Author + information
- Received September 28, 2007
- Revision received March 3, 2008
- Accepted March 15, 2008
- Published online June 1, 2008.
- Atsunori Okamura, MD,
- Hiroshi Ito, MD, FACC⁎ (, )
- Katsuomi Iwakura, MD,
- Toshiya Kurotobi, MD,
- Yasushi Koyama, MD,
- Motoo Date, MD,
- Yoshiharu Higuchi, MD,
- Koichi Inoue, MD and
- Kenshi Fujii, MD
- ↵⁎Reprint requests and correspondence:
Dr. Hiroshi Ito, Division of Cardiology, Sakurabashi Watanabe Hospital, 2-4-32 Umeda, Kita-ku, Osaka 530-0001, Japan.
Objectives This study sought to investigate the timing and amount of embolic particles generation during the percutaneous coronary intervention (PCI) procedure and studied the relationship between embolic burden and coronary blood flow and myocardial damage.
Background Distal embolization is a major complication of PCI. The Doppler guidewire (DGW) can detect the embolic particles as high-intensity transient signals (HITS) during the PCI procedure.
Methods We prospectively studied 37 patients with acute myocardial infarction (MI). Under monitoring with the DGW, we performed first and second balloon angioplasty, followed by stenting and post-high-pressure dilatation. Left ventricular ejection fraction (LVEF) (%) and regional wall motion (RWM) (standard deviation/chord) were measured on days 1 and 22.
Results The HITS were detected in 35 of 37 patients. The number of HITS was the greatest after stenting (16 ± 18) followed by first balloon inflation (5 ± 4). There was a significant correlation between the total number of HITS and the corrected Thrombolysis In Myocardial Infarction frame count (r = 0.52, p = 0.003) and a significant weak inverse correlation between the total number of HITS and changes in LVEF and RWM (r = 0.37, p = 0.03 and r = 0.35, p = 0.04, respectively).
Conclusions Distal embolization is common during PCI in patients with acute MI, and the majority of HITS were observed after stenting. An increase in the total number of HITS is associated with reduced coronary blood flow, and is weakly associated with poor recovery of left ventricular function.
Percutaneous coronary intervention (PCI) substantially improves the functional and clinical outcomes of patients with acute myocardial infarction (MI). One of the major complications of PCI, however, is distal embolization caused by thrombi and plaque components liberated during balloon inflation. Abrupt blood flow reduction is sometimes observed immediately after the inflation–deflation procedure of vulnerable plaque and lesions in saphenous vein grafts. The quantitative relationships between embolic burden and the reduction of coronary blood flow or myocardial damage remain unknown because of the lack of methods to detect and quantify embolic particles in vivo. Recently, Bahrmann et al. (1) reported that the intracoronary Doppler guidewire (DGW) can detect embolic particles as high-intensity transient signals (HITS), which was confirmed with a distal protection device by our group (2). They reported that sand particles with a mean diameter of 300 μm can be detected as HITS with the DGW, but they did not determine the minimal detectable size (1). The HITS signals are detected during PCI of the stable plaque, and an increase in number of HITS signals is associated with minor myocardial injury (3,4). We also experienced a case of acute MI in which a cluster of HITS caused angiographic no-reflow during stent implantation into the culprit lesion (5).
In the experimental study, we studied the size of particles detectable as HITS with the DGW using an experimental circuit system and microspheres of 3 different diameters. In the clinical study, we quantified the embolic burden and the timing of liberation during the PCI procedure in patients with acute MI and assessed the relationships between the embolic burden and reduction of epicardial coronary blood flow and myocardial damage. We also examined whether it is possible to predict the embolic burden from the information obtained with coronary angiography or the DGW.
