Mild, Nonprogressive Disease

Wright, in his original series, identified 2 populations of patients—a group with severe disease onset followed by a complete and lasting recovery, and a more severe progressive group. Although they did not present exact figures, Wright stated that cases with mild disability outnumbered those with severe disability. In that series, 74% of the total group presented with oligoarthritis. They confirmed the observations of predominantly limited disease in a subsequent article that reported radiologic changes. A later review of these patients confirmed the largely benign nature of the condition: 78% of the 178 patients were classified as mild, with only 11% of this group deteriorated over a follow-up period of more than 10 years.

Later studies, particularly those from Toronto, also confirmed the presence of a milder, nonprogressive, oligoarticular population. For example, the first complete series published by Gladman and colleagues included a large nonerosive group. Subsequent series reported at damage progression in a group of 33% to 36% of patients who had no evidence of damaged joints at either presentation or follow-up. A similar proportion of patients—28%—remained without disability over a 10-year period.

In unselected patients, approximately one-quarter demonstrate at least 1 erosion at presentation. Data from a study in Dublin found this proportion increased to 41% after 2 years ; a more recent study from an early (<2 years) cohort found that at least 27% had 1 erosion at presentation, increasing to 31% at 1 year. This erosive burden will, of course, depend on treatment; psoriatic arthritis patients develop less erosive disease (at least in the hands and feet) than rheumatoid arthritis patients at all stages of disease.

Prospective, community-based data are not available, but a single report identified psoriatic arthritis from health records over a 9-year period. Despite the large population (population 124,277) numbers of cases were small at 66 (expected numbers in the region of 250). Only 10 people in this largely oligoarticular group received disease-modifying antirheumatic drugs (DMARDs) and only 8% developed erosions in the hands.

Progressive Deforming Arthritis

Wright reported a group of patients who developed a severely deforming arthritis, termed arthritis mutilans, with osteolysis of the digits, polyarticular involvement, and frequent spinal involvement. Affected individuals often developed telescoping joints and flail digits in both hands and feet, resulting in severe disability. However, this distinctive group only represents a small proportion of affected people, less than 5% in most series. The majority of patients with psoriatic arthritis develop polyarticular arthritis with progressive deformity and damage. This mutilans group slowly accrues damaged joints and becomes progressively more disabled, as demonstrated by several cross-sectional and longitudinal surveys.

Erosive disease of the peripheral joints is an indicator of progression, but other manifestations of psoriatic arthritis, particularly spinal disease, are important in the natural history. The prevalence of spinal involvement, identified by signs and symptoms, is 40%. In many patients, involvement is asymptomatic. The Toronto group noted radiologic progression of spinal disease in the absence of clinical progression (as indicated by symptoms and spinal mobility) over a mean follow-up period of 57 months—clearly important to document when defining the natural history of the disease.

Cross-Sectional Studies

Cross-sectional studies provide limited data about prognostic factors owing to potential bias and confounding factors. Roberts and associates identified a more severe deforming group with of psoriatic arthritis patients with an acute onset, systemic features, onset before the age of 20, and spinal involvement. Kammer and co-workers found that a polyarticular group was more likely to progress and to have nail involvement. More recently, a number of studies have indicated that delay in diagnosis is associated with greater joint damage, suggesting that earlier diagnosis may improve outcomes.

Longitudinal Studies

In longitudinal studies, significant polyarticular inflammation at onset, represented by more than 5 swollen joints and an increased erythrocyte sedimentation rate, predicts joint damage. In an extension of this study, existing damage and previous steroids use were also predictive of joint damage. Two other longitudinal series reported similar results. Kane and colleagues, studying an inception cohort, found that patients with an oligoarticular disease onset were more likely to be in remission at 2 years. Conversely, in a separate Spanish study, polyarticular disease was the main predictor of erosions.

Taken together, and recognizing that effective treatment can delay or prevent progression, these longitudinal studies suggest that earlier intervention in patients with more active disease might improve outcomes. Indeed, recent data from the Swedish early arthritis register indicates better outcomes for people with shorter symptom duration at diagnosis, confirming the data from cross-sectional studies. It is likely, albeit not yet proven, that these improved outcomes with earlier referral and diagnosis are the result of earlier treatment. In addition, recent data indicate that a treat-to-target approach also improves outcomes, but additional studies are needed to examine the efficacy of this strategy in different phenotypes.

Other Clinical Prognostic Factors

Dactylitis, a hallmark clinical feature of psoriatic arthritis, represents inflammation in articular, osseous, and extraarticular tissues of a digit. It is present in about 30% of cases of psoriatic arthritis at some time and may be associated with marked disability, particularly when present in the hand. Dactylitic digits are more likely to develop progressive radiologic changes compared with nondactylitic digits, thus making dactylitis an important marker of disease severity and progression in psoriatic arthritis.

In cross-sectional studies, nail disease has been associated with more severe disease. In one study, describing the development of a new instrument to quantify nail involvement. A Psoriatic Nail Severity Score (PNSS) score of more than 16 was associated with more hospital admissions for psoriasis and greater radiologic progression of the arthritis.

Early hip involvement is typical of ankylosing spondylitis; in psoriatic arthritis, a series from the Mayo clinic showed that patients with hip disease are significantly younger at onset (29 vs 40 years of age) and are more likely to have spondylitis. Although only a few patients with established psoriatic arthritis have clinical and radiologic hip involvement, hip arthritis is often disabling and may require early arthroplasty.

