Trabecular bone score as an assessment tool to identify the risk of osteoporosis in axial spondyloarthritis: a case–control study

Trabecular bone score as an assessment tool to identify the risk of osteoporosis in axial... Abstract Objectives To compare the trabecular bone score (TBS) between patients with axial spondyloarthritis (axSpA) and matched normal controls and identify risk factors associated with a low TBS. Methods TBS and BMD were assessed in the two groups (axSpA and control) using DXA. Osteoporosis risk factors and inflammatory markers were also assessed. Disease activity and radiographic progression in the sacroiliac joint and spine were evaluated in the axSpA group. Multivariate linear regression analysis was performed to identify risk factors associated with TBS. Results In the axSpA group, 248 subjects were enrolled; an equal number of age- and sex-matched subjects comprised the control group. The mean TBS was 1.43 (0.08) and 1.38 (0.12) in the control and axSpA groups, respectively (P < 0.001); BMD at the lumbar spine did not differ between the two groups. The TBS was negatively correlated with ESR and CRP levels in the axSpA group only (P < 0.001 and P = 0.007, respectively). Syndesmophytes in the axSpA group was associated with lower TBS (P < 0.001) but higher lumbar BMD (P = 0.021) vs controls. In the multivariate analyses, ESR, CRP and spinal radiographic progression were significantly associated with TBS. Conclusion TBS assessments revealed poor bone quality in patients with axSpA compared with the matched controls. In axSpA, systemic inflammatory markers were negatively correlated with TBS and spinal radiographic progression and inflammatory markers were independently correlated with low TBS. TBS may, therefore, be a useful clinical tool to identify the risk of osteoporosis in patients with axSpA. axial spondyloarthritis, trabecular bone score, bone mineral density, bone microarchitecture, bone quality Rheumatology Key messages Trabecular bone score assessment revealed poor bone quality in patients with axial SpA when compared with matched controls. Spinal radiographic progression and inflammatory markers were independently related to low trabecular bone score. Introduction Axial spondyloarthritis (axSpA) is a chronic inflammatory disease that primarily affects the axial skeleton. AxSpA includes patients with structural damage in the sacroiliac joints or spine visible on radiographs (radiographic axSpA, or AS) and also patients without such structural damage, defined as non-radiographic axSpA (nr-axSpA). Patients present with chronic inflammatory pain, predominantly of the pelvis and lower back. Inflammation and structural damage can occur in the axial skeleton as a consequence of inflammation [1]. SpA exerts a paradoxical situation with respect to bone metabolism as it is associated with generalized bone loss (osteoporosis), as well as localized inflammatory bone resorption and new bone formation, such as syndesmophytes. It has been established that generalized bone loss may be due to systemic inflammation and disease activity. Increased inflammation and pro-inflammatory cytokines have been implicated in the development of secondary osteoporosis [2]. Bone mineral density (BMD) and fracture risk in axSpA patients are lower and higher, respectively, than in age- and sex-matched general populations [3, 4]. Osteoporosis is a skeletal disorder characterized by compromised bone strength, which predisposes with this disease to an increased risk of fracture. Bone strength can be considered to be primarily determined by bone density and quality (e.g. microarchitecture) [5]. A deterioration of trabecular microarchitecture with loss of connectivity between trabeculae and cortical thinning is typically observed in patients with osteoporosis [6]. Increased fracture risk in axSpA is likely to be multifactorial, resulting from traditional osteoporosis risk factors and disease-related factors such as systemic inflammation, which affects not only BMD, but also bone quality properties [7, 8]. It is unclear which imaging approach is most effective for diagnosing and monitoring osteoporosis in axSpA. DXA measurement of BMD at the hip and spine is most often used to establish or confirm a diagnosis of osteoporosis and to monitor patients [9]. Most data related to bone loss in AS are also based on studies using DXA measurement of BMD. However, defining osteoporosis using DXA has several limitations, including the confounding effects of the surrounding soft tissue, bone artifacts caused by osteoarthritis, degenerated discs, aortic calcification and vertebral compression fractures. Syndesmophytes, as a result of inflammation, may increase BMD as measured by DXA despite the presence of osteoporosis. BMD may, therefore, not be an optimal marker for diagnosing osteoporosis in patients with axSpA. In addition, DXA cannot be used to measure bone microarchitecture or quality, which are believed to influence fracture risk [6]. In patients with axSpA, bone microarchitecture alterations, as well as BMD reduction, might be responsible for the increased risk of fractures. Quantitative CT (QCT) and high resolution peripheral QCT (HR-pQCT) have been developed to measure bone microarchitecture and assess bone quality. Although evidence suggests that these methods can predict fracture, they are not routinely available in practice [10]. Therefore, a clinically available, non-invasive technique for the accurate evaluation of bone microarchitecture in axSpA is required. The trabecular bone score (TBS) is a novel tool used to evaluate bone microarchitecture. It is a textural index that evaluates pixel grey-level variations in the lumbar spine DXA image, thereby providing an indirect index of trabecular microarchitecture [11]. The TBS derived from DXA images correlates with the 3D microarchitecture parameters measured by QCT [12]. A previous study showed that TBS is not influenced by spinal osteophytes, which may lead to an overestimation of BMD among patients with lumbar spine OA [13]. This suggests that TBS can assess the bone microarchitecture among patients with axSpA regardless of syndesmophytes. To date, no studies have compared TBS between patients with axSpA and matched controls. Therefore, the present study was conducted to assess TBS in these two groups and to identify the risk factors related to a low TBS in patients with axSpA. Methods Study population Between August 2013 and December 2016, consecutive axSpA patients from Incheon St Mary’s Hospital (Incheon, Korea) were recruited to this case–control study. All enrolled patients fulfilled the imaging arm of the Assessment of SpondyloArthritis International Society (ASAS) axSpA criteria [14]. All patients were aged between 20 and 69 years, and all female patients were premenopausal. Age- and sex-matched subjects who attended routine health check-up examinations during the same period were recruited as controls. Exclusion criteria included patients or controls with thyroid or parathyroid disorders, presence of chronic renal or liver disease, and use of corticosteroid, calcium or bisphosphonate agents. No one received vitamin D agents. TBS and BMD were assessed at the time of enrolment. At the time of TBS assessment, demographic data were collected, osteoporosis risk factors were assessed (smoking status and alcohol consumption) and ESR and CRP were measured. All participants provided written consent according to the Declaration of Helsinki. The study was approved by the ethics committee at Incheon St Mary’s Hospital (study number: OC16OISI0138). Clinical data In patients with axSpA, disease-related data and disease activity scores were collected. Clinical data included time after symptom onset, the presence of HLA B27, family history and the use of medications such as anti-inflammatory drugs, SSZ and TNF inhibitors. Measures of disease activity were collected using the BASDAI [15]. All scores were recorded on a visual analogue scale from 0 to 10. The BASFI [16] and patient global assessment scores were also recorded. The ASDAS was calculated as described previously [17]. BMD assessment Areal BMD was measured at the lumbar spine (L1–L4) and left hip using DXA (Lunar Prodigy densitometer, GE Healthcare, Madison, WI, USA). All measurements were taken by experienced operators using the same machine and standardized procedures for participant positioning. BMD was measured at the lumbar spine (L1–L4) and the left hip (femoral neck and total proximal femur) and expressed as the number of grams of bone mineral per square centimetre (g/cm2). TBS assessment TBS was analysed using DXA images of the lumbar spine (L1–L4). Lumbar spine DXA images were reanalysed in an operator-independent automated manner using TBS iNight software version 2.1 (Med-Imaps, Merignac, France). The software uses the posterior–anterior images, including the BMD region of interest and edge detection; thus the TBS is calculated over exactly the same region as the lumbar BMD assessment. Patients in the axSpA group were divided into three TBS groups according to the risk of fracture in a recent meta-analysis; high risk: TBS below 1.23; medium risk: TBS 1.23–1.31; low risk: TBS above 1.31 [18]. Radiographic scoring For all patients, radiographs of the cervical spine, lumbar spine and pelvis were obtained at the time of the TBS assessment. Lateral views of the cervical and lumbar spine were scored according to the modified Stoke AS Spinal Score (mSASSS) [19]. Sacroiliitis was scored from right and left sided pelvic radiographs using the modified New York criteria [20]. The average score for both sides was used for analysis. Sacroiliitis and the mSASSS were scored by a single trained expert who was blinded to the patient characteristics. Statistical analysis Statistical analyses were performed using SPSS Statistics (version 21.0; IBM Corp., Armonk, NY, USA). Continuous data were expressed as the mean (s.d.), and categorical data were expressed as