Research ArticleArticle

Risk Factors for Axial Inflammatory Arthritis in Patients with Psoriatic Arthritis

VINOD CHANDRAN, DAVID C. TOLUSSO, RICHARD J. COOK and DAFNA D. GLADMAN
The Journal of Rheumatology April 2010, 37 (4) 809-815; DOI: https://doi.org/10.3899/jrheum.091059

Abstract

Objective. Axial involvement is an important manifestation of psoriatic arthritis (PsA). We aimed to identify risk factors associated with the presence of axial PsA (AxPsA) in patients with PsA.

Methods. Patients with AxPsA (bilateral sacroiliitis ≥ grade 2/unilateral sacroiliitis ≥ 3 and inflammatory neck/back pain or limited spinal mobility) at first clinic visit were identified from the University of Toronto PsA clinic database. Risk factors associated with the presence of AxPsA were determined. Subsequently, patients without AxPsA at first clinic visit were identified. Under a multistate framework, the proportion of patients with PsA who subsequently developed AxPsA was estimated robustly using marginal methods and a Markov model. Risk factors at baseline that were associated with future development of AxPsA were identified through multiplicative time-homogeneous Markov models.

Results. Our study included 206 patients. Fifty patients had AxPsA at first clinic visit. HLA-B*27, radiographic damage to peripheral joints, and elevated erythrocyte sedimentation rate (ESR) increased odds of having AxPsA, while family history of PsA decreased the odds. One hundred fifty-six patients did not have AxPsA at first clinic visit. On followup, 28 developed AxPsA, and 11 died. We estimated that after 10 years of followup, 15% would develop AxPsA. Nail dystrophy, number of radiographically damaged joints, periostitis, and elevated ESR increased the risk of developing AxPsA, while swollen joints decreased the risk.

Conclusion. These results suggest that severe peripheral arthritis and HLA-B*27 are risk factors for AxPsA.

Psoriatic arthritis (PsA) is defined as an inflammatory arthritis associated with psoriasis, usually seronegative for rheumatoid factor1. PsA affects peripheral and/or axial joints and inflammatory axial disease is an important manifestation1. Axial inflammation is characteristic of ankylosing spondylitis (AS), the prototype spondyloarthritis (SpA). AS is classified according to the modified New York criteria and is strongly associated with HLA-B*272. Although the inflammatory axial disease in PsA may be indistinguishable from AS, it may also differ from AS in several respects. These differences were originally described by McEwen, et al in 1971, subsequently in 1993 by Gladman, et al, and most recently in 1998 by Helliwell, et al3–5. HLA-B*27 is associated with PsA, but the association is not as strong as that with AS4,6. In cross-sectional studies, HLA-B*27 was shown to be associated with earlier age of psoriasis and arthritis onset, bilateral sacroiliitis, and male gender, but an association with axial PsA (AxPsA) was not always found7,8. These inconsistencies could be due to the fact that these studies may have included patients with classical AS and psoriasis, as well as those with AxPsA having the distinguishing features described first by McEwen, et al3.

We have noted that patients with PsA develop new-onset axial inflammatory arthritis on longitudinal followup9,10. We therefore aimed to identify factors associated with the presence of AxPsA at first clinic visit, to determine the proportion of patients newly developing AxPsA after clinic entry, and to identify risk factors at first clinic visit for future development of AxPsA in patients with peripheral PsA.

MATERIALS AND METHODS

We conducted our study at the University of Toronto PsA clinic. In this clinic, we assess patients every 6–12 months according to a standard protocol that includes history, assessment of comorbidities, pharmacotherapy, examination (including detailed assessment of peripheral joints, spine, and extraarticular features), laboratory evaluation including HLA-B*27 typing, radiographic evaluation of hands, feet, pelvis, cervical, thoracic and lumbar spine (every 2 years), and patient questionnaires on function, fatigue, and quality of life11. Data are tracked on a computerized database. The study is approved by the Research Ethics Board of the University Health Network, Toronto, Ontario, Canada.

