Cardiovascular Scintigraphy

Myocardial Perfusion Imaging (MPI)

(Cold spot imaging). Comparison between stress and rest requires identical repositioning. Stress >85% max predicted HR (220-age) or double/triple product (DP, systolic pressure x HR) cf rest to ~20,000; by exercise or infusion of dipyridamole/adenosine (coronary vasodilatation, but not if >50% stenosis) or dobutamine (direct inotrope/chronotrope). Label injected at peak (exercise continued for further 30-60s) and attatches to muscle on first pass. SPECT with LV myocardium in short-axis (apex/apical button to base; anterior, septal, lateral and inferior/posterior walls), vertical long-axis and horizontal long axis. 180% anterior acquisition (asymmetry of heart and spine attenuation), ECG gated (allows wall motion analysis and functional information). Prone imaging after standard supine reduces false-positives from breast/diaphragm atenuation, hot bowel loops, motion artifacts.

Thallium/Tl-201 – HL 73h, 69-83keV (low energy, poor penetration, high dose), cyclotron. K+ analogue pumped into viable cells by Na/K-ATP pump, hence more physiological than Tc-99m agents. Normal myocardial 50% washout is 4h. Image within 5-10min of injection. Rest/redistribution image at 3-4h fills in ischaemic-type defects. Reinjection just before delayed/rest can increase sensitivity for very high-grade stenosis. High lung activity at stress (lung:heart ratio) or post-stress dilatation indicates LVF during exercise (poor prognosis for subsequent cardiac event).

Tc-99m sestamibi (Cardiolite) and Tc-99m tetrofosmin (Myoview, faster background clearance) – Passive diffusion, bound to myocytes mostly in mitochondria, no significant redistribution, negligible washout. Delayed imaging at 30-60min to allow biliary and background clearance. Rest imaging performed another day, or initial small dose rest scan before larger dose stress 4hrs later. Dual isotope scan has initial Tl-201 rest followed by Tc-99m stress.

Bull’s-eye maps and 3D reconstructions are more sensitive, but true abnormalities should also be visible on planar or SPECT images. Sensitivity and specificity ~90%.

Reversibly ischaemic areas have low uptake at stress but uptake at rest; usually associated with angiographic abnormality and relieved by treatment. Peri-infarct ischaemia adjacent to true infarct is of less significance. Ischaemic defects may be from CAD or LBBB, vasculitis, vasospasm, cardiomyopathy (dilated or hypertrophic) and small vessel disease (eg diabetes). Ischaemia may not be detected if there was inadequate stress or if there is balanced triple vessel disease (usually has post-stress dilatation).

Hibernating myocardium simulates infarction, doesn’t contract at rest but is still viable; from severe ischaemia, high-grade stenosis. Slow to reverse on Tl-201, may show on reinjection or 24h delayed; more sensitive than Tc99m rest images.

Myocardial infarction has layers of nonperfused scar tissue, thin myocardium, decreased stress and rest uptake; may be subendocardial throught to transmural. Beware attenuation artifacts (breast tissue, subdiaphragmatic structures) causing “defects” on both images; reduced with repeat prone post-stress scan with Tc-99m, gating (normal perfused tissue moves inwards, thickens and contracts becoming brighter), or attenuation correction with SPECT-CT +/- scatter correction (scattered photons from emission source still accepted).

Stunned myocardium – Temporarily damaged cells around infarction, hypokinetic/akinetic and does not hold Tl-201 until recovery several weeks later (hence infarct on Tl-201 appears worse on rest cf stress = “reverse redistribution”).


Higher resolution and fewer attenuation artifacts than standard MPI. Stress usually performed with pharmacologic agents. Perfusion imaging with rubidium-82 or ammonia-13. Hibernating myocardium may have low uptake on rest (simulating infarction), but normal or increased FDG uptake (shift from free fatty acid uptake to glucose metabolism) whereas true infarction has no significant FDG uptake.

