Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

CHAPTER ELEVEN

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease Z.H. Endre, BScMed, MBBS, PhD, RACP, FASN*,**,† and R.J. Walker, MBChB, MD, FRACP, FASN, FAHA‡

*Department of Nephrology, Prince of Wales Hospital and Clinical School, University of New South Wales, Sydney, NSW, Australia **Department of Medicine, University of Otago, Christchurch, New Zealand † School of Medicine, University of Queensland, Brisbane, QLD, Australia ‡ Department of Nephrology, Dunedin Hospital and University of Otago, Dunedin, New Zealand

Contents Introduction The Definition of CKD and the Risk of Cardiovascular Diseases: GFR, Albuminuria, and Proteinuria Biomarkers of Progression of CKD Summary Cardiac Biomarkers in CKD Troponins Natriuretic Peptides C-Reactive Protein Fibroblast Growth Factor 23 Placental Growth Factor Galectin 3 Neutrophil Gelatinase-Associated Lipocalin Summary References

485 486 493 496 497 497 499 502 503 505 505 506 507 508

INTRODUCTION Chronic kidney disease (CKD) is an established independent risk factor for cardiovascular disease. Likewise cardiovascular disease is the major cause of morbidity and mortality in patients with CKD [1]. With the progression of CKD, there is a change from the traditional atherosclerotic related cardiovascular events, such as stroke and myocardial infarction, to increased arterial stiffness, arteriosclerosis, left ventricular hypertrophy, cardiac fibrosis, and heart failure. This change in pathophysiology is associated with increased risk of arrhythmias and sudden cardiac death. As a consequence Biomarkers of Kidney Disease. http://dx.doi.org/10.1016/B978-0-12-803014-1.00011-X Copyright © 2017 Elsevier Inc. All rights reserved.

485

486

Z.H. Endre and R.J. Walker

of the altered pathophysiology, biomarkers of cardiac injury are more detectable in patients with CKD. These changes in the biomarkers need to be interpreted in the clinical context associated with CKD. This chapter will review traditional markers of kidney disease, traditional markers of cardiovascular disease, and novel markers of kidney damage as markers of cardiovascular risk in subjects with CKD.

THE DEFINITION OF CKD AND THE RISK OF CARDIOVASCULAR DISEASES: GFR, ALBUMINURIA, AND PROTEINURIA CKD is common. Defined as a glomerular filtration rate of less than 60 mL/ min per 1.73 m2, or the presence of kidney damage regardless of cause for more than 3 months, the prevalence of CKD in western populations is of the order of 10–16% and increasing [2–4]. In 2013, there were approximately 470 million cases of CKD globally, and, depending on cause, the prevalence had increased from 1995 by 26.8% for CKD from hypertension to 82.5% for CKD due to diabetes mellitus, with intermediate increases for CKD arising from other causes [5]. Most subjects with CKD never reach dialysis. It has long been appreciated that the majority of these subjects with CKD (50%) will die of cardiovascular disease before ever requiring renal replacement therapy [6]. The association between CKD and cardiac pathology (hypertrophy) at post mortem was first reported by Richard Bright in 1836 (cited by Gansevoort [7]). Most recently, it has become accepted that all severity stages of CKD from mild disease to the dialysis-dependent end-stage renal disease (ESRD) are associated with an increase in cardiovascular risk and mortality [1,7–9], with the absolute risk of death increasing exponentially with decreasing kidney function [1,10]. Nevertheless, guidelines on the management of cardiovascular risk have paid limited attention to CKD as a major risk factor. This section will address how the development of the definition of CKD has allowed cardiovascular disease risk to be quantified and examine the robustness of this definition as a risk factor for mortality, particularly cardiovascular mortality. Nephrologists and others have become increasingly aware of the frequency of cardiovascular disease in CKD, recognizing that this represents the major cause of death among CKD patients [3]. This recognition was assisted by the introduction of a conceptual model in 2002 for definition and classification of CKD by the National Kidney Foundation’s Kidney

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

487

Disease Outcomes Quality Initiative (KDOQI), which was subsequently endorsed by the Kidney Disease: Improving Global Outcomes (KDIGO) [11]. CKD was defined as kidney damage or a glomerular filtration rate, GFR, <60 mL/min per 1.73 m2 for 3 months or more, irrespective of cause. Kidney damage was defined by structural or functional abnormalities that could lead to decreased GFR, manifest either by pathologic abnormalities or markers of kidney damage, including abnormalities in the composition of blood or urine, or abnormalities in imaging tests. Albuminuria was accepted as a marker of kidney damage at a threshold albumin-to-creatinine ratio (ACR) of 30 mg/g. Estimates of GFR (eGFR) based on serum creatinine and calculated using estimating equations were also accepted. CKD was subdivided into five stages defined respectively by the eGFR bands 1, ≥90; 60–89; 30–59; 15–29, and <15 mL/min per 1.73 m2. Early application of the classification to risk assessment revealed that there was an independent, graded association between a reduced eGFR and the risk of death, cardiovascular events and hospitalization [1]. The high prevalence of CKD revealed by this classification has enormous implications for clinical practice, research and public health policy. In the United States, some 12% of the population (25 million) was estimated to have CKD, while <0.2% of the population (>500,000) was treated with dialysis or transplantation [11]. Similar prevalence data have been obtained in other countries. For instance, a cross-sectional survey of Australian adults over 25 years of age revealed a high prevalence of 16% with either kidney damage as shown by proteinuria, or hematuria, and/or a reduced GFR, with over 50% of subjects over 65 years of age having an eGFR <60 mL/min per 1.73 m2 [4]. While this was revealing, the public health and cost implications of the high prevalence of earlier stages of CKD compared with the incidence of treated kidney failure, and the concern that thresholds of albuminuria and GFR might vary according to age, sex, or race were issues that made the definition and classification of CKD controversial. This led to a reassessment of the CKD definitions at a KDIGO controversies conference in London in 2009 [12]. This meeting provided an opportunity to revisit the definitions after analysis of a large data set for all-cause and cardiovascular mortality, and for kidney outcomes with data pooled from 45 large cohorts totaling 1,555,332 subjects including general population, high risk, and CKD cohorts. The subdivision of CKD into five stages allowed assessment of the relative risk of five outcomes to declining eGFR and three severity grades of albuminuria (an ACR < 30 mg/g, 30–299 mg/g, and ≥300  mg/g). The

488

Z.H. Endre and R.J. Walker

Figure 11.1  KDIGO Consensus Classification of Chronic Kidney Disease by Severity. The figure shows the composite ranking for relative risks by glomerular filtration rate (GFR) and albuminuria (KDIGO, 2009 reported in Ref. [12]) reproduced with permission. Colors reflect the ranking of adjusted relative risk. The ranks assigned in Figure 5 were averaged across all 5 outcomes for the 28 GFR and albuminuria categories. The categories with mean rank numbers 1–8 are black (green in the web version), mean rank numbers 9–14 are light gray (yellow in the web version), mean rank numbers 15–21 are gray (orange in the web version), and mean rank numbers 22–28 are dark gray (red in the web version). Color for twelve additional cells with diagonal hash marks is extrapolated based on results from the metaanalysis of chronic kidney disease cohorts [12]. The highest level of albuminuria is termed “nephrotic” to correspond with nephrotic range albuminuria and expressed as ×2000 mg/g.

outcomes assessed were all-cause mortality, cardiovascular mortality, ESRD, acute kidney injury, and progressive kidney disease. Cardiovascular mortality was defined as death due to myocardial infarction, heart failure, sudden cardiac death, or stroke. ESRD was defined as initiation of dialysis or transplantation or death coded as due to kidney disease other than acute kidney injury (AKI). AKI was defined ICD-9 code 584 as primary or additional discharge code. Kidney disease progression was defined as an average annual decline in eGFR during follow-up of at least 2.5 mL/min per 1.73 m2 per year and a last eGFR value of < 45 mL/min per 1.73 m2 (Fig. 11.1).