For our study, we classified patients as having AxPsA when the following criteria were fulfilled: (1) psoriasis, (2) bilateral sacroiliitis ≥ grade 2 or unilateral sacroiliitis ≥ grade 3 (New York radiographic criteria), and (3) inflammatory back/neck pain (low back pain or neck pain and stiffness for more than 3 months that improves with exercise but is not relieved by rest), or limitation in spinal mobility (qualitative restriction of cervical mobility, Schober’s test < 5 cm, chest expansion < 5 cm)12.

From the database, we identified patients who fulfilled criteria for AxPsA at first clinic visit. Factors associated with the presence of AxPsA were determined by comparison with those patients who did not have AxPsA at first clinic visit. We then identified patients who did not have AxPsA at first clinic visit. The proportion of patients from this group satisfying the criteria for AxPsA at various timepoints following entry to clinic was determined. Risk factors present at clinic entry associated with the future development of AxPsA were then identified. The risk factors studied included sex, age, ethnicity, family history of psoriasis and PsA, duration of psoriasis and PsA, number of joints actively inflamed, swollen, clinically damaged and radiographically damaged, presence of dactylitis, psoriatic nail dystrophy, radiographic periostitis, enthesitis, calcaneal spurs, prior nonsteroidal antiinflammatory agent use, prior disease-modifying antirheumatic drug use, smoking, hypertension (defined as systolic blood pressure ≥ 140 or diastolic blood pressure ≥ 90 mm Hg or treatment with antihypertensive medications), erythrocyte sedimentation rate (ESR), and HLA-B*27 status. Actively inflamed joints were defined as joints with swelling, stress pain, and/or joint-line tenderness (out of a total of 66 swollen and 68 tender joints scored according to protocol)11,13. The number of swollen joints was assessed specifically. Clinical damage was defined as the presence of joints with fixed deformities, flail joints, fused joints, or joints that had undergone surgery, and the total number out of a maximum of 68 was counted11,13. Radiographic damage was determined according to the modified Steinbrocker method, and the number of joints with at least stage 2 changes (presence of erosions) out of a maximum of 42 joints was assessed14. Dactylitis was defined as diffuse swelling of a whole digit15. Enthesitis was defined as tenderness at tendon/ligament insertion of Achilles tendon and plantar fascia. Periostitis was defined as ill-defined ossification near joint margins (but excluding osteophyte formation) on plain radiographs of the hands or feet16.

Statistical analyses

We first created a dataset with complete covariate information. We compared patients with and without AxPsA at first clinic visit and identified factors associated with the presence of AxPsA using logistic regression. To determine the proportion of patients newly developing AxPsA after clinic entry, and to identify risk factors at first clinic visit for future development of AxPsA in patients with peripheral PsA, we analyzed data in the framework of a 3-state illness-death model (alive with PsA, alive with AxPsA, and dead), where individuals in state 0 (PsA) can make transitions to either state 1 (AxPsA) or 2 (death), while individuals in state 1 can only make transitions to state 2. Thus, the event “death” is a competing risk event for the event “alive with AxPsA” (Figure 1)17. The data on the subsequent occurrence of AxPsA in patients presenting with peripheral PsA are examples of interval-censored data since the precise time of occurrence of the event of interest (AxPsA) following clinic entry is unknown, and is only known to have occurred between 2 timepoints when assessments were done. We obtained robust prevalence estimates (Pepe local likelihood) of AxPsA with the associated risk of mortality based on a difference in distribution functions of entry times18,19. We employed the local likelihood method with a nearest neighbor’s bandwidth of 0.1, an Epanechnikov kernel, a locally constant approximation, and 400 grid points19. Ninety-five percent bootstrap CI were obtained by the nonparametric bootstrap method based on 500 bootstrap samples. We identified factors that affected the risk of transition between states through multiplicative time-homogeneous Markov models. We included univariate effects that were significant at the 0.25 level in a multivariate model. The model was then reduced by backward elimination until the remaining effects were significant at the 0.05 level. At that point, we tested all other effects for reentry to the model (no effects were added at this stage). We obtained estimates of regression coefficients from this final model.

Figure 1.
Figure 1.

Three-state “Illness–Death” model for onset of axial psoriatic arthritis.

