Article Text
PDF
Abstract
Objective The authors aimed to assess gender-related differences in severity of psoriatic arthritis (PsA) as reflected by measures of disease activity, joint damage, quality of life and disability.
Methods A cross-sectional analysis was performed among patients who have been followed in a large PsA clinic. Demographic, clinical and radiographic data as well as information about quality of life and function were retrieved from the clinic database. Radiographic damage was assessed according to modified Steinbrocker score (mSS). The association between gender and the following outcome variables, radiographic joint damage, axial involvement and measures of quality of life and function, was assessed by multivariate regression analysis after adjustment for potential confounders.
Results Three hundred and forty-five men and 245 women were included in the study. Axial involvement was more frequent in men (42.9% vs 31%, p=0.003). In multivariate analysis, adjusting for potential confounders, men were more likely to develop axial involvement (OR 1.8, p=0.003). Men were also more likely to develop more severe radiographic damage in the peripheral joints as evident by mSS. Men were more likely to be in a higher mSS damage category compared with women after adjusting for potential confounders in multivariate analysis (OR 1.6, p=0.007). Women suffered from more severe limitations in function and worse quality of life compared with men based on several patients' reported outcomes.
Conclusions Men with PsA are more likely to develop axial involvement and radiographic joint damage, while women are more likely to report about limitation in function and impaired quality of life.
Statistics from Altmetric.com
Seronegative spondyloarthritis (SpA) describes a group of diseases that share certain genetic, clinical and radiographic features. Ankylosing spondylitis (AS), the prototype of the SpA with a predominant spinal involvement, is a male predominant disease,1 while psoriatic arthritis (PsA) is generally considered a disease that affects men and women equally.2,–,6 However, a few studies suggested that PsA is more prevalent in male subjects7,–,9 while others have shown a female predominance.10 ,11
There is limited information about gender difference in PsA. Gladman et al reported that men with PsA have more severe axial involvement as documented by restriction of back movements and more common sacroiliac joint ankylosis than women.12 Similarly, Queiro et al reported that men were more likely to develop isolated axial disease and were more likely to be HLA-B*27 carriers while women had higher swollen joint count and higher functional disability scores.13 These studies were limited by their small sample size and by their inclusion of only patients with radiographic sacroiliitis.
In the present study, we aimed at assessing gender related differences in severity of PsA as reflected by measures of disease activity, joint damage, quality of life and disability in a large well phenotyped cohort.
Methods
Patients and setting
For this cross-sectional analysis, patients were selected from the University of Toronto PsA cohort. The cohort was established in 1978 as part of an ongoing prospective study. Approximately 98% of the patients in the cohort satisfy the Classification of Psoriatic Arthritis (CASPAR) criteria.14 The patients are followed in the clinic according to a standardised protocol every 6–12 months.15 Since questionnaires on quality of life and function have been collected routinely only since 2006, patients were selected for inclusion in this study only if they have visited the clinic as of 1 January 2006. Since we aimed to assess gender-related differences among patients with established disease, information from the last clinic visit was selected for analysis.
The study was approved by the University Health Network Research Ethics Board and all patients gave their informed consent.
Data collection
Clinical information
All study patients underwent assessment by a rheumatologist which included a complete history, physical examination and laboratory evaluation at 6–12-month intervals and radiological assessment at 2-year intervals. The last included radiographs of hands, feet, spine and sacroiliac joints. The following information was collected: demographic data, smoking status, alcohol consumption, comorbid medical conditions and medications. Disease related information included: age at diagnosis of psoriasis and PsA, family history of psoriasis and SpA in a first-degree relative and the presence of an inflammatory back pain. Past and current medication history included the use of disease modifying antirheumatic drugs (DMARDs) and anti-tumour necrosis factor (TNF)α agents.
