- Email: [email protected]
The clinical implications of no reflow demonstrated with intravenous perfluorocarbon containing microbubbles following restoration of Thrombolysis in Myocardial Infarction (TIMI) 3 flow in patients with acute myocardial infarction
Demonstrat;?d With Intravenous Perfluorocarbon Containing Microbubbles Following Restoration of Thrombolysis In Myocardial Infarction (TIM) 3 Flow in Patients With Acute Myocardial Infarction Thomas
Li, MD, Rebecca R. Porter, MD, Shoupin Ubeydulla if Deligonul, MD
Intravenous injections or infusions of perfluorocarbonexposed sonicated dextrose albumin microbubbles were given 2.4 + 1.6 days following acute myocardial infarction to 45 consecutive patients. Patients were divided into 3 groups: patients with Thrombolysis In Myocardial Infarction (TIMI) grade 3 an iographic flow but persistent m ocardial contrast de9ects by echocardiography (no rek ow), patients with TtMI 3 flow and myocardial contrast enhancement (reflow), and patients with TtMl grade 0 to 2 flow in the infarct vessel. Thirty-five patients had TtMl3 flow at the time of contrast study. Of these, 25 had evidence of reflow with intravenous contrast, whereas 10 (29%) still had contrast defects. At follow-
Oster,
ARDCS,
and
up, end-systolic volume index decreased significantly in patients who exhibited reflow (21 + 8 ml/m2 at baseline to 18 + 8 ml/m2 at follow-up; p = 0.04), whereas those with no reflow had a significant increase (26 -C 9 ml/m2 at baseline to 32 f 9 ml/m2 at follow-up; p = 0.006). A persistent contrast defect in the infarct zone demonstrated with intravenous ultrasound contrast following restoration of TtMl grade 3 flow in the infarct vessel identified patients likely to have deterioration in both regional and global systolic function. 01998 by Excerpta Medica, Inc. (Am J Cardiol 1998;82: 1173-l 177)
had an angiographic evaluation of the infarct vessel. Mean age of the patients was 59 + 12 years; 13 patients were women. Of the infarctions, 35 were myocardial infarction have been utilized to identify associated with ST elevation on the electrocardiothe “no-reflow” phenomena.’ Its presence has been gram. The infarct vessel by angiography was left associated with persistent left ventricular dysfunction anterior descending in 21 patients, right coronary arin the infarct zone at follow-up and a higher incidence tery in 11 patients, and left circumflex artery in 13 of complications following infarction.‘J The purpose patients. Twenty-six patients received thrombolytic of this study was to determine (1) whether intravenous therapy and 12 underwent successful angioplasty (n = ultrasound contrast could effectively define the inci- 6) or intravascular stenting (n = 6) before contrast dence of microvascular flow abnormalities within the studies. All patients signed informed consent, and the infarct zone following restoration of normal epicardial entire study protocol was approved by the Institutional flow in a consecutive series of patients, and (2) Review Board at our institution. whether this noninvasive method of identifying no Image acquisition: The intravenous contrast used reflow would identify patients more likely to exhibit for this study was perfluorocarbon-exposed sonicated deterioration of regional and global systolic function dextrose albumin (PESDA).4--6 The ultrasound conat follow-up. trast studies were performed 2.4 ? 1.6 days after the myocardial infarction (range 1 to 9) and within 1.3 + METHODS 1.2 days of cardiac catheterization (range 6 hours to 5 The study involved 45 consecutive patients who days). Intravenous PESDA was given to patients using experienced an acute myocardial infarction that was either a repeated bolus injection of either 0.00125 to diagnosed by enzyme criteria (serial elevation in cre- 0.0025 ml/kg in 28 patients, or as a continuous infuatine phosphokinase levels), and who subsequently sion of 0.05 or 0.1 ml/kg of PESDA in 80 ml of normal saline in 17 patients at 2 to 4 ml/min. UltraFrom +he Uriverslh/ of Nebraska Mecical Cewe-, &aha, Nebraska. sound images were obtained in parasternal (long and Maluscriot received Mar& 30, 1998; revised mowscript received short axis) and apical views (apical 4 chamber, 2 and acceoted June 26, 1998 chamber, and long axis) after each injection or during Address for reorilts: Tlomos R. Porter, ME, Universiy 0: ncnroskcl the continuous infusion. When using a bolus injection, Medical Cewer, 600 South L21d Street, Box 981 16.5, Omohc, ultrasound triggering was set to 1 frame every 2 carNebraska 68 198-226.5. E-mail: +rpcr:erQ~.nmc.edu.
ntracoronary injections of sonicated radiographic contrast following restoration of Thrombolysis In IMyocardial Infarction (TIMI) grade 3 flow in acute
01998 by Excerpt0 Medica, All riahts reserved.
Inc.
