CARDIOVASCULAR SCORE RISK CALCULATORS IN DIALYSIS PATIENTS: CAN WE USE THEM?

Abstract BACKGROUND AND AIMS Cardiovascular disease is the most significant cause of morbidity and mortality in dialysis patients. Despite significant information regarding the utility of cardiovascular risk score charts in the general population for prediction of major cardiovascular events, their usefulness in dialysis patients was rarely analysed. These scores may underestimate the risk since they were not validated in chronic kidney disease and non-traditional risk factors are also involved in this population. A J-relationship was demonstrated between total cholesterol and cardiovascular death in dialysis, and some data suggest that these charts are useless in dialysis since total cholesterol is included in calculators. The aim of the study was to assess the predictive capacity of risk scores for developing a major cardiovascular event, myocardial infarction or cardiovascular death, in dialysis. METHOD We performed a prospective study in which we evaluated 2 score charts, the Framingham risk score (FRS) for hard coronary heart disease and the new SCORE 2, in haemodialysis (HD) and peritoneal dialysis (PD) patients. We excluded patients with diabetes because FRS is not recommended in these patients. Predictive power for myocardial infarction (MI) and cardiovascular death (CvD) of both scores was evaluated using receiver operating characteristic curve analysis by SPSS ver. 20.0. RESULTS We included 232 PD patients (114 F), mean age 58.4 + 7.2 years, and 260 HD patients (137 F), mean age 59.7 + 3.9 years. Mean follow-up was 5.5 years. Majority of these patients were classified as high cardiovascular risk (CvR) by both charts: 65% by FRS and 54% by SCORE2 in PD, 69% by FRS and 58% by SCORE2 in HD. Both scores were significantly higher in patients with renal hypertensive disease compared with other causes of kidney disease (Table 1). We found no significant differences regarding these scores in HD versus PD. The best predictive value for developing MI during follow up was obtained by FRS (AUC 0.885 for HD, AUC 0.887 for PD) compared with SCORE2 (AUC 0.776 for HD, AUC 0.724 for PD). FRS had also the best predictive value for CvD (AUC 0.812 for HD, 0.808 for PD) compared with SCORE2 (AUC 0.701 for HD, 0.694 for PD). Regarding MI, all events occurred in high-CvR patients during follow-up according to these two risk scores: 22 events (9.5%) occurred in PD and 26 events (10%) in HD. All CvD events also occurred in high-CvR category, with 36 events (15.5%) in PD and 39 events (15%) in HD. After adjustment for age, gender and previous cardiovascular events, both scores were independent predictors for MI and CvD- FRS had an adjusted HR 4.3 [95% confidence interval (95% CI) 1.8–8.6; P = 0.005] for HD, HR 3.7 (95% CI 1.5–6.6; P = 0.001) for PD; SCORE2 had an adjusted HR 4.9 (95% CI 2.8–9.3; P = 0.003) for HD, HR 5.1 (95% CI 1.4–7.7; P = 0.001) for PD. CONCLUSION We found that both risk scores, the Framingham risk score and SCORE2, are useful to predict major heart events (MI and CvD) in dialysis, both HD and PD. These charts may be used as a tool in order to improve patients’ management. Our results need to be investigated and validated in larger prospective studies.