Prospective Study of Neuropeptide Y as an Adverse Cardiovascular Risk Factor in End-Stage Renal Disease

Study Protocol

The study protocol confirmed to the ethical guidelines of our institutions, and informed consent was obtained from each participant. All studies were performed during a nondialysis day, between 8 a.m. and 1 p.m.

Study Cohort

Two hundred seventy-seven patients (154 male and 123 female patients, 226 receiving hemodialysis and 51 receiving chronic ambulatory peritoneal dialysis [CAPD]) who had been receiving regular dialysis treatment for at least 6 mo (average duration of regular dialysis treatment, 67 mo; range, 6 to 98 mo) without history of congestive heart failure and without concomitant illnesses were eligible for the study. The main demographic and clinical characteristics of the patients included in the study are listed in Table 1. The prevalence of diabetes mellitus in this cohort was 16% (44 of 277 patients).

Hemodialysis patients were virtually anuric (24 h urine volume <200 ml/d) and were being treated three times a week with standard bicarbonate dialysis (Na 138 mmol/L, HCO₃ 35 mmol/L, K 1.5 mmol/L, Ca 1.25 mmol/L, Mg 0.75 mmol/L) and cellulose or semisynthetic membranes (dialysis filters surface area, 1.1 to 1.7 m²). The average urea Kt/V in these patients was 1.21 ± 0.27. Patients receiving CAPD were all on a four-exchange-a-day schedule and used standard dialysis bags. The average weekly Kt/V in these patients was 1.67 ± 0.32. Thirty-four patients had a history of myocardial infarction, and 79 had clinical and electroencephalogram evidence of cardiac ischemia. Twenty-five patients had had a stroke, and 29 had experienced transient ischemic attacks. One hundred thirteen patients were habitual smokers (22 ± 17 cigarettes a day). One hundred forty-four patients received erythropoietin therapy. One hundred sixteen patients were being treated with antihypertensive drugs (82 receiving monotherapy with angiotensin-converting enzyme inhibitors, AT-1 antagonists, calcium channel blockers, and α- and β-blockers, and 34 on double- or triple-drug therapy with various combinations of these drugs).

Follow-up

After the initial assessment, patients were followed up for an average of 34 mo (range, 0.2 to 52.0 mo). During the follow-up, CV events (electrocardiogram-documented anginal episodes and myocardial infarction, heart failure, electrocardiogram-documented arrhythmia, transient ischemic attacks, stroke, and other thrombotic events) and death were accurately recorded. Each death was reviewed and assigned an underlying cause by a panel of five physicians. As a part of the review process, all available medical information about each death was collected. This information always included study and hospitalization records. In instances where the patient died when he or she was not hospitalized, family members were interviewed by telephone to better ascertain the circumstances surrounding death.

Laboratory Measurements

Blood sampling was performed between 8.00 a.m. and 10.00 a.m. in steady-state conditions, during a nondialysis day for hemodialysis patients and at empty abdomen for CAPD patients. Samples for the measurement of plasma NPY were taken after 20 to 30 min of quiet resting in semirecumbent position. Plasma NPY was measured after extraction by Sep-Pac C18 cartridges (Waters Associates, Milford, MA) by means of a commercially available RIA kit (Peninsula Laboratories, Merseyside, England). The intra- and interassay variations were 10% and 13%, respectively. The upper limit of the normal range of plasma NPY in a group of 53 healthy control subjects was 8 pmol/L (average: 2.5 ± 2.0 pmol/L). The plasma concentrations of NE and epinephrine (E) were measured by a RIA kit (Amicyl-test; Immunological Laboratories, Hamburg, Germany). Serum lipids, albumin, calcium, phosphate, and hemoglobin measurements were made by standard methods in the routine clinical laboratory. C-reactive protein (CRP) and plasma homocysteine were measured by previously described methods (16).

Statistical Analyses

Data are expressed as mean ± SD (normally distributed data), geometric mean ± SD (nonnormally distributed data), or as percentage of frequency, and comparisons between groups were made by one-way ANOVA or χ² test, as appropriate.

Probability of survival was analyzed by the Kaplan-Meier survival analysis and by the multivariate Cox proportional-hazard model. For patients who experienced multiple events, survival analysis was restricted to the first event. There was no missing value for plasma NPY. There were only 30 missing values (hemoglobin, cholesterol, albumin, calcium, phosphate, and CRP, five per each variable). Missing values for these covariates were set to the mean value. To identify the independent prognostic power of plasma NPY for all-cause mortality and fatal and nonfatal CV events, we started with saturated Cox proportional-hazard models and included all covariates that were associated (P < 0.10) with plasma NPY (Table 1) as well as covariates that were related (P < 0.10) to all-cause death or CV outcomes at univariate analysis. To obtain parsimonious models, variables in the final models were selected by a stepwise approach. By means of this strategy, we constructed models of adequate statistical power (at least 19 events for each variable in the final model).

The assumption of linearity for the Cox proportional-hazard models was examined through visual inspection, and no violation of proportional hazard was found. Hazard ratios (HR) and their 95% confidence intervals (95% CIs) were calculated by the estimated regression coefficients and their standard errors in the Cox regression analysis. All calculations were made by a standard statistical software package (SPSS for Windows, version 9.0.1).

NPY, NE and E, and Plasma Urea

Plasma NPY was directly related to plasma NE (Figure 1). A similar but weaker relationship was also found between plasma NPY and plasma E (r = 0.17, P = 0.005). NPY was largely unrelated to plasma urea (r = −0.06, P = 0.32).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Relationship between plasma norepinephrine (NE) and plasma neuropeptide Y (NPY). Because NE and NPY showed positively skewed distributions, these data were log transformed (log₁₀). Data are Pearson product moment correlation coefficient and P value.

