Page 127 - THE EVOLUTION OF EARLY ARTHRITIS AND CARDIOVASCULAR RISK Samina A. Turk

inflammation generally leads to an increased TC:HDL ratio, a higher CV risk is expected if markers of inflammation are high(18;19). This association was opposite for DAS44, a possibility is that other components of the DAS like the visual analogue scale and/ or tender joint count disturb this association. The association between measures of inflammation and cholesterol levels, especially TC:HDL ratio was strongest at baseline, when all patients had a high disease activity. At four weeks, nearly all patients had low disease activity, which explains why this association was not present anymore after four weeks. Although there was no unambiguous association between disease activity and CV risk score, we do think that it is important to calculate CV risk during a time of low disease activity.
The change in CV risk score and so the advice about preventive treatment is probably correlated with the reduction of disease activity or the initiation of anti-rheumatic treatment. Previous literature described a reduction in acute myocardial infarction with the use of methotrexate (RR 0.81), but a dose-dependent increase with glucocorticoid use (RR 1.32)(36-38). An improvement in TC:HDL ratio was found after one and two years of COBRA-light treatment, however this did not had a favorable effect on CV risk prediction(18;19). Furthermore, lower disease activity (obtained with anti-rheumatic treatment) was associated with a lower blood pressure(39;40).
Unfortunately, we could not take into account some additional CV risk factors, such as 6 renal function, physical activity and family history of CVD. These factors are considered
as CV risk modifying factors which can be an additional reason to give CV risk prevention treatment. Therefore, the lack of these factors may have influenced our results.
In addition to the Dutch SCORE and Heart SCORE, different CV risk scores are available. For example, the Framingham risk score is commonly used in the United States. However, this score is not generally applied in Europe and is limited to estimating the 10-year risk of a myocardial infarction and coronary heart disease-related death, thus underestimating the total atherosclerotic vascular disease risk(31;41). Especially, risk calculators that correct for the systemic inflammation are interesting. For example, the QRISK-2 and QRISK-3 calculators are used to predict CV risk in the United Kingdom, these calculators take RA into account as a separate CV risk factor. As a zip code is also a component of these algorithms, these calculators are not feasible in other countries(5;42). The Reynolds Risk score includes high-sensitivity CRP levels into the risk model, however this risk score is not recommended for patients with a systemic inflammatory disease, as CRP levels will be increased due to the inflammatory disease. (5) Further efforts were already performed to develop a RA-specific risk calculator, however, a new calculator (including DAS or HAQ) did not demonstrate an improvement compared to the current CV risk models which are used in the general population(43;44). The influence of fluctuations in disease activity over time in RA patients is difficult to incorporate in risk prediction models; single disease activity measurements (as DAS, CRP and ESR) are maybe not good enough, a biomarker that measures cumulative RA disease activity might fulfill this unmet need(44).
CHANGE IN CARDIOVASCULAR RISK
125