No difference in antidepressant prescription in rheumatoid arthritis and controls. Results from a population-based, matched inception cohort.

OBJECTIVE: Depression occurs at least two times more often in rheumatoid arthritis (RA) patients than in controls, but little is known about the treatment of depression in RA. The primary objective of this study was to compare the 1 year period prevalence of antidepressant prescription in patients with RA versus controls. METHOD: We included a retrospective inception cohort of 509 patients with incident RA and 2545 frequency-matched population controls ascertained from 1995 to 2002. The cohort was followed until 31 December 2017 and linked with nationwide Danish registers. From the Danish National Prescription Register, we obtained information on redeemed prescriptions of antidepressants (Anatomical Therapeutic Chemical code N06A). RESULTS: We did not demonstrate significant differences in the 1 year period prevalence ratios and the incidence rate ratios for either antidepressant prescription or the frequency of hospital admissions with depressive episode. The most frequent indication for antidepressant prescription in patients with RA was depression. Cox regression analyses showed no association between RA and antidepressant prescription. CONCLUSION: We found no significant differences in the occurrence of antidepressant prescription in patients with RA versus matched controls. The main indication for antidepressant prescription in RA was depression.

Mortality and its predictors in patients with rheumatoid arthritis: a Danish population-based inception cohort study.

OBJECTIVES: To investigate mortality and its predictors in a retrospectively defined population-based rheumatoid arthritis (RA) inception cohort Method: We included patients ascertained with incident RA from a region in the southern part of Denmark from 1995 to 2002. All patients fulfilled the 1987 American College of Rheumatology criteria for RA. The patients were followed from RA classification until death, emigration, or end of follow-up on 31 December 2013. We used personal record linkage with national public registers to obtain information on education, employment, cohabitation, comorbidity, and vital status. RESULTS: The cohort comprised 509 patients, of whom 200 (39%) died during 6079 person-years. The most frequent underlying causes of death were cardiovascular disease (34%), neoplasms (26%), and respiratory disease (12%). In rheumatoid factor (RF)-positive males, the standardized mortality ratio (95% confidence interval) from all causes was 1.47 (1.15-1.88), from cardiovascular disease 1.63 (1.09-2.46), from respiratory disease 2.03 (1.06-3.90), and from neoplasms 2.26 (1.02-5.03) in the age group < 70 years, and 2.45 (1.23-4.90) in the age group > 79 years. On applying Cox models after multiple imputations by chained equations, we found that RF modified the effect of age. Employment status, comorbidity, and gender were independent baseline predictors of subsequent mortality. CONCLUSION: In this cohort, significant excess mortality was confined to RF-positive males. The effect of age was modified by RF, and employment status and comorbidity were independent predictors of mortality.

Effect of a treat-to-target strategy based on methotrexate and intra-articular betamethasone with or without additional cyclosporin on MRI-assessed synovitis, osteitis, tenosynovitis, bone erosion, and joint space narrowing in early rheumatoid arthritis: results from a 2-year randomized double-blind placebo-controlled trial (CIMESTRA).

OBJECTIVES: To investigate whether a treat-to-target strategy based on methotrexate (MTX) and intra-articular (IA) betamethasone suppresses magnetic resonance imaging (MRI)-determined measures of disease activity and reduces joint destruction in early rheumatoid arthritis (eRA) patients, and to investigate whether concomitant cyclosporin A (CyA) provides an additional effect. METHOD: In the 2-year randomized, double-blind, treat-to-target trial CIMESTRA, 160 patients with eRA (< 6 months) were randomized to MTX, intra-articular betamethasone and CyA, or placebo CyA. A total of 129 patients participated in the MRI substudy, and had contrast-enhanced MR images of the non-dominant hand at months 0, 6, 12, and 24. MR images were evaluated for osteitis, synovitis, tenosynovitis, bone erosion, and joint space narrowing (JSN), using validated scoring methods. RESULTS: Significant reductions were seen at 6 months in all inflammatory parameters [synovitis, mean change -1.6 (p < 0.001, Wilcoxon), tenosynovitis, -3.5 (p < 0.001), and osteitis, -1.3 (p < 0.05)] and at 12/24 months in synovitis and tenosynovitis [-1.6/-2.2 and -3.6/-3.8, respectively; all p < 0.001]. MRI signs of inflammation were not fully eliminated, and increases in erosion and JSN scores were observed at 6 months [0.4 (p < 0.01)/0.1 (p < 0.05)], 12 months [0.8 (p < 0.001)/0.3 (p < 0.01)], and 24 months [1.0 (p < 0.001)/0.4 (p < 0.001)]. Clinical measures decreased significantly (p < 0.001) at all time points. There were no consistent statistically significant differences between treatment groups. CONCLUSIONS: In this eRA treat-to-target trial, MTX and intra-articular glucocorticoids markedly reduced, but did not eliminate, MRI osteitis, synovitis, and tenosynovitis. Accordingly, minimal but statistically significant increases in bone erosion and JSN were observed. No additional effect of CyA was demonstrated.