Additional information was collected on physical examination: tender and swollen joint count, number of dactylitic digits and measurements of spinal mobility. An ‘actively inflamed’ joint was defined by the presence of stress pain and/or effusion in 64 joints. Measurements of spinal mobility included: forward flexion (modified Schober's test), lateral flexion (Domjan method),16 chest expansion, cervical rotation and occiput-to-wall distance. Psoriasis activity was determined by the PASI score17 and the modified Nail Psoriasis Severity Index.18 Additional laboratory information included human leucocyte antigen (HLA)-B*27 and HLA-C*06 status.
Radiographic damage
Radiographic damage in 42 joints was assessed according to modified Steinbrocker score (mSS).19 Radiographic lumbar and cervical spine damage was scored according to the modified Stokes Ankylosing Spondylitis Spine Score (mSASSS).20 Sacroiliitis was defined according to the New York criteria.21 The presence of axial involvement was defined by radiological evidence of either bilateral at least grade 2 sacroiliitis or unilateral grade 3 or 4 sacroiliitis.
Functional status
The study participants were asked to complete the following questionnaires: Bath Ankylosing Spondylitis Functional Index (BASFI), Health Assessment Questionnaire (HAQ), The Medical Outcome Study Short Form Health Survey (SF-36), Disease Life Quality Index (DLQI) and Fatigue Severity Score (FSS). The BASFI is a 10-item scale on which respondents rate the degree of difficulty they have in performing certain tasks, using visual analogue scales from 0 (easy) to 100 (impossible). The mean of the 10 responses is the BASFI score.22 The BASFI has been validated in patients with PsA.23 The HAQ is a 20-item scale which measures function. The HAQ asks questions about pain and the ability to perform activities of daily living. The derived score ranges from 0 to 3 with higher numbers reflecting greater disability. The HAQ has been validated for patients with PsA and has been shown to be related to disease activity.24 ,25 The SF-36 is a generic quality of life questionnaire26 consisting of eight sections: physical functioning, pain, vitality, social functioning, mental health, general health perception, role limitations due to physical problems and role limitation due to personal or emotional problems. It has been validated for PsA.27 The SF-36 may be collapsed into two components, the physical component score (PCS) and the mental component score (MCS). Lower scores in the SF-36 reflect decreased quality of life. The Dermatology Life Quality Index (DLQI)28 has been validated for psoriasis.29 Higher scores reflect a decreased quality of life. The FSS is a nine-item scale assessing the severity of fatigue during the past week. The FSS has been validated in PsA with higher scores reflecting increased fatigue.30
Statistical analysis
Baseline descriptive statistics were computed with continuous variables summarised by their means and SD and categorical variables summarised by their proportions. For the unadjusted comparison across the genders, t test was used to compare continuous variables that were normally distributed while Wilcoxon's rank sum test was used for continuous measures that were not normally distributed and χ2 test was used for categorical variables. The distribution of the radiographic damage score by mSS was heavily skewed to the left with a significant proportion of the observations scored 0. Therefore, these data were categorised into the following ordered categories: 0, 1–10, 11–30 and >30. The association of gender and outcome variables was calculated for each variable using regression analysis. The following outcome variables were assessed: the presence of axial involvement, mSS category and patients reported outcomes (HAQ, SF-36, DLQI, BASFI and FSS). Multivariate modelling of the association of gender with the above-mentioned outcomes included the duration of PsA as a confounding variable. Additional variables were then added to the model including: age, smoking, BMI, current use of DMARDs and anti-TNFα. The effect of gender was considered as statistically significant if the p value from the two-sided Wald test was <0.05.