RESULTS

The dataset created had 206 patients with complete covariate information. Of these patients, 50 had evidence of AxPsA at first clinic visit. The demographic and disease characteristics of these patients are given in Table 1. Table 2 shows the results of the logistic regression analyses on the factors associated with AxPsA at first clinic visit. HLA-B*27 (OR 5.75), the number of peripheral joints with radiographic damage (OR 1.12 for each damaged joint), and elevated ESR (1.02 for each 1 mm/h increase) were associated with increased odds of having AxPsA, while a family history of PsA (OR 0.1) decreased the odds.

View this table:
Table 1.

Demographic and disease characteristics of patients at study entry.

View this table:
Table 2.

Risk factors associated with the presence of AxPsA at first clinic visit.

We then determined the incidence and risk factors for future development of AxPsA in patients who did not satisfy criteria for AxPsA at first clinic visit. The demographics and disease characteristics of the 156 patients who did not have AxPsA at clinic entry are also given in Table 1. After a median followup duration of 6.4 (range 0.5 to 31) years, 28 patients were observed to develop AxPsA, and 11 died. Table 3 shows the baseline characteristics of the 28 patients who developed AxPsA on followup, as well as the remaining 128 who were not observed to develop AxPsA. Figure 2 depicts the Pepe local likelihood estimates and the 95% CI of the prevalence of AxPsA from clinic entry. We estimate that after 10 years of followup, 15% of patients with PsA who did not have AxPsA at first clinic visit are alive with AxPsA (95% CI 0.06–0.22). The apparent dip in the prevalence of AxPsA around 10 years of followup is due to a rise in mortality in the cohort around that time period.

Figure 2.
Figure 2.

Robust estimates and 95% bootstrap CI of remaining alive and having axial psoriatic arthritis (AxPsA) over 15 years from clinic entry. The area between the dotted lines represents the 95% CI of the prevalence of AxPsA from clinic entry. We estimated that after 10 years of followup, 15% of patients with PsA who did not have AxPsA at first clinic visit are alive with AxPsA (95% CI 0.06–0.22). The lower broken line depicts the proportion of patients dying after study entry. The apparent dip in the prevalence of AxPsA around 10 years of followup is due to a rise in mortality in the cohort around that time.

View this table:
Table 3.

Baseline characteristics of patients followed longitudinally (n = 156) who developed AxPsA (n = 28) and (n = 128) who did not develop AxPsA.

We then identified risk factors at baseline associated with future development of AxPsA. Table 4 shows the results of this analysis. Male sex, family history of PsA, nail dystrophy, number of swollen joints, enthesitis, number of radiographically damaged joints, periostitis, calcaneal spurs, elevated ESR, and presence of hypertension were associated with the development of AxPsA in univariate analyses (p < 0.25). Nail involvement [relative risk (RR) = 7.07], periostitis (RR = 5.67), number of radiographically damaged joints (RR = 1.11 for each damaged joint), and ESR (RR = 1.03 for each 1-mm increase in ESR) were associated with increased risk of AxPsA. Swollen joints reduced the risk (RR = 0.83 for each swollen joint). Interestingly, HLA-B*27 was not identified as a risk factor.

View this table:
Table 4.

Risk factors at first clinic visit associated with future development of AxPsA .

It is recognized that the clinical criteria (inflammatory back pain, restricted spinal mobility) have poor sensitivity and specificity20,21. We therefore repeated the analyses after redefining the criteria for AxPsA. We redefined the criteria basing it only on the New York radiographic criteria (AxPsA-R, i.e., psoriasis and bilateral sacroiliitis ≥ grade 2 or unilateral sacroiliitis ≥ grade 3)12. Male sex, radiographic joint damage, elevated ESR, and HLA-B*27 were associated with the presence of AxPsA-R at first clinic visit. In those without AxPsA-R at baseline, 23% were estimated to develop AxPsA-R after 10 years of followup. The risk factors at baseline that increased risk of future development of AxPsA-R were male sex, psoriatic nail dystrophy, periostitis, number of radiographically damaged joints, and elevated ESR (data not shown). Number of swollen joints no longer decreased the risk.

