Calcified Tissue International

, Volume 79, Issue 2, pp 69–75

Methotrexate, Azathioprine, Cyclosporine, and Risk of Fracture

Authors

    • Department of Endocrinology and Metabolism CAarhus Amtssygehus, Aarhus University Hospital
  • L. Rejnmark
    • Department of Endocrinology and Metabolism CAarhus Amtssygehus, Aarhus University Hospital
  • L. Mosekilde
    • Department of Endocrinology and Metabolism CAarhus Amtssygehus, Aarhus University Hospital
Article

DOI: 10.1007/s00223-006-0060-0

Cite this article as:
Vestergaard, P., Rejnmark, L. & Mosekilde, L. Calcif Tissue Int (2006) 79: 69. doi:10.1007/s00223-006-0060-0

Abstract

We studied the fracture risk associated with use of methotrexate, azathioprine, and cyclosporine. The study was designed as a case-control study. All patients with a fracture (n = 124,655) in the year 2000 in Denmark served as cases. Information on fractures and confounders was retrieved from the National Hospital Discharge Register and a number of other national registers. For each case, three age- and gender-matched controls were randomly drawn from the general population (n = 373,962). Exposure was use of the drugs and a number of covariates including other immunosuppressive drugs, corticosteroids, any cancer, Crohn’s disease, ulcerative colitis, rheumatoid arthritis, psoriasis, liver and kidney disease, prior fracture, and alcoholism. Azathioprine was associated with an increase in overall fracture risk, but besides this, none of the drugs was significantly associated with overall fracture risk or risk of hip, spine, or forearm fracture. Liver [odds ratio (OR) = 1.55, 95% confidence interval (CI) 1.42–1.69] and kidney (OR = 1.26, 95% CI 1.16–1.37) diseases were significantly associated with increased risk of fractures. Azathioprine was associated with an increase in overall fracture risk but not in the risk of spine, hip, or forearm fractures. Methotrexate and cyclosporine were not associated with fracture risk. It thus seems that the underlying disease for which the treatment is administered may be responsible for much of the increase in fracture risk rather than the drugs used to treat the disorder in question.

Keywords

MethotrexateAzathioprineCyclosporineFracture risk

Immunosuppressive therapy is used in a number of conditions ranging from inflammatory diseases (rheumatoid arthritis, asthma, inflammatory bowel disease, etc.) to malignancies and prevention of graft rejection in subjects undergoing organ transplantation [1]. However, they are assumed to be associated with a number of serious adverse reactions, including osteoporosis and fractures [1].

On the other hand, they might possess a corticoid-sparing effect [25], thus sparing the users the negative effects of corticosteroids on fracture risk [69] and bone mineral density (BMD) [1012]. However, the corticosteroid-sparing potential of methotrexate in conditions as different as polymyalgia [13] and asthma [4] has been questioned.

The diseases for which the drugs are used may also increase fracture risk, and any fracture risk seen with the drugs may be due to confounding by indication due to the severity of the underlying disease. This is the case with inhaled corticosteroids [14], where more severe cases of asthma require more corticosteroids for the control of the disease; however, the more severe cases of asthma are per se also more likely to have osteoporosis. In the case of immunosuppressants, concomitant use of corticosteroids may increase fracture risk. However, concomitant use of pain medication such as opiates may also lead to dizziness and falls and, thus, fractures.

The immunosuppressant drugs may have a number of effects besides their effects on the disease being treated. A number of diseases may be associated with hypogonadism through, e.g., direct destruction of the gonads or the pituitary or through emaciation. However, the treatment for the disease in question may also lead to hypogonadism by the effects of chemotherapy, corticosteroids, or irradiation on the gonads [15, 16]. Besides its effects on the gonadal system, chemotherapy may exert negative effects on osteoclasts and osteoblasts [15].