0002.9149/98/S 19.00 PII 50002-9149198100597-9
1173
TABLEI Ultrasound
Demographics of the Different Contrast and Angiography
Men/women Infarct location w anterior) Q waves No. receiving thrombolysis % Residual diameter stenosis TIMI frame count (s) Peak CPK (U) No. multivessel CAD Baseline LVEF p) CAD
= coronary
artery
disease;
Patient Groups Undergoing Following Acute Myocardial
TIMI 3 Reflow Group
TIMI 3 No Reflow Group
25 582 13 17/8 13 (52%) 16 (64%) 18 (72%) 51 236 302 10 1,188? 2,047 9 (36%) 582 10
10 59i13 7/3 7 (70%) 8 (73%) 7 (70%) 35 + 36 372 12 1,217 2 1,828 6 (60%) 53 + 11
CPK = creatine
kinase;
LVEF = left ventricular
disc cycles when measuring peak myocardial contrast enhancement. When using a continuous infusion, contrast enhancement was assessedat incremental triggering intervals beginning at 1 frame every cardiac cycle, and then once every 2, 4, and 6 cardiac cycles, with the longest pulsing interval equivalent to peak myocardia1 video intensity.7 These images were then stored on videotape for analysis of wall motion and contrast. Contrast analysis: Myocardial contrast enhancement inside and outside the infarct region following injections or infusions of PESDA was scored qualitatively and quantitatively. Two independent experienced observers reviewed the videotapes and scored peak contrast enhancement in 4 different regions: inferoposterior, lateral, anteroseptal, and apical regions. A visual score of 1 was given to any region exhibiting bright contrast enhancement, whereas a score of 2 was given to regions with reduced or patchy myocardial contrast enhancement compared with other regions. Regions that exhibited virtually no contrast enhancement were scored as a 3. A score of 2 or 3 within the infarct zone, despite TIM1 3 angiographic flow, was considered no reflow. The spatial extent of any contrast defect was planimetered using an offline review station (Tom-Tee, Louisville, Colorado). Background subtracted peak myocardial video intensity was also measured from digitized videotapes using a Tom-Tee review station. This software measures contrast enhancement over a 1 to 255 gray scale from a 150 square pixel region of interest placed within the mid-portion of each region (anteroseptal, apical, inferoposterior, and lateral). Wall motion was also graded in each region by each independent reviewer using a scoring system of 1 = normal, 2 = hypokinetic, 3 = akinetic, and 4 = dyskinetic, and a 16-segment model as described elsewhere.8 A wall motion score index was then computed by dividing the total wall motion score by the number of analyzable segments. Follow-up studies: Patients were asked to return for a repeat echocardiogram within 3 months of their hospital discharge. The follow-up study consisted of a repeat wall motion score and repeat biplane measure1174
THE AMERICAN
JOURNAL
OF CARDIOLOGYa
VOL.
ments of end-diastolic and end-systolic volume (Simpson’s Method of Discs) using the same views as the original study before hospital disTIMI O-2 charge. Each of these volume mea10 surements were then indexed to body 702 10 surface area. 8/2 Angiographic studies: Coronary 1 (10%) angiograms were performed at a time 6 (60%) 7 (70%) interval of 2.6 2 1.6 days following 97 f 6 the acute myocardial infarction. Measurements of stenosis diameter 1,220 2 1,726 were made in the infarct vessel as 9 (90%) well as in any of the noninfarct ves51 2 11 sels. Quantitative measurements on ejection fraction. the diagnostic angiograms were made with a hand-held electronic digital caliper (Tesa S.A., Renens, Switzerland) operated with custom-developed PC software. Twenty-six patients had angioplasty or intravascular stenting of the infarct vessel before the contrast study. In these patients, stenosis diameters before and after the intervention were analyzed with an automated border detection computer program (ARTREK, ImageComm, Santa Clara, California). The TIM1 grade of the angiographic flow in the infarct vessel was also determined using previously described criteria.9 In addition to this visual analysis of angiographic flow in the infarct vessel, a TIM1 frame count was also measured.lo This is computed by determining the number of cineframe counts needed for dye to reach a standardized distal landmark for each vessel.lo Statistical analysis: Patients were divided into 3 groups: group I were patients who exhibited normal (score = 1) contrast enhancement within the occluded perfusion bed after restoration of TIM1 3 flow (TIM1 3/reffow group). Group II patients were those who had persistent contrast defects (score 2 or 3), despite restoration of TIM1 3 flow (TIM1 3/no reflow group). Group III patients were those with TIM1 0 (n = 7), TIM1 1 (n = 2), or TIM1 2 (n = 1) flow in the infarct vessel at the time of study. Comparisons of categorical variables between these 3 groups were made with contingency tables, whereas comparsions of discrete variables were made with analysis of variance. Paired t testing was used to compare differences within groups from their initial study to the follow-up study. Intravenous Infarction
82
RESULTS There was no difference in peak myocardial video intensity within the noninfarct segments in patients who received bolus injections versus continuous infusions (51 t 12 U vs 50 ? 10 U, respectively). Of the 35 patients with TIM1 3 angiographic flow at the time of study, 10 demonstrated persistent echo contrast defects within the infarct region (score 2 or 3) following intravenous PESDA, whereas the remaining 25 patients had contrast enhancement (score of 1; both reviewers). As can be seen in Table I, there were no differences between these 2 groups in incidence of Q waves on the electrocardiogram, infarct location, reNOVEMBER
15,
1998
