Background/Purpose: Cardiovascular (CVD) mortality rates are raised in patients with inflammatory polyarthritis (IP) but have been shown to be falling in the general population. This study aims to examine CVD mortality over time in a cohort of recent onset IP patients compared to the general population in Norfolk, UK.

Methods: Between 1990-2004, patients >16 years with ≥2 swollen joints for >4 weeks were registered to an inception cohort, the Norfolk Arthritis Register (NOAR), in Norfolk, UK. Three cohorts (limited to symptom onset <2 years before baseline assessment) were defined by year of baseline assessment: cohort 1 (1990-1994); cohort 2 (1995-1999); cohort 3 (2000-2004).  Rheumatoid factor (RF) and anti-citrullinated protein antibody (ACPA) status (+/-) were determined from blood samples obtained at baseline assessment. The 1987 ACR criteria for RA were also applied. Patients were tracked via Office for National Statistics (ONS) for notification of death. CVD as underlying cause of death was defined according to ICD-10 (Chapter “I”). ONS also provided CVD death rates for the Norfolk general population. For all cohorts, standardised mortality ratios (SMRs) for CVD were calculated over 5 years from baseline as well as for subgroups who fulfilled the 1987 ACR criteria for RA; those RF+ or ACPA+ at baseline. To ensure adequate statistical power, analyses were performed only for subgroups with at least 10 observed CVD deaths over the 5 year follow up.

Results: Cohort 1 comprised 1006 patients, cohort 2: 880 patients and cohort 3: 638 patients. Approximately two-thirds were women in each cohort and median age at symptom onset was 54 (IQR 42-67); 55 (IQR 44-67); 58 (IQR 47-70) years in each respective cohort. The percentages RF+ were 28% (249/877), 29% (225/788) and 36% (203/559) and ACPA+ were 23% (174/745), 26% (191/730) and 31% (160/517) in each respective cohort.

By 5 years from baseline there were 96 (3.8% of the total cohort) CVD deaths: 36 (3.6%); 34 (3.9%); 26 (4.1%) in each respective cohort. The overall crude CVD death rate in adults >16 years in Norfolk decreased over time: 3.3% (1990-1994); 3.0% (1990-1994) and 2.7% (2000-2004). Five year CVD SMRs were raised in all cohorts but only significant statistically in cohort 3: cohort 1: 1.13 [95% CI 0.82-1.57]; cohort 2: 1.29 [95% CI 0.92-1.81] and cohort 3: 1.51 [95% CI 1.03-2.22].  RF+ patients had significantly raised 5 year CVD SMRs (1.74 [95%CI 1.05-2.88]) as did ACPA+ patients (2.17 [95%CI 1.23-3.80]) in cohort 1; more recent cohorts were excluded as <10 CVD deaths were observed. CVD SMRs were not statistically significantly increased in the subgroup of patients who met the ACR 1987 criteria for RA at baseline. No consistent differences in CVD SMRs between men and women were observed across the cohorts.

Conclusion: Raised SMRs for IP patients in advancing cohort years may be due to the declining CVD deaths observed in the general population over the same time period. The status of RF+ and ACPA+ may also contribute to excess early CVD mortality in IP patients.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/widening-gap-between-cardiovascular-specific-mortality-in-patients-with-inflammatory-polyarthritis-compared-to-the-general-population/