Interactions between smoking, increased serum levels of anti-CCP antibodies, rheumatoid factors, and erosive joint disease in patients with early, untreated rheumatoid arthritis.

OBJECTIVES: To determine to what extent shared epitopes, smoking, and anti-cyclic citrullinated peptide (anti-CCP) antibodies are associated with disease activity and erosive disease in patients with rheumatoid arthritis (RA) at disease onset. METHOD: RA patients not previously treated with disease-modifying anti-rheumatic drugs (DMARDs) and with a disease duration of < 6 months (CIMESTRA study) were examined for shared epitopes, anti-CCP antibodies, immunoglobulin M rheumatoid factor (IgM-RF) and IgA-RF, radiographic erosive changes in hands and feet, and clinical disease activity. RESULTS: The study comprised 153 patients, of whom 104 (68%) were ever-smokers. The prevalence of patients with 0, 1, or 2 shared epitopes was 40 (48%), 71 (49%), and 33 (23%), respectively. Anti-CCP antibodies, IgM-RF, and IgA-RF were present in 89 (58%), 99 (65%), and 82 (54%) patients, respectively. Among smokers, erosive disease was associated with anti-CCP antibodies [odds ratio (OR) 3.9, 95% confidence interval (CI) 1.6-9.3], IgM-RF (OR 4.9, 95% CI 1.9-12), and IgA-RF (OR 2.8, 95% CI 1.2-6.4) but absent with regard to shared epitopes. Among never-smokers, erosive disease was not associated with either shared epitopes or antibodies. All antibody levels measured were associated with smoking and shared epitopes. CONCLUSIONS: Shared epitopes and smoking were associated with the production of anti-CCP antibodies and rheumatoid factors of IgM and IgA isotypes, which again were associated with erosive disease at presentation only in smokers. As shared epitopes and smoking were not directly associated with erosive disease, smoking may enhance the development of erosive disease in RA at different levels or through separate pathways.

Heritability assessment of cartilage metabolism. A twin study on circulating procollagen IIA N-terminal propeptide (PIIANP).

OBJECTIVE: The aim of this investigation was to estimate the heritability of circulating collagen IIA N-terminal propeptide (PIIANP) by studying mono- and dizygotic healthy twin pairs at different age and both genders. DESIGN: 598 monozygotic (MZ) and dizygotic (DZ) twin individuals aged 18-59 years were recruited from the Danish Twin Registry. PIIANP was measured by competitive ELISA. The similarity of circulating PIIANP among MZ and DZ twins was assessed by intraclass correlations according to traits. The heritability was estimated by variance component analysis accounting for additive and dominant genetic factors as well as shared and non-shared environment but ignoring epistasis (genetic inter-locus interaction) and gene-environment interaction. RESULTS: The intraclass correlation of PIIANP in MZ and DZ twins was 0.69 (0.60-0.76) and 0.46 (0.34-0.58) respectively indicating a significant genetic impact on PIIANP in serum. Additive genetic effects explained 45% (21-70%), shared environment 24% (7-53%) and non-shared environment 31% (24-39%) of the total variance. The heritability estimate did not differ across ages and between genders. CONCLUSIONS: The study shows that approximately 45% of the collagen IIA synthesis as assessed by the collagen IIA N-terminal propeptide in serum is attributable to genetic effectors while individual and shared environment account for 24% and 31% respectively. The heritability does not differ between genders or according to age.

Periarticular and generalised bone loss in patients with early rheumatoid arthritis: influence of alendronate and intra-articular glucocorticoid treatment. Post hoc analyses from the CIMESTRA trial.

