Abstract
Objective. To study the BMD of patients with SLE according to the age of disease onset.
Methods. Consecutive SLE patients were screened for BMD at the hip, lumbar spine and whole body by the dual-energy X-ray absorptiometry (DXA). Comparison was made between patients who had disease onset in childhood (<18 years) and adulthood (≥18 years). Factors associated with low BMD were studied by linear regression.
Results. A total of 395 SLE patients were studied (94% women; 11% childhood-onset disease). Osteoporosis of the lumbar spine and the hip/femoral neck was present in 20 and 10% of the patients, respectively. Childhood-onset SLE patients were less likely to be post-menopausal, but had significantly lower BMI, longer SLE duration and a higher frequency of ever use of high-dose CSs, CYC and AZA. Despite a significantly younger age, the BMD of the hip, femoral neck and lumbar spine was significantly lower in childhood- than adult-onset SLE patients. In linear regression models, childhood-onset disease was an independent factor for lower BMD at the lumbar spine (β = −0.18; P = 0.002), hip (β = −0.20; P = 0.001) and femoral neck (β = −0.16; P = 0.01) after adjustment for age, sex, BMI, smoking, menopause, SLE duration and damage index, duration and current dose of prednisolone treatment and the ever use of high-dose glucocorticoids, other immunosuppressive agents, calcium, vitamin D and the bisphosphonates.
Conclusions. In adult SLE patients, childhood-onset disease carries a higher risk of osteoporosis, which may possibly be related to a higher cumulative dose of glucocorticoids used for more active disease and failure to achieve a normal peak bone mass during puberty.
Introduction
Patients with SLE are prone to osteoporosis and its complications. Possible contributing factors include chronic glucocorticoid therapy, renal insufficiency, premature ovarian failure, avoidance of sun exposure, disabling arthritis and myopathy, failure to achieve a peak bone mass (PBM) during adolescence and the use of other medications such as anti-coagulants and anti-convulsants [1, 2]. Persistent disease activity with elevation of cytokines such as IL-6 and TNF-α may also activate the osteoclasts and aggravate the loss of BMD [3].
Several studies have demonstrated that BMD in pre-menopausal women with SLE was significantly lower than that of age- and gender-matched healthy controls [4–6]. Osteoporosis occurred in 3–42% of SLE patients, depending on study design, ethnicity, age and proportion of post-menopausal patients in the study cohorts, duration of SLE and CS treatment. A retrospective study in the USA described an incidence of 12.3% of self-reported fractures in women with SLE [7]. Fracture showed nearly a 5-fold increase compared with women of similar age in the general population. Another cross-sectional study in UK showed that fragility fractures occurred in 9.1% of patients since SLE diagnosis [8].
Information on osteoporosis in childhood or juvenile-onset SLE is scant. A case–control study reported lower BMD at the femoral neck and lumbar spine in JSLE patients than matched controls [9]. Another uncontrolled study also reported a high incidence of lumbar spine osteoporosis in JSLE patients [10]. In these two studies, osteopenia (Z-score between −1.0 and −2.5) of the hip and the lumbar spine ranged from 16–40 to 38–41%, respectively, in SLE patients. The corresponding figures for osteoporosis (Z-score < −2.5) of the hip and lumbar spine were 7–9 and 9–20%, respectively. A longitudinal study of 20 JSLE patients receiving glucocorticoid treatment revealed a 3.5% loss of BMD after 1 year [11]. Finally, a recent study of 26 patients with SLE or overlap syndromes who were initiated glucocorticoids within a 30-day period reported that vertebral fracture occurred in 8% of these patients [12].
In the current study, we examined the BMD of the hip and spine in a large number of adult patients with SLE and factors associated with low BMD in these patients. Comparison of BMD was also made between patients who had childhood- and adult-onset SLE.
Patients and methods
Consecutive patients who were followed in our lupus clinics and fulfilled at least four of the ACR criteria for the classification of SLE [13] were recruited for screening of BMD at the hip, femoral neck, lumbar spine and whole body by using the dual-energy X-ray absorptiometry (DXA) scan. Data on demographic characteristics, risk factors for osteoporosis and use of medications were collected. Comparison was made between patients who had onset of their SLE in childhood or adolescence (<18 years) and adulthood (≥18 years). Linear regression models were established to study the factors associated with low BMD. Informed consent was obtained from all the participants and the study was approved by the Research and Ethics Committee of Tuen Mun Hospital.
DXA procedure
BMD at various body sites [lumbar spine (L2–4), non-dominant hip, femoral neck and whole body] were measured by the DXA technique using Delphi densitometer (Hologic, Bedford, USA). The precision of BMD measurement in the lumbar spine (L2–4), left femoral neck, left total hip and whole body as determined by a pre-service study were 1.14, 1.91, 1.03 and 0.77%, respectively. For patients with avascular bone necrosis of the hip or joint replacement, the BMD of the other hip was used. When avascular necrosis or hip replacement occurred on both sides, BMD data were not used for analysis. This also applied to patients who had collapse or severe scoliosis/deformity of the lumbar spine. The reference ranges for the T and Z scores were derived from the third National Health and Nutrition Examination Survey (NHANES III) database (hip) and the device manufacturer's data set (lumbar spine) [14]. The technician who was responsible for measuring the BMD of all participants was blinded for the details of the current study and the subsequent data analyses. The BMD of all patients were measured by the same DXA machine and software throughout.
Assessment of SLE disease activity and damage
Disease activity of SLE at the time of DXA scan was assessed by the safety of oestrogens in lupus erythematosus national assessment (SELENA)-SLEDAI, a validated instrument employed in the SELENA trials [15]. Damage of SLE was measured by the systemic lupus international collaborating clinics damage index (SDI) [16], a validated instrument consisting of 41 items that measure irreversible organ damage unrelated to active inflammation in 12 organ systems. Each item should be present for at least 6 consecutive months in order to be scored.
Statistical analyses
Unless otherwise stated, values in this study were expressed as mean (s.d.). Comparison of continuous variables between childhood- and adult-onset SLE patients was performed by the independent Student's t-test. Categorical variables between the two groups of patients were compared by the chi-square test. Fisher's exact test was used when the frequency was <5. Multivariate analysis for factors associated with BMD at the lumbar spine, hip and femoral neck (outcome variables) was performed by linear regression. Factors considered to be associated with BMD in the regression models were: childhood- vs adult-onset disease, age, sex, BMI, SLE duration, smoking, menopause, duration of glucocorticoid treatment, current prednisolone dose, SLE damage index, diabetes mellitus, hypertension, serum creatinine level and ever use of non-glucocorticoid immunosuppressive agents, calcium, vitamin D and the bisphosphonates. The slope, s.e. and standardized regression coefficients (β) were calculated for each of the covariates tested. Statistical significance was defined as a two-tailed P-value of <0.05. All statistical analyses were performed using the SPSS program, version 11.5 (SPSS, Chicago, IL, USA) for Windows XP.
Results
Study population
A total of 395 SLE patients were studied. Forty-two (11%) patients had childhood-onset disease, whereas 353 (89%) had adult-onset SLE. Three hundred and seventy-one (94%) patients were women. The mean (s.d.) age at the time of DXA scan was 40.4 (13) years and the mean (s.d.) SLE duration was 7.8 (7.1) years. The distribution of the age of SLE diagnosis in childhood-onset patients was as follows: ≤10 years (5 patients; 12%), 11 or 12 years (5 patients; 12%), 13 or 14 years (12 patients; 29%), 15 or 16 years (16 patients; 38%) and 17 years (4 patients; 10%). The demographic and clinical characteristics of these two groups of patients are shown in Table 1. At the time of DXA scan, 126 (32%) patients were not taking CSs; 30 (8%) patients were receiving ≤2.5 mg/day of prednisolone; 146 (37%) were receiving >2.5–5 mg/day; 64 (16%) were receiving >5–7.5 mg/day and 29 (7%) patients were taking >7.5 mg/day of prednisolone. Fourteen (4%) patients had a serum creatinine level of >120 µmol/l.
