Calcif Tissue Int (2006) 79:27 36 DOI: 10.1007/s00223-006-0024-4
Statin but not Non-Statin Lipid-Lowering Drugs Decrease Fracture Risk: A Nation-Wide Case-Control Study L. Rejnmark, P. Vestergaard, L. Mosekilde Department of Endocrinology and Metabolism, Aarhus University Hospital, Aarhus Sygehus, Tage Hansens Gade 2, DK-8000 Aarhus C, Denmark
Received: 30 January 2006 / Accepted: 8 April 2006 / Online publication: 24 July 2006
Abstract Introduction: Discrepant results have been reported on association between treatment with lipid lowering drugs and fracture risk. Several studies have failed to demonstrate an effect of statins on bone mineral density. Therefore, the epidemiological findings of a reduced fracture risk may be due to selections bias, e.g. a healthy drug user effect. If so, the reduced fracture risk is most likely independent of type of lipid lowering drug. Aim: We assessed fracture risk in users of various lipidlowering drugs. Methods: In a case-control design, we compared 124,655 fracture cases with 373,962 age- and gender-matched controls. We used computerized registers to assess individual drug use and related these data to individual fracture data and information on potential confounders. Results: Use of statins was associated with a reduced risk of any fracture (adj. OR 0.87; 95% CI, 0.83 0.92) and hip fractures (adj. OR 0.57; 95% CI, 0.48 0.69). Risk of hip fracture decreased with increased accumulated dose of statins. This was true in men and in women and in subjects younger and older than 65 years of age. However, fracture risk was not reduced in patients treated with pravastatin (adj. OR 1.02; 95% CI, 0.89 1.17) or non-statin lipid lowering drugs (adj. OR 0.99; 95% CI, 0.86 1.15). Conclusions: The reduced fracture risk in users of lipid lowering drugs is apparently specifically related to users of non-pravastatin statins. Our findings do not support the hypothesis of a healthy drug user effect as an explanation for the reduced fracture risk in users of statins. Statins have been suggested as potential agents in the treatment of osteoporosis. In vitro studies in different cell lines and studies in experimental animals have suggested antiresorptive as well as bone anabolic effects of statins [1 3]. However, in humans discrepant results have been reported on the effect of statins on biochemical markers of bone turnover, bone mineral density (BMD), and fracture risk. In a retrospective cohort study in type 2 Correspondence to: L. Rejnmark, Dept of Endocrinology and Metabolism C, Aarhus University Hospital, Aarhus Sygehus, Tage-Hansens Gade 2, DK-8000 Aarhus C, Denmark; E-mail:
[email protected]
diabetic patients, Chung et al. [4] reported a 2.9% higher femoral neck BMD in 36 patients on statin therapy compared with 33 untreated patients. In contrast, in a recent randomized controlled trial, one-year of treatment with Simvastatin 40 mg/day did not affect bone turnover or BMD in 82 postmenopausal women with osteopenia [5]. Treatment with statins has been associated with a decreased fracture risk in some [6 13] but not all [14 17] epidemiological studies. As statins are re-circulated in the enterohepatic circulation, it has been suggested that drug concentration in human bone tissue may be too low to exert the effects found in experimental-studies. Accordingly, the positive effect of statins on bone observed by some investigators may merely be due to factors associated with statin treatment in general than a specific pharmacological drug effect, i.e. a healthy drug user effect. If so, it would be reasonable to expect a similar reduction in risk of fracture due to treatment with non-statin lipid lowering drugs. However, treatment with non-statin lipid lowering drugs has only been associated with a decreased fracture risk in one [13] of several [6 9, 15, 18] studies. Furthermore, differences in pharmacokinetic properties exist between the statins. Pravastatin is a hydrophilic drug whereas the other types of statins are lipophilic drugs. It has therefore been suggested that pravastatin to a lesser extent than other statins is able to enter bone cells. If the positive effect of statins on bone is caused by a pharmacological drug effect, the effect of pravastatin may be expected to be less pronounced than the effect of other statins. On the other hand, if the observed effects of statins are due to a healthy drug user effect, treatment with pravastatin should exert effects similar to those of other statins. In previous studies, discrepant results on the effect of pravastatin therapy on fracture risk have been reported [6 8, 10, 16]. In a nation-wide population-based pharmaco-epidemiological case-control design, we studied effects of statin and non-statin lipid lowering drugs on fracture risk and compared the effect of pravastatin with the effect of other statins on fracture risk.