Experimental circuit system and microspheres
We made an experimental circuit system (Fig. 1) made of a blood reservoir, polyethylene tubing of 3-mm internal diameter, and a motor to generate constant flow. The system was filled with 60 ml of heparinized blood obtained from a volunteer. Care was taken not to include bubbles in the blood. The DGW (FloMap, Volcano Therapeutics, Rancho Cordova, California) was inserted from the Y-connector connected to the circuit system and its tip was placed in the center of the tube. Constant blood flow with a mean flow velocity of 48 cm/s was made in the circuit system. We used microspheres of 3 different diameters (15, 50, and 80 μm) made from methylmethacrylate (Sekisui Plastics Co., Osaka, Japan). We calculated the number of microspheres that would pass the cross-sectional area of the tube in a period of 1 s from the concentrations of microspheres. For 15-μm microspheres, the initial dose was the dose at which 50 microspheres passed the tip of the DGW in 1 s, and the dose of microspheres increased in 6 steps to the maximum at which 500 microspheres passed the tip of the DGW in 1 s. At each dose, we examined whether it was possible to detect microspheres as HITS with DGW. If detected, we then counted the number of HITS in the blood flow velocity spectra within 1 s and the result was expressed as the average of the number of 5 arbitrarily selected 1-s intervals. After removing microspheres and blood from the system, it was refilled with blood and the same experimental procedure was repeated using microspheres of another size (50 or 80 μm).
We performed a prospective study in 40 consecutive patients with first acute MI referred to our hospital for PCI between December 2005 and April 2007. The patients fulfilled the following criteria: 1) chest pain >30 min in duration; 2) ST-segment elevation ≥2 mm in at least 2 contiguous electrocardiographic leads; 3) >3-fold increase in serum creatine kinase level; 4) total or subtotal occlusion at the initial coronary angiography; 5) reference lumen diameter ≥2.5 mm and lesion length <20 mm; and 6) successful coronary stenting with residual stenosis in the culprit lesion <25% achieved within 6 h after onset of symptoms. Three patients were excluded for the following reasons: cardiogenic shock, 2 patients; reinfarction during follow up, 1 patient. Finally, 37 patients were enrolled in the present study. This study complied with the Declaration of Helsinki with regard to investigations in human subjects, and the study protocol was approved by the ethics committee of our hospital. Written informed consent was obtained from all patients before cardiac catheterization.
Each patient was given aspirin and a bolus injection of nicorandil (4 mg) (Sigmat, Chugai, Tokyo) at least 30 min before PCI, and then nicorandil was injected continuously at 6 mg/h for 24 h to preserve microvascular integrity (6). After administration of heparin (100 U/kg), we performed coronary angiography and performed coronary thrombectomy with an aspiration catheter before balloon inflation to obtain information on the distal coronary artery and determine the appropriate diameter size of first balloon angioplasty (BA). We used the DGW (FloMap, Volcano Therapeutics) as the guidewire for PCI. The tip of the DGW was placed 2 cm distal to the target lesion, and the coronary blood flow velocity spectrum was recorded continuously on videotape throughout the subsequent PCI procedures. The PCI procedures consisted of consecutive 4 processes: first BA, second BA (the same balloon size and inflation pressure as in the first BA), stenting, and post-dilation with a high pressure of 20 atm (the same balloon size as used in the stenting). If angiographical coronary flow was reduced, we repeated coronary thrombectomy and/or injected nicorandil and nitroprusside into the coronary artery. Ten minutes after the final PCI procedure, we performed coronary angiography to evaluate the corrected Thrombolysis In Myocardial Infarction (TIMI) flow count and TIMI myocardial perfusion grade (TMPG). Then, we performed myocardial contrast echocardiography with intracoronary injection of contrast agent, as reported previously (7). Left ventriculography (right anterior oblique view) was performed on day 1 and at pre-discharge (22 ± 5 days).
Analysis of catheter data
The coronary blood flow velocity spectrum was digitized and analyzed with an off-line personal computer. We defined HITS as the following characteristics to distinguish emboli signals from artifacts, such as wire motion: 1) visual high intensity signals; and 2) unidirectional signals within the advancing velocity spectrum as we reported previously (3). We assessed coronary microvascular dysfunction by coronary angiography. The TIMI frame count, corrected TIMI frame count (8), and TMPG (9) were defined as reported previously.
The left ventriculogram (30° right anterior oblique projection) was analyzed by a physician who had no knowledge of the patients' data. Regional wall motion (RWM) (standard deviation/chord) was evaluated with the centerline method. Global left ventricular ejection fraction (LVEF) (%) and left ventricular end-diastolic index and end-systolic index (LVEDVI and LVESVI, respectively, ml/m2) were evaluated with the area–length method. Changes (Δ) in the values at pre-discharge minus those at admission were calculated.
