|Home | About | Journals | Submit | Contact Us | Français|
Coronary flow velocity pattern (CFVP) recorded within 3 days of percutaneous coronary intervention (PCI) has been reported to be useful in predicting left ventricular (LV) function. The aim of this prospective study was to investigate, via transthoracic Doppler echocardiography, whether the relationship between CFVP and recovery of LV function persists.
Our study group comprised 37 patients with 1st anterior-wall acute myocardial infarction who underwent successful PCI for lesions in the left anterior descending coronary artery (LAD). The CFVP in the LAD was recorded at 24–48 hours, 7 days, and 4 weeks after PCI. Myocardial contrast echocardiography was performed at 24–48 hours after PCI.
The diastolic deceleration time (DDT) at each stage correlated significantly with the regional LV wall-motion score index at 6-month follow-up (r=–0.58 at 24–48 hr, –0.57 at day 7, and –0.50 at week 4; P <0.01 for all). The mean DDT increased over time. Optimal cutoff values for DDT to predict regional LV wall-motion score indices of <2.0 were 327 ms at 24–48 hours (sensitivity, 0.78; specificity, 0.64), 495 ms at day 7 (sensitivity, 0.75; specificity, 0.69), and 525 ms at week 4 (sensitivity, 0.83; specificity, 0.69). The DDT at 24–48 hours significantly correlated, better than the peak creatine kinase value, with reperfusion (r=0.68, P <0.01) as defined by myocardial contrast echocardiography.
In conclusion, CFVP in the LAD can be used, within 4 weeks after PCI, to predict the recovery of regional LV function in patients with reperfused anterior-wall acute myocardial infarction.
Despite adequate epicardial artery reperfusion, not all reperfusion therapy can achieve tissue-level perfusion.1,2 The evidence has shown that coronary flow velocity pattern (CFVP), when viewed shortly after percutaneous coronary intervention (PCI), can be used to evaluate reperfusion of the microcirculation and predict the recovery of left ventricular (LV) function in patients who have experienced a 1st anterior-wall acute myocardial infarction (AMI).3–5 It has been inferred that flow measurements can be used to indicate microvascular resistance in infarcted myocardium. Most such studies have investigated CFVP within 3 days after PCI. However, sometimes the CFVP cannot be recorded in time. We wondered if there might still be a relationship between CFVP and LV function at a later time.
In this prospective study, we examined CFVP serially by transthoracic Doppler echocardiography (TTDE) to determine whether CFVP at a later time still correlates with regional LV function at 6-month follow-up.
The study included 38 patients referred to our hospital with a 1st anterior-wall AMI. All patients underwent successful PCI within 12 hours of the onset of symptoms, and the culprit artery, as determined by means of coronary angiography, was the left anterior descending coronary artery (LAD). The diagnosis of AMI was made if the patient had shown the following: prolonged chest pain (>30 min), ST-segment elevation of ≥0.2 mV in at least 2 contiguous precordial electrocardiograms, and ≥3-fold increase in serum creatine kinase (CK) level. Of the 38 patients, one was excluded due to a cardiac event during follow-up. Therefore, 37 patients formed the study group (30 men and 7 women; mean age, 63 ± 11 yr; age range, 49–79 yr). The study protocol was approved by the Hospital Ethics Committee. Written consent was obtained from all patients.
Because of the difficulty in acquiring CFVP in coronary arteries other than the LAD, this analysis included only patients with 1st anterior-wall AMI. Each patient received aspirin (300 mg) and clopidogrel (300 mg) as soon as the AMI diagnosis was made. Successful PCI, including stent placement, was performed in all patients. Flow in the infarct-related vessel was graded in accordance with the Thrombolysis in Myocardial Infarction (TIMI) flow classification after the vessel had been recanalized.6 If the reflow grade did not reach TIMI 3, the investigator could choose to administer adenosine or calcium-channel-blocking agents to the coronary artery to improve blood flow. In-hospital drug treatments after PCI included the following: aspirin, clopidogrel, β-adrenergic-blocking agents, angiotensin-converting enzyme inhibitors, calcium-channel-blocking agents, and statins. The choice of medications was up to the clinician in charge. Serum CK was sampled at 8, 16, 24, 48, and 72 hours after PCI. The peak CK value was recorded.