percentages. Clinical variables were compared using an independent t test, and the chi-squared test was used to compare categorical variables between axSpA patients and matched control subjects. Spearman’s correlation coefficient was used to analyse the correlation between variables. Clinical variables, and BMD and TBS values were compared in TBS subgroups across fracture risk using chi-squared tests for categorical variables and analysis of variance tests for continuous data. Multiple linear regression models were used to assess the association between TBS and clinical variables in the axSpA group. All variables with a P < 0.05 in univariate linear regression analyses were incorporated as explanatory variables (enter method). In the multivariate analysis, separate models for baseline ESR and CRP were performed due to multicollinearity. P < 0.05 was considered statistically significant. Results Patient characteristics Both study groups recruited 248 subjects, the demographic and clinical characteristics of which are shown in Table 1. The mean age of the patients in both groups was 39 (11) years, and 78% were male. Table 1 Characteristics of control subjects and patients with axial SpA Variable  Control (n = 248)  AxSpA (n = 248)  P-value  Age, mean (s.d.), years  39 (11)  39 (11)  1.000  Male  193 (78)  193 (78)  1.000  BMI, mean (s.d.), kg/m2  24.4 (3.6)  24.0 (3.7)  0.286  Current smoking  104 (42)  92 (37)  0.309  Alcohol ≥3 U/day  6 (2)  7 (3)  1.000  Diabetes mellitus  14 (6)  8 (3)  0.275  Hypertension  25 (10)  36 (15)  0.171  Dyslipidaemia  18 (7)  17 (7)  1.000  History of fracture  6 (2)  30 (12)  <0.001  Laboratory findings            ESR, mean (s.d.), mm/h  9.6 (8.9)  24.3 (21.9)  <0.001      CRP mg/l, mean (s.d.)  1.3 (1.9)  10.0 (16.7)  <0.001      Haemoglobin, mean (s.d.), g/dl  14.8 (1.4)  14.4 (1.6)  0.002      Platelet, mean (s.d.), 103/µl  249 (66)  282 (199)  0.013  Disease-related data            Symptom duration, mean (s.d.), years    10 (9)        Family history of axSpA    34 (14)        HLA B27-positive    215 (87)        Patient global assessment, mean (s.d.)    4.7 (2.6)        BASDAI score, range, 0–10, mean (s.d.)    4.1 (2.2)        BASFI score, mean (s.d.)    2.0 (2.4)        ASDAS-ESR, mean (s.d.)    2.8 (1.1)        ASDAS-CRP, mean (s.d.)    2.4 (1.3)        Grade of sacroiliitis on X-ray, mean (s.d.)    2.9 (1.0)    mSASSS, mean (s.d.)    13.1 (20.0)    Presence of spinal progression    142 (57)    Number of syndesmophytes, mean (s.d.)    4.4 (7.4)    Presence of syndesmophytes    108 (44)    Patients on NSAIDs    230 (93)    Patients on SSZ    64 (26)    Patients on TNF inhibitors    52 (21)    Variable  Control (n = 248)  AxSpA (n = 248)  P-value  Age, mean (s.d.), years  39 (11)  39 (11)  1.000  Male  193 (78)  193 (78)  1.000  BMI, mean (s.d.), kg/m2  24.4 (3.6)  24.0 (3.7)  0.286  Current smoking  104 (42)  92 (37)  0.309  Alcohol ≥3 U/day  6 (2)  7 (3)  1.000  Diabetes mellitus  14 (6)  8 (3)  0.275  Hypertension  25 (10)  36 (15)  0.171  Dyslipidaemia  18 (7)  17 (7)  1.000  History of fracture  6 (2)  30 (12)  <0.001  Laboratory findings            ESR, mean (s.d.), mm/h  9.6 (8.9)  24.3 (21.9)  <0.001      CRP mg/l, mean (s.d.)  1.3 (1.9)  10.0 (16.7)  <0.001      Haemoglobin, mean (s.d.), g/dl  14.8 (1.4)  14.4 (1.6)  0.002      Platelet, mean (s.d.), 103/µl  249 (66)  282 (199)  0.013  Disease-related data            Symptom duration, mean (s.d.), years    10 (9)        Family history of axSpA    34 (14)        HLA B27-positive    215 (87)        Patient global assessment, mean (s.d.)    4.7 (2.6)        BASDAI score, range, 0–10, mean (s.d.)    4.1 (2.2)        BASFI score, mean (s.d.)    2.0 (2.4)        ASDAS-ESR, mean (s.d.)    2.8 (1.1)        ASDAS-CRP, mean (s.d.)    2.4 (1.3)        Grade of sacroiliitis on X-ray, mean (s.d.)    2.9 (1.0)    mSASSS, mean (s.d.)    13.1 (20.0)    Presence of spinal progression    142 (57)    Number of syndesmophytes, mean (s.d.)    4.4 (7.4)    Presence of syndesmophytes    108 (44)    Patients on NSAIDs    230 (93)    Patients on SSZ    64 (26)    Patients on TNF inhibitors    52 (21)    All values given as n (%) unless otherwise stated. axSpA: axial spondyloarthritis; mSASSS: Modified Stokes Ankylosing Spondylitis Score. There were no significant differences between the groups in terms of BMI, smoking history, alcohol consumption, diagnosis of diabetes mellitus, hypertension or treatment with lipid-lowering agents. ESR, CRP and platelet levels were higher in the axSpA group than in controls (P < 0.001, P < 0.001 and P = 0.013, respectively). Haemoglobin levels were lower in the axSpA group than in matched controls (P = 0.002). Among the 248 patients with axSpA, 207 fulfilled the modified New York criteria for AS [20]. The mean symptom duration was 10 (9) years, and 215 (87%) were HLA B27 positive. The mean BASDAI and ASDAS-CRP scores were 4.1 (2.2) and 2.4 (1.3). The average grade of sacroiliitis on X-ray and mSASSS was 2.9 (1.0) and 13.1 (20.0), respectively. Of the 248 patients, 108 (44%) had one or more syndesmophytes. The majority of patients (93%) received NSAIDs, and 21% received TNF inhibitors. Assessment of TBS and BMD The mean TBS value was 1.43 (0.08) for controls and 1.38 (0.12) for axSpA patients (P < 0.001; Fig. 1A). BMD data are shown in Fig. 1B. BMD at the lumbar spine did not differ between the two groups. The mean BMD at the femoral neck was 0.98 (0.14) in controls and 0.93 (0.13) in the axSpA group; the mean BMD of total hip was 1.03 (0.14) in controls and 0.97 (0.14) in the axSpA group (both P < 0.001). Fig. 1 View largeDownload slide Trabecular bone score and BMD in patients with axial SpA and matched controls (A) Comparison of TBS among axSpA patients and controls matched for age and sex. (B) Comparison of BMD of lumbar spine, femoral neck and total hip. axSpA: axial spondylarthropathy; TBS: trabecular bone score. Fig. 1 View largeDownload slide Trabecular bone score and BMD in patients with axial SpA and matched controls (A) Comparison of TBS among axSpA patients and controls matched for age and sex. (B) Comparison of BMD of lumbar spine, femoral neck and total hip. axSpA: axial spondylarthropathy; TBS: trabecular bone score. Correlation between clinical data, inflammatory markers, BMD and TBS In both groups, the TBS showed a significant negative correlation with age and positive correlation with BMD of lumbar spine, femoral neck and total hip (Table 2). The TBS was negatively correlated with ESR and CRP in the axSpA group only. Table 2 Correlation between clinical data, inflammatory markers, BMD and trabecular bone score Group  Age, years  BMI, kg/m2  ESR, mm/h  CRP, mg/l  BMD, g/cm2  Lumbar spine  Femoral neck  Total hip  Control, TBS, L1–L4  −0.328 (<0.001)  −0.026 (0.685)  −0.032 (0.669)  0.023 (0.767)  0.373 (<0.001)  0.376 (<0.001)  0.336 (<0.001)  Axial SpA, TBS, L1–4  −0.206 (0.001)  −0.096 (0.132)  −0.230 (<0.001)  −0.170 (0.007)  0.207 (0.001)  0.296 (<0.001)  0.276 (<0.001)  Group  Age, years  BMI, kg/m2  ESR, mm/h  CRP, mg/l  BMD, g/cm2  Lumbar spine  Femoral neck  Total hip  Control, TBS, L1–L4  −0.328 (<0.001)  −0.026 (0.685)  −0.032 (0.669)  0.023 (0.767)  0.373 (<0.001)  0.376 (<0.001)  0.336 (<0.001)  Axial SpA, TBS, L1–4  −0.206 (0.001)  −0.096 (0.132)  −0.230 (<0.001)  −0.170 (0.007)  0.207 (0.001)  0.296 (<0.001)  0.276 (<0.001)  R coefficients (P-values); TBS: trabecular bone score. Comparisons of TBS and BMD according to the presence of syndesmophytes As shown in Fig. 2, TBS values were significantly lower in axSpA patients with and without syndesmophytes than in matched controls (P = 0.004 and P < 0.001, respectively). Lumbar spine BMD was significantly lower in the axSpA group without syndesmophytes than in the controls [1.13 (0.14) vs 1.18 (0.15) in controls; P = 0.008]. In contrast, the mean lumbar spine BMD was higher in the axSpA group with syndesmophytes [1.26 (0.28) vs 1.19 (0.16) in controls; P = 0.021]. Fig. 2 View largeDownload slide Comparison of trabecular bone score and BMD according to syndesmophytes (A) Comparison of trabecular bone score L1–L4 in axial spondyloarthritis patients and matched control groups according to the presence of syndesmophytes. (B) Comparison of BMD of lumbar spine in axial spondyloarthritis patients and matched control groups according to the presence of syndesmophytes. Fig. 2 View largeDownload slide Comparison of trabecular bone score and BMD according to syndesmophytes (A) Comparison of trabecular bone score L1–L4 in axial spondyloarthritis patients and matched control groups according to the presence of syndesmophytes. (B) Comparison of BMD of lumbar spine in axial spondyloarthritis patients and matched control groups according to the presence of syndesmophytes. AxSpA patient characteristics stratified according to TBS fracture risk groups Table 3 shows the demographic characteristics and disease-related variables between three TBS subgroups according to the risk of fracture [18]. Factors associated with a higher risk were male sex (P = 0.028) and higher BMI (P = 0.029). There were no significant differences among the TBS subgroups of patients in the fracture risk with respect to age, smoking status, alcohol consumption, symptom duration and HLA B27 status. Serum ESR and CRP were higher in the axSpA group with high fracture risk TBS values than in the group with low risk (P = 0.013 and P = 0.034, respectively). BASDAI, ASDAS-ESR and ASDAS-CRP did not differ between the three groups. The grade of sacroiliitis on X-ray, the mSASSS and the mean number of syndesmophytes were higher in patients with high fracture risk TBS (P = 0.006, P < 0.001 and P < 0.001, respectively). BMD levels at the lumbar spine did not differ between the three groups, although significant differences in BMD of the femoral neck and total hip were seen (P = 0.003 and P = 0.015, respectively). Table 3 Characteristics of axial SpA group stratified according to trabecular bone score risk group Variable  High risk (TBS < 