Empirical findings suggest that hyaluronan injections and diacerein are similarly effective. Yet the authors found that monetary savings could be realized when conservative treatment with diacerein is chosen.
Affecting approximately 70% of the population older than age 60 years, osteoarthritis (OA) is a leading cause of disability among adults in the United States.1 Lower productivity, performance, and general satisfaction with self-fullfillment are some of the effects on the workforce of this prevalent chronic disease. Job-related costs, which account for $3.4 billion to $13.2 billion per year, are the most significant costs associated with OA.2
To date, a large number of treatment modalities have been used, but they remain mainly palliative or symptom modifying. According to the Osteoarthritis Research Society International Treatment Guidelines Committee, 51 different treatment modes have been addressed by numerous guidelines, but only 20 have been universally recommended.3 Last updated in 2000, the American College of Rheumatology’s clinical practice guidelines for the treatment of OA of the knee and hip outlined the use of nonpharmacologic modalities and drug therapy, briefly mentioning that disease-modifying drugs, in addition to controlling the symptoms, slow the progression of structural damage in the affected joints.4 Although a core set of recommendations for OA management exist, the consensus recommendations are not always supported by the best available evidence.3 Lack of expert consensus on OA management necessitates further research on comparative efficacy of the existing interventions and evaluating their cost-effectiveness.
Slow-acting pharmacologic agents for OA treatment, including intra-articular hyaluronan products and diacerein, are attracting increasing attention. These drugs are characterized by a slow onset of efficacy (6-8 weeks) and a prolonged residual effect once treatment is stopped (about 2 months after withdrawal).5 Diacerein, known in the International Union of Pure and Applied Chemistry as 4,5-diacetyloxy-9,10-dioxo-anthracene-2-carboxylic acid, is an oral anthraquinone derivative. Diacerein and its active metabolite rhein inhibit the stimulation of interleukin-1 beta and metalloproteases production that are involved in cartilage degradation.6,7 They have no inhibitory effects on cyclooxygenase and increase the expression of transforming growth factor—beta 1 (TGF-beta 1) and TGF-beta 2 with potential cartilage-repairing properties.6,7
In a systematic review of 7 randomized clinical trials (RCTs) on diacerein in adults with knee and hip OA that included 2069 participants, Fidelix et al reported “gold” evidence that diacerein has a small, consistent effect in improving pain and slowing the progress of hip OA.8 In a meta-analysis of 19 RCTs, Rintelen et al demonstrated superiority of diacerein over placebo and similar efficacy for both diacerein and nonsteroidal anti-inflammatory drugs (NSAIDs).9 The latter study also focused on a carryover effect of diacerein persisting up to 3 months after treatment. In a 3-year study, Dougados et al reported a significantly lower rate of joint space narrowing in patients who completed the study and confirmed the structure-modifying effect of diacerein compared with placebo.10
Published research shows a good safety profile for diacerein, which can be considered advantageous if prolonged drug intake is required for symptom control.7-11 In contrast, NSAIDs are seldom used for more than 4 to 6 weeks in patients with OA. Selective cyclooxygenase-2 inhibitors are known to increase risk for thromboembolic disorders such as myocardial infarction or stroke.5 Cardiovascular adverse events in patients treated with diacerein have been observed very rarely, and no reports of acute coronary syndrome have been published. Although not available in the United States or Canada, diacerein is registered in more than 20 countries, including France (more
than 14 million annual prescriptions9), Brazil, Argentina, and India; it is in the filing process in the United Kingdom and Germany.12 Currently approved in the United States, hyaluronan preparations appear to be well tolerated, but discrepancies exist in the literature regarding long-term effectiveness.5,13-17
Although each of the therapies with slow-acting drugs has specific indications and contraindications for their administration, diacerein represents a promising treatment modality as a preferred conservative intervention. It has anti-inflammatory effects different from those of NSAIDs, as well as structure-modifying effects; these effects persist after therapy is discontinued.1 Studies on cost-effectiveness of OA drug therapies can influence further development of existing clinical practice guidelines.