region, whereas 7 did not. Table III demonstrates that there were no significant differences in resting enddiastolic or end-systolic volume index or ejection fraction in these 2 groups before hospital discharge. However, patients who exhibited reflow with intravenous ultrasound contrast had significant decreases in both end-systolic volume and wall motion score index at the follow-up study, whereas patients with TIM1 3/no reflow had significant increases in both of these parameters. Left ventricular ejection fraction and wall motion score indexes at follow-up in the patients with TIM1 3/no reflow were nearly identical to those who had TIM1 0 to 2 flow in the infarct vessel. Figure 2 demonstrates a patient who still had a posterior wall defect extending to the apex, despite restoration of TIM1 3 flow to this infarct zone. Note that at folAoteroseptal MI: Reflow Aateroseptal MI: No Reflow low-up there was no improvement in wall motion score or ejection fraction. FIGURE 1. Examples from 2 different patients who had anteroInterobserver agreement: Of the 180 regions that septal and apical myocardial infarctions and similar wall motion were compared in the 45 patients, there was agreescore indexes following restoration of epicardial flow in the let3 anterior descending artery. The patient on the ri ht had a persisment on contrast score between the 2 independent tent perfusion defect (no reflow, arrowheads], w 9,ereas the pareviewers in 175 of the 180 regions (97%). There was tient on the left demonstrated reflow in the same regions (arrow96% agreement in wall motion assignment as well heads). MI = myocardial infarction. between the 2 reviewers. The mean difference in peak myocardial video intensity measurements by 2 differsidual percent diameter stenosis in the infarct vessel, ent reviewers was 4 +- 4 U (r = 0.94; SEE 5 U; p or TIM1 frame count. There was, however, a larger -CO.001; n = 24 comparisons). percentage of patients with 2- or 3-vessel coronary disease in the TIM1 3/no reflow group compared with the TIM1 3/reflow patients (p = 0.02). Figure 1 is an DISCUSSION Perfluorocarbon-containing microbubbles have example of reflow and no reflow in 2 patients who had apical infarctions and were studied after TIM1 3 flow been shown to produce myocardial contrast following intravenous injection or infusion when using intermithad been established in the infarct vessel. Background subtracted peak myocardial video in- tent harmonic imaging in selected patients.4-7 In this tensity was significantly higher in the infarct regions study, we demonstrated for the first time that the that visually exhibited reflow compared with those no-reflow phenomenon can be observed with intravewith TIM1 3/no reflow and those with TIM1 0 to 2 nous ultrasound contrast and intermittent harmonic flow (Table II). Visual detection of reflow was seen imaging in a consecutive series of patients following only in regions with a peak myocardial video intensity acute myocardial infarction. We observed that a sig(PMVI) of >2.5 U, whereas those with TIM1 3/no nificant percentage of patients (29%) exhibited this reflow had a range of PMVIs that were between 6 microvascular flow abnormality despite TIM1 grade 3 (visual score of 3) and 32 units (visual score of 2). The flow in the infarct vessel. This study also demonspatial extent of the no-reflow zone in patients with strated that these contrast defects noninvasively idenTIM1 3 flow did not differ from those with TIM1 0 to tified which patients would experience both failure of recovery of regional systolic function, as well as de2 angiographic flow. terioration in global systolic function. Changes in left ventricular regional and global sysThe incidence of no reflow in TIM1 3 patients in tolic function at follow-up: Twenty-three of the 35 patients who had TIM1 3 angiographic flow could be our study was larger than the incidence described in reached for follow-up studies at a mean time interval the original studies examining patients with TIM1 3 of 9 +- 4 weeks following their infarction. Twelve flow using intracoronary ultrasound contrast.‘,2 One patients did not have follow-up studies because of reason for this may be that we chose consecutive inability to reach the patient (n = S), death (n = 3), or patients and thus had patients with more advanced reinfarction within the original infarct zone (n = 1). disease. Only 20% of the patients with single-vessel Four of the 9 patients who had TIM1 0 or 2 flow in the disease in our study exhibited no reflow. This is siminfarct vessel at the time of study also had follow-up ilar to what intracoronary contrast studies involving studies. The remaining 5 did not have repeat evalua- patients with exclusively single-vessel coronary distions because they either had bypass surgery of the ease have shown. However, 40% of the patients with infarct vessel (n = 2) or could not be reached for multivessel coronary disease in our study had no reflow after TIM1 3 flow was restored (Table I). This follow-up (n = 3). Sixteen of the 23 patients with TIM1 3 angio- parallels more recent studies with intracoronary ultragraphic how who had follow-up echocardiograms had sound contrast that have observed a higher incidence evidence of contrast enhancement within the infarct of contrast defects in the infarct zone following corCORONARY ARTERY DISEASE/NO
REFLOW WITH INTRAVENOUS ULTRASOUND CONTRAST
1175
TABLE II Differences Groups.
in Peak Myocardial Video Intensity in the Infarct Zone and Adjacent Slope of Contrast Enhancement (bubble velocity) Is Shown for Patients
The Initial
Perfusion Bed in the Different Patient Who Had o Continuous Infusion Infarct Zone
TIMI 3: reflow TIMI 3: no reflow TIMI O-2 *p < 0.05
compared
with
TABLE Ill Baseline Contrast
n
Spatial Extent of No Reflow
PMVI
25 10 10
0 k o* 4.1 k 1.7 4.6 f 1.2
40 k 7* 20+ 10 18+7
the other
group
(l-way
and Follow-up Indexes After Acute Myocardial
Findings
analysis
Initial Slope
of variance).
of Regional Infarction
and Globol
Systolic
Function
in Patients
With
TIMI 3: reflow TIMI 3: no reflow TIMI O-2 ‘p < 0.05 +p
<
0.05
compared
EDVI = left ventricular area;
WMSI
= wall
20 -+ 8 26 k 9 26 k 11
49 k 11 60% 17 57* 12 with
compared with
no reflow other
enddiastolic motion
EF
ESVI
group
group
(l-way
(l-way
volume
indexed
WMSI
592 10 55r 13 53% 15 analysis
analysis
Different
Angiogrophic
and
Follow-up
Baseline
EDVI
0.0197 + 0.0084* 0.0089 k 0.0045 0.0064 k 0.0020
1.3 1.6 1.3
+ .3 + .3 k .5
EDVI
ESVI
EF
WMSI
492 12 602 15 54 t 8
18?8* 32 f 9 29t 14
63 2 9* 46 +- 5 51 t 17
1.1 + 2+ 1.72 .3 1.5 k .7
of variance).
of variance).
to body surface
oreo;
EF = ejection
fraction;
ESVI = left ventricular
end-systolic
volume
indexed
to body surface
score index.