Methotrexate has been associated with a detrimental effect on trabecular bone in rats [17, 18] and in humans with osteosarcoma [19]. However, no major effect on BMD has been observed after adjustment for concomitant use of corticosteroids in patients with rheumatoid arthritis [2026]. Fractures have been described in children with leukemia [27, 28] and in patients with rheumatoid arthritis [29] treated with methotrexate. Addition of cyclosporine in patients with methotrexate-resistant rheumatoid arthritis has led to suppression of inflammation and an increase in BMD [30]. Especially patients undergoing organ transplantation may have significant bone loss and increased risk of fractures [31]. No systematic studies on the effects of azathioprine, methotrexate, or cyclosporine on BMD or fracture risk seem to exist.

We therefore undertook a population-based study to assess the risk of fractures associated with azathioprine, cyclosporine, and methotrexate after adjustment for a number of comorbid conditions.

Subjects and Methods

Study Design

The study was designed as a case-control study. All subjects sustaining a fracture during the year 2000 in Denmark were included as cases (n = 124,655), and for each case three subjects of the same age (same birth year) and gender were randomly selected from the background population as controls (n = 373,962). Controls were selected using the incidence-density sampling technique [32]; i.e., they had to be alive and at risk for fracture diagnosis at the time the corresponding case was diagnosed. The follow-up time was time before fracture in cases and a corresponding dummy date among controls. Cases occurred only once in the analyses with the first occurrence of an incident fracture during the year 2000.

End Points

The study end points were occurrence of any fracture (ICD-10 codes S02.0-S02.9, S07.0-S07.9, S12.0-S12.9, S22.0-S22.9, S32.0-S32.8, S42.0-S42.9, S52.0-S52.9, S62.0-S62.9, S72.0-S72.9, S82.0-S82.9, S92.0-S92.9) between January 1, 2000, and December 31, 2000. In Denmark, almost all patients with fractures are managed in the hospital system (including emergency rooms) [33]; even fractures sustained abroad are registered upon return for insurance reasons. The capture of fractures is thus very high.

Exposure Variables

The exposure variables were ever use of (1) methotrexate, (2) cyclosporine, (3) azathioprine, or (4) other immunosuppresants/chemotherapeutics.

The confounders were (1) ever use of other drugs important to fracture risk (corticosteroids [6], anxiolytics and sedatives, neuroleptics, antidepressants, antiepileptics [34], diuretics [35, 36], opiates, and nonsteroidal anti-inflammatory drugs); (2) comorbidity (cancer, inflammatory bowel disease such as Crohn’s disease and ulcerative colitis, rheumatoid arthritis, psoriasis, any liver or kidney disease), number of bed days in 1999 (the year before the fracture), and number of contacts to general practitioner or practising specialist; (3) social variables (income, education level, living alone or with someone, working or not); (4) organ transplantation; and (5) other important variables associated with fracture risk (prior fracture [37], alcoholism).

Prior fracture was defined as a prior contact under such a diagnosis from 1977 to the date of censoring from the National Hospital Discharge Register. Alcoholism was defined as a prior contact under a diagnosis of alcoholism from 1968 to the date of censoring from the Psychiatric Central Register or at least one prescription of disulfiram from 1996 to the date of censoring from the National Pharmacoepidemiological Database [38].

Registers Used

We retrieved the information on fracture occurrence and occurrence of other diseases, prior fractures, and diagnosis of alcoholism from the National Hospital Discharge Register between 1977 and the date of censoring [39] and the Psychiatric Central Register for the period 1968 to the date of censoring [40]. We obtained information on the use of prescribed drugs from the Danish Medicines Agency for the period 1996–2000. Exposure was calculated as cumulated number of defined daily dosages (DDDs) redeemed of the drug group in question from January 1, 1996, to the date of fracture or the date of censoring among cases or controls. DDD in this report is thus the sum of drugs used. For methotrexate, DDD could not be used as the drug often is administered intermittently. Instead, the cumulated number of milligrams ingested was used.

The National Hospital Discharge Register was founded in 1977 [39]. It covers all inpatient contacts from 1977 to 1994 and, from 1995, also all outpatient visits to hospitals, outpatient clinics, and emergency rooms [39]. Upon discharge, the physician codes the reason for the contact using the International Classification of Diseases (ICD) system. The code used is at the discretion of the individual physician. The register has nationwide coverage and almost 100% capture of contacts [39]. In general, the validity of registrations is high [41], especially for fractures, where a precision of 97% has been reported for fractures treated both on an inpatient and on an outpatient basis via emergency rooms (e.g., forearm fracture) [42].

