Prevalence of CVD-related comorbidity in ankylosing spondylitis, psoriatic arthritis and psoriasis: A matched retrospective cohort study

Conference: 
SAPC ASM 2015 - Oxford - Evidence and innovation in primary care
Talk Code: 
EP1A.5

The problem

Patients with ankylosing spondylitis (AS) and psoriatic arthritis (PsA) have an increased risk of mortality through excess cardiovascular disease (CVD). However, the prevalence of CVD-related comorbidity, in patients with AS and PsA is still unclear in UK primary care. We examined the prevalence of CVD-related comorbidities in patients with AS, PsA and psoriasis (Ps), each compared with controls from UK primary care.

The approach

We conducted a matched retrospective cohort study using the consultation in primary care (CiPCA) database. Three cohorts were constructed using all patients with a diagnosis of AS, PsA or Ps (without known PsA), identified using Read codes between 1999 and 2009. Each cohort was matched in a 1:4 ratio to controls by age, gender and general practice. CVD-related comorbidities including, hypertension, ischaemic heart disease (IHD), hyperlipidaemia and diabetes mellitus (DM), were also identified using Read codes. The prevalence of each of these comorbidities were recorded and compared to their control group using chi square test. The prevalence of each CVD-related comorbidity were also stratified into pre or post AS, PsA or Ps diagnosis.

Findings

Patients with AS (n=94), PsA (n=106), Ps (n=290) were identified and matched to controls for age, gender and general practice. The proportion of each CVD-related comorbidity was higher in AS and PsA cohorts compared with their controls, yet this was only significantly different for hypertension (AS35 (37.2%) vs. controls 96 (25.5%), p = 0.029; PsA 41 (38.7%) vs. controls 114 (26.9%), p = 0.023). No difference was seen in the prevalence of any of the CVD-related comorbidities in those with Ps compared to controls. Interestingly, there was a higher prevalence of CVD-related comorbidities after the diagnosis of AS and PsA, compared to before the diagnosis, including IHD (AS 8 (80.0%) vs. controls 2 (20.0%) and PsA 6 (60.0%) vs. controls 4 (40.0%)). This was also the case for hyperlipidaemia (AS 7 (63.6%) vs. controls 4 (36.4%) and PsA 9 (75.0%) vs. controls 3 (25.0%)).

Consequences

Our findings provide a novel comparison of the prevalence of CVD-related comorbidities in three inflammatory conditions, compared with controls, in a primary care population. We found the prevalence of these CVD-related comorbidities to be lower than we would have expected. However, a higher proportion of certain CVD-related comorbidities have been shown to occur after the diagnosis of the inflammatory condition but further analysis was limited by small sample sizes. This study provides an interesting comparison of different inflammatory conditions, for which increased systemic inflammation and CVD comorbidity would be expected. These findings merit further investigation, particularly into the incidence of CVD-related comorbidities post-inflammatory condition diagnosis and how such rates may vary across conditions such as AS and PsA.

Credits

  • Nadeem Ahmed, Haywood Rheumatology Centre, Haywood Hospital, Stoke-on-Trent, UK
  • James Prior, Haywood Rheumatology Centre, Haywood Hospital, Stoke-on-Trent, UK
  • Ying Chen, Haywood Rheumatology Centre, Haywood Hospital, Stoke-on-Trent, UK
  • Richard Haywood, Haywood Rheumatology Centre, Haywood Hospital, Stoke-on-Trent, UK
  • Samantha Hider, Haywood Rheumatology Centre, Haywood Hospital, Stoke-on-Trent, UK