28
L. Rejnmark et al.: Lipid-lowering drugs and fracture risk
Subjects and Methods
Data on Confounding Factors
In Denmark the extensive nature of registers covering contacts to the health sector offers good possibilities for studies on the occurrence of fractures [19]. The Danish National Health Service provides tax-supported healthcare for all inhabitants, allowing free access to general practitioners and hospitals. In addition, reimbursement is provided for a variable proportion of the costs of prescribed drugs. Using the unique 10-digit civil registry number that is assigned to all Danish citizens shortly after birth, a complete hospital discharge and prescription history can be established for each individual, and valid linkage between population-based registries can be obtained. This case-control study was performed within the Danish population that constituted approximately 5.3 million individuals during the study period
In order to adjust for potential differences in co-morbidity between cases and controls, we used the Charlson co-morbidity index [25]. The index includes 19 diseases, which have been selected and weighted on the basis of the strength of their association with mortality [26]. Thus, for all cases and controls we identified all discharge diagnoses covering the 19 diseases included in the co-morbidity index using The National Hospital Discharge Register. Moreover, the numbers of days spend in hospital the year preceding fracture (year 1999) and a history of a prior fractures in the period 1977 2000 was also included as confounders. In addition, the prescription database was used to identify use of other medications believed to affect fracture risk. We retrieved data on whether studied subjects, during the period of 1996 to 2000 had redeemed a prescription (‘‘yes’’ or ‘‘no’’) for the following drugs: diuretic, antihypertensive drugs (except diuretics i.e., betablockers, calcium channel blockers, angiotensin-converting enzyme inhibitors, and other types of antihypertensive drugs), antiresorptive drugs (estrogens, bisphosphonates and/or raloxifene), antiepileptic drugs, anxiolytics, sedatives, neuroleptics, antidepressants, systemic and topical corticosteroids, thyroid hormones, and antithyroid drugs. Furthermore, our confounder analysis included data from the National Bureau of Statistics on income in 1999, social status in 1999, working status in 1999, educational status in 1999, and data from The National Health Organisation Register on number of contacts to general practitioners and practising specialists for the period 1996 to 2001.
Identification of fracture cases In Denmark, The National Hospital Discharge Register covers all contacts (on in- or out-patient basis) to the hospitals [20]. The register was founded in 1977, but outpatient records were first completely incorporated from 1995. The files of The National Hospital Discharge Register include information on the civil registry number of the patient, date of discharge, and discharge diagnoses, assigned exclusively by the physician at discharge according to the Danish version of the International Classification of Diseases, 8th revision (ICD-8) until the end of 1993, and to the Danish version of the International Classification of Diseases, 10th revision (ICD-10) from 1994. The register has nationwide coverage of public hospitals with an almost 100% completeness of recordings and a high precision of diagnoses [20], especially of fracture diagnoses [21]. Using The National Hospital Discharge Register we identified all subjects who had sustained a fracture between 1 January 2000 and 31 December 2000 (n = 124,655). The cases occurred only once in the analyses with the first occurrence of an incident fracture during the year 2000. Selection of Population-based Controls Using the Civil Registration System, which has electronic records on all changes in vital status, including change of address and date of death for the entire Danish population since 1968, we randomly selected 3 controls for each case, matched by gender and year of birth. The controls were selected using the incidence-density sampling technique [22], i.e., the controls had to be alive and at risk for fracture diagnosis at the time the corresponding case was diagnosed. A total of 373,962 controls were included in the study. Data on Use of Lipid Lowering Drugs In Denmark, pharmacies are equipped with a computerised accounting systems through which data are sent directly to a Register of Medicinal Product Statistics (i.e., a prescription database) at The Danish Medicines Agency with key information on prescriptions for refundable drugs. The prescription database includes information on the patient’s civil registry number, the type and amount of drug prescribed according to the Anatomical Therapeutical Chemical classification system (ATC) [23, 24], and the date the prescription was filled. Within the Prescription Database, we identified all prescriptions for lipid lowering drugs among cases and controls redeemed within five years before the date of hospitalisation of cases. The drugs studied were statins (atorvastatin, fluvastatin, lovastatin, pravastatin, and simvastatin) as well as non-statin lipid lowering drugs (cholestyramin, fibrate, and nicotinic acid). In Denmark, lipid- lowering drugs are available only by prescription, and the public health authority provides reimbursement for expenses on lipid lowering drugs.
