SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Objective

To investigate whether women with systemic sclerosis (SSc) have an increased risk of osteoporosis (OP) and related fractures compared to a high-risk population with rheumatoid arthritis (RA) and also healthy controls, and to determine putative specific OP and fracture risk factors.

Methods

We performed a cross-sectional study with successive inclusion of age-matched healthy women and women with SSc and RA. Risk factors for OP and fracture were collected for all patients. Bone mineral density (BMD) was systematically measured at the lumbar spine and total hip region with dual x-ray absorptiometry.

Results

We included 71 women with SSc, 139 women with RA, and 227 healthy women. The prevalence of OP and fracture was similar in SSc and RA, and was for both diseases higher than in healthy controls (OP: 30% in SSc, 32% in RA, and 11% in controls; fracture: 35% in SSc, 33% in RA, and 10% in controls). Multivariate analysis identified age as a risk factor of OP in SSc. Age and low 25-hydroxyvitamin D (25[OH]D) levels were recognized as risk factors of fracture in SSc. In comparison, age and corticosteroid treatment were associated with OP in RA. Multivariate analysis confirmed age, OP, and low 25(OH)D levels as independent risk factors of fractures in RA.

Conclusion

The prevalence of OP and fracture in SSc was increased compared to healthy women and reached the high prevalence associated with RA. Age and vitamin D deficiency were identified as risk factors of fracture in SSc. Therefore, increasing the awareness and performance of BMD measurements together with the vitamin D supply in patients with SSc is warranted.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Osteoporosis (OP) is a disease of the bone characterized by low bone mass and loss of normal bony architecture and therefore structural support (1). OP is a major cause of morbidity, and furthermore, with hip or even vertebral fracture, mortality can increase (2). An individual's bone density and risk for developing OP are influenced by a number of factors, including peak bone mass, race, older age, family history of OP, decreased sex steroid activity, corticosteroid use, certain chronic diseases that affect intestinal absorption or vitamin D metabolism, smoking, and excessive alcohol use (3).

Many patients with chronic inflammatory diseases have an increased risk of OP and fracture, related in particular to systemic inflammation and/or exposition to long-term corticosteroid therapy. Rheumatoid arthritis (RA) is a chronic inflammatory disorder that is characterized by an increased risk of OP that is not sufficiently explained by corticosteroid use (4). RA is the only secondary cause of OP that is considered independent of bone density in the FRAX algorithm (5). OP is recognized as the main extraarticular complication of RA and results in an increased risk of fractures (6). There is at least a 2-fold increase of fractures in RA, and a higher risk of up to 6-fold has been reported in patients with longstanding disease (7). Systemic sclerosis (SSc) is an orphan and severe connective tissue disorder causing vascular, immune, and fibrotic changes in the skin and some internal organs (8). Theoretically, patients with SSc may have an increased risk of OP as a result of a chronic inflammatory state, occult malabsorption or malnutrition, immobilization, and use of corticosteroid therapy. However, results regarding the risk of OP in SSc are still conflicting in SSc, since previous studies involved different SSc populations, study designs, and generally a relatively small sample size (3, 9–14). Therefore, we aimed to investigate whether patients with SSc have an increased risk of OP and fractures compared to a high-risk population with RA (positive controls) and a population of healthy controls. We also aimed at the identification of potential associations between OP/fracture and patients' phenotypes in order to identify subsets at risk.

Significance & Innovations

  • Point prevalence of osteoporosis and fracture in women with systemic sclerosis is similar to that of rheumatoid arthritis and higher than that of healthy controls.

  • Age and vitamin D deficiency are important risk factors of fracture in systemic sclerosis and rheumatoid arthritis.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Population.

We performed a single-center, cross-sectional, controlled study, including consecutive age-matched women with SSc and RA who had been hospitalized in the Rheumatology A Department for systematic followup in an 18-month period. All SSc patients were classified as having either limited cutaneous or diffuse cutaneous SSc according to the criteria by LeRoy et al (15). All RA patients fulfilled the American College of Rheumatology classification criteria for RA (16). We also included consecutive healthy women, matched for age with women with SSc and RA, referred to our outpatient clinic from their gynecologist for suspicion of OP. The local ethics committee approved the study and all of the subjects gave written informed consent.

Clinical assessment.

Detailed information was collected in all of the participating patients and controls. Demographic data included age, sex, height, weight, and body mass index (BMI; kg/m2). Disease duration was defined as the time elapsed between the onset of the first disease-related symptoms and enrollment. Current use of disease-modifying antirheumatic drugs, including biologic agents or immunosuppressive drugs, was collected. Cutaneous SSc subset was assessed as defined by LeRoy et al (15). SSc activity was assessed according to the preliminary composite index proposed by the European Scleroderma Study Group, and disease severity was assessed according to the severity score by Medsger et al (17, 18). RA activity was quantified by the Health Assessment Questionnaire (HAQ) and the Disease Activity Score using a 28-joint count (DAS28).

Risks factors of OP were also systematically collected, including menopause status, alcohol intake, smoking status, calcium/vitamin D intake, comorbidity, and steroid use (daily and cumulative doses).

Laboratory and systemic assessments.

