Underestimation of Vasodilator Effects of Nitroglycerin by Upper Limb Blood Pressure
Abstract To determine why upper limb blood pressure measurement underestimates the vasodilator effects of nitroglycerin on lowering ascending aortic systolic pressure, we studied 24 patients (58±11 years, mean±SD). Ascending aortic pressure and radial artery pulse calibrated by cuff blood pressure measurement at the brachial artery were recorded simultaneously before and 5 minutes after sublingual administration of 0.3 mg nitroglycerin. Waves were analyzed by a signal processor, and the fourth derivative wave was used to find the early (S1) and late (S2) systolic shoulders (S1 corresponds to the second zero crossing and S2 to the third zero crossing). Before nitroglycerin administration, maximal systolic pressure in the ascending aorta (141±21 mm Hg) coincided with the late systolic peak in all patients, and in most patients (21 of 24) maximal systolic pressure in the radial artery (140±19 mm Hg) coincided with the early systolic peak. Maximal systolic pressure decreased more in the ascending aorta than in the radial artery (22±13 and 11±11 mm Hg, respectively; P<.001). However, the reduction in the shoulder of late systolic pressure in the radial artery (24±13 mm Hg) clearly indicated the reduction in maximal systolic pressure (late systolic peak) in the ascending aorta. The augmentation index of the ratio of the height of late systolic pressure to early systolic pressure fell proportionally (r=.74, P<.001) in the radial artery (from 0.88±0.13 to 0.60±0.11) and in the ascending aorta (from 1.57±0.25 to 1.26±0.24), which indicated the reduction in late systolic pressures. These data suggest that ordinary peripheral artery blood pressure measurement underestimates the vasodilation effects of nitroglycerin on the ascending aorta because without arterial pulse measurement it cannot show changes in late systolic pressure.
Arterial pulse examinations have been used since ancient times for the diagnosis of various diseases and physiological conditions.1 Discrepancies of ascending aortic and upper limb systolic pressures after vasodilator agents have been reported2 3 with regard to differences in maximal systolic pressure. These discrepancies were caused by the use of maximal systolic pressure for the assessment of differences in the pressure wave. Data on pulse waves in the literature show a reduction in the late systolic shoulders of upper limb pressure waves, but no numerical data concerning the late systolic shoulders have yet been introduced. The most important point is that systolic pressure has two components. We usually refer to systolic pressure as maximal systolic pressure. However, the maximal systolic pressures in the ascending aorta and brachial artery are completely different in terms of systolic timing. The ascending aortic pressure wave can be divided into two components at an early systolic inflection point that coincides with the ascending aortic flow peak.4 The early systolic component is mainly caused by ejection from the left ventricle and the late systolic component by the reflection wave from the periphery. These two components are transmitted to peripheral arteries undergoing certain modifications according to location and show characteristic waveforms. Maximal systolic pressure in the ascending aorta usually corresponds to the late systolic peak in adults, but maximal systolic pressure in the brachial artery corresponds to the early systolic peak. This is the main reason for the discrepancy between ascending aortic and brachial arterial systolic pressures.
The purpose of this study was to compare the differences of the early and late systolic pressures in the ascending aorta and radial artery with a newly developed automatic detection system and explore the reasons why peripheral blood pressure measurements underestimate the vasodilator effects of nitroglycerin on lowering ascending aortic systolic pressure.
Twenty-four patients (21 men, 3 women) with a mean age of 58±11 years (±SD) who underwent diagnostic cardiac catheterization were studied: 12 with myocardial infarction, 9 with angina pectoris, and 3 with chest pain syndrome without organic cardiac abnormalities. Ascending aortic pressure was measured by a microtip catheter (SVPC 684D, Millar Instruments) and radial arterial pulse by multisensor tonometry (JENTOW-7000, Nippon Colin Co); both were recorded simultaneously on an FM tape recorder (XR-30, TEAC Co) and a chart recorder (MICOR, Siemens). Recordings were made before and after sublingual administration of 0.3 mg nitroglycerin. Waves were analyzed by a signal processor (San-ei 7T-18A, Nippon Electric Co) with a digital derivative method. Waveforms of the ascending aorta and radial artery were digitized at 0.005-second intervals and plotted from the first to the fourth derivatives of pressure waves. For smoothing of the waves, smoothing points of five points and five times averaging was used. The final frequency response of the derivative method was 8.16 Hz. The algorithm first found the peak and nadir of the pressure wave and then determined maximal systolic and diastolic pressures. If the slope of the fourth derivative wave corresponding to maximal systolic pressure is positive, this point is assigned as the time of the late systolic peak on the raw wave, and the second zero crossing from above to below is then used to determine the time of the early systolic shoulder on the raw wave. If the slope of the fourth derivative wave corresponding to maximal systolic pressure is negative, this point is assigned as the time of the early systolic peak on the raw wave, and the third zero crossing from below to above is then used to determine the time of the late systolic shoulder on the raw wave. The augmentation index was defined as the ratio of the height of the late systolic shoulder/peak to that of the early systolic shoulder/peak in the pulse. All procedures were approved by the ethics committee of Tokyo Medical College Hospital. Informed consent was obtained from all patients.