Focusing on comparison of OA treatments in adults with intra-articular hyaluronan product injections and oral intake of diacerein, the present study addresses 3 research questions: (1) Are the 2 treatments similarly effective in terms of pain reduction, avoidance of comedications, joint function improvement, and subjective ratings by patients/investigators? (2) Are they equally safe for patients and similarly tolerated? (3) What is the cost-effectiveness of invasive treatment with hyaluronan products versus conservative treatment with diacerein? These assessments can improve decision making among different payers for health services, formulary managers, and clinicians.
Combined data on health outcomes from independent placebo-controlled RCTs with intra-articular hyaluronan injections and diacerein in the treatment of knee/hip OA were analyzed by applying meta-analytic techniques to assess the interventions’ efficacy and safety. A cost-effectiveness analysis was done to evaluate each treatment’s medico-economic impact.
Selection of Trials
Search of RCTs on the treatment of adults (age >18 years) with knee/hip OA in MEDLINE, PREMEDLINE, Cochrane Central Register of Controlled Trials, Current Contents, Google, and Yahoo was carried out using the terms “diacerein,” “diacerhein,” “diacetylrhein,” “intra-articular,” “hyaluronan,” “hyaluronate,” “sodium hyaluronate,” and “hylan.” Publications in English up to January 2008 were used.
Single-blind and double-blind RCTs on intra-articular hyaluronan preparations injection versus placebo and diacerein versus placebo in adults with primary or secondary knee/hip OA fulfilling the American College of Rheumatology criteria were eligible for selection. To improve the homogeneity of the analysis, RCTs with extractable data on relevant outcome measures were selected. A methodologic quality score was assigned to each RCT by using the validated scale of Jadad et al.18 The scale assesses the method of randomization, blinding of the study to investigators and patients, the percentage of dropouts and withdrawals, and the availability of intention-to-treat, and assigns a score from 0 = weakest to 5 = strongest. The main criteria for exclusion from the meta-analysis were evidence of secondary disease and lack of the proper statistics needed for the effect size calculation.
Five treatment outcomes—pain on active movement, improvement in joint function, escape of need for comedication, global efficacy (subjective ratings by patients/investigators), and global safety (subjective ratings by patients/investigators, adverse events)—were included in the assessment of the efficacy and safety for each therapy option.
The visual analog scale (VAS) and Western Ontario and McMaster Universities’ Osteoarthritis Index (WOMAC) were used for measuring pain and functional impairment. The 100-mm VAS on pain has a range from 0 = no pain to 100 = unbearable pain. The summed WOMAC has possible scores of 5-25 for pain, 2-10 for stiffness, and 17-85 for function.15 The VAS on handicap has a 0-100 scale, 100 being the worst. The subjective grading of treatment was done according to a 4-point Likert scale (1 = ineffective, 2 = less effective, 3 = effective, 4 = very effective); these ratings, which were done both by the subjects themselves and by the investigators, were used to assess global efficacy. The number of adverse events reported (eg, injection-site swelling, skin reactions) was used to assess the drug safety. This research protocol also included clinicians’ assessment scores for global tolerance using a 5-point scale (very good, good, moderate, bad, very bad), based on the patients’ responses as documented by the investigators. To determine the escape medication outcome, the quantity of concurrent medication was measured.
Economic effectiveness was expressed in terms of the average annual direct cost per patient whose OA improved. The Medicare payment schedule was used to estimate the costs of physician and clinic visits, procedures, and diagnostic tests. Health resource utilization during a 1-year period was computed as the weighted average of the assessment visits from included RCTs. Data from the Agency for Healthcare Research and Quality19 and Centers for Medicare & Medicaid Services20 were major sources of cost information.
For the cost-effectiveness analysis, average intake of diacerein for 4 months and the price data for diacerein tablets found in online pharmacies were used in the calculations. Published findings proposing that hyaluronan products with different molecular weights possess similar clinical efficacy and safety profiles21,22 supported the computation of outcome measures for all hyaluronan preparations, which were included in the meta-analysis of the RCTs as weighted averages. The treatment cost for hyaluronan products was based on recommended treatment regimes. Loperamide was selected for controlling the most common adverse event of diacerein—diarrhea. Acetaminophen was the drug of choice for rescue medication. The costs of generic co-medications were used in the analysis. The similar nature of the treatments allowed us to infer that indirect costs relating to travel, caregiver payments, reduced productivity, etc, were the same across the 2 treatments; hence, it was safe to assume such costs away. Utilizing the Consumer Price Index for medical care, all calculated costs were adjusted to 2006 US dollars.23 The cost-saving ratio was computed for patients whose OA improved as the annual treatment costs associated with hyaluronan injections divided by the annual treatment costs associated with diacerein treatment.