Laboratory studies were obtained in SSc and RA patients on the morning of the hospital admission, including complete blood cell count, Westergren erythrocyte sedimentation rate (considered elevated at >28 mm/hour), C-reactive protein (CRP) concentration (considered elevated at >10 mg/liter), serum creatinine concentration, calcium, phosphorus, and tests for antinuclear and anticentromere antibodies (both detected in immunofluorescence on HEp-2 cells), anti–topoisomerase I antibodies (by counterimmunoelectrophoresis), rheumatoid factor (by enzyme-linked immunosorbent assay [ELISA]), and second-generation anti–cyclic citrullinated peptide antibody (by ELISA). Level of 25-hydroxyvitamin D (25[OH]D) was measured with the radioimmunoassay DiaSorin kit (Octeia, IDS). According to current recommendations, 25(OH)D concentrations <30 ng/ml were considered as indicating insufficiency, whereas values <10 ng/ml were classified as deficiency (19, 20). Intact parathyroid hormone was assessed with a chemiluminescent enzyme-labeled immunometric assay (Immulite 2000, Siemens Diagnostics) (21). Cardiopulmonary assessment in patients with SSc was performed as previously described (22). Standard radiographs of the hands and wrists were obtained for all patients.

Bone mineral density (BMD) measurements.

For patients with SSc and RA as well as healthy controls, BMDs of the hip (total hip region) and the anteroposterior lumbar spine (L1–L4) were measured by dual x-ray absorptiometry Prodigy (GE Lunar) or QDR4500 (Hologic), which were cross-calibrated. All BMD measurements were carried out by 2 experienced technicians (MM, EM). The World Health Organization (WHO) classification system was applied, defining osteoporosis as a T score less than −2.5 SDs at either the lumbar spine or the hip.

Fracture assessment.

Major nonvertebral osteoporotic fractures were defined as fractures of the hip, wrist, pelvis and sacrum, ribs and sternum, humerus, or femur. These sites are the most relevant for OP-related fractures and important in terms of disability and pain duration (23). Vertebral fractures were systematically assessed with lateral spinal radiographs of the thoracic and lumbar spines, obtained using standardized procedures of acquisition.

Statistical analysis.

The data are shown as means ± SDs for continuous variables and as numbers and percentages for categorical variables, unless stated otherwise. Differences in frequency were examined using the chi-square test. Data were analyzed with the Mann-Whitney unpaired data test. Spearman's rank correlation test was used to assess the relationship between quantitative variables. A multivariable stepwise logistic regression analysis was also performed to determine whether OP or fractures were associated with the various phenotypes of SSc. We included in our model all variables identified with a P value less than or equal to 0.1 in the single-variable analysis (24), and calculated odds ratios (ORs) and 95% confidence intervals (95% CIs). A P value less than 0.05 was considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Study population.

We included 71 women with SSc, with a mean ± SD age of 62 ± 12 years and a mean ± SD disease duration of 10 ± 9 years. All of the SSc patients had Raynaud's phenomenon, 20 (28%) had the diffuse cutaneous subset, and 51 (72%) had the limited cutaneous subset (Table 1). We included 139 women with RA, with a mean ± SD age of 61 ± 11 years and a mean ± SD disease duration of 18 ± 13 years. Detailed RA patient characteristics are provided in Table 2.

Table 1. Characteristics of patients with systemic sclerosis*
 Systemic sclerosis (n = 71)
  • *

    Values are the number (percentage). CT = computed tomography; FVC = forced vital capacity; DLCO/VA = diffusing capacity for carbon monoxide corrected for alveolar volume; RHC = right-sided heart catheterization.

Limited cutaneous/diffuse cutaneous subset51 (72)/20 (28)
Digital ulcers11 (15)
Pulmonary fibrosis on CT scan25 (35)
FVC <75% of predicted18 (25)
DLCO/VA <75% of predicted22 (31)
Pulmonary arterial hypertension on RHC4 (6)
Calcinosis on hand/wrist radiograph14 (20)
Acroosteolysis on hand/wrist radiograph10 (14)
Positive antinuclear antibodies57 (80)
Positive anticentromere antibodies20 (28)
Positive anti–topoisomerase I antibodies11 (15)
Calcium-channel inhibitors71 (100)
Immunosuppressive drugs25 (35)
Biologic agents9 (13)
 Rituximab5 (7)
 Abatacept1 (1)
 Tocilizumab3 (4)
Active disease (Valentini's activity score ≥3)25 (35)
Severe disease (Medsger's severity score ≥3)15 (21)
Table 2. Characteristics of patients with rheumatoid arthritis*
 Rheumatoid arthritis (n = 139)
  • *

    Values are the number (percentage) unless otherwise indicated. Anti–CCP-2 = anti–cyclic citrullinated peptide 2; DMARD = disease-modifying antirheumatic drug; TNFα = tumor necrosis factor α; HAQ = Health Assessment Questionnaire; DAS28 = Disease Activity Score in 28 joints.

Positive rheumatoid factor85 (61)
Positive anti–CCP-2 antibodies73 (53)
Erosions104 (75)
Conventional DMARD133 (96)
Biotherapy55 (40)
 TNFα inhibitors23 (17)
 Rituximab19 (13)
 Abatacept4 (3)
 Tocilizumab9 (6)
HAQ score, mean ± SD1.5 ± 0.9
DAS28 score, mean ± SD3.9 ± 1.6
DAS28 score >3.293 (67)

The SSc and RA patients had a higher BMI (mean ± SD 26 ± 4 and 26 ± 7 kg/m2, respectively, versus 24 ± 4 kg/m2; P < 0.001 for both comparisons) and were more likely to receive vitamin D (69% and 76%, respectively, versus 25%; P < 0.001 for both comparisons) and calcium supplementation (48% and 72%, respectively, versus 14%; P < 0.001 for both comparisons) than healthy controls.