nificantly increased left ventricular ejection fraction at 28 days after myocardial infarction in the patients who demonstrated reflow with ultrasound contrast early after infarction. Kenner et al11found that all segments that exhibited reflow with intracoronary contrast following reperfusion were functioning normally 1 month after infarction. Ragosta et al3 demonstrated that follow-up segmental wall motion correlated directly with the degree of contrast enhancement sterior perfusion abnormality (arrow/tea& in a paFIGURE 2. An example of a tient who had TiMI 3 epicar r lal flow in the left circumflex artery followin an acute in the infarct zone early after infarcposterior m ocardial infarction. As can be seen, the wall motion score dl*% not imtion. In this study, the contrast enprove at fo Kow-up. ESV = end-systolic volume; EF = ejection fraction; WMSI = wall hancement in the reperfused bed folmotion score index. lowing intravenous PESDA indicated not only that regional wall motion would improve, but also that global onary reperfusion in patients with multivessel coro- left ventricular systolic function would improve. Connary artery disease.3J1 versely, lack of contrast enhancement was predictive This higher incidence of no reflow in patients with of no improvement or even deterioration in both remultivessel disease may indicate the importance of gional and global systolic function at follow-up. The collateral flow from adjacent perfusion beds in deter- TIM1 3/no reflow patient group actually behaved in a mining whether microvascular necrosis will occur in manner that was similar to patients with less than the distribution of the infarct vessel. Previous studies TIM1 3 flow in the infarct-related arteries. These findutilizing intracoronary ultrasound contrast have dem- ings indicate the predictive power of intravenous myoonstrated that nearly 80% of patients who exhibit cardial contrast echocardiography over angiographic collateral perfusion to the infarct bed will have resto- assessmentsof flow in defining the eventual outcome ration of function within this region at follow-up.12 of patients following acute myocardial infarction. The follow-up changes in wall motion score and In conclusion, intravenous perfluorocarbon microleft ventricular ejection fraction of patients with TIM1 bubbles can noninvasively detect microvascular flow 3 reflow versus those with TIM1 3 no reflow (Table abnormalities that exist following adequate angioIII) following intravenous contrast are also similar to graphic reperfusion in patients with acute myocardial what has been observed with intracoronary injections infarction. Similar to studies involving intracoronary of ultrasound contrast.2JJ1 Ito et al2 observed a sig- microbubbles, no reflow demonstrated with intrave1176
THE
AMERICAN
JOURNAL
OF
CARDIOLOGY@’
VOL.
82
NOVEMBER
15,
1998
nous microbubbles identifies patients who are likely to experience further deterioration in regional and global systolic function at follow-up.
6. Porter TR, Li S, Kricsfeld D, Armbmster RW. Detection of myocardial perfusion in multiple echocardiographic windows with one intravenous injection of microbubbles using transient response second harmonic imaging. J Am Co/l 1997;29:791-799. Cardiol 7. Wei K, Jayaweera AR, Firoozan S, Linka A, Skyba DM, Kaul S. Quantification of myocardial blood flow with ultrasound-induced destruction of microbubbles administered as a constant venous infusion. Circularion 1998;97:473-
1. Ito H, Okamura A, Iwakura K, Masuyama T, Hori M, Takiuchi S, Negoro S, Nakatsuchi Y, Taniyama Y, Higashino Y, Fujii K, Minamino T. Myocardial perfusion patterns related to thrombolysis in myocardial infarction perfusion grades after coronary angioplasty in patients with acute anterior wall myocardial infarction. Circularion 1996;93:1993-1999, 2. Ito H, Maruyama A, Iwakura K, Takiuchi S, Masuyama T, Hori M, Higashino Y, Fujii K, Minamino T. Clinical implications of the ‘no reflow’ phenomenon. A predictor of complications and left ventricular remodeling in rep&used anterior wall myocardial infarction. Circularion 1996; 93:223-228. 3. Ragosta M, Camarano G, Kaul S, Powers ER, Sarembock IJ, Gimple LW. Microvascular integrity indicates myocellular viability in patients with recent myocardial infarction. Circulation 1994;89:2562-2569. 4. Porter TR, Xie F. Transient myocardial contrast following initial exposure to diagnostic ultrasound pressures with minute doses of intravenously injected 1995;92: microbubbles: demonstration and potential mechanisms. Circulation
483.
2391-2395.
5. Porter TR, Xie F, Kricsfeld D, Armbruster RW. Improved myocardial contrast with second harmonic transient ultrasound contrast response imaging in humans using intravenous perfluorocarbon-exposed sonicated dextrose albumin. J Am Coil Cardiol 1996;27: 1497-1501.
CORONARY ARTERY DISEASE/NO
8. Sawada SG, Segar DS, Ryan T, Brown SE, Dohan AM, Williams R, Fineberg NS, Armstrong WF, Feigenbaum H. Echocardiographic detection of coronary artery disease during dobutamine infusion. Circularion 1991;83: 1605-1614. 9. Chesebro JH, Knattemd G, Roberts R, Borer J, Cohen LS, Dale” J, Dodge HT, Francis CK, Hillis D. Ludbrook P, et al. Thrombolysis in myocardial infarction (TIMI) trial, Phase 1: a comparison between intravenous tissue plasminogen activator and intravenous streptokinase. Circularion 1987;76: 142-154. 10. Gibson CM, Cannon CP, D&y WL, Dodge Jr, JT, Alexander B, Marble SJ, McCabe CH, Raymond L. Fortin T, Poole WK, Braunwald E, for the TlMl 4 Study Group. TIM1 Frame Count: a quantitative method of assessing coronary artery flow. Circulation 1996:93:879-888. 11. Kenner MD, Zajac EJ, Kondos GT, Dave R. Winkelmann JW, Joftus J, Laucevicius A, Kybarskis A, Berukstis E, Urbonas A, F&stein SB. Ability of the no-reflow phenomenon during an acute myocardial infarction to predict left ventricular dysfunction at one-month follow-up. Am / Cardiol 1995:76:861-868. 12. Sabia PJ. Powers ER, Ragosta M, Sarembock IJ, BunveIl LR, Kaul S. An association between collateral blood flow and myocardial viability in patients with recent myocardial infarction. N Eng .I Med 1992:327:1825-1831.