DISCUSSION

Axial inflammatory arthritis is an important manifestation of PsA. Most patients with AxPsA also have peripheral arthritis. Only 2% of patients have axial arthritis alone1,21. However, AxPsA has not been well defined. Patients with classical AS with psoriasis and those with axial arthritis with features that are different from classical AS have been included in this category3,5. The association of HLA-B*27 with AxPsA is not as strong as that with AS. That may be because the association is stronger in that subset of AxPsA that has features similar to AS. HLA-B*27 association is stronger in patients with younger age at onset who have disease features similar to classic AS (positive family history, bilateral sacroiliitis, isolated axial pattern, and enthesitis)22.

In our longitudinal observational cohort we have observed that patients have AxPsA at first assessment, or develop it on followup. We show that HLA-B*27, radiographic damage to peripheral joints, and elevated ESR increased odds of having AxPsA at first clinic visit, while family history of PsA decreased the odds. On followup, the presence of nail dystrophy, number of radiographically damaged joints, periostitis, and elevated ESR at baseline increased the risk of future development of AxPsA, while swollen joints decreased risk. Thus, severe peripheral arthritis in addition to HLA-B*27 is a risk factor for AxPsA. We also show that 15% of patients with peripheral PsA without axial involvement are expected to develop AxPsA over 10 years of followup. When the criteria for AxPsA were relaxed to include radiographic criteria alone (AxPsA-R), then male sex, number of radiographically damaged joints, and elevated ESR increased the odds of having AxPsA-R at first clinic visit and also increased the risk of future AxPsA-R in those patients who did not have AxPsA-R at first visit. Other risk factors showed a differential association; HLA-B*27 was associated with AxPsA only at first clinic visit, and psoriatic nail dystrophy and radiographic periostitis increased the risk of future development of AxPsA-R.

Our study is unique in many respects. Our study has been possible because we collected data relevant to assessment of peripheral and axial arthritis regularly and performed radiographs of the spine even in the absence of symptoms or signs of axial arthritis in all patients with PsA. We have thus been able to detect radiographic changes in the spine in the absence of symptoms or signs, since it is recognized that a significant proportion of patients with AxPsA have silent axial involvement23,24. The various clinical, laboratory, and radiographic risk factors are recorded systematically in a longitudinal fashion, and tracked on a computerized database. HLA typing was done on all patients. The outcome (presence of AxPsA) was defined using strict definitions similar to the modified New York criteria for AS, and statistical analyses were done using robust methods. Drawbacks include small numbers of patients in the group presenting with peripheral arthritis alone who subsequently developed AxPsA, reflected in the wide CI of the estimate of AxPsA prevalence on followup (Figure 2). Moreover, this cohort of patients already had average disease duration of 6.6 years prior to clinic entry, and thus it was not an inception cohort. An inception cohort with disease duration < 1 year before clinic entry would be the ideal cohort for such studies.

One could argue that the definition of AxPsA that we used for this study does not reflect all manifestations of spinal involvement in PsA5. However, there is currently no validated or widely accepted definition for AxPsA25. We chose a rather strict definition based on criteria for AS, but recognize that this may not be adequate. Moreover, we repeated the analyses using radiographic criteria alone as the outcome measure. The results obtained were similar, except that family history of PsA and swollen joint count were no longer “protective” in the 2 AxPsA groups, respectively, and male sex and raised ESR were identified as additional risk factors.

We demonstrate that severe peripheral arthritis as indicated by number of joints with radiographic damage and elevated ESR is associated with the presence of AxPsA at first clinic visit as well as future development of AxPsA. Thus, severe peripheral arthritis is a risk factor for AxPsA. With larger numbers and advanced statistical methods, these results robustly confirm reports of the association between peripheral arthritis and AxPsA26,27. The process of damage in the peripheral joints, particularly ankylosis, may reflect the same pathological process seen in the sacroiliac joints and syndesmophytes.

