Patient centric performance and interpretation of SPECT and SPECT/CT myocardial perfusion imaging: a clinical consensus statement of the European Association of Cardiovascular Imaging of the ESC

Abstract

The non-invasive assessment of ischaemic heart disease with myocardial perfusion imaging remains an integral part of modern cardiology. This modality has been used for decades, but improving technology has maintained its relevance today. This document describes the fundamentals of single-photon emission computed tomography, including stress protocols, tracer pharmacodynamics, camera settings and capabilities, post-acquisition processing, and clinical translation in an easy to read and highly pictorial manner to be applicable to not only healthcare providers of all levels, but patients as well.

Keywords: SPECT.

Figure 1

Exercise and pharmacologic stress protocols.

Figure 2

Flowchart to help decide best stress modality in the setting of conduction disease.

Figure 3

Patient preparation: examples of medications, caffeine-containing foods, and caffeine-containing drinks that can interfere study quality. *Some providers may wish to continue beta-blockers or vasodilators when the purpose of the exam is to evaluate the extent of ischaemia in the setting of medical treatment.

Figure 4

Standard 1-day stress/rest protocol. Note there is a two- to three-fold increase in dose with the second injection to avoid shine through, thus increasing the radiation exposure. Compared with 2-day protocols, 1-day protocols may be more convenient for patients. Prone imaging can be performed for attenuation correction in the absence of a hybrid SPECT/CT system. With hybrid systems, CT for attenuation correction or calcium scoring can be performed prior to the SPECT acquisition. *Note: tracer dose may differ based on type of SPECT system.

Figure 5

Standard 1-day rest/stress protocol. Note there is a two- to three-fold increase in dose with the second injection to avoid shine through, thus increasing the radiation exposure. Compared with 2-day protocols, 1-day protocols may be more convenient for patients. Prone imaging can be performed for attenuation correction in the absence of a hybrid SPECT/CT system. With hybrid systems, CT for attenuation correction or calcium scoring can be performed prior to the SPECT acquisition. *Note: tracer dose may differ based on type of SPECT system.

Figure 6

Standard 2-day protocol. Both rest and stress tracer injections are the same dose. Two-day testing limits radiation exposure but may be less convenient for patients, as it requires two trips to the imaging centre. Two-day protocols may be more logistically challenging as they require appointment times to remain open until a decision can be made about the need for rest imaging. Alternatively, 2-day protocols may decrease the number of hours a patient is in the imaging facility and may reduce camera downtime. Prone imaging can be performed for attenuation correction in the absence of a hybrid SPECT/CT system. With hybrid systems, CT for attenuation correction or calcium scoring can be performed prior to the SPECT acquisition. *Note: tracer dose may differ based on type of SPECT system.

Figure 7

Two-day protocols have lower radiation dose, less time in the hospital, and less camera downtime compared with 1-day protocols. Depending on type of practice, 2-day protocols may present logistical challenges to coordinate.

Figure 8

Typical gamma camera with collimator, scintillation crystals, photomultiplier tubes, and dedicated electronics.

Figure 9

Unfiltered back projection (A) leads to smearing of the original view along the path it was originally acquired, resulting in a blurry image. During filtered back projection (B), each view is filtered before being back projected, resulting in an exact and sharp reconstruction of the original image. Reproduced with permission.

Figure 10

Filtered back projection (FBP) vs. ordered subset expectation–maximization (OSEM) iterative reconstruction for different total counts and with a butterworth post-processing filter at different intensities. Low dose  =  300 MBq 99mTc-tetrofosmin; high dose  =  900 MBq 99mTc-tetrofosmin. Reproduced with permission.

Figure 11

Myocardial perfusion SPECT imaging of a female patient with a body mass index of 29 kg/m². Selected slices (stress in the top row, rest in the bottom row) and corresponding polar plots of the reconstruction without attenuation correction (A) depict a reduction in counts particularly affecting the inferior wall. In contrast, attenuation corrected reconstructions (B) lead to a complete normalization, unmasking the apparent defect as an attenuation artefact.

Figure 12

Panel (A) shows a fixed defect in the inferior wall. However, after CT attenuation correction, the fixed defect is shown to be artefact from diaphragmatic attenuation in Panel (B).

Figure 13

The initial images (A) suggest a large defect in the inferior wall. However, evaluation of the fusion images shows a misregistration artefact. When properly aligned (B), the perfusion is normal.

Figure 14

The D-SPECT is a CZT-system with nine rotating detector columns arranged over a stationary gantry (left). CZT-based cameras have the capability to reduce radiation dose and shorten imaging time. The D-SPECT system can also improve comfort by imaging patients while sitting (right).