REFLOW WITH INTRAVENOUS ULTRASOUND CONTRAST
1177
Li, MD, Rebecca R. Porter, MD, Shoupin Ubeydulla if Deligonul, MD
Intravenous injections or infusions of perfluorocarbonexposed sonicated dextrose albumin microbubbles were given 2.4 + 1.6 days following acute myocardial infarction to 45 consecutive patients. Patients were divided into 3 groups: patients with Thrombolysis In Myocardial Infarction (TIMI) grade 3 an iographic flow but persistent m ocardial contrast de9ects by echocardiography (no rek ow), patients with TtMI 3 flow and myocardial contrast enhancement (reflow), and patients with TtMl grade 0 to 2 flow in the infarct vessel. Thirty-five patients had TtMl3 flow at the time of contrast study. Of these, 25 had evidence of reflow with intravenous contrast, whereas 10 (29%) still had contrast defects. At follow-
Oster,
ARDCS,
and
up, end-systolic volume index decreased significantly in patients who exhibited reflow (21 + 8 ml/m2 at baseline to 18 + 8 ml/m2 at follow-up; p = 0.04), whereas those with no reflow had a significant increase (26 -C 9 ml/m2 at baseline to 32 f 9 ml/m2 at follow-up; p = 0.006). A persistent contrast defect in the infarct zone demonstrated with intravenous ultrasound contrast following restoration of TtMl grade 3 flow in the infarct vessel identified patients likely to have deterioration in both regional and global systolic function. 01998 by Excerpta Medica, Inc. (Am J Cardiol 1998;82: 1173-l 177)
had an angiographic evaluation of the infarct vessel. Mean age of the patients was 59 + 12 years; 13 patients were women. Of the infarctions, 35 were myocardial infarction have been utilized to identify associated with ST elevation on the electrocardiothe “no-reflow” phenomena.’ Its presence has been gram. The infarct vessel by angiography was left associated with persistent left ventricular dysfunction anterior descending in 21 patients, right coronary arin the infarct zone at follow-up and a higher incidence tery in 11 patients, and left circumflex artery in 13 of complications following infarction.‘J The purpose patients. Twenty-six patients received thrombolytic of this study was to determine (1) whether intravenous therapy and 12 underwent successful angioplasty (n = ultrasound contrast could effectively define the inci- 6) or intravascular stenting (n = 6) before contrast dence of microvascular flow abnormalities within the studies. All patients signed informed consent, and the infarct zone following restoration of normal epicardial entire study protocol was approved by the Institutional flow in a consecutive series of patients, and (2) Review Board at our institution. whether this noninvasive method of identifying no Image acquisition: The intravenous contrast used reflow would identify patients more likely to exhibit for this study was perfluorocarbon-exposed sonicated deterioration of regional and global systolic function dextrose albumin (PESDA).4--6 The ultrasound conat follow-up. trast studies were performed 2.4 ? 1.6 days after the myocardial infarction (range 1 to 9) and within 1.3 + METHODS 1.2 days of cardiac catheterization (range 6 hours to 5 The study involved 45 consecutive patients who days). Intravenous PESDA was given to patients using experienced an acute myocardial infarction that was either a repeated bolus injection of either 0.00125 to diagnosed by enzyme criteria (serial elevation in cre- 0.0025 ml/kg in 28 patients, or as a continuous infuatine phosphokinase levels), and who subsequently sion of 0.05 or 0.1 ml/kg of PESDA in 80 ml of normal saline in 17 patients at 2 to 4 ml/min. UltraFrom +he Uriverslh/ of Nebraska Mecical Cewe-, &aha, Nebraska. sound images were obtained in parasternal (long and Maluscriot received Mar& 30, 1998; revised mowscript received short axis) and apical views (apical 4 chamber, 2 and acceoted June 26, 1998 chamber, and long axis) after each injection or during Address for reorilts: Tlomos R. Porter, ME, Universiy 0: ncnroskcl the continuous infusion. When using a bolus injection, Medical Cewer, 600 South L21d Street, Box 981 16.5, Omohc, ultrasound triggering was set to 1 frame every 2 carNebraska 68 198-226.5. E-mail: +rpcr:erQ~.nmc.edu.
ntracoronary injections of sonicated radiographic contrast following restoration of Thrombolysis In IMyocardial Infarction (TIMI) grade 3 flow in acute
01998 by Excerpt0 Medica, All riahts reserved.
Inc.