The Psychiatric Central Register was founded in 1968 and covers all in- and outpatient contacts to Danish mental hospitals [40]. It has nationwide coverage and high validity of diagnoses [33]. This register also uses the ICD system for coding contacts.

The Danish Medicines Agency keeps a nationwide register of all drugs sold at pharmacies throughout the country from 1996 on (National Pharmacological Database run by Danish Medicines Agency, http://www.dkma.dk). Any drugs bought are registered with ATC (Anatomical therapeutic chemical classification system) code, dosage sold, and date of sale for the period January 1, 1996, to December 31, 2000. As all sales are registered to the individual who redeemed the prescription, the capture and validity are high.

We linked these sources of information through the Central Person Register Number, which is a unique registration code given to every inhabitant—to some degree similar to the U.S. social security number—that allows registration on an individual basis.

The project was approved by the National Board of Health and subject to control by the National Data Protection Agency.

Statistical Analyses

Mean and standard deviation (SD) were used as descriptive statistics. Crude and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. A conditional logistic regression analysis was used to assess the association between any fracture and the exposure variable.

Analyses were performed using STATA 8.1 (StataCorp, College Station, TX) and SPSS 13.0 (SPSS Inc., Chicago IL), both in the UNIX version.

Results

Table 1 shows the baseline characteristics of fracture cases and controls. Cases and controls were well matched for age and gender. Fracture cases more often were retired and thus had a lower income. Cases also more often were unmarried. The comorbidity and use of drugs in general were higher among cases than controls. Only methotrexate, cyclosporine, and azathioprine were used in numbers that allowed individual analysis. The other immunosuppressants were put into one group. Due to differences in defined dosages, it was not possible to compute a common dose-response relationship for this group. The total number of person-years of observation was 10.2 million for cases and controls combined from 1977 on (National Hospital Discharge Register on comorbidity) and 2.5 million from 1996 on (drug use from the National Pharmacological Database).
Table 1

Characteristics of patients and controls – any fracture

Variable

Cases (n = 124,655)

Controls (n = 373,962)

P

Age (years)

43.44 ± 27.39

43.44 ± 27.39

Gender

   

  Men

60,107 (48.2%)

180,321 (48.2%)

  Women

64,548 (51.8%)

193,641 (51.8%)

Annual income (DKR)

161,036 ± 138,789

172,322 ± 193,704

<0.01

Living with someone

35,922 (29.0%)

123,925 (33.4%)

<0.01

Working

41,380 (40.4%)

137,751 (44.9%)

<0.01

Previous fracture

41,315 (33.1%)

56,200 (15.0%)

<0.01

Number of bed days in hospital in 1999

9.7 ± 39.7

4.2 ± 20.3

<0.01

Contacts to GP or specialists in 1999

23.9 ± 43.3

18.1 ± 31.4

<0.01

Alcoholism

8,863 (7.1%)

9,473 (2.5%)

<0.01

Rheumatoid arthritis

2,870 (2.3%)

5,485 (1.5%)

<0.01

Psoriasis

474 (0.4%)

939 (0.3%)

<0.01

Crohn’s disease

332 (0.3%)

763 (0.2%)

<0.01

Ulcerative colitis

514 (0.4%)

1,272 (0.3%)

<0.01

Any liver disease

1,327 (1.1%)

1,138 (0.3%)

<0.01

Any kidney disease

1,106 (0.9%)

1,774 (0.5%)

<0.01

Cancer (any type)

7,555 (6.1%)

18,331 (4.9%)

<0.01

Organ transplantation

102 (0.1%)

90 (<0.1%)

<0.01

Antiepileptic drugs

7,091 (5.7%)

10,974 (2.9%)

<0.01

Sedatives, anxiolytics, and hypnotics

35,840 (28.8%)

82,766 (22.1%)

<0.01

Neuroleptics

9,738 (7.8%)

17,243 (4.6%)

<0.01

Antidepressants

18,511 (14.8%)

34,521 (9.2%)

<0.01

Ever use of any corticosteroid

67,695 (54.3%)

