Background/Purpose:

A greater than three-fold increase in cardiovascular mortality as been reported within the first five years of diagnosis of anti-neutrophil cytoplasmic antibody (ANCA) associated systemic vasculitis (AASV) compared to the general population.  A 5 year predictive cardiovascular risk model was developed using data derived from a trial cohort, to stratify patients into a high, moderate or low risk of a cardiovascular (CV) event.

We describe the CV events occurring in an inception cohort of patients with a new diagnosis of AASV within the first 5 years of diagnosis and test the validity of the existing predictive risk model in parallel.

Methods:

A retrospective case note review was performed on 387 patients recruited to a vasculitis registry with a diagnosis of granulomatosis with polyangiitis (GPA) or microscopic polyangiitis (MPA).

CV events following the diagnosis of GPA or MPA in this cohort were analysed over time and results compared to the predicted rate per 1000 general population matched for age and gender.  CV events were then stratified according to proteinase-3 (PR3) or myeloperoxidase (MPO) ANCA positivity. All results were then compared to the data from the published trials cohort.

Patients without pre-existing CV events were used to validate the existing predictive cardiovascular risk model.

Results:

39/387 (10.1%) patients experienced at least one CV event during follow up (0.02 – 27.8 years, median 3.6 years).

The CV event free-survival time was significantly reduced in the MPA patients vs GPA patients (p=0.0077) over 120 months (figure 1). However, no statistical difference was found between MPO or PR3 ANCA positive groups (p= 0.2767).

A trend towards more CV events was observed within the AASV group compared to the predicted rates per 1000 general population although this did not reach statistical significance.

Regarding the predictive risk model, the inception cohort data was similar to the trial cohort data in predicting CV events based on age, hypertension and the absence of PR3 status when comparing the percentage of patients per risk group (p=0.5309) and CV events by risk group (p=0.9278).

Conclusion:

We demonstrated an overall trend towards more CV events in vasculitis patients compared to the general population, although this does not reach statistical significance.

MPA is associated with significantly more CV events than GPA with events occurring earlier in the disease course.  However, there is no difference in CV events between MPO or PR3 positive patients suggesting that clinical phenotype may be more relevant in identifying patients at risk of CV events than serological status.

We have provided further validation for the published 5 year cardiovascular model in this inception cohort, proving its transferability to clinical practice to allow stratification of individual CV risk.