aDepartment of Radiology, St. Paul's Hospital
Departments of bNuclear Medicine
dRadiology, and Seoul St. Mary's Hospital, The Catholic University of Korea, The College of Medicine, Seoul, Korea
eDepartment of Radiology, Division of Nuclear Medicine, Emory Hospital, Atlanta, USA
Correspondence to Dr Hyeon-Sook Kim, MD, Department of Radiology, St. Paul's Hospital, The Catholic University of Korea, The College of Medicine, Seoul, Korea
Tel: +82 2 958 2084; fax: +82 2 960 4568;
Received 26 May 2009 Revised 12 July 2009 Accepted 13 July 2009
Objective: The location of a myocardial lesion on a wall thickening polar map often does not coincide with the location of the lesion on a perfusion polar map, especially when the myocardial lesion is located at the mid cardiac region. The purpose of this study was to determine the frequency and extent of discrepancy in the location of the lesion between perfusion and wall thickening polar maps on gated single photon emission computed tomography (SPECT) using lesion axis angle (LAA).
Methods: One hundred and forty-seven consecutive patients (male : female = 80 : 67, age range: 41–96 years) who underwent myocardial gated 99mTc-tetrofosmin SPECT on the suspicion of myocardial ischemia or infarct between September 2003 and September 2008 and showed both reduced myocardial perfusion and wall thickening on gated SPECT at mid cardiac region were reviewed. LAA, which is the angle between the lesion axis on perfusion and wall thickening polar maps, was measured for the patients who showed a discrepancy in lesion axis between the two polar maps. LAA was said to have a positive value when the lesion axis of the wall thickening polar map showed a counterclockwise angular rotation as compared with that of a perfusion polar map. The patients with LAA of less than 10° were considered as having no lesion axis discrepancy between perfusion and wall thickening polar maps. LAA was correlated with left ventricular ejection fraction (LVEF) on gated SPECT using Pearson's correlation. Furthermore, two groups, one with LAA of ≥10° and the other with LAA less than 10° were correlated with dichotomous groups with ≥50% and less than 50% LVEF, using the χ2 test. Then, 35 patients with acute coronary syndrome (ACS group) were analyzed separately for correlation between LAA and LVEF.
Results: The mean ± SD of LAA was 44.31±30.77° (range: 0–145°). LAA was 0–10° in 25 patients, 11–30° in 24 patients, 31–60° in 58 patients, 61–90° in 30 patients, and >90° in 10 patients. In addition, the lesion axis of the wall thickening polar map as compared with that of the perfusion polar map was rotated in the counterclockwise direction in 122 patients and not rotated in 25 patients. LVEF on gated SPECT showed positive correlation with LAA (P = 0.000147). In addition, there was statistically significant correlation (P = 0.001) when the two groups with LAA of ≥10° and less than 10°, respectively, were correlated with the groups of ≥50% and less than 50% LVEF. For the ACS group, the mean ± SD of LAA was 45.88±30.30° (range: 0–135°) and LVEF showed positive correlation with LAA (P = 0.0001). There was no significant statistical difference concerning LAA and LVEF between ACS group and non-ACS group (P = 0.725 and P = 0.473, respectively).
Conclusion: In most of our patients with coronary artery disease, the lesion axis of reduced wall thickening was rotated in the counterclockwise direction as compared with that of reduced perfusion on SPECT polar maps, especially when the myocardial lesion was at mid cardiac region. The LAA decreased as the LVEF decreased. This might be related to spatiotemporal distortion of myocardial contraction mentioned in the helical heart concept.
Since Torrent-Guasp introduced the concept of helical motion of the heart, understanding of alteration in cardiac wall motion in the diseased heart as well as normal cardiac movement have become possible . There are many imaging methods that prove helical motion of the heart such as MRI or echocardiogram [2–6]. In addition, there is one previous study that proves helical motion of the heart with gated single photon emission computed tomography (SPECT) . This study examined whether torsion of the heart can be estimated with gated SPECT data and whether torsion corrections affect cardiac measurements.