The comparison between the SSc and RA populations for OP risk factors showed that RA patients were more likely to receive calcium supplementation (100 [72%] versus 34 [48%]; P < 0.001) and corticosteroids (129 [93%] versus 41 [58%]; P < 0.001) than SSc patients. The cumulative dose of corticosteroids was significantly higher in patients with RA than SSc (mean ± SD 39,554 ± 29,661 mg versus 19,392 ± 19,333 mg; P < 0.001). CRP levels were also significantly more elevated in RA than in SSc (mean ± SD 12 ± 16 mg/liter versus 7 ± 7.9 mg/liter; P = 0.01). There was no difference for the other OP risk factors (Table 3).

Table 3. Characteristics of SSc and RA patients*
 SSc (n = 71)RA (n = 139)Healthy controls (n = 227)P
SSc vs. controlsRA vs. controlsSSc vs. RA
  • *

    Values are the number (percentage) unless otherwise indicated. SSc = systemic sclerosis; RA = rheumatoid arthritis; NA = not applicable; BMI = body mass index; ND = no data; PTH = parathyroid hormone; CS = corticosteroids; CRP = C-reactive protein.

Age, mean ± SD years62 ± 1261 ± 1160 ± 80.090.30.5
Disease duration, mean ± SD years10 ± 918 ± 13NANANA< 0.001
Menopause63 (89)129 (93)216 (95)0.20.60.6
Age at menopause, mean ± SD years48 ± 449 ± 649 ± 40.070.90.2
Earlier menopause (age ≤40 years)3 (4)8 (5)9 (4)0.70.80.9
BMI, mean ± SD kg/m226 ± 426 ± 724 ± 4< 0.001< 0.0010.9
BMI <19 kg/m23 (4)11 (8)12 (5)0.90.30.8
Current smoker3 (4)17 (12)38 (17)0.010.30.1
Alcohol intake ≥3 units/day1 (1)5 (4)15 (6)0.20.60.4
Vitamin D deficiency40 (56)97 (70)NDNDND0.1
Vitamin D supplementation49 (69)105 (76)56 (25)< 0.001< 0.0010.3
Elevated intact PTH >65 ng/liter12 (17)35 (25)NDNDND0.3
Calcium, mean ± SD mmoles/liter2.33 ± 0.192.33 ± 0.10NDNDND0.9
Phosphorus, mean ± SD mmoles/liter1.16 ± 0.221.11 ± 0.18NDNDND0.08
Calcium supplementation34 (48)100 (72)32 (14)< 0.001< 0.001< 0.001
CS41 (58)129 (93)NANANA< 0.001
Cumulative dose of CS, mean ± SD mg19,392 ± 19,33339,554 ± 29,661NANANA< 0.001
Biologic agents7 (10)55 (40)NANANA< 0.001
CRP level, mean ± SD mg/liter7.0 ± 7.912 ± 16NDNDND0.01
CRP level >10 mg/liter11 (15)42 (30)NDNDND0.03

At the time of inclusion, 23 SSc patients (32%) and 42 RA patients (30%) were receiving treatment for OP: bisphosphonates in 22 SSc patients and 38 RA patients, strontium ranelate in 1 SSc patient and 3 RA patients, and teriparatide in 1 RA patient. Eleven healthy controls (5%) were receiving treatment for OP: bisphosphonates in 8 patients, selective estrogen receptor modulator in 2 patients, and strontium ranelate in 1 patient.

Point prevalence of OP and fractures.

The point prevalence of OP according to the WHO classification was 30% in SSc (21 of 71 patients) and 32% in RA (45 of 139 patients; P = 0.9), which was significantly higher than the prevalence observed in the control group (26 [11%] of 227 patients; P < 0.001 versus SSc and P < 0.001 versus RA).

The frequency of total hip and lumbar spine OP was increased in SSc and RA patients compared to healthy controls (total hip: 12 [17%] of 71 and 26 [19%] of 139, respectively, versus 13 [6%] of 227; P = 0.009 versus SSc and P < 0.001 versus RA, and lumbar spine: 18 [25%] of 71 and 29 [21%] of 139, respectively, versus 19 [8%] of 227; P < 0.001 for both comparisons), but was not significantly different between SSc and RA patients (P = 0.9 and P = 0.6 for the total hip and lumbar spine, respectively).

In addition, BMD measured on the total hip (mean ± SD 0.817 ± 0.148 gm/cm2 versus 0.788 ± 0.168 gm/cm2; P = 0.2) and lumbar spine (mean ± SD 0.963 ± 0.188 gm/cm2 versus 0.987 ± 0.197 gm/cm2; P = 0.4) did not differ between SSc and RA patients, but was significantly decreased compared to those measures in the healthy controls (mean ± SD 0.901 ± 0.136 gm/cm2 in the total hip; P < 0.001 versus SSc and RA, and mean ± SD 1.058 ± 0.169 gm/cm2 in the lumbar spine; P < 0.001 versus SSc and RA) (Figure 1A and B).

thumbnail image

Figure 1. Bone mineral density (BMD) measured on A, the anteroposterior lumbar spine (L1–L4), and B, the total hip region by dual x-ray absorptiometry in patients with systemic sclerosis (SSc) and rheumatoid arthritis (RA) and controls. * = P < 0.05 versus SSc and RA.