Demographic and clinical characteristics
| Covariates | Childhood onset (n = 42) | Adult onset (n = 353) | P-value |
|---|---|---|---|
| Gender: female, n (%) | 40 (95) | 331 (94) | >0.99 |
| Age at DXA scan, years | 23.8 (6.0) | 42.3 (12) | <0.001 |
| Age at SLE onset, years | 13.9 (2.6) | 34.7 (11) | <0.001 |
| SLE duration, years | 10.0 (6.4) | 7.5 (7.1) | 0.02 |
| Body weight, kg | 52.2 (8.8) | 55.0 (9.7) | 0.07 |
| Body height, m | 1.57 (0.08) | 1.57 (0.07) | 0.77 |
| BMI, kg/m2 | 21.1 (3.3) | 22.3 (3.8) | 0.03 |
| Diabetes mellitus, n (%) | 0 (0) | 12 (3) | 0.63 |
| Hypertension, n (%) | 5 (12) | 93 (26) | 0.04 |
| Chronic smoking, n (%) | 1 (2) | 42 (12) | 0.07 |
| Habitual drinking, n (%) | 0 (0) | 1 (0.3) | >0.99 |
| Post-menopausal, n (%) | 3 (7) | 127 (36) | <0.001 |
| Covariates | Childhood onset (n = 42) | Adult onset (n = 353) | P-value |
|---|---|---|---|
| Gender: female, n (%) | 40 (95) | 331 (94) | >0.99 |
| Age at DXA scan, years | 23.8 (6.0) | 42.3 (12) | <0.001 |
| Age at SLE onset, years | 13.9 (2.6) | 34.7 (11) | <0.001 |
| SLE duration, years | 10.0 (6.4) | 7.5 (7.1) | 0.02 |
| Body weight, kg | 52.2 (8.8) | 55.0 (9.7) | 0.07 |
| Body height, m | 1.57 (0.08) | 1.57 (0.07) | 0.77 |
| BMI, kg/m2 | 21.1 (3.3) | 22.3 (3.8) | 0.03 |
| Diabetes mellitus, n (%) | 0 (0) | 12 (3) | 0.63 |
| Hypertension, n (%) | 5 (12) | 93 (26) | 0.04 |
| Chronic smoking, n (%) | 1 (2) | 42 (12) | 0.07 |
| Habitual drinking, n (%) | 0 (0) | 1 (0.3) | >0.99 |
| Post-menopausal, n (%) | 3 (7) | 127 (36) | <0.001 |
Values were expressed as mean (S.D.) unless otherwise stated.
Demographic and clinical characteristics
| Covariates | Childhood onset (n = 42) | Adult onset (n = 353) | P-value |
|---|---|---|---|
| Gender: female, n (%) | 40 (95) | 331 (94) | >0.99 |
| Age at DXA scan, years | 23.8 (6.0) | 42.3 (12) | <0.001 |
| Age at SLE onset, years | 13.9 (2.6) | 34.7 (11) | <0.001 |
| SLE duration, years | 10.0 (6.4) | 7.5 (7.1) | 0.02 |
| Body weight, kg | 52.2 (8.8) | 55.0 (9.7) | 0.07 |
| Body height, m | 1.57 (0.08) | 1.57 (0.07) | 0.77 |
| BMI, kg/m2 | 21.1 (3.3) | 22.3 (3.8) | 0.03 |
| Diabetes mellitus, n (%) | 0 (0) | 12 (3) | 0.63 |
| Hypertension, n (%) | 5 (12) | 93 (26) | 0.04 |
| Chronic smoking, n (%) | 1 (2) | 42 (12) | 0.07 |
| Habitual drinking, n (%) | 0 (0) | 1 (0.3) | >0.99 |
| Post-menopausal, n (%) | 3 (7) | 127 (36) | <0.001 |
| Covariates | Childhood onset (n = 42) | Adult onset (n = 353) | P-value |
|---|---|---|---|
| Gender: female, n (%) | 40 (95) | 331 (94) | >0.99 |
| Age at DXA scan, years | 23.8 (6.0) | 42.3 (12) | <0.001 |
| Age at SLE onset, years | 13.9 (2.6) | 34.7 (11) | <0.001 |
| SLE duration, years | 10.0 (6.4) | 7.5 (7.1) | 0.02 |
| Body weight, kg | 52.2 (8.8) | 55.0 (9.7) | 0.07 |
| Body height, m | 1.57 (0.08) | 1.57 (0.07) | 0.77 |
| BMI, kg/m2 | 21.1 (3.3) | 22.3 (3.8) | 0.03 |
| Diabetes mellitus, n (%) | 0 (0) | 12 (3) | 0.63 |
| Hypertension, n (%) | 5 (12) | 93 (26) | 0.04 |
| Chronic smoking, n (%) | 1 (2) | 42 (12) | 0.07 |
| Habitual drinking, n (%) | 0 (0) | 1 (0.3) | >0.99 |
| Post-menopausal, n (%) | 3 (7) | 127 (36) | <0.001 |
Values were expressed as mean (S.D.) unless otherwise stated.
Compared with adult-onset patients, childhood-onset SLE patients were significantly younger [23.8 (6.0) vs 42.3 (12) years; P < 0.001], less likely to be hypertensive (12 vs 26%; P = 0.04) and post-menopausal (7 vs 36%; P < 0.001); but had a lower BMI [21.1 (3.3) vs 22.3 (3.8) kg/m2; P = 0.03] and longer SLE duration at the time of DXA examination [10.0 (6.4) vs 7.5 (7.1) years; P = 0.02].
Clinical manifestations and therapies ever received
Table 2 shows the cumulative clinical features and therapies ever received by the two groups of patients since SLE diagnosis. There were no significant differences in the prevalence of clinical manifestations between the two groups of patients. Renal disease appeared to be numerically more common in childhood-onset SLE patients. Among patients with LN, the proportion of diffuse proliferative GN (World Health Organization Class IV disease) was also slightly higher in childhood-onset patients.