Analysis of myocardial contrast echocardiography data
Myocardial contrast echocardiography images were analyzed using an off-line computer system, and the presence or absence of the no-reflow phenomenon was determined based on the size of contrast perfusion defects after PCI (10). Briefly, the risk area was defined as an area of akinesis or dyskinesia at baseline. If the endocardial length of contrast perfusion defect exceeded a quarter of that of the risk area, the microvascular integrity was considered imperfect and this was classified as the no-reflow phenomenon. Other cases were classified as good reflow.
All data are given as mean ± SD. All statistical analyses were performed using StatView (Abacus Concepts Inc., Berkeley, California). Simple linear regression analysis was performed to examine the relationship between the average number of HITS and the estimated number of microspheres that passed the cross-section of the tube in 1 s. The unpaired Student t test was used to compare the procedural results of continuous variables among the 4 PCI procedures. One-way analysis of variance and the post hoc test were used to compare the number of HITS among 4 PCI procedures. Simple linear regression was also performed to examine the relationship in the number of HITS between after the first BA and after stenting or throughout the PCI procedures, between the total number of HITS and lesion characteristics, and between the total number of HITS and peak creatine kinase or left ventricular functional and morphological outcomes. A value of p < 0.05 was considered statistically significant.
The detectable size of microspheres
Figure 2 shows the blood flow velocity spectra in the concentration at which 100 microspheres were calculated to pass the cross-section of the tube in 1 s. No visible HITS were observed when microspheres 15 μm in diameter were used. In contrast, HITS were detected when microspheres 50 and 80 μm in diameter were used. However, there were no differences in morphology or appearance of HITS between microspheres 50 and 80 μm in diameter. In each experiment using microspheres 50 and 80 μm in diameter, there was an excellent linear relationship between the average number of HITS and the estimated number of microspheres that passed the cross-section of the tube in 1 s (Fig. 3).
Baseline characteristics and the procedural results of PCI
Table 1 summarizes the patient backgrounds and lesion characteristics as well as the PCI procedural results. Reference diameter, minimal luminal diameter, diameter stenosis, and lesion length were measured after the first thrombectomy. Post-dilation balloon inflation pressure was higher than those in first BA, second BA, or stenting. The ratio of balloon to reference artery diameter were comparable among the 4 PCI procedures.
Detection of HITS during the PCI procedures
Figure 4 shows the coronary blood flow velocity spectra after each PCI procedure in a patient with anterior acute MI. Figure 5A shows a comparison with the number of HITS after each PCI procedure. The HITS were detected throughout the PCI procedures in 35 (95%) of the 37 patients, and all of the HITS signals were only observed within the initial 4 beats after balloon deflation. The number of HITS was greatest after stenting (16 ± 18), followed by first BA (5 ± 4), second BA (0.4 ± 0.9), and post-dilation (1 ± 3). There were no differences in the total number of HITS between PCI to the left coronary artery (n = 20) and that to the right coronary artery (n = 17) (17 ± 17 vs. 29 ± 31, p = 0.13). There was a significant correlation between the number of HITS after the first BA and that after stenting (r = 0.66, p = 0.0002) (Fig. 5B). There was also a significant correlation between the number of HITS after the first BA and that throughout the PCI procedures (r = 0.79, p < 0.0001). There were no significant correlations between the total number of HITS and the pre-procedural lesion characteristics (Table 2).
Correlation of the total number of HITS to coronary blood flow and myocardial damage
All of the patients showed TIMI flow grade 3 before stenting, but 5 patients (13%) showed substantial reduction in coronary blood flow after stenting without residual stenosis (Fig. 6) (TIMI flow grade 0: n = 1, TIMI flow grade 1: n = 2, TIMI flow grade 2: n = 2) (the right coronary artery: n = 4, the left coronary artery: n = 1). All 5 patients showed TIMI flow grade 2 after additional coronary thrombectomy and/or intracoronary injection of nicorandil and nitroprusside. The number of HITS after stenting was greater in patients with than in those without reduction of coronary blood flow (48 ± 16 vs. 11 ± 13, p < 0.0001). The frequency of TMPG-1 was significantly higher in patients with than in those without reduction of coronary blood flow (60% vs. 94%, p = 0.001) (Table 3). However, all of the 5 patients with reduced coronary blood flow showed good reflow with myocardial contrast echocardiography performed at the end of the PCI procedure.