1.23, n = 28)  Medium risk (TBS = 1.23–1.31, n = 27)  Low risk (TBS > 1.31, n = 193)  P-value  Age, years  43 (12)  42 (13)  38 (11)  0.061  Male, n (%)  26 (93)  24 (89)  143 (74)  0.028  BMI, kg/m2  25.7 (2.8)  24.1 (3.9)  23.8 (3.7)  0.029  Current smoking, n (%)  13 (46)  13 (48)  66 (34)  0.108  Alcohol ≥3 U/day, n (%)  3 (11)  3 (11)  22 (11)  0.994  Symptom duration, years  10.7 (11.2)  10.0 (8.3)  9.8 (9.2)  0.891  Family history of axial SpA, n (%)  2 (7)  5 (19)  27 (14)  0.458  HLA B27-positive, n (%)  27 (96)  23 (85)  165 (67)  0.273  ESR, mm/h  34.3 (24.9)  21.6 (4.2)  22.2 (21.1)  0.013  CRP, mg/l  12.0 (15.5)  17.2 (20.6)  8.6 (16.0)  0.034  Patient global assessment  5.3 (2.5)  5.1 (2.6)  4.6 (2.6)  0.301  BASDAI score, range, 0–10  4.1 (2.1)  4.4 (1.9)  4.1 (2.3)  0.728  BASFI score  2.3 (2.6)  2.7 (2.5)  1.9 (2.3)  0.197  ASDAS-ESR  3.1 (1.1)  3.0 (1.1)  2.7 (1.1)  0.126  ASDAS-CRP  2.7 (1.2)  2.9 (1.2)  2.3 (1.3)  0.055  Grade of sacroiliitis on X-ray  3.4 (0.8)  3.1 (1.0)  2.8 (1.0)  0.006  mSASSS  28.1 (22.9)  15.6 (18.7)  10.6 (18.8)  <0.001  Presence of spinal progression, n (%)  24 (86)  18 (67)  100 (52)  0.002  Number of syndesmophytes  9.5 (9.0)  4.9 (6.5)  3.6 (7.0)  <0.001  Presence of syndesmophytes, n (%)  20 (71)  13 (48)  75 (30)  0.004  Patients on NSAIDs, n (%)  28 (100)  22 (82)  180 (93)  0.025  Patients on SSZ, n (%)  5 (18)  7 (26)  52 (27)  0.590  Patients on TNF inhibitors, n (%)  8 (15)  5 (19)  39 (20)  0.565  TBS, L1–L4  1.14 (0.07)  1.27 (0.02)  1.43 (0.07)  <0.001  BMD, g/cm2              Lumbar spine  1.17 (0.26)  1.13 (0.21)  1.20 (0.22)  0.231      Femoral neck  0.88 (0.14)  0.87 (0.18)  0.94 (0.12)  0.003      Total hip  0.92 (0.13)  0.92 (0.19)  0.98 (0.13)  0.015  Variable  High risk (TBS < 1.23, n = 28)  Medium risk (TBS = 1.23–1.31, n = 27)  Low risk (TBS > 1.31, n = 193)  P-value  Age, years  43 (12)  42 (13)  38 (11)  0.061  Male, n (%)  26 (93)  24 (89)  143 (74)  0.028  BMI, kg/m2  25.7 (2.8)  24.1 (3.9)  23.8 (3.7)  0.029  Current smoking, n (%)  13 (46)  13 (48)  66 (34)  0.108  Alcohol ≥3 U/day, n (%)  3 (11)  3 (11)  22 (11)  0.994  Symptom duration, years  10.7 (11.2)  10.0 (8.3)  9.8 (9.2)  0.891  Family history of axial SpA, n (%)  2 (7)  5 (19)  27 (14)  0.458  HLA B27-positive, n (%)  27 (96)  23 (85)  165 (67)  0.273  ESR, mm/h  34.3 (24.9)  21.6 (4.2)  22.2 (21.1)  0.013  CRP, mg/l  12.0 (15.5)  17.2 (20.6)  8.6 (16.0)  0.034  Patient global assessment  5.3 (2.5)  5.1 (2.6)  4.6 (2.6)  0.301  BASDAI score, range, 0–10  4.1 (2.1)  4.4 (1.9)  4.1 (2.3)  0.728  BASFI score  2.3 (2.6)  2.7 (2.5)  1.9 (2.3)  0.197  ASDAS-ESR  3.1 (1.1)  3.0 (1.1)  2.7 (1.1)  0.126  ASDAS-CRP  2.7 (1.2)  2.9 (1.2)  2.3 (1.3)  0.055  Grade of sacroiliitis on X-ray  3.4 (0.8)  3.1 (1.0)  2.8 (1.0)  0.006  mSASSS  28.1 (22.9)  15.6 (18.7)  10.6 (18.8)  <0.001  Presence of spinal progression, n (%)  24 (86)  18 (67)  100 (52)  0.002  Number of syndesmophytes  9.5 (9.0)  4.9 (6.5)  3.6 (7.0)  <0.001  Presence of syndesmophytes, n (%)  20 (71)  13 (48)  75 (30)  0.004  Patients on NSAIDs, n (%)  28 (100)  22 (82)  180 (93)  0.025  Patients on SSZ, n (%)  5 (18)  7 (26)  52 (27)  0.590  Patients on TNF inhibitors, n (%)  8 (15)  5 (19)  39 (20)  0.565  TBS, L1–L4  1.14 (0.07)  1.27 (0.02)  1.43 (0.07)  <0.001  BMD, g/cm2              Lumbar spine  1.17 (0.26)  1.13 (0.21)  1.20 (0.22)  0.231      Femoral neck  0.88 (0.14)  0.87 (0.18)  0.94 (0.12)  0.003      Total hip  0.92 (0.13)  0.92 (0.19)  0.98 (0.13)  0.015  All values given as mean (s.d.) unless otherwise stated. mSASSS: modified stokes AS score; TBS: trabecular bone score. Linear regression analysis for TBS among axSpA patients Table 4 shows the results of univariate and multivariate analyses of TBS in the axSpA group. Univariate analysis revealed that TBS was associated with age, current smoking status, HLA B27, ESR, CRP, the grade of sacroiliitis and the mSASSS. In the multivariate analyses, ESR, CRP and mSASSS were significantly associated with the TBS value. Table 4 Linear regression analysis of the trabecular bone score in the lumbar spine among patients with axial SpA (n = 248) Variable  Univariate model β (95% Cl)  Multivariate ESR modela β (95% Cl)  Multivariate CRP modela β (95% Cl)  Demographic            Age, years  −0.002 (−0.004, −0.001)*          Gender, male  0.033 (−0.003, 0.069)          Current smoking  −0.033 (−0.064, −0.002)*          Alcohol ≥3 U/day  0.012 (−0.036, 0.059)          BMI, kg/m2  −0.003 (−0.007, 0.001)          HLA B27  −0.050 (−0.094, −0.006)*      Medication            NSAIDs  −0.016 (−0.073, 0.042)          SSZ  0.029 (−0.005, 0.063)          TNF inhibitor  −0.024 (−0.061, 0.013)      Disease activity            ESR, mm/h  −0.001 (−0.002, −0.001)*  −0.001 (−0.002, −0.001)*        CRP, mg/l  −0.001 (−0.002, 0.000)*    −0.001 (−0.002, 0.000)*      BASDAI  0.001 (−0.006, 0.007)          ASDAS  −0.010 (−0.022, 0.002)      Radiographic scores            Sacroiliitis on X-ray  −0.029 (−0.043, −0.015)*          mSASSS  −0.002 (−0.003, −0.001)*  −0.001 (−0.002, −0.000)*  −0.001 (−0.002, −0.000)*  Adjusted R2    0.434  0.396  Variable  Univariate model β (95% Cl)  Multivariate ESR modela β (95% Cl)  Multivariate CRP modela β (95% Cl)  Demographic            Age, years  −0.002 (−0.004, −0.001)*          Gender, male  0.033 (−0.003, 0.069)          Current smoking  −0.033 (−0.064, −0.002)*          Alcohol ≥3 U/day  0.012 (−0.036, 0.059)          BMI, kg/m2  −0.003 (−0.007, 0.001)          HLA B27  −0.050 (−0.094, −0.006)*      Medication            NSAIDs  −0.016 (−0.073, 0.042)          SSZ  0.029 (−0.005, 0.063)          TNF inhibitor  −0.024 (−0.061, 0.013)      Disease activity            ESR, mm/h  −0.001 (−0.002, −0.001)*  −0.001 (−0.002, −0.001)*        CRP, mg/l  −0.001 (−0.002, 0.000)*    −0.001 (−0.002, 0.000)*      BASDAI  0.001 (−0.006, 0.007)          ASDAS  −0.010 (−0.022, 0.002)      Radiographic scores            Sacroiliitis on X-ray  −0.029 (−0.043, −0.015)*          mSASSS  −0.002 (−0.003, −0.001)*  −0.001 (−0.002, −0.000)*  −0.001 (−0.002, −0.000)*  Adjusted R2    0.434  0.396  a Adjusted for age, smoking status, HLA B27, sacroiliitis on X-ray and mSASSS. * P < 0.05. mSASSS: modified Stokes Ankylosing Spondylitis Score. Discussion In the present study, TBS was used to assess bone microarchitecture and investigate the relationship between systemic inflammation, radiographic progression and bone quality in patients with axSpA. When assessed by TBS, patients with axSpA were seen to have poor bone quality in the lumbar spine compared with matched controls. Markers of systemic inflammation, such as ESR or CRP, correlated with TBS in patients with axSpA, but not in controls. Also, systemic inflammatory markers and spinal radiographic progression scores were independently associated with low TBS. AxSpA is characterized by ossification of the entheses and formation of syndesmophytes, but paradoxically there is also loss of trabecular bone adjacent to sites of inflammation. Trabecular and cortical compartments appear to have different reactions to inflammation; in AS, inflammation has a direct effect on the trabecular bone of the vertebrae, but not the cortical bone [21], and therefore AS patients show low trabecular BMD in the spine QCT [8]. DXA-based BMD cannot distinguish between cortical and trabecular compartments and cannot reflect trabecular bone microarchitecture alterations, and therefore potentially result in inconsistent data in patients with AS [22–24]. TBS has the advantage of being able to directly assess the microarchitecture of trabecular bone at the lumbar spine, which is the main area of involvement in axSpA patients. A previous study showed that sacroiliitis on MRI correlates with BMD at the femoral neck and total hip; however, sacroiliitis on MRI was not associated with lumbar BMD [25]. This indicates that bone loss at the spine in axSpA patients may be due to the localized effects of inflammation. Our data showed that BMD at the lumbar spine did not differ between axSpA and non-axSpA subjects despite differences in bone microarchitecture measured by TBS. This may be related to both an overestimation of BMD due to the presence of syndesmophytes and insufficient reflection of changes in the quality of trabecular bone caused by chronic inflammation. In the subgroup analysis according to the presence of syndesmophytes, we found that the presence of syndesmophytes was associated with lower TBS but higher BMD in the lumbar spine. This result is consistent with earlier studies also reporting lower trabecular volumetric BMD measured by QCT but higher cortical volumetric BMD in AS [8, 26]. The EULAR taskforce made a series of recommendations with respect to the use of imaging for the management of SpA. They proposed that, in patients with axSpA without syndesmophytes in the lumbar spine on conventional radiography, osteoporosis should be assessed using hip DXA and anterior–posterior spine DXA; in patients with syndesmophytes in the lumbar spine on conventional radiography, osteoporosis should be assessed using hip DXA supplemented by either spine DXA (lateral projection) or possibly QCT of the spine [27]. Although spinal DXA using a lateral view has the advantage of allowing selection of a trabecular zone in the centre of the vertebral bodies, it has poor precision due to the difficulty in positioning the patients. Moreover, due to the overlap of the last rib and iliac crest, only L3 can be measured in some cases [13, 28]. QCT cannot be performed routinely in a clinical setting due to the high radiation dose required, cost and limited access to such devices [10]. Therefore, a tool that is unaffected by spinal progression is needed to assess osteoporosis and to predict fracture risk in axSpA. Our findings indicate that TBS may be a valuable alternative tool for the assessment of bone quality influenced by inflammation in patients with axSpA, regardless of spinal progression. A recent small study also showed