Impact of Time to Treatment on Myocardial Reperfusion and Infarct Size With Primary Percutaneous Coronary Intervention for Acute Myocardial Infarction (from the EMERALD Trial)

Impact of Time to Treatment on Myocardial Reperfusion and Infarct Size With Primary Percutaneous Coronary Intervention for Acute Myocardial Infarction (from the EMERALD Trial) Bruce R. Brodie, MDa,*, John Webb, MDd, David A. Cox, MDb, Mansoor Qureshi, MDe, Anna Kalynych, MDg, Mark Turco, MDh, Heinz P. Schultheiss, MDi, Daniel Dulas, MDf, Barry Rutherford, MDj, David Antoniucci, MDk, Tom Stuckey, MDa, Mitch Krucoff, MDc, Raymond Gibbons, MDl, Alexandra Lansky, MDm, Yingbo Na, MDm, Roxana Mehran, MDm, and Gregg W Stone, MDm, for the EMERALD Investigators The impact of time to treatment on outcomes after primary percutaneous coronary intervention (PCI) is controversial, and there are few data about time to treatment and infarct size. The EMERALD trial randomly assigned 501 high-risk patients with ST-elevation myocardial infarction undergoing primary PCI to stenting with or without GuardWire (Medtronic, Santa Rosa, California) distal protection. Infarct size using sestamibi imaging at 5 to 14 days and clinical outcomes were examined by time to treatment. There were no differences in outcomes between distal protection and control patients. Shorter time to reperfusion (<2 vs 2 to 3 vs >3 to 4 vs >4 hours) was associated with smaller infarct size (2% vs 9% vs 12% vs 11%, p ⴝ 0.026), trends for better myocardial blush (p ⴝ 0.08), and lower 6-month mortality rates (0% vs 0% vs 2.4% vs 5.3%, p ⴝ 0.06). Incremental delays in reperfusion after 2 hours had little impact on infarct size. Shorter time to reperfusion impacted on infarct size in patients with anterior infarction (0% vs 17% vs 20.5% vs 30.5%, p ⴝ 0.026), but not nonanterior infarction (3% vs 7% vs 7.5% vs 10%, p ⴝ 0.23, p ⴝ 0.022 for interaction). In conclusion, very early reperfusion with primary PCI is associated with smaller infarct size and has a much greater impact in anterior versus nonanterior infarction. Incremental delays in reperfusion after 2 hours have less effect on infarct size. These data have implications regarding the triage of patients for primary PCI. © 2007 Elsevier Inc. All rights reserved. (Am J Cardiol 2007;99:1680 –1686)

The relation between time to treatment and clinical outcomes after primary percutaneous coronary intervention (PCI) for patients with ST-segment elevation acute myocardial infarction (STEMI) has been controversial,1–7 and there are few data regarding the relation between time to treatment with primary PCI and infarct size.8 –11 Studies that evaluated the impact of time to treatment with primary PCI on infarct size had only small numbers of patients and yielded conflicting results.8 –10 The Enhanced Myocardial Efficacy and Removal by Aspiration of Liberated Debris

(EMERALD) Trial evaluated the impact of primary PCI with and without distal coronary protection on infarct size measured using technetium-99m sestamibi scintigraphy.12 This study did not show an effect of distal protection on infarct size, but provided an opportunity to evaluate the impact of time to treatment with primary PCI on infarct size in a large patient population. We evaluated the impact of time to treatment with primary PCI on outcomes in the EMERALD Trial with special emphasis on myocardial infarct size.

a LeBauer Cardiovascular Research Foundation and Moses Cone Heart and Vascular Center, Greensboro; bMid Carolina Cardiology, Charlotte; c Duke Clinical Research Institute, Durham, North Carolina; dSt. Paul’s Hospital, Vancouver, Canada; eSt. Joseph Mercy Hospital, Ann Arbor; f Mercy Hospital, Coon Rapids, Michigan; gCardiology of Georgia, Atlanta, Georgia; hWashington Adventist Hospital, Tacoma Park, Maryland; iUniversity Hospital Benjamin Franklin, Berlin, Germany; jSaint Luke’s Hospital Kansas City–Mid America Heart Institute, Kansas City, Missouri; k Policlinico Careggi, Florence, Italy; lMayo Clinic Foundation, Rochester, Minnesota; and mCardiovascular Research Foundation and Lenox Hill Heart and Vascular Institute, New York, New York. Manuscript received November 9, 2006; revised manuscript received and accepted January 17, 2007. This study was supported by Medtronic Inc, Minneapolis, Minnesota. *Corresponding author: Tel.: 336-547-1756; fax: 336-851-8427. E-mail address: [email protected] (B.R. Brodie).