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

489

As summarized by Levey et al. [12], the incidence rates were higher for mortality than kidney outcomes, but the risk relationships had similar shape and albuminuria patterns, with higher relative risks associated with lower eGFR and higher levels of albuminuria, suggesting that groups at risk for one outcome were at risk for all outcomes. The graded increase in risk for higher albuminuria categories was independent of eGFR, without an apparent threshold value. The increased relative risk for all outcomes was significant for eGFR of <60 mL/min per 1.73 m2 in the continuous analysis and in the range 45–59 mL/min per 1.73 m2 in the categorical analysis, consistent with the threshold value for the definition of CKD (<60 mL/ min per 1.73 m2). The predictive value of albuminuria at all levels of eGFR supported the need to add albuminuria stages to all eGFR stages. Agestratified analyses showed quantitatively similar patterns in subjects with age <65 and ≥65 years. With the exception of all-cause mortality, the relative hazards were similar above and below 65 years of age, which argued against the use of age-specific GFR thresholds for the definition of CKD. Multiple analyses support and extend these conclusions. The Chronic Kidney Disease Prognosis Consortium was established by KDIGO to compile and metaanalyze the best available data to provide a more comprehensive assessment of the independent and combined associations of eGFR and albuminuria with mortality and kidney outcomes. Metaanalysis of over 100,000 subjects in 21 general population cohorts in 2009 for all-cause and cardiovascular mortality confirmed that both eGFR and albuminuria were independent predictors of mortality risk in the general population [13]. An exponential increase in mortality risk became significant around an eGFR of 60 mL/min per 1.73 m2 and was twofold higher around an eGFR of 30–45 mL/min per 1.73 m2 compared with optimum eGFR levels of 90–104 mL/min per 1.73 m2 independently of albuminuria. In contrast to the threshold effect observed with eGFR, the association between albuminuria and mortality was linear on the log–log scale with a twofold higher risk at a urinary albumin-creatinine ratio (ACR) of approximately 11.3 mg/mmol (∼microalbuminuria) compared with an optimal ACR (0.6 mg/mmol) independently of eGFR and conventional risk factors. Mortality risk was significant even at ACR = 1.1 mg/mmol (∼10 mg/g) compared with 0.6 mg/mmol. Further metaanalysis with individual level data from 637,315 adults (>18 years) in 24 cohorts followed for a median of 4 years, comprising 19 general-population cohorts, three high risk cohorts of individuals with diabetes, and two CKD cohorts (Fig. 11.2) [14] demonstrated that the addition of eGFR and ACR significantly improved the

490

Z.H. Endre and R.J. Walker

Figure 11.2  Cardiovascular outcomes according to estimated glomerular filtration rate (eGFR) and albumin-to-creatinine ratio (ACR) in combined general-population and high-risk cohorts. The shaded areas represent 95% CI for cardiovascular mortality, coronary heart disease, stroke and heart failure. Reproduced with permission from Matsushita K, Coresh J, Sang Y, et al. Estimated glomerular filtration rate and albuminuria for prediction of cardiovascular outcomes: a collaborative meta-analysis of individual participant data. Lancet Diabetes Endocrinol 2015;3:514–525 [14].

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

491

discrimination of cardiovascular outcomes beyond traditional risk factors in general populations, with a greater improvement coming from ACR than eGFR and more evident for cardiovascular mortality and heart failure, than coronary disease and stroke (ibid). Adjusted cardiovascular risk was constant for eGFR values of 75–105 mL/min per 1.73 m2, and increased steadily below this range (Fig. 11.2). The area under the ROC curve (AUC or C statistic) for cardiovascular outcomes based on traditional risk factors ranged from 0.729 to 0.838 and these values were significantly improved by the addition of either measure of kidney disease. Discrimination improvement was greater with ACR than eGFR and especially evident in individuals with diabetes or hypertension, but still significant in the absence of these disorders [14]. Dipstick proteinuria improved risk prediction to a lesser extent than ACR. As identified previously, these conclusions remain valid regardless of associated comorbidities and the increased cardiovascular risk in patients with CKD is not the result of these underlying diseases [9,15]. This conclusion is strengthened by the observation [10] that 40–50% of patients with low eGFR and albuminuria do not have diabetes or hypertension, and the associations of eGFR and albuminuria with cardiovascular mortality are similar in these patients to individuals with CKD with diabetes or hypertension. The effect of age was further assessed by the consortium in an individual-level metaanalysis including 2,051,244 participants from 33 general population and high risk (of vascular disease) cohorts and 13 CKD cohorts from Asia, Australasia, Europe, and North and South America followed up for up to 31 years (mean 5.8 years) [16]. Mortality and ESRD risk increased independently with decreasing eGFR below 60 mL/min per 1.73 m2 regardless of age. Mortality showed a lower relative risk but higher absolute risk differences at older age. In cohorts specifically selected for CKD, age did not modify the mortality associations. The authors further discuss the definition of CKD, highlighting that this could be based on a “normal” distribution in apparently healthy individuals or on kidney markers associated with increased risk of future adverse outcomes. Supporters of the former approach argue for age-specific cut-offs for defining normal kidney function and note an increasing prevalence of nephrosclerosis in renal biopsies with age, even among healthy kidney donors and suggest the concept of “natural inevitable senescence” of the kidneys [17]. However, repeated very large-scale metaanalyses (with over 1 million subjects) conducted by the CKD Consortium document that the risk of mortality and ESRD is increased at a threshold of 60 mL/min per 1.73 m2 across all age groups. The relative mortality risk associated with low eGFR attenuates with increasing

492

Z.H. Endre and R.J. Walker

age, as with traditional risk factors and in assessing the interaction with age, the slopes of the hazard ratios versus eGFR at less than 45 mL/min per 1.73 m2 were largely parallel across the age categories. Similar results were obtained with albuminuria, except, as observed and discussed previously; there was no threshold effect with albuminuria. Consistent with these observations, CKD confers increased short and long-term cardiovascular mortality risk after surgery. Review of 260,352 evaluable cases in the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) data sets for 2005–07 demonstrated that increasing severity of CKD stage was associated with progressive increases in short term (30 day) mortality after any Surgery [18]. Recent studies have also demonstrated increased long-term mortality after surgery irrespective of whether acute kidney injury (AKI) was a complication of that surgery [8,19]. While previous studies were focused on all-cause mortality rather than cardiovascular causes, on the assumption that progression to ESRD was the underlying mechanism for observed long-term mortality, these recent studies have examined the association between kidney disease and cause-specific mortality, particularly cardiovascular-specific mortality. In a single-center cohort of 51,457 adult surgical patients, cardiovascular-specific mortality was modeled using a multivariable subdistributional hazards model, treating any other cause of death as a competing risk [8]. This demonstrated that cardiovascular-specific mortality was significantly higher among patients with kidney disease, independently of the progression to ESRD. Preexisting CKD and ESRD and postoperative AKI were the main independent predictors. After a median follow-up of 7 years (IQR: 5–10 years), the two commonest causes of death were cancer (5051/15247, i.e., 33%) and cardiovascular disease (4269/15247, i.e., 28%). Cardiovascular disease accounted for more deaths among patients with any type of kidney disease compared to 18% among patients with no kidney disease (P < 0.0001), ranging from 29% for patients with AKI and no CKD, 35% for patients with CKD but no AKI, 45% for those with CKD and AKI during admission, and 55% for ESRD patients). In contrast, cancer accounted for fewer deaths among patients with kidney disease compared to those without. At 10-year follow-up, adjusted cardiovascular-specific mortality estimates were 6, 11, 12, 19, and 27% for patients with no kidney disease, AKI with no CKD, CKD with no AKI, AKI with CKD, and ESRD, respectively (P < 0.001). This association remained after excluding 916 patients who progressed to ESRD after discharge although it was significantly amplified among them. Compared to patients with no kidney disease, adjusted hazard