In patients presenting with peripheral PsA, in addition to the number of radiographically damaged joints and increased ESR, we identified psoriatic nail dystrophy and periostitis as risk factors associated with future AxPsA. Enthesitis has been proposed as a mechanism underlying psoriatic nail dystrophy28. Periostitis is a specific manifestation of PsA and indicates new bone formation16,29. Thus, these 2 markers together could indicate the presence of pathogenetic events that may occur in the spine leading to axial arthritis, since ankylosis in the spine is a reflection of both entheseal inflammation and new bone formation. These risk factors, thus, probably have biologic relevance. The presence of swollen joints was associated with reduced risk. Because swollen joints predict future peripheral joint damage, one would expect swollen joints to be a risk factor for AxPsA if damaged joints are, since swollen joints predict future peripheral joint damage30,31. It is possible that there is an inverse relationship between the degree and severity of inflammation in the peripheral joints and the spine. However, when we defined AxPsA based on radiographic criteria alone, swollen joints were no longer associated, and therefore this association may be due to problems in defining AxPsA.

Radiographic damage, nail dystrophy, and periostitis might indicate severity of the disease instead of being markers of pathogenic events. The more severe the disease, the more structures are involved. Thus, our data may also be interpreted as follows: patients with peripheral arthritis as well as axial arthritis, periostitis, and nail dystrophy have the complete disease spectrum, and the rest have incomplete PsA in an atypical and usually incomplete manifestation.

HLA-B*27 is associated with the presence of AxPsA at first clinic visit. This reflects shared genetic links between AxPsA and AS32. The association of HLA-B*27 with AxPsA is controversial. In SpA, HLA-B*27 is associated with the presence of spondylitis33. In cross-sectional studies in PsA, HLA-B*27 was shown to be associated with earlier age of psoriasis and arthritis onset, bilateral sacroiliitis, and male gender, but this association with AxPsA in PsA was not always found7,8. We have shown that HLA-B*27 is associated with AxPsA defined on the basis of presence of ≥ grade 2 sacroiliitis alone, or grade 1 sacroiliitis accompanied by presence of syndesmophytes and/or inflammatory back pain at first clinic visit6. In our study, we have demonstrated that HLA-B*27 is associated with AxPsA at first clinic visit but not with AxPsA that develops on followup. In addition to being associated with axial arthritis, HLA-B*27 may be associated with the severity of peripheral arthritis, since patients with AxPsA at first clinic visit also had severe peripheral arthritis (Table 1). We have shown that, when HLA-DR*07 was present, HLA-B*27 is associated with disease progression in PsA34. HLA-B*27 may be an important factor that helps distinguish AS from AxPsA.

There are few prospective longterm studies on SpA evaluating future development of definite spinal involvement (radiographic sacroiliitis and or fulfillment of diagnostic criteria for AS). Studies have been done on juvenile idiopathic arthritis (JIA) and undifferentiated SpA (USpA)35–40. In JIA, radiographically evident sacroiliitis developed in 6% of patients after median disease duration of 14.9 years35. HLA-B*27, absence of HLA-DPB1*02, late onset of disease, hip arthritis within the first 6 months, and elevated ESR for at least 6 months were found to be associated with the development of sacroiliitis35,36. In USpA, buttock pain and low-grade sacroiliitis were associated with future development of AS, and acute-phase reactant levels, juvenile onset, and HLA-B*27 showed a trend37,40. Our study shows similar results in patients with PsA.

Thus, we identified risk factors for AxPsA. Our results will need to be validated by studies in inception cohorts of patients with PsA.

Acknowledgments

The authors acknowledge the contribution of Catherine T. Schentag to this study.

Footnotes

  • V. Chandran is supported by a Canadian Institutes of Health Research (CIHR) Clinical Research Initiative Fellowship. R. Cook is a Canada Research Chair in Statistical Methods for Health Research and is supported by the CIHR. The University of Toronto PsA clinic is supported by the Krembil Foundation.

    • Accepted for publication November 20, 2009.