Methods Study population: The EMERALD trial was a prospective randomized multicenter trial of distal microcirculatory protection using the GuardWire Plus (Medtronic, Santa Rosa, California) balloon occlusion and aspiration system as an adjunct to mechanical reperfusion therapy in patients with STEMI. Patients with STEMI ⬍6 hours in duration with high-risk electrocardiographic criteria (ⱖ2 mm of STsegment elevation in ⱖ2 contiguous leads or left branch bundle block) targeted for primary or rescue PCI after failed thrombolytic therapy were considered for enrollment. Patients with shock and serum creatinine ⬎2.5 mg/dl were excluded. Of 501 patients randomly assigned in the EMERALD Trial, 450 patients had adequate sestamibi scans and

0002-9149/07/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.amjcard.2007.01.047

www.AJConline.org

Coronary Artery Disease/Time to Treatment and Infarct Size

1681

Table 1 Baseline clinical, angiographic, and procedural variables and median infarct size Variable

Age ⬎70 yrs Women Diabetes mellitus Previous myocardial infarction Previous coronary bypass surgery Hypertension Current smoker Killip class ⬎II Anterior infarction Baseline TIMI flow 0/1 Collateral flow 0/1 3-Vessel coronary disease Rescue PCI Time to reperfusion ⬍2 hrs

Median Infarct Size (% of left ventricle) Variable Present

Variable Absent

13.0 (2.0–36.0) (98) 2.0 (0.0–14.0) (91) 19.5 (6.0–48.0) (50) 12.0 (6.0–34.0) (49) 12.0 (0.0–30.0) (15) 11.0 (2.0–31.0) (161) 11.5 (1.0–23.0) (190) 25.0 (0.0–48.0) (53) 22.0 (4.0–43.0) (175) 13.0 (4.0–28.0) (279) 11.0 (2.0–27.0) (314) 11.0 (3.0–29.0) (67) 13.0 (3.0–25.0) (81) 2.0 (0.0–13.0) (21)

11.0 (1.0–23.0) (339) 13.0 (3.0–27.0) (346) 11.0 (1.0–23.0) (387) 11.0 (0.5–23.5) (388) 11.0 (2.0–25.0) (422) 11.0 (0.0–23.0) (275) 11.0 (2.0–27.0) (247) 11.0 (2.0–22.0) (378) 8.0 (0.0–17.0) (262) 5.5 (0.0–18.0) (142) 12.0 (1.5–20.0) (108) 11.0 (1.0–24.0) (370) 11.0 (0.0–24.0) (356) 11.0 (2.0–24.0) (373)

p Value

0.09 ⬍0.0001 0.005 0.14 0.97 0.31 0.91 0.04 ⬍0.0001 ⬍0.0001 0.70 0.40 0.21 0.008

Values expressed as median (interquartile range) (number of patients).

complete time-to-reperfusion data and form our study group. Study protocol: Patients received aspirin 324 mg (chewed), clopidogrel 300 mg, intravenous heparin 70-U/kg bolus, and intravenous ␤ blockers in the absence of contraindications. Glycoprotein IIb/IIIa inhibitors were used at the discretion of the operator, and additional heparin was given before PCI to prolong the activated clotting time to ⱖ300 seconds (200 to 300 seconds with glycoprotein IIb/IIIa inhibitors). Eligible patients were randomly assigned to primary PCI with stent implantation if appropriate, with or without distal protection. End points and definitions: Primary end points of the EMERALD trial were (1) infarct size measured using technetium-99m sestamibi imaging with sestamibi 20 to 30 mCi injected on days 5 to 14, and (2) frequency of complete electrocardiographic ST-segment resolution (STR) measured using continuous Holter monitoring after PCI. Secondary end points included Thrombolysis In Myocardial Infarction (TIMI) flow rates after PCI, restoration of normal (grade 3) myocardial blush, and a composite of death, reinfarction, ischemic target-vessel revascularization, and stroke at 1 and 6 months. Reinfarction, ischemic targetvessel revascularization, and disabling stroke were defined as previously reported.10 Time to reperfusion was defined as time from symptom onset until balloon inflation. Door-toballoon time was time from arrival at the hospital until balloon inflation. All clinical end points were adjudicated by a clinical events committee. Infarct size was measured using technetium-99m sestamibi imaging at 5 to 14 days after PCI. Single-photon emission computed tomographic imaging equipment and image acquisition procedures for each site were tested and quality controlled using a cardiac phantom. Data were processed by staff blinded to treatment assignment at a central core laboratory, as previously described.13,14 Infarct size was expressed as percentage of the left ventricle. Infarct size for patients who died before 5 to 14 days was imputed as the