Clinical features and therapies ever received
| Clinical characteristics | Childhood onset (n = 42) | Adult onset (n = 353) | P-value |
|---|---|---|---|
| Clinical features | |||
| Musculoskeletala | 28 (67) | 249 (71) | 0.60 |
| Mucocutaneousb | 33 (79) | 235 (67) | 0.12 |
| Hematologicalc | 27 (64) | 196 (56) | 0.28 |
| Renald | 27 (64) | 179 (51) | 0.10 |
| Proportion Class IV nephritis | 14/26 (54) | 57/150 (38) | 0.13 |
| Neuropsychiatrice | 8 (19) | 44 (12) | 0.23 |
| Serositis | 5 (12) | 39 (11) | 0.87 |
| SLEDAI score at DXA scan | |||
| Mean (s.d.) | 5.6 (6.0) | 3.8 (3.0) | 0.07 |
| Median (IQR) | 4.0 (2.0–6.0) | 4.0 (2.0–4.0) | |
| SDI score | |||
| Mean (s.d.) | 0.81 (1.3) | 0.93 (1.3) | 0.57 |
| Median (IQR) | 0.0 (0.0-1.0) | 0.0 (0.0-1.0) | |
| Autoantibodies ever | |||
| Anti-dsDNA | 33 (79) | 242 (69) | 0.18 |
| Anti-Sm | 5 (12) | 43 (12) | 0.96 |
| Anti-Ro | 19 (45) | 192 (54) | 0.26 |
| Anti-La | 6 (14) | 53 (15) | 0.90 |
| Anti-nRNP | 6 (14) | 92 (26) | 0.10 |
| Treatment ever | |||
| Prednisolone ≥3 years | 34 (81) | 215 (61) | 0.01 |
| Duration of prednisolone, mean (s.d.), months | 94.2 (54) | 64.4 (46) | 0.001 |
| High-dose prednisolonef | 30 (71) | 84 (24) | <0.001 |
| Current prednisolone dose, mean (s.d.), mg/day | 4.6 (4.4) | 4.2 (4.7) | 0.61 |
| CYC | 22 (52) | 63 (18) | <0.001 |
| AZA | 33 (79) | 206 (58) | 0.01 |
| MMF | 13 (31) | 65 (18) | 0.05 |
| CNI | 7 (17) | 47 (13) | 0.55 |
| HCQ | 27 (64) | 198 (56) | 0.31 |
| Calcium | 25 (60) | 182 (52) | 0.33 |
| Vitamin D | 15 (36) | 104 (29) | 0.40 |
| Bisphosphonates | 6 (14) | 40 (11) | 0.57 |
| Clinical characteristics | Childhood onset (n = 42) | Adult onset (n = 353) | P-value |
|---|---|---|---|
| Clinical features | |||
| Musculoskeletala | 28 (67) | 249 (71) | 0.60 |
| Mucocutaneousb | 33 (79) | 235 (67) | 0.12 |
| Hematologicalc | 27 (64) | 196 (56) | 0.28 |
| Renald | 27 (64) | 179 (51) | 0.10 |
| Proportion Class IV nephritis | 14/26 (54) | 57/150 (38) | 0.13 |
| Neuropsychiatrice | 8 (19) | 44 (12) | 0.23 |
| Serositis | 5 (12) | 39 (11) | 0.87 |
| SLEDAI score at DXA scan | |||
| Mean (s.d.) | 5.6 (6.0) | 3.8 (3.0) | 0.07 |
| Median (IQR) | 4.0 (2.0–6.0) | 4.0 (2.0–4.0) | |
| SDI score | |||
| Mean (s.d.) | 0.81 (1.3) | 0.93 (1.3) | 0.57 |
| Median (IQR) | 0.0 (0.0-1.0) | 0.0 (0.0-1.0) | |
| Autoantibodies ever | |||
| Anti-dsDNA | 33 (79) | 242 (69) | 0.18 |
| Anti-Sm | 5 (12) | 43 (12) | 0.96 |
| Anti-Ro | 19 (45) | 192 (54) | 0.26 |
| Anti-La | 6 (14) | 53 (15) | 0.90 |
| Anti-nRNP | 6 (14) | 92 (26) | 0.10 |
| Treatment ever | |||
| Prednisolone ≥3 years | 34 (81) | 215 (61) | 0.01 |
| Duration of prednisolone, mean (s.d.), months | 94.2 (54) | 64.4 (46) | 0.001 |
| High-dose prednisolonef | 30 (71) | 84 (24) | <0.001 |
| Current prednisolone dose, mean (s.d.), mg/day | 4.6 (4.4) | 4.2 (4.7) | 0.61 |
| CYC | 22 (52) | 63 (18) | <0.001 |
| AZA | 33 (79) | 206 (58) | 0.01 |
| MMF | 13 (31) | 65 (18) | 0.05 |
| CNI | 7 (17) | 47 (13) | 0.55 |
| HCQ | 27 (64) | 198 (56) | 0.31 |
| Calcium | 25 (60) | 182 (52) | 0.33 |
| Vitamin D | 15 (36) | 104 (29) | 0.40 |
| Bisphosphonates | 6 (14) | 40 (11) | 0.57 |
Values are expressed as n (%) unless otherwise noted. aMusculoskeletal manifestations included arthritis and myositis. bMucocutaneous manifestations included oral ulcers, malar rash, discoid lesions, photosensitivity, RP and alopecia. cHaematological manifestations included haemolytic anaemia, leucopenia, thrombocytopenia and lymphadenopathy. dRenal involvement was defined by the ACR classification criteria. eNeuropsychiatric manifestations included seizure, acute-confusional state, psychosis, myelopathy, neuropathy, mononeuritis multiplex, myasthenia gravis, pseudotumour cerebri and aseptic meningitis. fHigh-dose prednisolone: ≥1 mg/kg/day for at least 6 weeks ± pulse methylprednisolone. IQR: interquartile range; CNI: calcineurin inhibitors.
Clinical features and therapies ever received
| Clinical characteristics | Childhood onset (n = 42) | Adult onset (n = 353) | P-value |
|---|---|---|---|
| Clinical features | |||
| Musculoskeletala | 28 (67) | 249 (71) | 0.60 |
| Mucocutaneousb | 33 (79) | 235 (67) | 0.12 |
| Hematologicalc | 27 (64) | 196 (56) | 0.28 |
| Renald | 27 (64) | 179 (51) | 0.10 |
| Proportion Class IV nephritis | 14/26 (54) | 57/150 (38) | 0.13 |
| Neuropsychiatrice | 8 (19) | 44 (12) | 0.23 |
| Serositis | 5 (12) | 39 (11) | 0.87 |
| SLEDAI score at DXA scan | |||
| Mean (s.d.) | 5.6 (6.0) | 3.8 (3.0) | 0.07 |
| Median (IQR) | 4.0 (2.0–6.0) | 4.0 (2.0–4.0) | |
| SDI score | |||
| Mean (s.d.) | 0.81 (1.3) | 0.93 (1.3) | 0.57 |
| Median (IQR) | 0.0 (0.0-1.0) | 0.0 (0.0-1.0) | |
| Autoantibodies ever | |||
| Anti-dsDNA | 33 (79) | 242 (69) | 0.18 |
| Anti-Sm | 5 (12) | 43 (12) | 0.96 |
| Anti-Ro | 19 (45) | 192 (54) | 0.26 |
| Anti-La | 6 (14) | 53 (15) | 0.90 |
| Anti-nRNP | 6 (14) | 92 (26) | 0.10 |
| Treatment ever | |||
| Prednisolone ≥3 years | 34 (81) | 215 (61) | 0.01 |
| Duration of prednisolone, mean (s.d.), months | 94.2 (54) | 64.4 (46) | 0.001 |
| High-dose prednisolonef | 30 (71) | 84 (24) | <0.001 |
| Current prednisolone dose, mean (s.d.), mg/day | 4.6 (4.4) | 4.2 (4.7) | 0.61 |
| CYC | 22 (52) | 63 (18) | <0.001 |
| AZA | 33 (79) | 206 (58) | 0.01 |
| MMF | 13 (31) | 65 (18) | 0.05 |
| CNI | 7 (17) | 47 (13) | 0.55 |
| HCQ | 27 (64) | 198 (56) | 0.31 |
| Calcium | 25 (60) | 182 (52) | 0.33 |
| Vitamin D | 15 (36) | 104 (29) | 0.40 |
| Bisphosphonates | 6 (14) | 40 (11) | 0.57 |
| Clinical characteristics | Childhood onset (n = 42) | Adult onset (n = 353) | P-value |
|---|---|---|---|
| Clinical features | |||
| Musculoskeletala | 28 (67) | 249 (71) | 0.60 |
| Mucocutaneousb | 33 (79) | 235 (67) | 0.12 |
| Hematologicalc | 27 (64) | 196 (56) | 0.28 |
| Renald | 27 (64) | 179 (51) | 0.10 |
| Proportion Class IV nephritis | 14/26 (54) | 57/150 (38) | 0.13 |
| Neuropsychiatrice | 8 (19) | 44 (12) | 0.23 |
| Serositis | 5 (12) | 39 (11) | 0.87 |
| SLEDAI score at DXA scan | |||
| Mean (s.d.) | 5.6 (6.0) | 3.8 (3.0) | 0.07 |
| Median (IQR) | 4.0 (2.0–6.0) | 4.0 (2.0–4.0) | |
| SDI score | |||
| Mean (s.d.) | 0.81 (1.3) | 0.93 (1.3) | 0.57 |
| Median (IQR) | 0.0 (0.0-1.0) | 0.0 (0.0-1.0) | |
| Autoantibodies ever | |||
| Anti-dsDNA | 33 (79) | 242 (69) | 0.18 |
| Anti-Sm | 5 (12) | 43 (12) | 0.96 |
| Anti-Ro | 19 (45) | 192 (54) | 0.26 |
| Anti-La | 6 (14) | 53 (15) | 0.90 |
| Anti-nRNP | 6 (14) | 92 (26) | 0.10 |
| Treatment ever | |||
| Prednisolone ≥3 years | 34 (81) | 215 (61) | 0.01 |
| Duration of prednisolone, mean (s.d.), months | 94.2 (54) | 64.4 (46) | 0.001 |
| High-dose prednisolonef | 30 (71) | 84 (24) | <0.001 |
| Current prednisolone dose, mean (s.d.), mg/day | 4.6 (4.4) | 4.2 (4.7) | 0.61 |
| CYC | 22 (52) | 63 (18) | <0.001 |
| AZA | 33 (79) | 206 (58) | 0.01 |
| MMF | 13 (31) | 65 (18) | 0.05 |
| CNI | 7 (17) | 47 (13) | 0.55 |
| HCQ | 27 (64) | 198 (56) | 0.31 |
| Calcium | 25 (60) | 182 (52) | 0.33 |
| Vitamin D | 15 (36) | 104 (29) | 0.40 |
| Bisphosphonates | 6 (14) | 40 (11) | 0.57 |
Values are expressed as n (%) unless otherwise noted. aMusculoskeletal manifestations included arthritis and myositis. bMucocutaneous manifestations included oral ulcers, malar rash, discoid lesions, photosensitivity, RP and alopecia. cHaematological manifestations included haemolytic anaemia, leucopenia, thrombocytopenia and lymphadenopathy. dRenal involvement was defined by the ACR classification criteria. eNeuropsychiatric manifestations included seizure, acute-confusional state, psychosis, myelopathy, neuropathy, mononeuritis multiplex, myasthenia gravis, pseudotumour cerebri and aseptic meningitis. fHigh-dose prednisolone: ≥1 mg/kg/day for at least 6 weeks ± pulse methylprednisolone. IQR: interquartile range; CNI: calcineurin inhibitors.