189,636 (50.7%)

<0.01

Ever use of NSAIDs

59,690 (47.9%)

142,274 (38.0%)

<0.01

Ever use of opiates

38,896 (31.2%)

64,522 (17.3%)

<0.01

Ever use of azathioprine

486 (0.4%)

771 (0.2%)

<0.01

Ever use of cyclosporine

82 (0.1%)

120 (<0.1%)

<0.01

Ever use of methotrexate

712 (0.6%)

1,361 (0.4%)

<0.01

Other immunosuppressants

137 (0.1%)

180 (<0.1%)

<0.01

Melphalan

20 (<0.1%)

17 (<0.1%)

<0.01

Cyclophosphamide

43 (<0.1%)

54 (<0.1%)

<0.01

Chlorambucil

58 (<0.1%)

113 (<0.1%)

<0.01

Estramustin

5 (<0.1%)

0 (0%)

<0.01

Mycophenolat

18 (<0.1%)

7 (<0.1%)

<0.01

Tacrolimus

0 (0%)

1 (<0.1%)

0.56

DKR, Danish kroner; NSAID, nonsteroidal anti-inflammatory drug; GP, general practitioner.

Table 2 shows the crude and adjusted relative fracture risks associated with use of various immunosuppressants. All drugs were associated with an increased fracture risk in the crude analysis. However, risk estimates were attenuated after adjustment for other drugs and further attenuated upon adjustment for the other covariates, only azathioprine and the heterogeneous group of other immunosuppressants being significantly associated with risk of any fracture. Liver and kidney diseases and rheumatoid arthritis were associated with an increased fracture risk
Table 2

Relative risk of any fracture: OR and 95% CI

Variable

Crude OR

Mutually adjusted ORa

Multiply adjusted ORb

Azathioprine

1.90 (1.69–2.12)*

1.31 (1.16–1.48)*

1.17 (1.03–1.33)*

Cyclosporine

2.05 (1.55–2.72)*

1.33 (1.00–1.78)

1.20 (0.88–1.64)

Methotrexate

1.57 (1.44–1.72)*

1.12 (1.01–1.23)*

1.01 (0.91–1.12)

Other immunosuppressants

2.29 (1.83–2.85)*

1.67 (1.33–2.10)*

1.49 (1.17–1.88)*

Crohn’s disease

1.31 (1.15–1.49)*

1.18 (1.03–1.35)*

0.98 (0.85–1.13)

Ulcerative colitis

1.21 (1.09–1.34)*

1.11 (0.99–1.23)

0.93 (0.83–1.04)

Rheumatoid arthritis

1.58 (1.51–1.66)*

1.45 (1.38–1.52)*

1.08 (1.02–1.14)*

Psoriasis

1.52 (1.36–1.69)*

1.30 (1.16–1.45)*

1.03 (0.91–1.16)

Any cancer

1.25 (1.22–1.29)*

1.21 (1.18–1.25)*

0.97 (0.94–1.00)*

Any liver disease

3.53 (3.26–3.82)*

3.35 (3.10–3.63)*

1.55 (1.42–1.69)*

Any kidney disease

1.88 (1.74–2.03)*

1.71 (1.58–1.85)*

1.26 (1.16–1.37)*

a Adjusted for the other immunosuppressive drugs

b Adjusted for organ transplantation, alcoholism, working or not, Crohn‘s disease, ulcerative colitis, rheumatoid arthritis, psoriasis, any cancer, any liver disease, any kidney disease, thiazide diuretics, loop diuretics, potassium-sparing diuretics, other diuretics, ever use of antiepileptic drugs, ever use of anxiolytics or sedatives, ever use of antidepressants, ever use of neuroleptics, ever use of corticosteroids, number of bed days in 1999, number of contacts to general practitioner or specialist in 1999, living with someone or living alone, prior fracture, education level, income in 1999, and ever use of nonsteroidal anti-inflammatory drugs or opiates

* P < 0.05

Table 3 shows the relative fracture risk at typical osteoporotic fracture sites. In this analysis, none of the drugs was significantly associated with an increased fracture risk. Actually, a small trend toward a decrease in forearm fracture risk was seen for methotrexate. Again, liver and kidney diseases were significantly associated with fracture risk.
Table 3