0002.9149/98/S 19.00 PII 50002-9149198100597-9
1173
TABLEI Ultrasound
Demographics of the Different Contrast and Angiography
Men/women Infarct location w anterior) Q waves No. receiving thrombolysis % Residual diameter stenosis TIMI frame count (s) Peak CPK (U) No. multivessel CAD Baseline LVEF p) CAD
= coronary
artery
disease;
Patient Groups Undergoing Following Acute Myocardial
TIMI 3 Reflow Group
TIMI 3 No Reflow Group
25 582 13 17/8 13 (52%) 16 (64%) 18 (72%) 51 236 302 10 1,188? 2,047 9 (36%) 582 10
10 59i13 7/3 7 (70%) 8 (73%) 7 (70%) 35 + 36 372 12 1,217 2 1,828 6 (60%) 53 + 11
CPK = creatine
kinase;
LVEF = left ventricular
disc cycles when measuring peak myocardial contrast enhancement. When using a continuous infusion, contrast enhancement was assessedat incremental triggering intervals beginning at 1 frame every cardiac cycle, and then once every 2, 4, and 6 cardiac cycles, with the longest pulsing interval equivalent to peak myocardia1 video intensity.7 These images were then stored on videotape for analysis of wall motion and contrast. Contrast analysis: Myocardial contrast enhancement inside and outside the infarct region following injections or infusions of PESDA was scored qualitatively and quantitatively. Two independent experienced observers reviewed the videotapes and scored peak contrast enhancement in 4 different regions: inferoposterior, lateral, anteroseptal, and apical regions. A visual score of 1 was given to any region exhibiting bright contrast enhancement, whereas a score of 2 was given to regions with reduced or patchy myocardial contrast enhancement compared with other regions. Regions that exhibited virtually no contrast enhancement were scored as a 3. A score of 2 or 3 within the infarct zone, despite TIM1 3 angiographic flow, was considered no reflow. The spatial extent of any contrast defect was planimetered using an offline review station (Tom-Tee, Louisville, Colorado). Background subtracted peak myocardial video intensity was also measured from digitized videotapes using a Tom-Tee review station. This software measures contrast enhancement over a 1 to 255 gray scale from a 150 square pixel region of interest placed within the mid-portion of each region (anteroseptal, apical, inferoposterior, and lateral). Wall motion was also graded in each region by each independent reviewer using a scoring system of 1 = normal, 2 = hypokinetic, 3 = akinetic, and 4 = dyskinetic, and a 16-segment model as described elsewhere.8 A wall motion score index was then computed by dividing the total wall motion score by the number of analyzable segments. Follow-up studies: Patients were asked to return for a repeat echocardiogram within 3 months of their hospital discharge. The follow-up study consisted of a repeat wall motion score and repeat biplane measure1174
THE AMERICAN
JOURNAL
OF CARDIOLOGYa
VOL.
ments of end-diastolic and end-systolic volume (Simpson’s Method of Discs) using the same views as the original study before hospital disTIMI O-2 charge. Each of these volume mea10 surements were then indexed to body 702 10 surface area. 8/2 Angiographic studies: Coronary 1 (10%) angiograms were performed at a time 6 (60%) 7 (70%) interval of 2.6 2 1.6 days following 97 f 6 the acute myocardial infarction. Measurements of stenosis diameter 1,220 2 1,726 were made in the infarct vessel as 9 (90%) well as in any of the noninfarct ves51 2 11 sels. Quantitative measurements on ejection fraction. the diagnostic angiograms were made with a hand-held electronic digital caliper (Tesa S.A., Renens, Switzerland) operated with custom-developed PC software. Twenty-six patients had angioplasty or intravascular stenting of the infarct vessel before the contrast study. In these patients, stenosis diameters before and after the intervention were analyzed with an automated border detection computer program (ARTREK, ImageComm, Santa Clara, California). The TIM1 grade of the angiographic flow in the infarct vessel was also determined using previously described criteria.9 In addition to this visual analysis of angiographic flow in the infarct vessel, a TIM1 frame count was also measured.lo This is computed by determining the number of cineframe counts needed for dye to reach a standardized distal landmark for each vessel.lo Statistical analysis: Patients were divided into 3 groups: group I were patients who exhibited normal (score = 1) contrast enhancement within the occluded perfusion bed after restoration of TIM1 3 flow (TIM1 3/reffow group). Group II patients were those who had persistent contrast defects (score 2 or 3), despite restoration of TIM1 3 flow (TIM1 3/no reflow group). Group III patients were those with TIM1 0 (n = 7), TIM1 1 (n = 2), or TIM1 2 (n = 1) flow in the infarct vessel at the time of study. Comparisons of categorical variables between these 3 groups were made with contingency tables, whereas comparsions of discrete variables were made with analysis of variance. Paired t testing was used to compare differences within groups from their initial study to the follow-up study. Intravenous Infarction
82
RESULTS There was no difference in peak myocardial video intensity within the noninfarct segments in patients who received bolus injections versus continuous infusions (51 t 12 U vs 50 ? 10 U, respectively). Of the 35 patients with TIM1 3 angiographic flow at the time of study, 10 demonstrated persistent echo contrast defects within the infarct region (score 2 or 3) following intravenous PESDA, whereas the remaining 25 patients had contrast enhancement (score of 1; both reviewers). As can be seen in Table I, there were no differences between these 2 groups in incidence of Q waves on the electrocardiogram, infarct location, reNOVEMBER
15,
1998