that TBS is not affected by syndesmophytes in patients with axSpA [29]. Including the assessment of bone quality using TBS, as well as BMD measurement, may improve the predictive value for the development of new vertebral fractures. The data presented here also show that ESR and CRP levels were independently associated with TBS after adjusting for confounding factors. In contrast with ESR and CRP, we found that BASDAI and ASDAS did not correlate with TBS. In general, clinical disease activity scores such as BASDAI and ASDAS comprise self-reported measures. These scores can be influenced by ankylosis and secondary changes in AS. Therefore, bony changes caused by inflammation may be more closely related to objective inflammatory markers than to clinical disease activity scores. The association between TBS and inflammatory markers suggests that in patients with axSpA TBS may reflect bone quality alteration in the presence of systemic inflammation. The correlation between TBS and inflammatory markers among axSpA patients suggests that TBS reflects inflammation-mediated changes in trabecular bone in the lumbar spine, and therefore, assessment of bone quality in the spine using the TBS may be a better predictor of the risk of osteoporosis and vertebral fracture in patients with axSpA. In addition, spinal radiographic progression score (mSASSS) was an independent risk factor for low TBS. Spinal radiographic progression reflects the severity of cumulative inflammation in the spine, and our data suggest that spinal radiographic progression may increase the risk of osteoporosis as it is associated with poor bone quality, as well as stiffening of the spine, in patients with axSpA. Longitudinal studies are required to examine whether the TBS is predictive of vertebral fracture. Lumbar TBS is associated with an age-dependent variable, and a significant decrease in TBS has been observed with ageing in both male and female subjects [11, 30]. In the present study, age was negatively correlated with TBS in both the axSpA and control groups. In the axSpA group, age was associated with TBS in univariate analysis, but the association was not observed in multivariate analysis after adjusting for confounding factors. Therefore, it appears that TBS in axSpA is more influenced by the severity of systemic and local inflammation than ageing, and the strict control of inflammation may help to decrease the risk of osteoporosis among patients. This finding should be assessed further in prospective longitudinal studies. This study has a few limitations. First, vitamin D and PTH were not measured. These variables are important for skeletal health and could affect the risk of osteoporosis. Second, only 52 (21%) patients received TNF inhibitors. This is a relatively small number when considered alongside the total number of axSpA patients; therefore, the regression analysis could be underpowered. In conclusion, TBS assessment revealed poor bone quality in the lumbar spine in patients with axSpA when compared with matched controls. Systemic inflammatory markers were negatively correlated with TBS in axSpA and spinal radiographic progression, and inflammatory markers were independently related to low TBS. These data support the use of TBS as a useful clinical tool to identify the risk of osteoporosis in patients with axSpA. Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript. Disclosure statement: The authors have declared no conflicts of interest. References 1 Sieper J, Poddubnyy D. Axial spondyloarthritis. Lancet  2017; 390: 73– 84. Google Scholar CrossRef Search ADS PubMed  2 Szentpetery A, Horvath A, Gulyas K et al.   Effects of targeted therapies on the bone in arthritides. Autoimmun Rev  2017; 16: 313– 20. Google Scholar CrossRef Search ADS PubMed  3 Kang KY, Kwok SK, Ju JH, Hong YS, Park SH. Assessment of fracture risk in patients with axial spondyloarthritis: a case-control study using the fifth Korean National Health and Nutrition Examination Survey (KNHANES V). Scand J Rheumatol  2015; 45: 23– 31. Google Scholar CrossRef Search ADS PubMed  4 Magrey M, Khan MA. Osteoporosis in ankylosing spondylitis. Curr Rheumatol Rep  2010; 12: 332– 6. Google Scholar CrossRef Search ADS PubMed  5 NIH Consensus Development Panel on Osteoporosis Prevention Diagnosis, and Therapy. Osteoporosis prevention, diagnosis, and therapy. JAMA  2001; 285: 785– 95. CrossRef Search ADS PubMed  6 Lorentzon M, Cummings SR. Osteoporosis: the evolution of a diagnosis. J Intern Med  2015; 277: 650– 61. Google Scholar CrossRef Search ADS PubMed  7 Geusens P, Lems WF. Osteoimmunology and osteoporosis. Arthritis Res Ther  2011; 13: 242. Google Scholar CrossRef Search ADS PubMed  8 Klingberg E, Lorentzon M, Gothlin J et al.   Bone microarchitecture in ankylosing spondylitis and the association with bone mineral density, fractures, and syndesmophytes. Arthritis Res Ther  2013; 15: R179. Google Scholar CrossRef Search ADS PubMed  9 Cosman F, de Beur SJ, LeBoff MS et al.   Clinician's Guide to Prevention and Treatment of Osteoporosis. Osteoporos Int  2014; 25: 2359– 81. Google Scholar CrossRef Search ADS PubMed  10 Harvey NC, Gluer CC, Binkley N et al.   Trabecular bone score (TBS) as a new complementary approach for osteoporosis evaluation in clinical practice. Bone  2015; 78: 216– 24. Google Scholar CrossRef Search ADS PubMed  11 Silva BC, Leslie WD, Resch H et al.   Trabecular bone score: a noninvasive analytical method based upon the DXA image. J Bone Miner Res  2014; 29: 518– 30. Google Scholar CrossRef Search ADS PubMed  12 Silva BC, Walker MD, Abraham A et al.   Trabecular bone score is associated with volumetric bone density and microarchitecture as assessed by central QCT and HRpQCT in Chinese American and white women. J Clin Densitom  2013; 16: 554– 61. Google Scholar CrossRef Search ADS PubMed  13 Kolta S, Briot K, Fechtenbaum J et al.   TBS result is not affected by lumbar spine osteoarthritis. Osteoporos Int  2014; 25: 1759– 64. Google Scholar CrossRef Search ADS PubMed  14 Rudwaleit M, Landewe R, van der Heijde D et al.   The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part I): classification of paper patients by expert opinion including uncertainty appraisal. Ann Rheum Dis  2009; 68: 770– 6. Google Scholar CrossRef Search ADS PubMed  15 Garrett S, Jenkinson T, Kennedy LG et al.   A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index. J Rheumatol  1994; 21: 2286– 91. Google Scholar PubMed  16 Calin A, Garrett S, Whitelock H et al.   A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath Ankylosing Spondylitis Functional Index. J Rheumatol  1994; 21: 2281– 5. Google Scholar PubMed  17 van der Heijde D, Lie E, Kvien TK et al.   ASDAS, a highly discriminatory ASAS-endorsed disease activity score in patients with ankylosing spondylitis. Ann Rheum Dis  2009; 68: 1811– 8. Google Scholar CrossRef Search ADS PubMed  18 McCloskey EV, Oden A, Harvey NC et al.   A meta-analysis of trabecular bone score in fracture risk prediction and its relationship to FRAX. J Bone Miner Res  2016; 31: 940– 8. Google Scholar CrossRef Search ADS PubMed  19 Wanders AJ, Landewe RB, Spoorenberg A et al.   What is the most appropriate radiologic scoring method for ankylosing spondylitis? A comparison of the available methods based on the Outcome Measures in Rheumatology Clinical Trials filter. Arthritis Rheum  2004; 50: 2622– 32. Google Scholar CrossRef Search ADS PubMed  20 van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum  1984; 27: 361– 8. Google Scholar CrossRef Search ADS PubMed  21 Schett G. Structural bone changes in spondyloarthritis: mechanisms, clinical impact and therapeutic considerations. Am J Med Sci  2011; 341: 269– 71. Google Scholar CrossRef Search ADS PubMed  22 Donnelly S, Doyle DV, Denton A et al.   Bone mineral density and vertebral compression fracture rates in ankylosing spondylitis. Ann Rheum Dis  1994; 53: 117– 21. Google Scholar CrossRef Search ADS PubMed  23 Ghozlani I, Ghazi M, Nouijai A et al.   Prevalence and risk factors of osteoporosis and vertebral fractures in patients with ankylosing spondylitis. Bone  2009; 44: 772– 6. Google Scholar CrossRef Search ADS PubMed  24 Klingberg E, Geijer M, Gothlin J et al.   Vertebral fractures in ankylosing spondylitis are associated with lower bone mineral density in both central and peripheral skeleton. J Rheumatol  2012; 39: 1987– 95. Google Scholar CrossRef Search ADS PubMed  25 Kim HN, Jung JY, Hong YS, Park SH, Kang KY. Severe bone marrow edema on sacroiliac joint MRI increases the risk of low BMD in patients with axial spondyloarthritis. Sci Rep  2016; 6: 22158. Google Scholar CrossRef Search ADS PubMed  26 Lange U, Teichmann J, Strunk J, Muller-Ladner U, Schmidt KL. Association of 1.25 vitamin D3 deficiency, disease activity and low bone mass in ankylosing spondylitis. Osteoporos Int  2005; 16: 1999– 2004. Google Scholar CrossRef Search ADS PubMed  27 Mandl P, Navarro-Compan V, Terslev L et al.   EULAR recommendations for the use of imaging in the diagnosis and management of spondyloarthritis in clinical practice. Ann Rheum Dis  2015; 74: 1327– 39. Google Scholar CrossRef Search ADS PubMed  28 Larnach TA, Boyd SJ, Smart RC et al.   Reproducibility of lateral spine scans using dual energy X-ray absorptiometry. Calcif Tissue Int  1992; 51: 255– 8. Google Scholar CrossRef Search ADS PubMed  29 Wildberger L, Boyadzhieva V, Hans D, Stoilov N, Rashkov R, Aubry-Rozier B. Impact of lumbar syndesmophyte on bone health as assessed by bone density (BMD) and bone texture (TBS) in men with axial spondyloarthritis. Joint Bone Spine  2017; 84: 463– 466. Google Scholar CrossRef Search ADS PubMed  30 Bazzocchi A, Ponti F, Diano D et al.   Trabecular bone score in healthy ageing. Br J Radiol  2015; 88: 20140865. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Rheumatology Oxford University Press

Trabecular bone score as an assessment tool to identify the risk of osteoporosis in axial spondyloarthritis: a case–control study

Loading next page...