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

493

ratios for cardiovascular mortality were significantly higher among patients with kidney disease ranging from 1.95 (95% CI: 1.80–2.11) for patients with de novo AKI to 5.70 (CI: 5.00–6.49) for patients with preexisting ESRD. The increasing cardiovascular mortality was proportional to the severity of kidney disease independent of patients’ age, gender, comorbidity burden on admission, other postoperative complications or the type of operation in the cohort that included a wide range of major surgical procedures, including noncardiac and specialty surgeries. Thus there is a similar and significant negative impact of both AKI and CKD on long-term survival after surgery due to increased cardiovascular-specific mortality [8]. Similar conclusions were drawn by Huber et al. [19], who examined the association between kidney disease and cardiovascular-specific mortality in a recent large cohort analysis of 3646 subjects undergoing cardiac surgery at a single center from 2000 to 2010. The two commonest causes of the 1577 deaths in the cohort were cardiovascular disease (845 deaths, 56.4%) and cancer (173 deaths, 11.0%). Adjusted cardiovascular mortality estimates at 10 years (after adjusting for demographic variables including age, surgery type, and admission hemoglobin levels) were 17, 31, 30, and 41% respectively for patients with no kidney disease, AKI without CKD, CKD without AKI, and AKI with CKD. The adjusted hazard ratios for cardiovascular mortality were significantly greater for all types of kidney disease than for no kidney disease, even for CKD without AKI (2.01; CI: 1.46–2.78). These were significantly greater than for other risk factors including increasing age, emergent surgery and lower admission hemoglobin levels (less than 10 g/dL versus greater than 12 g/dL).

Biomarkers of Progression of CKD CKD clearly predicts increased mortality and increasing severity of CKD identifies increasing cardiovascular mortality risk. This raises the question of what predicts progression of CKD. As identified by the earlier largescale metaanalyses of the CKD Prognosis Consortium, reduced eGFR and severity of albuminuria or proteinuria predict ESRD and AKI but reduced eGFR does not alone predict progression of CKD [12] (Fig. 11.3). Mortality is associated with the rate of GFR function decline [20] although the mechanism remains unknown [21]. In turn, proteinuria of increasing severity is associated with a faster rate of function decline, regardless of baseline GFR [22]. While past decline in renal function contributes more to the absolute risk of ESRD at lower levels of eGFR, current eGFR was more strongly associated with future ESRD risk than the magnitude of

494

Z.H. Endre and R.J. Walker

Figure 11.3  Kidney outcomes in general population cohorts as a function of eGFR and albuminuria or proteinuria. eGFR is expressed as a continuous variable. The three lines represent urine ACR of <30 mg/g or dipstick negative and trace (light gray; blue in the web version), urine ACR 30–299 mg/g or dipstick 1+ positive (gray; green in the web version), and urine ACR ≥300 mg/g or dipstick ≥2+ positive (dark gray; red in the web version). All results are adjusted for covariates and compared with reference point of eGFR of 95 mL/min per 1.73 m2 and ACR of 30 mg/g or dipstick negative (diamond). Each point represents the pooled relative risk from a metaanalysis. Solid circles indicate statistical significance compared with the reference point (P < 0.05); triangles indicate nonsignificance. Arrows indicate eGFR of 60 mL/min per 1.73 m2, threshold value of eGFR for the current definition of CKD. HR, hazards ratio; OR, odds ratio. Modified from Levey AS, de Jong PE, Coresh J, et al. The definition, classification, and prognosis of chronic kidney disease: a KDIGO Controversies Conference report. Kidney Int 2011;80:17–28 [12].

prior decrease, with both factors contributing substantially to ESRD risk when eGFR was <30 mL/min per 1.173 m2 [23]. The inability to identify patients at high risk of progression while in early stages of CKD confounds effective management. Several novel biomarkers of tubular injury have also been examined to see if these predict progression of CKD. These include evaluation of the association of liver fatty acid binding protein (L-FABP), kidney injury molecule 1 (KIM-1), N-acetyl-β-d-glucosaminidase (NAG), and neutrophil gelatinase–associated lipocalin (NGAL) with incident ESRD in a community-based sample from the Atherosclerosis Risk in Communities Study [24]. This was a matched case-control study of 135 patients with ESRD and 186 controls who were matched on sex, race, kidney function, and diabetes status at baseline. The biomarkers were indexed to urinary creatinine. No significant associations were observed for NAG/Cr, NGAL/Cr, or L-FABP/Cr with ESRD.Those in the highest category for KIM-1/Cr had a higher risk of ESRD compared with those with undetectable biomarker levels (reference group) in unadjusted models (odds ratio, 2.24; 95% confidence interval, 1.97–4.69; P = 0.03) or adjustment for age (odds ratio, 2.23; 95% confidence interval,

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

495

1.06–4.67; P = 0.03). No association between KIM-1/Cr and ESRD was found when KIM-1/Cr was analyzed as a continuous variable. In a prospective observational Canadian cohort study in 2544 CKD patients with eGFR of 15–45 mL/min per 1.73 m2, the utility of the newer biomarkers, cystatin C, high sensitivity C-reactive protein (hsCRP), interleukin 6 (IL6), transforming growth factor β1 (TGFβ1), fibroblast growth factor 23 (FGF23), N-terminal probrain natriuretic peptide (NT-proBNP), troponin I and asymmetric dimethylarginine (ADMA), was assessed with respect to prediction of renal and mortality outcomes during the first year of enrolment [25]. The biomarkers did not improve prediction of dialysis, when added to conventional risk factors, such as eGFR, uACR, hemoglobin, phosphate, and albumin. However, in predicting death within 1 year, cystatin C, NT-proBNP, hsCRP and FGF23 values significantly improved model discrimination and reclassification: increasing the c statistic for absolute risk by 4.3% and Net Reclassification Improvement for categories of low, intermediate and high risk by 11.2%. While CKD has been presumed to follow an unremittingly progressive decline over time, increasing evidence demonstrates that many subjects remain stable over time while a significant minority of 3–9% actually have improving function [25–27]. Several prospective cohort studies of CKD subjects have demonstrated that the majority of subjects do not progress to more severe stages of CKD or end stage kidney disease. For example, among the recruited Sunnybrook cohort of 1607 subjects referred with Stage 3 CKD, 21% progressed to Stage 4 CKD over a median follow-up time of 3 years, whereas only 3% of patients developed ESRD [28]. Progression to Stage 4 CKD was associated with a nearly twofold higher risk of mortality, AKI and all-cause hospitalizations when compared with patients who did not progress by the end of follow-up, and these risks were similar in subgroups of patients with either Stage 3A or 3B CKD at the origin. The risk of cardiovascular-related hospitalization after progression to Stage 4 was higher than noncardiovascular-related hospitalization. Importantly, the observed risks of mortality and morbidity were due to progression to Stage 4 CKD itself rather than as a result of subsequent ESRD. The risk of subsequent mortality and hospitalization remained significant after adjustment for baseline eGFR, age, and other potential confounders, suggesting that progression to Stage 4 was independently associated with long-term adverse clinical outcomes. Survivor bias was thoughtfully avoided in this analysis (in patients presenting with Stage 4 CKD, who would have had to endure a longer duration of CKD and consequently, a

496

Z.H. Endre and R.J. Walker

longer period of elevated mortality risk, compared with patients with Stage 3 CKD) by treating the progression to Stage 4 CKD as a time-dependent variable, comparing hospitalization and mortality risks subsequent to the progression, for those who progressed to Stage 4, with otherwise similar patients who did not progress to Stage 4 during the observation period. In the CanPREDDICT study, a prospective multicentre cohort of 2544 adult CKD patients with an eGFR of 15–45 mL/min per 1.73 m2 enrolled across Canada from 2008 to 2009 over 18 months, were followed for 3 years or until death [25]. After 12 months, a total of 142 patients (5.9%) had progressed to RRT and 137 patients (5.7%) had died [15 (0.6%) after starting RRT]. Twelve months, 10% progressed rapidly (defined as decline of renal function of more than 30%), 13% had slower decline of renal function (20–30%), and 69% were stable, as defined by the change within 20% of baseline, while 9% improved (defined as increase in eGFR > 20%). Progression from Stage 3 to Stage 4 CKD or to ESRD identifies patients at especially high risk of cardiovascular adverse events who require extra vigilance and some of whom might benefit from early instigation of treatment based on guidelines for CKD 4 subjects [28]. However, there are currently no biomarkers to identify subjects who will progress. This has important clinical and public health implications, some of which have been reviewed (ibid). Consider, for example, the cost-benefit implications of treating all CKD Stage 3 subjects, given the proportion of the population with Stage 3 CKD is 15 times as large as that for Stage 4 CKD (Fig. 11.4).

