Objective: To investigate factors affecting X-ray structure of the spine in patients with ankylosing spondylitis (AS). Methods: A total of 206 AS patients were recruited. Clinical and laboratory parameters in AS patients were recorded in detail. Disease activity index [Bath ankylosing spondylitis disease activity index (BASDAI), ankylosing spondylitis disease activity score (ASDAScrp)], X-ray structural damage index-modified stoke ankylosing spondylitis spine score (mSASSS) and grading results of radiographic examination of sacroiliac joint were calculated. Statistical analysis using Statistical Package form Soci-science(SPSS) 17.0 Chi-square test, rank test, Logistics regression analysis and other statistical methods were used. Differences of mSASSS levels, spine involvement (mSASSS>0) and rates of bone bridge formation were compared between different groups. Results: Incidences of spine involvement (100%) and bone bridge formation(65.2%) in AS patients ≥40 years old were significantly higher than those in AS patients <40 years old (90.6%、31.9%)(χ²=4.651, P=0.031; χ²=16.647, P<0.01), and the level of mSASSS was also higher (Z=5.575, P<0.01). In AS patients with BMI ≥24 kg/m², disease duration ≥5 years (49.2%, 50.4%), rates of bone bridge formation was significantly higher than those in AS with BMI <24 kg/m², but the disease duration (34.5%, 19.7%)(χ²=4.014, P=0.045; χ²=18.173, P=0.03), and mSASSS values were significantly higher (Z=2.281, P=0.023, Z=4.828, P<0.01). Bone bridge formation rate in smoking patients (50.6%) was significantly higher than that in non-smoking patients (31.0%) (χ²=7.346, P=0.007) and mSASSS value was significantly higher (Z=2.045, P=0.041). Bone bridge formation rates in AS with high-ESR and high-CRP(48.6%, 49.0%) were significantly higher than those in patients with normal-ESR and normal-CRP(25.6%, 28.9%)(χ²=10.784, P=0.001; χ²=8.102, P=0.004) and mSASSS value was clearly higher(Z=2.379, P<0.01; Z=3.112, P<0.01). Bone bridge formation rate in AS with BASDAI≥4 or ASDAScrp≥2.1 groups (52.8%, 46.4%) were significantly higher than that in AS with BASDAI<4 or ASDAScrp<2.1 groups (34.2%, 30.7%) (χ²=5.681, P=0.017; χ²=4.646, P=0.031) and mSASSS values were significantly higher (Z=3.887, P<0.01; Z=3.895, P=0.004). Rates of bone bridge formation among different X-ray grading of sacroiliac joint (10.8%, 35.6%, 60.3%) and MRI findings (33.3%, 50.0%, 15.4%) differed with each other (χ²=25.714, P<0.01; χ²=6.855, P=0.032). Logistics regression analysis showed that BMI [OR(95%CI)=1.145(1.037, 1.265), P<0.01], disease duration [OR(95%CI)=1.144(1.055, 1.239), P<0.01], smoking [OR(95%CI)=2.832(1.343, 5.969), P<0.01] and sacroiliac joint X-ray staging [OR(95%CI)=2.584(1.337, 4.997), P<0.01] were risk factors for the bone bridge formation in spine of AS. Conclusion Spinal involvement in AS is related to disease activity. Bone bridge formation correlateswith disease duration, BMI and disease-status, especially with smoking.