When we analyzed the myocardial gated SPECT images of the patients with coronary artery disease (CAD), we often noticed a discrepancy in the location as well as extent and severity of the myocardial lesion between the perfusion polar map and wall thickening polar map. Interestingly, the axis of the myocardial lesion on the wall thickening polar map was often rotated in the counterclockwise direction when compared with the axis of lesion on the perfusion polar map, especially when the myocardial lesion was located at mid cardiac region (mid-anterior, inferior, septal, or lateral segments). This phenomenon on SPECT images, to the best of our knowledge, has not been described in the literature so far and may be caused by the rotational myocardial motion in short-axis orientation during the cardiac cycle.
We have demonstrated the difference in the lesion axis between perfusion and wall thickening polar maps on gated SPECT in patients with CAD using lesion axis angle (LAA). The lesion axis was defined as a line crossing the center of the myocardial lesion and the center of a polar map. LAA was defined as the angle formed by the lesion axis on perfusion polar map and lesion axis on wall thickening polar map.
Myocardial gated SPECT images of 786 consecutive patients in whom myocardial ischemia or infarct was suspected between 2 September 2003 and 26 September 2008 were reviewed. The patients with reduction of both the myocardial perfusion and wall thickening on gated SPECT were included and their medical records including clinical history, ECG, or echocardiogram and conventional coronary angiography or CT coronary angiography were reviewed. Patients with myocardial lesions in the apical or basal cardiac region (basal-anterior, inferior, septal, or lateral segments) without abnormality at mid cardiac region were excluded. Patients with CAD of all three coronary vessels or diffusely decreased wall thickening and thus the lesion axis could not be drawn were also excluded. Finally, 147 patients (80 male, 67 female, age range: 41–96 years) were included in our study. Clinical characteristics of 147 patients are presented on Table 1. The cases were further validated by conventional coronary angiography (n = 33), CT coronary angiography (n = 13), or clinical follow-up (n = 101). The time interval between gated SPECT and conventional coronary angiography or coronary CT angiography ranged from 0 to 30 days.
Gated SPECT imaging
One-day rest–stress protocol was used to obtain myocardial perfusion gated SPECT images. Image acquisition began 60–90 min after intravenous injection of 20–30 mCi 99mTc-tetrofosmin (Myoview; GE Healthcare Ltd., Amersham, UK) at rest and stress, respectively. The stress study was performed 4 h after the rest study. Gated SPECT data were obtained using a Millenium VG (GE Healthcare Ltd., Haifa, Israel) dual head tomographic system and step-and-shoot mode. Sixty projections were obtained over a semicircular 90° arch starting at 45° and 315° right anterior oblique position with 30 s detections every 3°. At each projection angle, image data of eight frames per cardiac cycle were acquired for relative percent peak count of radioisotope on rest and stress. The camera was equipped with 3/8 inch thick NaI(Tl) crystals and a high resolution, low energy, parallel hole collimator. A 20% energy window centered on the 140 keV photopeak of 99mTc was used. The stress study was performed in all patients following intravenous infusion of 0.14 mg/kg/min adenosine (Adenocor, Sanofi-Synthelabo Ltd., Nortre Dame De Bondeville, France) for 6 min. Before reconstruction, all studies were corrected for nonuniformity with a 100 million counts matrix, obtained weekly from a 99mTc flood source.