Results of pressures and times before and after nitroglycerin are shown in the Table⇓, with descriptive statistics and the results of paired t tests. Maximal systolic pressure in the radial artery coincided with the early systolic peak in 21 patients before nitroglycerin and all 24 patients after nitroglycerin. Maximal systolic pressure decreased from 141±21 to 118±16 mm Hg in the ascending aorta and from 140±19 to 128±17 mm Hg in the radial artery. The reduction in maximal systolic pressure was greater in the ascending aorta (22±13 mm Hg) than the radial artery (11±11 mm Hg). The reduction in the radial arterial late systolic shoulder/peak was 24±13 mm Hg, and there was no statistical difference compared with the reduction in maximal systolic pressure in the ascending aorta. Fig 1⇓ shows a sample tracing that reveals a marked reduction in maximal systolic pressure or late systolic peak in the ascending aorta, but maximal systolic pressure or early systolic peak in the radial artery did not change, with a marked reduction in the late systolic shoulder. Fig 2⇓ shows how to identify the shoulder/peak. Fig 3⇓ shows the relationship between the ascending aortic augmentation index (AoP-AI) and radial arterial augmentation index (RaP-AI). AoP-AI increased with increasing RaP-AI (r=.74, P<.001).
It is obvious that the marked difference in the ascending aortic systolic pressure waveform and upper limb pressure waveform is one of the major causes of the underestimation of vasodilator effects of nitroglycerin by upper limb blood pressure measurements. Although maximal systolic pressure usually coincides with the late systolic peak in the ascending aorta, it usually coincides with the first systolic peak in the radial artery. Since nitroglycerin reduces late systolic pressure in the ascending aorta by reducing the reflection wave,6 the reduction in the second systolic peak is the same as the reduction in maximal systolic pressure in the ascending aorta. However, in the brachial and radial arteries, this reduction, even when marked, does not appear as a reduction in maximal systolic pressure because the second systolic peak occurs between the first systolic peak (maximal systolic pressure) and diastolic pressure. We can easily recognize such differences of pressure waveforms when we record each pulse. Recently developed arterial tonometry4 7 provides us with stable and accurate pressure waveforms. However, it is not easy to detect the early and late systolic shoulders, especially when the shoulder fails to show a distinct inflection point. Murgo et al5 were the first to describe the concept of the augmentation index, which they referred to as the reflection return point in the ascending aorta. Kelly et al4 introduced the term augmentation index. The augmentation index we used was the ratio of the height of the late systolic shoulder/peak to the early systolic shoulder/peak in the pulse, which was simpler than that in previous reports but essentially the same. Kelly et al4 pointed out that a means for objective recognition of the early systolic shoulder was needed. They determined the early systolic inflection point using the fourth derivative wave of the second zero crossing, which coincides with peak flow velocity. This is very useful for detection of the early systolic shoulder, but for analysis of the upper limb pulse waveform it is necessary to clarify the late systolic shoulder, and for this it is important to know where the late systolic inflection point is. The fourth derivative wave has four major segments, and the third zero crossing from below to above the fourth derivative indicates the late systolic inflection point. We developed the automatic detection of the late systolic shoulder using the third zero crossing from below to above the fourth derivative wave, which corresponds to the late systolic shoulder. The limitation of this method is the lower frequency response of 8.16 Hz, which is not used in the determination of the maximal pressure rise of the wave, such as maximal dP/dt. Gain loss occurs with the lower frequency derivative method. However, automatic analysis of pressure waveforms requires stable waveforms. The major characteristics of the radial pulse are included within 4 to 5 Hz,8 whereas the dicrotic notch contour can be characterized within 8 Hz. The frequency response (8.16 Hz) of this system was considered to keep the characteristic frequencies of radial pulse and maintain the dicrotic notch while eliminating the increased signal-to-noise ratio at higher frequencies. Recently, frequency analyses of upper limb pressure waves for determination of the ascending aortic pressure wave have been developed by Karamanoglu et al8 as a transfer function method, which is one of the best estimates based on wave analysis. However, it is possible to compare the ascending aortic pressure waveform and the upper limb pressure waveform in a time-domain relationship. This may be the one approach for exploring the importance of the recognition of the two systolic components and contributing to the understanding of not only the discrepancies between ascending aortic pressure and upper limb pressure but also the effects of vasodilators on lowering ascending aortic systolic pressure with the use of measurement of the reduction in the second systolic component in the radial artery.
- Received January 31, 1995.
- Revision received February 21, 1995.
- Accepted March 30, 1995.
O’Rourke MF, Kelly RP, Avolio AP. History. In: The Arterial Pulse. Philadelphia, Pa: Lea & Febiger; 1992:3-14.
Kelly RP, Gibbs HH, Morgan JJ, Daley JE, Mang K, Avolio AP, O’Rourke MF. Nitroglycerin has a more favorable effect on left ventricular afterload than apparent from measurement of pressure in a peripheral artery. Eur Heart J. 1990;11:138-144.
Kelly RP, Hayward CS, Avolio AP, O’Rourke MF. Non-invasive determination of age-related changes in the human arterial pulse. Circulation. 1989;80:1652-1659.
Murgo JP, Westerhof N, Giolma JP, Altobelli SA. Aortic input impedance in normal man: relationships to pressure wave forms. Circulation. 1980;62:105-116.
Yaginuma T, Takazawa K, O’Rourke MF. Effects of nitroglycerin on ascending aortic impedance and on left ventricular load in normal subjects and patients with coronary artery disease. In: O’Rourke MF, Safar ME, Dzau VJ, eds. Arterial Vasodilation: Mechanisms and Therapy. London, UK: Edward Arnold; 1993:78-90.
Karamanoglu M, O’Rourke MF, Avolio AP, Kelly RP. An analysis of the relationship between central aortic and peripheral upper limb pressure waves in man. Eur Heart J. 1993;14:160-167.