Childhood-onset SLE patients tended to have a higher mean SLEDAI score at the time of DXA scan, and were significantly more likely to have ever received high-dose prednisolone (71 vs 24%), CYC (52 vs 18%) and AZA (79 vs 58%). The mean SLE damage index (SDI), however, was not significantly different between the two groups. The duration of prednisolone treatment was significantly higher in childhood- than adult-onset SLE patients [94.2 (54) vs 64.4 (46) months; P = 0.001], and the proportion of patients who had ever used prednisolone for ≥3 years was also significantly higher in childhood-onset patients (81 vs 61%; P = 0.01). The current prednisolone dose and frequencies of ever use of calcium, vitamin D and the bisphosphonates were similar between the two groups.
BMD and associated factors
Table 3 shows the BMD of the lumbar spine and the hip in the participants. The BMD data of the hip could not be used for evaluation in 3 and 14, respectively, of the childhood- and adult-onset SLE patients because of avascular necrosis of the hip, severe degeneration or joint replacement. On the other hand, the BMD data of the lumbar spine were dropped in one childhood- and eight adult-onset SLE patients because of known compression fractures, deformities or degenerative changes. Overall, 76 (20%) patients had osteoporosis (defined as T-scores of −2.5 or less) at the lumbar spine and 39 (10%) patients were osteoporotic at the hip/femoral neck. The corresponding figures for osteopenia (T-scores of between −1.0 and −2.5) of the spine and hip/femoral neck were 48 and 62%, respectively. The BMD of the lumbar spine, hip, femoral neck and the whole body was significantly lower in childhood- than adult-onset SLE patients, even after adjustment for age, sex, SLE duration and damage index, BMI, menopausal status, smoking, diabetes mellitus, hypertension, duration of CS treatment and whether other immunosuppressive drugs, calcium, vitamin D and the bisphosphonates were ever used. Moreover, the proportions of childhood-onset SLE patients who had osteoporosis of either the lumbar spine or total hip/femoral neck were also significantly higher than those of adult-onset SLE patients.
BMD and osteoporosis
| BMD, g/cm2 | Childhood-onset SLE | Adult-onset SLE | P-value | Adjusted P-value* |
|---|---|---|---|---|
| Lumbar (L2–4) | 0.864 (0.13) | 0.929 (0.13) | 0.004 | 0.002 |
| T-score | −1.92 (1.2) | −1.38 (1.2) | 0.01 | – |
| Z-score | −1.65 (1.3) | −0.90 (1.2) | 0.006 | – |
| Total hip | 0.776 (0.11) | 0.818 (0.12) | 0.03 | 0.001 |
| T-score | −1.35 (0.84) | −1.04 (0.92) | 0.04 | – |
| Z-score | −1.47 (0.72) | −0.75 (0.90) | <0.001 | – |
| Femoral neck | 0.685 (0.10) | 0.717 (0.10) | 0.07 | 0.02 |
| T-score | −1.45 (0.86) | −1.21 (0.92) | 0.12 | – |
| Z-score | −1.57 (0.81) | −0.71 (0.93) | <0.001 | – |
| Whole body | 1.009 (0.10) | 1.053 (0.10) | 0.02 | 0.01 |
| T-score | −1.06 (1.0) | −0.66 (1.06) | 0.03 | – |
| Osteoporosisa (lumbar spine), n (%) | 14/41 (34) | 62/345 (18) | 0.01 | – |
| Osteoporosisa (hip/femoral neck), n (%) | 8/39 (21) | 31/339 (9) | 0.03 | – |
| BMD, g/cm2 | Childhood-onset SLE | Adult-onset SLE | P-value | Adjusted P-value* |
|---|---|---|---|---|
| Lumbar (L2–4) | 0.864 (0.13) | 0.929 (0.13) | 0.004 | 0.002 |
| T-score | −1.92 (1.2) | −1.38 (1.2) | 0.01 | – |
| Z-score | −1.65 (1.3) | −0.90 (1.2) | 0.006 | – |
| Total hip | 0.776 (0.11) | 0.818 (0.12) | 0.03 | 0.001 |
| T-score | −1.35 (0.84) | −1.04 (0.92) | 0.04 | – |
| Z-score | −1.47 (0.72) | −0.75 (0.90) | <0.001 | – |
| Femoral neck | 0.685 (0.10) | 0.717 (0.10) | 0.07 | 0.02 |
| T-score | −1.45 (0.86) | −1.21 (0.92) | 0.12 | – |
| Z-score | −1.57 (0.81) | −0.71 (0.93) | <0.001 | – |
| Whole body | 1.009 (0.10) | 1.053 (0.10) | 0.02 | 0.01 |
| T-score | −1.06 (1.0) | −0.66 (1.06) | 0.03 | – |
| Osteoporosisa (lumbar spine), n (%) | 14/41 (34) | 62/345 (18) | 0.01 | – |
| Osteoporosisa (hip/femoral neck), n (%) | 8/39 (21) | 31/339 (9) | 0.03 | – |
Values are expressed as mean (s.d.) unless otherwise noted. aOsteoporosis defined as a T-score of less than −2.5. *Values compared between the two groups, with adjustment for age, sex, SLE duration and damage index, BMI, duration of CS treatment, ever use of other immunosuppressive therapies, calcium, vitamin D and the bisphosphonates, menopausal status, smoking, drinking, diabetes mellitus and hypertension (by analysis of covariance method).