largest infarct for any patient in the study (72% of the left ventricle). TIMI flows at baseline and after PCI were graded on a scale of 0 to 3, and myocardial blush after PCI was graded on a scale of 0 to 3 using methods previously described.15,16 Collateral flow to the infarct zone was assessed on the initial angiogram before PCI and graded on a scale of 0 to 3 using the method of Rentrop et al.17 All measurements were made by an angiographic core laboratory blinded to randomization and patient outcomes. Summed STR was measured from continuous 24-hour digital 12-lead electrocardiographic recordings at a central core laboratory as previously described.18,19 Percentage of STR from baseline to 60 minutes after the last contrast injection was calculated using Schroder’s classification as complete (⬎70%), partial (30% to 70%), or absent (⬍30%).20 Statistical analysis: Categorical data were compared using Fisher’s exact test, and continuous variables are presented as median with interquartile range and compared using Kruskal-Wallis nonparametric test. Six-month outcomes are summarized as Kaplan-Meier estimates and compared using log-rank test. The relation between time to treatment expressed as a continuous variable and infarct size was evaluated using linear regression. Predictors of infarct size were compared using multiple linear regression. All variables listed in Table 1 were entered into the model. All comparisons are 2 sided, and p ⬍0.05 was defined as significant, except when otherwise stated. Results Baseline variables by treatment times: Frequencies of baseline clinical and angiographic variables were similar across the 4 categories of time to reperfusion, although there was a trend for patients with longer times to reperfusion to be older (Table 2). Frequencies of baseline clinical and angiographic variables were similar across categories of door-to-balloon time, except that patients with shorter door-

1682

The American Journal of Cardiology (www.AJConline.org)

Table 2 Baseline clinical and angiographic variables by time to reperfusion Variable

Age (yrs) Women Diabetes mellitus Previous myocardial infarction Previous coronary bypass surgery Hypertension Current smoker Killip class ⱖII Infarct artery Left anterior descending Right Circumflex Saphenous vein graft TIMI 3 flow before PCI Collateral flow 2–3 Index ejection fraction (%) 3-Vessel coronary disease

Time to Reperfusion (hrs)

p Value

⬍2 (n ⫽ 24)

2–3 (n ⫽ 111)

⬎3–4 (n ⫽ 130)

⬎4 (n ⫽ 185)

58.8 (53.9–67.3) 25.0% 8.3% 12.5% 4.2% 16.7% 21.7% 41.7%

57.9 (52.6–64.9) 12.6% 13.5% 12.6% 2.7% 8.4% 38.7% 40.5%

58.0 (50.2–66.7) 23.1% 14.6% 6.2% 2.3% 11.8% 36.2% 43.1%

61.8 (52.4–71.2) 22.7% 12.7% 13.0% 3.8% 12.0% 38.9% 42.7%

0.08 0.13 0.59 0.24 0.87 0.64 0.43 0.98

33.3% 62.5% 4.2% 0% 13.6% 22.7% 66.5 (60.0–74.0) 12.5%

36.0% 50.5% 13.5% 1.8% 18.3% 29.6% 65.0 (55.0–72.0) 19.8%

44.6% 42.3% 13.1% 0.8% 21.1% 21.9% 61.5 (53.0–71.0) 13.1%

39.5% 53.3% 7.6% 1.6% 20.6% 26.7% 62.0 (53.0–70.0) 17.8%

0.50 0.16 0.19 0.82 0.83 0.57 0.21 0.48

Table 3 Baseline clinical and angiographic variables by door-to-balloon time Variable

Age (yrs) Women Diabetes mellitus Previous myocardial infarction Previous coronary bypass surgery Hypertension Current smoker Killip class ⱖII Infarct artery Left anterior descending Right Circumflex Saphenous vein graft TIMI 3 flow before PCI Collateral flow 2–3 Index ejection fraction (%) 3-Vessel coronary disease

Door-to-Balloon Time (hrs)

p Value

⬍1 (n ⫽ 33)