Figure 15

Shows the quality review panel with the orthogonal projection views of the raw-data (left) with an elliptical region of interest circumscribing the heart. This figure also displays the sinogram and the transaxial images for axis orientation and ventricular limits determination.

Figure 16

Shows the standard review panel of the spread-out tomographic slices of the left ventricle, conventionally the stress images are located above the rest images in the three axes. SA, short axis that runs from apex (left) to base of de ventricle (right); HLA, horizontal long axis that runs from inferior (left) to anterior wall of de ventricle (right); VLA, vertical long axis that runs from septal (left) to lateral wall of de ventricle (right).

Figure 17

Shows the standard segmentation used to name the 17 segments of the left ventricle, extrapolated to the right on a generic polar map, in turn correlating each arterial territory. LAD, left anterior descending artery; RCA, right coronary artery; LCX, circumflex artery.

Figure 18

Polar maps showing a perfusion defect at peak stress in the circumflex artery territory which reverses at rest. Ischaemia (reversibility of the defect) is shown in the lower map. On the right you can see the correlation in the 3D rendering of the myocardium adjusted to perfusion.

Figure 19

Shows the level of myocardial involvement (arrows) according to the visual hypoperfusion score, where 0 corresponds to normal perfusion, 1 is equivocal, 2 is moderate, 3 is severe hypoperfusion, and 4 is absent.

Figure 20

Two different types of acquisitions of gated imaging for quantification of ejection fraction and volumes are shown: on the left, volume curves acquired with 16 intervals between R-R of the ECG and on the right with 8, reconstructed with different commercial software.

Figure 21

Panel (A) shows a gated SPECT with normal function and motility, with preserved thickening in a hypertrophic left ventricle. Panel (B) shows a patient with an anterior myocardial infarction and an anterior and septal dyskinesia (arrows) with severely depressed ejection fraction.

Figure 22

Supine (left) and Prone (right) post-effort SPECT images acquisitions, showing diaphragmatic attenuation (small white arrows) in sections of the LV (basal) short axis and in the VLA, also observable in the polar map (long arrows).

Figure 23

In this CZT acquisition, motion artefact is present on the left, showing a misleading inferior defect (arrows). The images were reacquired with the patient moving from a supine to prone position. This change in position can shift the diaphragm away from the heart and eliminate diaphragmatic attenuation. In this patient, prone imaging was more comfortable, leading to less motion during the acquisition. The motion artefact is no longer present on the right images. Note the increase in EF and the marked decrease in LV volumes that the artefact falsely presents.

Figure 24

Three different cases of motion artefact, where a pattern often seen as a ‘double inverted V’ (arrows) is observed. On the right, the fourth graph corresponds to the schematic representation of the phenomenon.

Figure 25

Left ventricle polar map views (middle) and its correspondent tomographic non-gated slices (left) and 3D rendering (right) with end diastole and perfusion layers. This shows how the poor positioning of the valve plane line marked as V (left ventricular base), has an influence on the size of the defect with the consequent under (∼7% of LV in upper PM) or overestimation (∼12% of LV in lower PM) of an area of hypoperfusion. In this case, there was a true defect in the territory of the diagonal artery (long arrows), but artefact in the base (short arrows).

Figure 26

A comparison of radiation exposure for various cardiac imaging modalities. The typical background radiation exposure is 3.1 mSv. Of note, there is a wide variation in radiation exposure among the modalities based on technique, equipment, protocols, and patient factors. The radiation dose of PET MPI will be larger if viability imaging is performed. SPECT MPI, single-photon emission computed tomography myocardial perfusion imaging; PET MPI, positron emission computed tomography myocardial perfusion imaging; CCTA, coronary computed tomography angiography; CMR, cardiac magnetic resonance; Echo, echocardiography.

Figure 27

Radiation reduction strategies and best practices for SPECT MPI.^,

Figure 28

Incidental cardiovascular findings in CT performed for attenuation correction. (A) Distinct coronary calcifications should be noted. (B) Aortic valve and (C) mitral annular calcifications should prompt further evaluation with echocardiography. (D) Ascending aortic enlargement should also lead to dedicated evaluations.

Figure 29

Incidental lung findings on CT performed for attenuation correction. Small bilateral pleural effusions are noted (A). A lung mass in the right upper lobe in a background of emphysema is evident (B). A lung nodule is observed in the left upper lobe (C).

Figure 30

Dedicated coronary artery calcium scanning as part of a rest–stress myocardial perfusion study. No (A), mild (B), and severe (C) coronary calcifications are noted.

Figure 31

Components of a clinically meaningful report.