Statistic Data from the different registers were merged at the National Bureau of Statistics, and for each subject the 10 digit civil registry number was substituted by a unique case number i.e., as investigators we had no access to personally identifiable information. Mean and standard deviation were used as descriptive statistics. Crude odds ratios (OR’s) were calculated and 95% confidence intervals approximated using the method of Miettinen [27]. A conditional logistic regression analysis was used to assess the association between any fracture and the exposure variable. Analyses were also performed gender- and age stratified. Analyses were performed using STATA 8.1 (STATA Corp., College Station, TX) and SPSS 10.1.0 (SPSS Inc., Chicago Ill.) both in the Unix version. Results
Characteristics of the 124,655 cases that sustained a fracture during year 2000, and the 373,962 age- and gender-matched controls are shown in Table 1. Among the cases, a total of 10,530 subjects sustained a hip fracture (71.7% were women) and 20,035 sustained a forearm fracture (61.6% were women). Moreover, 3,364 were diagnosed with a fracture at the spine (55.2% were women). The mean age of studied subjects was 43 years. Socioeconomic and health related characteristics differed significantly between cases and controls (Table 1). The cases more often were retired, were more likely to be divorced or unmarried, and had a lower income than controls. In addition, compared to controls, cases tended to have a higher frequency of comorbidity (fewer with a Charlson index of 0) and a higher number of comorbid conditions, including a higher frequency of
L. Rejnmark et al.: Lipid-lowering drugs and fracture risk
Table 1. Characteristics of patients and controls
29
any fracture
Variable Age (years) Gender Men Women Annual income (Danish Crowns) Marital status Widowed Divorced Married Unmarried Other Occupational status Independent Assisting wife/husband Working Retired Other Charlson index* 0 (no comorbidity) 1 2 3 4 ‡5 Previous fracture Number of bed days in 1999 Contacts to GP or specialists in 1999 Use of antiresorptive drug Ever use of corticosteroids Ever use of levothyroxine Ever use of antithyroid drugs Anxiolytics, sedatives and hypnotics Neuroleptics Ever use of antidepressants Ever use antiepileptic drug Ever use of diuretics Ever use of other diuretics Ever use of potassium sparing diuretics Ever use of loop diuretics Ever use of thiazide diuretics Ever use of antihypertensives (other than diuretics) Ever use of statins Ever use of non-statin cholesterol-lowering drugs Antihypertensive drugs (other than diuretics) Alcoholism Cholestyramin Fibrate Nicotinic acid Atorvastatin Fluvastatin Lovastatin Pravastatin Simvastatin
Cases (n = 124,655)
Controls (n = 373,962)
43.44 ± 27.39
43.44 ± 27.39
60,107 (48.2%) 64,548 (51.8%) 161,036 ± 138,789
180,321 (48,2%) 193,641 (51.8%) 172,322 ± 193,704
18,365 10,423 35,859 59,335 90
(14.8%) (8.4%) (28.9%) (47.8%) (0.1%)
52,550 23,239 123,719 171,349 264
(14.2%) (6.3%) (33.3%) (46.2%) (0.1%)
3,374 209 37,797 40,201 20,752
(3.3%) (0.2%) (36.9%) (39.3%) (20.3%)
11,816 951 124,984 109,447 59,278
(3.9%) (0.3%) (40.8%) (35.7%) (19.3%)
P — —