 
/lp/ou_press/trabecular-bone-score-as-an-assessment-tool-to-identify-the-risk-of-EH1n7VLH4V
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
1462-0324
eISSN
1462-0332
D.O.I.
10.1093/rheumatology/kex377
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives To compare the trabecular bone score (TBS) between patients with axial spondyloarthritis (axSpA) and matched normal controls and identify risk factors associated with a low TBS. Methods TBS and BMD were assessed in the two groups (axSpA and control) using DXA. Osteoporosis risk factors and inflammatory markers were also assessed. Disease activity and radiographic progression in the sacroiliac joint and spine were evaluated in the axSpA group. Multivariate linear regression analysis was performed to identify risk factors associated with TBS. Results In the axSpA group, 248 subjects were enrolled; an equal number of age- and sex-matched subjects comprised the control group. The mean TBS was 1.43 (0.08) and 1.38 (0.12) in the control and axSpA groups, respectively (P < 0.001); BMD at the lumbar spine did not differ between the two groups. The TBS was negatively correlated with ESR and CRP levels in the axSpA group only (P < 0.001 and P = 0.007, respectively). Syndesmophytes in the axSpA group was associated with lower TBS (P < 0.001) but higher lumbar BMD (P = 0.021) vs controls. In the multivariate analyses, ESR, CRP and spinal radiographic progression were significantly associated with TBS. Conclusion TBS assessments revealed poor bone quality in patients with axSpA compared with the matched controls. In axSpA, systemic inflammatory markers were negatively correlated with TBS and spinal radiographic progression and inflammatory markers were independently correlated with low TBS. TBS may, therefore, be a useful clinical tool to identify the risk of osteoporosis in patients with axSpA. axial spondyloarthritis, trabecular bone score, bone mineral density, bone microarchitecture, bone quality Rheumatology Key messages Trabecular bone score assessment revealed poor bone quality in patients with axial SpA when compared with matched controls. Spinal radiographic progression and inflammatory markers were independently related to low trabecular bone score. Introduction Axial spondyloarthritis (axSpA) is a chronic inflammatory disease that primarily affects the axial skeleton. AxSpA includes patients with structural damage in the sacroiliac joints or spine visible on radiographs (radiographic axSpA, or AS) and also patients without such structural damage, defined as non-radiographic axSpA (nr-axSpA). Patients present with chronic inflammatory pain, predominantly of the pelvis and lower back. Inflammation and structural damage can occur in the axial skeleton as a consequence of inflammation [1]. SpA exerts a paradoxical situation with respect to bone metabolism as it is associated with generalized bone loss (osteoporosis), as well as localized inflammatory bone resorption and new bone formation, such as syndesmophytes. It has been established that generalized bone loss may be due to systemic inflammation and disease activity. Increased inflammation and pro-inflammatory cytokines have been implicated in the development of secondary osteoporosis [2]. Bone mineral density (BMD) and fracture risk in axSpA patients are lower and higher, respectively, than in age- and sex-matched general populations [3, 4]. Osteoporosis is a skeletal disorder characterized by compromised bone strength, which predisposes with this disease to an increased risk of fracture. Bone strength can be considered to be primarily determined by bone density and quality (e.g. microarchitecture) [5]. A deterioration of trabecular microarchitecture with loss of connectivity between trabeculae and cortical thinning is typically observed in patients with osteoporosis [6]. Increased fracture risk in axSpA is likely to be multifactorial, resulting from traditional osteoporosis risk factors and disease-related factors such as systemic inflammation, which affects not only BMD, but also bone quality properties [7, 8]. It is unclear which imaging approach is most effective for diagnosing and monitoring osteoporosis in axSpA. DXA measurement of BMD at the hip and spine is most often used to establish or confirm a diagnosis of osteoporosis and to monitor patients [9]. Most data related to bone loss in AS are also based on studies using DXA measurement of BMD. However, defining osteoporosis using DXA has several limitations, including the confounding effects of the surrounding soft tissue, bone artifacts caused by osteoarthritis, degenerated discs, aortic calcification and vertebral compression fractures. Syndesmophytes, as a result of inflammation, may increase BMD as measured by DXA despite the presence of osteoporosis. BMD may, therefore, not be an optimal marker for diagnosing osteoporosis in patients with axSpA. In addition, DXA cannot be used to measure bone microarchitecture or quality, which are believed to influence fracture risk [6]. In patients with axSpA, bone microarchitecture alterations, as well as BMD reduction, might be responsible for the increased risk of fractures. Quantitative CT (QCT) and high resolution peripheral QCT (HR-pQCT) have been developed to measure bone microarchitecture and assess bone quality. Although evidence suggests that these methods can predict fracture, they are not routinely available in practice [10]. Therefore, a clinically available, non-invasive technique for the accurate evaluation of bone microarchitecture in axSpA is required. The trabecular bone score (TBS) is a novel tool used to evaluate bone microarchitecture. It is a textural index that evaluates pixel grey-level variations in the lumbar spine DXA image, thereby providing an indirect index of trabecular microarchitecture [11]. The TBS derived from DXA images correlates with the 3D microarchitecture parameters measured by QCT [12]. A previous study showed that TBS is not influenced by spinal osteophytes, which may lead to an overestimation of BMD among patients with lumbar spine OA [13]. This suggests that TBS can assess the bone microarchitecture among patients with axSpA regardless of syndesmophytes. To date, no studies have compared TBS between patients with axSpA and matched controls. Therefore, the present study was conducted to assess TBS in these two groups and to identify the risk factors related to a low TBS in patients with axSpA. Methods Study population Between August 2013 and December 2016, consecutive axSpA patients from Incheon St Mary’s Hospital (Incheon, Korea) were recruited to this case–control study. All enrolled patients fulfilled the imaging arm of the Assessment of SpondyloArthritis International Society (ASAS) axSpA criteria [14]. All patients were aged between 20 and 69 years, and all female patients were premenopausal. Age- and sex-matched subjects who attended routine health check-up examinations during the same period were recruited as controls. Exclusion criteria included patients or controls with thyroid or parathyroid disorders, presence of chronic renal or liver disease, and use of corticosteroid, calcium or bisphosphonate agents. No one received vitamin D agents. TBS and BMD were assessed at the time of enrolment. At the time of TBS assessment, demographic data were collected, osteoporosis risk factors were assessed (smoking status and alcohol consumption) and ESR and CRP were measured. All participants provided written consent according to the Declaration of Helsinki. The study was approved by the ethics committee at Incheon St Mary’s Hospital (study number: OC16OISI0138). Clinical data In patients with axSpA, disease-related data and disease activity scores were collected. Clinical data included time after symptom onset, the presence of HLA B27, family history and the use of medications such as anti-inflammatory drugs, SSZ and TNF inhibitors. Measures of disease activity were collected using the BASDAI [15]. All scores were recorded on a visual analogue scale from 0 to 10. The BASFI [16] and patient global assessment scores were also recorded. The ASDAS was calculated as described previously [17]. BMD assessment Areal BMD was measured at the lumbar spine (L1–L4) and left hip using DXA (Lunar Prodigy densitometer, GE Healthcare, Madison, WI, USA). All measurements were taken by experienced operators using the same machine and standardized procedures for participant positioning. BMD was measured at the lumbar spine (L1–L4) and the left hip (femoral neck and total proximal femur) and expressed as the number of grams of bone mineral per square centimetre (g/cm2). TBS assessment TBS was analysed using DXA images of the lumbar spine (L1–L4). Lumbar spine DXA images were reanalysed in an operator-independent automated manner using TBS iNight software version 2.1 (Med-Imaps, Merignac, France). The software uses the posterior–anterior images, including the BMD region of interest and edge detection; thus the TBS is calculated over exactly the same region as the lumbar BMD assessment. Patients in the axSpA group were divided into three TBS groups according to the risk of fracture in a recent meta-analysis; high risk: TBS below 1.23; medium risk: TBS 1.23–1.31; low risk: TBS above 1.31 [18]. Radiographic scoring For all patients, radiographs of the cervical spine, lumbar spine and pelvis were obtained at the time of the TBS assessment. Lateral views of the cervical and lumbar spine were scored according to the modified Stoke AS Spinal Score (mSASSS) [19]. Sacroiliitis was scored from right and left sided pelvic radiographs using the modified New York criteria [20]. The average score for both sides was used for analysis. Sacroiliitis and the mSASSS were scored by a single trained expert who was blinded to the patient characteristics. Statistical analysis Statistical analyses were performed using SPSS Statistics (version 21.0; IBM Corp., Armonk, NY, USA). Continuous data were expressed as the mean (s.d.), and categorical