to very low quality. The main reasons for downgrading evidence were inconsistency and risk of bias. We summarised the following data from the important outcomes of this review.

Lifestyle intervention versus control group

For the mother:

There was no clear evidence of a difference between lifestyle intervention and control groups for the risk of hypertensive disorders of pregnancy (pre-eclampsia) (average risk ratio (RR) 0.70; 95% confidence interval (CI) 0.40 to 1.22; four trials, 2796 women; I2 = 79%, Tau2 = 0.23; low-quality evidence); caesarean section (average RR 0.90; 95% CI 0.78 to 1.05; 10 trials, 3545 women; I2 = 48%, Tau2 = 0.02; low-quality evidence); development of type 2 diabetes (up to a maximum of 10 years follow-up) (RR 0.98, 95% CI 0.54 to 1.76; two trials, 486 women; I2 = 16%; low-quality evidence); perineal trauma/tearing (RR 1.04, 95% CI 0.93 to 1.18; one trial, n = 1000 women; moderate-quality evidence) or induction of labour (average RR 1.20, 95% CI 0.99 to 1.46; four trials, n = 2699 women; I2 = 37%; high-quality evidence).

More women in the lifestyle intervention group had met postpartum weight goals one year after birth than in the control group (RR 1.75, 95% CI 1.05 to 2.90; 156 women; one trial, low-quality evidence). Lifestyle interventions were associated with a decrease in the risk of postnatal depression compared with the control group (RR 0.49, 95% CI 0.31 to 0.78; one trial, n = 573 women; low-quality evidence).

For the infant/child/adult:

Lifestyle interventions were associated with a reduction in the risk of being born large-for-gestational age (LGA) (RR 0.60, 95% CI 0.50 to 0.71; six trials, 2994 infants; I2 = 4%; moderate-quality evidence). Birthweight and the incidence of macrosomia were lower in the lifestyle intervention group.

Exposure to the lifestyle intervention was associated with decreased neonatal fat mass compared with the control group (mean difference (MD) -37.30 g, 95% CI -63.97 to -10.63; one trial, 958 infants; low-quality evidence). In childhood, there was no clear evidence of a difference between groups for body mass index (BMI) ≥ 85th percentile (RR 0.91, 95% CI 0.75 to 1.11; three trials, 767 children; I2 = 4%; moderate-quality evidence).

There was no clear evidence of a difference between lifestyle intervention and control groups for the risk of perinatal death (RR 0.09, 95% CI 0.01 to 1.70; two trials, 1988 infants; low-quality evidence). Of 1988 infants, only five events were reported in total in the control group and there were no events in the lifestyle group. There was no clear evidence of a difference between lifestyle intervention and control groups for a composite of serious infant outcome/s (average RR 0.57, 95% CI 0.21 to 1.55; two trials, 1930 infants; I2 = 82%, Tau2 = 0.44; very low-quality evidence) or neonatal hypoglycaemia (average RR 0.99, 95% CI 0.65 to 1.52; six trials, 3000 infants; I2 = 48%, Tau2 = 0.12; moderate-quality evidence).

Diabetes and adiposity in adulthood and neurosensory disability in later childhoodwere not prespecified or reported as outcomes for any of the trials included in this review.