BMD and osteoporosis
| BMD, g/cm2 | Childhood-onset SLE | Adult-onset SLE | P-value | Adjusted P-value* |
|---|---|---|---|---|
| Lumbar (L2–4) | 0.864 (0.13) | 0.929 (0.13) | 0.004 | 0.002 |
| T-score | −1.92 (1.2) | −1.38 (1.2) | 0.01 | – |
| Z-score | −1.65 (1.3) | −0.90 (1.2) | 0.006 | – |
| Total hip | 0.776 (0.11) | 0.818 (0.12) | 0.03 | 0.001 |
| T-score | −1.35 (0.84) | −1.04 (0.92) | 0.04 | – |
| Z-score | −1.47 (0.72) | −0.75 (0.90) | <0.001 | – |
| Femoral neck | 0.685 (0.10) | 0.717 (0.10) | 0.07 | 0.02 |
| T-score | −1.45 (0.86) | −1.21 (0.92) | 0.12 | – |
| Z-score | −1.57 (0.81) | −0.71 (0.93) | <0.001 | – |
| Whole body | 1.009 (0.10) | 1.053 (0.10) | 0.02 | 0.01 |
| T-score | −1.06 (1.0) | −0.66 (1.06) | 0.03 | – |
| Osteoporosisa (lumbar spine), n (%) | 14/41 (34) | 62/345 (18) | 0.01 | – |
| Osteoporosisa (hip/femoral neck), n (%) | 8/39 (21) | 31/339 (9) | 0.03 | – |
| BMD, g/cm2 | Childhood-onset SLE | Adult-onset SLE | P-value | Adjusted P-value* |
|---|---|---|---|---|
| Lumbar (L2–4) | 0.864 (0.13) | 0.929 (0.13) | 0.004 | 0.002 |
| T-score | −1.92 (1.2) | −1.38 (1.2) | 0.01 | – |
| Z-score | −1.65 (1.3) | −0.90 (1.2) | 0.006 | – |
| Total hip | 0.776 (0.11) | 0.818 (0.12) | 0.03 | 0.001 |
| T-score | −1.35 (0.84) | −1.04 (0.92) | 0.04 | – |
| Z-score | −1.47 (0.72) | −0.75 (0.90) | <0.001 | – |
| Femoral neck | 0.685 (0.10) | 0.717 (0.10) | 0.07 | 0.02 |
| T-score | −1.45 (0.86) | −1.21 (0.92) | 0.12 | – |
| Z-score | −1.57 (0.81) | −0.71 (0.93) | <0.001 | – |
| Whole body | 1.009 (0.10) | 1.053 (0.10) | 0.02 | 0.01 |
| T-score | −1.06 (1.0) | −0.66 (1.06) | 0.03 | – |
| Osteoporosisa (lumbar spine), n (%) | 14/41 (34) | 62/345 (18) | 0.01 | – |
| Osteoporosisa (hip/femoral neck), n (%) | 8/39 (21) | 31/339 (9) | 0.03 | – |
Values are expressed as mean (s.d.) unless otherwise noted. aOsteoporosis defined as a T-score of less than −2.5. *Values compared between the two groups, with adjustment for age, sex, SLE duration and damage index, BMI, duration of CS treatment, ever use of other immunosuppressive therapies, calcium, vitamin D and the bisphosphonates, menopausal status, smoking, drinking, diabetes mellitus and hypertension (by analysis of covariance method).
In both groups of patients, those with history of renal and/or major neuropsychiatric manifestations had lower BMD values at any sites compared with those without, but the differences were not statistically significant (data not shown). Linear regression models showed that childhood-onset SLE (compared with adult-onset SLE) was a risk factor for lower BMD at the lumbar spine (β = −0.18; P = 0.002), hip (β = −0.20; P = 0.001) and the femoral neck (β = −0.16; P = 0.01) after adjustment for other confounding variables that included age, sex, BMI, chronic smoking, menopause, SLE duration and damage index, diabetes mellitus, hypertension, serum creatinine level, duration of prednisolone treatment, current prednisolone dose and whether high-dose glucocorticoids, other immunosuppressive drugs, calcium, vitamin D and the bisphosphonates were ever used (Tables 4 and 5). Other factors significantly associated with lower BMD at the lumbar spine were female sex, lower BMI, higher SLE damage score, menopause and the use of bisphosphonates. On the other hand, female sex, lower BMI, higher SLE damage score, higher serum creatinine level and the ever use of calcium and bisphosphonates were other independent factors significantly associated with lower BMD at the hip.
Multivariate analysis of factors associated with BMD of the lumbar spine by linear regression
| Covariates | Slope (s.e.) | β | P-value |
|---|---|---|---|
| Women | −0.14 (0.06) | −0.11 | 0.02 |
| Childhood-onset disease (vs adult-onset disease) | −0.08 (0.03) | −0.18 | 0.002 |
| Age at DXA screening, per year | <0.001 (0.001) | 0.02 | 0.85 |
| BMI, per kg/m2 | 0.005 (0.002) | 0.14 | 0.009 |
| SLE duration, per year | −0.001 (0.001) | −0.04 | 0.45 |
| SLE damage index, per point | −0.01 (0.006) | −0.11 | 0.047 |
| Chronic smoking | −0.01 (0.02) | −0.02 | 0.69 |
| Menopause | −0.09 (0.02) | −0.31 | <0.001 |
| Duration of prednisolone treatment, per year | <−0.001 (0.002) | −0.009 | 0.88 |
| Ever use of high-dose glucocorticoidsa | −0.04 (0.02) | −0.01 | 0.83 |
| Current daily dose of prednisolone, per mg | <0.001 (0.001) | 0.02 | 0.75 |
| Diabetes mellitus | −0.01 (0.04) | −0.01 | 0.79 |
| Hypertension | 0.009 (0.02) | 0.03 | 0.60 |
| Serum creatinine at DXA scan, per µmol/l | <0.001 (0.001) | −0.07 | 0.15 |
| AZA ever | −0.02 (0.02) | −0.07 | 0.17 |
| CYC ever | 0.01 (0.02) | 0.04 | 0.49 |
| MMF ever | −0.008 (0.02) | −0.03 | 0.64 |
| CNI ever | 0.02 (0.02) | 0.05 | 0.39 |
| HCQ ever | −0.001 (0.01) | −0.004 | 0.93 |
| Calcium ever | −0.02 (0.02) | −0.06 | 0.35 |
| Vitamin D ever | −0.03 (0.02) | −0.12 | 0.07 |
| Bisphosphonates ever | −0.07 (0.02) | −0.16 | 0.005 |
| Covariates | Slope (s.e.) | β | P-value |
|---|---|---|---|
| Women | −0.14 (0.06) | −0.11 | 0.02 |
| Childhood-onset disease (vs adult-onset disease) | −0.08 (0.03) | −0.18 | 0.002 |
| Age at DXA screening, per year | <0.001 (0.001) | 0.02 | 0.85 |
| BMI, per kg/m2 | 0.005 (0.002) | 0.14 | 0.009 |
| SLE duration, per year | −0.001 (0.001) | −0.04 | 0.45 |
| SLE damage index, per point | −0.01 (0.006) | −0.11 | 0.047 |
| Chronic smoking | −0.01 (0.02) | −0.02 | 0.69 |
| Menopause | −0.09 (0.02) | −0.31 | <0.001 |
| Duration of prednisolone treatment, per year | <−0.001 (0.002) | −0.009 | 0.88 |
| Ever use of high-dose glucocorticoidsa | −0.04 (0.02) | −0.01 | 0.83 |
| Current daily dose of prednisolone, per mg | <0.001 (0.001) | 0.02 | 0.75 |
| Diabetes mellitus | −0.01 (0.04) | −0.01 | 0.79 |
| Hypertension | 0.009 (0.02) | 0.03 | 0.60 |
| Serum creatinine at DXA scan, per µmol/l | <0.001 (0.001) | −0.07 | 0.15 |
| AZA ever | −0.02 (0.02) | −0.07 | 0.17 |
| CYC ever | 0.01 (0.02) | 0.04 | 0.49 |
| MMF ever | −0.008 (0.02) | −0.03 | 0.64 |
| CNI ever | 0.02 (0.02) | 0.05 | 0.39 |
| HCQ ever | −0.001 (0.01) | −0.004 | 0.93 |
| Calcium ever | −0.02 (0.02) | −0.06 | 0.35 |
| Vitamin D ever | −0.03 (0.02) | −0.12 | 0.07 |
| Bisphosphonates ever | −0.07 (0.02) | −0.16 | 0.005 |
aHigh-dose glucocorticoids = prednisolone: ≥1 mg/kg/day for at least 6 weeks ± pulse methylprednisolone. β: standardized regression coefficient; CNI: calcineurin inhibitor.
