American Association for Clinical Chemistry
Better health through laboratory medicine
July 2007 Clinical Laboratory News: Diagnostic Profiles

July 2007: Volume 33, Number 7

BNP Levels in ESRD Seem Unrelated
to Certain Clinical Parameters

Values of BNP are elevated in patients with end-stage renal disease (ESRD) and decline after each dialysis session, but changes in BNP do not correlate with changes in clinical parameters related to volume status, including ultrafiltration volume and blood pressure, according to recent research (American Heart Journal 2007; 153: 244e1–244e5). To explore the possibility that BNP may have some utility in elucidating the volume status and its implications on cardiac function in ESRD patients, researchers from Veterans Affairs Medical Center (San Diego) and University of California, San Diego studied data from 39 ESRD patients who underwent hemodialysis three times a week for at least 30 days at the San Diego Veterans Healthcare System. Samples were collected at the start and end of each of three consecutive sessions, while left ventricular ejection fractions were obtained from echocardiograms performed within 1 year of enrollment. The first session was the dialysis session, which occurred after a 72-hour interdialytic period, while the second and third sessions were after a 48-hour period. The researchers measured plasma volume changes in a subset of 13 patients. Pre-and postdialysis BNP levels for each of the three sessions were 434 and 343 pg/mL, 3347 and 231 pg/mL, and 249 and 202 pg/mL, respectively. The values for body weights were 82.6 and 78.6 kg, 81.5 and 78.2 kg, and 81.5 and 78.3 kg, respectively. The values of mean systolic blood pressures were 150 and 134 mm Hg, 142 and 134 mm Hg, and 142 and 131 mm Hg, respectively. The respective values for mean diastolic blood pressures were 81 and 70 mm Hg, 74 and 72.1 mm Hg, and 76 and 72 mm Hg. They found no correlation between changes in intradialytic BNP values and other measured parameters. Plasma volume changed minimally during dialysis. The lack of correlation between reduced BNP levels and clinical parameters related to volume status may be due to lack of plasma volume change during dialysis, the fact that post-BNP levels were sampled immediately after the end of the session and lacked adequate time for achieving a steady state of intravascular volume, or that changes in BNP are related to the dialysis procedure itself, the researchers postulated.

BNP with Heart Failure Predicts In-Hospital Mortality

Because BNP levels have predicted outcomes in patients with acute decompensated heart failure independent of other clinical and laboratory values in a large cohort, hospitals should consider adding BNP assays to assessment of patients presenting with the condition, according to researchers from University of California, Los Angeles, Los Angeles Medicine Center, The Cleveland Clinic Foundation, Brigham & Women’s Hospital in Boston, and Scios, Inc. (Mountain View, Calif.) (JACC 2007; 49: 1943–1950). While prior studies had examined use of natriuretic peptide levels to diagnose heart failure only in small populations, the current study aimed to assess BNP levels within 24 hours of presentation in 48,629 episodes entered in the Acute Decompensated Heart Failure National Registry (ADHERE) between April 2003 and December 2004. Researchers assessed in-hospital mortality by BNP quartiles and in 19,544 patients with reduced left ventricular systolic function and 18,164 with preserved ventricular systolic function using chi-square and logistic regression models. The quartiles of BNP were <430 pg/ml (Q1), 430-839 pg/ml (Q2), 840-1,729 pg/ml (Q3), and ≥1,730 pg/ml. In 3.3% of the cohort, levels were <100 pg/ml. Researchers found a near-linear relationship between BNP quartiles and in-hospital mortality, with rates of 1.9% for Q1, 2.8% for Q2, 3.8% for Q3, and 6.0% for Q4. After adjustment for age, sex, systolic blood pressure, blood urea nitrogen, creatinine, sodium, pulse, and dyspnea at rest, the BNP quartiles remained highly predictive of mortality, with Q4 versus Q1 adjusted odds ratio at 2.23 (95% CI, 1.91-2.62). The BNP quartiles independently predicted mortality in patients with reduced and preserved systolic function. Researchers noted that current American College of Cardiology and American Heart Association heart failure guidelines recommend considering BNP solely for diagnostic purposes and only in patients with uncertain diagnoses, but these results suggest that BNP levels could be useful to stratify risk.

EPCA-2 Shows Promise as Prostate Cancer Assay

EPCA may be a novel prostate cancer biomarker that offers high sensitivity and specificity and accurately differentiates between men whose disease is organ-confined and those whose cancer has spread, according to recent research. (Urology 2007; 69: 714–720) Investigators from the Brady Urological Institute at Johns Hopkins University in Baltimore, Md., studied serum samples from 385 men from five groups: those with PSA <2.5 ng/mL, PSA levels ≥2.5ng/mL, with negative biopsy findings, those with benign prostatic hyplasia, organ-confined prostate cancer, non-organ confined disease, and prostate cancer with PSA levels < 2.5 ng/mL. In addition, a diverse group of controls was assessed with an enzyme-linked immunosorbent assay to detect an epitope of the EPCA-w protein, EPCA-2.22. Using a cutoff of 30 ng/mL, the EPCA-2.22 assay had a specificity of 92% for healthy men and men with benign prostatic hyperplasia (95% CI, 85%–96%) and 94% sensitivity (95% CI, 93%–99%) for overall prostate cancer. The specificity of PSA in these selected groups was 65% (95% CI, 55%–75%). EPCA-2.22 had high accuracy in differentiating between localized and extra capsular disease (area under the curve, 0.89, 95% CI, 0.82–0.97), in contrast to PSA (area under the curve, 0.62, 95% CI, 0.50–0.75). The researchers noted that their results did not determine the efficacy of EPCA 2.22 in screening men for prostate cancer or differentiate between men with organ-confined versus extracapsular disease, but that they suggest the potential of the marker to do so. Researchers called for a study that would specifically answer that question.