Fundamental 2-dimensional (2-D) echocardiography, TTDE, and myocardial contrast echocardiography (MCE) were performed in the coronary care unit 24 to 48 hours after PCI (mean, 30 ± 5 hr) had successfully recanalized the infarct-related coronary vessel. The CFVP was recorded at that time and again at 7 days and at 4 weeks after PCI. Left ventricular function was evaluated at 6-month follow-up by means of 2-D echocardiography. All studies were performed using an ACUSON Sequoia™ 512 (Siemens Medical Solutions USA, Inc.; Mountain View, Calif) equipped with an ultrawide-band-frequency transducer (3–8 MHz). All echocardiographic data were stored digitally on magneto-optical disc and on S-VHS videotape for subsequent offline analysis and blinded review.
The “intramural” pre-setting was chosen, and LAD flow was sought under the guidance of color-flow mapping. To record coronary flow velocity using the pulsed Doppler, a sample volume (5·2.5 mm) was set on the color signal in the mid to distal LAD. Only when the Doppler incident angle was greater than 30° was an angle correction made. The CFVP was analyzed by a person who was blinded to other patient data. Peak diastolic velocity, mean diastolic velocity, diastolic velocity time integral, and diastolic deceleration time (DDT) (Fig. 1) were measured. The mean of each of these values was calculated from 2 or 3 cardiac cycles.
Intravenous MCE was then performed by injection of 2 mL of SonoVue® (Bracco Diagnostics Inc.; Milan, Italy) followed by a 2-mL saline flush. Image acquisition was performed using the Sequoia 512 with contrast pulse sequencing: image frequency, 2.0 MHz; low mechanical index, 0.16; frame rate, 26 Hz; and gain, –10. The instrument settings for MCE were consistent for all patients. Contrast enhancement was analyzed on end-diastolic frames and graded in each segment as follows: 0 for no opacification; 0.5 for a patchy pattern in the entire segment or for opacification only in the epicardial portion of the segment (Fig. 2); and 1 for homogeneous and complete enhancement. A mean perfusion score was calculated for each patient by dividing the sum of the contrast scores for individual segments within the infarct bed by the number of infarct segments that displayed abnormal wall motion. A patient was considered to have no reflow if his or her mean perfusion score was 0.5 or less.7,8
At each study to determine systolic LV function, the echocardiographic wall motion9 of 16 myocardial segments was evaluated by an independent observer who was blinded to the patients' data. The wall-motion score for each myocardial segment was recorded as follows: 1, normal; 2, mild hypokinesis; 3, severe hypokinesis; and 4, akinesis or dyskinesis. Nine of the 16 segments were determined to be in LAD territory.10 Global and regional infarct-zone wall-motion score indices (WMSI) were calculated by dividing the sum of the scores of the analyzed segments by the number of analyzed segments.
Continuous variables were expressed as mean ± SD. Variables between the groups of reflow and no reflow were compared by means of Student's 2-tailed t test, the Wilcoxon rank sum test, and the χ2 test, as appropriate. Analysis of variance and the post-hoc t test with the Bonferroni correction were used to compare repeated measures of CFVP values at different stages. Pearson's correlation was applied to evaluate the correlation between peak CK value and reperfusion as determined by MCE examination. Multiple linear regression analysis was performed to evaluate the relationship between CFVP values and regional WMSI at 6-month follow-up, and the relationship between CFVP values and mean perfusion scores. Partial correlation coefficients were calculated between CFVP values and regional WMSI at 6-month follow-up. The ability of the CFVP values to predict regional WMSI of <2.0 at 6-month follow-up was analyzed by use of receiver operating characteristic curves. All P values of <0.05 were considered statistically significant. Statistical analysis was performed with SPSS version 11.0 (SPSS Inc.; Chicago, Ill).
Of the 37 patients, 28 had complete recordings of CFVP values in the LAD both at day 7 and at week 4.
The mean time from the onset of symptoms to coronary perfusion was 7.1 ± 3 hours. The mean peak CK level was 3,104 ± 1,860 U/L. There was no clinical evidence—angiographic, electrocardiographic, or enzymatic—of reinfarction in the 37 patients during the 6-month follow-up.
Global and regional WMSI of the 37 patients decreased significantly from testing at 24–48 hours to testing at 6-month follow-up (from 1.72 ± 0.39 to 1.56 ± 0.39, and from 2.19 ± 0.62 to 1.91 ± 0.62, respectively; both P <0.05). The study patients were divided into 2 groups, reflow and no reflow, according to their MCE results. Table I shows a comparison between the 2 groups at baseline. Except for peak CK value, diastolic deceleration time, and global WMSI, there were no significant differences in baseline characteristics.
Table II shows CFVP values at different stages. Mean DDT, compared with that at 24–48 hours, increased by 43% at day 7 and by 48% at week 4. There was no significant difference between CFVP values at day 7 and week 4, except for DDT.