Download figure to PowerPoint

The point prevalence of osteoporotic fractures was 35% (25 of 71) and 33% (46 of 139) in the SSc and RA patients, respectively (P = 0.9), and was 10% (24 of 227) in the healthy controls (P < 0.001 versus SSc and RA). The frequency of vertebral (18 [25%] of 71 SSc patients versus 26 [19%] of 139 RA patients) and nonvertebral (16 [23%] of 71 SSc patients versus 31 [22%] of 139 RA patients) osteoporotic fractures did not differ significantly between SSc and RA patients, but was significantly increased compared to the controls (2 [1%] of 227 individuals with vertebral fractures; P < 0.001 versus SSc and RA, and 22 [10%] of 227 individuals with nonvertebral fractures; P = 0.008 versus SSc and P = 0.003 versus RA).

Risk factors for OP and fractures in SSc patients.

SSc patients with OP (n = 21) were significantly older (mean ± SD age 67 ± 12 years versus 60 ± 10 years; P = 0.03) and had a significantly higher disease duration (mean ± SD 15 ± 10 years versus 8 ± 8 years; P < 0.001) than patients without OP. No association was identified between OP and any other SSc-related characteristics in univariate analysis, especially with the cutaneous subset, heart/lung involvement, calcinosis, and acroosteolysis on hand radiographs. In addition, we found no association between OP and BMI, vitamin D deficiency, cumulative dose of corticosteroids, and systemic inflammation (Table 4). There was no correlation between BMD measured on the lumbar spine and hip and cumulative dose of corticosteroids (r = −0.28, P = 0.09 and r = −0.15, P = 0.3, respectively).

Table 4. Factors associated with osteoporosis in SSc and RA patients*
 Patients with osteoporosisPatients without osteoporosisPStepwise logistic regression, OR (95% CI)
  • *

    SSc = systemic sclerosis; RA = rheumatoid arthritis; OR = odds ratio; 95% CI = 95% confidence interval; PTH = parathyroid hormone; CS = corticosteroids; CRP = C-reactive protein; BMI = body mass index; HAQ = Health Assessment Questionnaire; DAS28 = Disease Activity Score in 28 joints.

  • P ≤ 0.1. Included in the stepwise logistic regression analysis.

  • Disease duration was not included in the multivariate model, since this parameter correlated with age and was dependent of age for the risk of osteoporosis.

SSc patients with osteoporosis (n = 21)    
 Age, mean ± SD years67 ± 1260 ± 100.031.16 (1.01–1.23)
 Disease duration, mean ± SD years15 ± 108 ± 8< 0.001 
 Menopause, no. (%)19 (90)44 (88)0.9 
 Age at menopause, mean ± SD years49 ± 948 ± 90.3 
 Vitamin D deficiency, no. (%)15 (71)25 (50)0.1 
 Calcium supplementation, no. (%)14 (67)20 (40)0.07 
 Elevated intact PTH >65 ng/liter, no. (%)5 (24)7 (14)0.5 
 CS, no. (%)14 (67)27 (54)0.5 
 Cumulative dose of CS, mean ± SD mg24,274 ± 13,45817,541 ± 10,8790.3 
 CRP level >10 mg/liter, no. (%)4 (19)7 (14)0.9 
RA patients with osteoporosis (n = 45)    
 Age, mean ± SD years65 ± 1359 ± 11< 0.0011.12 (1.02–1.26)
 Disease duration, mean ± SD years17 ± 1218 ± 130.7 
 Menopause, no. (%)44 (98)85 (90)0.2 
 Age at menopause, mean ± SD years51 ± 849 ± 70.05 
 BMI <19 kg/m2, no. (%)3 (7)8 (9)0.9 
 Vitamin D deficiency, no. (%)37 (82)60 (64)0.05 
 Calcium supplementation, no. (%)39 (87)61 (65)0.01 
 Elevated intact PTH >65 ng/liter, no. (%)12 (27)23 (24)0.8 
 CS, no. (%)40 (89)89 (95)0.3 
 Cumulative dose of CS, mean ± SD mg40,051 ± 19,67228,574 ± 16,8900.0043.21 (1.12–10.44)
 CRP level >10 mg/liter, no. (%)15 (33)27 (29)0.8 
 HAQ score, mean ± SD1.80 ± 0.581.43 ± 0.770.1 
 DAS28 score, mean ± SD3.87 ± 1.123.89 ± 1.140.9 

In stepwise logistic regression analysis including all variables identified with a P value ≤0.1 in the single-variable analysis except disease duration, which was not included because it correlated with age in SSc patients (r = 0.4, P = 0.003), only age was identified as an independent risk factor of OP (OR 1.16, 95% CI 1.01–1.23).