data were expressed as percentages. Clinical variables were compared using an independent t test, and the chi-squared test was used to compare categorical variables between axSpA patients and matched control subjects. Spearman’s correlation coefficient was used to analyse the correlation between variables. Clinical variables, and BMD and TBS values were compared in TBS subgroups across fracture risk using chi-squared tests for categorical variables and analysis of variance tests for continuous data. Multiple linear regression models were used to assess the association between TBS and clinical variables in the axSpA group. All variables with a P < 0.05 in univariate linear regression analyses were incorporated as explanatory variables (enter method). In the multivariate analysis, separate models for baseline ESR and CRP were performed due to multicollinearity. P < 0.05 was considered statistically significant. Results Patient characteristics Both study groups recruited 248 subjects, the demographic and clinical characteristics of which are shown in Table 1. The mean age of the patients in both groups was 39 (11) years, and 78% were male. Table 1 Characteristics of control subjects and patients with axial SpA Variable  Control (n = 248)  AxSpA (n = 248)  P-value  Age, mean (s.d.), years  39 (11)  39 (11)  1.000  Male  193 (78)  193 (78)  1.000  BMI, mean (s.d.), kg/m2  24.4 (3.6)  24.0 (3.7)  0.286  Current smoking  104 (42)  92 (37)  0.309  Alcohol ≥3 U/day  6 (2)  7 (3)  1.000  Diabetes mellitus  14 (6)  8 (3)  0.275  Hypertension  25 (10)  36 (15)  0.171  Dyslipidaemia  18 (7)  17 (7)  1.000  History of fracture  6 (2)  30 (12)  <0.001  Laboratory findings            ESR, mean (s.d.), mm/h  9.6 (8.9)  24.3 (21.9)  <0.001      CRP mg/l, mean (s.d.)  1.3 (1.9)  10.0 (16.7)  <0.001      Haemoglobin, mean (s.d.), g/dl  14.8 (1.4)  14.4 (1.6)  0.002      Platelet, mean (s.d.), 103/µl  249 (66)  282 (199)  0.013  Disease-related data            Symptom duration, mean (s.d.), years    10 (9)        Family history of axSpA    34 (14)        HLA B27-positive    215 (87)        Patient global assessment, mean (s.d.)    4.7 (2.6)        BASDAI score, range, 0–10, mean (s.d.)    4.1 (2.2)        BASFI score, mean (s.d.)    2.0 (2.4)        ASDAS-ESR, mean (s.d.)    2.8 (1.1)        ASDAS-CRP, mean (s.d.)    2.4 (1.3)        Grade of sacroiliitis on X-ray, mean (s.d.)    2.9 (1.0)    mSASSS, mean (s.d.)    13.1 (20.0)    Presence of spinal progression    142 (57)    Number of syndesmophytes, mean (s.d.)    4.4 (7.4)    Presence of syndesmophytes    108 (44)    Patients on NSAIDs    230 (93)    Patients on SSZ    64 (26)    Patients on TNF inhibitors    52 (21)    Variable  Control (n = 248)  AxSpA (n = 248)  P-value  Age, mean (s.d.), years  39 (11)  39 (11)  1.000  Male  193 (78)  193 (78)  1.000  BMI, mean (s.d.), kg/m2  24.4 (3.6)  24.0 (3.7)  0.286  Current smoking  104 (42)  92 (37)  0.309  Alcohol ≥3 U/day  6 (2)  7 (3)  1.000  Diabetes mellitus  14 (6)  8 (3)  0.275  Hypertension  25 (10)  36 (15)  0.171  Dyslipidaemia  18 (7)  17 (7)  1.000  History of fracture  6 (2)  30 (12)  <0.001  Laboratory findings            ESR, mean (s.d.), mm/h  9.6 (8.9)  24.3 (21.9)  <0.001      CRP mg/l, mean (s.d.)  1.3 (1.9)  10.0 (16.7)  <0.001      Haemoglobin, mean (s.d.), g/dl  14.8 (1.4)  14.4 (1.6)  0.002      Platelet, mean (s.d.), 103/µl  249 (66)  282 (199)  0.013  Disease-related data            Symptom duration, mean (s.d.), years    10 (9)        Family history of axSpA    34 (14)        HLA B27-positive    215 (87)        Patient global assessment, mean (s.d.)    4.7 (2.6)        BASDAI score, range, 0–10, mean (s.d.)    4.1 (2.2)        BASFI score, mean (s.d.)    2.0 (2.4)        ASDAS-ESR, mean (s.d.)    2.8 (1.1)        ASDAS-CRP, mean (s.d.)    2.4 (1.3)        Grade of sacroiliitis on X-ray, mean (s.d.)    2.9 (1.0)    mSASSS, mean (s.d.)    13.1 (20.0)    Presence of spinal progression    142 (57)    Number of syndesmophytes, mean (s.d.)    4.4 (7.4)    Presence of syndesmophytes    108 (44)    Patients on NSAIDs    230 (93)    Patients on SSZ    64 (26)    Patients on TNF inhibitors    52 (21)    All values given as n (%) unless otherwise stated. axSpA: axial spondyloarthritis; mSASSS: Modified Stokes Ankylosing Spondylitis Score. There were no significant differences between the groups in terms of BMI, smoking history, alcohol consumption, diagnosis of diabetes mellitus, hypertension or treatment with lipid-lowering agents. ESR, CRP and platelet levels were higher in the axSpA group than in controls (P < 0.001, P < 0.001 and P = 0.013, respectively). Haemoglobin levels were lower in the axSpA group than in matched controls (P = 0.002). Among the 248 patients with axSpA, 207 fulfilled the modified New York criteria for AS [20]. The mean symptom duration was 10 (9) years, and 215 (87%) were HLA B27 positive. The mean BASDAI and ASDAS-CRP scores were 4.1 (2.2) and 2.4 (1.3). The average grade of sacroiliitis on X-ray and mSASSS was 2.9 (1.0) and 13.1 (20.0), respectively. Of the 248 patients, 108 (44%) had one or more syndesmophytes. The majority of patients (93%) received NSAIDs, and 21% received TNF inhibitors. Assessment of TBS and BMD The mean TBS value was 1.43 (0.08) for controls and 1.38 (0.12) for axSpA patients (P < 0.001; Fig. 1A). BMD data are shown in Fig. 1B. BMD at the lumbar spine did not differ between the two groups. The mean BMD at the femoral neck was 0.98 (0.14) in controls and 0.93 (0.13) in the axSpA group; the mean BMD of total hip was 1.03 (0.14) in controls and 0.97 (0.14) in the axSpA group (both P < 0.001). Fig. 1 View largeDownload slide Trabecular bone score and BMD in patients with axial SpA and matched controls (A) Comparison of TBS among axSpA patients and controls matched for age and sex. (B) Comparison of BMD of lumbar spine, femoral neck and total hip. axSpA: axial spondylarthropathy; TBS: trabecular bone score. Fig. 1 View largeDownload slide Trabecular bone score and BMD in patients with axial SpA and matched controls (A) Comparison of TBS among axSpA patients and controls matched for age and sex. (B) Comparison of BMD of lumbar spine, femoral neck and total hip. axSpA: axial spondylarthropathy; TBS: trabecular bone score. Correlation between clinical data, inflammatory markers, BMD and TBS In both groups, the TBS showed a significant negative correlation with age and positive correlation with BMD of lumbar spine, femoral neck and total hip (Table 2). The TBS was negatively correlated with ESR and CRP in the axSpA group only. Table 2 Correlation between clinical data, inflammatory markers, BMD and trabecular bone score Group  Age, years  BMI, kg/m2  ESR, mm/h  CRP, mg/l  BMD, g/cm2  Lumbar spine  Femoral neck  Total hip  Control, TBS, L1–L4  −0.328 (<0.001)  −0.026 (0.685)  −0.032 (0.669)  0.023 (0.767)  0.373 (<0.001)  0.376 (<0.001)  0.336 (<0.001)  Axial SpA, TBS, L1–4  −0.206 (0.001)  −0.096 (0.132)  −0.230 (<0.001)  −0.170 (0.007)  0.207 (0.001)  0.296 (<0.001)  0.276 (<0.001)  Group  Age, years  BMI, kg/m2  ESR, mm/h  CRP, mg/l  BMD, g/cm2  Lumbar spine  Femoral neck  Total hip  Control, TBS, L1–L4  −0.328 (<0.001)  −0.026 (0.685)  −0.032 (0.669)  0.023 (0.767)  0.373 (<0.001)  0.376 (<0.001)  0.336 (<0.001)  Axial SpA, TBS, L1–4  −0.206 (0.001)  −0.096 (0.132)  −0.230 (<0.001)  −0.170 (0.007)  0.207 (0.001)  0.296 (<0.001)  0.276 (<0.001)  R coefficients (P-values); TBS: trabecular bone score. Comparisons of TBS and BMD according to the presence of syndesmophytes As shown in Fig. 2, TBS values were significantly lower in axSpA patients with and without syndesmophytes than in matched controls (P = 0.004 and P < 0.001, respectively). Lumbar spine BMD was significantly lower in the axSpA group without syndesmophytes than in the controls [1.13 (0.14) vs 1.18 (0.15) in controls; P = 0.008]. In contrast, the mean lumbar spine BMD was higher in the axSpA group with syndesmophytes [1.26 (0.28) vs 1.19 (0.16) in controls; P = 0.021]. Fig. 2 View largeDownload slide Comparison of trabecular bone score and BMD according to syndesmophytes (A) Comparison of trabecular bone score L1–L4 in axial spondyloarthritis patients and matched control groups according to the presence of syndesmophytes. (B) Comparison of BMD of lumbar spine in axial spondyloarthritis patients and matched control groups according to the presence of syndesmophytes. Fig. 2 View largeDownload slide Comparison of trabecular bone score and BMD according to syndesmophytes (A) Comparison of trabecular bone score L1–L4 in axial spondyloarthritis patients and matched control groups according to the presence of syndesmophytes. (B) Comparison of BMD of lumbar spine in axial spondyloarthritis patients and matched control groups according to the presence of syndesmophytes. AxSpA patient characteristics stratified according to TBS fracture risk groups Table 3 shows the demographic characteristics and disease-related variables between three TBS subgroups according to the risk of fracture [18]. Factors associated with a higher risk were male sex (P = 0.028) and higher BMI (P = 0.029). There were no significant differences among the TBS subgroups of patients in the fracture risk with respect to age, smoking status, alcohol consumption, symptom duration and HLA B27 status. Serum ESR and CRP were higher in the axSpA group with high fracture risk TBS values than in the group with low risk (P = 0.013 and P = 0.034, respectively). BASDAI, ASDAS-ESR and ASDAS-CRP did not differ between the three groups. The grade of sacroiliitis on X-ray, the mSASSS and the mean number of syndesmophytes were higher in patients with high fracture risk TBS (P = 0.006, P < 0.001 and P < 0.001, respectively). BMD levels at the lumbar spine did not differ between the three groups, although significant differences in BMD of the femoral neck and total hip were seen (P = 0.003 and P = 0.015, respectively). Table 3 Characteristics of axial SpA group stratified according to trabecular bone score