Assessment of Global Left Ventricular Function with 64-MSCT: Comparison with Cineventriculography

Transcription

1 中華放射醫誌 Chin J Radiol 08; 33: Assessment of Global Left Ventricular Function with 64-MSCT: Comparison with Cineventriculography Ping-Hua Tsai Gigin Lin Yuan-Chang Liu Ming-Jer Hsieh 2 Chun-Chi Chen 2 Ming-Shien Wen 2 Yun-Liang Wan Department of Medical Imaging and Intervention, 2nd Section of Cardiology, Department of Internal Medicine 2, Chang Gung Memorial Hospital at Linkou, Chang Gung University The purpose of this study was to evaluate the role of multi-slice computed tomography (MSCT) in assessing global left ventricular (LV) function. Twenty patients (6 male, 4 female) with angina were enrolled, the ages ranged from 42 to 78 years. Data from standardized 64-MSCT scan was analyzed with dedicated analysis software to evaluate the global LV function. Conventional angiography was used as the standard for comparison. Statistical analysis with Pearson s correlation was used. For LV volumes, there was excellent correlation for end-systolic volume (r = 0.83, p < 0.00) and moderate correlations for end-diastolic volume (r = 0.57, p = 0.04). Good correlation was also observed (r = 0.73, p < 0.00) in the evaluation of global ejection fraction. Our result indicates that it is reliable to assess global LV function with 64-MSCT. Functional analysis in additional to anatomical evaluation may increase the clinical efficacy of MSCT. Reprint requests to: Dr. Yun-Liang Wan Department of Medical Imaging and Intervention, Chang Gung Memorial Hospital at Linkou, Chang Gung University. No. 5, Fu Hsing Road, Gueishan, Taoyuan 333, Taiwan, R.O.C. Morbidity and mortality for a patient with angina correlate with global left ventricular (LV) function [], therefore LV function could serve as an important prognostic parameter and is routinely used for therapeutic decision []. A variety of diagnostic tools have been implemented for assessment of ventricular function [2], such as echocardiography, electronic beam computed tomography (CT), single photon emission CT, and conventional angiography. Cardiac magnetic resonance imaging (MRI) has been accepted as the gold standard during the past 0 years in the evaluation of LV systolic function due to improved temporal and spatial resolution, its consistency and reproducibility [, 3], but inferior image quality about coronary arteries limited its ultimate success as a one-stop shop in cardiac imaging. There is a good correlation between CT angiography (CTA) and conventional coronary angiography (CCA) in quantifying coronary arterial stenosis [4-8], which enables CTA as a noninvasive method for anatomical evaluation. Recently, ret rospectively ECG-gated multislice computed tomography (MSCT) was introduced for the assessment of cardiac function [9]. With the same data set acquiring for CTA and CCA, functional and anatomical data can be obtained simutanously from 64-MSCT and conventional cineventriculography (CVG). In the present study, we attempt to assess the correlation between CVG and MSCT in evaluation the LV function. MATERIALS AND METHODS Subjects From December 05 to November 06, a total of patients who suffered from angina were enrolled in our study. The subjects included 6 men and 4 women with range of age from 42 to 78 years.

2 24 Assessment of global left ventricular function with 64-MSCT a b 2a 2a 3a 3b Figures -3. After identifying the a. end-systolic phase and b. end-diastolic phase, standard. short-axis, 2. four-chamber and 3. two-chamber views were created. Using the integrated computer software, the left ventricle volume was calculated with the Simpson s method by slice summation. Semi automated contour detection and manual cor rection of contours were performed.

