Effect of high-performance liquid chromatography on plasma angiotensin II measurements in treated and untreated normotensive and hypertensive patients

Abstract

Background: Angiotensin II (Ang II) levels are normally very low in human plasma, approximately 5 pg/ml. They are usually measured by radioimmunoassay after extraction and concentration. An additional high-performance liquid chromatography (HPLC) step is reportedly necessary for accurate measurement but it is laborious and time-consuming, severely limiting the number of samples that can be assayed.

Objective: To investigate whether the HPLC step was necessary for measuring Ang II in human plasma samples in our laboratory using our own Ang II antiserum.

Design: Human plasma Ang II levels, measured with and without the HPLC step, were compared in two different studies. Since the action of renin is the rate-limiting step in the production of Ang II in plasma, the relationships of plasma renin activity (PRA) to Ang II levels measured with and without HPLC were also evaluated. In the first study, 108 blood samples were collected from 29 hypertensive patients during placebo or treatment with the Ang II antagonist BMS-186295. In the second study blood samples were collected from 12 normal subjects before and during beta-adrenergic blockade.

Results: In samples collected during angiotensin II antagonism, which predictably increased plasma Ang II levels, a highly significant relationship between the Ang II measurements with and without HPLC was found (y = 0.99x + 1.7; r = 0.97, P < 0.001). The y intercept of 1.7 pg/ml suggested that the nonspecific immunoreactivity was close to 2 pg/ml in samples assayed without the HPLC step. During beta-adrenergic blockade, which predictably suppressed plasma renin levels, highly significantly linear relationships between HPLC and non-HPLC Ang II measurements (y = 1.3x + 1.6; r = 0.93. P < 0.001, n = 16) and between non HPLC Ang II and PRA (y = 1.9x + 1.7; r = 0.73, P < 0.001, n = 108) were again found. The relationship between PRA and HPLC Ang II was also highly significant (y = 1.4x + 0.04; r = 0.92, P < 0.001, n = 16), but the y intercept was significantly lower (P < 0.001), approaching zero, indicating the removal of nonspecific immunoreactivity during the HPLC step.

Conclusions: These results demonstrate once more that, when using polyclonal antibody 182, the accuracy of the Ang II measurement in human plasma is improved by the inclusion of a HPLC step, especially for samples with Ang II levels in the normal-to-low range. They also show that plasma Ang II and PRA increase or decrease proportionally during treatment with Ang II antagonists or beta-adrenergic blockade, respectively.