Summary In conclusion, no biomarker convincingly predicts progression of CKD. However, in both general and high-risk populations, the risk of cardiovascular disease and cardiovascular-specific mortality increases with stage of kidney disease and loss of GFR. Importantly, from a population perspective, individuals with earlier stages of CKD are more likely to die of cardiovascular disease than to develop ESRD and require dialysis. The definition of CKD utilizing both eGFR and albuminuria appears sound [7]. Incorporation of eGFR or ACR or both into prediction models provides similar or better risk prediction than most of the traditional risk factors including blood pressure, lipids, and smoking [14] and supports the use of both measures to improve assessment of cardiovascular risk in the general population. The next section examines what is the role of other biomarkers of cardiac disease (both established and novel newer markers) in predicting

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

497

Figure 11.4  Cardiovascular outcomes with omission of kidney disease measures and traditional risk factors in a CKD population. The figure shows the difference in C statistics and 95% CIs for four cardiovascular outcomes with omission of kidney disease measures and traditional risk factors from a model with all risk factors in a CKD population. Systolic blood pressure includes two different coefficients for those with and without antihypertensive drugs. Total and HDL cholesterol refers to simultaneous omission of these two predictors. From Matsushita K, Coresh J, Sang Y, et al. Estimated glomerular filtration rate and albuminuria for prediction of cardiovascular outcomes: a collaborative meta-analysis of individual participant data. Lancet Diabetes Endocrinol 2015;3:514–525 [14].

cardiovascular outcomes in individuals with CKD compared to those with normal kidney function.

CARDIAC BIOMARKERS IN CKD Troponins Elevated troponins are the biomarkers of choice for cardiac injury. The cardiac troponin complex plays a critical role in the regulation of excitation– contraction coupling in the heart. The release of cardiac specific troponins into the circulation are highly specific for cardiac myocyte injury [29] and have become the biomarker of choice for the detection of cardiac injury but do not define the mechanism of injury [30]. With the use of highly sensitive cardiac troponin assays (hs-cTn), low hs-cTn concentrations can

498

Z.H. Endre and R.J. Walker

be detected in 50–90% of the healthy population, such that an elevation of hs-cTn above the 99th percentile provides a sensitive and diagnostic marker of cardiac injury [31]. Elevated cTn values in patients with acute ischemic presentations, with and without ECG changes, identify patients at higher risk for the development of cardiac events both during short- and long-term follow-up. The short-term outcome is closely associated with the higher acute thrombotic risk of the underlying unstable plaque, where as the long-term prognosis most likely reflects the higher prevalence of a more severe and complex coronary anatomy [29]. CKD is associated with elevated levels of cardiac troponins (both cTnI and cTnT) which in the absence of symptoms of an acute coronary syndrome most likely reflect subtle myocardial injury facilitated by the impact of the CKD [32,33]. Michos EA and coworkers have undertaken a systematic review and metaanalysis of troponin testing in patients with CKD without evidence of an acute coronary syndrome [34]. They analyzed papers that investigated troponin testing in patients on dialysis and for those not on dialysis. For those on dialysis, they identified 11 studies, measuring cTnT where they were able to metaanalyze the pool hazard ratios that had been adjusted for age and coronary artery disease or a risk equivalent. They found that there was a threefold increased risk of all-cause mortality [HR = 3.0 (95% CI: 2.4–4.3)]. For cardiovascular mortality, adjusted for age and coronary artery disease, there was a metaanalysis of the pooled hazards ratio for five studies which demonstrated a threefold increase in cardiovascular mortality [HR = 3.3 (95% CI: 1.8–5.4)] but there was significant heterogeneity between the studies. When they undertook a metaanalysis of the odds risk, there were nine comparable studies that demonstrated an odds risk of cardiovascular mortality of 4.3 (95% CI: 3.0–6.4). Similar results were seen for cTnI assays [34]. A smaller number of studies investigated the predictive value of troponins in patients with CKD not on dialysis [34]. There were two studies measuring TnT levels and all-cause mortality adjusted for age and coronary artery disease and the metaanalysis of the pooled hazards ratio demonstrated a 3.4 fold risk (95% CI: 1.1–11.0) and a metaanalysis of the odds risk from 5 studies demonstrated an OR = 3.0 ( 95% CI: 1.4–6.7).They concluded that an elevated baseline troponin was associated with a higher risk (approximately two- to threefold) for all-cause mortality, cardiovascular mortality, and major adverse cardiac events, for individuals on dialysis without suspected acute coronary syndrome. There was a similar trend for those with CKD not on dialysis but there were fewer studies available for analysis [34].

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

499

A few studies have compared cardiac troponins alone or in conjunction with other biomarkers to better clarify cardiovascular risk in CKD or using cardiac troponins to identify intervention strategies to reduce cardiovascular risk. In the Atherosclerosis Risk in the Community (ARIC) study, the associations of cTnT and NT-proBNP with CVD were independent of kidney function and actually stronger in those with CKD than in those without CKD [35]. Similar findings have reported in the Chronic Renal Insufficiency Cohort (CRIC) study which demonstrated that hsTnT and NT-proBNP were strongly associated with incident heart failure among a diverse cohort of individuals with mild to severe CKD. Elevations in these biomarkers may indicate subclinical changes in volume and myocardial stress that subsequently contribute to the development of clinical heart failure [36]. In CKD patients presenting with symptoms of an acute coronary syndrome, the clinical utility of cardiac troponins is less clear as there is limited data to establish the diagnostic cutoff values for patients with CKD or the required magnitude of change in serial values to define an acute myocardial event [33]. In a subgroup analysis of a prospective multicenter study investigating the utility of different cardiac troponin assays for patients presenting with symptoms of an acute coronary syndrome, for patients with CKD, where the cardiac troponins were greater than the 99th percentile (normal range), these supported the adjudicated diagnosis of an acute myocardial infarction in 45–80% cases [37]. However, the optimal receiver–operator characteristic curve derived cTn cutoff levels in patients with renal dysfunction were significantly higher compared with those in patients with normal renal function by a factor of 1.9–3.4 [37]. In another study investigating weekly variations of cardiac troponins in stable healthy controls and hemodialysis patients, the investigators suggested that a greater than 25% change for hs cTnT concentrations in the hemodialysis group was diagnostic with a low probability for a false positive result. Using change would reduce the need for diagnostic cutoffs for the hs cTnT assay [38,39].

Natriuretic Peptides Natriuretic peptides play an important role in volume homeostasis via vasodilation, natriuresis, and diuresis to protect the cardiovascular system from volume overload.There are three natriuretic peptides, atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), and C-type natriuretic peptide (CNP). ANP is produced and secreted from atria, BNP is predominantly synthesized in the cardiac ventricles with myocardial wall stress and stretch