1–1.5 (n ⫽ 70)

⬎1.5–2 (n ⫽ 98)

⬎2 (n ⫽ 195)

59.9 (53.9–72.5) 27.3% 18.2% 9.1% 3.0% 43.8% 33.3% 6.1%

60.5 (54.4–65.3) 12.9% 12.9% 20.0% 5.7% 38.6% 42.9% 9.1%

57.6 (50.5–67.0) 23.5% 10.2% 6.1% 3.1% 33.7% 48.0% 11.6%

57.2 (50.4–67.4) 20.0% 10.8% 8.2% 1.0% 33.3% 42.6% 13.4%

0.16 0.26 0.61 0.02 0.19 0.62 0.52 0.57

24.2% 60.6% 15.2% 0% 12.9% 30.0% 66.5 (58.0–73.0) 15.2%

37.1% 51.4% 7.1% 4.3% 13.6% 30.3% 63.0 (56.0–71.0) 22.9%

40.4% 45.5% 12.1% 2.0% 17.3% 26.5% 65.0 (57.0–72.0) 11.2%

43.4% 46.5% 9.6% 0.5% 21.5% 24.1% 61.5 (52.5–71.0) 17.9%

0.20 0.41 0.56 0.13 0.41 0.71 0.22 0.24

to-balloon times had a higher frequency of previous infarction (Table 3). Relations among treatment times and coronary flow, myocardial reperfusion, and clinical outcomes: Patients with shorter times to reperfusion had trends for a higher frequency of final grade 3 myocardial blush, but there were no significant correlations between time to reperfusion and final TIMI 3 flow or STR (Table 4). Patients with longer times to reperfusion had trends for higher mortality at 6 months and significantly higher frequencies of death or reinfarction at 6 months (Table 4). There were no other significant correlations between time to reperfusion and clinical outcomes. There were no significant correlations between door-to-balloon time and myocardial reperfusion or clinical outcomes (Table 5).

Relation between treatment times and infarct size: Time to reperfusion expressed as a continuous variable significantly correlated with imputed infarct size using linear regression, but the correlation was very weak because much of the variation was explained by other variables (p ⫽ 0.031, r ⫽ 0.012; Figure 1). Correlation between time to reperfusion and infarct size was significant, but also very weak, in patients with anterior infarction (p ⫽ 0.017, r ⫽ 0.035; Figure 2) and was not significant in patients with nonanterior infarction (p ⫽ 0.35, r ⫽ 0.004; Figure 2). Time to reperfusion was a significant predictor of infarct size using multiple linear regression in patients with anterior infarction (p ⫽ 0.042), but not in patients with nonanterior infarction (p ⫽ 0.73). There was a significant interaction between time to reperfusion and infarct location (p ⫽

Coronary Artery Disease/Time to Treatment and Infarct Size

1683

Table 4 Myocardial reperfusion, infarct size, and clinical outcomes by time to reperfusion Time to Reperfusion (hrs) ⬍2 (n ⫽ 24) Myocardial reperfusion Final TIMI-3 flow Final grade 3 blush Compete STR (60 min) Infarct size All (% of LV) Left anterior descending (% of LV) Right, circumflex (% of LV) Clinical outcomes (6 mos) Death Reinfarction Death or reinfarction Stroke Target-vessel revascularization Major adverse clinical events

95.5% 66.7% 62.5% 2 (0–13) 0 (0–14) 3 (0–13) 0% 0% 0% 0% 0% 0%

2–3 (n ⫽ 111)

⬎3–4 (n ⫽ 130)

p Value ⬎4 (n ⫽ 185)

91.7% 65.3% 67.3%

89.8% 57.1% 69.4%

87.8% 50.3% 67.5%

0.58 0.08 0.92

9 (0–18) 17 (3.5–35.5) 7 (0–14)

12 (2–24) 20.5 (6–40) 7.5 (2–16.5)

11 (2–29) 30.5 (3–54) 10 (1–19)

0.026 0.026 0.23

5.3% 5.2% 9.8% 1.2% 7.1% 13.3%

0.06 0.42 0.036 0.91 0.54 0.15

0% 2.8% 2.8% 0.9% 6.5% 8.3%

2.4% 2.4% 4.0% 1.6% 5.2% 8.3%

LV ⫽ left ventricle. Table 5 Myocardial reperfusion, infarct size, and clinical outcomes by door-to-balloon time Door-to-Balloon Time (hrs) ⬍1 (n ⫽ 33) Myocardial reperfusion Final TIMI 3 flow Final grade 3 blush Complete STR (60 min) Infarct size All (% of LV) Left anterior descending (% of LV) Right, circumflex (% of LV) Clinical outcomes (6 mos) Death Reinfarction Death or reinfarction Stroke Target-vessel revascularization Major adverse clinical events

