Based on our previous studies of left ventricular function, four serial radionuclide scans (at shock onset, and at three, seven, and ten days following shock) were planned for each patient. In the present study, at least two radionuclide ventriculography studies were performed on each patient. The first scan was done as near to the onset of shock as technically feasible, always within the first 24 hours. The final study in survivors was done when they were hemodynamically stable and not receiving any vasopressors, usually 6 to 14 days after the onset of shock. The final study in nonsurvivors was done within 24 hours of death.
Using a portable gamma camera, radionuclide ventriculography was done at each patients bedside in the Intensive Care Unit after in vivo labeling of the patients erythrocytes with stannous pyrophosphate followed by 20 mCi of technetium-99m. The camera was positioned in a 35° left anterior oblique orientation with a 15° caudal tilt to isolate the left ventricle. Left ventricular and background regions of interest were labeled, and background-corrected left ventricular time-activity curves were generated from the image sequence. The left ventricular ejection fraction was calculated as left ventricular end-diastolic counts minus left ventricular end- systolic counts divided by left ventricular end-diastolic counts, all corrected for background. erectalis tablets
Right ventricular ejection fraction was calculated from an equilibrium scan using the two regions of interest method of Maddahi et al. In our laboratory, the two regions of interest method produced good interobserver and intraobserver reproducibility for measuring right ventricular ejection fraction. Each study was analyzed independently by two observers (M.M.P and K.E.M.), who determined right ventricular ejection fractions to be within 10 percent of each other on 89 percent of the studies. These observers also found changes in the same direction in all patients. Right ventricular end-diastolic and end-systolic regions of interest and a paraventricular background region of interest were drawn. The right ventricular ejection fraction was calculated from the same equation used for the left ventricle, substituting right ventricular for left ventricular counts.
Results are reported as the mean ± SEM. A paired sample t test was used to compare initial with final values. Spearman coefficients of rank correlation were used to assess the relation between right ventricular ejection fraction or end-diastolic volume index and pulmonary artery mean pressure or pulmonary vascular resistance index. A p value of less than 0.05 was considered statistically significant.