OBJECTIVES: The aims of this study were to investigate the influence of alendronate and intra-articular betamethasone treatment on bone mineral density (BMD) changes in hand, lumbar spine and femoral neck during 1 year of a treat-to-target study (Cyclosporine, Methotrexate, Steroid in RA (CIMESTRA)). PATIENTS AND METHODS: A hundred and sixty patients with early, active rheumatoid arthritis (RA) received methotrexate, intra-articular betamethasone and ciclosporin /placebo-ciclosporin. Patients with Z-score ≤0 also started alendronate 10 mg/day. BMD of the hand (digital x-ray radiogrammetry (DXR-BMDhand)), BMD of lumbar spine and femoral neck (dual x-ray absorptiometry (DXA-BMDlumbar spine and DXA-BMDfemoral neck)) and x-rays of hands, wrists and forefeet (modified Sharp-van der Heijde score) were measured at baseline and 1 year, with complete data available in 107 patients. RESULTS: The change in BMD in hand, lumbar spine and femoral neck was negatively associated with the dose of intra-articular betamethasone (p<0.01 for all), but the bone loss in hand was modest and in the axial skeleton comparable with that of healthy individuals. Alendronate did not influence changes in DXR-BMDhand, which averaged -2.8%, whereas significant changes were observed in DXA-BMDlumbar spine and DXA-BMDfemoral neck in alendronate-treated patients (1.8% and 0.8%) compared with untreated patients (-1.8% and -2.2%) (p<0.01 and 0.02). Alendronate did not affect the radiographic progression (alendronate-treated patients: 0 (range 0-19), non-alendronate: 0 (0-18)). CONCLUSIONS: In early active RA, intra-articular betamethasone injections added to disease-modifying antirheumatic drug (DMARD) treatment led to minimal loss of hip and lumbar BMD, and the loss could be prevented by treatment with alendronate. Alendronate treatment did not affect radiographic progression.

MRI bone oedema is the strongest predictor of subsequent radiographic progression in early rheumatoid arthritis. Results from a 2-year randomised controlled trial (CIMESTRA).

OBJECTIVE: To identify predictors of radiographic progression in a 2-year randomised, double-blind, clinical study (CIMESTRA) of patients with early rheumatoid arthritis (RA). METHODS: Patients with early RA (n = 130) were treated with methotrexate, intra-articular betamethasone and ciclosporin/placebo-ciclosporin. Baseline magnetic resonance imaging (MRI) of the wrist (wrist-only group, n = 130) or MRI of wrist and metacarpophalangeal (MCP) joints (wrist+MCP group, n = 89) (OMERACT RAMRIS), x-ray examination of hands, wrists and forefeet (Sharp/van der Heijde Score (TSS)), Disease Activity Score (DAS28), anti-cyclic citrullinated peptide antibodies (anti-CCP), HLA-DRB1-shared epitope (SE) and smoking status were assessed. Multiple regression analysis was performed with delta-TSS (0-2 years) as dependent variable and baseline DAS28, TSS, MRI bone oedema score, MRI synovitis score, MRI erosion score, anti-CCP, smoking, SE, age and gender as explanatory variables. RESULTS: Baseline values: median DAS28 5.6 (range 2.4-8.0); anti-CCP positive 61%; radiographic erosions 56%. At 2 years: DAS28 2.0 (0.5-5.7), in DAS remission: 56%, radiographic progression 26% (wrist+MCP group, similar for wrist-only group). MRI bone oedema score was the only independent predictor of delta-TSS (wrist+MCP group: coefficient = 0.75 (95% CI 0.55 to 0.94), p<0.001; wrist-only group: coefficient = 0.59 (95% CI 0.40 to 0.77), p<0.001). Bone oedema score explained 41% of the variation in the progression of TSS (wrist+MCP group), 25% in wrist-only group (Pearson's r = 0.64 and r = 0.50, respectively). Results were confirmed by sensitivity analyses. CONCLUSION: In a randomised controlled trial aiming at remission in patients with early RA, baseline RAMRIS MRI bone oedema score of MCP and wrist joints (and of wrist only) was the strongest independent predictor of radiographic progression in hands, wrists and forefeet after 2 years. MRI synovitis score, MRI erosion score, DAS28, anti-CCP, SE, smoking, age and gender were not independent risk factors. TRIAL REGISTRATION NUMBER: NCT00209859.