93.5% 64.5% 54.8% 4.5 (0–14) 0 (0–14) 5 (3–16)

1–1.5 (n ⫽ 70)

⬎1.5–2 (n ⫽ 98)

p Value ⬎2 (n ⫽ 195)

83.3% 68.3% 64.6%

93.9% 61.4% 75.0%

90.5% 52.5% 70.1%

0.14 0.12 0.17

11 (2–22) 22 (0–38) 9 (2–14)

12 (0–22) 21.5 (9–42) 5 (0–16)

11 (0.5–23) 22 (3–48) 8 (2–18)

0.37 0.15 0.63

0% 0%

0% 5.7%

1.0% 3.2%

4.6% 3.3%

0.11 0.52

0% 6.7% 6.7%

1.5% 5.7% 8.6%

1.0% 5.4% 7.3%

0.5% 6.9% 11.3%

0.83 0.99 0.78

Abbreviation as in Table 4.

0.022). There were no interactions between time to reperfusion and other predictor variables. Time to reperfusion expressed as a categorical variable (⬍2 vs 2 to 3 vs ⬎3 to 4 vs ⬎4 hours) significantly correlated with median infarct size (2% vs 9% vs 12% vs 11%, p ⫽ 0.026; Table 4 and Figure 3). The relation did not appear linear. The smallest infarct size was seen in patients with very early reperfusion (⬍2 hours; Table 4 and Figure 3). After 2 hours, incremental delays in time to reperfusion had only a small impact on infarct size. Time to reperfusion significantly correlated with infarct size in patients with anterior infarction (p ⫽ 0.026; Table 4 and Figure 4), but not in patients with nonanterior infarction (p ⫽ 0.16; Table 4 and Figure 4). Time to reperfusion expressed as a categorical variable was a significant independent predictor of infarct size using multiple linear regression (Table 6). Door-to-balloon time expressed as a continuous variable significantly correlated with infarct size (p ⫽ 0.008, r ⫽

Figure 1. Plot of imputed infarct size by time to reperfusion in all patients.

1684

The American Journal of Cardiology (www.AJConline.org)

Figure 2. Plot of imputed infarct size by time to reperfusion in patients with (A) anterior and (B) nonanterior infarction.

Figure 3. Imputed infarct size (median values) by time to reperfusion.

0.020), but the correlation was very weak. The correlation between door-to-balloon time and infarct size was significant in patients with anterior infarction (p ⫽ 0.007, r ⫽ 0.051), but not in patients with nonanterior infarction (p ⫽ 0.52, r ⫽ 0.002). Door-to-balloon time expressed as a categorical variable did not significantly correlate with infarct size (Table 5). Other predictors of infarct size: Relations among baseline clinical, angiographic, and procedural variables and infarct size are listed in Table 1. Infarct size was significantly greater in men and patients with diabetes, Killip class ⱖII, anterior infarction, baseline TIMI flow 0/1, and time to reperfusion ⬎2 hours. Using multiple linear regression, the variables anterior infarction, diabetes, baseline TIMI flow 0/1, rescue PCI, male gender, longer time to reperfusion, and older age were significant independent predictors of larger infarct size (Table 6). Infarct location, diabetes, and baseline TIMI flow accounted for most of the variability in infarct size. Time to reperfusion, gender, and age accounted for less of the variability in infarct size. Discussion Major findings of this study are that (1) time to reperfusion has a significant impact on infarct size in patients with