SSc patients with fractures (n = 25) were significantly older (mean ± SD age 69 ± 11 years versus 58 ± 10 years; P < 0.001) and were more likely to have a higher disease duration (mean ± SD 13 ± 11 years versus 9 ± 9 years; P = 0.03) than patients without fractures. Other risk factors for fractures in univariate analysis were the presence of osteoporosis (13 [52%] of 25 versus 8 [17%] of 46; P = 0.005), low 25(OH)D levels (<30 ng/ml; 19 [76%] of 25 versus 21 [46%] of 46; P = 0.03), calcium intake (17 [68%] of 25 versus 17 [37%] of 46; P = 0.02), and current antiosteoporotic treatment (14 [56%] of 25 versus 9 [20%] of 46; P = 0.01) (Table 5). No association was found between fracture and the cutaneous subset, heart/lung involvement, calcinosis, and acroosteolysis on hand radiographs. In multivariate analysis (after exclusion of disease duration, which was dependent of age for the risk of osteoporosis), only age (OR 1.18, 95% CI 1.03–1.26) and low 25(OH)D levels (<30 ng/ml; OR 6.19, 95% CI 1.66–23.11) were identified as independent risk factors for fractures.

Table 5. Factors associated with fractures in SSc and RA patients*
 Patients with fracturePatients without fracturePStepwise logistic regression, OR (95% CI)
  • *

    SSc = systemic sclerosis; RA = rheumatoid arthritis; OR = odds ratio; 95% CI = 95% confidence interval; PTH = parathyroid hormone; CS = corticosteroids; CRP = C-reactive protein; BMI = body mass index; HAQ = Health Assessment Questionnaire; DAS28 = Disease Activity Score in 28 joints.

  • P ≤ 0.1. Included in the stepwise logistic regression analysis.

  • Disease duration was not included in the multivariate model, since this parameter correlated with age and was dependent of age for the risk of osteoporosis.

SSc patients with fractures (n = 25)    
 Age, mean ± SD years69 ± 1158 ± 10< 0.0011.18 (1.03–1.26)
 Disease duration, mean ± SD years13 ± 119 ± 90.03 
 Menopause, no. (%)25 (100)38 (83)0.07 
 Age at menopause, mean ± SD years48 ± 1049 ± 80.7 
 Osteoporosis, no. (%)13 (52)8 (17)0.005 
 Vitamin D deficiency, no. (%)19 (76)21 (46)0.036.19 (1.66–23.11)
 Elevated intact PTH >65 ng/liter, no. (%)4 (16)8 (17)0.8 
 Calcium supplementation, no. (%)17 (68)17 (37)0.02 
 CS, no. (%)16 (64)25 (54)0.6 
 Cumulative dose of CS, mean ± SD mg24,258 ± 12,25716,772 ± 10,3270.2 
 CRP level >10 mg/liter, no. (%)3 (12)8 (17)0.8 
 Antiosteoporotic treatment, no. (%)14 (56)9 (20)0.01 
RA patients with fractures (n = 46)    
 Age, mean ± SD years66 ± 1058 ± 11< 0.0011.09 (1.03–1.18)
 Disease duration, mean ± SD years18 ± 1118 ± 100.9 
 Menopause, no. (%)44 (96)85 (91)0.5 
 Osteoporosis, no. (%)24 (52)21 (23)0.0013.22 (1.18–8.75)
 Age at menopause, mean ± SD years49 ± 849 ± 90.9 
 BMI <19 kg/m2, no. (%)3 (7)8 (9)0.9 
 Vitamin D deficiency, no. (%)40 (87)57 (61)0.0034.31 (1.29–14.41)
 Elevated intact PTH >65 ng/liter, no. (%)13 (28)22 (24)0.9 
 Calcium supplementation, no. (%)41 (89)59 (63)0.003 
 CS, no. (%)43 (93)86 (92)0.9 
 Cumulative dose of CS, mean ± SD mg42,785 ± 21,36937,848 ± 20,6100.4 
 CRP level >10 mg/liter, no. (%)18 (39)24 (26)0.2 
 Biologic agents, no. (%)12 (26)43 (46)0.030.34 (0.12–0.95)
 Antiosteoporotic treatment, no. (%)12 (26)30 (32)0.6 
 HAQ score, mean ± SD1.71 ± 0.461.46 ± 0.310.4 
 DAS28 score, mean ± SD3.92 ± 1.093.69 ± 0.870.8 

Risk factors for OP and fractures in RA patients.

In univariate analysis, patients with RA and OP (n = 45) were more likely to be older (mean ± SD age 65 ± 13 years versus 59 ± 11 years; P < 0.001), receive calcium supplementation (39 [87%] of 45 versus 61 [65%] of 94; P = 0.01), and have been treated with higher cumulative doses of corticosteroids (mean ± SD 40,051 ± 19,672 mg versus 28,574 ± 16,890 mg; P = 0.004) than patients without OP (Table 4). In addition, cumulative dose of corticosteroids negatively correlated with BMD measured at the lumbar spine (r = −0.38, P = 0.01) and total hip (r = −0.49, P = 0.008) in RA patients. HAQ score, DAS28, and CRP levels were not associated with OP. In multivariate analysis, age (OR 1.12, 95% CI 1.02–1.26) and cumulative dose of corticosteroids (OR 3.21, 95% CI 1.12–10.44) were confirmed as independent risk factors of OP in RA (Table 4).