region, whereas 7 did not. Table III demonstrates that there were no significant differences in resting enddiastolic or end-systolic volume index or ejection fraction in these 2 groups before hospital discharge. However, patients who exhibited reflow with intravenous ultrasound contrast had significant decreases in both end-systolic volume and wall motion score index at the follow-up study, whereas patients with TIM1 3/no reflow had significant increases in both of these parameters. Left ventricular ejection fraction and wall motion score indexes at follow-up in the patients with TIM1 3/no reflow were nearly identical to those who had TIM1 0 to 2 flow in the infarct vessel. Figure 2 demonstrates a patient who still had a posterior wall defect extending to the apex, despite restoration of TIM1 3 flow to this infarct zone. Note that at folAoteroseptal MI: Reflow Aateroseptal MI: No Reflow low-up there was no improvement in wall motion score or ejection fraction. FIGURE 1. Examples from 2 different patients who had anteroInterobserver agreement: Of the 180 regions that septal and apical myocardial infarctions and similar wall motion were compared in the 45 patients, there was agreescore indexes following restoration of epicardial flow in the let3 anterior descending artery. The patient on the ri ht had a persisment on contrast score between the 2 independent tent perfusion defect (no reflow, arrowheads], w 9,ereas the pareviewers in 175 of the 180 regions (97%). There was tient on the left demonstrated reflow in the same regions (arrow96% agreement in wall motion assignment as well heads). MI = myocardial infarction. between the 2 reviewers. The mean difference in peak myocardial video intensity measurements by 2 differsidual percent diameter stenosis in the infarct vessel, ent reviewers was 4 +- 4 U (r = 0.94; SEE 5 U; p or TIM1 frame count. There was, however, a larger -CO.001; n = 24 comparisons). percentage of patients with 2- or 3-vessel coronary disease in the TIM1 3/no reflow group compared with the TIM1 3/reflow patients (p = 0.02). Figure 1 is an DISCUSSION Perfluorocarbon-containing microbubbles have example of reflow and no reflow in 2 patients who had apical infarctions and were studied after TIM1 3 flow been shown to produce myocardial contrast following intravenous injection or infusion when using intermithad been established in the infarct vessel. Background subtracted peak myocardial video in- tent harmonic imaging in selected patients.4-7 In this tensity was significantly higher in the infarct regions study, we demonstrated for the first time that the that visually exhibited reflow compared with those no-reflow phenomenon can be observed with intravewith TIM1 3/no reflow and those with TIM1 0 to 2 nous ultrasound contrast and intermittent harmonic flow (Table II). Visual detection of reflow was seen imaging in a consecutive series of patients following only in regions with a peak myocardial video intensity acute myocardial infarction. We observed that a sig(PMVI) of >2.5 U, whereas those with TIM1 3/no nificant percentage of patients (29%) exhibited this reflow had a range of PMVIs that were between 6 microvascular flow abnormality despite TIM1 grade 3 (visual score of 3) and 32 units (visual score of 2). The flow in the infarct vessel. This study also demonspatial extent of the no-reflow zone in patients with strated that these contrast defects noninvasively idenTIM1 3 flow did not differ from those with TIM1 0 to tified which patients would experience both failure of recovery of regional systolic function, as well as de2 angiographic flow. terioration in global systolic function. Changes in left ventricular regional and global sysThe incidence of no reflow in TIM1 3 patients in tolic function at follow-up: Twenty-three of the 35 patients who had TIM1 3 angiographic flow could be our study was larger than the incidence described in reached for follow-up studies at a mean time interval the original studies examining patients with TIM1 3 of 9 +- 4 weeks following their infarction. Twelve flow using intracoronary ultrasound contrast.‘,2 One patients did not have follow-up studies because of reason for this may be that we chose consecutive inability to reach the patient (n = S), death (n = 3), or patients and thus had patients with more advanced reinfarction within the original infarct zone (n = 1). disease. Only 20% of the patients with single-vessel Four of the 9 patients who had TIM1 0 or 2 flow in the disease in our study exhibited no reflow. This is siminfarct vessel at the time of study also had follow-up ilar to what intracoronary contrast studies involving studies. The remaining 5 did not have repeat evalua- patients with exclusively single-vessel coronary distions because they either had bypass surgery of the ease have shown. However, 40% of the patients with infarct vessel (n = 2) or could not be reached for multivessel coronary disease in our study had no reflow after TIM1 3 flow was restored (Table I). This follow-up (n = 3). Sixteen of the 23 patients with TIM1 3 angio- parallels more recent studies with intracoronary ultragraphic how who had follow-up echocardiograms had sound contrast that have observed a higher incidence evidence of contrast enhancement within the infarct of contrast defects in the infarct zone following corCORONARY ARTERY DISEASE/NO
REFLOW WITH INTRAVENOUS ULTRASOUND CONTRAST
1175
TABLE II Differences Groups.
in Peak Myocardial Video Intensity in the Infarct Zone and Adjacent Slope of Contrast Enhancement (bubble velocity) Is Shown for Patients
The Initial
Perfusion Bed in the Different Patient Who Had o Continuous Infusion Infarct Zone
TIMI 3: reflow TIMI 3: no reflow TIMI O-2 *p < 0.05
compared
with
TABLE Ill Baseline Contrast
n
Spatial Extent of No Reflow
PMVI
25 10 10
0 k o* 4.1 k 1.7 4.6 f 1.2
40 k 7* 20+ 10 18+7
the other
group
(l-way
and Follow-up Indexes After Acute Myocardial
Findings
analysis
Initial Slope
of variance).
of Regional Infarction
and Globol
Systolic
Function
in Patients
With
TIMI 3: reflow TIMI 3: no reflow TIMI O-2 ‘p < 0.05 +p
<
0.05
compared
EDVI = left ventricular area;
WMSI
= wall
20 -+ 8 26 k 9 26 k 11
49 k 11 60% 17 57* 12 with
compared with
no reflow other
enddiastolic motion
EF
ESVI
group
group
(l-way
(l-way
volume
indexed
WMSI
592 10 55r 13 53% 15 analysis
analysis
Different
Angiogrophic
and
Follow-up
Baseline
EDVI
0.0197 + 0.0084* 0.0089 k 0.0045 0.0064 k 0.0020
1.3 1.6 1.3
+ .3 + .3 k .5
EDVI
ESVI
EF
WMSI
492 12 602 15 54 t 8
18?8* 32 f 9 29t 14
63 2 9* 46 +- 5 51 t 17
1.1 + 2+ 1.72 .3 1.5 k .7
of variance).
of variance).
to body surface
oreo;
EF = ejection
fraction;
ESVI = left ventricular
end-systolic
volume
indexed
to body surface
score index.