500

Z.H. Endre and R.J. Walker

the major stimuli for release. CNP is released from the endothelium as well as the myocardium [40]. Natriuretic peptides, in particular BNP, are useful markers for the diagnosis and evaluation of patients with heart failure and volume overload as well as having prognostic value in a number of cardiovascular conditions in both the general population as well as in patients with CKD [40,41]. There have been few studies that have investigated the role of ANP as a biomarker of cardiovascular disease in patients with CKD. Initial studies investigated the role of ANP as a biomarker of volume status in dialysis patients [42–44]. These studies demonstrated that ANP concentrations were closely related to atrial volume and or cardiac filling pressures. However cardiac function was a major confounder in the interpretation of the ANP concentrations documented in patients with CKD. Zoccali and coworkers examined both the predictive value of both BNP and ANP for all-cause mortality and cardiovascular mortality [44]. They demonstrated that BNP was a better indicator of cardiovascular outcomes compared to ANP, and in a Cox’s model controlling for left ventricular mass and ejection fraction, only BNP remained an independent predictor of death [44]. Although C-natriuretic peptide has been linked to vascular and cardiac function, there have been no studies examining the predictive role of CNP for cardiovascular outcomes in patients with CKD. Brain Natriuretic Peptide Brain natriuretic peptide (BNP) arises from pre-pro BNP, which is synthesized in the cardiomyocytes and is initially cleaved to pro-BNP, which is further cleaved to the active BNP and the inactive N terminal pro BNP (NT proBNP). NT-proBNP is more stable with a longer half-life and is considered a better marker for heart failure and volume overload, as well as for hard outcomes of death and cardiovascular events (Fig. 11.3) [33]. BNP and NT-proBNP plasma concentrations are expressed as pmol/L or pg/ mL. The conversion factor for BNP is 1 pg/mL = 0.289 pmol/L. For NTproBNP the conversion is 1 pg/mL = 0.118 pmol/L [45] (Fig. 11.5). Although there has been some concern that natriuretic peptides are metabolized and cleared by the kidneys [45,46], a number of studies have demonstrated that left ventricular dysfunction is the major determinant of elevated BNP and NT-proBNP concentrations independent of the severity of kidney dysfunction [35,36,47]. In the CRIC study, elevated NT-proBNP levels were a strong predictor of incipient heart failure. When compared to individuals in the lowest quartile for NT-proBNP (<47.6 ng/mL)

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

501

Figure 11.5  Brain natriuretic peptides (BNP) in patients with CKD. In response to increased stretch or tension, left ventricular myocytes release BNP and N-terminal-proBNP (NT-pro-BNP) from precursors. BNP is an active molecule with a short plasma halflife and is degraded in the circulation by enzymatic action. NT-pro-BNP is the inactive form of BNP, with a longer half-life. It is primarily cleared by the kidneys. Reduced eGFR correlates to a greater extent with elevated plasma NT-pro-BNP than to BNP levels. Increased NT-pro-BNP/BNP ratio correlates with advancing CKD stages, especially if the eGFR is 30 mL/min per 1.73 m2. However, both BNP and NT-pro-BNP are associated with surrogate and hard clinical outcomes in asymptomatic patients with CKD. From Colbert G, Jain N, de Lemos JA, Hedayati SS. Utility of traditional circulating and imaging-based cardiac biomarkers in patients with predialysis CKD. CJASN 2015;10:515–529 [33].

participants in the highest quartile (>433.0 ng/mL) had a significantly higher rate of heart failure (HR 9.57: 95% CI: 4.40–20.53) [36].This association remained strong, even after the adjustment for a broad range of traditional and nontraditional cardiovascular risk factors The risk of heart failure with elevated NT-proBNP was similar in CKD patients in both those with incipient heart failure and those with normal cardiac function [36]. Similar

502

Z.H. Endre and R.J. Walker

findings have been reported in the Atherosclerosis Risk in the Community (ARIC) study, which demonstrated that cardiac markers NT-proBNP and cardiac troponin T were more sensitive for predicting cardiovascular disease in patients with CKD than kidney markers (cystatin C, β2 microglobulin, β-trace protein) or hsCRP [35]. Of interest, the prediction of cardiovascular disease in individuals with CKD by cTNT and NT-proBNP was considerably higher than that reported for traditional risk factors of total cholesterol, LDL-cholesterol and hsCRP [35,47,48]. This is not surprising in that both elevations in hs cTnT and NT-proBNP reflect subclinical changes in cardiac structure and function which are highly likely in patients with CKD [35]. For further review of the various studies reporting the strong association of NT-proBNP and cardiovascular events, see Table 2 in the well-summarized review by Colbert and coworkers [33]. For patients with end stage kidney disease on dialysis, (hemodialysis or peritoneal dialysis), BNP and NT-proBNP are associated with left ventricular structural and functional abnormalities, which are prevalent in almost all ESKD patients. In a comprehensive review by Wang and Lai [41] they commented that BNP and NT-proBNP have a limited role in assessing acute changes in extracellular volume and the diagnostic utility for BNP is to rule out systolic dysfunction and predict the presence of left ventricular hypertrophy.Whereas BNP and NT-proBNP consistently have been shown to have powerful prognostic value for mortality and cardiovascular death in both hemodialysis patients [49,50] and peritoneal dialysis patients [49], this association is independent of and well beyond that contributed by alterations in left ventricular mass and function [40,41,44,50]. With regard to clinical utility, both BNP and NT-proBNP can be used for prognosis in patients with CKD with respect to cardiovascular events and death.What is less clear is the role of NT-proBNP as a useful diagnostic test for acute congestive heart failure in patients with CKD [33]. In the PRIDE study, the investigators, using a higher cut off value for NT-proBNP of >1200 pg/mL (>142 pmol/L) for patients with an eGFR <60 mL/min per 1.73 m2, NT-proBNP was found to be a useful diagnostic tool, for acute decompensated heart failure with a sensitivity of 89% and specificity of 72% [51]. Further prospective studies are required to corroborate these findings.

C-Reactive Protein C-reactive protein (CRP) is an acute phase reactant and is measured as high-sensitivity C-reactive protein (hs-CRP). hs-CRP is an established marker of systemic inflammation as well as a predictor of future cardiac

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

503

events, in the general population and in patients with CKD [40,52–54]. A number of studies have reported a significant association between hs-CRP and all-cause mortality in ESKD patients. In a study of 399 hemodialysis patients, 46% of patients had an elevated concentration of hs-CRP, which was associated with a 2 year mortality rate of 44% [48]. In the majority of dialysis patients hs-CRP values can be variable related to a number of different inflammatory stimuli, and a number of authors have suggested that serial measurements of hs-CRP may be more meaningful, with only those individuals with a persistently elevated hs-CRP (>10 mg/L) at increased risk of death [52,53]. In CKD patients, similarly hs-CRP levels can be quite variable, but elevated levels of hs-CRP have been associated with a higher risk for cardiovascular events in a community study of elderly patients with CKD [55]. When compared to patients with low hs-CRP and no CKD, the adjusted hazard ratio for those individuals with a high hs-CRP (>2.1 mg/L) and CKD (eGFR < 60 mL/min) was 1.93 (95% CI: 1.45–2.89) [55]. In the Chronic Renal Insufficiency Cohort study, higher levels of hs-CRP were associated with left ventricular hypertrophy and impaired function [53]. Overall, hs-CRP may be useful as an additional assessment of cardiovascular risk, but is of less value if there is already established coronary artery disease. Also on its own, hs-CRP is less predictive of cardiovascular mortality than NT-BNP or cTnT levels [40]. Interestingly, the predictive power of ACR and eGFR either alone or together was substantially higher than that observed in a metaanalysis using hs-CRP as a predictor of cardiovascular outcomes [14,47].

Fibroblast Growth Factor 23 Components of the CKD–mineral bone disorder (CKD–MBD), phosphate, parathyroid hormone, vitamin D, and more recently fibroblast growth factor 23 (FGF-23) have been independently associated with the risk of all-cause mortality in patients with CKD and on dialysis, as well as heart failure, cardiovascular events and death in the general population [40]. FGF-23 levels increase in CKD as a physiological adaptation to maintain neutral phosphate balance and normal serum phosphate concentrations in the setting of reduced phosphate excretion. In the Chronic Renal Insufficiency Cohort (CRIC) study, elevated FGF-23 was independently associated with a greater risk of death and this mortality risk increased by each quartile of measured FGF-23 with the hazard ratio (HR) for the second quartile 1.3 (95% CI: 0.8–2.2), for the third quartile HR 2.0 (95% CI: 1.2–3.3) and for