Myocardial Infarct Scan

(Hot spot imaging). Tc-PYP (pyrophosphate), Tc-tetracycline, Tc-GH (glucoheptonate), In-111-antimyosin antibodies or F-18-sodium fluorine. Pyrophosphate taken up in myocardial necrosis, complexing with calcium deposits. PYP scan positive at 12hrs, peak sensitivity 48-72hrs, normal at 14/7; persistent uptake implies poor prognosis or developing aneurysm. Cardiomyopathies and diffuse myocarditis show diffuse uptake, contusions and radiation myocarditis regional uptake.

Gated Blood Pool Scans

Radionuclide ventriculogram (RVG) uses Tc-99m-labelled RBC and ECG gating (to R waves) into minimum 32 frames/cycle (several hundred cardiac cycles averaged, taking 5-20min) to evaluated LV size, wall motion, LV volume curve (V = LV volume vs T = time; best obtained from septal LAO view) and functional parameters (LVEF and first derivative = dV/DT). Volume curve from counts in computerised edge detection of LV ROI. LVEF is the single best parameter of LV function = EDV-ESV (normal 55-75%); falsely lowered with arrythmias (detected with R-R interval histogram; most systems allow selected analysis of cycles with same R-R interval). CO = HR x LVEF x EDV; EDV calculated by comparing with count rate in blood sample of known volume. Count-based ratio method for CO uses total counts in LV ROI, maximum pixel count and pixel size.

  • LVEF elevated with mitral/aortic regurgitation, hypertrophic cardiomyopathy, high CO states (eg hyperthyroidism).
  • LVEF reduced with previous MI, ischaemia (with CHF), cardiomyopathy. LVEDV and RVEDV usually appear equal on best-septal LAO (RVEDV normally greater to give equal stroke volume, compensating for smaller EF). LV dilates and becomes rounder in failure.

Wall motion assessed in cine on LAO/best septal, anterior and LPO. Areas of hypokinesia/akinesia indicate damage/infarction. Aneurysm causes dyskinesia. Fourier phase analysis includes amplitude (wall motion) and phase (relative timing). Amplitude image useful for confirming hypokinesia/akinesia. Phase image with phase angle histogram (phase vs pixels) useful for tardive kinesis (slowly contracting myocardium) and dramatic in dyskinesia (aneurysmal areas).

Valvular regurgitation calculated with each pixel value coded with blood volume change during cardiac cycle. Total values in LV/RV ROI produces LV-to-RV stroke volume fraction, thus regurgitant fraction (only works when regurgitant valve on one side only, and cannot distinguish AR from MR).

Exercise RVG performed for 2-3min periods at each exercise level and each compared to rest. Relative CO compared between stages and workload. Normal patients increase/augment LVEF and dV/DT white reducing LVESV. Abnormalities include increase of LVEDV >10%, lack of increase/fall in LVEF, or wall motion abnormalities caused by exercise-induced ischaemia.

Right Ventricular Studies

RV is unable to be isolated from other chambers as per LV.

  • First pass function study – RAO, bolus of high-specific-activity isotope injected rapidly followed immediately by nonradioactive flush. Activity passes through RV in 3-8 heartbeats, time-activity curve used to measure RVEF for each beat; average RVEF calculated. Alternatively RVEF measured using Xe-133 slow venous infusion, which is exhaled without entering the left heart; gated study over many seconds as per standard RVG. Normal RVEF 32-52%.
  • First pass flow studies – Anterior projection comparing blood flow between lungs (for extrinsic PA compression, pulmonary sequestration) or transit time between ventricles (delay in CHF). SVC obstruction demonstrates numerous collaterals on initial 1-second frames. Activity in lungs should reduce exponentially. L->R intracardiac shunts show prolongation of washout as blood reenters the right side; quantified with gamma-variate curve-fitting method (for shunts ratios as low as 1.2:1 cf 2:1 detected by CXR).
  • Right-to-left shunts detected with IV macroaggregated albumin, normal <10% activity found in systemic circulation via AV shunts in the lungs. Static images taken of the whole body, with ROI over lungs, head, neck, abdo and extremities to quantify the shunt.