Of the CFVP values, DDT measured serially after PCI significantly correlated with regional WMSI at the 6-month follow-up. Figures 3, ,4,4, and and55 show the relationship between DDT at each stage and regional WMSI at the 6-month follow-up. There was no significant correlation between other CFVP values and regional WMSI at the 6-month follow-up.
Optimal cutoff values for DDT as determined by receiver operating characteristic curve analysis were 327 ms (sensitivity, 0.78; specificity, 0.64) at 24–48 hours, 495 ms at day 7 (sensitivity, 0.75; specificity, 0.69), and 525 ms at week 4 (sensitivity, 0.83; specificity, 0.69). These cutoff values were chosen to predict regional WMSI of <2.0.
The DDT measured at 24–48 hours was significantly correlated with the mean perfusion score as determined by MCE (r=0.68; P <0.01). There was no significant relationship between other CFVP values and the mean perfusion score.
Although peak CK value was significantly correlated with the mean perfusion score (r=–0.513; P <0.01), there was no significant correlation between peak CK value and WMSI at the 6-month follow-up.
This study showed a significant correlation between DDT recorded at each stage and regional left ventricular WMSI at 6-month follow-up. This means that DDT recorded within 4 weeks after PCI can be used to predict the recovery of regional LV function. The study also showed a trend of prolonged diastolic deceleration over time. Consequently, if we use DDT within 7 days after PCI for prediction of LV function recovery, the trend of increase in DDT should be considered. Between day 7 and week 4, there was no significant difference in DDT, although the mean value increased slightly. At day 7, the cutoff value for DDT to predict regional WMSI of <2.0 was also close to that at week 4.
Flow measurements have been thought to reflect microvascular resistance in the infarcted myocardium. In this study, DDT was found to be the best indicator of microvascular resistance, among the several variables that we used to define CFVP. Rapid DDT after reperfusion indicates high microvascular resistance. The increase in microvascular resistance is thought to be caused by obstructive perivascular edema and leukocyte plugging of the capillaries after acute myocardial ischemia.11–16 The extent of microvascular obstruction increases over the first 48 hours after AMI,17 reaching its peak 2 days after reperfusion.18 After that, microvascular resistance might decrease progressively, as the occluded microvessels gradually recanalize. Accordingly, in our study, the DDT at day 7 and at week 4 was longer than at 24–48 hours. But DDT at a later time, even at 4 weeks after PCI, still correlated with regional LV function at the 6-month follow-up. A clinical study19 reported that, in patients with chronic myocardial infarction, microvascular resistance is still significantly higher in the infarct area than in the reference area. In a histopathologic study of chronic myocardial infarction,20 an inverse relationship was found between the reduction in capillary density and the transmurality of the infarct scar. The amount of viable myocardial tissue within an infarct area correlated with the number of capillaries. These reports might explain our finding.
Our study also showed a significant correlation between DDT at 24–48 hours and myocardial perfusion as defined by MCE. According to this study, the DDT at 24–48 hours was more useful in assessing myocardial reperfusion than was the peak CK value.
Because our analysis included only patients with 1st anterior-wall AMI, our conclusion applies only to patients who experienced anterior-wall AMI due to an LAD lesion. Systolic velocity was not assessed because it was too difficult to differentiate, by TTDE, a systolic coronary flow velocity signal from an artifact signal.
Second, we judged reflow and no reflow only by MCE, uncorroborated by other imaging methods (such as cardiac magnetic resonance imaging or emission computed tomographic imaging). This might have made the reperfusion analysis less accurate.
Third, our relatively small sample size might have restricted our introducing more variables into the multivariate analysis. However, the study showed a strong relationship between DDT and regional LV function. Future clinical study with a greatly increased sample size may show even more clearly the prognostic ability of CFVP.
In conclusion, the use of CFVP in the LAD within 4 weeks after PCI appears to predict the recovery (or nonrecovery) of regional LV function in patients with reperfused anterior-wall AMI.
The authors wish to acknowledge the involvement of Dr. Hua Sun throughout the design, analysis, and publication of this study. Also, we thank Xu-Hong Hou for her statistical programming expertise and Bruce Smith for his expertise in editing medical English.
The study was funded (Grant No. 064119507) by the Science & Technology Committee of Shanghai Municipal Government, People's Republic of China.
Address for reprints: Yue-Li Zhang, MD, Department of Ultrasound in Medicine, Shanghai Sixth People Hospital Affiliated to Shanghai Jiao Tong University, Yi Shan Road 600, Shanghai 200233, PRC