The efficacy and safety of the 2 treatments were evaluated by using Meta Analysis Version 2 software developed by Borenstein et al.24 Data on selected outcomes were manually entered as weighted mean differences with 95% confidence intervals. When different means were entered for measurement of the same variables, standardized mean differences were applied. The statistical analyses were based on the intention-to-treat principle. The software allowed computation of odds ratios and correlation with calculated effect sizes.
Effect sizes were calculated as Cohen’s d, where a positive effect size represents improvement and a negative effect size represents a “worsening of symptoms.” Cohen’s d is a descriptive measure, and Cohen interpreted the effect size as small if d <0.2, medium if d <0.5, and large if d >0.5.25 Cohen’s d was obtained according to the formula
d = M1 − M2
[(σ2 1 + σ2 2)/2]
where M1 and M2 are the means of the intervention and control groups, respectively, divided by the pooled standard deviation. Hedges’s g was derived from Cohen’s d as its corrected measure. The binominal effect size display was used to illustrate the percentage of persons who experienced symptoms improvement in the treatment and control groups.26 The odds ratios were obtained through conversion of the calculated effect size. By convention, an odds ratio of 1 indicates that the outcome under study is equally likely in both intervention and control groups. An odds ratio greater than 1 indicates that the outcome is more likely in the first group, whereas an odds ratio less than 1 indicates the opposite.
Of the 51 publications retrieved, 18 clinical trials on diacerein versus placebo and hyaluronan products versus placebo met the inclusion criteria. These RCTs were coded and further evaluated for applicable data on outcome measures. One of these studies was in French; 2 others lacked the required statistical information. Thus, the meta-analysis was done using 10 RCTs with administration of hyaluronan injections and 5 RCTs with diacerein intake (Table 1 and Table 2). The average Jadad scores for studies on hyaluronan products and diacerein were 4.0 and 4.2 points, respectively, which indicate good quality of the selected RCTs.
Efficacy and Safety
At the end of active treatment, the effects of hyaluronan injections on reduction of joint pain and the global efficacy of the product as assessed by patients and investigators were considerable, as evident from the estimated medium effect size (Table 3). However, a small effect was observed in terms of function improvement, global safety, and reduced need for additional medication.
The results indicated that about 20% of the patients treated with hyaluronan preparations would not require additional NSAIDs or analgesics. An even smaller percentage would be expected to be free of adverse effects during the course of treatment. Compared with placebo, diacerein showed small effects on all 5 variables at the end of active treatment (Table 4). Global efficacy for diacerein was statistically superior to that for placebo, but the drug was neither safer nor better tolerated.
The odds ratio of 1.33 for global safety pointed out that adverse events were more likely to develop in the drug intervention group. The odds ratio of 0.76 for the co-medication variable indicated reduced use of analgesics or NSAIDs in the diacerein group and a greater need for rescue medication in the placebo group. Although the highest level of diacerein efficacy was observed between 2 and 6 months of therapy, the mean length of treatment for the selected RCTs of patients receiving diacerein was 12.1 months. Such prolonged intake of the drug could have influenced the obtained parameters of global safety, which reflected a higher incidence of adverse events for diacerein compared with placebo.
The mean effect sizes for the 2 interventions indicated different levels of treatment impact on the health outcomes, as is evident from the medium effect of hyaluronan injections with effect size = 0.33 and the small effect of diacerein therapy with effect size = 0.26 (Table 5). The mean success rate for hyaluronan products was 58%, while the mean success rate for diacerein intake was 56%. The results showed that with hyaluronan injections, 57% of intention-to-treat adults had functional improvement in the joints and 58% of the patients experienced relief from pain and were satisfied with the therapy. With diacerein treatment, 56% had improved joint function and less pain and 57% were satisfied with the treatment.
