This study tested the hypothesis that the angle-dependent nature of ultrasonic backscatter can differentiate
nontransmural from transmural myocardial infarction. The relationship between the amplitude of integrated
backscatter (IB) and the angle of incidence was studied in vitro with normal canine myocardium (n = 5) and infarcted
myocardium 4 weeks after coronary ligation, including nontransmural (n = 5) and transmural infarcts (n = 2). A
wide-band transducer was used (center frequency = 4.0 MHz). IB (dB) from the mid-portion of the myocardium was
measured from a normal angle (0°) to 30° of incidence, by rotating a sample holder in 5° increments. Then the
angular dependency of IB was determined as the slop (dB/degree) of a regression line plotted from IB versus the angle
of incidence. IB was shown to be influenced by a function of the angle of incidence in all three groups. The amplitude
of IB in nontransmural and transmural myocardial infarction was significantly greater than in normal myocardium
over the range of angles studied (p < 0.01). However, IB at a normal angle could not distinguish nontransmural from
transmural myocardial infarction (-43.2 ± 2.3 vs. -40.2 ± 4.0 dB, mean ± SD, p = NS). It became obvious that the
angular dependency of IB could reliably distinguish nontransmural from transmural myocardial infarction (-0.189
± 0.032 vs. -0.121 ± 0.012 dB/degree, p < 0.01). Furthermore, the angular dependency of IB in normal myocardium
(-0.087 ± 0.024 dB/degree) was substantially smaller than in nontransmural (p < 0.01) or transmural myocardial
infarction (p < 0.05). This analytic technique was highly reproducible. In conclusion, the angular dependency
of IB allowed satisfactory differentiation between nontransmural and transmural myocardial infarction,
although the amplitude of IB at a normal angle could not distinguish between the two groups. These findings suggest
that quantitative analysis of the angle-dependence of ultrasonic backscatter offers considerable promise for cardiac
tissue characterization.