Objective To explore the changes of serum vitamin D binding protein (VDBP) levels in patients with rheumatoid arthritis (RA) and the clinical significance of association between serum VDBP with secondary osteoporosis (OP) in RA.Methods One hundred and sixty patients with RA were enrolled in the study.Eighty-three normal subjects were recruited as the control group.The concentration of serum VDBP was determined by enzyme-linked immuno sorbent assay (ELISA),and bone mineral density (BMD) was measured by dual energy X-ray absorptiometry.Clinical and laboratory indexes of RA patients were recorded in detail and disease activity (DAS28) score,health assessment questionnaire (HAQ) and Sharp score according to X-ray examination of both hands were calculated simultaneously.The t test was used to compare the metrological data between the two groups,and the x2 test was used to compare the intergroup rate.The correlation analysis was tested by Pearson correlation analysis,the ROC curve was used to analyze the threshold of the serum VDBP,and the multivariate logistic regression analysis was used for multivariate analysis.Results ① Serum levels of VDBP in RA patients were higher than that in control group [(414±12) ng/ml vs (79±12) ng/ml,t=20.082,P＜0.01].Positive rate of serum levels of VDBP was 67.0％(118/176) in RA patients,which was higher than that in the control group (4.8％)(x2 =87.651,P＜0.01).② The threshold of serum VDBP levels for diagnosing RA was 193.74 ng/ml (AUC=0.943,Youden index=0.796,P＜0.01).③ DAS28 sore in group with positive VDBP was significantly higher than that in group with negative VDBP (P9=0.025).There was significant difference regarding on the incidence of OP in female RA patients between groups with positive and negative VDBP [41.9％(54/129) vs 31.8％(14/44),x2=4.325,P=0.038].④ Linear correlation analysis found that DAS28in RA patients was positively correlated with serum VDBP levels (r=0.252,P=0.019).And anti-CCP was negatively correlated with serum VDBP levels (r=-0.150,P=0.049).⑤ Results of logistic regression analysis showed that sex [OR=9.841,95％CI (1.349,71.810),P=0.024],age [OR=1.154,95 ％CI (1.069,1.245),P＜0.01] and Sharp score [OR=1.102,95％CI (1.002,1.021),P=0.018] were risk factors for OP in RA patients.Conclusion Serum VDBP levels are significantly higher in patients with RA.Meanwhile,serum VDBP levels are correlated with disease activity and secondary OP in RA.

Objective To explore the value of patient global assessment (PGA) on evaluating disease activity in patients with axial spondyloarthritis (SpA),Methods A total of 222 patients with axial SpA were recruited.Scores of PGA,disease activity index [Bath ankylosing spondylitis disease activity index (BASDAI),ankylosing spondylitis disease activity score (ASDAS)crp] and spondyloarthritis research consortium of Canada (SPARCC) were calculated.Differences of PGA scores between different disease activity groups in axial SpA were compared and correlations between different disease activity index with PGA scores were analyzed.Statistical analyses were performed using Statistical Product and Service Solutions (SPSS) software (version 17.0).Comparison of frequency among different groups was performed by x2 test.Rank-sum test was used to compare the median of measurement data in different groups when the data were skewed in distribution.Cut-off value of PGA for assessing disease activity in axial SpA was calculated by ROC curve.Results Medians of PGA score in groups with BASDAI remission[3(1,4) vs 5(4,7)] and ASDAScrp remission [1(1,2) vs 4(2,5)] were lower than that in disease activity group (P＜0.01).BASDAI scores [1.80(1.20,2.90) vs 3.40(2.28,4.63) vs 5.15 (4.08,5.88)] and ASDAScrp scores [2.19(1.34,2.76) vs 2.86(2.08,3.54) vs 4.08(2.96,4.41)] were significant different among PGA groups (≤3,4-6 and ≥7) (P＜0.01).Differences of SPARCC scores [6.00(0,18.00) vs 7.50(3.75,18.00) vs 18.50(6.75,24.50)] were statistically significant among PGA groups (Z=7.427,P=0.037).Erythrocyte sedimentation rate (ESR) [12.00(5.00,23.00) mm/1 h vs 19.50(7.00,44.50) mm/1 h vs 18.00(7.75,54.75) mm/1 h],C-reactive protein (CRP) [7.85(2.37,22.49) mg/L vs 10.07(3.02,28.51) mg/L vs 21.28(7.14,37.74) mg/L] and Bath ankylosing spondylitis functional index (BASFI) [0.70(0.10,1.30) vs 2.25(0.60,3.30) vs 2.85(0.83,6.53)] were also different among PGA groups (P＜0.01,separately).Proportion of axial SpA patients in BASDAI disease activity group or ASDAScrp higher disease activity group were different among PGA groups (P＜0.01,separately),while represented as positive correlations (P＜0.01,separately).Correlation analyses revealed that PGA was positively correlated with ASDAScrp (r=0.694),BASDAI(r=0.616),SPARCC (r=0.271),ESR (r=0.288),CRP(r=0.215),occipital wall distance (r=0.196),finger-floor distance (r=0.385) and negatively correlated with Sschober's test (r=-0.195) (P＜0.05).Receiver operator characteristic (ROC) curve analysis found that PGA-BASDAI AUC was 0.813,the cut off value of PGA was 3.5 and PGA-ASDAScrp AUC was 0.860,the cut off value of PGA was 2.5.Conclusion PGA has good correlations with the disease activity indexes in axial SpA patients.It can also reflect the degree of inflammation in iconography.PGA may reflect disease activity especially when the value of PGA is around 3.

