Seismocardiographic changes associated with obstruction of coronary blood flow during balloon angioplasty

Abstract

Seismocardiography, a new noninvasive technique, detects low-frequency cardiac vibrations on the chest wall during ventricular contraction and during both early and late ventricular filling. To evaluate the ability of seismocardiography to detect ischemia caused by decreased coronary blood flow, 35 patients were studied during coronary angioplasty. Seismocardiograms and electrocardiograms were recorded twice at baseline, with the catheter across the lesion before first inflation (n = 15), every 30 seconds during the first inflation, 1 and 2 minutes after the first inflation and ≥5 minutes after the final inflation. For comparison, sequential seismocardiograms were also obtained from 15 healthy volunteers. Electrocardiograms were blindly scored for ST change from baseline (0 = none, 1 = 0.5 mm ST depression, 2 = ≥1.0 mm ST depression, 3 = ST elevation). Seismocardiograms were blindly scored for change from baseline (0 = none, 1 = mild, 2 = moderate, 3 = marked) for both the systolic and diastolic waves. The average maximal systolic seismocardiographic score was 2.5 ± 0.8 for patients who had undergone angioplasty and 1.0 ± 0.9 for volunteers (p < 0.001). The average maximal diastolic seismocardiographic score was 2.3 ± 0.8 for angioplasty patients and 0.7 ± 0.9 for volunteers (p < 0.001). The percentage of angioplasty patients with electrocardiographic, systolic and diastolic seismocardiographic scores ≥2 was, respectively: 0, 11 and 14% at second baseline; 23, 67 and 53% with catheter across the lesion; 44, 75 and 59% after 30 seconds of inflation; 42, 71 and 61% after 60 seconds of inflation; 23, 74 and 61% after 1 minute of deflation; and 0, 71 and 47% 5 minutes after final inflation. The percentage of volunteers with seismocardiographic score ≥2 never exceeded 13% for systole or 14% for diastole at any time point. Thus, both electrocardiograms and seismocardiograms changed rapidly during coronary angioplasty, whereas Seismocardiograms did not change in volunteers. Most changes developed when the catheter crossed the lesion, before balloon inflation. These findings are consistent with the hypothesis that the seismocardiographic changes are due to ischemic changes in ventricular wall motion or compliance. Seismocardiographic changes persisted longer than electrocardiographic changes, possibly due to delayed recovery in ventricular function.