Reconstruction was done by means of filtered back projection technique and a Butterworth filter (order 5, cutoff frequency 0.52 cycle/pixel for the stress study; order 10, cutoff frequency 0.40 cycle/pixel for the rest study) without attenuation and scatter correction. Trans-axial tomograms were reoriented into vertical and horizontal, long axis, and short axis planes. Regional perfusion on both rest and stress gated SPECT images were shown as relative percent peak count. Data were displayed on a 20-segmented polar map and automatically graded with a 5-point scale and summed scores were calculated. Twenty segments were assigned to the three coronary territories: left anterior descending coronary artery (LAD) territory, right coronary artery (RCA) territory, and left circumflex artery (LCx) territory. Emory Cardiac Toolbox (ECToolbox; GE Medical Systems, Triat, Hacarmel, Israel) software package was used to generate three kinds of perfusion polar maps: raw polar map, blackout polar map, and standard deviation polar map with the gated SPECT data. The blackened area of the blackout polar map represented pixels with the raw profiles below the threshold compared with the corresponding normal database. ECToolbox was also used to analyze wall thickening image with stress gated SPECT data. Percent wall thickening was calculated by the following formula:
Equation (Uncited)Image Tools
where ES represents end-systolic and ED represents end-diastolic . Count based percent regional wall thickening was displayed on the polar map and automatically graded with a 6-point scale: >40; 25–40; 10–25; 0–10; −10 to 0; and less than −10%. The epicenter and the number of slices, 14–18 slices, slice thickness of 1 cm, from apex to base of the heart were adjusted to be the same on both perfusion polar map and wall thickening polar map. A three-dimensional surface display, cinematic, eight frame SPECT images were used for the assessment of percent wall thickening and regional wall motion. Furthermore, left ventricular (LV) volume and left ventricular ejection fraction (LVEF) were calculated from ECToolbox stress gated SPECT data. LVEF was derived from the volumes using the following formula:
Equation (Uncited)Image Tools
Imaging analysis and definition of lesion axis and lesion axis angle
The imaging interpretations were confirmed by consensus of two experienced observers. The lesion axis on perfusion or wall thickening polar maps was defined as a line crossing the center of the myocardial lesion and the center of the polar map. The gated SPECT data obtained from the stress study were used to draw the lesion axis on both perfusion and wall thickening polar maps. The blackened area of the stress blackout perfusion polar map was used to define a lesion of reduced perfusion. Estimated percent wall thickening less than 25% was considered as reduced wall thickening. We obtained the lesion axis from the lesions located at mid cardiac regions (mid-anterior, inferior, septal, or lateral region of the heart). First, we drew two circles to exclude the lesions in the apical and basal cardiac regions from the 20 segments of polar map (Fig. 1a and c). Two circles had circumferences ½ and 3/4 the radius of the polar map. Second, we drew two lines along the lateral borders of a lesion from the center of the polar map (Fig. 1b and d). Then, the lesion axis was obtained by drawing another line equally dividing the angle formed by the above two lines at the lesion borders. Finally, the LAA, which is the angle formed by the lesion axis on perfusion polar map and lesion axis on wall thickening polar map, was measured for patients with a discrepancy in the lesion axis between the two polar maps. The most severely involved areas with a 5-point scale on perfusion polar map and with 6-point scale on wall thickening polar map were taken into consideration in drawing lesion axis on polar maps of gated SPECT when the extent of reduced perfusion and reduced wall thickening were not well matched using blackout polar maps. The LAA has a positive value when the lesion axis of wall thickening polar map showed counterclockwise angular rotation compared with the perfusion polar map. An LAA of less than 10° was considered as having no lesion axis discrepancy between perfusion and wall thickening polar maps.
The LAA was correlated with the LVEF on gated SPECT using Pearson's correlation. Furthermore, the group with LAA of ≥10° and the other group with LAA less than 10° were correlated with dichotomous groups with LVEF ≥50% and LVEF less than 50% using χ2 test. Thirty-five patients with acute coronary syndrome (ACS) were analyzed separately for correlation between LAA and LVEF using Pearson's correlation. The patients with ACS (ACS group) were clinically diagnosed as unstable angina (n = 10), ST elevation myocardial infarct (n = 22) or non-ST elevation myocardial infarct (n = 3). And then, Student's t test was used to evaluate whether there is statistical difference concerning LAA and LVEF between ACS group and non-ACS group. All data entry and analysis were performed with SPSS 10.0 software (Apache Software Foundation, Chicago, Illinois, USA). A probability value of 0.05 was considered to be statistically significant.