Ankylosing spondylitis in Danish and Norwegian twins: occurrence and the relative importance of genetic vs. environmental effectors in disease causation.

OBJECTIVE: To estimate the influence of genetic effects in the aetiology and pathogenesis of ankylosing spondylitis (AS). METHODS: The study comprised one Norwegian and two Danish nationwide twin surveys. In 1994 and 2002, respectively, 37,388 and 46,331 Danish twin individuals were asked by questionnaire if they had AS. Similarly, in 1998, 12,718 Norwegian twins were asked if they had AS using a questionnaire phrased according to the Danish survey. Twins reporting AS were categorized according to the modified New York criteria. RESULTS: A total of 113 twin individuals reported AS, of whom 81 (72.3%) participated in validation of the diagnosis. After validation, 39 probands were diagnosed with AS. Subsequent invitation of co-twins resulted in 27 complete pairs. The point prevalence and the annual incidence of AS was 0.1% and 3/100,000 person-years (pyr) among the Danish twins. The positive predictive value of self-reported AS was 49.3%. Probandwise concordance rates on AS were (2/5) 40% in monozygotic (MZ) and (1/23) 4% in dizygotic (DZ) twins [difference 35% (95% CI 2.9-72.8), p = 0.26]. Heredity analysis including previously published and the present HLA-B27-positive twin pairs indicated that additive genetic effects account for 94% (95% CI 0.56-0.99) of the variance in the causation of AS. CONCLUSION: Self-reported AS needs careful validation. The occurrence of AS in a Danish twin population was 0.1% and accords well with previous studies on singletons in hospital settings. The present study adds to previous evidence of a major genetic effect in the pathogenesis of AS.

The occurrence of psoriatic arthritis in Denmark.

OBJECTIVE: To apply and compare different classification criteria on a representative nationwide sample of psoriatic arthritis (PsA) twins and to estimate the prevalence and incidence of PsA. METHODS: The study comprised three Danish nationwide twin cohorts. In 1994 37 388 Danish twin individuals and in 2002 46 418 twin individuals received a questionnaire, including questions on rheumatic diseases. Twins reporting PsA and their co-twins were classified according to the Moll and Wright and CASPAR (ClASsification criteria for Psoriatic ARthritis) criteria based on interview, clinical examination and scrutiny of medical records. RESULTS: 228 twin individuals reported PsA and 164 (72%) participated in clinical validation. By using the Moll and Wright and CASPAR criteria, 54 and 50 cases were diagnosed with PsA respectively. The positive predictive value of self-reported PsA was 31%. According to the Moll and Wright and CASPAR criteria the prevalence was 0.15% (95% CI: 0.13%, 0.22%) and 0.14% (95% CI: 0.11%, 0.19%) respectively. The annual incidence rate based on new self-reported cases in 2002 was 6/100 000 person-years (95% CI: 3/100 000 person-years, 11/100 000 person-years). CONCLUSIONS: The positive predictive value of self-reported PsA was 31%. The prevalence and incidence figures of PsA were equivalent to the previously reported occurrence in population- and hospital-based studies.

On the heritability of psoriatic arthritis. Disease concordance among monozygotic and dizygotic twins.

OBJECTIVE: A nationwide unselected twin population to estimate the relative importance of genetic and environmental effectors in the aetiopathogenesis of psoriatic arthritis (PsA). METHODS: The study comprised three Danish nationwide twin cohorts. In 1994 and 2002 a total of 37 388 and 46 418 Danish twin individuals respectively were asked by questionnaire if they had PsA. Twins reporting PsA were invited to participate in a clinical examination. Patients were classified according to the Moll and Wright and the CASPAR (ClASsification criteria for Psoriatic ARthritis) criteria. Heritability was estimated by probandwise concordance rates and variance component analysis. RESULTS: 228 twin individuals reported PsA. Following diagnostic validation in 164 (70%), 50 probands were diagnosed with PsA according to the Moll and Wright criteria. Five of their co-twins were either dead, had emigrated, or did not participate in the twin study and nine did not respond, resulting in 36 complete pairs. A total of one of 10 monozygotic pairs and one of 26 dizygotic pairs were concordant for PsA, yielding a 6.2% difference in proportions (95% CI: -11%, 37%). Five of 10 monozygotic pairs and four of 26 dizygotic pairs were concordant for psoriatic skin disease implying a 35% difference (95% CI: 2%, 60%, p<0.05). CONCLUSIONS: This first twin study on PsA confirms that genes are important in the causation of psoriatic skin disease. Despite the limited statistical power, the almost identical concordance rates for PsA in monozygotic and dizygotic twins stresses the importance of the continued search for non-genetic effectors in PsA.