Multivariate analysis of factors associated with BMD of the lumbar spine by linear regression
| Covariates | Slope (s.e.) | β | P-value |
|---|---|---|---|
| Women | −0.14 (0.06) | −0.11 | 0.02 |
| Childhood-onset disease (vs adult-onset disease) | −0.08 (0.03) | −0.18 | 0.002 |
| Age at DXA screening, per year | <0.001 (0.001) | 0.02 | 0.85 |
| BMI, per kg/m2 | 0.005 (0.002) | 0.14 | 0.009 |
| SLE duration, per year | −0.001 (0.001) | −0.04 | 0.45 |
| SLE damage index, per point | −0.01 (0.006) | −0.11 | 0.047 |
| Chronic smoking | −0.01 (0.02) | −0.02 | 0.69 |
| Menopause | −0.09 (0.02) | −0.31 | <0.001 |
| Duration of prednisolone treatment, per year | <−0.001 (0.002) | −0.009 | 0.88 |
| Ever use of high-dose glucocorticoidsa | −0.04 (0.02) | −0.01 | 0.83 |
| Current daily dose of prednisolone, per mg | <0.001 (0.001) | 0.02 | 0.75 |
| Diabetes mellitus | −0.01 (0.04) | −0.01 | 0.79 |
| Hypertension | 0.009 (0.02) | 0.03 | 0.60 |
| Serum creatinine at DXA scan, per µmol/l | <0.001 (0.001) | −0.07 | 0.15 |
| AZA ever | −0.02 (0.02) | −0.07 | 0.17 |
| CYC ever | 0.01 (0.02) | 0.04 | 0.49 |
| MMF ever | −0.008 (0.02) | −0.03 | 0.64 |
| CNI ever | 0.02 (0.02) | 0.05 | 0.39 |
| HCQ ever | −0.001 (0.01) | −0.004 | 0.93 |
| Calcium ever | −0.02 (0.02) | −0.06 | 0.35 |
| Vitamin D ever | −0.03 (0.02) | −0.12 | 0.07 |
| Bisphosphonates ever | −0.07 (0.02) | −0.16 | 0.005 |
| Covariates | Slope (s.e.) | β | P-value |
|---|---|---|---|
| Women | −0.14 (0.06) | −0.11 | 0.02 |
| Childhood-onset disease (vs adult-onset disease) | −0.08 (0.03) | −0.18 | 0.002 |
| Age at DXA screening, per year | <0.001 (0.001) | 0.02 | 0.85 |
| BMI, per kg/m2 | 0.005 (0.002) | 0.14 | 0.009 |
| SLE duration, per year | −0.001 (0.001) | −0.04 | 0.45 |
| SLE damage index, per point | −0.01 (0.006) | −0.11 | 0.047 |
| Chronic smoking | −0.01 (0.02) | −0.02 | 0.69 |
| Menopause | −0.09 (0.02) | −0.31 | <0.001 |
| Duration of prednisolone treatment, per year | <−0.001 (0.002) | −0.009 | 0.88 |
| Ever use of high-dose glucocorticoidsa | −0.04 (0.02) | −0.01 | 0.83 |
| Current daily dose of prednisolone, per mg | <0.001 (0.001) | 0.02 | 0.75 |
| Diabetes mellitus | −0.01 (0.04) | −0.01 | 0.79 |
| Hypertension | 0.009 (0.02) | 0.03 | 0.60 |
| Serum creatinine at DXA scan, per µmol/l | <0.001 (0.001) | −0.07 | 0.15 |
| AZA ever | −0.02 (0.02) | −0.07 | 0.17 |
| CYC ever | 0.01 (0.02) | 0.04 | 0.49 |
| MMF ever | −0.008 (0.02) | −0.03 | 0.64 |
| CNI ever | 0.02 (0.02) | 0.05 | 0.39 |
| HCQ ever | −0.001 (0.01) | −0.004 | 0.93 |
| Calcium ever | −0.02 (0.02) | −0.06 | 0.35 |
| Vitamin D ever | −0.03 (0.02) | −0.12 | 0.07 |
| Bisphosphonates ever | −0.07 (0.02) | −0.16 | 0.005 |
aHigh-dose glucocorticoids = prednisolone: ≥1 mg/kg/day for at least 6 weeks ± pulse methylprednisolone. β: standardized regression coefficient; CNI: calcineurin inhibitor.
Multivariate analysis of factors associated with BMD of the hip and femoral neck by linear regression
| Total hip | Femoral neck | |||||
|---|---|---|---|---|---|---|
| Covariates | Slope (s.e.) | β | P-value | Slope (s.e.) | β | P-value |
| Women | −0.14 (0.05) | −0.14 | 0.005 | −0.06 (0.05) | −0.06 | 0.24 |
| Childhood-onset disease (vs adult-onset disease) | −0.08 (0.02) | −0.20 | 0.001 | −0.05 (0.02) | −0.16 | 0.01 |
| Age at DXA screening, per year | −0.001 (0.001) | −0.13 | 0.15 | −0.001 (0.001) | −0.08 | 0.41 |
| BMI, per kg/m2 | 0.008 (0.002) | 0.26 | <0.001 | 0.006 (0.002) | 0.23 | <0.001 |
| SLE duration, per year | −0.002 (0.001) | −0.10 | 0.09 | −0.002 (0.001) | −0.14 | 0.03 |
| SLE damage index, per point | −0.02 (0.005) | −0.18 | 0.002 | −0.001 (0.005) | −0.007 | 0.91 |
| Chronic smoking | −0.008 (0.02) | −0.02 | 0.69 | −0.008 (0.02) | −0.02 | 0.66 |
| Menopause | −0.03 (0.02) | −0.11 | 0.13 | −0.02 (0.02) | −0.11 | 0.15 |
| Duration of prednisolone treatment, per year | <0.001 (0.002) | 0.02 | 0.78 | <0.001 (0.002) | 0.02 | 0.77 |
| Ever use of high-dose glucocorticoidsa | 0.004 (0.01) | 0.02 | 0.78 | 0.002 (0.01) | 0.007 | 0.91 |
| Current daily dose of prednisolone, per mg | 0.001 (0.001) | 0.04 | 0.44 | <0.001 (0.001) | 0.02 | 0.80 |
| Diabetes mellitus | 0.01 (0.04) | 0.02 | 0.71 | −0.007 (0.03) | −0.01 | 0.82 |
| Hypertension | 0.01 (0.02) | 0.04 | 0.50 | −0.002 (0.01) | −0.01 | 0.87 |
| Serum creatinine at DXA scan, per µmol/l | <0.001 (0.001) | −0.12 | 0.02 | <0.001 (0.001) | −0.11 | 0.06 |
| AZA ever | −0.02 (0.01) | −0.09 | 0.09 | −0.004 (0.01) | −0.02 | 0.73 |
| CYC ever | 0.02 (0.02) | 0.06 | 0.33 | 0.01 (0.02) | 0.06 | 0.33 |
| MMF ever | 0.001 (0.02) | 0.002 | 0.97 | 0.001 (0.02) | 0.005 | 0.93 |
| CNI ever | 0.01 (0.02) | 0.04 | 0.41 | 0.02 (0.02) | 0.05 | 0.34 |
| HCQ ever | 0.01 (0.01) | 0.04 | 0.46 | 0.02 (0.01) | 0.09 | 0.09 |
| Calcium ever | −0.03 (0.01) | −0.13 | 0.04 | −0.03 (0.01) | −0.16 | 0.01 |
| Vitamin D ever | −0.005 (0.02) | −0.02 | 0.76 | 0.006 (0.02) | 0.03 | 0.71 |
| Bisphosphonates ever | −0.05 (0.02) | −0.15 | 0.01 | −0.04 (0.02) | −0.14 | 0.03 |
| Total hip | Femoral neck | |||||
|---|---|---|---|---|---|---|
| Covariates | Slope (s.e.) | β | P-value | Slope (s.e.) | β | P-value |
| Women | −0.14 (0.05) | −0.14 | 0.005 | −0.06 (0.05) | −0.06 | 0.24 |
| Childhood-onset disease (vs adult-onset disease) | −0.08 (0.02) | −0.20 | 0.001 | −0.05 (0.02) | −0.16 | 0.01 |
| Age at DXA screening, per year | −0.001 (0.001) | −0.13 | 0.15 | −0.001 (0.001) | −0.08 | 0.41 |
| BMI, per kg/m2 | 0.008 (0.002) | 0.26 | <0.001 | 0.006 (0.002) | 0.23 | <0.001 |
| SLE duration, per year | −0.002 (0.001) | −0.10 | 0.09 | −0.002 (0.001) | −0.14 | 0.03 |
| SLE damage index, per point | −0.02 (0.005) | −0.18 | 0.002 | −0.001 (0.005) | −0.007 | 0.91 |
| Chronic smoking | −0.008 (0.02) | −0.02 | 0.69 | −0.008 (0.02) | −0.02 | 0.66 |
| Menopause | −0.03 (0.02) | −0.11 | 0.13 | −0.02 (0.02) | −0.11 | 0.15 |
| Duration of prednisolone treatment, per year | <0.001 (0.002) | 0.02 | 0.78 | <0.001 (0.002) | 0.02 | 0.77 |
| Ever use of high-dose glucocorticoidsa | 0.004 (0.01) | 0.02 | 0.78 | 0.002 (0.01) | 0.007 | 0.91 |