The likelihood of fracture in RA patients was higher in older patients (mean ± SD age 66 ± 10 years versus 58 ± 11 years; P < 0.001) and in patients with OP (24 [52%] of 46 versus 21 [23%] of 93; P = 0.001), low 25(OH)D levels (<30 ng/ml; 40 [87%] of 46 versus 57 [61%] of 93; P = 0.003), or calcium supplementation (41 [89%] of 46 versus 59 [63%] of 93; P = 0.003). Moreover, RA patients receiving biologic agents were less likely to have fracture (12 [26%] of 46 versus 43 [46%] of 93; P = 0.03). HAQ, DAS28, systemic inflammation, and current antiosteoporotic treatment were not associated with the presence of fractures. Multivariate analysis confirmed age (OR 1.09, 95% CI 1.03–1.18), OP (OR 3.22, 95% CI 1.18–8.75), and low 25(OH)D levels (<30 ng/ml; OR 4.31, 95% CI 1.29–14.41) as independent risk factors of fractures, and treatment with biologic agents as a protective factor of fractures (OR 0.34, 95% CI 0.12–0.95).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

We systematically evaluated, in a controlled study, OP and fracture prevalence in a cohort of hospitalized women with SSc compared to age-matched healthy controls and a high-risk population of patients with RA. We found that the prevalence of OP and fracture in our cohort of patients with SSc was increased compared to our population of healthy controls and similar to our positive RA control group, highlighting an increased risk of OP and fracture in SSc. We also identified age and vitamin D deficiency as independent risk factors of fracture in both SSc and RA populations.

The prevalence of OP in our SSc population was 30%, which is consistent with the current literature (range 7–51%, mean 24%), and significantly increased compared to a group of healthy age-matched women referred for OP suspicion. The prevalence of OP in women with SSc was similar to a large group of age-matched women with RA who might have more active and severe disease because they were recruited from a tertiary referral center, highlighting the high risk of OP in SSc (3, 11–14). As in our subsets, several authors reported no difference in OP between patients with the limited cutaneous or diffuse cutaneous subset (12, 14). Despite the fact that 90% of SSc patients were postmenopausal, we did not observe an association between age at menopause and OP diagnosis, as observed in a previous study (9). In addition, we did not find an earlier age at menopause in our SSc cohort as compared to a previous study (9).

Consistent with the frequency reported in the German network for SSc, 58% of SSc patients received corticosteroids (25). Despite their large prescription, this therapy did not influence the outcome of the diagnosis of OP. In addition, cumulative dose of steroids did not differ significantly between women with SSc with or without OP and did not correlate with lumbar spine and hip BMD. This is consistent with the results of an analysis of 61 women with SSc that found no statistical association between BMD values and previous use of corticosteroids (12). This result differed from the RA population: cumulative doses of corticosteroids were associated with OP in multivariate analysis and negatively correlated with BMD measurements. Since patients with RA received significantly higher cumulative doses of corticosteroids than patients with SSc, it may be speculated that SSc patients who took steroids were exposed for a relatively short period of time and/or in low doses, which had no influence on the development of OP.

The fracture rate in SSc was similar to RA (35% versus 33%) and higher than in healthy controls. This prevalence was increased compared to the prevalence previously reported in SSc (3). This may be explained by the relatively young age of the patients previously studied as compared to our population. OP is one of the most common comorbidities in patients with RA, and has been reported in 10–56% of patients depending on the populations studied, which is in accordance with our results (26). As a consequence, patients with RA are at a substantially higher risk of experiencing fracture compared with the general population. The similar frequency of fracture in our SSc and RA population underlines the high risk of fracture in SSc and supports the systematic screening for this complication, especially for vertebral fractures, which may be ignored because of other symptoms in patients with painful diseases such as SSc or RA (4).

As expected, our study confirmed that age was an independent risk factor of fracture, which is consistent with literature (27, 28). The diagnosis of SSc and RA was related to the presence of fractures independently of corticosteroid use. A lack of relationship between cumulative corticosteroid dose and presence of fractures was also observed in these 2 diseases. This was in accordance with a previous prospective study conducted in 255 RA patients that failed to demonstrate a link between incident fractures and baseline or followup use of corticosteroids (29). In SSc, no publication has ever reported any association between corticosteroid and the risk of fracture, and further studies are warranted to confirm this absence of a link.

Previous studies have highlighted the relationship between fractures and systemic inflammation in RA and other inflammatory diseases, such as ankylosing spondylitis (30, 31). In addition, an optimal control of inflammation has been shown to reduce bone loss in these inflammatory diseases (32). Even in the general population, the highly sensitive CRP level is a predictor of the risk of fracture (33). Our study failed to identify CRP level as a risk factor of OP and fracture in SSc and RA patients. This might be explained by the cross-sectional design of this study, with a single measure of CRP at the time of BMD measurement, and by the low number of patients with systemic inflammation, especially in the SSc subgroup (15% and 30% of patients with a CRP level >10 mg/liter in SSc and RA, respectively). The absence of an association between fracture and steroids or systemic inflammation was not explained by concomitant use of antiosteoporotic treatment, since this parameter was not significant in the final statistical model.

Low 25(OH)D levels were identified in SSc as an independent risk factor of fracture. Although patients with RA were more likely to receive calcium and vitamin D supplementation than SSc patients, since the risk of OP in RA is better recognized, low 25(OH)D levels were also predictive of fracture in RA. Low 25(OH)D levels have been linked to the risk of fracture in white women (34). Vitamin D deficiency can cause secondary hyperparathyroidism, high bone turnover, low BMD, and mineralization defects, all of which could contribute to an increased factor risk (34). Several studies have recently reported a high prevalence of vitamin D deficiency in SSc and RA (21, 35–38). In SSc, vitamin D deficiency was independent of geographic origin and vitamin D supplementation. This suggests that common vitamin D supplementation does not correct the deficiency in SSc patients, and that higher doses are probably required. Considering the prevalence of vitamin D deficiency and the risk of fracture related to this condition in patients with SSc and RA, screening and supplementation should be performed routinely. In addition, a higher dose of vitamin D is probably needed in SSc patients because common vitamin D supplementation does not correct the deficiency (21, 37).