Results
Gender differences in clinical variables
Out of the 636 patients who visited the PsA clinic between 2006 and 2011, a total of 345 men and 245 women were included in the study. Forty-six patients were excluded due to missing radiographic data. There were no differences across the genders in the age at onset of PsA (table 1); however, women had earlier onset of psoriasis that may be explained by the higher frequency of HLA-C*06 allele in women. This allele has been associated with early onset psoriasis.31 There was no difference across the genders in the frequency of HLA-B*27. Women were more likely to report positive family history of psoriasis and PsA. There were no significant differences in most clinical variables including: active and swollen joint counts, the presence of dactylitis, inflammatory spinal pain and PASI score. However, men were more likely to have psoriatic nail lesions. There were differences across the genders in the pattern of arthritis at the time of the diagnosis: women tended to present with polyarticular involvement while the most common pattern in men was oligoarthritis. Furthermore, women were more likely to use DMARDs compared with men; there was no difference in the usage of anti-TNFα agents. In the metrology indices, men had more limitation in cervical spine movements (abnormal occiput to wall distance 25.2% vs 12%, p<0.001) while abnormal chest wall expansion was more prevalent in females (35.1% vs 25.1%, p=0.01).
Comparison of clinical features between male and female subjects with PsA
Radiographic damage
Men suffered from more severe radiographic damage compared with women in both axial and peripheral joints (table 2). In agreement with previous observations in PsA, axial involvement was more frequent in men (42.9% vs 31%, p=0.003). After adjusting for duration of PsA, men were 1.8 times more likely to develop axial involvement (table 3). In multivariate analysis adjusting for potential confounders, men were still more likely to develop axial involvement (OR 1.8, p=0.003).
Comparison of radiographic features between male and female subjects with PsA
The associations of male gender with axial involvement compared with female gender using univariate and multivariate logistic regression analysis
Furthermore, men were more likely to develop sacroiliitis grade 3 or 4 and syndesmophytes in the cervical, thoracic and lumbar spine. In a subset of patients for whom radiographic axial damage was scored using mSASSS, men had higher mSASSS compared with women. Men were also more likely to develop more severe radiographic damage in the peripheral joints as evident by mSS. Women were more likely to have non-erosive disease (mSS=0). Although only trending for statistical significance, the unadjusted median of mSS in men was numerically higher than for women (8 vs 6, p=0.08). Men were more likely to be in a higher mSS damage category as compared with women after adjusting for duration of PsA (OR 1.6, p=0.005). This association between male gender and a higher mSS category remained statistically significant in multivariate analysis after adjusting for other potential confounders (table 4).
The associations of male gender with modified Steinbrocker score compared with female gender using univariate and multivariate logistic regression analysis*
Function and quality of life
Women suffered from more severe limitations in function and worse quality of life compared with men based on several patients' reported outcomes (table 5). Women scored higher in HAQ and BASFI and lower in the SF-36-PCS suggesting that women experience more limitations in daily physical function, although there was no difference in SF-36-MCS that scores mental and social function. Furthermore, women scored higher in the FSS, suggesting more severe fatigue in women. However, no difference was noted in DLQI which emphasises the effect of skin disease on quality of life. These differences remained statistically significant in the multivariate analyses (table 6), suggesting that women suffer from more limitation in function and worse quality of life despite the fact that men suffered from worse radiographic damage in their axial and peripheral joints. These findings suggest that quality of life and function scores do not necessarily reflect previous joint damage.
Comparison of function and quality of life measures between male and female subjects with PsA
The associations of male gender with functional and quality of life scores compared with female gender using univariate and multivariate linear regression analysis*
Discussion
Gender medicine is a new science focusing on the differences between men and women in both health and disease. Over the past decade, gender has been shown to affect clinical presentation, natural history and response to medications in several rheumatic diseases.32,–,34 We have shown that men with PsA develop more severe radiographic axial and peripheral joint damage while women suffer from more severe disability and impaired quality of life.
