Povilas Žudys, Eglė Kazakauskaitė, prof. habil. dr. Renaldas Jurkevičius
Lithuanian University of Health Sciences, Department of Cardiology
Abstract
Aim. Evaluate myocardial viability, volumes of heart chambers and systolic function with different modalities: echocardiography, SPECT, PET/CT, and MRI.
Objectives. 1) Assess and compare myocardium movement, perfusion, and viability with echocardiography, SPECT, PET/CT, and MRI.
2) Compare ejection fraction (EF), end-diastolic volume (EDV), and end-systolic volume (ESV) results between different modalities.
3) Establish which diagnostic method is similar to PET/CT („gold standard“) the most.
4) Compare MRI („gold standard“) EF, EST, EDT results with results of other modalities.
Methods. Prospective trial. The statistical analysis was performed by SPSS software.
Study participants. Thirty patients who were treated in the Department of Cardiology, Hospital of Lithuanian University of Health Sciences Kaunas Clinics in 2012-2016. They have significantly lowered EF (EF ≤40%), severely damaged coronary arteries after acute myocardial infarction, that is why the benefit of revascularization for these patients is questionable because myocardium could have scars.
Results. Average patient age was 67,8 (interval 43–84 year), there was 25 (83,3%) men and 5 (16,6%) women. All patients had risk factors. Between echocardiography and PET/CT paired t test results of EF (OR 0,73 [95% CI -2,04-3,51], p=0,59), EDV (OR 13,36 [95% CI -10,18-36,92], p=0,25), ESV (OR 0,76[95% CI -16,93-18,46], p=0,93) were not significantly different (p>0,05). Although SPECT compared with PET/CT (paired t test) between EF (OR 2,76 [95% CI 0,88-4,64], p<0,05), EDV (OR 37,7 [95% CI 20,08-55,39], p<0,05), ESV (OR 23,26 [95% CI 11,34-35,19], p<0,05) and paired t test results between CMR and PET/CT (OR 3,43 [95% CI -5,68-(-1,18)], p<0,05), GDT (OR -43,23 [95% CI -66,82 (-19,64)], p<0,05) ir GST (OR -23,13 [95% CI -40,35-(-5,90)], p<0,05), it is clear that difference is significant. While evaluating miocard segment with Spearman corelation coeficient (r>0,05) and tested with McNemar test (p>0,05) it is clear, that in thirteen from sixteen miocardium segments there was found no significant difference between different modalities. Calculated Kappa coeficient shows agreement from fair to great. On the other hand, three of rest segments (lateral basal (18), lateral middle (12) and upper anterior(1)) statisticaly significantly were different (r<0.05, κ<0.20), McNemar test was negative with only few PET/CT with ultrasound and PET/CT with SPECT (p<0.05).
Conclusion. 1) Non-invasive cardiovascular imaging modalities (echocardiography, (99mTc)-MIBI SPECT, (18F)-FDG PET/CT and MRI) correlate well because 13 of 16 segment parameters match between types of examination. Spearman correlation coefficient (r>0,05) and Kappa value (κ>0,05) was counted, McNemar test was performed (p>0,05).
2) The agreement is seen between ultrasound and PET/CT when EF, ESV, and EDV is compared. Agreement between MRI and SPECT is seen too. Only EF (without EDV and ESV) agreement is seen between echocardiography and MRI. A paired t-test was performed (p>0,05).
3) PET/CT (“gold standard”) and CMR damage of segments (viability) correlate best of all because most of the segments (10 of 16) Spearman coefficients and 11 of 16 segments Kappa values are close to one.
4) Comparing MRI (“gold standard”) EF, EDV, ESV with other modalities, significantly similar data could be obtained only with MRI and SPECT. EF (without EDT and ESV) agreement is seen between echocardiography and MRI alone (p>0,05).
Keywords: PET/CT, MRI, SPECT, myocardial infarction