The review concluded that treatment with fluoride increased spine and hip bone mass density. A low fluoride dose (20 mg per day or less of fluoride equivalents) was associated with a significant reduction in fracture risk. Given several limitations of the review the authors’ conclusions should be treated with caution.

To evaluate the effectiveness and safety of treatment with fluoride on bone mineral density and fracture risk.

EMBASE, PubMed and ISI Web of Science databases were searched to September 2006. Search terms were reported.

Randomised controlled trials (RCTs) evaluating all formulations of fluoride compared to placebo, calcium and/or vitamin D or other anti-resorptive drugs for the treatment of adults (older than 18 years) were eligible for inclusion if the trial duration was six months or more. The outcomes of interest were fracture occurrence or changes in bone mineral density (BMD). To be included, studies had to measure BMD at the lumbar spine or femoral neck on contemporary DEXA-scanners. Spine fractures had to be verified on X-rays. The review also assessed side effects. Studies on secondary osteoporosis and glucocorticoid-induced osteoporosis were eligible for inclusion.

The primary studies evaluated a variety of different formulations of fluoride (including sodium fluoride, slow release fluoride and monofluorophosphate) different doses of fluoride (≤20 mg and >20 mg) and combinations of treatments (regimens included hormone replacement therapy, etidronate, calcium and vitamin D). There were also differences in the underlying disease of patients (details reported in review). Age of participants ranged from 18 years to 76 years (when reported). Duration of treatment ranged from 12 months to 56 months.

The authors state neither how papers were selected for the review nor how many reviewers performed the selection.

It appeared that validity was assessed using the Jadad scale (maximum score of 5 points). The authors stated neither how papers were assessed for validity nor how many reviewers performed the validity assessment.

Data were extracted on number of participants with one or more vertebral or non-vertebral fractures during follow-up and lumbar spine and hip area BMD. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for each outcome. The authors stated neither how data were extracted for the review nor how many reviewers performed the data extraction.

For continuous outcomes including BMD the weighted mean difference (WMD) was calculated. A fixed-effect model was used for homogenous outcomes. A random-effects model was utilised for heterogeneous outcomes. Heterogeneity was assessed using the Χ² test. Meta-regression analysis and subgroup analyses were used to examine the influence of fluoride formulation, dose and treatment duration. Publication bias was assessed using funnel plots, Egger's regression method and the trim and fill method.

Twenty five RCTs (n=2,348) were included in the review: 12 RCTs scored 5 points on the Jadad scale; two scored 4 points; 10 scored 3 points; and one scored 2 points.

Effects on BMD: Spine BMD increased by 7.9 per cent (95% CI: 5.4%, 10.5%, p<0.001, n=1,774) and hip BMD by 2.1 per cent (95% CI: 0.9%, 3.4%, p<0.01, n=1,434) after treatment with fluoride, but with evidence of significant heterogeneity (p<0.01).

Meta-regression analysis showed an increase in spine BMD with increasing duration of treatment (5.04 +/- 2.16 per cent per year of treatment).

Effects on fractures: There were no statistically significant effects of fluoride treatment on the risk of vertebral or non-vertebral fracture risk, but with evidence of significant heterogeneity (p<0.01).

Meta-regression analysis for non-vertebral fracture risk showed an increase of fracture risk with increasing fluoride dose (OR 0.14 per mg, p<0.01). There was no effect of treatment duration. Sub-group analysis showed that with a daily dose of 20 mg or less of fluoride equivalents there was a statistically significant reduction for both vertebral fracture risk (OR 0.28; 95% CI: 0.09, 0.87, six studies, n=593) and for non-vertebral fracture risk (OR 0.52; 95% CI: 0.28, 0.76, six studies, n=768). A daily dose of 20 mg or more of fluoride equivalents showed a non-statistically significant increase in the risk of both vertebral fractures (OR 1.26; 95% CI: 0.78, 2.04, 11 studies) and non-vertebral fracture (OR 1.46, 95% CI: 0.77, 2.76, seven studies).

Funnel plots showed no signs of significant publication bias.

Adverse events
There were no significant differences for the frequency of gastrointestinal symptoms between fluoride and control groups (14 studies). There was a significantly higher risk of pain for fluoride treated groups compared to controls (OR 2.76; 95% CI: 1.35, 5.65). There were similar increases in pain with doses of 20 mg fluoride or less and doses of 20 mg or more.

Treatment with fluoride increased spine and hip BMD, dependent on treatment duration. Overall there was no evidence of effect on hip or spine fracture risk. However, subgroup analyses found that a low fluoride dose (20 mg per day or less of fluoride equivalents) was associated with a significant reduction in fracture risk.

Inclusion criteria were specified for study design and outcomes. Criteria for intervention were somewhat broad since combinations of various treatments were permitted and no criteria were defined for participants. Some relevant sources were searched, but no attempts were made to minimise language bias. No attempts were made to locate unpublished studies, but appropriate methods were used to assess publication bias and no evidence of its presence was found. Methods used to select studies, assess validity and extract data were not described, so it was unknown whether efforts were made to reduce reviewer errors and bias. Validity was assessed using an established checklist, although only the composite score was presented making it difficult to judge study validity. Combining the studies in a meta-analysis may not have been appropriate due to the considerable differences between studies. The authors concluded that treatment with fluoride increased spine and hip BMD, dependent on treatment duration. Given the identified weaknesses of the review the authors’ conclusions should be treated with caution.

Practice: The authors did not state any implications for practice.

Research: The authors appeared to state that additional studies were required to evaluate the effects of fluoride combined with vitamin D treatment.

Laura og Jens Veng Christensens Foundation and Bagermester August H. Jenson og Hustrus Legat.

Haguenauer D, Welch V, Shea B et al. Fluoride of the treatment of osteo-postmenopausal osteoporotic fractures: a meta analysis. Osteoporos Int 2000;11:727-738.

Haguenauer D, Welch V, Shea B, Tugwell P, Wells G. Fluoride for treating postmenopausal osteoporosis. Cochrane Database Syst Rev CD002825.

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.