The per patient average annual direct cost of treatment with diacerein was $1921.90, and the average direct cost for hyaluronan injections was $3032.10 (Table 6 and Table 7). The costs of hyaluronan products and required arthrocentesis were the main contributors to the difference in the costs of the 2 interventions. The weighted cost per patient whose OA improved on diacerein therapy was $1076.26. For hyaluronan injections, the weighted cost per patient whose OA improved was $1758.62, or $682.36 more than the cost of diacerein (Table 8).
The difference between the 2 therapies in the proportion of patients improved was not statistically significant (P >.05). The results also showed that if the diacerein treatment option is selected, it can yield a cost-saving ratio of 1.63.
Published meta-analytic research studies vary in the number of clinical trials included in the analysis; hence, they may differ in their findings because of eligibility criteria for selecting trials for inclusion as well as different strategies for the literature search.5,9,13,17 Meta-analysis in this study provided evidence of statistically significant clinical efficacy of intra-articular hyaluronan preparations and diacerein for treatment of OA in adults.
The analysis demonstrated that for pain reduction during active movements, joint function improvement, and subjective ratings of global efficacy, both pharmacologic agents are similarly effective, albeit with a slightly higher success rate (58%) for hyaluronan therapy than for diacerein (56%). Global safety assessment revealed that hyaluronan injections did not cause more adverse events than intra-articular placebo, indicating that administration of hyaluronan products is reasonably safe. Global safety results for diacerein confirmed the higher incidence of adverse events in patients receiving diacerein than in patients receiving placebo. The dependent variable representing additional intake of medications for symptom reduction yielded a small effect size for both hyaluronan products and diacerein.
Few research studies regarding the medico-economic impact of knee/hip OA treatment in a large population have been undertaken. Torrance et al demonstrated that including Hylan GF-20 in “appropriate care” resulted in significantly higher costs, but more quality-adjusted life-years for patients.34 Kahan et al reported that Hylan GF-20 was more effective than conventional treatment and at no additional cost to the public health insurance system.35
The search of accessible databases did not reveal any published materials on economic evaluation of diacerein administration. However, diacerein is proven to be effective for relieving OA symptoms,7-11,33,36,37 and hyaluronan preparations are recommended by the clinical practice guidelines only when all attempts of treatment with oral pharmacologic agents have failed. Therefore, consideration should be given to making diacerein available as a potential conservative treatment choice and to undertaking further OA studies regarding the potential disease-modifying effects of diacerein.
The obtained cost-saving ratio of 1.63 indicates that the conservative treatment with diacerein is cost-efficient and should be included as an option. Whether this ratio represents a good value depends on the decision makers responsible for optimal allocation of healthcare resources.
Four of the RCTs on hyaluronan products included in the meta-analysis were single-blind. The protocols of these RCTs may have carried biases in terms of objectivity of the clinical assessment data. Only 5 RCTs on diacerein met the inclusion criteria. Although all 5 trials were double-blind, a larger number of the studies might provide more information on health outcomes, thus adding to the validation of the computed effect size, success rate, and calculated cost-effectiveness. Access to the treatment cost data was restricted in government databases and healthcare statistical reports. In addition, the cost-effectiveness analysis did not consider the cost of more than 1 course of hyaluronan injections, and only most frequent adverse events reported in the RCTs were accounted for.
The present study evaluating efficacy of invasive treatment with hyaluronan products and conservative treatment with diacerein is one of the first to compare the cost-effectiveness of these 2 interventions in adults with symptomatic knee/hip OA. Hyaluronan injections were slightly more effective in terms of health outcomes, yielding medium effects on pain reduction during active movements and subjective assessment of global efficacy. Diacerein administration was characterized by small effects for these variables.
The difference between the 2 interventions in the proportions of patients whose OA improved was not statistically significant. The difference in the annual costs per patient whose OA improved was $682.36; the difference in Medicare payments was $981.61 per patient per year as measured in 2006 US dollars. The cost-saving ratio of 1.63 with diacerein therapy and the advantages of conservative treatment suggest that decision makers should consider use of diacerein therapy.
Based on the results of this study, clinicians as well as healthcare managers can assist patients with symptomatic knee/hip OA in balancing the potential benefits of continuation of conservative treatment against the benefits of invasive therapy. Future studies using measurements of health-related quality of life are needed to support the findings that diacerein therapy for OA patients has benefits from the social perspective.