STEMI treated with primary PCI; (2) time to reperfusion impacts on infarct size in patients with anterior infarction, but has less effect in patients with nonanterior infarction; (3) infarct size is smallest when reperfusion is achieved at ⬍2 hours, after which there is less change in infarct size with incremental delays in time to reperfusion; and (4) time to reperfusion has little effect on measures of myocardial reperfusion. Previous studies evaluating time to reperfusion and infarct size had small numbers of patients and yielded conflicting results.8 –10 O’Keefe et al8 evaluated predictors of myocardial salvage in 59 patients with STEMI treated using primary PCI with acute and follow-up sestamibi imaging and found that shorter times to reperfusion (⬍2 vs 2 to 4 vs ⬎4 hours) were associated with better myocardial salvage index (80% vs 47% vs 44%, p ⫽ 0.004). Similarly, Milavetz et al9 studied 55 patients with anterior STEMI undergoing successful reperfusion and found that time to reperfusion ⬍2 hours was associated with the highest myocardial salvage. Conversely, Schomig et al10 found no significant relation between time to reperfusion with primary PCI and infarct size. However, this study did not evaluate the effect of very early reperfusion (⬍2 hours) on infarct size. Our study with larger numbers of patients documents the nonlinear relation between time to reperfusion and infarct size that was shown using both primary PCI13,14 and fibrinolytic therapy.21 Early reperfusion (⬍2 hours) was associated with smaller infarct size, but incremental delays in reperfusion after 2 hours appeared to have less effect on infarct size. This is consistent with animal models of reperfusion described by Reimer et al22 in which the time window for myocardial salvage was limited to the first 3 hours. Several previous studies showed that the presence of residual flow to the infarct zone (baseline TIMI flow 2 to 3 or collateral flow 2 to 3) was associated with smaller infarct size.8,9,23 We found that baseline TIMI 2 to 3 flow, but not collateral flow, was associated with smaller infarct size. In contrast to previous studies,9,23 we did not find a significant interaction between time to reperfusion and baseline TIMI flow or collateral flow on infarct size. Although time to reperfusion impacts on infarct size, most of the variability of infarct size is explained by other variables (infarct location, diabetes, baseline TIMI flow, rescue PCI, and male gender).

Coronary Artery Disease/Time to Treatment and Infarct Size

1685

Figure 4. Imputed infarct size (median values) by time to reperfusion in patients with (A) anterior and (B) nonanterior infarction. Table 6 Predictors of infarct size by multiple linear regression Multivariable Predictors Anterior infarction Diabetes mellitus Baseline TIMI flow 0/1 Rescue PCI Male sex Time to reperfusion (per category)* Age (per 10 yrs)

Estimate

SE

p Value

15.38 10.00 8.43 7.75 5.44 2.55 2.05

1.79 2.74 1.90 2.53 2.23 1.01 0.76

⬍0.0001 0.0003 ⬍0.0001 0.002 0.02 0.01 0.007

* Time onset to first balloon was categorized as ⬍2 versus 2 to 3 versus 3 to 4 versus ⬎4 hours.

Most,6,7,24,25 but not all,26 previous studies with primary PCI showed that longer time to reperfusion was associated with worse STR and worse myocardial blush scores. Our study showed trends for worse blush scores with later reperfusion, but no significant correlation between time to reperfusion and STR. Reasons for these differences from previous studies are not clear, but our study has a smaller number of patients than some previous studies. Some,2,5–7 but not all,1,3,4 previous studies showed a significant correlation between time to reperfusion with primary PCI and mortality. We found a trend for lower mortality at 6 months and a significantly lower incidence of mortality or reinfarction at 6 months with early reperfusion. Mortality appeared to increase with increasing time to reperfusion out to 6 hours. This is in contrast to some previous studies2,7 that showed that incremental time delays after 2 to 3 hours had little impact on mortality. Our study was not powered to detect differences in clinical events, and these differences may be related to our small sample size. Our study has several limitations. Although this is the largest primary PCI trial with data for infarct size using sestamibi imaging, the small number of patients treated very early limits the power to perform subgroup analyses. This is an observational study, and there may be differences in baseline variables across categories of time to treatment (not adjusted for by multivariable analyses) that could influence our results. Time to reperfusion was defined as time from symptom onset until balloon inflation, but is uncertain in patients with TIMI 2 to 3 flow on initial angiography. There are limitations in the measurement of infarct size using sestamibi imaging described previously.14 Measurement of