Only two studies thus far have specifically aimed to compare gender differences in PsA; one of them was from our group.12 ,13 These studies were limited in several aspects: first by their inclusion criteria that were limited to patients with axial involvement while our study included patients who satisfied the CASPAR criteria; therefore, a wider spectrum of the disease was included. Second, previous studies included 194 and 100 PsA patients while our study consists of a much larger sample of patients thus increasing the precision of the results. Last, to the best of our knowledge the present study was the first to assess the impact of gender on radiographic peripheral joint damage and on several patient-reported outcomes related to function and quality of life among PsA patients. Despite its strengths, the present study was limited in several aspects. Its cross-sectional nature precluded a firm conclusion regarding causal associations, and as in any observational study we cannot rule out the possibility that some of the associations were confounded by unmeasured variables. Although this is a specialty clinic and the generalisability of the results may be questioned, it should be noted that the clinic includes a wide spectrum of patients from very mild to very severe disease.
In the present study, men were more likely to have radiographic axial involvement, developed more severe radiographic sacroiliitis and syndesmophytes and had higher mSASSS. Previous studies in AS34,–,36 and PsA12 ,13 support our finding of more severe axial disease in men, irrespective of HLA-B*27 status. Most studies in AS indicate that women tend to have less lumbar and thoracic spinal radiographic damage. Moreover, the development of extensive ankylosis is significantly more common in men (14%–34%) compared with women (7%–12%).36 ,37 Similarly, our study found that men also suffered from more severe radiographic damage to the peripheral joints and were more likely to develop erosive disease compared with women. Since the mSS does not score erosions separately from joint space narrowing we are unable to disentangle the gender differences of these two components of mSS. No previous study has assessed gender differences in radiographic damage to peripheral joints in PsA; however, in rheumatoid arthritis (RA), several studies have noted worse radiographic joint damage in women.38 ,39 PsA and RA clearly have different underlying pathogenic mechanisms with regard to joint damage.
The underlying causes of this discrepancy in radiographic damage are unknown. Sex hormones may play a role, but environmental factors may also be important. Trauma has been suggested as a risk factor for PsA40,–,42 and engagement in physically demanding occupational activities is associated with more severe radiographic damage in AS patients.43 The more frequent involvement of men in heavy labour may explain these differences; however, prospective information about occupation and physical activity that might have affected joint damage was not available for all patients in this study. Another potential mechanism that may explain the difference in radiographic damage may be the different treatment regimens across the genders, and while there was no difference in the usage of anti-TNFα agents, women were more likely use DMARDs compared with men. This difference may reflect the higher tender joint counts and polyarticular pattern of arthritis in women which might have affected their physicians' decision to prescribe DMARDs which in turn may have led to less radiographic damage progression. However, there is no evidence that any of the traditional DMARDs actually prevent joint damage progression in PsA.
Although men suffered from more severe radiographic damage, women were more disabled in their daily function and suffered from more severe fatigue and impaired quality of life. This finding is not unique to PsA and has also been demonstrated in RA and AS. Lee et al reported that women with AS had higher functional limitation scores for each radiographic level of axial damage.34 Women with PsA are more likely to suffer from functional limitations as measured by higher HAQ score after adjusting for age and disease related variables.44 Fatigue scores are also higher among female patients with PsA and are related to physical limitations, mental dysfunction and fibromyalgia but not to disease activity.30 A potential explanation for this observation is the more pronounced pain perception in women than in men45 which can lead to limitation in function and poorer quality of life that do not reflect inflammation or damage. Furthermore, the impact of pain on disability indices may be less pronounced in men due to their greater muscle strength, making it easier to perform daily tasks. Muscle strength has been shown to significantly affect the HAQ.46 Last, it has been reported that men with RA tend to underscore their disability, as they overestimated their functional activity by 0.21 HAQ units.47
In summary, there are several gender related differences in clinical expression, joint damage and function in patients with PsA. Men are more likely to develop axial involvement and radiographic joint damage, while women are more likely to report about limitation in function and impaired quality of life. Future studies are required to assess the progression of these changes over the course of the disease and the underlying mechanisms that lead to these differences.
Acknowledgments
The Psoriatic Arthritis Program is funded in part by The Arthritis Society, Canadian Institutes of Health Research and the Krembil Foundation.
References
Footnotes
-
Competing interests None.
-
Provenance and peer review Not commissioned; externally peer reviewed.