Table 4 shows the correlations between the total number of HITS and the coronary blood flow or myocardial damage. The no-reflow phenomenon was observed with myocardial contrast echocardiography in 2 patients; however, their total numbers of HITS were low (10 and 17 counts). There was no difference in the total number of HITS between cases with and without the no-reflow phenomenon. There was a significant correlation between the total number of HITS and the corrected TIMI frame count. There was a weak inverse significant correlation between the total number of HITS and improvement in LVEF or RWM between day 22 and day 1. However, there were no significant correlations between the total number of HITS and peak creatine kinase level, changes (Δ) in LVEDVI and LVESVI, or left ventricular function and remodeling (LVEF, RWM, LVEDVI, and LVESVI) on day 22.
This esperimental study shows that: 1) microspheres ≥50 μm in diameter can be detected as HITS; and 2) there is a close relationship between number of HITS and number of microspheres. This clinical study shows that: 1) distal embolization is commonly found during PCI in patients with acute MI, and the number of embolic particles is greatest after stenting; 2) the number of embolic particles after first BA can predict that after stenting; and 3) coronary flow is reduced when the total number of HITS is great, but the capillary integrity is not necessarily impaired. Increases in the total number of HITS are weakly associated with poor recovery of left ventricular function.
The detectable size of microspheres with the DGW
This result regarding the detectable particle size is important with respect to whether the embolic particles detected as HITS can actually cause myocardial injury. Bahrmann et al. (1) reported that sand particles with a mean diameter of 300 μm (range 100 to 500 μm) can be detected as HITS with the DGW, but they did not determine the minimal detectable size. We found that DGW can detect microspheres at least 50 μm in diameter. Particles ≥50 μm in diameter can obstruct the arterioles and small arteries that regulate coronary blood flow, rather than the perforator or epicardial arteries (11). Hori et al. (12) performed intracoronary injection of microspheres in dogs. They reported that intracoronary injection of microspheres 15 μm in diameter hardly reduced resting coronary blood flow and was not associated with myocardial damage unless huge numbers of microspheres were injected. On the other hand, injection of microspheres 100 or 300 μm in diameter can easily reduce coronary blood flow and cause myocardial injury. These results supported the observation that the embolic particles detected as HITS with the DGW are clinically significant, influence coronary blood flow, and could cause myocardial damage.
The DGW detects Doppler signals originating from the clusters of red blood cells and not from the cells themselves, a phenomenon called Raleigh scattering. The diameter of a red blood cell is around 8 μm and is too small to be detected by high-frequency ultrasound signals. Similarly, individual microspheres 15 μm in diameter are too small to be detected as HITS with DGW until they are present in sufficiently large numbers to reflect ultrasound signals.
Liberation timing and prediction of the number of embolic particles
The important finding of this study was that the majority of distal embolization occurs after stenting, and coronary blood flow is reduced at this time, if it occurs. Al-Mubarak et al. (13) reported that transcranial Doppler ultrasound study also detects the majority of HITS after stenting in carotid stenting. Stone et al. (14) reported that decreases in coronary blood flow occur more frequently after stenting as compared with BA. Because there were no significant differences between balloon/reference artery diameter and stent-balloon/reference artery diameter, the stent structure itself can crush plaque more easily than the balloon and the liberated particles are squeezed into the coronary artery through the gaps between stent struts.
To achieve optimal PCI, it is important to predict patients who are likely to suffer from angiographic slow flow or no-reflow after stenting. We found that there is a correlation between the number of HITS after first BA and that after stenting. The number of embolic particles produced after stenting is likely to be around 3-fold greater than that after first BA (Fig. 6B). This relationship is clinically important because we can select lesions that will require distal protection after first BA, which may be associated with better results as compared with stenting alone.
The impact of distal embolization on coronary blood flow and myocardial damage
There was a significant correlation between the total number of HITS and the corrected TIMI frame count, indicating that embolic particles can cause coronary flow reduction. The total number of HITS showed weak correlations with changes in LVEF or RWM, but not with peak creatine kinase level or left ventricular function and volume on day 22. Thus, distal embolization sometimes reduces coronary blood flow transiently, but is not likely to have a deleterious effect on the recovery of left ventricular function. It is not associated with left ventricular remodeling. This is in agreement with the results of the randomized multicenter Enhanced Myocardial Efficacy and Removal by Aspiration of Liberalized Debris trial (15).