risk group Variable  High risk (TBS < 1.23, n = 28)  Medium risk (TBS = 1.23–1.31, n = 27)  Low risk (TBS > 1.31, n = 193)  P-value  Age, years  43 (12)  42 (13)  38 (11)  0.061  Male, n (%)  26 (93)  24 (89)  143 (74)  0.028  BMI, kg/m2  25.7 (2.8)  24.1 (3.9)  23.8 (3.7)  0.029  Current smoking, n (%)  13 (46)  13 (48)  66 (34)  0.108  Alcohol ≥3 U/day, n (%)  3 (11)  3 (11)  22 (11)  0.994  Symptom duration, years  10.7 (11.2)  10.0 (8.3)  9.8 (9.2)  0.891  Family history of axial SpA, n (%)  2 (7)  5 (19)  27 (14)  0.458  HLA B27-positive, n (%)  27 (96)  23 (85)  165 (67)  0.273  ESR, mm/h  34.3 (24.9)  21.6 (4.2)  22.2 (21.1)  0.013  CRP, mg/l  12.0 (15.5)  17.2 (20.6)  8.6 (16.0)  0.034  Patient global assessment  5.3 (2.5)  5.1 (2.6)  4.6 (2.6)  0.301  BASDAI score, range, 0–10  4.1 (2.1)  4.4 (1.9)  4.1 (2.3)  0.728  BASFI score  2.3 (2.6)  2.7 (2.5)  1.9 (2.3)  0.197  ASDAS-ESR  3.1 (1.1)  3.0 (1.1)  2.7 (1.1)  0.126  ASDAS-CRP  2.7 (1.2)  2.9 (1.2)  2.3 (1.3)  0.055  Grade of sacroiliitis on X-ray  3.4 (0.8)  3.1 (1.0)  2.8 (1.0)  0.006  mSASSS  28.1 (22.9)  15.6 (18.7)  10.6 (18.8)  <0.001  Presence of spinal progression, n (%)  24 (86)  18 (67)  100 (52)  0.002  Number of syndesmophytes  9.5 (9.0)  4.9 (6.5)  3.6 (7.0)  <0.001  Presence of syndesmophytes, n (%)  20 (71)  13 (48)  75 (30)  0.004  Patients on NSAIDs, n (%)  28 (100)  22 (82)  180 (93)  0.025  Patients on SSZ, n (%)  5 (18)  7 (26)  52 (27)  0.590  Patients on TNF inhibitors, n (%)  8 (15)  5 (19)  39 (20)  0.565  TBS, L1–L4  1.14 (0.07)  1.27 (0.02)  1.43 (0.07)  <0.001  BMD, g/cm2              Lumbar spine  1.17 (0.26)  1.13 (0.21)  1.20 (0.22)  0.231      Femoral neck  0.88 (0.14)  0.87 (0.18)  0.94 (0.12)  0.003      Total hip  0.92 (0.13)  0.92 (0.19)  0.98 (0.13)  0.015  Variable  High risk (TBS < 1.23, n = 28)  Medium risk (TBS = 1.23–1.31, n = 27)  Low risk (TBS > 1.31, n = 193)  P-value  Age, years  43 (12)  42 (13)  38 (11)  0.061  Male, n (%)  26 (93)  24 (89)  143 (74)  0.028  BMI, kg/m2  25.7 (2.8)  24.1 (3.9)  23.8 (3.7)  0.029  Current smoking, n (%)  13 (46)  13 (48)  66 (34)  0.108  Alcohol ≥3 U/day, n (%)  3 (11)  3 (11)  22 (11)  0.994  Symptom duration, years  10.7 (11.2)  10.0 (8.3)  9.8 (9.2)  0.891  Family history of axial SpA, n (%)  2 (7)  5 (19)  27 (14)  0.458  HLA B27-positive, n (%)  27 (96)  23 (85)  165 (67)  0.273  ESR, mm/h  34.3 (24.9)  21.6 (4.2)  22.2 (21.1)  0.013  CRP, mg/l  12.0 (15.5)  17.2 (20.6)  8.6 (16.0)  0.034  Patient global assessment  5.3 (2.5)  5.1 (2.6)  4.6 (2.6)  0.301  BASDAI score, range, 0–10  4.1 (2.1)  4.4 (1.9)  4.1 (2.3)  0.728  BASFI score  2.3 (2.6)  2.7 (2.5)  1.9 (2.3)  0.197  ASDAS-ESR  3.1 (1.1)  3.0 (1.1)  2.7 (1.1)  0.126  ASDAS-CRP  2.7 (1.2)  2.9 (1.2)  2.3 (1.3)  0.055  Grade of sacroiliitis on X-ray  3.4 (0.8)  3.1 (1.0)  2.8 (1.0)  0.006  mSASSS  28.1 (22.9)  15.6 (18.7)  10.6 (18.8)  <0.001  Presence of spinal progression, n (%)  24 (86)  18 (67)  100 (52)  0.002  Number of syndesmophytes  9.5 (9.0)  4.9 (6.5)  3.6 (7.0)  <0.001  Presence of syndesmophytes, n (%)  20 (71)  13 (48)  75 (30)  0.004  Patients on NSAIDs, n (%)  28 (100)  22 (82)  180 (93)  0.025  Patients on SSZ, n (%)  5 (18)  7 (26)  52 (27)  0.590  Patients on TNF inhibitors, n (%)  8 (15)  5 (19)  39 (20)  0.565  TBS, L1–L4  1.14 (0.07)  1.27 (0.02)  1.43 (0.07)  <0.001  BMD, g/cm2              Lumbar spine  1.17 (0.26)  1.13 (0.21)  1.20 (0.22)  0.231      Femoral neck  0.88 (0.14)  0.87 (0.18)  0.94 (0.12)  0.003      Total hip  0.92 (0.13)  0.92 (0.19)  0.98 (0.13)  0.015  All values given as mean (s.d.) unless otherwise stated. mSASSS: modified stokes AS score; TBS: trabecular bone score. Linear regression analysis for TBS among axSpA patients Table 4 shows the results of univariate and multivariate analyses of TBS in the axSpA group. Univariate analysis revealed that TBS was associated with age, current smoking status, HLA B27, ESR, CRP, the grade of sacroiliitis and the mSASSS. In the multivariate analyses, ESR, CRP and mSASSS were significantly associated with the TBS value. Table 4 Linear regression analysis of the trabecular bone score in the lumbar spine among patients with axial SpA (n = 248) Variable  Univariate model β (95% Cl)  Multivariate ESR modela β (95% Cl)  Multivariate CRP modela β (95% Cl)  Demographic            Age, years  −0.002 (−0.004, −0.001)*          Gender, male  0.033 (−0.003, 0.069)          Current smoking  −0.033 (−0.064, −0.002)*          Alcohol ≥3 U/day  0.012 (−0.036, 0.059)          BMI, kg/m2  −0.003 (−0.007, 0.001)          HLA B27  −0.050 (−0.094, −0.006)*      Medication            NSAIDs  −0.016 (−0.073, 0.042)          SSZ  0.029 (−0.005, 0.063)          TNF inhibitor  −0.024 (−0.061, 0.013)      Disease activity            ESR, mm/h  −0.001 (−0.002, −0.001)*  −0.001 (−0.002, −0.001)*        CRP, mg/l  −0.001 (−0.002, 0.000)*    −0.001 (−0.002, 0.000)*      BASDAI  0.001 (−0.006, 0.007)          ASDAS  −0.010 (−0.022, 0.002)      Radiographic scores            Sacroiliitis on X-ray  −0.029 (−0.043, −0.015)*          mSASSS  −0.002 (−0.003, −0.001)*  −0.001 (−0.002, −0.000)*  −0.001 (−0.002, −0.000)*  Adjusted R2    0.434  0.396  Variable  Univariate model β (95% Cl)  Multivariate ESR modela β (95% Cl)  Multivariate CRP modela β (95% Cl)  Demographic            Age, years  −0.002 (−0.004, −0.001)*          Gender, male  0.033 (−0.003, 0.069)          Current smoking  −0.033 (−0.064, −0.002)*          Alcohol ≥3 U/day  0.012 (−0.036, 0.059)          BMI, kg/m2  −0.003 (−0.007, 0.001)          HLA B27  −0.050 (−0.094, −0.006)*      Medication            NSAIDs  −0.016 (−0.073, 0.042)          SSZ  0.029 (−0.005, 0.063)          TNF inhibitor  −0.024 (−0.061, 0.013)      Disease activity            ESR, mm/h  −0.001 (−0.002, −0.001)*  −0.001 (−0.002, −0.001)*        CRP, mg/l  −0.001 (−0.002, 0.000)*    −0.001 (−0.002, 0.000)*      BASDAI  0.001 (−0.006, 0.007)          ASDAS  −0.010 (−0.022, 0.002)      Radiographic scores            Sacroiliitis on X-ray  −0.029 (−0.043, −0.015)*          mSASSS  −0.002 (−0.003, −0.001)*  −0.001 (−0.002, −0.000)*  −0.001 (−0.002, −0.000)*  Adjusted R2    0.434  0.396  a Adjusted for age, smoking status, HLA B27, sacroiliitis on X-ray and mSASSS. * P < 0.05. mSASSS: modified Stokes Ankylosing Spondylitis Score. Discussion In the present study, TBS was used to assess bone microarchitecture and investigate the relationship between systemic inflammation, radiographic progression and bone quality in patients with axSpA. When assessed by TBS, patients with axSpA were seen to have poor bone quality in the lumbar spine compared with matched controls. Markers of systemic inflammation, such as ESR or CRP, correlated with TBS in patients with axSpA, but not in controls. Also, systemic inflammatory markers and spinal radiographic progression scores were independently associated with low TBS. AxSpA is characterized by ossification of the entheses and formation of syndesmophytes, but paradoxically there is also loss of trabecular bone adjacent to sites of inflammation. Trabecular and cortical compartments appear to have different reactions to inflammation; in AS, inflammation has a direct effect on the trabecular bone of the vertebrae, but not the cortical bone [21], and therefore AS patients show low trabecular BMD in the spine QCT [8]. DXA-based BMD cannot distinguish between cortical and trabecular compartments and cannot reflect trabecular bone microarchitecture alterations, and therefore potentially result in inconsistent data in patients with AS [22–24]. TBS has the advantage of being able to directly assess the microarchitecture of trabecular bone at the lumbar spine, which is the main area of involvement in axSpA patients. A previous study showed that sacroiliitis on MRI correlates with BMD at the femoral neck and total hip; however, sacroiliitis on MRI was not associated with lumbar BMD [25]. This indicates that bone loss at the spine in axSpA patients may be due to the localized effects of inflammation. Our data showed that BMD at the lumbar spine did not differ between axSpA and non-axSpA subjects despite differences in bone microarchitecture measured by TBS. This may be related to both an overestimation of BMD due to the presence of syndesmophytes and insufficient reflection of changes in the quality of trabecular bone caused by chronic inflammation. In the subgroup analysis according to the presence of syndesmophytes, we found that the presence of syndesmophytes was associated with lower TBS but higher BMD in the lumbar spine. This result is consistent with earlier studies also reporting lower trabecular volumetric BMD measured by QCT but higher cortical volumetric BMD in AS [8, 26]. The EULAR taskforce made a series of recommendations with respect to the use of imaging for the management of SpA. They proposed that, in patients with axSpA without syndesmophytes in the lumbar spine on conventional radiography, osteoporosis should be assessed using hip DXA and anterior–posterior spine DXA; in patients with syndesmophytes in the lumbar spine on conventional radiography, osteoporosis should be assessed using hip DXA supplemented by either spine DXA (lateral projection) or possibly QCT of the spine [27]. Although spinal DXA using a lateral view has the advantage of allowing selection of a trabecular zone in the centre of the vertebral bodies, it has poor precision due to the difficulty in positioning the patients. Moreover, due to the overlap of the last rib and iliac crest, only L3 can be measured in some cases [13, 28]. QCT cannot be performed routinely in a clinical setting due to the high radiation dose required, cost and limited access to such devices [10]. Therefore, a tool that is unaffected by spinal progression is needed to assess osteoporosis and to predict fracture risk in axSpA. Our findings indicate that TBS may be a valuable alternative tool for the assessment of bone quality influenced by inflammation in patients with axSpA, regardless of spinal