Markers of inflammation and cardiovascular disease in dialysis patients : variability and prognostic value

Abstract

Patients with end-stage renal disease are burdened with a high cardiovascular morbidity and mortality. Inflammation is a common feature in these patients and strongly associates with cardiovascular complications and outcome. Variability in markers of inflammation is known to a certain degree but associated factors and the difference between dialysis modalities have been scarcely studied. N-terminal pro-brain natriuretic peptide (NT-proBNP) is a predictor of cardiac events and death in the general population and in chronic kidney disease. It is known to be greatly elevated in dialysis patients but little is known about variation over time within individuals, the difference between individuals and the prognostic effect of NT-proBNP variation. Similarly, data on clinical factors relating to increases in NT- proBNP in end-stage renal disease is restricted. High-sensitivity cardiac troponin I (hs-cTnI) and T (hs- cTnT) are elevated in dialysis patients even without signs of cardiac ischemia. Cardiac troponins have been shown to predict prognosis but very little is known about variability of hs-cTnI and T in dialysis patients. Our aim was to characterize short-term variability of high-sensitivity CRP (hs-CRP), interleukin- 6 (IL-6), NT-proBNP, hs-cTnI and hs-cTnT and to explore factors associated with variability of these markers in prevalent dialysis patients. Another aim was to evaluate the prognostic value of serial measurements of hs-CRP, NT-proBNP, hs-cTnI and hs-cTnT.

In paper I, weekly hs-CRP was measured in 228 prevalent hemodialysis (HD) patients during three months. Results showed large intra-and interindividual variation in hs-CRP and associations with 7 out of 12 clinical symptoms/events, comorbidity, male sex and age. hs-CRP at baseline correlated to the time averaged hs-CRP (Spearman’s rho 0.76) and the individual median hs-CRP (Spearman’s rho 0.78) both p<0.001. Hazards ratios for all-cause mortality were significant for age, comorbidity, dialysis vintage and serial hs-CRP values. Median hs-CRP associated with a higher hazard ratio for death than baseline hs-CRP, maximum hs-CRP, minimum hs-CRP and average hs-CRP. In paper II, repeated values of IL-6 (monthly) and hs-CRP (weekly) were measured in 228 prevalent HD (same as in paper I) and 80 prevalent peritoneal dialysis (PD) patients. IL-6 was higher in HD than PD patients; median IL-6 8.3 (IQR, 5.3-14.5) vs. 6.7 (IQR, 4.2-10.0) pg/ml; and median hs-CRP 6.1 (IQR, 2.5-14.0) vs. 5.4 (IQR, 1.6-9.0) mg/l; p<0.001 for both. Clinical events, age, male sex, protein-energy wasting (PEW) and HD predicted increased hs-CRP and IL-6 variability. Increased comorbidity predicted IL-6, but not hs-CRP, variability. In paper III, monthly measurements of NT-proBNP were obtained from 211 prevalent HD patients (same cohort as in paper I and II). Inflammation, age, PEW and comorbidity were predictors of NT-proBNP variability. Patients with constantly high NT-proBNP (above 18443) during three months had a significantly increased risk for death adjusting for age, sex, vintage and comorbidity but not when also adjusting for PEW. In paper IV, hs-cTnI and hs-cTnT monthly measurements were obtained from 198 prevalent HD and 78 PD patients (same cohorts as in paper II). The levels of troponins were similar in HD and PD patients; median (IQR) hs-cTnI; 25ng/L (14–43) vs. 21ng/L (11–37), hs-cTnT: 70ng/L (44–129) vs. 67ng/L (43–123). 42% of hs-cTnI and 98% of hs-cTnT measurements were above the decision level for myocardial infarction. Variability of troponins was associated with age, male sex, PEW and congestive heart failure but not ischemic heart disease or dialysis modality. Constantly high hs-cTnT levels (above 108 ng/L) predicted death (HR 2.09 95% CI 1.03-4.26) after adjusting for confounders, whereas hs-cTnI levels did not.

Levels of IL-6 and hs-CRP are increased and have a high variability in dialysis patients, more so in HD than PD. Comorbidity, PEW and acute clinical events are strongly related to inflammatory activity in dialysis patients. In spite of the variability, inflammation as assessed by hs-CRP is a strong predictor of mortality in HD patients. Serial measurements provide additional information compared to single measurements. Longitudinal changes of NT-proBNP are associated with inflammation, PEW, age and comorbidity. Repeatedly high NT-proBNP levels predict mortality. Dialysis patients without signs of ischemic cardiac events have elevated hs-cTnI and hs-cTnT. The large intra-individual differences in cardiac troponins support the use of reference change levels when assessing a patient with a possible acute cardiac event. Constantly high levels only of hs-cTnT and not hs-cTnI predicted risk of death.