|
Side effects of phytoestrogens: a meta-analysis of randomized trials |
Tempfer CB, Froese G, Heinze G, Bentz EK, Hefler LA, Huber JC |
|
|
CRD summary This review concluded that phytoestrogen supplementation had a safe side-effect profile with moderately increased rates of gastrointestinal side-effects when compared with placebo or no treatment, in women. Supplementation over a period of two years was recommended. This conclusion reflected the evidence presented, but weaknesses in the review methods might limit the reliability of the findings. Authors' objectives To ascertain the incidence of side-effects and adverse occurrences associated with phytoestrogen supplementation in women. Searching PubMed and the Cochrane Central Register of Controlled Trials (CENTRAL) were searched for articles in English. Search terms were reported, but the dates were not. Study selection Randomised controlled trials (RCTs) assessing the effects of treatment with phytoestrogens (defined as substances containing a defined amount of isoflavones, coumestans, or lignans), compared with placebo or no treatment, in women, were eligible for inclusion. Trials evaluating women and men were eligible if they included subgroup analyses on women.
Across the included trials there was variation in their inclusion criteria, sample sizes (range seven to 389), and, where reported, the mean ages of participants (26.3 to 70.6 years), mean body mass indices (21.3 to 32.2), trial durations (one day to five years, median 6.2 months), control group designs, sources of phytoestrogens (compound used), and doses. Multiple studies assessing the same population were included only once (earliest publication date). Two reviewers selected trials for inclusion in the review. Assessment of study quality Trial quality was assessed according to adequacy of treatment allocation concealment, blinding of participants and outcome assessors to treatment, use of intention-to-treat analyses, and the number of participants lost to follow-up or withdrawals from the trial.
The authors did not state how the validity assessment was performed. Data extraction The incidence of side-effects and adverse occurrences in the treatment and control groups were extracted. An incident rate ratio (number of side-effects divided by person-time) was computed for each category of side-effect in the control and treatment groups. Events were counted per side-effect and in crossover designs per side-effect in all the treatment groups. To adjust for the trials that reported no side-effects 0.5 was added to the number of side-effects and one was added to the number of patients in each group.
Two reviewers extracted data for inclusion in the review. Methods of synthesis Reported side-effects and adverse events were categorised into six groups: gynaecological or urinary, gastrointestinal, musculoskeletal, neurological or sensory, nonspecific or other, and undefined.
Incident rate ratios were combined using a weighted mean of log of the incident rate ratio (with weights given by the reciprocal variance of each trial), using fixed-effect meta-analysis. For specific side-effects exact p values were computed and corrected for multiple testing. The effects of clinical and epidemiological variables, on the occurrence of reported adverse outcomes, were also investigated. Publication bias was explored using funnel plot analyses. Results of the review Ninety-two RCTs (n=9,629 participants) were included in the review (treatment group n=5,502, control group n=4,806, including crossover designs).
Eighty-two trials (89%) had adequate concealment of treatment allocation; 21 (23%) had centralised randomisation. Blinding of both participants and outcome assessors was carried out in 81 trials (88%); outcome assessors were not blinded in 11 trials. In 58 trials (63%) over 10% of participants withdrew or were lost to follow-up and intention-to-treat analysis was performed in 21 trials (23%).
The overall incidence of side-effects was 36.7% (2,019 out of 5,502) in the phytoestrogen groups and 38.0% (1,824 out of 4,806) in the control groups (IRR 1.01, 95% CI 0.95 to 1.08; 92 RCTs). The incident rate ratio for these side-effects was influenced by the regional location of the trial and the participant age, with phytoestrogen-treated women older than 55 years experiencing higher rates. Phytoestrogen users had significantly higher rates of gastrointestinal side-effects (IRR 1.28, 95% CI 1.08 to 1.50; 92 RCTs). Overall and within categories there were no significant differences in the rates of side-effects for gynaecological, musculosketetal, neurological, and unspecified categories, for phytoestrogen and control groups. Funnel plot analysis showed no publication bias for the incidence of side-effects. Authors' conclusions Phytoestrogen supplements had a safe side-effect profile with moderately increased rates of gastrointestinal side-effects, such as abdominal pain, as well as myalgia and sleepiness. They were not associated with increased risks of breast cancer, endometrial cancer, vaginal bleeding, and endometrial hyperplasia. CRD commentary This review had clearly focused inclusion and exclusion criteria for study design, participants, and intervention, with broader criteria for outcomes. The authors searched two relevant databases one of which was CENTRAL which has multiple sources and reduces the risk of trials being missed. Funnel plot analysis showed no evidence of publication bias despite the exclusion of unpublished trials. The search dates were not reported, but the year of publication of included trials ranged from 1995 to 2008. Studies that were not in English were excluded and so language bias cannot be ruled out. It was unclear whether appropriate methods were used to minimise reviewer bias and error at all stages of the review process. Studies were assessed and data were extracted by two reviewers, but it was not reported whether these activities were carried out independently or how disagreements were resolved. Standard methods were used to pool the data. There was no statistical heterogeneity, but the trials might have been too clinically heterogeneous to be pooled.
The authors' conclusions reflected the evidence presented, but the weaknesses in the review methods may limit the reliability of the findings. Implications of the review for practice and research Practice: The authors stated that phytoestrogen supplements had a safe side-effect profile, with moderate increases in gastrointestinal side-effects. They might, therefore, be safe alternative treatments to hormone replacement therapy. The use of phytoestrogens over a period of two years could be recommended, but the side-effects were significantly more often reported as the reason for drop-out when compared with controls. The potential risk of rare side-effects in women on long-term phytoestrogen treatment could not be ruled out.
Research: The authors did not state any implications for research. Bibliographic details Tempfer CB, Froese G, Heinze G, Bentz EK, Hefler LA, Huber JC. Side effects of phytoestrogens: a meta-analysis of randomized trials. American Journal of Medicine 2009; 122(10): 939-946.e9 Indexing Status Subject indexing assigned by NLM MeSH Breast Neoplasms /epidemiology /etiology; Dietary Supplements; Dose-Response Relationship, Drug; Drug Administration Schedule; Drug-Related Side Effects and Adverse Reactions; Endometrial Hyperplasia /epidemiology /etiology; Endometrial Neoplasms /epidemiology /etiology; Estrogen Replacement Therapy /adverse effects /methods; Female; Follow-Up Studies; Gastrointestinal Diseases /epidemiology /etiology; Humans; Incidence; Phytoestrogens /therapeutic use; Plant Extracts; Postmenopause /drug effects; Randomized Controlled Trials as Topic; Risk Assessment; Uterine Hemorrhage /epidemiology /etiology AccessionNumber 12009109300 Date bibliographic record published 10/03/2010 Date abstract record published 05/05/2010 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. |
|
|
|