Survival Analysis

No patient was lost to follow-up. One hundred sixty-one fatal and nonfatal CV events occurred in 113 patients; 112 patients died, 66 of them (59%) of CV causes (Table 2). Plasma NPY failed to predict survival either in the Kaplan-Meier analysis (log-rank test, 1.75; P = 0.19) or in a multivariate Cox regression model (P = 0.29). Multivariate analysis found that age (P < 0.001), previous CV events (P = 0.001), male gender (P = 0.002), CRP (P = 0.003), diabetes (P = 0.01), and serum albumin (P = 0.02) were significantly associated with all-cause mortality.

In the Kaplan-Meyer analysis, the CV event-free survival was progressively and significantly lower from the first tertile of plasma NPY onward (log-rank test, 9.64; P = 0.002) (Figure 2). In a Cox regression model that tested all covariates associated to incident CV events as well as factors associated to plasma NPY (Table 1) by univariate analysis, plasma NPY was found to be an independent predictor of CV events (HR [10 pmol/L increase in plasma NPY], 1.32; 95% CI, 1.09 to 1.60; P = 0.004) (Table 3, model 1). This relationship was unaffected by treatment modality (hemodialysis versus CAPD) (Table 3). Plasma NPY maintained its predictive power for CV events (HR [10 pmol/L increase in plasma NPY], 1.25; 95% CI, 1.02 to 1.54; P = 0.03) in a statistical model including plasma NE (Table 3, model 2). In this model, plasma NE failed to predict CV outcomes (P = 0.11), and it became a significant predictor of fatal and nonfatal CV events (HR [1 nmol/L increase in plasma NE], 1.07; 95% CI, 1.01 to 1.13; P = 0.01) when plasma NPY was excluded from the model.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Kaplan-Meier cardiovascular event-free survival curves in patients divided into three tertiles of plasma neuropeptide Y (NPY). First tertile of plasma NPY, <7.41 pmol/L; second tertile, 7.41 to 18.26 pmol/L; third tertile, >18.26 pmol/L.

NPY in ESRD

It is well demonstrated that nerves in the gut are an important source of NPY in humans (1). NPY has a biphasic disappearance from plasma, and the corresponding half-lives are 4 to 6 min and 20 to 40 min (23). The metabolism of NPY in patients with renal failure has not been studied, but it seems likely that enzyme (peptidase) activities (24) that degrade this neuromediator are altered in renal failure (25). NPY was measured in several studies in patients with ESRD and found to be mildly (6,26⇓) to markedly (12,13,27,28⇓⇓⇓) elevated. The pathophysiological implications of increased NPY in these patients are unclear because NPY was reported to be related to arterial pressure (26,29⇓) and to fluid overload (26) in some studies, whereas no such relationships were observed in the other studies (6,12,13,27,28⇓⇓⇓⇓). Furthermore, no consistent evidence has emerged that this peptide participates in CV regulation during dialysis treatment (27,30⇓). Overall, these studies dealt with a fairly small number of patients (8 to 25 patients), which precluded our studying the relationship between this substance and clinically important outcome measures. In this study, we found that NPY was unrelated to arterial pressure, but patients with high NPY (third tertile) had faster heart rates and were more likely to be receiving antihypertensive treatment. The possibility that NPY may be involved in CV complications is suggested by our recent observation that high NPY is strongly associated to left ventricular concentric hypertrophy in patients with ESRD (15).

NPY, Survival, and Incident CV Events

Approximately 39% of patients in our study had had one or more CV events during a 3-yr follow-up, and the mean death rate was 12% a year, which is analogous to that typically observed in European dialysis registries (31). Such a high CV morbidity and mortality in part depends on the fact that traditional (i.e., Framingham) risk factors are highly prevalent in the dialysis population. Interestingly, in this study, NPY was independently related to survival and CV outcomes independently of traditional risk factors (32) as well as of emerging risk factors, such as CRP (33,34⇓) and homocysteine (16,35⇓), and risk factors peculiar to ESRD, such as anemia (36) and hyperphosphatemia (37).

Sympathetic overactivity is an established trigger of CV structural alterations such as left ventricular hypertrophy and arterial remodeling (38), and we have recently reported that high NE in patients with ESRD is associated to left ventricular concentric hypertrophy (39) and that it predicts shorter survival and adverse CV outcomes (40). Because NPY is coreleased with NE, the link between this peptide and survival and incident CV events may well represent an epiphenomenon of sympathetic overactivity. In this study, NPY predicted incident CV events independently of plasma NE (Table 3, model 2). It is therefore possible that this neuropeptide is, at least in part, an event trigger independent of NE. NPY is present in sympathetic innervation of all parts of the conduction system of the heart but also in nerve fibers in the heart that do not represent sympathetic fibers (41). Mechanistic and intervention studies are required to answer this important question.

Plasma NPY predicts CV complications in ESRD. This phenomenon likely depends on the fact that this neuropeptide reflects sympathetic activity, but the notion that high NPY per se might be an event trigger cannot be excluded. Controlled trials with antiadrenergic drugs and with NPY antagonists (42) are needed to establish whether interference with the sympathetic system, with NPY, or both may reduce the high CV morbidity and mortality of dialysis patients. In this regard, it is of interest that high flux dialysis reduces plasma concentration of NPY (43).