acute and convalescent images, which was not done for logistical reasons, would have allowed assessment of area at risk and calculation of myocardial salvage index and added power to our study. Infarct size was measured at 5 to 14 days. Measurements at a later time may have detected additional myocardial salvage. Our data may have implications for the triage of patients for primary PCI, especially patients presenting at noninterventional hospitals. In patients who present with STEMI early after the onset of symptoms (⬍2 hours), time is important for myocardial salvage. If substantial delays to primary PCI are expected, alternative reperfusion strategies may be considered. In patients who present later after the onset of symptoms (⬎2 hours) and patients with nonanterior infarction, delays are less important for myocardial salvage and transfer for primary PCI may be the best option, even with longer delays. Ongoing trials with facilitated PCI (pharmacologic reperfusion therapy followed by emergent PCI) may help define the best reperfusion strategy for patients in whom substantial delays to PCI are expected. Finally, our data show that only a small proportion of patients with STEMI treated with primary PCI (5.3%) undergo reperfusion early enough (⬍2 hours) to achieve a substantial decrease in infarct size. This emphasizes the need for improvement in processes to shorten time to reperfusion. 1. Berger PB, Ellis SG, Holmes DR, Granger CB, Criger DA, Betriu A, Topol EJ, Califf RM. Relationship between delay in performing direct coronary angioplasty and early clinical outcomes in patients with acute myocardial infarction: results from the GUSTO-IIb trial. Circulation 1999;100:14 –20. 2. Brodie BR, Stuckey TD, Wall TC, Kissling G, Hansen CJ, Muncy DB, Weintraub RA, Kelly TA. Importance of time to reperfusion for 30 day and late survival and recovery of left ventricular function after primary angioplasty for acute myocardial infarction. J Am Coll Cardiol 1998; 32:1312–1319. 3. Cannon CP, Gibson CM, Lambrew CT, Shoultz DA, Levy D, French WJ, Gore JM, Weaver WD, Rogers WJ, Tiefenbrunn AJ. Relationship of symptom-onset-to-balloon time and door-to-balloon time with mortality in patients undergoing angioplasty for acute myocardial infarction. JAMA 2000;283:2941–2947. 4. Brodie BR, Stone GW, Morice MC, Cox DA, Garcia E, Mattos LA, Boura J, O’Neill WW, Stuckey TD, Milks S, Lansky AJ, Grines CL, for the Stent Primary Angioplasty in Myocardial Infarction Study Group. Importance of time to reperfusion on outcomes with primary coronary angioplasty for acute myocardial infarction: results from the

1686

5.

6.

7.

8. 9. 10.

11.

12.

13. 14.

15.

The American Journal of Cardiology (www.AJConline.org)

Stent Primary Angioplasty in Myocardial Infarction Trial. Am J Cardiol 2001;88:1085–1090. Antoniucci D, Valenti R, Migliorini A, Moschi G, Trapani M, Buonamici P, Cerisano G, Bolognese L, Santoro GM. Relation of time to treatment and mortality in patients with acute myocardial infarction undergoing primary coronary angioplasty. Am J Cardiol 2002;89: 1248 –1252. De Luca G, Suryapranata H, Zijlstra F, van’t Hof AW, Hoorntje JC, Gosselink AT, Dambrink JH, de Boer MJ, for the ZWOLLE Myocardial Infarction Study Group. Symptom-onset-to-balloon time and mortality in patients with acute myocardial infarction treated by primary angioplasty. J Am Coll Cardiol 2003;42:991–997. Brodie BR, Stone GW, Cox DA, Stuckey TD, Turco M, Tcheng JE, Berger P, Mehran R, McLaughlin M, Costantini C, Lansky AJ, Grines CL. Impact of treatment delays on outcomes of primary percutaneous coronary intervention for acute myocardial infarction: analysis from the CADILLAC Trial. Am Heart J 2006;151:1231–1238. O’Keefe JH, Grines CL, DeWood MA, Bateman TM, Christian TF, Gibbons RJ. Factors influencing myocardial salvage with primary angioplasty. J Nucl Cardiol 1995;2:35– 41. Milavetz JJ, Giebel DW, Christian TF, Schwartz RS, Holmes DR, Gibbons RJ. Time to therapy and salvage in myocardial infarction. J Am Coll Cardiol 1998;31:1246 –1251. Schomig A, Ndrepepa G, Mehilli J, Schwaiger M, Schuhlen H, Nekolla S, Pache J, Martinoff S, Bollwein H, Kastrati A. Therapydependent influence of time-to-treatment interval on myocardial salvage in patients with acute myocardial infarction treated with coronary artery stenting or thrombolysis. Circulation 2003;108:1084 –1088. Schomig A, Mehilli J, Antoniucci D, Ndrepepa G, Markwardt C, Di Pede F, Nekolla SG, Schlotterbeck K, Schuhlen H, Pache J, et al, for the Beyond 12 hours Reperfusion AlternatiVe Evaluation (BRAVE-2) Trial Investigators. Mechanical reperfusion in patients with acute myocardial infarction presenting more than 12 hours from symptom onset: a randomized controlled trial. JAMA 2005;293:2865–2872. Stone GW, Webb J, Cox DA, Brodie BR, Qureshi M, Kalynych A, Turco M, Schultheiss HP, Dulas D, Rutherford BD, et al, for the Enhanced Myocardial Efficacy and Recovery by Aspiration of Liberated Debris (EMERALD) Investigators. Distal microcirculatory protection during percutaneous coronary intervention in acute ST-segment elevation myocardial infarction: a randomized controlled trial. JAMA 2005;293:1063–1072. Christian TF, Schwartz RS, Gibbons RJ. Determinants of infarct size in reperfusion therapy for acute myocardial infarction. Circulation 1992;86:81–90. Gibbons RJ, Miller TD, Christian TF. Infarct size measured by SPECT imaging with technetium-99m sestamibi—a measure of the efficacy of therapy in acute myocardial infarction. Circulation 2000;101:101– 108. Stone GW, Peterson MA, Lansky AJ, Dangas G, Mehran R, Leon MB. Impact of normalized myocardial perfusion after successful angio-