progression. A recent small study also showed that TBS is not affected by syndesmophytes in patients with axSpA [29]. Including the assessment of bone quality using TBS, as well as BMD measurement, may improve the predictive value for the development of new vertebral fractures. The data presented here also show that ESR and CRP levels were independently associated with TBS after adjusting for confounding factors. In contrast with ESR and CRP, we found that BASDAI and ASDAS did not correlate with TBS. In general, clinical disease activity scores such as BASDAI and ASDAS comprise self-reported measures. These scores can be influenced by ankylosis and secondary changes in AS. Therefore, bony changes caused by inflammation may be more closely related to objective inflammatory markers than to clinical disease activity scores. The association between TBS and inflammatory markers suggests that in patients with axSpA TBS may reflect bone quality alteration in the presence of systemic inflammation. The correlation between TBS and inflammatory markers among axSpA patients suggests that TBS reflects inflammation-mediated changes in trabecular bone in the lumbar spine, and therefore, assessment of bone quality in the spine using the TBS may be a better predictor of the risk of osteoporosis and vertebral fracture in patients with axSpA. In addition, spinal radiographic progression score (mSASSS) was an independent risk factor for low TBS. Spinal radiographic progression reflects the severity of cumulative inflammation in the spine, and our data suggest that spinal radiographic progression may increase the risk of osteoporosis as it is associated with poor bone quality, as well as stiffening of the spine, in patients with axSpA. Longitudinal studies are required to examine whether the TBS is predictive of vertebral fracture. Lumbar TBS is associated with an age-dependent variable, and a significant decrease in TBS has been observed with ageing in both male and female subjects [11, 30]. In the present study, age was negatively correlated with TBS in both the axSpA and control groups. In the axSpA group, age was associated with TBS in univariate analysis, but the association was not observed in multivariate analysis after adjusting for confounding factors. Therefore, it appears that TBS in axSpA is more influenced by the severity of systemic and local inflammation than ageing, and the strict control of inflammation may help to decrease the risk of osteoporosis among patients. This finding should be assessed further in prospective longitudinal studies. This study has a few limitations. First, vitamin D and PTH were not measured. These variables are important for skeletal health and could affect the risk of osteoporosis. Second, only 52 (21%) patients received TNF inhibitors. This is a relatively small number when considered alongside the total number of axSpA patients; therefore, the regression analysis could be underpowered. In conclusion, TBS assessment revealed poor bone quality in the lumbar spine in patients with axSpA when compared with matched controls. Systemic inflammatory markers were negatively correlated with TBS in axSpA and spinal radiographic progression, and inflammatory markers were independently related to low TBS. These data support the use of TBS as a useful clinical tool to identify the risk of osteoporosis in patients with axSpA. Funding: No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript. Disclosure statement: The authors have declared no conflicts of interest. References 1 Sieper J, Poddubnyy D. Axial spondyloarthritis. Lancet  2017; 390: 73– 84. Google Scholar CrossRef Search ADS PubMed  2 Szentpetery A, Horvath A, Gulyas K et al.   Effects of targeted therapies on the bone in arthritides. Autoimmun Rev  2017; 16: 313– 20. Google Scholar CrossRef Search ADS PubMed  3 Kang KY, Kwok SK, Ju JH, Hong YS, Park SH. Assessment of fracture risk in patients with axial spondyloarthritis: a case-control study using the fifth Korean National Health and Nutrition Examination Survey (KNHANES V). Scand J Rheumatol  2015; 45: 23– 31. Google Scholar CrossRef Search ADS PubMed  4 Magrey M, Khan MA. Osteoporosis in ankylosing spondylitis. Curr Rheumatol Rep  2010; 12: 332– 6. Google Scholar CrossRef Search ADS PubMed  5 NIH Consensus Development Panel on Osteoporosis Prevention Diagnosis, and Therapy. Osteoporosis prevention, diagnosis, and therapy. JAMA  2001; 285: 785– 95. CrossRef Search ADS PubMed  6 Lorentzon M, Cummings SR. Osteoporosis: the evolution of a diagnosis. J Intern Med  2015; 277: 650– 61. Google Scholar CrossRef Search ADS PubMed  7 Geusens P, Lems WF. Osteoimmunology and osteoporosis. Arthritis Res Ther  2011; 13: 242. Google Scholar CrossRef Search ADS PubMed  8 Klingberg E, Lorentzon M, Gothlin J et al.   Bone microarchitecture in ankylosing spondylitis and the association with bone mineral density, fractures, and syndesmophytes. Arthritis Res Ther  2013; 15: R179. Google Scholar CrossRef Search ADS PubMed  9 Cosman F, de Beur SJ, LeBoff MS et al.   Clinician's Guide to Prevention and Treatment of Osteoporosis. Osteoporos Int  2014; 25: 2359– 81. Google Scholar CrossRef Search ADS PubMed  10 Harvey NC, Gluer CC, Binkley N et al.   Trabecular bone score (TBS) as a new complementary approach for osteoporosis evaluation in clinical practice. Bone  2015; 78: 216– 24. Google Scholar CrossRef Search ADS PubMed  11 Silva BC, Leslie WD, Resch H et al.   Trabecular bone score: a noninvasive analytical method based upon the DXA image. J Bone Miner Res  2014; 29: 518– 30. Google Scholar CrossRef Search ADS PubMed  12 Silva BC, Walker MD, Abraham A et al.   Trabecular bone score is associated with volumetric bone density and microarchitecture as assessed by central QCT and HRpQCT in Chinese American and white women. J Clin Densitom  2013; 16: 554– 61. Google Scholar CrossRef Search ADS PubMed  13 Kolta S, Briot K, Fechtenbaum J et al.   TBS result is not affected by lumbar spine osteoarthritis. Osteoporos Int  2014; 25: 1759– 64. Google Scholar CrossRef Search ADS PubMed  14 Rudwaleit M, Landewe R, van der Heijde D et al.   The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part I): classification of paper patients by expert opinion including uncertainty appraisal. Ann Rheum Dis  2009; 68: 770– 6. Google Scholar CrossRef Search ADS PubMed  15 Garrett S, Jenkinson T, Kennedy LG et al.   A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index. J Rheumatol  1994; 21: 2286– 91. Google Scholar PubMed  16 Calin A, Garrett S, Whitelock H et al.   A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath Ankylosing Spondylitis Functional Index. J Rheumatol  1994; 21: 2281– 5. Google Scholar PubMed  17 van der Heijde D, Lie E, Kvien TK et al.   ASDAS, a highly discriminatory ASAS-endorsed disease activity score in patients with ankylosing spondylitis. Ann Rheum Dis  2009; 68: 1811– 8. Google Scholar CrossRef Search ADS PubMed  18 McCloskey EV, Oden A, Harvey NC et al.   A meta-analysis of trabecular bone score in fracture risk prediction and its relationship to FRAX. J Bone Miner Res  2016; 31: 940– 8. Google Scholar CrossRef Search ADS PubMed  19 Wanders AJ, Landewe RB, Spoorenberg A et al.   What is the most appropriate radiologic scoring method for ankylosing spondylitis? A comparison of the available methods based on the Outcome Measures in Rheumatology Clinical Trials filter. Arthritis Rheum  2004; 50: 2622– 32. Google Scholar CrossRef Search ADS PubMed  20 van der Linden S, Valkenburg HA, Cats A. Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum  1984; 27: 361– 8. Google Scholar CrossRef Search ADS PubMed  21 Schett G. Structural bone changes in spondyloarthritis: mechanisms, clinical impact and therapeutic considerations. Am J Med Sci  2011; 341: 269– 71. Google Scholar CrossRef Search ADS PubMed  22 Donnelly S, Doyle DV, Denton A et al.   Bone mineral density and vertebral compression fracture rates in ankylosing spondylitis. Ann Rheum Dis  1994; 53: 117– 21. Google Scholar CrossRef Search ADS PubMed  23 Ghozlani I, Ghazi M, Nouijai A et al.   Prevalence and risk factors of osteoporosis and vertebral fractures in patients with ankylosing spondylitis. Bone  2009; 44: 772– 6. Google Scholar CrossRef Search ADS PubMed  24 Klingberg E, Geijer M, Gothlin J et al.   Vertebral fractures in ankylosing spondylitis are associated with lower bone mineral density in both central and peripheral skeleton. J Rheumatol  2012; 39: 1987– 95. Google Scholar CrossRef Search ADS PubMed  25 Kim HN, Jung JY, Hong YS, Park SH, Kang KY. Severe bone marrow edema on sacroiliac joint MRI increases the risk of low BMD in patients with axial spondyloarthritis. Sci Rep  2016; 6: 22158. Google Scholar CrossRef Search ADS PubMed  26 Lange U, Teichmann J, Strunk J, Muller-Ladner U, Schmidt KL. Association of 1.25 vitamin D3 deficiency, disease activity and low bone mass in ankylosing spondylitis. Osteoporos Int  2005; 16: 1999– 2004. Google Scholar CrossRef Search ADS PubMed  27 Mandl P, Navarro-Compan V, Terslev L et al.   EULAR recommendations for the use of imaging in the diagnosis and management of spondyloarthritis in clinical practice. Ann Rheum Dis  2015; 74: 1327– 39. Google Scholar CrossRef Search ADS PubMed  28 Larnach TA, Boyd SJ, Smart RC et al.   Reproducibility of lateral spine scans using dual energy X-ray absorptiometry. Calcif Tissue Int  1992; 51: 255– 8. Google Scholar CrossRef Search ADS PubMed  29 Wildberger L, Boyadzhieva V, Hans D, Stoilov N, Rashkov R, Aubry-Rozier B. Impact of lumbar syndesmophyte on bone health as assessed by bone density (BMD) and bone texture (TBS) in men with axial spondyloarthritis. Joint Bone Spine  2017; 84: 463– 466. Google Scholar CrossRef Search ADS PubMed  30 Bazzocchi A, Ponti F, Diano D et al.   Trabecular bone score in healthy ageing. Br J Radiol  2015; 88: 20140865. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oup.com

Journal

RheumatologyOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off