Our study has some limitations that deserve consideration. This was an observational study. The disease phenotype associations have to be interpreted cautiously since the study was cross-sectional. Moreover, the different disease characteristics identified were associated with prevalent fractures, and not with incident fractures. Some factors, including early menopause and malabsorption, were not studied because the number of patients with these complications was too limited. The influence of proton-pump inhibitors (PPIs) as a risk factor of OP and fracture was not assessed, since all SSc patients received high doses of PPIs.

It is noteworthy that BMD incompletely detected patients with fractures, since only 52% of SSc and RA patients with fractures had OP. These data suggest that other parameters, especially the bone microarchitecture, have to be taken into account and support the evaluation of other techniques to assess the risk of fracture in SSc and RA, such as the trabecular bone score (39).

In conclusion, our study showed that OP and fracture point prevalence in women with SSc was similar to that of RA and higher than that of healthy controls, highlighting an increased risk of OP and fracture. Age and vitamin D deficiency were found to be important risk factors of fracture in SSc and RA. Systematic screening for OP based on BMD measurements, detection of fracture, and supply of vitamin D is warranted based on our results in patients with SSc, especially for the elderly.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Avouac had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Avouac, Koumakis, Kahan, Allanore.

Acquisition of data. Avouac, Koumakis, Toth, Meunier, Maury, Kahan.