Objective To investigate the value of serum levels of peroxisome proliferater-activated receptor (PPAR)γin patients with rheumatoid arthritis (RA) and to explore the associations between serum PPARγwith disease activity of RA and RA-associated osteoporosis (OP).Methods One hundred and one cases of hospitalized patients with RA were enrolled.A total of 88 normal subjects during the same period were recruited as the control group.Levels of serum PPARγwere detected by enzyme linked immunosorbent assay (ELISA).Bone mineral density (BMD) was measured by dual energy X-ray absortiometry.All the clinical and laboratory indexes of RA patients were recorded in detail.T-test (or t'-test) was used for comparison of measurement data between the two groups,non-parametric test was applied for skewed distribution data.Comparison of incidence was analyzed with x2 test,correlation analysis was represented ascorrelation coefficient (r).Results Binary logistic regression analysis was used for multivariate regression analysis.Serum levels of PPARγ(3.38/4.00 ng/ml) in RA patients were higher than thosein normal subjects (2.63/1.76) ng/ml (Z=3.204,P=0.001).The positive rate of serum levels of PPARγin RA was 35.6％ (36/101),while the positive rate in the controls was (2.3％,2/88,x2=32.602,P＜0.01).The incidence of OP in RA was 34.7％(35/101;while the levels of serum PPARγ in RA without OP at femur area (femoral neck,total hip) and lumbar spine (L1,L1-4) were higher than that in RA with OP (P＜0.05).Serum levels of PPARγ between groups with different disease activity had no significant difference (P＞0.05).Serum levels of PPARγ in RA with negative RF or negative anti-CCP were higher than thosetin the RA group with positive RF or positive anti-CCP(P＜0.05).Serum levels of PPARγ in RA were negatively correlated with serum RF,anti-CCP,hemoglobin (P ＜0.05-0.001),and positively correlated with erythrocyte sedimentation rate,platelet,BMD at sites of femur neck andWard (P＜0.05-0.001).Results of binary logistic regression analysis showed that levels of serum PPARγwere protective factors in RA for OP at femurneck [OR=0.577,P=0.005,95％CI (0.394-0.846)],and total hip [OR=0.754,P=0.033,95％CI (0.581-0.978)].Condusion Serum levels of PPARγ in patients with RA are significantly increased,and negatively correlate with autoantibodies.Serum levels of PPARγare protective factors for OP in RA.