On stress and rest perfusion polar maps, 129 patients showed myocardial lesions at mid cardiac region with involvement of single vascular territory: 107 patients at LAD territory, nine patients at RCA territory, 13 patients at LCx territory. Eighteen patients showed myocardial lesions with involvement of two-vessel territories: nine patients at LAD and RCA territories, five patients at LAD and LCx territories, four patients at RCA and LCx territories.
The mean±SD of LAA was 44.31±30.77° (range: 0–145°). LAA was 0–10° in 25 patients, 11–30° in 24 patients, 31–60° in 58 patients, 61–90° in 30 patients, and >90° in 10 patients (Table 2). Out of a total of 147 patients, the lesion axis of wall thickening polar map was rotated in the counterclockwise direction compared with that of the perfusion polar map in 122 patients (Fig. 1) and not rotated in 25 patients.
LVEF on gated SPECT showed positive correlation with LAA (P = 0.000147) (Fig. 2). In addition, there was statistically significant correlation (P = 0.001) when the group with LAA of ≥10° and the group with LAA less than 10° were correlated with groups with LVEF ≥50% and LVEF less than 50%, respectively. The mean LVEF on gated SPECT was 60.52±13.53% (range: 15–86%).
For the patients with ACS, the mean±SD of LAA was 45.88±30.30° (range: 0–135°) and LVEF showed positive correlation with LAA (P = 0.0001). There was no significant statistical difference concerning LAA and LVEF between ACS group and non-ACS group (P = 0.725 and P = 0.473, respectively).
ECG gated myocardial SPECT is an accurate method to simultaneously evaluate myocardial perfusion and ventricular function in patients with ischemic heart disease [9,10]. Regional wall motion and wall thickening can be evaluated in addition to myocardial perfusion by this imaging method and these functional parameters play an important role in the diagnosis of myocardial viability. Some studies have reported that gated SPECT with technetium compounds is a method with excellent reproducibility for the clinical monitoring of LV volumes and LVEF [11,12]. Because of this advantage, gated SPECT has gained popularity in many hospitals and is currently one of the most commonly performed imaging studies for evaluating risks, making therapeutic decisions and monitoring sequential follow-up of CAD.
Percent wall thickening contains information about the regional myocardial function while myocardial perfusion image represents the status of cardiac vessels. So, the region of reduced wall thickening showing discrepancy in extent and severity compared with the region of reduced perfusion on gated SPECT polar maps is not an uncommon finding. However, we observed a discrepancy in the location of the lesion between myocardial perfusion and wall thickening polar maps in addition to a discrepancy in extent and severity. In our study, most of the patients (122 patients among a total of 147 patients) showed counterclockwise rotation of the lesion axis on the wall thickening polar map compared with the lesion axis drawn on the perfusion polar map for myocardial lesions at mid cardiac region. It was the same in the patients with ACS and there was no significant statistical difference concerning LAA between ACS group and non-ACS group. We excluded cases with myocardial lesions at apical or basal cardiac regions. Defining the lesion axis for lesions at the apical region would not be reliable because these segments are located in the central portion of polar map. Furthermore, in the cases of the lesions at the basal segments, ECToolbox algorithm is vulnerable to an incorrect estimation.
We think that the discrepancy of the lesion axis between perfusion and wall thickening polar maps in CAD patients may be related to rotational myocardial motion in short-axis orientation during the cardiac cycle. The helical movement of systolic LV motion in normal heart described by Torrant-Guasp  is the most likely explanation for the counterclockwise rotation of the myocardial lesion on the wall thickening polar map compared with that of the myocardial perfusion polar map on gated SPECT. Recently, advanced imaging techniques have been introduced in cardiac imaging and have become useful tools for the study of cardiac anatomy and function [2–6]. Lorenz et al. and Jung et al. [4,5] described the complete systolic time course of myocardial motion using tagged cine-MRI. Investigators observed that during LV isovolumetric contraction, the myocardial twist is counterclockwise when seen from the apex. Later in systolic phase, the base of the LV changes the rotation to the clockwise direction, whereas the apical region continues with counterclockwise rotation. Torsion increased steadily throughout systole after isovolumic contraction, whereas twist displayed rate changes. Fogel et al.  analyzed LV diastolic motion by myocardial tissue tagging MRI and concluded that diastolic mechanics is not homogeneous and LV diastolic rotation becomes more clockwise during motion from the atrioventricular valve to the apex. This kind of integrated mechanical behavior of the left ventricular myocardium can be explained by a double-loop spatial arrangement of the myocardial fibers in an oblique pattern described by Torrent-Guasp  and Torrent-Guasp et al. .
