BACKGROUND

Increased aortic stiffness markers--aortic pulse wave velocity (PWV) and augmentation index (AIx)--have emerged as powerful predictors of survival in haemodialysis (HD). Various and often contradictory abnormalities of endothelium-dependent (ED) and endothelium-independent (EID) vasomotor function, have been described in dialysis subjects, pre- and post-dialysis, using methods that are difficult to export to the clinical setting or to large prospective trials assessing their relevance. Therefore, we determined the influence of a HD session on PWV and the ED and EID vascular reactivity, employing pulse wave analysis (PWA) of the aortic waveforms, combined with provocative pharmacological stimuli known to reduce wave reflection.

METHODS

PWV and aortic AIx (difference between the first and second systolic peak on the aortic pressure waveform divided by the pulse wave height) were determined from PWA of arterial waveforms recorded by applanation tonometry using a SphygmoCor device in 41 HD (20 males, age 41.8 years) and in 20 controls with essential hypertension (10 males, age 43.6 years). ED and EID vascular reactivity were assessed by changes in AIx following inhaled salbutamol and sublingual nitroglycerin (GTN), respectively, pre- and post-dialysis session. Echocardiography was performed in all patients, pre-HD and before the PWV recordings.

RESULTS

Pre-HD AIx (27.9+/-11.9%) was significantly higher compared with hypertensive patients with normal renal function (16.5+/-17%, P<0.05). Dialysis significantly reduced AIx to 18.2+/-18.3% (P<0.05 compared with pre-HD AIx), a level comparable with non-renal subjects (P = NS). Overall, PWV increased following HD to 7.89+/-2.09 m/s (P = 0.004 vs pre-HD, 6.34+/-1.32 m/s in essential hypertensive patients, P<0.05); however, a 19.1% increase was seen in 29 subjects and a 9.1% decrease in the remaining 12 subjects, both P<0.05. In HD patients, either pre- or post-HD, the EID vascular reactivity is significantly greater than the ED vasodilatation elicited by a beta 2-agonist. Moreover, when compared with hypertensive patients with normal renal function, the dialysis session only improved the EID abnormality (post-HD GTN AIx(HD) = -20.8+/-22.9% vs post-GTN AIx(hypertensive) = -14.2+/-5.7%, P = NS), while it had a non-additive effect on the ED response. A smaller response to a GTN challenge was associated with a greater left ventricular mass: r =-0.42, P = 0.007. In contrast, a diminished response to a beta 2-agonist did not represent a marker for cardiac abnormalities.

CONCLUSIONS

The HD session acutely restores EID but not ED vasomotor function comparable with essential hypertensive patients. Pulse-wave analysis methodology, combined with provocative pharmacological testing may be used to unveil subsets of patients with more severe cardiac structural abnormalities.