504

Z.H. Endre and R.J. Walker

the fourth quartile HR 3.0 (95% CI: 1.85–5.1) [56]. In addition there was a graded and independent association with heart failure and atherosclerotic events [57]. A more recent publication from the CRIC study demonstrated that FGF-23 is associated with left ventricular hypertrophy but does not promote vascular calcification. In contrast, higher serum phosphate levels were associated with the prevalence and severity of coronary artery calcium on CT coronary angiography, even after adjustment for FGF23 [58]. The association of FGF-23 with total mortality was robust across all eGFR strata, supporting the notion that FGF-23 mediates total mortality independently of bone mineral disease and other CKD-mediated comorbidities [59]. In the Cardiovascular Health Study of older people living in the community [60], they examined the strength of the association of FGF-23 with kidney function at the baseline visit and the extent to which associations of FGF-23 with longitudinal outcomes were dependent on the presence of concomitant CKD. Modest decrements in eGFR and elevations in urine Albumin Creatinine Ratios are each independently associated with higher FGF-23. In the CKD group (n = 1,128), the highest FGF-23 quartile had adjusted hazards ratios (HR) of 1.87 (95% confidence interval, CI: 1.47– 2.38) for all-cause death, 1.94 (95% CI: 1.32–2.83) for incident HF, and 1.49 (95% CI: 1.02–2.18) for incident CVD events compared with the lowest quartile. These results support the findings in the CRIC study demonstrating an association between, CKD, elevated FGF-23 concentrations and heart failure as well as for all-cause mortality [60]. Similar findings have also been reported in the Homocysteine in Kidney and End Stage Renal Disease (HOST) which also demonstrated in patients with advanced CKD, FGF-23 was strongly and independently associated with all-cause mortality, cardiovascular events, and initiation of chronic dialysis [61]. Thus measurement of FGF-23 may contribute to our understanding of mechanisms linking CKD to cardiovascular disease but the clinical utility for FGF 23 is yet to be established. FGF-23 is not routinely available in clinical practice, nor has the predictive value been evaluated against other well-documented cardiovascular risk assessment tools. Further studies are required to compare FGF-23 to other biomarkers either alone or in combination to identify its’ true predictive value [62]. A number of newer markers have been identified that are associated with the increased cardiovascular risk in the setting of CD and possibly relate to the overall increased proinflammatory state of progressive CKD. Most of these newer markers have been identified from prospective longitudinal cohort studies where these has been a single baseline sample taken.

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

505

As such, causality cannot be demonstrated merely an association with increased mortality and cardiovascular risk. The markers discussed later require further clinical studies to establish a true predictive role and as such are not yet in routine clinical use.

Placental Growth Factor Placental Growth Factor (PlGF) is a pleiotropic cytokine, similar to vascular endothelial growth factor (VEGF), which binds to its membrane bound receptor—fms tyrosine kinase 1 (Flt-1) to stimulate angiogenesis. PlGF expression is increased in atherosclerotic plaques with activation of monocytes and macrophages and subsequent increased release of inflammatory angiogenic mediators, increasing the risk of plaque rupture [63]. In the Novel Assessment of Risk management for Atherosclerotic disease in CKD (NARA-CKD) cohort of 1351 individuals followed for a median of 3 years, Matsui and coworkers demonstrated that serum PlGF concentrations are strongly and independently associated with all-cause mortality and cardiovascular events in CKD patients [63]. Using the median of the lowest quartile for serum PlGF (7.4 pg/mL) as the reference, in adjusted analyses the hazard ratio for mortality and cardiovascular risk increased in each successive quartile of serum PlGF level; hazard ratios (HRs) [95% confidence intervals (95% CIs)] for mortality and cardiovascular risk, respectively, were 1.59 (0.83–3.16) and 1.55 (0.92–2.66) for the second quartile, 2.97 (1.67– 5.59) and 3.39 (2.20–5.41) for the third quartile, and 3.87 (2.24–7.08) and 8.42 (5.54–13.3) for the fourth quartile (see figure) [63]. With stratification for age, sex, traditional risk factors, and eGFR, serum PlGF still remained an independent risk factor for all-cause mortality and cardiovascular events in this cohort of patients with CKD [63]. There are some limitations to the interpretation of these results. It is implied that all the cardiovascular events are related to ischemic cardiac events and plaque rupture. While this might hold true for acute coronary syndromes, it is not necessarily a valid assumption for sudden cardiac death and heart failure in patients with CKD, where there are other contributing factors. Likewise, it is a single time point measurement at baseline in an observational study. It is not clear what happens to PlGF over time and with progression of CKD (Fig. 11.6).

Galectin 3 Galectin 3 is a member of the multifunctional galectin family, which is ubiquitously expressed in the heart, the kidney, blood vessels, and macrophages

506

Z.H. Endre and R.J. Walker

Figure 11.6  Multivariable-adjusted hazard ratios (HRs) for all-cause mortality and for cardiovascular events over the distribution of serum placental growth factor (PlGF) concentrations. The median of the lowest quartile of PlGF concentration served as the reference group (7.4 pg/mL). The Cox model was adjusted for age, sex, diabetes, hypertension, dyslipidemia, obesity, smoking, previous coronary artery disease, previous stroke, CKD stage, hemoglobin, serum albumin, calcium, phosphorus, C-reactive protein, HbA1c, LDL cholesterol, HDL cholesterol, triglycerides, and use of RAS blockers, calcium channel blockers, β-blockers, mineralo-corticoid receptor antagonists, lipidlowering agents, diuretics, and antiplatelet agents. The solid lines show HRs for mortality and cardiovascular events. Dashed lines indicate 95% CIs. From Matsui M, Uemura S, Takeda Y, et al. Placental growth factor as a predictor of cardiovascular events in patients with CKD from the NARA-CKD study. J Am Soc Nephrol 2015;26:2871–2881 [63].

and plays a role in tissue fibrosis, immunity, and the inflammatory response. In a large combined cohort of patients from the 4D (German Diabetes Dialysis) and the Ludwigshafen Risk and Cardiovascular Health (LURIC) study, circulating galectin-3 concentrations increase in parallel with decreasing kidney function and were markedly elevated in dialysis patients with type 2 diabetes mellitus [64]. In this study, Galectin-3 was significantly associated with clinical end points, but only in patients with reduced kidney function [64]. It remains unclear if galectin-3 adds to eGFR as a marker of cardiovascular risk.

Neutrophil Gelatinase-Associated Lipocalin Neutrophil Gelatinase-Associated Lipocalin (NGAL) is a member of the lipocalin iron-carrying protein family expressed in a variety of tissues including neutrophils, kidney, and cardiomyocytes. At steady state plasma and urinary NGAL concentrations are negligible. Plasma and urinary levels are increased in both acute kidney injury and CKD. Elevated plasma NGAL is a strong predictor of acute kidney injury, the subsequent need for acute renal

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

507

replacement therapy and in-hospital mortality [65]. Urinary NGAL concentrations are increased with CKD [66], however the actual discrimination value of urinary NGAL as a predictor of CKD progression is marginal [67]. The role of NGAL as a marker of cardiovascular risk is much more limited, with only a few studies demonstrating a small predictive benefit of elevated serum NGAL concentrations for cardiac events. In a study of 673 patients presenting with a ST segment elevation myocardial infarction (STEMI), and who underwent percutaneous coronary intervention, a serum NGAL was measured on hospital admission [68]. These patients all had a normal serum creatinine on admission. A serum NGAL > 110 pg/mL was associated with a 1 year mortality of 20% suggesting that serum NGAL measurement can enhance the prognostic score for 1 year mortality rates in patients presenting with STEMI [68]. Liu and coworkers examined the predictive value of urinary NGAL in the Chronic Renal Insufficiency Cohort study. This demonstrated that higher urinary NGAL concentrations adjusted hazard ratio (HR) for the highest (>49 ng/mL) versus the lowest (<6.9 ng/mL) quartile: 1.83 (95% CI: 1.20–2.81), was associated independently with ischemic atherosclerotic events but not heart failure events or death [69]. The different predictive values of serum NGAL or urinary NGAL for cardiovascular events particularly in the setting of CKD where the NGAL values are already elevated, needs clarification as to what they are actually measuring. For example, are the elevated urinary NGAL concentrations reflecting changes in kidney function in the setting of acute or chronic cardiovascular events or actually reflecting the cardiac events themselves. Given that serum NGAL concentrations can be derived from cardiac tissue, neutrophils as well as kidneys, makes it less clear as to the role of NGAL as a predictive tool. At present, studies are based on a single baseline sample. Further studies are required to determine the importance of changes over time, as well as the actual renal clearance of serum NGAL.

