Baby hopes and vitamins

1 September 2013

CLINICALLY APPRAISED TOPIC (CAT): Does taking vitamin supplements help women with fertility problems?

CLINICAL BOTTOM:

Taking oral antioxidants is unlikely to improve the chances of getting pregnant for subfertile women.

CLINICAL SCENARIO:

You and your partner have been trying for more than one year to get pregnant. You are increasingly worried that you might have fertility problems. Friends have told you that vitamin supplements (antioxidants) may increase your chances of getting pregnant. You decide to have a look at the literature.

QUESTION:

Among women with fertility problems, does taking antioxidant supplements increase rates of pregnancy?

SEARCH STRATEGY:

PubMed clinical queries (Broad): Antioxidants AND subfertility

CITATION:

Showell MG, Brown J, Clark J, Hart RJ. Antioxidants for female subfertility. Cochrane Database Sys Rev 2013, Issue 8. Art No.: CD007807. DOI:10.1002/14651858.CD007807.pub2

STUDY SUMMARY:

Systematic review of oral antioxidants compared to placebo, no treatment, standard treatment or another antioxidant. Inclusion criteria were:

[1] type of study was randomised controlled trials;

[2] Patients were subfertile women attending a reproductive clinic that may or may not be using assisted reproductive technology. Subfertile was defined as couple trying to conceive for more than one year without success;

[3] Interventions: any type of antioxidant supplementation including pentoxifylline used in addition to subfertility treatment;

[4] Outcomes were live birth, rate of pregnancy and adverse effects.

In all 2197 study titles were screened, 67 were retrieved for further review and 28 trials were included.

STUDY VALIDITY:

Search strategy – electronic searches were made of Cochrane Menstrual Disorders and Subfertility Group Specialised trials register, Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE, PsycINFO, CINAHL, LILACS and OpenSIGLE. Conference abstracts and citation lists on ISI Web of Knowledge were searched, reference lists from retrieved articles checked and a Google search conducted. There were no language limitations.

Risk of bias – two reviewers independently screened and selected included studies and extracted data. Disagreements were resolved by discussion or a third reviewer – authors were contacted for additional information if data was not described. Only where clinical heterogeneity was sufficiently low were studies combined in meta-analysis.

Quality assessment – Methodological quality in each study was independently assessed by two reviewers. Elements assessed were randomisation generation, allocation concealment, blinding of intervention, incomplete outcome data reported, selective outcome reporting and other sources of bias.

Publication bias – was assessed using funnel plots.

Overall – a high quality review of generally low quality studies.

RESULTS:

Twenty eight trials involving 3548 women were identified. Nine trial compared antioxidants to placebo, nine trials compared antioxidant to no treatment or standard treatment, two trials compared antioxidant to antioxidant, two trials compared pentoxifylline to placebo, one trial compared pentoxifylline and vitamin E to no treatment, four trials compared antioxidants and a co-intervention with placebo and a co-intervention, and in one trial the control was not specified. Nine trials did not report randomisation sequence, twenty-two trials did not adequately report allocation concealment, Eleven trial studies were not blinded or did not report blinding and eleven trials used incomplete outcome data. Antioxidants did not increase live birth rates or pregnancy rates (Table). Adverse events included miscarriage, multiple pregnancy, gastrointestinal disturbances and ectopic pregnancy. There were no significant differences on each of these types of adverse event.

COMMENTS:

Overall the trials were considered low quality because of the number of smaller studies included, the high risk of bias within the studies, and amount of statistical heterogeneity in the main analyses.

Reviewer:

Dr Andrew Jull, RN PhD, Associate Professor (School of Nursing, University of Auckland), Nurse Advisor – Quality & Safety, Auckland District Health Board.

 

Table. Results with 95% confidence intervals

  Antioxidant Control Odds Ratio
(95% CI)
% heterogeneity
(I2)
Pregnancy
(13 trials,
n=2441)
48%
(23/
48)
37%
18/
49
1.25
(0.19 to
8.26)
 75%
Pregnancy
(13 trials,
n=2441)
25%
(343/
1383)
23%
(244/
1058)
1.30
(0.92 to
1.85)
 55%