|
Glucocorticoid-induced osteoporosis: a systematic review and cost-utility analysis |
Kanis J A, Stevenson M, McCloskey E V, Davis S, Lloyd-Jones M |
|
|
CRD summary Extensive searching revealed only small individual data sets and the review was largely inconclusive. The conclusion that there was insufficient evidence to judge the clinical effectiveness of most interventions for glucocorticoid-induced osteoporosis was reasonable, but the conclusion that risedronate was cost effective in a limited population was based on very sparse effectiveness data and should be viewed with caution. Authors' objectives To assess the clinical effectiveness and cost effectiveness of pharmacological interventions for the prevention and treatment of glucocorticoid-induced osteoporosis. Searching MEDLINE, PreMEDLINE, EMBASE, BIOSIS Previews, CINAHL, The Cochrane Library, EMB Reviews, HEED, DARE, NHS EED, NHS HTA, Science Citation Index and Social Science Citation Index were searched from inception. The date on which the original searches were conducted was not reported, but update searches were conducted in The Cochrane Library, MEDLINE and EMBASE in October 2002. No language restrictions were applied. The bibliographies of relevant articles were checked for additional studies. A range of other sources were searched (listed in full in appendix 3 of the report), including research-funding agencies, academic departments, Royal Colleges, national guidelines agencies and patient groups. Study selection Published RCTs conducted in patients who had been treated with glucocorticoids, irrespective of baseline bone mineral density were eligible for inclusion. The following interventions were eligible for inclusion: bisphosphonates; vitamin D with and without calcium; derivatives of vitamin D; calcitonin; pharmacological doses of calcium; oestrogens and oestrogen-like molecules; anabolic steroids; fluoride salts; thiazide diuretics; selective estrogen receptor modulators (SERMs); testosterone; and parathyroid hormone. Acceptable comparators were placebo, any of the specified interventions and no treatment. Included studies were required to report incident vertebral or non-vertebral fracture or associated effects as outcome measures; continuance and compliance were also assessed.
The mean age of study participants, where reported, ranged from 35 to 72 years. Most studies were in mixed male and female populations. The underlying conditions (reasons for glucocorticoid treatment) varied and included rheumatological, pulmonary, dermatological, gastrointestinal and renal conditions, and myasthenia/myasthenia gravis; pre-trial duration of glucocorticoid treatment and treatment regimens also varied (full details given in Appendix 9 of the report). The osteoporosis treatment regimens and fracture definitions used in the trials also varied considerably (full details given in Appendix 9 of the report).
Title screening was used to eliminate clearly irrelevant citations and then abstracts of all studies that used relevant interventions and populations were screened for inclusion. The authors' did not state how many reviewers were involved in the process. Assessment of study quality The methodological quality of included studies was assessed using a published topic-specific tool (Gillespie et al) to assess adequacy of randomisation, blinding of outcome assessors, handling of withdrawals, comparability of groups at baseline, confirmation of diagnosis of hip or other appendicular skeleton fracture and method of diagnosis of vertebral fracture.
Quality assessment was carried out by one reviewer who was not blinded to the author, institution or journal of publication. Data extraction Data were extracted on the numbers of incident vertebral fractures, incident non-vertebral fractures, incident hip fractures, incident wrist fractures and associated adverse and beneficial effects in each group; relative risks and 95% confidence intervals (CIs) were calculated. Data were also extracted on continuance and compliance, where available.
Data were extracted by one reviewer using standardised forms. Methods of synthesis The results of included studies were described in text and tables grouped by osteoporosis intervention and comparator. Studies were eligible for inclusion in meta-analyses if they reported sufficient data for the calculation of relative risks (fracture incidence in terms of the number of participants suffering fracture, rather than overall fracture rate). It appeared that statistical as well as clinical heterogeneity was taken into consideration when decisions were made on the advisability of pooling data, but no details of an assessment of statistical heterogeneity were provided. A fixed-effect model was used to generate pooled relative risks for groups of studies with common definitions of fracture and objectives (treatment or prevention). Results of the review Forty five RCTs (total number of participants unclear) were included in the review. Included studies assessed the effectiveness of bisphosphonates (one study each of alendronate, clondronate and ibandronate, four studies each of pamidronate and risedronate and 12 studies of etidronate; parathyroid hormone (one study), vitamin D alone or in combination with calcium (two studies), alfacalcidol (five studies), calcitriol (eight studies), calcidiol (two studies), calcitonin (10 studies), oestrogen replacement (one study), fluoride (five studies), and thiazide diuretics (one study) in reducing incident fractures.