SUMMARY CKD is an established independent risk factor for cardiovascular disease. Likewise cardiovascular disease is the major cause of morbidity and mortality in patients with CKD [1]. There are a number of biomarkers of both cardiac and kidney in origin that have predictive value in identifying the increased risk of cardiovascular events in individuals with CKD. Biomarkers can be useful both to add diagnosis of cardiovascular events and as a prognostic tool with respect to long-term outcomes. To date, the addition

508

Z.H. Endre and R.J. Walker

of eGFR and urinary albumin/creatinine ratio (ACR) significantly improves the discrimination of cardiovascular outcomes beyond traditional risk factors in general populations, with a greater improvement coming from ACR than eGFR and more evident for cardiovascular mortality and heart failure, than coronary disease and stroke [14]. The associations of eGFR and albuminuria with cardiovascular mortality are even stronger in those patients with CKD [10] and the predictive power of ACR and eGFR either alone or together is substantially higher than that observed in a metaanalysis using hs-CRP as a predictor of cardiovascular outcomes [14,47]. While newer biomarkers have been demonstrated to have predictive value for cardiovascular outcomes, these require larger clinical studies to really identify their diagnostic and prognostic roles in clinical management.

REFERENCES [1] Go AS, Chertow GM, Fan D, Mcculloch CE, Hsu C-Y. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004;351:1296–305. [2] Coresh J, Selvin E, Stevens LA, et al. Prevalence of chronic kidney disease in the United States. JAMA 2007;298:2038–47. [3] Sarnak MJ, Levey AS, Schoolwerth AC, et al. Kidney disease as a risk factor for development of cardiovascular disease: a statement from the American Heart Association Councils on Kidney in Cardiovascular Disease, High Blood Pressure Research, Clinical Cardiology, and Epidemiology and Prevention. Circulation 2003;108:2154–69. [4] Chadban SJ, Briganti EM, Kerr PG. Prevalence of kidney damage in Australian adults: the AusDiab kidney study. J Am Soc Nephrol 2003;14:S131–8. [5] Iburg KM,Voss T, Salomon J, Murray CJ. Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2015;386:743–800. [6] Levey AS, Beto JA, Coronado BE, Eknoyan G. Controlling the epidemic of cardiovascular disease in chronic renal disease: what do we know? What do we need to learn? Where do we go from here? Am J Kidney Dis 1998;32:853–906. [7] Gansevoort RT, Correa-Rotter R, Hemmelgarn BR, et al. Chronic kidney disease and cardiovascular risk: epidemiology, mechanisms, and prevention. Lancet 2013;382:339–52. [8] Ozrazgat-Bislanti T, Matthew H, Kent B, et al. Acute and chronic kidney disease and cardiovascular mortality after major surgery. Ann Surg 2016. (Epub ahead of print). [9] Fox CS, Matsushita K, Woodward M, Bilo H. Associations of kidney disease measures with mortality and end-stage renal disease in individuals with and without diabetes: a meta-analysis. Lancet 2012;380:1662–73. [10] Tonelli M, Wiebe N, Culleton B, et al. Chronic kidney disease and mortality risk: a systematic review. J Am Soc Nephrol 2006;17:2034–47. [11] Levey AS, Eckardt KU, Tsukamoto Y, et al. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2005;67:2089–100. [12] Levey AS, de Jong PE, Coresh J, et al. The definition, classification, and prognosis of chronic kidney disease: a KDIGO Controversies Conference report. Kidney Int 2011;80:17–28.

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

509

[13] Chronic Kidney Disease Prognosis Consortium. Association of estimated glomerular filtration rate and albuminuria with all-cause and cardiovascular mortality in general population cohorts: a collaborative meta-analysis. Lancet 2010;375:2073–81. [14] Matsushita K, Coresh J, Sang Y, et al. Estimated glomerular filtration rate and albuminuria for prediction of cardiovascular outcomes: a collaborative meta-analysis of individual participant data. Lancet Diabetes Endocrinol 2015;3:514–25. [15] Mahmoodi BK, Matsushita K, Woodward M, et al. Associations of kidney disease measures with mortality and end-stage renal disease in individuals with and without hypertension: a meta-analysis. Lancet 2012;380:1649–61. [16] Hallan SI, Matsushita K, Sang Y, et al. Age and association of kidney measures with mortality and end-stage renal disease. JAMA 2012;308:2349–60. [17] Glassock RJ, Rule AD. The implications of anatomical and functional changes of the aging kidney: with an emphasis on the glomeruli. Kidney Int 2012;82:270–7. [18] Gaber AO, Moore LW, Aloia TA, et al. Cross-sectional and case-control analyses of the association of kidney function staging with adverse postoperative outcomes in general and vascular surgery. Ann Surg 2013;258:169–77. [19] Huber M, Ozrazgat-Baslanti T, Thottakkara P, et al. Mortality and cost of acute and chronic kidney disease after vascular surgery. Ann Vasc Surg 2015;30. 72.e1-2–81.e1-2. [20] Al-Aly Z, Zeringue A, Fu J, et al. Rate of kidney function decline associates with mortality. J Am Soc Nephrol 2010;21:1961–9. [21] Kovesdy CP. Rate of kidney function decline associates with increased risk of death. J Am Soc Nephrol 2010;21:1814–6. [22] Turin TC, James M, Ravani P, et al. Proteinuria and rate of change in kidney function in a community-based population. J Am Soc Nephrol 2013;24:1661–7. [23] Kovesdy CP, Coresh J, Ballew SH. Past decline versus current eGFR and subsequent ESRD risk. J Am Soc Nephrol 2016;27:2447–55. [24] Foster MC, Coresh J, Bonventre JV, et al. Urinary biomarkers and risk of ESRD in the atherosclerosis risk in communities study. CJASN 2015;10:1956–63. [25] Levin A, Rigatto C, Brendan B, et al. Cohort profile: Canadian study of prediction of death, dialysis and interim cardiovascular events (CanPREDDICT). BMC Nephrol 2013;14P 121. [26] Levin A, Rigatto C, Barrett B, et al. Biomarkers of inflammation, fibrosis, cardiac stretch and injury predict death but not renal replacement therapy at 1 year in a Canadian chronic kidney disease cohort. Nephrol Dial Transplant 2014;29:1037–47. [27] Hu B, Gadegbeku C, Lipkowitz MS, et al. Kidney function can improve in patients with hypertensive CKD. J Am Soc Nephrol 2012;23:706–13. [28] Sud M, Tangri N, Pintilie M, Levey AS, Naimark DMJ. Progression to Stage 4 chronic kidney disease and death, acute kidney injury and hospitalization risk: a retrospective cohort study. Nephrol Dial Transplant 2015. [29] Thygesen K, Mair J, Katus H, et al. Recommendations for the use of cardiac troponin measurement in acute cardiac care. Eur Heart J 2010;31:2197–204. [30] Babuin L, Jaffe AS.Troponin: the biomarker of choice for the detection of cardiac injury. CMAJ 2005;173:1191–202. [31] Reichlin T, Hochholzer W, Bassetti S, et al. Early diagnosis of myocardial infarction with sensitive cardiac troponin assays. N Engl J Med 2009;361:858–67. [32] Mallamaci F,Tripepi G.Value of troponin T as a screening test for left ventricular hypertrophy in CKD. Am J Kidney Dis 2013;61:689–91. [33] Colbert G, Jain N, de Lemos JA, Hedayati SS. Utility of traditional circulating and imaging-based cardiac biomarkers in patients with predialysis CKD. CJASN 2015;10:515–29. [34] Michos ED,Wilson LM,Yeh H-C, et al. Prognostic value of cardiac troponin in patients with chronic kidney disease without suspected acute coronary syndrome: a systematic review and meta-analysis. Ann Intern Med 2014;161:491–501.