Myocardial contrast echocardiography findings could address the mechanism of the above results. It uses the microbubbles (mean size 12 μm) as tracers and can assess the integrity of the microcirculation at the capillary level (16). In the present study, 5 patients showed substantial reduction of coronary blood flow after stenting, and the total number of HITS of the 5 patients was great (48 ± 16) but they showed good reflow with myocardial contrast echocardiography performed at the end of the PCI procedure. On the other hand, 2 patients showed the no-reflow phenomenon; however, the total numbers of HITS in these cases was low (10 and 17 counts, respectively). This implies that distal embolization is not the main cause of the no-reflow phenomenon assessed by myocardial contrast echocardiography. The no-reflow phenomenon with myocardial contrast echocardiography is caused by extensive capillary obstruction associated with severe myocardial damage and is progressive after coronary reperfusion. It is not necessarily associated with the embolization of resistance arteries. In contrast, the embolization to small arteries and/or arterioles is usually transient and is not associated with capillary obstruction, and causes spotty microvascular damage in accordance with the embolized area in the risk area. If the embolized area is not too large, it can be compensated by the nonembolized area, including the risk area, as shown in the previous experimental model (12), and myocardial perfusion at the capillary level is preserved. Therefore, distal embolization does not have a marked impact on infarct size or recovery of left ventricular function and remodeling.
The small size of the study population limited the reliability of data regarding the incidence of substantial reduction of coronary blood flow, and the no-reflow phenomenon may have reduced the statistical power of the analyses. In the experimental study, we only used methylmethacrylate microspheres but not material from atheromatous plaques, which means that it is not clear that HITS accurately detect embolic particles. The DGW can quantify the number of embolic particles, but may not be able to assess their size. We performed intravenous administration of nicorandil and thrombectomy before the first BA in all patients, and additional thrombectomy and/or intracoronary injections of nicorandil and nitroprusside in cases of coronary flow reduction for the clinical reason, which would weaken the effect of embolization on myocardial damage. Pre-dilatation of BA may cause the friability of the lesion, which increases embolic particles after stenting. Further study is needed, such as comparison of the number of HITS between coronary thrombectomy plus direct stenting and pre-dilatation plus stenting.
The authors thank Mr. Shingo Takahara (Goodman Co., Ltd., Osaka, Japan) for the preparation of the experimental circuit system.
- Abbreviations and Acronyms
- balloon angioplasty
- Doppler guidewire
- high-intensity transient signals
- left ventricular end-diastolic index
- left ventricular ejection fraction
- left ventricular end-systolic index
- myocardial infarction
- percutaneous coronary intervention
- regional wall motion
- Thrombolysis In Myocardial Infarction
- Thrombolysis In Myocardial Infarction myocardial perfusion grade
- Received September 28, 2007.
- Revision received March 3, 2008.
- Accepted March 15, 2008.
- American College of Cardiology Foundation
- Bahrmann P.,
- Figulla H.R.,
- Wagner M.,
- et al.
- Okamura A.,
- Ito H.,
- Iwakura K.,
- et al.
- Bahrmann P.,
- Werner G.S.,
- Heusch G.,
- et al.
- Ito H.,
- Taniyama Y.,
- Iwakura K.,
- et al.
- Ragosta M.,
- Camarano G.,
- Kaul S.,
- et al.
- Gibson C.M.,
- Cannon C.P.,
- Daley W.L.,
- et al.
- van 't Hof A.W.,
- Liem A.,
- Suryapranata H.,
- et al.
- Ito H.,
- Tomooka T.,
- Sakai N.,
- et al.
- Hori M.,
- Inoue M.,
- Kitakaze M.,
- et al.
- Al-Mubarak N.,
- Roubin G.S.,
- Vitek J.J.,
- et al.
- Stone G.W.,
- Brodie B.R.,
- Griffin J.J.,
- on behalf of the Primary Angioplasty in Myocardial Infarction (PAMI) Investigators
- Stone G.W.,
- Webb J.,
- Cox D.A.,
- et al.,
- Enhanced Myocardial Efficacy and Recovery by Aspiration of Liberated Debris (EMERALD) Investigators
- Keller M.W.,
- Segal S.S.,
- Kaul S.,
- et al.