Relative risk of osteoporotic fractures: multiply adjusted OR and 95% CIa

Variable

Hip fracture

Forearm fracture

Spine fracture

Azathioprine

0.82 (0.55–1.22)

0.79 (0.53–1.17)

0.93 (0.50–1.72)

Cyclosporine

0.59 (0.21–1.64)

0.75 (0.29–1.94)

2.93 (0.32–27.2)

Methotrexate

1.13 (0.87–1.47)

0.74 (0.55–1.00)*

1.15 (0.71–1.86)

Other immunosuppressants

1.97 (1.12–3.46)*

1.40 (0.71–2.74)

2.70 (0.83–8.84)

Crohn’s disease

1.61 (1.04–2.49)*

0.71 (0.45–1.11)

1.96 (0.90–4.26)

Ulcerative colitis

0.89 (0.62–1.27)

0.89 (0.65–1.22)

0.64 (0.33–1.26)

Rheumatoid arthritis

1.18 (1.05–1.32)*

0.91 (0.79–1.05)

1.09 (0.85–1.39)

Psoriasis

1.28 (0.90–1.81)

1.05 (0.76–1.45)

1.25 (0.69–2.26)

Any cancer

1.10 (1.03–1.18)*

0.99 (0.92–1.07)

0.93 (0.80–1.08)

Any liver disease

1.88 (1.47–2.41)*

1.50 (1.19–1.91)*

2.35 (1.48–3.73)*

Any kidney disease

1.57 (1.30–1.90)*

1.35 (1.07–1.72)*

1.77 (1.14–2.75)*

a Adjusted for organ transplantation, alcoholism, working or not, Crohn’s disease, ulcerative colitis, rheumatoid arthritis, psoriasis, any cancer, any liver disease, any kidney disease, thiazide diuretics, loop diuretics, potassium-sparing diuretics, other diuretics, ever use of antiepileptic drugs, ever use of anxiolytics or sedatives, ever use of antidepressants, ever use of neuroleptics, ever use of corticosteroids, number of bed days in 1999, number of contacts to general practitioner or specialist in 1999, living with someone or living alone, prior fracture, education level, income in 1999, and ever use of nonsteroidal anti-inflammatory drugs or opiates

* P < 0.05

The increase in overall fracture risk with azathioprine did not stem from fractures of the hip, forearm, or spine but was the combined result of fractures in the rest of the skeleton. No specific location could be pointed out as the main reason.

Table 4 shows the dose-response relationship with the drugs. No general association with fracture risk was seen for any of the drugs. For overall fracture risk, low doses of azathioprine were associated with increased fracture risk. However, this increase disappeared at high doses. Again, liver and kidney diseases were significantly associated with fracture risk.
Table 4

Dose response and fracture risk: mutually adjusted OR and 95% CIa

Variable

Any fracture

Hip fracture

Forearm fracture

Spine fracture

Azathioprine

  <60 DDD

1.30 (1.03–1.63)*

0.64 (0.32–1.27)

0.95 (0.48–1.87)

1.08 (0.40–2.88)

  60–249 DDD

1.25 (1.00–1.57)

1.09 (0.58–2.05)

0.85 (0.43–1.71)

1.59 (0.50–5.04)

  ≥ 250 DDD

1.04 (0.86–1.29)

0.78 (0.39–1.57)

0.64 (0.34–1.20)

0.43 (0.14–1.28)

Cyclosporine

  <60 DDD

1.13 (0.64–1.99)

0.21 (0.03–1.85)

1.06 (0.22–5.11)

2.73 (0.25–29.3)

  60–249 DDD

1.21 (0.73–1.99)

0.68 (0.15–3.08)

0.64 (0.13–3.18)

  ≥ 250 DDD

1.28 (0.74–2.21)

1.41 (0.16–2.5)

0.67 (0.12–3.66)

Methotrexate

  <500 mg

0.87 (0.72–1.06)

0.62 (0.36–1.06)

0.59 (0.32–1.07)

3.87 (1.44–10.4)

  500–1,499 mg

0.98 (0.83–1.15)