3 Assessment of global left ventricular function with 64-MSCT 25 They underwent MSCT imaging for evaluation of coronary arterial condition, and all patients received CCA within two days after MSCT. Exclusion criteria included renal insufficiency (serum creatinine 2.0 mg/dl), previous allergic reaction to iodinated contrast media, and arrhythmia. MSCT All MSCT scanning were performed with a 64-slice multi-detector spiral CT scanner (Aquilion 64, Toshiba, Japan). Four of subjects had oral administration of 0-30 mg of Propranolol to achieve targetted heart rate under 70 beats per minute. Sublingual nitroglycerin spray (0.4mg) was routinely administered 5-7 minutes before the scan. The parameters used for MSCT included 64 x 0.5 mm collimation, pitch 2.8 cm, kv, and 0mA for coronary evaluation. Retrospectively electrocardiography gated analysis was performed. Region of interest was placed at the proximal descending thoracic aorta around the carina level for automated peak enhancement detection used for determining the time of starting scan. While continuous monitoring the ROI breaking through the presetted threshold of Hounsfield Unit, 90 to 95 ml of non-ionic contrast media was administered via right antecubital vein at the rate of 5 ml/s followed immediatedly with 50mL normal saline chaser bolus. Scan was initiated 4 seconds after contrast medium injection. Patients were instructed to strictly hold their breath for about 0 seconds. Imaging analysis The same data set for CT angiography were retrospectively retrieved for evaluation of global LV function. The images were reconstructed with mm thickness and retrospectively electrocardiography gated at every 0% of the RR interval. These ten reconstructed MSCT images data were transferred to off-line workstation (Vitrea v. 3.0, Vital imges, Minnetonka, MN, USA) for evaluating LV function. Two chambers, four chambers, and short axis views were created for analysis of LV function. Enddiastolic volume (EDV) and end-systolic volume (ESV) were calculated by slice summation using semiautomated contour detection with manual correction if necessary (Fig. -3). Papillary muscles were regarded as part of the LV cavity. Global ejection fraction (EF) was generated by an equation: EF = (EDV-ESV) / EDV. Biplane Cineventriculography Biplane cineventriculography was performed using standardized 30 right anterior oblique projection and left anterior oblique, with injection of at least 30 ml of contrast medium at a flow rate of 2 ml/s, using a 6 French pigtail catheter. Semiautomatic contour-tracking was used to define the endocardial borders. The end-diastolic and end-systolic images are based on the EKG-guided frames on left ventriculograms. EDV and ESV were determined using area-length method. Statistical Data Data were expressed as mean ± SD, and compared with paired 2-tail Student s t test. The Pearson s coefficient r calculated according SPSS (Version 3.0; SPSS Inc, 233 S Wacker Dr, th Fl, Chicago, IL) analysis software was used to determine the correlation of parameters between MSCT and CVG. The Table. Comparison of end diastolic volume (EDV), end systolic volume (ESV) and ejection fraction (EF) using multislice computed tomography (MSCT) and cineventriculography (CVG). Pearson correlation MSCT CVG r value P value EDV (ml) Mean ± SD* 23.9 ± ± = 0.04 Range ESV (ml) Mean ± SD 45.9 ± ± < 0.00 Range EF (%) Mean ± SD 64.6 ± ± < 0.00 * SD = standard deviation Range

4 26 Assessment of global left ventricular function with 64-MSCT agreement was defined as follow: poor correlateion, r = 0; slight correlation, r = ; fair correlation, r = ; moderate correlation, r = ; good correlation, r = ; and excellent correlation, r = Bland-Altman analysis was performed for each pair of values of EDV, ESV and EF, to calculate limits of agreement and systematic errors between the two modalities. The software MedCalc (MedCalc Version , MedCalc Software, Mariakerke, Belgium) was used to perform this statistical analysis. RESULTS Sixteen out of patients had coronary artery disease on conventional angiography, including single vessel disease in 7 patients, 2-vessel disease in 3 patients, and triple vessel disease in 6 patients. The imaging quality for MSCT was excellent to acceptable in all exams, and could also be used for evaluating function. MSCT Data For MSCT, the range of EDV and ESV were -84 (mean ± SD = 23.9 ± 26.3) ml and 7-38 (45.9 ± 26.4) ml, respectively. The range of LV ejection fraction was (64.6 ± 2.3) %. Conventional angiography data For CVG, the range of EDV and ESV were (.7 ± 29.9) ml and 2-44 (.0 ± 27.) ml, respectively, and the range of LV ejection fraction was (.0 ± 2.7) % (Fig. 4). Comparing MSCT and CVG There was a significant correlation between the MSCT and CVG with respect to EDV, ESV, EF. Between the MSCT and CVG, there was good correlation with respect to the global LVEF (r = 0.73, P < Figure 4. Calculation of left ventricular volume with cineventriculography. The endocardial borders are traced semi-automatically. CathEDV (ml) 5a EDV (ml) CathESV (ml) 5b 00 0 ESV (ml) CathEF % 5c EF % Figure 5. Scatter plots showing correlation between left ventricular parameters end diastolic volume (EDV) a. end systolic volume (ESV) b. and ejection fraction (EF) c. as determined by CVG and multislice computed tomography.

