乌鸦传媒

Swedes with psoriatic arthritis (PsA) tended to die earlier than other people, a national case-control study indicated, but the risk varied considerably among different subgroups.

Overall all-cause mortality was 11% higher for PsA patients compared with individuals drawn from the general population (95% CI 7%-16%), according to Johan Wallman, MD, PhD, of Lund University in Sweden, and colleagues; the incidence rate ratio (IRR) was 1.18 (95% CI 1.13-1.22).

As groups, the people with PsA faced a 10% mortality rate after 9.3 years, versus 10.5 years for controls.

But the researchers also found disparities by sex, age, and disease duration, they .

Male PsA patients faced a substantially smaller increase in risk compared with women (9% vs 27%, P<0.05). When people were stratified by age, the greatest increase in mortality risk relative to controls was seen in PsA patients in the 40-49 age range (IRR 1.37, 95% CI 1.08-1.73).

Sex and age appeared to be at least additive: mortality rates among female patients ages 40-49 were 73% higher than in similarly age controls (95% CI 21%-149%). For every age group examined in men, on the other hand, mortality was not significantly greater for patients versus controls.

Meanwhile, long-standing PsA (defined as first diagnosed prior to 2007) carried a 25% increase in mortality versus controls (95% CI 19%-32%), whereas those diagnosed from 2007-2017 had only a barely significant 6% increase (95% CI 0%-13%).

Causes of death were similar between patients and controls.

Wallman and colleagues were hoping to resolve conflicts among previous studies regarding mortality in PsA: some had found substantial increases relative to the general population while others had not. This was a bit surprising insofar as increased mortality is well established for people with plaque psoriasis. PsA, however, is less common and many of the earlier studies would have been underpowered to detect the increases found by Wallman's group.

For this study, the researchers combed Swedish national registry data to identify PsA patients diagnosed from 2001 to 2017, ultimately finding 33,026 with sufficient data for analysis; 38% of these were diagnosed prior to 2007. Each was matched by sex, age, and county of residence to five people from the general population. Median follow-up was just under 9 years.

Mean age at diagnosis was 52 and 45% were men.

Besides sex, age, and disease duration, other factors were also associated with increased mortality relative to controls. Chief among them were comorbidities. Chronic kidney disease brought the biggest increase, with a hazard ratio for death of 3.90 (95% CI 3.48-4.38) in patients relative to controls. Others included cardiovascular disease, diabetes, pulmonary disease, cancer, and anxiety/depression, each of which came with hazard ratios in the range of 1.65-2.21. Patients who had had joint surgery also faced a smaller but still significant 15% increase in mortality risk.

The researchers had no firm explanations for the disparities by sex, age, and disease duration. Wallman and colleagues speculated that women may be more burdened than men by comorbidities. Indeed, when the group adjusted the data for comorbidities, "the elevated mortality risk associated with PsA disappeared." Just about every type of comorbidity was more common in patients versus controls; cardiovascular disease, for example, was recorded for 41% of patients compared with 31% of controls.

As well, Wallman and colleagues argued that longer disease duration would be expected to raise mortality risk because of cumulative effects from chronic systemic inflammation. But it remains a mystery why, in the age stratification, patients in their 40s had the highest increase in risk relative to controls.

Limitations to the study included reliance on administrative data and a lack of information on potential confounders such as body mass index or smoking status. Wallman and colleagues noted that PsA cases treated exclusively in private clinics or in primary care may not be counted in the national registry. No attempt was made to evaluate possible impacts of different treatment regimens. Also, it's possible that some PsA diagnoses were incorrect, although the researchers said this would not influence the results substantially.

"[F]urther studies are warranted to disentangle the effects of disease activity and medication on mortality risk," the group concluded.

Comment