Failure of normalized ventricular filling rates to predict true filling rates

Ventricular filling rates derived from radionuclide angiographic (RNA) time-activity curves are commonly expressed as normalized values. The assumption that normalized filling rates have a relationship to the actual filling rates was tested. RNA and contrast angiography were performed within 20 min of one another in 21 patients with widely disparate volumes. The RNA time-activity curve was converted from counts to milliliters by equating the contrast angiography derived end diastolic volume to the end diastolic count rate determined by RNA. Peak filling rates were normalized to end diastolic volume (EDV), stroke volume (SV, and peak ejection rate (ER). No significant correlation between the normalized filling rates and the true filling rate was found. Significant correlations were found between the EDV normalized filling rate and the EDV (r = -0.70) and the ejection fraction (r = 0.89). Normalized filling rates are dependent upon the normalizing variable and are not a pure measure of ventricular filling rates. As the technique of gated radionuclide angiography has matured, it has become apparent that there is more information in the time-activity curve than just the ejection fraction. The emptying rates, filling rates, time to peak emptying, and time to peak filling are parameters that are also available from the time-activity curve. Several authors have used this information to quantitate ventricular ejection and filling rates [1-6]. Since the contrast angiography literature would indicate that in some disease states ventricular filling is impaired [7,8], attempts have been made to identify impaired filling rates by radionuclide techniques. Using this method, decreased normalized filling rates have been found in groups of patients with coronary artery disease and it has been suggested that the observed decrease is due to impairment of active relaxation and/or to reduced compliance [2,3,5,6]. It has even been suggested that the decrease seen in the normalized filling rates may be a reflection of ischemia in the resting patient [6]. While these RNA derived parameters have been normalized to end diastolic volume by most authors, normalization to stroke volume and maximum ejection rate have also been suggested [9,10]. A possible rationale for normalization is that the activity measured over the ventricle is dependent upon many factors, including radionuclide dose, attenuation from the patient's chest wall, the specific type of collimator used, thickness of the crystal, and the window width.(ABSTRACT TRUNCATED AT 400 WORDS)