Baroreflex gain measured by the phenylephrine method and by the Robbe spectral technique in post-myocardial infarction patients

In this study we compared the baroreflex gain (BRG) measured by the spectral method proposed by Robbe with the standard phenylephrine method in a population of 43 post-infarctuated patients (mean age ± SD: 52±8). The agreement between the two techniques was assessed computing the bias (i.e. the mean value of the difference) and the 95% limits of agreement, whereas the degree of linear association was assessed by correlation analysis. The Pearson correlation coefficient was 0.67. The estimated limits of agreement were: -10.5÷8.5 ms/mmHg with a bias of -1 ms/mmHg (N.S.). As the difference between the two measurements was found to depend on their mean value (which is an estimate of the true BRG), we performed the same analysis on the percentage difference. The corresponding limits of agreement were: -93.6÷88.5% with a bias of -2.6%. According to these results, the difference between the two measurements can be almost as large as the BRG of the patient. Our conclusion is that the spectral technique can substitute the pharmacological measurement for screening purposes but not in the assessment of patient prognosis where a depressed phenylephrine BRG (<3 ms/mmHg) has an important independent prognostic value.