5 Assessment of global left ventricular function with 64-MSCT 27 (EDV - CathEDV) / Average % SD 44.0 Mean SD (ESV - CathESV) / Average % SD 54.9 Mean SD (EF - CathEF) / Average % SD 37.5 Mean SD a AVERAGE of EDV and CathEDV 6b AVERAGE of ESV and CathESV 6c AVERAGE of EF and CathEF Figure 6. Bland Altman plot of MSCT end diastolic volume (EDV) a. end systolic volume (ESV) b. and ejection fraction (EF) c. showing relationship between differences and means with biplane cineventriculography. The difference (y-axis) between each pair (MSCT value - CVG value) is plotted against the average value (x-axis) of the same pair (MSCT value + CVG value divided by 2) (solid line = mean value of differences) 0.00), moderate correlation with EDV (r = 0.57, p = 0.04) and excellent correlation with ESV (r = 0.83, p < 0.00) (Table, Fig. 5). The differences between MSCT and CVG in systolic, diastolic volumes, and ejection fraction are illustrated with Bland-Altman plots (Fig. 6). The mean differences were 5,.4, and 7.8; 95% confidence interval (.96 SD = 95% CIs): -85 to 54.9, -4. to 44, and -22 to The ESV was underestimated with MSCT with mean difference of 5 ml. For EDV and EF, both parameters were shown slightly overestimated by MSCT with mean difference of.4 and 7.8 ml. These differences were all within 95% confidence interval, and that means we could use the two measurement methods interchangeably [0]. DISCUSSION The present study shows the potential clinical applications of 64-MSCT for functional evaluation. Previous studies had described the use of MSCT in evaluating LV function [2, -3]. Juergeus et al. and Heuschmid et al. used 4-slice MSCT and demonstrated well correlation with CVG findings in EF values (r = 0.8 and r = 0.79) [4, 5] However, Heuschmid s reported that the correlation between MSCT and CVG was less satisfactory with r = 0.8 for mean ESV and r = 0.5 for mean EDV. Such a correlation was also described by Martine et al. by using 6-slice MSCT (r = 0.74 and 0.4, respectively) [3, 5]. Schuijf et al. reported good correlation of 6-slice MSCT and 2-dimentional echocardiography in 70 patients, with a satisfactory correlation of EDV (r = 0.97), ESV (r = 0.98) and EF (r = 0.9) [6]. Similar findings were also found in comparing 64-MSCT and single photon emission CT (SPECT), with good correlation in EDV (r = 0.898), ESV (r = 0.956) and EF (r = 0.825) [7]. The global and regional LV function investigated by 64-MSCT and 2-dimentional echocardiography were comparable to that assessed by magnetic resonance imaging which was used as golden standard [8]. Our study demonstrated that between the 64-MSCT and CVG, there is an excellent correlation for ESV (r = 0.83), a moderate correlation for EDV (r = 0.57), and a good correlation for LVEF (r = 0.73). In this study, the correlation for EDV is lower than ESV between the MSCT and CVG. This finding could probably be attributed to the volume calculation method. The marginal contour of the chamber could be over or under estimated with semiautomated detection. Such a volume s discrepancy could be greater in diastolic phase than in systolic phase with slice summation. For precise qualitative and quantitative evaluation of LV global function, imaging with high temporal and spatial resolution is needed. ECGgated MSCT provides excellent spatial resolution to differentiate the LV cavity and myocardium [9-2]. In addition, differentiation between systolic and diastolic images is possible with a temporal resolution of milliseconds [22]. The temporal resolution of echocardiography, cardiac MRI, and electron beam CT is approximately 50 milliseconds and optimal temporal resolution is mandatory for reproducible global LV function assessment [9]. The 64-MSCT scanner we used offers a temporal resolution of milliseconds with partial-scan and multisegmental reconstruction algorithms and a spatial