16.

17.

18.

19.

20.

21. 22. 23. 24. 25.

26.

plasty in acute myocardial infarction. J Am Coll Cardiol 2002;39:591–597. Costantini CO, Stone GW, Mehran R, Aymong E, Grines CL, Cox DA, Stuckey T, Turco M, Gersh BJ, Tcheng JE, et al. Frequency, correlates, and clinical implications of myocardial perfusion after primary angioplasty and stenting with and without glycoprotein IIb/ IIIa inhibition in acute myocardial infarction. J Am Coll Cardiol 2004;44:305–312. Rentrop KP, Cohen M, Blanke H, Phillips RA. Changes in collateral channel filling immediately after controlled coronary artery occlusion by an angioplasty balloon in human subjects. J Am Coll Cardiol 1985;5:587–592. Krucoff MW, Croll MA, Pope JE, Pieper KS, Kanani PM, Granger CB, Veldkamp RF, Wagner BL, Sawchak ST, Califf RM. Continuously updated 12-lead ST-segment recovery analysis for myocardial infarct artery patency assessment and its correlation with multiple simultaneous early angiographic observations. Am J Cardiol 1993;71: 145–151. Roe MT, Green CL, Giugliano RP, Gibson CM, Baran K, Greenberg M, Palmeri ST, Crater S, Trollinger K, Hannan K, Harrington RA, Krucoff MW, for the INTEGRITI Investigators. Improved speed and stability of ST-segment recovery with reduced-dose tenecteplase and eptifibatide compared with full-dose tenecteplase for acute ST-segment elevation myocardial infarction. J Am Coll Cardiol 2004;43: 549 –556. Schroder R, Wegscheider K, Schroder K, Dissmann R, Meyer-Sabellek W. Extent of early ST segment elevation resolution: a strong predictor of outcome in patients with acute myocardial infarction and a sensitive measure to compare thrombolytic regimens. J Am Coll Cardiol 1995;26:1657–1664. Chareonthaitawee P, Gibbons RJ, Roberts RS, Christian TF, Burns R, Yusuf S. The impact of time to thrombolytic treatment on outcome in patients with acute myocardial infarction. Heart 2000;84:142–148. Reimer KA, Vander Heide RS, Richard VJ. Reperfusion in acute myocardial infarction: effect of timing and modulating factors in experimental models. Am J Cardiol 1993;72(suppl):13G–21G. Clements IP, Christian TF, Higano ST, Gibbons RJ, Gersh BJ. Residual flow to the infarct zone as a determinant of infarct size after direct angioplasty. Circulation 1993;88:1527–1533. van’t Hof AW, Liem A, de Boer MJ, Zijlstra F. Clinical value of 12-lead electrocardiogram after successful reperfusion therapy for acute myocardial infarction. Lancet 1997;350:615– 619. Fu Y, Goodman S, Chang WC, Van De Werf F, Granger CB, Armstrong PW. Time to treatment influences the impact of ST-segment resolution on 1-year prognosis: insights from the ASSENT-II trial.Circulation2001;104:2653–2659. Schroder K, Wegscheider K, Zeymer U, Tebbe U, Schroder R. Extent of ST-segment deviation in a single electrocardiogram lead 90 minutes after thrombolysis as a predictor of medium term mortality in acute myocardial infarction. Lancet 2001;358:1479 –1486.