Analysis and interpretation of data. Avouac, Kahan, Cormier, Allanore.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES
  • 1
    La Montagna G, Baruffo A, Abbadessa S, Maja L, Tirri R. Evidence for bone resorption in systemic sclerosis. J Rheumatol 1995; 22: 7979.
  • 2
    Loucks J, Pope JE. Osteoporosis in scleroderma. Semin Arthritis Rheum 2005; 34: 67882.
  • 3
    Yuen SY, Rochwerg B, Ouimet J, Pope JE. Patients with scleroderma may have increased risk of osteoporosis: a comparison to rheumatoid arthritis and noninflammatory musculoskeletal conditions. J Rheumatol 2008; 35: 10738.
  • 4
    Ghazi M, Kolta S, Briot K, Fechtenbaum J, Paternotte S, Roux C. Prevalence of vertebral fractures in patients with rheumatoid arthritis: revisiting the role of glucocorticoids. Osteoporos Int 2012; 23: 5817.
  • 5
    Broy SB, Tanner SB. Official positions for FRAX clinical regarding rheumatoid arthritis from Joint Official Positions Development Conference of the International Society for Clinical Densitometry and International Osteoporosis Foundation on FRAX. J Clin Densitom 2011; 14: 1849.
  • 6
    Van Staa TP, Geusens P, Bijlsma JW, Leufkens HG, Cooper C. Clinical assessment of the long-term risk of fracture in patients with rheumatoid arthritis. Arthritis Rheum 2006; 54: 310412.
  • 7
    Spector TD, Hall GM, McCloskey EV, Kanis JA. Risk of vertebral fracture in women with rheumatoid arthritis. BMJ 1993; 306: 558.
  • 8
    Allanore Y, Avouac J, Wipff J, Kahan A. New therapeutic strategies in the management of systemic sclerosis. Expert Opin Pharmacother 2007; 8: 60715.
  • 9
    La Montagna G, Vatti M, Valentini G, Tirri G. Osteopenia in systemic sclerosis: evidence of a participating role of earlier menopause. Clin Rheumatol 1991; 10: 1822.
  • 10
    Frediani B, Baldi F, Falsetti P, Acciai C, Filippou G, Spreafico A, et al. Bone mineral density in patients with systemic sclerosis [letter]. Ann Rheum Dis 2004; 63: 3267.
  • 11
    Souza RB, Borges CT, Takayama L, Aldrighi JM, Pereira RM. Systemic sclerosis and bone loss: the role of the disease and body composition. Scand J Rheumatol 2006; 35: 3847.
  • 12
    Sampaio-Barros PD, Costa-Paiva L, Filardi S, Sachetto Z, Samara AM, Marques-Neto JF. Prognostic factors of low bone mineral density in systemic sclerosis. Clin Exp Rheumatol 2005; 23: 1804.
  • 13
    Frediani B, Baldi F, Falsetti P, Acciai C, Filippou G, Spreafico A, et al. Clinical determinants of bone mass and bone ultrasonometry in patients with systemic sclerosis. Clin Exp Rheumatol 2004; 22: 3138.
  • 14
    Neumann K, Wallace DJ, Metzger AL. Osteoporosis: less than expected in patients with scleroderma? [Letter]. J Rheumatol 2000; 27: 18223.
  • 15
    LeRoy EC, Black C, Fleischmajer R, Jablonska S, Krieg T, Medsger TA Jr, et al. Scleroderma (systemic sclerosis): classification, subsets and pathogenesis. J Rheumatol 1988; 15: 2025.
  • 16
    Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988; 31: 31524.
  • 17
    Valentini G, Silman AJ, Veale D. Assessment of disease activity. Clin Exp Rheumatol 2003; 21 Suppl: S3941.
  • 18
    Medsger TA Jr, Bombardieri S, Czirjak L, Scorza R, Della Rossa A, Bencivelli W. Assessment of disease severity and prognosis. Clin Exp Rheumatol 2003; 21 Suppl: S426.
  • 19
    Hollis BW, Wagner CL. Normal serum vitamin D levels [letter]. N Engl J Med 2005; 352: 5156.
  • 20
    Holick MF. Vitamin D deficiency. N Engl J Med 2007; 357: 26681.
  • 21
    Vacca A, Cormier C, Piras M, Mathieu A, Kahan A, Allanore Y. Vitamin D deficiency and insufficiency in 2 independent cohorts of patients with systemic sclerosis. J Rheumatol 2009; 36: 19249.
  • 22
    Avouac J, Meune C, Ruiz B, Couraud PO, Uzan G, Boileau C, et al. Angiogenic biomarkers predict the occurrence of digital ulcers in systemic sclerosis. Ann Rheum Dis 2012; 71: 3949.
  • 23
    Reginster JY, Seeman E, De Vernejoul MC, Adami S, Compston J, Phenekos C, et al. Strontium ranelate reduces the risk of nonvertebral fractures in postmenopausal women with osteoporosis: Treatment of Peripheral Osteoporosis (TROPOS) study. J Clin Endocrinol Metab 2005; 90: 281622.
  • 24
    Allanore Y, Meune C, Vonk MC, Airo P, Hachulla E, Caramaschi P, et al. Prevalence and factors associated with left ventricular dysfunction in the EULAR Scleroderma Trial and Research group (EUSTAR) database of systemic sclerosis patients. Ann Rheum Dis 2010; 69: 21821.
  • 25
    Hunzelmann N, Moinzadeh P, Genth E, Krieg T, Lehmacher W, Melchers I, et al. High frequency of corticosteroid and immunosuppressive therapy in patients with systemic sclerosis despite limited evidence for efficacy. Arthritis Res Ther 2009; 11: R30.
  • 26
    Edwards CJ, Williams E. The role of interleukin-6 in rheumatoid arthritis-associated osteoporosis. Osteoporos Int 2010; 21: 128793.
  • 27
    Looker AC, Wahner HW, Dunn WL, Calvo MS, Harris TB, Heyse SP, et al. Proximal femur bone mineral levels of US adults. Osteoporos Int 1995; 5: 389409.
  • 28
    Orwoll ES, Bauer DC, Vogt TM, Fox KM, for the Study of Osteoporotic Fractures Research Group. Axial bone mass in older women. Ann Intern Med 1996; 124: 18796.
  • 29
    Orstavik RE, Haugeberg G, Uhlig T, Mowinckel P, Falch JA, Halse JI, et al. Incidence of vertebral deformities in 255 female rheumatoid arthritis patients measured by morphometric x-ray absorptiometry. Osteoporos Int 2005; 16: 3542.
  • 30
    Vosse D, Landewe R, van der Heijde D, van der Linden S, van Staa TP, Geusens P. Ankylosing spondylitis and the risk of fracture: results from a large primary care-based nested case-control study. Ann Rheum Dis 2009; 68: 183942.
  • 31
    Gough AK, Lilley J, Eyre S, Holder RL, Emery P. Generalised bone loss in patients with early rheumatoid arthritis. Lancet 1994; 344: 237.
  • 32
    Wijbrandts CA, Klaasen R, Dijkgraaf MG, Gerlag DM, van Eck-Smit BL, Tak PP. Bone mineral density in rheumatoid arthritis patients 1 year after adalimumab therapy: arrest of bone loss. Ann Rheum Dis 2009; 68: 3736.
  • 33
    Schett G, Kiechl S, Weger S, Pederiva A, Mayr A, Petrangeli M, et al. High-sensitivity C-reactive protein and risk of nontraumatic fractures in the Bruneck study. Arch Intern Med 2006; 166: 2495501.
  • 34
    Cauley JA, Lacroix AZ, Wu L, Horwitz M, Danielson ME, Bauer DC, et al. Serum 25-hydroxyvitamin D concentrations and risk for hip fractures. Ann Intern Med 2008; 149: 24250.
  • 35
    Caramaschi P, Dalla Gassa A, Ruzzenente O, Volpe A, Ravagnani V, Tinazzi I, et al. Very low levels of vitamin D in systemic sclerosis patients. Clin Rheumatol 2011; 29: 141925.
  • 36
    Kerr GS, Sabahi I, Richards JS, Caplan L, Cannon GW, Reimold A, et al. Prevalence of vitamin D insufficiency/deficiency in rheumatoid arthritis and associations with disease severity and activity. J Rheumatol 2011; 38: 539.
  • 37
    Rios Fernandez R, Fernandez Roldan C, Callejas Rubio JL, Ortego Centeno N. Vitamin D deficiency in a cohort of patients with systemic scleroderma from the south of Spain [letter]. J Rheumatol 2010; 37: 13556.
  • 38
    Belloli L, Ughi N, Marasini B. Vitamin D in systemic sclerosis [letter]. Clin Rheumatol 2011; 30: 1456.
  • 39
    Hans D, Goertzen AL, Krieg MA, Leslie WD. Bone micro-architecture assessed by TBS predicts osteoporotic fractures independent of bone density: the Manitoba study. J Bone Miner Res 2011; 26: 27629.