REFERENCES

  1. 1.↵
    1. Harris ED,
    2. Budd RC,
    3. Firestein GS,
    4. et al.
    1. Gladman DD
    . Psoriatic arthritis. In: Harris ED, Budd RC, Firestein GS, et al. editors. Kelley’s textbook of rheumatology. 7th ed. Philadelphia: W.B. Saunders Co.; 2005:1155–64.
  2. 2.↵
    1. Mease PJ
    . Spondyloarthritis update: new insights regarding classification, pathophysiology, and management. Bull NYU Hosp Jt Dis 2008;66:203–9.
    OpenUrlPubMed
  3. 3.↵
    1. McEwen C,
    2. DiTata D,
    3. Lingg C,
    4. Porini A,
    5. Good A,
    6. Rankin T
    . Ankylosing spondylitis and spondylitis accompanying ulcerative colitis, regional enteritis, psoriasis and Reiter’s disease. A comparative study. Arthritis Rheum 1971;14:291–318.
    OpenUrlPubMed
  4. 4.↵
    1. Gladman DD,
    2. Brubacher B,
    3. Buskila D,
    4. Langevitz P,
    5. Farewell VT
    . Differences in the expression of spondyloarthropathy: a comparison between ankylosing spondylitis and psoriatic arthritis. Clin Invest Med 1993;16:1–7.
    OpenUrlPubMed
  5. 5.↵
    1. Helliwell PS,
    2. Hickling P,
    3. Wright V
    . Do the radiological changes of classic ankylosing spondylitis differ from the changes found in the spondylitis associated with inflammatory bowel disease, psoriasis, and reactive arthritis? Ann Rheum Dis 1998;57:135–40.
    OpenUrlAbstract/FREE Full Text
  6. 6.↵
    1. Gladman DD,
    2. Anhorn KA,
    3. Schachter RK,
    4. Mervart H
    . HLA antigens in psoriatic arthritis. J Rheumatol 1986;13:586–92.
    OpenUrlPubMed
  7. 7.↵
    1. Queiro R,
    2. Sarasqueta C,
    3. Belzunegui J,
    4. Gonzalez C,
    5. Figueroa M,
    6. Torre-Alonso JC
    . Psoriatic spondyloarthropathy: a comparative study between HLA-B27 positive and HLA-B27 negative disease. Semin Arthritis Rheum 2002;31:413–8.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Mazzanti G,
    2. Coloni L,
    3. DeSabbata G,
    4. Paladini G
    . Is HLA B27 a true marker of axial involvement in psoriatic arthropathy? Acta Derm Venereol Suppl 1994;186:71–2.
    OpenUrlPubMed
  9. 9.↵
    1. Hanly JG,
    2. Russell ML,
    3. Gladman DD
    . Psoriatic spondyloarthropathy: a long term prospective study. Ann Rheum Dis 1988;47:386–93.
    OpenUrlAbstract/FREE Full Text
  10. 10.↵
    1. Gladman DD,
    2. Brubacher B,
    3. Buskila D,
    4. Langevitz P,
    5. Farewell VT
    . Psoriatic spondyloarthropathy in men and women: a clinical, radiographic, and HLA study. Clin Invest Med 1992;15:371–5.
    OpenUrlPubMed
  11. 11.↵
    1. Gladman DD,
    2. Shuckett R,
    3. Russell ML,
    4. Thorne JC,
    5. Schachter RK
    . Psoriatic arthritis (PSA) — an analysis of 220 patients. Q J Med 1987;62:127–41.
    OpenUrlAbstract/FREE Full Text
  12. 12.↵
    1. van der Linden S,
    2. Valkenburg HA,
    3. Cats A
    . Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum 1984;27:361–8.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Gladman DD,
    2. Cook RJ,
    3. Schentag C,
    4. Feletar M,
    5. Inman RD,
    6. Hitchon C,
    7. et al.
    The clinical assessment of patients with psoriatic arthritis: results of a reliability study of the Spondyloarthritis Research Consortium of Canada. J Rheumatol 2004;31:1126–31.
    OpenUrlAbstract/FREE Full Text
  14. 14.↵
    1. Rahman P,
    2. Gladman DD,
    3. Cook RJ,
    4. Zhou Y,
    5. Young G,