| Current daily dose of prednisolone, per mg | 0.001 (0.001) | 0.04 | 0.44 | <0.001 (0.001) | 0.02 | 0.80 |
| Diabetes mellitus | 0.01 (0.04) | 0.02 | 0.71 | −0.007 (0.03) | −0.01 | 0.82 |
| Hypertension | 0.01 (0.02) | 0.04 | 0.50 | −0.002 (0.01) | −0.01 | 0.87 |
| Serum creatinine at DXA scan, per µmol/l | <0.001 (0.001) | −0.12 | 0.02 | <0.001 (0.001) | −0.11 | 0.06 |
| AZA ever | −0.02 (0.01) | −0.09 | 0.09 | −0.004 (0.01) | −0.02 | 0.73 |
| CYC ever | 0.02 (0.02) | 0.06 | 0.33 | 0.01 (0.02) | 0.06 | 0.33 |
| MMF ever | 0.001 (0.02) | 0.002 | 0.97 | 0.001 (0.02) | 0.005 | 0.93 |
| CNI ever | 0.01 (0.02) | 0.04 | 0.41 | 0.02 (0.02) | 0.05 | 0.34 |
| HCQ ever | 0.01 (0.01) | 0.04 | 0.46 | 0.02 (0.01) | 0.09 | 0.09 |
| Calcium ever | −0.03 (0.01) | −0.13 | 0.04 | −0.03 (0.01) | −0.16 | 0.01 |
| Vitamin D ever | −0.005 (0.02) | −0.02 | 0.76 | 0.006 (0.02) | 0.03 | 0.71 |
| Bisphosphonates ever | −0.05 (0.02) | −0.15 | 0.01 | −0.04 (0.02) | −0.14 | 0.03 |
aHigh-dose glucocorticoids = prednisolone: ≥1 mg/kg/day for at least 6 weeks ± pulse methylprednisolone. β: standardized regression coefficient; CNI: calcineurin inhibitor.
Multivariate analysis of factors associated with BMD of the hip and femoral neck by linear regression
| Total hip | Femoral neck | |||||
|---|---|---|---|---|---|---|
| Covariates | Slope (s.e.) | β | P-value | Slope (s.e.) | β | P-value |
| Women | −0.14 (0.05) | −0.14 | 0.005 | −0.06 (0.05) | −0.06 | 0.24 |
| Childhood-onset disease (vs adult-onset disease) | −0.08 (0.02) | −0.20 | 0.001 | −0.05 (0.02) | −0.16 | 0.01 |
| Age at DXA screening, per year | −0.001 (0.001) | −0.13 | 0.15 | −0.001 (0.001) | −0.08 | 0.41 |
| BMI, per kg/m2 | 0.008 (0.002) | 0.26 | <0.001 | 0.006 (0.002) | 0.23 | <0.001 |
| SLE duration, per year | −0.002 (0.001) | −0.10 | 0.09 | −0.002 (0.001) | −0.14 | 0.03 |
| SLE damage index, per point | −0.02 (0.005) | −0.18 | 0.002 | −0.001 (0.005) | −0.007 | 0.91 |
| Chronic smoking | −0.008 (0.02) | −0.02 | 0.69 | −0.008 (0.02) | −0.02 | 0.66 |
| Menopause | −0.03 (0.02) | −0.11 | 0.13 | −0.02 (0.02) | −0.11 | 0.15 |
| Duration of prednisolone treatment, per year | <0.001 (0.002) | 0.02 | 0.78 | <0.001 (0.002) | 0.02 | 0.77 |
| Ever use of high-dose glucocorticoidsa | 0.004 (0.01) | 0.02 | 0.78 | 0.002 (0.01) | 0.007 | 0.91 |
| Current daily dose of prednisolone, per mg | 0.001 (0.001) | 0.04 | 0.44 | <0.001 (0.001) | 0.02 | 0.80 |
| Diabetes mellitus | 0.01 (0.04) | 0.02 | 0.71 | −0.007 (0.03) | −0.01 | 0.82 |
| Hypertension | 0.01 (0.02) | 0.04 | 0.50 | −0.002 (0.01) | −0.01 | 0.87 |
| Serum creatinine at DXA scan, per µmol/l | <0.001 (0.001) | −0.12 | 0.02 | <0.001 (0.001) | −0.11 | 0.06 |
| AZA ever | −0.02 (0.01) | −0.09 | 0.09 | −0.004 (0.01) | −0.02 | 0.73 |
| CYC ever | 0.02 (0.02) | 0.06 | 0.33 | 0.01 (0.02) | 0.06 | 0.33 |
| MMF ever | 0.001 (0.02) | 0.002 | 0.97 | 0.001 (0.02) | 0.005 | 0.93 |
| CNI ever | 0.01 (0.02) | 0.04 | 0.41 | 0.02 (0.02) | 0.05 | 0.34 |
| HCQ ever | 0.01 (0.01) | 0.04 | 0.46 | 0.02 (0.01) | 0.09 | 0.09 |
| Calcium ever | −0.03 (0.01) | −0.13 | 0.04 | −0.03 (0.01) | −0.16 | 0.01 |
| Vitamin D ever | −0.005 (0.02) | −0.02 | 0.76 | 0.006 (0.02) | 0.03 | 0.71 |
| Bisphosphonates ever | −0.05 (0.02) | −0.15 | 0.01 | −0.04 (0.02) | −0.14 | 0.03 |
| Total hip | Femoral neck | |||||
|---|---|---|---|---|---|---|
| Covariates | Slope (s.e.) | β | P-value | Slope (s.e.) | β | P-value |
| Women | −0.14 (0.05) | −0.14 | 0.005 | −0.06 (0.05) | −0.06 | 0.24 |
| Childhood-onset disease (vs adult-onset disease) | −0.08 (0.02) | −0.20 | 0.001 | −0.05 (0.02) | −0.16 | 0.01 |
| Age at DXA screening, per year | −0.001 (0.001) | −0.13 | 0.15 | −0.001 (0.001) | −0.08 | 0.41 |
| BMI, per kg/m2 | 0.008 (0.002) | 0.26 | <0.001 | 0.006 (0.002) | 0.23 | <0.001 |
| SLE duration, per year | −0.002 (0.001) | −0.10 | 0.09 | −0.002 (0.001) | −0.14 | 0.03 |
| SLE damage index, per point | −0.02 (0.005) | −0.18 | 0.002 | −0.001 (0.005) | −0.007 | 0.91 |
| Chronic smoking | −0.008 (0.02) | −0.02 | 0.69 | −0.008 (0.02) | −0.02 | 0.66 |
| Menopause | −0.03 (0.02) | −0.11 | 0.13 | −0.02 (0.02) | −0.11 | 0.15 |
| Duration of prednisolone treatment, per year | <0.001 (0.002) | 0.02 | 0.78 | <0.001 (0.002) | 0.02 | 0.77 |
| Ever use of high-dose glucocorticoidsa | 0.004 (0.01) | 0.02 | 0.78 | 0.002 (0.01) | 0.007 | 0.91 |
| Current daily dose of prednisolone, per mg | 0.001 (0.001) | 0.04 | 0.44 | <0.001 (0.001) | 0.02 | 0.80 |
| Diabetes mellitus | 0.01 (0.04) | 0.02 | 0.71 | −0.007 (0.03) | −0.01 | 0.82 |
| Hypertension | 0.01 (0.02) | 0.04 | 0.50 | −0.002 (0.01) | −0.01 | 0.87 |
| Serum creatinine at DXA scan, per µmol/l | <0.001 (0.001) | −0.12 | 0.02 | <0.001 (0.001) | −0.11 | 0.06 |
| AZA ever | −0.02 (0.01) | −0.09 | 0.09 | −0.004 (0.01) | −0.02 | 0.73 |
| CYC ever | 0.02 (0.02) | 0.06 | 0.33 | 0.01 (0.02) | 0.06 | 0.33 |
| MMF ever | 0.001 (0.02) | 0.002 | 0.97 | 0.001 (0.02) | 0.005 | 0.93 |
| CNI ever | 0.01 (0.02) | 0.04 | 0.41 | 0.02 (0.02) | 0.05 | 0.34 |
| HCQ ever | 0.01 (0.01) | 0.04 | 0.46 | 0.02 (0.01) | 0.09 | 0.09 |
| Calcium ever | −0.03 (0.01) | −0.13 | 0.04 | −0.03 (0.01) | −0.16 | 0.01 |
| Vitamin D ever | −0.005 (0.02) | −0.02 | 0.76 | 0.006 (0.02) | 0.03 | 0.71 |
| Bisphosphonates ever | −0.05 (0.02) | −0.15 | 0.01 | −0.04 (0.02) | −0.14 | 0.03 |
aHigh-dose glucocorticoids = prednisolone: ≥1 mg/kg/day for at least 6 weeks ± pulse methylprednisolone. β: standardized regression coefficient; CNI: calcineurin inhibitor.
Discussion
This is a cross-sectional study of the BMD of a large group of adult patients with SLE. We demonstrated that the overall incidence of osteoporosis was 20% at the lumbar spine and 10% at the hip/femoral neck. Despite having a significantly younger age, childhood-onset SLE patients had significantly lower BMD at the lumbar spine, hip and the femoral neck compared with patients with adult-onset disease. The difference remained significant after adjustment by linear regression for age, sex, duration and damage index of SLE, duration of CS treatment, current CS dose, ever use of high-dose CSs, BMI, chronic smoking, menopause, serum creatinine level and the ever use of calcium, vitamin D, bisphosphonates and other immunosuppressive medications.