nificantly increased left ventricular ejection fraction at 28 days after myocardial infarction in the patients who demonstrated reflow with ultrasound contrast early after infarction. Kenner et al11found that all segments that exhibited reflow with intracoronary contrast following reperfusion were functioning normally 1 month after infarction. Ragosta et al3 demonstrated that follow-up segmental wall motion correlated directly with the degree of contrast enhancement sterior perfusion abnormality (arrow/tea& in a paFIGURE 2. An example of a tient who had TiMI 3 epicar r lal flow in the left circumflex artery followin an acute in the infarct zone early after infarcposterior m ocardial infarction. As can be seen, the wall motion score dl*% not imtion. In this study, the contrast enprove at fo Kow-up. ESV = end-systolic volume; EF = ejection fraction; WMSI = wall hancement in the reperfused bed folmotion score index. lowing intravenous PESDA indicated not only that regional wall motion would improve, but also that global onary reperfusion in patients with multivessel coro- left ventricular systolic function would improve. Connary artery disease.3J1 versely, lack of contrast enhancement was predictive This higher incidence of no reflow in patients with of no improvement or even deterioration in both remultivessel disease may indicate the importance of gional and global systolic function at follow-up. The collateral flow from adjacent perfusion beds in deter- TIM1 3/no reflow patient group actually behaved in a mining whether microvascular necrosis will occur in manner that was similar to patients with less than the distribution of the infarct vessel. Previous studies TIM1 3 flow in the infarct-related arteries. These findutilizing intracoronary ultrasound contrast have dem- ings indicate the predictive power of intravenous myoonstrated that nearly 80% of patients who exhibit cardial contrast echocardiography over angiographic collateral perfusion to the infarct bed will have resto- assessmentsof flow in defining the eventual outcome ration of function within this region at follow-up.12 of patients following acute myocardial infarction. The follow-up changes in wall motion score and In conclusion, intravenous perfluorocarbon microleft ventricular ejection fraction of patients with TIM1 bubbles can noninvasively detect microvascular flow 3 reflow versus those with TIM1 3 no reflow (Table abnormalities that exist following adequate angioIII) following intravenous contrast are also similar to graphic reperfusion in patients with acute myocardial what has been observed with intracoronary injections infarction. Similar to studies involving intracoronary of ultrasound contrast.2JJ1 Ito et al2 observed a sig- microbubbles, no reflow demonstrated with intrave1176
THE
AMERICAN
JOURNAL
OF
CARDIOLOGY@’
VOL.
82
NOVEMBER
15,
1998
nous microbubbles identifies patients who are likely to experience further deterioration in regional and global systolic function at follow-up.
6. Porter TR, Li S, Kricsfeld D, Armbmster RW. Detection of myocardial perfusion in multiple echocardiographic windows with one intravenous injection of microbubbles using transient response second harmonic imaging. J Am Co/l 1997;29:791-799. Cardiol 7. Wei K, Jayaweera AR, Firoozan S, Linka A, Skyba DM, Kaul S. Quantification of myocardial blood flow with ultrasound-induced destruction of microbubbles administered as a constant venous infusion. Circularion 1998;97:473-
1. Ito H, Okamura A, Iwakura K, Masuyama T, Hori M, Takiuchi S, Negoro S, Nakatsuchi Y, Taniyama Y, Higashino Y, Fujii K, Minamino T. Myocardial perfusion patterns related to thrombolysis in myocardial infarction perfusion grades after coronary angioplasty in patients with acute anterior wall myocardial infarction. Circularion 1996;93:1993-1999, 2. Ito H, Maruyama A, Iwakura K, Takiuchi S, Masuyama T, Hori M, Higashino Y, Fujii K, Minamino T. Clinical implications of the ‘no reflow’ phenomenon. A predictor of complications and left ventricular remodeling in rep&used anterior wall myocardial infarction. Circularion 1996; 93:223-228. 3. Ragosta M, Camarano G, Kaul S, Powers ER, Sarembock IJ, Gimple LW. Microvascular integrity indicates myocellular viability in patients with recent myocardial infarction. Circulation 1994;89:2562-2569. 4. Porter TR, Xie F. Transient myocardial contrast following initial exposure to diagnostic ultrasound pressures with minute doses of intravenously injected 1995;92: microbubbles: demonstration and potential mechanisms. Circulation
483.
2391-2395.
5. Porter TR, Xie F, Kricsfeld D, Armbruster RW. Improved myocardial contrast with second harmonic transient ultrasound contrast response imaging in humans using intravenous perfluorocarbon-exposed sonicated dextrose albumin. J Am Coil Cardiol 1996;27: 1497-1501.
CORONARY ARTERY DISEASE/NO
8. Sawada SG, Segar DS, Ryan T, Brown SE, Dohan AM, Williams R, Fineberg NS, Armstrong WF, Feigenbaum H. Echocardiographic detection of coronary artery disease during dobutamine infusion. Circularion 1991;83: 1605-1614. 9. Chesebro JH, Knattemd G, Roberts R, Borer J, Cohen LS, Dale” J, Dodge HT, Francis CK, Hillis D. Ludbrook P, et al. Thrombolysis in myocardial infarction (TIMI) trial, Phase 1: a comparison between intravenous tissue plasminogen activator and intravenous streptokinase. Circularion 1987;76: 142-154. 10. Gibson CM, Cannon CP, D&y WL, Dodge Jr, JT, Alexander B, Marble SJ, McCabe CH, Raymond L. Fortin T, Poole WK, Braunwald E, for the TlMl 4 Study Group. TIM1 Frame Count: a quantitative method of assessing coronary artery flow. Circulation 1996:93:879-888. 11. Kenner MD, Zajac EJ, Kondos GT, Dave R. Winkelmann JW, Joftus J, Laucevicius A, Kybarskis A, Berukstis E, Urbonas A, F&stein SB. Ability of the no-reflow phenomenon during an acute myocardial infarction to predict left ventricular dysfunction at one-month follow-up. Am / Cardiol 1995:76:861-868. 12. Sabia PJ. Powers ER, Ragosta M, Sarembock IJ, BunveIl LR, Kaul S. An association between collateral blood flow and myocardial viability in patients with recent myocardial infarction. N Eng .I Med 1992:327:1825-1831.
REFLOW WITH INTRAVENOUS ULTRASOUND CONTRAST
1177