    6. Salonen D
    . Radiological assessment in psoriatic arthritis. Br J Rheumatol 1998;37:760–5.
    OpenUrlAbstract/FREE Full Text
  15. 15.↵
    1. Brockbank JE,
    2. Stein M,
    3. Schentag CT,
    4. Gladman DD
    . Dactylitis in psoriatic arthritis: a marker for disease severity? Ann Rheum Dis 2005;64:188–90.
    OpenUrlAbstract/FREE Full Text
  16. 16.↵
    1. Taylor WJ,
    2. Porter GG,
    3. Helliwell PS
    . Operational definitions and observer reliability of the plain radiographic features of psoriatic arthritis. J Rheumatol 2003;30:2645–58.
    OpenUrlAbstract/FREE Full Text
  17. 17.↵
    1. Satagopan JM,
    2. Ben-Porat L,
    3. Berwick M,
    4. Robson M,
    5. Kutler D,
    6. Auerbach AD
    . A note on competing risks in survival data analysis. Br J Cancer 2004;91:1229–35.
    OpenUrlCrossRefPubMed
  18. 18.↵
    1. Pepe MS,
    2. Longton G,
    3. Thornquist M
    . A qualifier Q for the survival function to describe the prevalence of a transient condition. Stat Med 1991;10:413–21.
    OpenUrlPubMed
  19. 19.↵
    1. Betensky RA,
    2. Lindsey JC,
    3. Ryan LM,
    4. Wand MP
    . Local EM estimation of the hazard function for interval-censored data. Biometrics 1999;55:238–45.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. Rudwaleit M,
    2. Metter A,
    3. Listing J,
    4. Sieper J,
    5. Braun J
    . Inflammatory back pain in ankylosing spondylitis: a reassessment of the clinical history for application as classification and diagnostic criteria. Arthritis Rheum 2006;54:569–78.
    OpenUrlCrossRefPubMed
  21. 21.↵
    1. Taylor WJ,
    2. Zmierczak HG,
    3. Helliwell PS
    . Problems with the definition of axial and peripheral disease patterns in psoriatic arthritis. J Rheumatol 2005;32:974–7.
    OpenUrlFREE Full Text
  22. 22.↵
    1. Queiro R,
    2. Alperi M,
    3. Lopez A,
    4. Sarasqueta C,
    5. Riestra JL,
    6. Ballina J
    . Clinical expression, but not disease outcome, may vary according to age at disease onset in psoriatic spondylitis. Joint Bone Spine 2008;75:544–7.
    OpenUrlCrossRefPubMed
  23. 23.↵
    1. Queiro R,
    2. Belzunegui J,
    3. González C,
    4. De DJ,
    5. Sarasqueta C,
    6. Torre JC,
    7. et al.
    Clinically asymptomatic axial disease in psoriatic spondyloarthropathy. A retrospective study. Clin Rheumatol 2002;21:10–3.
    OpenUrlCrossRefPubMed
  24. 24.↵
    1. Gladman DD
    . Psoriatic arthritis. Rheum Dis Clin North Am 1998;24:829–44.
    OpenUrlCrossRefPubMed
  25. 25.↵
    1. Gladman DD
    . Axial disease in psoriatic arthritis. Curr Rheumatol Rep 2007;9:455–60.
    OpenUrlCrossRefPubMed
  26. 26.↵
    1. Marsal S,
    2. Armadans-Gil L,
    3. Martínez M,
    4. Gallardo D,
    5. Ribera A,
    6. Lience E
    . Clinical, radiographic and HLA associations as markers for different patterns of psoriatic arthritis. Rheumatology 1999;38:332–7.
    OpenUrlAbstract/FREE Full Text
  27. 27.↵
    1. Suarez-Almazor ME,
    2. Russell AS
    . Sacroiliitis in psoriasis: relationship to peripheral arthritis and HLA-B27. J Rheumatol 1990;17:804–8.
    OpenUrlPubMed
  28. 28.↵
    1. Tan AL,
    2. Benjamin M,
    3. Toumi H,
    4. Grainger AJ,
    5. Tanner SF,
    6. Emery P,
    7. et al.
    The relationship between the extensor tendon enthesis and the nail in distal interphalangeal joint disease in psoriatic arthritis — a high-resolution MRI and histological study. Rheumatology 2007;46:253–6.