Adaptation of the Bath measures on disease activity and function in ankylosing spondylitis into Danish.

OBJECTIVE: Adaptation of the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), the Bath Ankylosing Spondylitis Global Score (BASG), and the Bath Ankylosing Spondylitis Functional Index (BASFI) for defining disease status in ankylosing spondylitis (AS) and related diseases for use in Denmark. METHODS: The BASDAI, BASFI, and BASG were translated into Danish and subsequently field-tested among 17 AS patients for relevance, face, and content validity. Reliability and validity were assessed by administering the new measures and a comparator instrument to 113 AS patients on two occasions. RESULTS: Test-retest reliability was high (>0.90) and the random measurement error was within+/-2.0 for the BASG and within approximately+/-1.5 for BASDAI and BASFI, which is acceptable for most clinical settings. The measures have good internal consistency and are able to discriminate between functional impairment and disease activity according to the Nottingham Health Profile (NHP) and the Stanford Health Assessment Questionnaire (HAQ). CONCLUSION: Danish versions of the BASG, BASDAI, and BASFI are feasible for application in clinical trials and epidemiological studies on AS in Denmark.

[Incidence of juvenile-onset diabetes in Denmark. A prospective registration in the counties of Fyn, Ribe, Sønderjylland and Vejle]. / Incidensen af børnediabetes i Danmark. En prospektiv registrering i Fyns, Ribe, Sønderjyllands og Vejle amter.

To provide contemporary figures of the incidence of childhood onset diabetes mellitus in Denmark, a prospective routine registration system has been established. The present study covered the population of children aged 0-14 years in four Danish counties, observed during the calendar years 1989 through 1993 concerning new cases of the disease. A total of 201 cases (113 boys and 88 girls) were registered, corresponding with a sex- and age-adjusted incidence rate of 17.4 per 100,000 person-years. Based on a validation analysis the study material is considered virtually complete. Overall, the incidence among boys was slightly higher than among girls, but not statistically significantly so. The incidence rate increased with age for both boys and girls. The results confirm that the incidence of childhood onset diabetes in Denmark is relatively high. Compared with previous Danish studies there has been an increase of about 20-25% in the incidence during the last 20 years. Our study stresses the importance of a continuous registration of new cases of childhood diabetes in Denmark, thereby providing clues for further aetiological studies as well as providing reference values for the planning of the diabetes health care. Our results indicate that at least two independent ascertainment sources must be employed to ensure complete registration in studies that rely upon voluntary registration as the primary reporting source.

Increasing incidence of early onset type 1 (insulin-dependent) diabetes mellitus: a study of Danish male birth cohorts.

To provide information of the incidence trends of Type 1 (insulin-dependent) diabetes mellitus we performed a cohort study of a series of Danish male birth cohorts. All male livebirths in Denmark between 1 January 1949 and 31 December 1964, were investigated regarding the development of Type 1 diabetes during the first 20 years of life using the files of the Danish Conscript Board, supplemented by a search in the Danish National Registry of Deaths. Diagnosis was verified and clinical information obtained from medical records. The material is estimated to be more than 95% complete. A total of 1705 diabetic subjects were identified of whom 23 were not representative of idiopathic Type 1 diabetes. The cumulative rate of Type 1 diabetes development during the first 20 years of life increased from 2.37 to 2.90 per 1000 for the first eight and last eight birth cohorts, respectively. A log-linear analysis of the incidence revealed significantly increasing incidence rates, independent of age effects, with a maximum in the late 1970's. About 75% of the cases presented short duration of symptoms prior to diagnosis, and only very few cases were diagnosed incidentally. Cases were diagnosed relatively less frequently during summer months than during winter months.