510

Z.H. Endre and R.J. Walker

[35] Matsushita K, Sang Y, Ballew SH, et al. Cardiac and kidney markers for cardiovascular prediction in individuals with chronic kidney disease: the Atherosclerosis Risk in Communities Study. Arterio Thromb Vasc Biol 2014;34:1770–7. [36] Bansal N, Hyre Anderson A,Yang W, et al. High-sensitivity troponin T and N-terminal pro-B-type natriuretic peptide (NT-proBNP) and risk of incident heart failure in patients with CKD: the Chronic Renal Insufficiency Cohort (CRIC) Study. J Am Soc Nephrol 2015;26:946–56. [37] Twerenbold R, Wildi K, Jaeger C, et al. Optimal cutoff levels of more sensitive cardiac troponin assays for the early diagnosis of myocardial infarction in patients with renal dysfunction. Circulation 2015;131:2041–50. [38] Aakre KM, Røraas T, Petersen PH, et al. Weekly and 90-minute biological variations in cardiac troponin T and cardiac troponin I in hemodialysis patients and healthy controls. Clin Chem 2014;60:838–47. [39] Khalili H, de Lemos JA. What constitutes a relevant change in high-sensitivity troponin values over serial measurement? Clin Chem 2014;60:803–5. [40] D’Marco L, Bellasi A, Raggi P. Cardiovascular biomarkers in chronic kidney disease: state of current research and clinical applicability. Dis Markers 2015;2015:586569–616. [41] Wang AY-M, Lai K-N. Use of cardiac biomarkers in end-stage renal disease. J Am Soc Nephrol 2008;19:1643–52. [42] Zoccali C, Ciccarelli M, Mallamaci F, et al. Effect of ultrafiltration on plasma concentrations of atrial natriuretic peptide in haemodialysis patients. Nephrol Dial Transplant 1986;1:188–91. [43] Yoshihara F, Horio T, Nakamura S, et al. Adrenomedullin reflects cardiac dysfunction, excessive blood volume, and inflammation in hemodialysis patients. Kidney Int 2005;68:1355–63. [44] Zoccali C, Mallamaci F, Benedetto FA, et al. Cardiac natriuretic peptides are related to left ventricular mass and function and predict mortality in dialysis patients. J Am Soc Nephrol 2001;12:1508–15. [45] Almirez R, Protter AA. Clearance of human brain natriuretic peptide in rabbits; effect of the kidney, the natriuretic peptide clearance receptor, and peptidase activity. J Pharmacol Exp Ther 1999;289:976–80. [46] Takami Y, Horio T, Iwashima Y, Takiuchi S. Diagnostic and prognostic value of plasma brain natriuretic peptide in non-dialysis-dependent CRF. Am J Kidney Dis 2004;44:420–8. [47] Kaptoge S, Di Angelantonio E, Emerging Risk Factors Collaboration. et al. Creactive protein, fibrinogen, and cardiovascular disease prediction. N Engl J Med 2012;367:1310–20. [48] Apple FS, Murakami MM, Pearce LA, Herzog CA. Multi-biomarker risk stratification of N-terminal pro-B-type natriuretic peptide, high-sensitivity C-reactive protein, and cardiac troponin T and I in end-stage renal disease for all-cause death. Clin Chem 2004;50:2279–85. [49] Wang AY-M, Lam CW-K, Yu C-M, et al. N-terminal pro-brain natriuretic peptide: an independent risk predictor of cardiovascular congestion, mortality, and adverse cardiovascular outcomes in chronic peritoneal dialysis patients. J Am Soc Nephrol 2007;18:321–30. [50] Satyan S, Light RP, Agarwal R. Relationships of N-terminal pro-B-natriuretic peptide and cardiac troponin T to left ventricular mass and function and mortality in asymptomatic hemodialysis patients. Am J Kidney Dis 2007;50:1009–19. [51] Anwaruddin S, Lloyd-Jones DM, Baggish A, et al. Renal function, congestive heart failure, and amino-terminal pro-brain natriuretic peptide measurement: results from the ProBNP Investigation of Dyspnea in the Emergency Department (PRIDE) Study. J Amer Coll Cardiol 2006;47:91–7.

Biomarkers of Cardiovascular Risk in Chronic Kidney Disease

511

[52] Nascimento MM, Pecoits-Filho R, Qureshi AR, et al. The prognostic impact of fluctuating levels of C-reactive protein in Brazilian haemodialysis patients: a prospective study. Nephrol Dial Transplant 2004;19:2803–9. [53] Snaedal S, Heimbürger O, Qureshi AR, et al. Comorbidity and acute clinical events as determinants of C-reactive protein variation in hemodialysis patients: implications for patient survival. Am J Kidney Dis 2009;53:1024–33. [54] Gupta J, Dominic EA, Fink JC, et al. Association between Inflammation and Cardiac Geometry in Chronic Kidney Disease: Findings from the CRIC Study. PLoS One 2015;10:e0124772. [55] Jalal D, Chonchol M, Etgen T, Sander D. C-reactive protein as a predictor of cardiovascular events in elderly patients with chronic kidney disease. J Nephrol 2012;25:719–25. [56] Isakova T, Xie H,Yang W, et al. Fibroblast growth factor 23 and risks of mortality and endstage renal disease in patients with chronic kidney disease. JAMA 2011;305:2432–9. [57] Isakova T,Wahl P,Vargas GS, et al. Fibroblast growth factor 23 is elevated before parathyroid hormone and phosphate in chronic kidney disease. Kidney Int 2011;79:1370–8. [58] Scialla JJ, Lau WL, Reilly MP, et al. Fibroblast growth factor 23 is not associated with and does not induce arterial calcification. Kidney Int 2013;83:1159–68. [59] Denker M, Boyle S, Anderson AH, et al. Chronic Renal Insufficiency Cohort Study (CRIC): overview and summary of selected findings. CJASN 2015;10:2073–83. [60] Ix JH, Katz R, Kestenbaum BR, et al. Fibroblast growth factor-23 and death, heart failure, and cardiovascular events in community-living individuals: CHS (Cardiovascular Health Study). J Amer Coll Cardiol 2012;60:200–7. [61] Kendrick J, Cheung AK, Kaufman JS, et al. FGF-23 associates with death, cardiovascular events, and initiation of chronic dialysis. J Am Soc Nephrol 2011;22:1913–22. [62] Larsson TE. The role of FGF-23 in CKD–MBD and cardiovascular disease: friend or foe? Nephrol Dial Transplant 2010;25:1376–81. [63] Matsui M, Uemura S, Takeda Y, et al. Placental growth factor as a predictor of cardiovascular events in patients with CKD from the NARA-CKD study. J Am Soc Nephrol 2015;26:2871–81. [64] Drechsler C, Delgado G, Wanner C, et al. Galectin-3, renal function, and clinical outcomes: results from the LURIC and 4D studies. J Am Soc Nephrol 2015;26:2213–21. [65] Haase M, Bellomo R, Devarajan P, Schlattmann P, Haase-Fielitz A. NGAL Meta-analysis Investigator Group. Accuracy of neutrophil gelatinase-associated lipocalin (NGAL) in diagnosis and prognosis in acute kidney injury: a systematic review and meta-analysis. Am J Kidney Dis 2009;54:1012–24. [66] Fassett RG, Venuthurupalli SK, Gobe GC, Coombes JS, Cooper MA, Hoy WE. Biomarkers in chronic kidney disease: a review. Kidney Int 2011;80:806–21. [67] Liu KD, Yang W, Anderson AH, et al. Urine neutrophil gelatinase-associated lipocalin levels do not improve risk prediction of progressive chronic kidney disease. Kidney Int 2013;83:909–14. [68] Helanova K, Littnerova S, Kubena P, et al. Prognostic impact of neutrophil gelatinaseassociated lipocalin and B-type natriuretic in patients with ST-elevation myocardial infarction treated by primary PCI: a prospective observational cohort study. BMJ Open 2015;5:e006872. [69] Liu KD,Yang W, Go AS, et al. Urine neutrophil gelatinase-associated lipocalin and risk of cardiovascular disease and death in CKD: results from the Chronic Renal Insufficiency Cohort (CRIC) Study. Am J Kidney Dis 2015;65:267–74.