Author + information
- H. William Strauss, MD⁎ (, )
- Heiko Schoder, MD and
- Heinrich R. Schelbert, MD
- ↵⁎Memorial Sloan-Kettering Cancer Center, Nuclear Medicine Service, 1275 York Avenue, ROOM S212, New York, New York 10021
We applaud the additional measurements made by Dr. Gewirtz et al. to support their hypothesis that 4-[18F]-tetraphenylphosphonium (18F-TPP) can be used to measure relative, but not absolute, myocardial perfusion. The investigators analyzed a total of 112 samples to verify the relationship of microsphere determined–perfusion to 18F-TPP activity (1). The additional data support the results reported in the paper, namely that 18F-TPP cannot be used as an indicator of absolute myocardial blood flow.
Because these results were unexpected, we asked a consultant, Dr. Heinrich R. Schelbert, to independently review the data. Dr. Schelbert asks whether this conclusion is universally valid. In their reply, the authors present a data plot indicating first-pass extraction fractions (E) of 20% to 40% for flows (F) ranging from 0.4 to 2.0 ml/min/g. Yet, the experimental approach, as adopted in the current investigation from Di Rocco et al. (2), might underestimate E. In the Di Rocco study (2), the nonlinear regression equations between F and the radiotracer net uptake in canine myocardium predict a flow of 0.76 ml/min/g, an E value of only 0.32. This value is substantially lower than the E of 0.66 reported in another study (3) and more consistent with observations in human myocardium. Second, the inverse relationship between E and F, as shown in Figure 1C in the letter by Gewirtz et al., offers a means for quantifying F, provided that the relationship between E and F is constant between subjects. There is, however, considerable scatter of the data around the regression line. Inspection of the data suggests several “families” of data points. If each of these data families were obtained from different animals, it might indicate considerable interanimal variations in E (or the net retention of tracer in myocardium). If the tracer retention depends on the mitochondrial membrane potential, where a potential-dependent forward transport of radiotracer from cytosol into mitochondria competes with flow-dependent back diffusion of radiotracer, is it possible that the observed variability resulted from interanimal differences in mitochondrial membrane potentials that ranged from about −65 to −112 mV, and thus varied considerably between animals, as shown in Figure 1 of Gurm et al. (1)? Importantly, did the experimental study conditions—including anesthetics, positive inotropic drugs, and vasodilator agents—account for this variability? If this variability does not exist under normal conditions in animals and in humans, and E for 18F-TTP would be more stable, it might thus enable quantitative measurements of myocardial blood flow.
Despite the very low extraction, the myocardial images in Figure 6 of Gurm et al. (1) have a very good myocardial-to-background ratio, and excellent contrast between regions of ischemia and normal myocardium. The image quality is particularly surprising because the positron emission tomography imaging device was manufactured more than 20 years ago.
As a result, we must conclude that the additional analysis supports the comments of Dr. Gewirtz et al. that in these experiments 18F-TPP distribution does not reflect absolute myocardial perfusion. However, viewing the flow data as “families” of curves, and considering the quality of the myocardial images as well as analysis of the normalized data, the tracer may be very useful in the clinic.
- American College of Cardiology Foundation