Our observation of counterclockwise rotation of a region of reduced wall thickening compared with that of reduced perfusion on gated SPECT polar maps might be explained by the ECToolbox algorithm. Percent wall thickening is calculated by the following formula:
Equation (Uncited)Image Tools
Therefore, misregistration of calculated wall thickening on the gated SPECT polar map may be inevitable in helically contracting heart. As a relative percent peak count of radioisotope is acquired using eight individual ECG gated bins per R-R interval on gated SPECT for a perfusion polar map, the count of radioisotope for one pixel is erroneously gathered from eight adjacent pixels in a continuously rotating heart. Those are illustrated on Fig. 3.
Considering an important element from the cardiac helical fiber concept which states that the ventricular size is the keynote for the helical architecture of diseased heart, we postulated that the counterclockwise rotation of reduced wall thickening region compared with the reduced perfusion region on SPECT polar maps would decrease and eventually become close to zero as the LVEF decreases. For this reason, we used LVEF as a variable of global myocardial function for correlation with LAA. In our study LAA turned out to decrease as LVEF decreased. In addition, we found out that the patients with LVEF ≥50% were more likely to show counterclockwise rotation of myocardial lesion on wall thickening polar map compared with the perfusion polar map than those with LVEF less than 50%. Our findings suggest that there is a relationship between cardiac helical fiber and myocardial function as observed in previous studies [14–17].
In our study, we tried to demonstrate the presence of discrepancy of the lesion axis between the perfusion polar map and the wall thickening polar map on gated SPECT in the patients with CAD and discuss the reason for the phenomenon. We introduced LAA to measure the discrepancy of the lesion axis between myocardial perfusion and wall thickening polar maps on gated SPECT and found out LAA to be a reproducible measurement method. LAA, which was closely correlated with myocardial function on gated SPECT may possibly be used as an additional cardiac measurement and may be used to compare changes in myocardial function following volume reduction cardiac surgery for the patients with both CAD and heart failure.
Limitations of the study
The principal limitations of this study are that it was retrospective and there was a small number of patients for the ACS group. Another limitation is that the myocardial lesions can be erroneously localized on both perfusion and wall thickening polar maps in a contracting heart and it is difficult to decide which lesions on perfusion or wall thickening polar maps are true. Therefore, it was difficult to validate our observation of counterclockwise rotation of reduced wall thickening compared with that of reduced perfusion on SPECT polar maps. However, the cinematic view of wall motion showed counterclockwise rotation of the region of reduced perfusion in most of our cases and we propose that the counterclockwise rotation of the heart may be a possible explanation for our observation (Fig. 3). Further studies with advanced cardiac imaging methods, such as MRI or PET to correlate with gated SPECT for this specific finding would be helpful. Although the helical motion of the heart is not accounted for radionuclide techniques because images on gated SPECT are not motion-blurred by this cardiac motion, there was an effort to correct this in a laboratory level .
In most of our patients with CAD, the lesion axis of reduced wall thickening was rotated in the counter-clockwise direction compared with the lesion of axis of reduced perfusion on SPECT polar maps, especially when the myocardial lesion was at mid cardiac region. LAA, the angle between the two lesion axis on perfusion and wall thickening polar maps, decreased as LVEF decreased. The correlation might be related to spatiotemporal distortion of myocardial contraction mentioned in the helical heart concept.
The authors would like to thank Hee-Seung Bom and Chan-Beom Park for their inspiration in preparing the manuscript.
