Dynamic Geometry During Srstole in Normally Functioning Left Ventricles Based on Thre*Dlmensional Studies with Technetiub99m lsonitrile Scans Shimon Rosenheck, MD, Henry Atlan, MD, PhD, Dan Sapoznikov, PhD, David Stone, MD, and Mervyn S. Gotsman, MD - T he left ventricular contraction during systole has the following 2 components: circumferential contraction and apex-to-base shortening.1 Apical morphology is difficult to assess by echocardiography*and contrast or nuclear ventriculography. Radioactive perfusion scans that image the moving myocardium in vivo are excel- lent methods of defining myocardium and its motion. The new isonitrile compounds labeled with technetium- 99m, which has high radiation intensity,senable precise imaging of the myocardium (which is superior to thalli- urn-2014)and allow gated single-photon emission com- puted tomography.S7In the present study, technetium- 99m isobutyl isonitrile gated single-photon emission computed tomography and computer-assisted 3-dimen- sional reconstructions of the left ventricle were used to describe the dynamics of the left ventricle during sys- tole in humans. Eight consecutive men who underwent technetium- 99m methylbutyl isonitrile examination were selected for the study. Four patients had significant coronary artery disease, underwent percutaneous transluminal coronary angioplasty without evidence of myocardial damage, had no history of myocardial infarction, and had a normal thallium-201 scan, follow-up coronary angiography and echocardiography. Four subjects had no structural heart disease, and had normal thallium- 201 scans and left ventricular junction. Six examina- tions were peeormed after a submaximal treadmill ex- ercise test, and methoxy isobutyl isonitrile labeled with 20 mCi of technetium-99m was injected after maxima.1 effort was achieved. After the injection, exercise was continued for 60 seconds. In other 8 examinations, me- From the Departmentsof Cardiology and Medical Biophysics, Hadas- sah University Hospital, Kiryat Hadassah, P.O.Box 12000, Jerusalem 91120, Israel. Manuscript receivedOctober 13,1992; revised manuscript received February 16, 1993, and accepted February 19. thoxy-isobutyl-isonitrile was injected with subjects at rest. Planar acquisition was begun 30 to 60 minutes after injection, with patients in the supine position. After planar acquisition, gated emission computerized tomography was pelformed with an Apex rotating gamma camera (Elscint, Haifa, Israel). Tomographic acquisition was perJormed over a range of 180°, be- ginning at the 45” right anterior oblique position and then rotating to the 135” left posterior oblique position, with recording in 30 positions (every 6”) for 30 seconds each. The gating was related to the RR interval (stan- dard monitor recording) and divided into 16 frames. The raw data consisted of 16 frames in each of 30 po- sitions (total of 480 frames). Every 2 consecutiveslices, beginning with the end-diastolic slice, were combined, and 8 groups of transaxial slices were derived with each group representing a consecutivephase on the RR in- terval. From these slices, long-axis and sagittal views were reconstructed. The first ji-ame was chosen as the end-diastolic frame, and the one in which the cavity of the left ventricle was the smallest was considered as end-systolic. The accuracy of end-diastolic and end-sys- tolic fiame delineation was verified using a visual tine presentation. The dimensions of the apex, septum, and lateral, anterior and inferior walls in end-diastole and end-systolewere determined using a digitized presenta- tion of the mid-long-axis and mid-sagittal views. The extent and direction of the apical transverse motion were also determined. The end-diastolic and end-sys- tolic frames in the 2 perpendicular long-axis views (mid-sagittal and mid-long-axis) were realigned so that the long axis was parallel, oriented vertically and with superimposed epicardial apexes in end-diastole. The common long axis was extendedfrom the epicardial apex to the most basal point for all 4 contours (epi- cardium and endocardium in endsystole and end-dias- tole). The common long axis was divided into 50 seg- I SAGGITAL LONG AXIS LATERAL BRIEF REPORTS 235