Objective To explore the serum levels of fibroblast growth factor 23 (FGF23) in patients with rheumatoid arthritis (RA) and to investigate the relationship between FGF23 and RA disease activity and the occurrence of osteoporosis (OP).Methods Serum levels of FGF23 from 174 cases of patients with RA and 88 normal subjects were detected by enzyme linked immunosorbent assay (ELISA).Bone mineral density (BMD) was measured by dual energy X-ray absorptiometry.All the clinical and laboratory indexes of RA patients were recorded in details,disease activity score (DAS28) and health assess questionnaire (HAQ) were also calculated in the meantime.Radiographic changes in both hands of RA patients were assessed by Sharp's method.T test,nonparametric test,x2 test,correlation analysis and Logistic regressive analysis were used for statistical analysis.Results Serum levels of FGF3 [145.46(67.67,245.93) pg/ml] in RA patients were higher than the control group [32.64(12.34,44.70) pg/ml,Z=11.416,P＜0.01].The positive rate of serum levels of FGF23 (≥71.95 pg/ml) in RA was 74.7％(130/174),while the positive rate in control was 4.5％(4/88,x2=115.16,P＜0.01).The threshold of FGF23 serum levels for diagnosing RA was 48.56 pg/ml (AUC=0.932,Youden index=0.743,P＜0.01,sensitivity 89.1％,specificity 85.2％).In RA patients with serum FGF23 ≥48.56 pg/ml,compared with negative FGF23 group,VAS,HAQ,number of joint swelling and BMD at femoral neck,Ward,GT,Total hip,L4 and L1-4 were significantly higher in FGF23 positive group (P＜0.05).Linear correlation analysis found that in RA patients with serum FGF23 ≥48.56 pg/ml,anti-CCP was negatively correlated with serum FGF23 levels (r=-0.171,P=0.035).And DAS28 was positively correlated with serum FGF23 (r=0.163,P=0.045).BMD at femoral neck,Ward,GT,Total hip,L4 and L1-4 were negatively correlated with serum FGF23 (P＜0.05).Results of logistic regression analysis showed that sex (OR=8.518,95％CI (2.636,27.522),P＜0.01,age [OR=1.129,95％CI (1.079,1.180),P＜0.01] and Sharp score [OR=1.008,95％CI(1.003,1.013),P=0.001]were risk factors for OP in RA patients.BMI[OR=0.801,95％CI(0.707,0.909),P=0.001] was a protective factor for OP in RA patients.Conclusion Serum FGF23 level is significantly higher in RA patients.Meanwhile,the serum FGF23 level correlates with RA disease activity and BMD.

Objective:To explore the clinical value of sarcopenia and vitamin D deficiency on gluco-corticoid induced osteoporosis (GIOP) in patients with rheumatoid arthritis (RA).Methods:Three hundred and eleven patients with RA from January 2017 to December 2018 were enrolled in the study. One hundred and fifty-eight sex, age-matched normal subjects were recruited as control group. Clinical and laboratory features, daily dosage and treatment duration of glucocorticoid (GC) were recorded in detail. Skeletal muscle mass was measured by biological electrical impedance. Serum levels of 25-hydroxy vitamin D [25(OH)D] were examined using electro-chemiluminescence. Bone mineral density (BMD) at total hip and lumbar vertebra were detected by dual energy X-ray absorptiometry (DEXA). Numerical data and categorical data comparisons were analyzed using χ2 test, non-parametric test, Logistic regression analysis test. Results:① The prevalence of osteoporosis (OP) in RA patients was 33.4%(104/311), which was higher than that in the control group 12.7%(20/158)( χ2=23.267, P<0.01). Percentage of GC taking in 311 RA patients was 56.6%(176/311), and the prevalence of GIOP was 40.9%(72/176). The prevalence of sarcopenia in RA patients was 61.7%(192/311), which was higher than that in the control group [9.0%(14/156), χ2=117.310, P<0.01]. The prevalence of vitamin D deficiency in RA patients was 81.7%(254/311), which was higher than that in control group [38.0%(60/158), χ2=90.415, P<0.01]. ② The prevalence of OP in RA without sarcopenia was 17.6% (21/119), which was lower than that in patients with sarcopenia [43.2%(83/192), χ2=21.601, P<0.01]. In condition without GC, the prevalence of OP in RA without sarcopenia was 9.8%(6/61), which was significantly lower than that in patients with sarcopenia [35.1%(26/74), χ2=11.834, P<0.01]. Under circumstances with GC, the prevalence of OP in RA without sarcopenia (25.9%, 15/58), which was significantly lower than that in patients with sarcopenia (48.3%, 57/118, χ2=8.103, P<0.01). ③ No matter whether existing vitamin D deficiency or not, the prevalence of OP in RA without GC was 23.7%(32/135), which was significantly lower than that in patients with GC [40.9%(72/176), χ2=10.161, P<0.01]. In patients without vitamin D deficiency, the prevalence of OP in RA without GC was 21.4%(6/28), which was similar to that in patients with GC [31.0%(9/29), χ2=0.678, P>0.05]. In the case of vitamin D deficiency, the prevalence of OP in RA without GC was 24.3%(24/107), which was significantly lower than that in patients with GC [42.9% (63/147), χ2=9.370 2, P<0.01]. ④ In RA patients with GC, age( t=5.313, P<0.01), Sharp score ( Z=2.999, P<0.01), disease duration ( Z=2.141, P<0.05) and treatment duration of GC ( Z=2.460, P<0.05) were higher in group with GIOP than that in group without GIOP, while erythrocyte sedimentation rate (ESR)( Z=2.262, P<0.05), C-reactive protein levels (CRP) ( Z=2.551, P<0.05) and body mass index (BMI) ( t=2.425, P<0.05) were lower and the composition ratio of X-ray staging was worse ( χ2=12.484, P<0.01).⑤ Logistic regression analysis (LR Backward) showed that female gender [ OR(95% CI)=14.240(3.878, 52.288), P<0.01], age [ OR(95% CI)=1.079(1.042, 1.118), P<0.01] and sarcopenia [ OR(95% CI)=2.470(1.192, 5.120), P<0.05] were the risk factors for GIOP in RA patients. Conclusion:The proportion of treatment with GC in RA patients is very high (about 60%), and the prevalence of GIOP is 40.9%, which is closely related to sarcopenia and vitamin D deficiency.

Objective To explore the prevalence and reference value of disease features of patients with spondyloarthritis. Methods Spondyioarthritis features and laboratory indexes and radiographic indexes of 505 patients with spondyloarthritis (SpA) including 353 patients with ankylosing spondylitis (AS), 62 patients with non-radiographic axial spondyloarthritis (nr-axSpA) and 90 patients with peripheral spondyloarthritis (pSpA) were recorded. One-way analysis of variance, Kruskal-Wallis test, x2-test, Logistic regression were used for statistical analysis. Results Sex ratio ( x2=20.673, P<0.01), age ( x2=22.258, P<0.01), disease duration ( x2=76.052, P<0.01) were different among AS, nr-axSpA and pSpA. Besides, Bath ankylosing spondylitis disease activity index (BASDAI), ankylosing spondylitis disease activity score (ASDAScrp), erythrocyte sedimentation rate (ESR), C-reactionprotein (CRP) and Bath ankylosing spondylitis functional index (BASFI)were different among SpA subgroups ( x2/F=13.196-40.028, P<0.01). Prevalence of inflammatory back pain, peripheral arthritis, preceding infection, positive human lymphocyte antigen (HLA)-B27 and elevated CRP were different among SpA subgroups ( x2=11.416, 32.657, P<0.01). Prevalence of dactylitis in SpA with positive HLA-B27 was lower than that in SpA with negative HLA-B27 ( x2=5.414, P=0.02). Prevalence of enthesitis and dactylitis in SpA patients with peripheral arthritis was higher than that in SpA without peripheral arthritis involvement ( x2=7.177, 14.428, P<0.01). Prevalence of good response to Non-steroid anti-inflammatory drugs. (NSAIDs) in patients with anterior uveitis involvement was higher than SpA without anterior uveitis involvement ( x2=4.578, P=0.032). SpA patients were stratified by total number of SpA features into 4 subgroups (n≤1, n=2, n=3, n≥4). Prevalence of inflammatory back pain, positive HLA-B27, good response to NSAIDs were the top three in all subgroups. Inflammatory back pain and HLA-B27 (+) were risk factors for axSpA (OR=3.254, 3.323, P<0.01). Peripheral arthritis, dactylitis, and preceding infection were risk factors for pSpA (OR=3.759, 4.134, 17.044, P<0.01). Conclusion Inflammatory back pain, HLA-B27 (+) and good response to NSAIDs should be emphasized for the diagnosis of SpA. Inflammatory back pain and HLA-B27(+) always means axSpA. Peripheral arthritis, dactylitis and preceding infection always indicates pSpA.