Study Links Uric Acid and Metabolic Syndrome

Serum concentrations of uric acid are strongly associated with the prevalence of the metabolic syndrome and several of its components among U.S. adolescents, according to a recent research (Circulation 2007; 115: 2526–2532). Researchers from the CDC, University of Rochester, Rochester, N.Y., University of British Columbia, Vancouver, Canada, and Brigham and Women’s Hospital in Boston performed a cross-sectional analysis of 1,370 adolescents of both sexes ages 12 to 17 using data from NHANES 1999–2002 to find participants who had three of five criteria for the metabolic syndrome (lipid triglycerides ≥110 mg/dL , HDL cholesterol ≥40 mg/dL, waist circumference ≥90th percentile for sex, glucose concentration ≥100 mg/dL, and systolic or diastolic blood pressure ≥90th percentile for age, height, and sex) and to understand the association between the syndrome and subjects’ uric acid levels. The prevalence of metabolic syndrome was <1% among participants in the lowest quartile of serum concentration of uric acid, 3.7% in the second quartile, 103% in the third quartile, and 21.1% in the highest quartile. Compared with the lowest two quartiles of uric acid together (≤291.5 μmol/L), the odds ratios were 5.80 (95% CI, 7.78–28.11) for those in the third quartile (2.91.5– ≤339 μmol/L or > 4.9 to ≤mg/dL) and 14.79 (95% CI, 7.78 –28.11) for those in the top quartile (>339 μmol/L), after adjustment for age, sex, race or ethnicity, and concentrations of CRP. Starting with the lowest quartile of concentration of uric acid, mean concentrations of serum insulin were 66.2, 66.7, 79.9, and 90.0 μmol/L for ascending quartiles, respectively. The researchers postulated that the association between metabolic syndrome and high concentrations of uric acid are due to the inverse relationship between renal clearance of urate and degree of insulin resistance or to increases in children’s consumption of fructose, which has been shown to elevate concentrations of uric acid that in turn impairs endothelial function. Impaired oxidative phosphorylation involved in the metabolic syndrome may increase systemic adenosine concentrations by increasing the intracellular concentrations of coenzyme A esters of long-chain fatty-acids, the researchers added.

Non-HDL Cholesterol Suggested to
Replace Lipoprotein Measurements

Although measurements of total cholesterol, LDL cholesterol, and HDL cholesterol are widely recommended, non HDL-cholesterol is as good or better than these markers at identifying subclinical atherosclerosis in young adults, according to recent research (The American Journal of Cardiology 2007; 100: 64–68). To augment scant comparative data on utility of non-HDL cholesterol in detecting increased carotid intima-media thickness (IMT) as a measurement of subclinical atherosclerosis, researchers from Tulane University (New Orleans, La.) and Wake Forest University (Winston-Salem, N.C.) examined data from 1,203 African American and white subjects enrolled in the Bogalusa Heart Study. The community-based study examined the natural history of CVD beginning in childhood. Subjects of the present study were 71% white, 43% male and between ages 24 and 43. Researchers used multivariate logistic regression analysis to estimate the odds ratio and 95% CI of each lipoprotein measurement—HDL cholesterol, LDL cholesterol, triglycerides, apolipoprotein B, apolipoprotein A-I,, ratio of total cholesterol to HDL cholesterol, and ratio of apo B to apo A-I—as a risk factor for carotid IMT in the top decile versus the lower 9 deciles specific for age, race, and sex. Only non-HDL cholesterol, total cholesterol/HDL cholesterol, and apo B emerged as significant correlates, with respective odds ratios of 1.75 (95% CI, 1.10-2.78), 2.02 (95% CI, 1.27–3.19), and 2.13 (95% CI, 1.38–3.29), after adjusting for body mass index, systolic blood pressure, and other lipoprotein measurements. In terms of discriminative value between predictive models, as assessed by under the receiver character curve analysis, none of the other lipoprotein measurements was significantly better than non-HDL cholesterol in detecting increased carotid IMT. C-value for non-HDL cholesterol (0.73) was similar to those for LDL cholesterol (0.76), total cholesterol/HDL cholesterol (0.72), apoB/apo A-1 (0.71), and HDL cholesterol (0.70), but significantly higher than that for apo A-I (0.69), triglycerides (0.64), and apo B (0.64). The researchers noted that their findings in young adult subjects are consistent with those from previous studies of middle-aged and older subjects and indicate the superiority of non-HDL cholesterol over other lipoprotein measurements. HDL-cholesterol measurement as a primary CAD risk-assessment lipoprotein measurement offers several advantages over the others, including its lower cost, no need for patient fasting, standardization, and clinically useful—but as yet unvalidated—cutpoints for the U.S. population.