Although a lower BMI, longer duration of SLE and CS treatment in childhood-onset patients may contribute to the lower BMD observed, their effects have been adjusted in multivariate regression models. Childhood-onset disease remains to be a significant independent risk factor for lower BMD at the hip and spine in adult SLE patients. This observation can possibly be explained by some factors that are not tested in the models. First, glucocorticoid therapy at the time of the pubertal growth spur could have caused a failure of achievement of a PBM expected for normal adolescents. The lower PBM attained in the adolescent years may account for a significantly lower BMD in adulthood even after adjustment for the duration of SLE and glucocorticoid treatment. Secondly, childhood-onset SLE had more serious disease manifestations as shown by a higher proportion of patients suffering from diffuse proliferative LN, severe neuropsychiatric and haematological disease and hence requiring high-dose glucocorticoid and CYC treatment. Persistent and more serious disease activity is associated with higher levels of pro-inflammatory cytokines that may lead to a more profound loss in BMD. However, we do not have data on the cumulative/mean disease activity (e.g. area under the curve of the SLE disease activity scores) of all our patients since their SLE diagnosis to confirm this postulate. Although the duration of glucocorticoid treatment, current prednisolone dose and ever use of high-dose glucocorticoids were not found to be independently associated with low BMD in SLE patients, the cumulative glucocorticoid dose received ever since disease diagnosis, which was unavailable in our study, might be a more important factor. We acknowledge that a higher cumulative dose of glucocorticoids received by childhood-onset SLE patients for their more active disease remains an explanation for the lower BMD observed. Lastly, the difference in diet, nutritional status and level of physical activity, which was not the focus of the current study, between childhood- and adult-onset SLE patients might have also led to a difference in the BMD observed.
The PBM is one of the most important determinants of post-menopausal osteoporosis and fragility fracture [17, 18]. Cross-sectional studies have indicated that PBM is not attained until late in the third decade in life [19], whereas other reports have demonstrated little bone mass accrual after the second decade in life [20]. In both men and women, longitudinal studies have shown that the rate of increase in bone mass of the femoral neck and lumbar spine slows down markedly in late adolescence [21]. Around 26% of adult bone mineral is acquired during the 2 adolescent years of peak height velocity [22] and up to 60% is accrued in the remaining peripubertal years [23]. The PBM is influenced by many factors that include genetics, physical activity, lifestyle factors, nutrition, hormonal status, presence of chronic medical illnesses and endocrinopathies and use of medications during the critical years of bone acquisition [24].
Glucocorticoids exhibit a number of deleterious effects on BMD by altering calcium metabolism and the function, differentiation and apoptosis of the osteoclasts, osteoblasts and osteocytes [25]. In addition, long-term glucocorticoid treatment reduces the pulsatility and secretion of growth hormone by increasing the somatostatin inhibitor tone [26]. Glucocorticoids also decrease the expression of growth hormone receptors on hepatocytes, leading to lower insulin growth factor-1 (IGF-1) mRNA levels [27]. The suppression of production of growth hormone and the sex steroids by glucocorticoid treatment during puberty may have profound negative effects on skeletal growth and the accrual of the PBM. However, our study was limited by the lack of information on the stage of puberty at the time of glucocorticoid administration in childhood-onset SLE patients.
Elevation of pro-inflammatory cytokines such as TNF-α, IL-1, IL-6, IL-7 and IL-17 induces the expression of receptor activator of nuclear factor κ-B ligand, which enhances osteoclast differentiation [28]. On the other hand, cytokines such as TNF-α induce proteins Dkk-1 and sclerostin, which are both potent inhibitors of the Wnt pathway and osteoblastogenesis [29]. Levels of cytokines such as TNF-α and IL-6 are elevated in patients with SLE and correlate with disease activity [30, 31]. Persistent SLE activity may therefore be involved in the acceleration of BMD loss as a result of elevation of these cytokines.
In conclusion, our study provides evidence to show that in adult patients with SLE, childhood-onset disease carries a higher risk of low BMD at the lumbar spine, hip and femoral neck compared with adult-onset disease, which is independent of age, sex, general risk factors for osteoporosis, organ damage and the use of non-CS immunosuppressive medications. This may possibly be contributed by the failure to acquire a normal PBM during puberty as a result of glucocorticoid treatment and more active disease in childhood-onset SLE patients that require a higher cumulative dose of glucocorticoids for control. Longitudinal studies on the rate of BMD loss in childhood- and adult-onset SLE patients, and its relationship with cumulative disease activity, glucocorticoid doses, nutritional status and physical activity should be pursued in future studies. More judicious use of glucocorticoids, early use of glucocorticoid-sparing agents and vigorous measures to prevent BMD loss should be undertaken in all SLE patients, in particular those with onset of disease in childhood.
Disclosure statement: The authors have declared no conflicts of interest.
Comments