HLA Studies and Serum Markers

In rheumatoid arthritis, specific HLA haplotypes are associated with disease severity and susceptibility. It seems reasonable, then, to seek such predictors in psoriatic arthritis, and genetic factors are associated with different subgroups of psoriatic arthritis.

Espinoza and associates first reported an association between HLA Bw38 and peripheral polyarthritis (the subgroup most likely to progress) and this was later confirmed in a larger series from Toronto. The Toronto group also found a higher prevalence of HLA DR4 in the polyarticular subgroup, providing distal interphalangeal joints were not involved. The role of HLA antigens in predicting disease progression was studied in 2 additional series from Toronto. Progression was predicted by the following HLA antigens: combination of HLA B27 and HLA DR7 (relative risk [RR], 2.47), DQw3 (RR, 1.63), and HLA B39 (RR, 7.05). The association of DQw3 with DR7 was, however, protective (RR, 0.54). In a later Swedish study, no HLA haplotypes were predictive; the only significant variables associated with articular destruction were distal interphalangeal involvement and a polyarticular pattern.

Data on cytokine polymorphisms indicate that erosive disease, and by inference the group with the worst prognosis, is associated with certain subtypes of tumor necrosis factor (TNF). Balding and colleagues examined 147 patients with psoriatic arthritis and compared them to 389 healthy controls. Fifty-one psoriatic arthritis patients had sequential radiographs to assess the progression of erosive disease. TNFα -308 and TNFβ +252 were associated with erosive disease and progression of erosions. It has also been suggested that a minor allele of the interleukin (IL) 12 receptor may also be ‘protective’ for disease severity, as indicated by peripheral erosions (odds ratio [OR], 0.47; 95% CI, 0.18–1.20).

Finally, anticitrullinated peptide antibodies, usually associated with rheumatoid arthritis, are present in low titer in less than 10% of patients with psoriatic arthritis. These antibodies may be associated with more erosive disease, based on data from 1 cohort. These data await confirmation.

Disability and Quality of Life

Psoriatic arthritis impacts significantly function and quality of life. Sokoll and Helliwell compared impairment, function, and quality of life in a series of 50 patients with psoriatic arthritis and patients with rheumatoid arthritis matched for age, sex, and duration of disease. Impairment, as measured by radiologic damage score, was greater in the rheumatoid arthritis group, but both disability and quality of life were similarly impaired in both groups. Some of the impairment in psoriatic arthritis may have been attributable to the associated skin disease; patients graded for severity of skin disease showed a decline in function and quality of life with increasing severity, independent of their joint disease. A similar study from Toronto found that rheumatoid arthritis patients had markedly higher Health Assessment Questionnaire disability index scores (mean, 1.11) than psoriatic arthritis patients (mean, 0.58). After adjustment for age, sex, and disease duration (the groups were not matched for these variables), only the vitality domain of the Short Form-36 separated the 2 groups. In contrast with the Leeds study, no correlation was identified between skin severity (as measured by the Psoriasis Area and Severity Index [PASI]) and quality of life.

Cardiovascular Morbidity

A number of studies have highlighted the association between cardiovascular morbidity, and associated risk factors (obesity, hypertension, type 2 diabetes mellitus, and hyperlipidemia) in psoriasis. These cardiovascular risk factors were correlated with increasing extent of psoriasis. The added burden of risk associated with joint disease is not completely clear, although it would be expected unresolved inflammation at a second site (joints) is significant. However, to fully understand the incremental contribution from the arthritis, it is essential to control for traditional risk factors and to acknowledge that many psoriatic arthritis patients have relatively mild psoriasis. Nevertheless, the increase in cardiovascular risk in psoriatic arthritis, largely similar to that found in rheumatoid arthritis, was demonstrated in a systematic literature review. This increased cardiovascular risk provides an additional domain to assess and treat in these patients. The mechanism of the association is not fully understood, but seems to be a combination of an increase in traditional risk factors and the presence of widespread systemic inflammation, a phenomenon termed the “psoriatic march.” Direct evidence of cardiovascular involvement comes from studies of subclinical carotid atherosclerosis that measure intimal media thickness with noninvasive methods.

Most of the studies discussed were performed in a secondary care setting. Do the same considerations apply to a community setting? Some evidence suggests this may not be the case. Gelfand and colleagues published widely on a UK primary care database. Allowing for the likely errors in coding in such a database, this group showed only a modest increase in major adverse cardiovascular events in psoriatic arthritis patients not taking a disease modifying drug (implying milder forms of the disease); the OR was significant at 1.24, whereas the OR of 1.17 for those on a DMARD was not significant. This findings contrasts with rheumatoid arthritis, where the OR for major adverse cardiovascular events, off and on treatment, were both significant at 1.39 and 1.58.

Other Comorbidities

As indicated, an increase in traditional cardiovascular risk factors (obesity, hypertension, type 2 diabetes mellitus, and hyperlipidemia) was demonstrated in psoriasis; not surprisingly, a similar increase was shown in psoriatic arthritis. As befits a condition grouped with other spondyloarthropathies, an increased prevalence of inflammatory bowel disease, either overt or covert, is observed. Finally, psychological morbidity is increased in psoriatic arthritis, particularly depression, an increase shared with psoriasis. An increased risk of non-Hodgkin’s lymphoma in rheumatoid arthritis was reported, but a population-based study in Sweden did not demonstrate a significantly increased risk in psoriatic arthritis.