1.44 (0.97–2.14)

0.78 (0.48–1.27)

0.71 (0.35–1.45)

  ≥ 1,500 mg

1.16 (0.98–1.37)

1.19 (0.77–1.82)

0.80 (0.50–1.29)

0.95 (0.38–2.38)

Other immunosuppressants

1.49 (1.17–1.88)*

1.96 (1.12–3.45)*

1.39 (0.71–2.72)

2.77 (0.86–9.00)

Crohn’s disease

0.99 (0.86–1.13)

1.63 (1.05–2.54)*

0.72 (0.45–1.12)

1.92 (0.88–4.19)

Ulcerative colitis

0.93 (0.83–1.04)

0.88 (0.62–1.25)

0.89 (0.65–1.22)

0.63 (0.32–1.23)

Rheumatoid arthritis

1.07 (1.02–1.13)*

1.18 (1.05–1.32)*

0.91 (0.79–1.05)

1.11 (0.86–1.42)

Psoriasis

1.03 (0.91–1.16)

1.29 (0.91–1.83)

1.05 (0.76–1.44)

1.23 (0.68–2.24)

Any cancer

0.97 (0.94–1.00)*

1.10 (1.03–1.18)*

0.99 (0.92–1.07)

0.93 (0.80–1.08)

Any liver disease

1.55 (1.42–1.69)*

1.90 (1.49–2.43)*

1.50 (1.19–1.91)*

2.46 (1.54–3.94)*

Any kidney disease

1.26 (1.16–1.37)*

1.58 (1.31–1.92)*

1.36 (1.07–1.72)*

1.79 (1.15–2.77)*

a Adjusted for organ transplantation, alcoholism, working or not, Crohn’s disease, ulcerative colitis, rheumatoid arthritis, psoriasis, any cancer, any liver disease, any kidney disease, thiazide diuretics, loop diuretics, potassium-sparing diuretics, other diuretics, ever use of antiepileptic drugs, ever use of anxiolytics or sedatives, ever use of antidepressants, ever use of neuroleptics, ever use of corticosteroids, number of bed days in 1999, number of contacts to general practitioner or specialist in 1999, living with someone or living alone, prior fracture, education level, income in 1999, and ever use of nonsteroidal anti-inflammatory drugs or opiates

* P <0.05

Stratifying the analyses by time of last use showed that azathioprine use less than 1 year ago was associated with an increase in overall fracture risk (OR = 1.24, 95% CI 1.04–1.48), while use of cyclosporine less than 1 year ago (OR = 1.33, 95% CI 0.84–2.12) and use of methotrexate less than 1 year ago (OR = 1.10, 95% CI 0.89–1.15) were not associated with an increase in fracture risk.

Age stratification did not change the overall results (data not shown).

Discussion

In this large-scale population-based case-control study, we found no increase in overall fracture risk, risk of hip, or risk of spine fractures with the use of methotrexate. There was a small trend toward a decrease in forearm fracture risk with methotrexate. There was no dose-response relationship for fractures with methotrexate. Azathioprine was associated with an increase in overall fracture risk but not in the risk of spine, hip, or forearm fractures. There was no association with fracture risk for cyclosporine.

A syndrome of methotrexate osteopathy with pain of the distal tibia, osteoporosis, and compression fractures of the distal tibia has been described [43]. Especially in children, chemotherapy with methotrexate may impair skeletal growth and increase skeletal fragility [44, 45]. However, this could be the result of the very high doses used and the underlying disease.

There was an increase in overall fracture risk with azathioprine. However, the increase did not come from traditional osteoporotic skeletal sites such as the hip, forearm and spine but was the combined result of fractures throughout the rest of the skeleton. The fact that an increase was seen only at low doses with azathioprine could point at a chance finding.

The decrease in fracture risk with increasing doses of azathioprine may be due to adaptation to the drug with increased metabolism but may also signal reduction of the fracture risk associated with the underlying disease because the underlying disease is controlled by treatment with azathioprine.