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    1. Taylor W,
    2. Gladman D,
    3. Helliwell P,
    4. Marchesoni A,
    5. Mease P,
    6. Mielants H,
    7. CASPAR Study Group
    . Classification criteria for psoriatic arthritis: development of new criteria from a large international study. Arthritis Rheum 2006;54:2665–73.
    OpenUrlCrossRefPubMed
  30. 30.↵
    1. Gladman DD,
    2. Farewell VT,
    3. Nadeau C
    . Clinical indicators of progression in psoriatic arthritis: multivariate relative risk model. J Rheumatol 1995;22:675–9.
    OpenUrlPubMed
  31. 31.↵
    1. Bond SJ,
    2. Farewell VT,
    3. Schentag CT,
    4. Gladman DD
    . Predictors for radiological damage in psoriatic arthritis: results from a single centre. Ann Rheum Dis 2007;66:370–6.
    OpenUrlAbstract/FREE Full Text
  32. 32.↵
    1. Sieper J,
    2. Braun J,
    3. Rudwaleit M,
    4. Boonen A,
    5. Zink A
    . Ankylosing spondylitis: an overview. Ann Rheum Dis 2002;61 Suppl 3:iii8–18.
    OpenUrlAbstract/FREE Full Text
  33. 33.↵
    1. Nash P,
    2. Mease PJ,
    3. Braun J,
    4. van der Heijde D
    . Seronegative spondyloarthropathies: to lump or split? Ann Rheum Dis 2005;64 Suppl 2:ii9–13.
    OpenUrlAbstract/FREE Full Text
  34. 34.↵
    1. Gladman DD,
    2. Farewell VT
    . The role of HLA antigens as indicators of disease progression in psoriatic arthritis. Multivariate relative risk model. Arthritis Rheum 1995;38:845–50.
    OpenUrlCrossRefPubMed
  35. 35.↵
    1. Flatø B,
    2. Smerdel A,
    3. Johnston V,
    4. Lien G,
    5. Dale K,
    6. Vinje O,
    7. et al.
    The influence of patient characteristics, disease variables, and HLA alleles on the development of radiographically evident sacroiliitis in juvenile idiopathic arthritis. Arthritis Rheum 2002;46:986–94.
    OpenUrlCrossRefPubMed
  36. 36.↵
    1. Flatø B,
    2. Hoffmann-Vold AM,
    3. Reiff A,
    4. Førre Ø,
    5. Lien G,
    6. Vinje O
    . Long-term outcome and prognostic factors in enthesitis-related arthritis: a case-control study. Arthritis Rheum 2006;54:3573–82.
    OpenUrlCrossRefPubMed
  37. 37.↵
    1. Sampaio-Barros PD,
    2. Bertolo MB,
    3. Kraemer MH,
    4. Marques-Neto JF,
    5. Samara AM
    . Undifferentiated spondyloarthropathies: a 2-year follow-up study. Clin Rheumatol 2001;20:201–6.
    OpenUrlCrossRefPubMed
  38. 38.↵
    1. Kumar A,
    2. Bansal M,
    3. Srivastava DN,
    4. Pandhi A,
    5. Menon A,
    6. Mehra NK,
    7. et al.
    Long-term outcome of undifferentiated spondylarthropathy. Rheumatol Int 2001;20:221–4.
    OpenUrlCrossRefPubMed
  39. 39.↵
    1. Jäntti JK,
    2. Kaarela K,
    3. Lehtinen KE
    . Seronegative oligoarthritis: a 23-year follow-up study. Clin Rheumatol 2002;21:353–6.
    OpenUrlCrossRefPubMed
  40. 40.↵
    1. Huerta-Sil G,
    2. Casasola-Vargas JC,
    3. Londoño JD,
    4. Rivas-Ruíz R,
    5. Chávez J,
    6. Pacheco-Tena C,
    7. et al.
    Low grade radiographic sacroiliitis as prognostic factor in patients with undifferentiated spondyloarthritis fulfilling diagnostic criteria for ankylosing spondylitis throughout follow up. Ann Rheum Dis 2006;65:642–6.
    OpenUrlAbstract/FREE Full Text
View Abstract
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

 Request Permissions

Bookmark this article

Jump to section