6 28 Assessment of global left ventricular function with 64-MSCT resolution of 0.35 mm. The advancement of MSCT technology has gradually improved the diagnostic performance of CTA. In cases of noninvasive coronary angiography with MSCT, the same dataset might be used for assessment of global LV function, This is particularly of benefit to obese patients that have poor acoustic window in sonography, or to patients that are contraindicated for MRI examinations, e.g. pacemaker placement, metallic devices implants. The valid data can be obtained without additional radiation exposure or administration of contrast media []. There are several limitations in this study. First, the case number is small. Second, oral beta-blocker was used for obtaining optimal image quality by reducing heart rate. It may have influence over the contractility of the myocarium and alter the functional parameters [9]. Third, routine usage of sublingual nitroglycerin may also alter LV function. However, nitroglycerin is routinely used during CCA as well, thus internal bios could be diminished. In conclusion, our results indicate that the assessment of global LV function with 64-MSCT is feasible and reliable. There was a good correlation between MSCT and CVG with respect to global ejection fraction. Functional analysis in addition to anatomical evaluation with the same data-set may increase the clinical efficacy of MSCT without additional radiation exposure and administration of contrast media [9]. REFERENCE. Bansal D, Singh RM, Sarkar M, et al. Assessment of left ventricular function: comparison of cardiac multidetector-row computed tomography with two-dimension standard echocardiography for assessment of left ventricular function. Int J Cardiovasc Imaging 08; 24: Juergens KU, Fischbach R. Left ventricular function studied with MDCT. Eur J Radiol 06; 6: Gilard M, Pennec PY, Cornily JC, et al. Multi-slice computer tomography of left ventricular function with automated analysis software in comparison with conventional ventriculography. Eur J Radiol 06; 59: Ong TK, Chin SP, Liew CK, et al. Accuracy of 64-row multidetector computed tomography in detecting coronary artery disease in 34 symptomatic patients: influence of calcification. Am Heart J 06; 5: 323, Cury RC, Pomerantsev EV, Ferencik M, et al. Comparison of the degree of coronary stenoses by multidetector computed tomography versus by quantitative coronary angiography. Am J Cardiol 05; 96: Leber AW, Knez A, von Ziegler F, et al. Quantification of obstructive and nonobstructive coronary lesions by 64-slice computed tomography: a comparative study with quantitative coronary angiography and intravascular ultrasound. J Am Coll Cardiol 05; 46: Raff GL, Gallagher MJ, O Neill WW, Goldstein JA. Diagnostic accuracy of noninvasive coronary angiography using 64-slice spiral computed tomography. J Am Coll Cardiol 05; 46: Caussin C, Larchez C, Ghostine S, et al. Comparison of coronary minimal lumen area quantification by sixtyfour-slice computed tomography versus intravascular ultrasound for intermediate stenosis. Am J Cardiol 06; 98: Mühlenbruch G, Das M, Hohl C, et al. Global left ventricular function in cardiac CT. Evaluation of an automated 3D region-growing segmentation algorithm. European Radiology 06; 6: Bland JM, Altman DG. Statistical methods fro assessing agreement between two methods for clinical measurement. Lancet 986; : Dewey M, Muller M, Eddicks S, et al. Evaluation of global and regional left ventricular function with 6-slice computed tomography, biplane cineventriculography, and two-dimensional transthoracic echocardiography: comparison with magnetic resonance imaging. J Am Coll Cardiol 06; 48: Henneman MM, Bax JJ, Schuijf JD, et al. Global and regional left ventricular function: a comparison between gated SPECT, 2D echocardiography and multi-slice computed tomography. Eur J Nucl Med Mol Imaging 06; 33: Henneman MM, Schuijf JD, Jukema JW, et al. Comprehensive cardiac assessment with multislice computed tomography: evaluation of left ventricular function and perfusion in addition to coronary anatomy in patients with previous myocardial infarction. Heart 06; 92: Juergens KU, Grude M, Fallenberg EM, et al. Using ECG-gated multidetector CT to evaluate global left ventricular myocardial function in patients with coronary artery disease. AJR Am J Roentgenol 02; 79: Heuschmid M, Kuttner A, Schroder S, et al. Left ventricular functional parameters using ECG-gated multidetector spiral CT in comparison with invasive ventriculography. Rofo 03; 75: Schuijf JD, Bax JJ, Jukema JW, et al. Assessment of left ventricular volumes and ejection fraction with 6-slice multi-slice computed tomography; comparison with 2D-echocardiography. Int J Cardiol 07; 6: Schepis T, Gaemperli O, Koepfli P, et al. Comparison of 64-slice CT with gated SPECT for evaluation of left ventricular function. J Nucl Med 06; 47: Annuar BR, Liew CK, Chin SP, et al. Assessment of global and regional left ventricular function using 64-slice multislice computed tomography and 2D echocardiography: A comparison with cardiac magnetic resonance. Eur J Radiol 08; 65: Cury RC, Nieman K, Shapiro MD, Nasir K, Cury RC, Brady TJ. Comprehensive cardiac CT study: Evaluation of coronary arteries, left ventricular function, and myocardial perfusion - Is it possible? Journal of Nuclear Cardiology 07; 4:

7 Assessment of global left ventricular function with 64-MSCT 29. Nikolaou K, Rist C, Wintersperger BJ, et al. Clinical value of MDCT in the diagnosis of coronary artery disease in patients with a low pretest likelihood of significant disease. AJR Am J Roentgenol 06; 86: Achenbach S, Moshage W, Ropers D, Nossen J, Daniel WG. Value of electron-beam computed tomography for the noninvasive detection of high-grade coronaryartery stenoses and occlusions. N Engl J Med 998; 339: Orakzai SH, MD, Orakzai RH, MD, Nasir K, Budoff MJ. Assessment of cardiac function using multidetector row computed tomography. J Comput Assist Tomogr 06; 30:

8 30 Assessment of global left ventricular function with 64-MSCT 使用多層次 64 切電腦斷層評估左心室功能 : 與傳統血管攝影心室造影測量之比較 蔡秉樺 林吉晉 劉原彰 謝明哲 2 陳俊吉 2 溫明賢 2 萬永亮 長庚大學林口長庚紀念醫院影像診療科部 心臟內二科 2 本研究的目的是為了評估 64 切多層次電腦斷層在評估左心室功能所扮演的角色 總共有 個 歲臨床上有心絞痛表現的病人 ( 含 6 個男性及 4 個女性 ) 接受此項檢查 收集到的資料以專屬的軟體來分析, 並使用傳統的血管攝影作為比較的標準, 統計分析採用 Pearson s 相關係數 對於左心室的容量的評估, 多層次電腦斷層得到的結果與傳統血管攝影相比, 在收縮末期容積得到極佳的相關性 (r = 0.83, p < 0.00); 在舒張末期的容積得到中度相關性 (r = 0.57, p = 0.04); 在心臟的射出率則可觀察到好的相關性 (r = 0.73, p < 0.00) 由此可知使用 64 切的多層次電腦斷層在評估左心室功能是可信賴的 所以多層次電腦斷層除了可用來做心臟血管解剖上的評估, 還可運用於功能性的評估, 增加了多層次電腦斷層的臨床應用