No dose response was seen for any of the immunosuppressants studied. This may strengthen the view that the drugs per se did not affect fracture risk. The absence of a dose-response effect may be due to the fact that several of these drugs may be administered intermittently (e.g., once a week for methotrexate), which may allow for regeneration of any damage to the bone cells induced by the drug. Furthermore, bone cells may not be susceptible to damage from the drugs at the concentrations used. Improvement of bone turnover by control of the underlying disease by the drugs may be another way that damage by the drugs themselves is countered.

Recency of use did not affect fracture risk associated with the immunosuppressive drugs. This further supports no detrimental effects on fracture risk of methotrexate and cyclosporine. Azathioprine was associated with an increase in fracture risk with both recent and prior use. This may indicate that the increase is a chance finding since an increase with both recent and prior use may indicate permanent damage to the skeleton, but no increase with dose was seen to counter this.

There were significant differences between fracture cases and controls, mainly linked to a higher degree of frailty among cases. Those treated with azathioprine, cyclosporine, or methotrexate were thus often more frail, possibly due to the underlying disease. More frail subjects may also be more susceptible to some of the diseases. The disease rather than the drug may thus be responsible for the increase in fracture risk.

We observed a consistent increase in overall fracture risk and risk of hip, forearm, and spine fractures with liver and kidney diseases. For kidney diseases, the increase is due to the reduced renal function that is associated with profound changes in calcium and phosphorus metabolism with hyperphosphatemia, hypocalcemia, increased calcium phosphorus product, secondary hyperparathyroidism, reduced activity of 1α-hydroxylase, and the effects on the bone cells of uremic toxins leading to uremic osteodystrophy. In chronic liver disease, reduced absorption by the gut of, e.g., vitamin D may follow from reduced bile excretion. Reduced hydroxylation of vitamin D in the liver may also contribute, along with toxic effects of compounds not removed from the circulation when liver function is failing.

The main strengths of our study are the uniformly organized health-care system, allowing a large-scale population-based design, and the use of data on exposure and confounders that are collected before the date of fracture. Thus, recall bias did not influence data collection.

The weaknesses include potential selection bias, e.g., use of routine hospital discharge diagnoses of fractures coded by hospital doctors to ascertain case status. Some coding errors probably occurred. However, misclassification of case status is unlikely to be related to prescription of immunosuppressive drugs before hospitalization, and any nondifferential misclassification will lead to underestimation of our risk estimates. Moreover, the positive predictive value of hip fracture discharge diagnoses from Danish hospitals has previously been shown to be as high as 93% [46], and the risk of misclassification of case status is thus most likely of minor importance.

We cannot exclude that information biases may have influenced our results; e.g., we had no information on patient compliance in our study since redeeming a prescription was used as a proxy for actual use of a drug. However, the patients paid for part of the cost of the drug, which increases the likelihood of compliance. Furthermore, data on drugs, including immunosuppressants and chemotherapeutics, administered during hospitalization are not registered in the prescription database and therefore not included in our study. Both of these uncertainties could have led to misclassification of exposure. Furthermore, several departments supply the patients with the immunosuppressants, and the drugs are thus not registered in the database. This would have led to an underestimation of the risk associated with the drugs in question.

Although we adjusted for several potential confounding factors in the statistical analyses, our results may still be influenced by potential confounding factors not included in the analyses, e.g., smoking, physical activity, differences in body weight, and use of calcium/vitamin D supplements, or by residual confounding due to the use of crude measures (risk of falling).

Our registers were based on data collected for administrative purposes by health-care professionals and not on the recollection of individual subjects regarding drug use. No differences in the accuracy of data on recent and past exposure were thus present.

Our study was designed as a case-control study. If a cohort design had been chosen instead, the effect of duration of treatment could also be evaluated in a follow-up design.

In conclusion, only weak associations with fracture risk were seen for azathioprine but not for the other immunosuppressants. It thus seems that the underlying disease for which the treatment is administered may be responsible for much of the increase in fracture risk.

Acknowledgment

Danmarks Statistik (Statistics Denmark) is acknowledged for the help without which this project would not have been possible. Research librarian Ms. Edith Clausen is acknowledged for invaluable help with the references. The Danish Medical Research Council granted financial support (grant 22-04-0495).

Copyright information

© Springer Science+Business Media, Inc. 2006