A Cochrane review included 5 studies with a total of 3130 subjects. All women were at low risk of potentially requiring a general anaesthesia. The comparisons were complete restriction vs freedom to eat and drink in 1 trial, water only vs specific fluids and foods in 2 trials, and water only vs carbohydrate drinks in 2 trials. When comparing any restriction of fluids and food vs some nutrition during active labour, the meta-analysis was dominated by one study undertaken in a highly medicalised environment (n=2400). There were no statistically significant differences in caesarean section (average risk ratio (RR) 0.89, 95% CI 0.63 to 1.25; 5 studies, n=3103), operative vaginal births (average RR 0.98, 95% CI 0.88 to 1.10; 5 studies, n=3103) and Apgar scores less than 7 at five minutes (average RR 1.43, 95% CI 0.77 to 2.68; 3 studies, n=2574), nor in any of the other outcomes assessed. Women's views were not assessed. No women suffered from regurgitation during general anaesthesia. However, the data were insufficient to assess the incidence of Mendelson's syndrome, an extremely rare outcome (increased risk of the stomach contents entering the lungs during general anaesthesia). Other comparisons showed similar findings, except an increased risk of caesarean sections in one study (n=88) for women taking carbohydrate drinks compared with water only.
Comment: The quality of evidence is downgraded by study quality (lack of blinding of mothers and midwives, in the dominating trial 20 % in the water group ate and 29 % in the food group did not eat), and by imprecise results for aspiration pneumonia.