Results indicated that the bisphosphonate risedronate 5mg/day (relative risk 0.33, 95% CI: 0.14 to 0.80; three studies) and calcidiol 40μg/day plus calcium 3g/day (relative risk 0.56, 95% CI: 0.39 to 0.80; one study) were effective in reducing the incidence of vertebral fracture compared with placebo. There were no significant effects on non-vertebral fracture, including hip fracture. No other agent had demonstrable efficacy for the reduction of either vertebral or non-vertebral fracture; data were often insufficient.
Data on adverse events, continuance and compliance were reported in the text, where available. No significant issues were noted by the authors. Cost information Economic modelling suggested that treatment with risedronate would be cost-effective in patients of 75 years of age and over who were receiving long-term glucocorticoid therapy, irrespective of bone mineral density. In younger patients without prior fracture, treatment would be cost-effective for those with a bone mineral density T-score of -2.0SD or less. Authors' conclusions Risedronate was considered to be cost effective for the management of glucocorticoid-induced osteoporosis, but only at extremes of age and T-score, such that less than 50% of patients aged 50 years or over would be eligible for treatment. Greater cost effectiveness was observed when the effects of bisphosphonates in glucocorticoid-induced osteoporosis were assumed to be similar to those seen in post-menopausal osteoporosis. Further research was needed on the clinical and cost-effectiveness of other interventions for glucocorticoid-induced osteoporosis and on adverse events associated with treatment. CRD commentary The review defined a clear research question using appropriate inclusion criteria. Despite extensive searching and the inclusion of a relatively large number of studies, individual data sets were small and the findings of the review were largely inconclusive. The exclusion of unpublished studies and consequent potential for publication bias was acknowledged by the authors. The review process was conducted by a single researcher, which increased the potential for error and/or bias. An appropriate assessment was made of the methodological quality of included studies and the results were tabulated in appendices to the report, but little use was made of quality assessment in the interpretation of results. Meta-analyses, where applied, were appropriate. The full reporting of the systematic review of clinical effectiveness in both the main body of the report and Appendix 1 did not aid clarity. The authors' conclusions were reasonable in respect of interventions for which there was insufficient evidence. However, conclusions on the cost-effectiveness of risedronate were based on very limited effectiveness data and, as such, should be viewed with caution. Implications of the review for practice and research Practice: The authors did not state any recommendations for practice.
Research: The authors stated that further primary research was required on the utilities and side effects of treatments for glucocorticoid-induced osteoporosis. Health economic assessments based on the probability of fracture and that included independent risk factors as well as the T-score were needed. Secondary research was recommended to evaluate the impact of all vertebral fractures (rather than clinically overt vertebral fractures) on cost effectiveness and methods of treatment monitoring. Funding NHS R&D Health Technology Assessment Programme. Bibliographic details Kanis J A, Stevenson M, McCloskey E V, Davis S, Lloyd-Jones M. Glucocorticoid-induced osteoporosis: a systematic review and cost-utility analysis. Health Technology Assessment 2007; 11(7): 1-256 Indexing Status Subject indexing assigned by NLM MeSH Aged; Aged, 80 and over; Body Mass Index; Female; Glucocorticoids /adverse effects; Great Britain; Humans; Male; Middle Aged; Osteoporosis, Postmenopausal /chemically induced /economics; Randomized Controlled Trials as Topic AccessionNumber 12007008407 Date bibliographic record published 02/03/2009 Date abstract record published 26/08/2009 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. |
|
|
|