Diagnostic potential of circulating natriuretic peptides in chronic kidney disease.

BACKGROUND Measurement of natriuretic peptides, particularly brain natriuretic peptide (BNP) is an established method for the diagnosis of cardiovascular disorders, chiefly left ventricular (LV) dysfunction. The influence of renal function on the diagnostic utility of natriuretic peptides is unclear. METHODS We performed a cross-sectional study of 296 patients with renal disease but no history of cardiac disease using echocardiography to assess LV mass and function. Circulating levels of atrial natriuretic peptide (ANP) and BNP were also measured. RESULTS The incidence of LV hypertrophy increased with progressive renal dysfunction; from 39% in patients with near-normal renal function, to 80% in renal transplant patients. There was a negative correlation between both ANP and BNP, and glomerular filtration rate (GFR) (ANP: r = -0.28, P<0.001; BNP: r = -0.40, P<0.001). Serum ANP and BNP had sensitivity and specificity for LV hypertrophy of 39.9%, 87.4% (ANP) and 61.4%, 67.6% (BNP) respectively. Sensitivity and specificity for LV dysfunction was 77.2%, 32.4% (ANP) and 71.8%, 40.0% (BNP). Significant confounders in determining serum ANP were haemoglobin, beta blockade and albumin, while serum BNP levels were significantly confounded by GFR, albumin, haemoglobin, beta blockade and age. CONCLUSIONS Across a spectrum of renal dysfunction, GFR is a more important determinant of serum BNP than ventricular function, and several factors are predictors of natriuretic peptide levels. In chronic kidney disease, the use of natriuretic peptides to diagnose LV hypertrophy must be interpreted in light of these other factors. The use of these peptides in renal dysfunction to diagnose LV dysfunction may be of limited value.