|
Maternal exercise and growth in breastfed infants: a meta-analysis of randomized controlled trials |
Daley AJ, Thomas A, Cooper H, Fitzpatrick H, McDonald C, Moore H, Rooney R, Deeks JJ |
|
|
CRD summary The authors concluded that mothers could exercise and breast-feed without reducing their infant's growth, but this was based on limited evidence and required further research. This was a well-conducted review and the authors' tentative conclusion and recommendations for further research appear to be reliable. Authors' objectives To evaluate the evidence on the effects of maternal exercise on the growth of breast-fed infants. Searching Five databases (including PubMed) were searched to June 2011, without date or language restrictions. Search terms were reported. Three clinical trial registers were searched for ongoing and completed trials. Relevant review articles and reference lists of retrieved studies were searched to locate further studies. Study selection Eligible for inclusion were randomised controlled trials (RCTs) that compared any type of exercise intervention, with other or no treatment, in women exclusively or predominantly breast-feeding their infant. Trials could be published or unpublished, and those of exercise, with other interventions, were eligible. Exercise interventions had to be delivered for a minimum of seven days, and data on infant weight or length gain had to be reported. The primary outcome was infant weight gain; other outcomes were infant gain in length, maternal body weight, and adherence (defined in the paper). All included trials were conducted in the USA and were published between 1994 and 2009. All women were recruited between three and 16 weeks after giving birth and were exclusively breast-feeding at baseline. Most were sedentary and most had delivered healthy infants at term. Exercise interventions, in most trials, consisted of moderate-to-high intensity aerobic exercise, three-to-five times per week; interventions lasted from 11 days to 16 weeks. Half of the trials included exercise as a co-intervention with energy intake restriction. Comparators were control groups or groups undertaking diet or energy restriction only. One reviewer screened titles and abstracts of retrieved articles; two reviewers independently screened the full papers and selected trials for inclusion. Assessment of study quality A modified version of the Delphi List was used to assess trial quality for randomisation, baseline comparability, eligibility criteria, blinding of outcome assessors, allocation concealment, reporting of outcomes, and intention-to-treat analysis. The maximum score was seven, indicating high quality. Two reviewers independently assessed quality; any discrepancies were resolved by consensus with a third reviewer. Data extraction The data were extracted by two reviewers independently to calculate mean differences and 95% confidence intervals. Where necessary, infant weight gain was converted from kilos or pounds to grams. In trials with multiple follow-up times, the data were extracted for the final assessment point. Trial authors were contacted for any missing data. Methods of synthesis The effect estimates were pooled using a fixed-effect or random-effects model, according to the level of heterogeneity, assessed using Ι². One three-arm trial assessed one control group and one diet only group, compared with one exercise intervention group; this was treated as two separate comparisons, and the sample size of the intervention group was halved to avoid double counting. For infant weight gain, the meta-analysis weighted trials by sample size; for all other meta-analyses, the generic inverse-variance method was used. Results of the review Four RCTs (five comparisons; 170 participants) were included in the review and meta-analysis (160 participants). None of the trials reported adequate blinding of outcome assessors or allocation concealment, and only one included an intention-to-treat analysis. All other quality criteria were adequately described across the trials. The total quality scores ranged from 4 to 5 (out of 7). Two trials reported infant weight gain at 14 weeks, or at 18 to 20 weeks, after birth. The other two did not explicitly report the follow-up times for this outcome. No statistically significant difference in infant weight gain was found between exercise interventions and their comparators (MD 18.6g, 95% CI -113.52 to 150.80; five comparisons; Ι²=0). Similar results were shown in sensitivity analyses removing the trial with the shortest intervention (MD 8.52g, 95% CI -211.33 to 228.38), and removing trials with additional co-interventions (MD 33.07g, 95% CI -323.88 to 257.74). One trial reported a statistically non-significant difference in infant gain in length at 10 weeks, between a diet plus exercise group and a control group. Maternal weight gain was statistically significantly lower for exercise intervention participants, compared with controls (MD 1.40kg, 95% CI -3.05 to -0.26; three RCTs, four comparisons). Significant statistical heterogeneity was shown (Ι²=96%). Adherence to exercise intervention guidelines, and attendance at exercise classes, were reported to be very good in all of the included trials. Authors' conclusions Mothers could exercise and breast-feed without reducing the growth of their infants, but the evidence was limited and further research was needed. CRD commentary The review question was clear and the inclusion criteria were well defined. Several relevant databases were searched, and efforts were made to locate grey literature. There were no restrictions on date, language, or publication status, reducing the risk of relevant trials being missed. Efforts were made to minimise reviewer error and bias during study selection and quality assessment, and apparently for data extraction. The quality criteria seem to have been suitable, and all the trials were adequate for more than half the criteria. Trial details were reported and the methods of synthesis seem to have been appropriate. Attempts were made to explore clinical and statistical heterogeneity in sensitivity analyses, with no substantial changes to the pooled results, but the trials all had small samples, limiting the generalisability of the findings to the general population. This was a well-conducted review and the authors’ tentative conclusion and recommendations for further research appear to be reliable. Implications of the review for practice and research Practice: The authors did not state any implications for clinical practice. Research: The authors stated that a large high-quality trial was required to confirm the evidence on infant weight gain. A trial investigating infant body length, breast-feeding frequency and duration, infant fussiness, milk volume, and milk composition was recommended. Funding No external funding received. Bibliographic details Daley AJ, Thomas A, Cooper H, Fitzpatrick H, McDonald C, Moore H, Rooney R, Deeks JJ. Maternal exercise and growth in breastfed infants: a meta-analysis of randomized controlled trials. Pediatrics 2012; 130(1): 108-114 Indexing Status Subject indexing assigned by NLM MeSH Body Height; Breast Feeding; Child Development; Exercise; Female; Humans; Infant; Infant, Newborn; Maternal Behavior; Weight Gain; Weight Loss AccessionNumber 12012034578 Date bibliographic record published 22/11/2012 Date abstract record published 05/03/2013 Record Status This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn. |
|
|
|