Heart failure (HF) complicates chronic kidney disease (CKD) despite improving medical therapy for co-morbidities like hypertension. We hypothesized that vascular stiffness, (aortic pulse wave velocity [PWV]), predicts incident HF in well controlled CKD participants free of HF at enrollment.

We performed aortic (i.e. carotid-femoral) PWV measurements (Sphygmocor) in 1889 participants enrolled in the multicenter Chronic Renal Insufficiency Cohort (CRIC) at their second year follow up visit and followed them prospectively (mean follow-up time of 1.4 years) for incident hospitalized HF events occurring before December 31, 2007 adjudicated by two independent Investigators. Mean age was 60 years, 44% women, 42% black, 6% Hispanic and 47% were Caucasian with mean(SD) blood pressure of 127(22)/70(13) mmHg (more than half of the participants were prescribed ≥3 antihypertensives) and a mean estimated glomerular filtration rate of 42.6 mL/min/1.73m2. Mean(SD) BMI was 31(6.6) kg/m2. Mean(SD) aortic PWV (adjusted for waist circumference) was 9.5(3.1) m/sec. There were 54 incident HF events. The unadjusted Cox proportional hazard ratio [95% CI] for incident HF for those with PWV >10m/s was 4.71 (2.59, 8.54; censored for deaths / withdrawals); adjusting for age, race, gender and mean arterial pressure it was 3.82[2.02,7.25]. We observed that for each 1 m/sec increase in aortic PWV there was a 11% (unadjusted) and 9% (after adjustment) increase in the hazard for HF.

In summary, aortic PWV independently predicted incident HF in CKD patients with well-controlled BP. Measures to better understand and improve PWV may compliment standard BP control to prevent HF in CKD.