1. Torrent-Guasp F. On cardiac morphology and functionalism. I. Rev Esp Cardiol 1966; 19:48–55.
2. Helm P, Beg MF, Miller ML, Winslow RL. Measuring and mapping cardiac fiber and laminar architecture using diffusion tensor MR imaging. Ann N Y Acad Sci 2005; 1047:296–307.
3. Boettler P, Claus P, Herbots L, McLaughlin M, D'hooge J, Bijnens B, et al. New aspect of the ventricular septum and its function: an echocardiographic study. Heart 2005; 91:1343–1348.
4. Lorenz CH, Pastorek JS, Bundy JM. Delineation of normal human left ventricular twist throughout systole by tagged cine magnetic resonance imaging. J Cardiovasc Magn Reson 2000; 2:97–108.
5. Jung B, Markl M, Foll D, Hennig J. Investigating myocardial motion by MRI using tissue phase mapping. Eur J Cardiothorac Surg 2006; 29 (Suppl 1):S150–S157.
6. Fogel MA, Weinberg PM, Hubbard A, Haselgrove J. Diastolic biomechanics in normal infants utilizing MRI tissue tagging. Circulation 2000; 102:218–224.
7. Nichols K, Kamran M, Cooke CD, Faber TL, Garcia EV, Bergmann SR, Depuey EG. Feasibility of detecting cardiac torsion in myocardial perfusion gated SPECT data. J Nucl Cardiol 2002; 9:500–507.
8. Cooke CD, Garcia EV, Cullom SJ, Faber TL, Pettigrew RI. Determining the accuracy of calculating systolic wall thickening using a fast Fourier transform approximation: a simulation study based on canine and patient data. J Nucl Med 1994; 35:1185–1192.
9. Germano G, Kiat H, Kavanagh PB, Moriel M, Mazzanti M, Su HT, et al. Automatic quantification of ejection fraction from gated myocardial perfusion SPECT. J Nucl Med 1995; 36:2138–2147.
10. Faber TL, Cooke CD, Folks RD, Vansant JP, Nichols KJ, DePuey EG, et al. Left ventricular function and perfusion from gated SPECT perfusion images: an integrated method. J Nucl Med 1999; 40:650–659.
11. Hyun IY, Kwan J, Park KS, Lee WH. Reproducibility of Tl-201 and Tc-99m sestamibi gated myocardial perfusion SPECT measurement of myocardial function. J Nucl Cardiol 2001; 8:182–187.
12. Hambye AS, Van Den Branden F, Vandevivere J. Diagnostic value of Tc-99m sestamibi gated SPECT to assess viability in a patient after acute myocardial infarction. Clin Nucl Med 1996; 21:19–23.
13. Torrent-Guasp F, Buckberg GD, Clemente C, Cox JL, Coghlan HC, Gharib M. The structure and function of the helical heart and its buttress wrapping. I. The normal macroscopic structure of the heart. Semin Thorac Cardiovasc Surg 2001; 13:301–319.
14. Hashimoto A, Nakata T, Wakabayashi T, Kyuma M, Takahashi T, Tsuchihashi K, Shimamoto K. Validation of quantitative gated single photon emission computed tomography and an automated scoring system for the assessment of regional left ventricular systolic function. Nucl Med Commun 2002; 23:887–898.
15. Di Donato M, Toso A, Maioli M, Sabatier M, Stanley AW Jr, Dor V, the Restore Group. Intermediate survival and predictors of death after surgical ventricular restoration. Semin Thorac Cardiovasc Surg 2002; 13:468–475.
16. Vanoverschelde JL, Depre C, Gerber BL, Borgers M, Wijns W, Robert A, et al. Time course of functional recovery after coronary artery bypass graft surgery in patients with chronic left ventricular ischemic dysfunction. Am J Cardiol 2000; 85:1432–1439.
17. Sallin EA. Fiber orientation and ejection fraction in the human ventricle. Biophys J 1969; 9:954–964.
© 2010 Lippincott Williams & Wilkins, Inc.