Times to Treatment in Transfer Patients Undergoing Primary Percutaneous Coronary Intervention in the United States
National Registry of Myocardial Infarction (NRMI)-3/4 Analysis
Background— Treatment delays in patients with ST-segment–elevation myocardial infarction (STEMI) transferred for primary percutaneous coronary intervention (PCI) may decrease the advantage of this strategy over on-site fibrinolytic therapy that has been demonstrated in recent clinical trials. Accordingly, we sought to describe patterns of times to treatment in patients undergoing interhospital transfer for primary PCI in the United States.
Methods and Results— We analyzed patients with STEMI undergoing interhospital transfer for primary PCI between January 1999 and December 2002 in the National Registry of Myocardial Infarction. The primary outcome was “total” door-to-balloon time measured from time of arrival at the initial hospital to time of balloon inflation at the PCI hospital. Multivariable hierarchical models were used to assess the relationship of total door-to-balloon time with patient and hospital characteristics. Among 4278 patients transferred for primary PCI at 419 hospitals, the median total door-to-balloon time was 180 minutes, with only 4.2% of patients treated within 90 minutes, the benchmark recommended by national quality guidelines. Comorbid conditions, absence of chest pain, delayed presentation after symptom onset, less specific ECG findings, and hospital presentation during off-hours were associated with longer total door-to-balloon times. Patients at teaching hospitals in rural areas also had significantly longer times to treatment.
Conclusions— Total door-to-balloon times for transfer patients undergoing primary PCI in the United States rarely achieve guideline-recommended benchmarks, and current decision making should take these times into account. For the full benefits of primary PCI to be realized in transfer patients, improved systems are urgently needed to minimize total door-to-balloon times.
Received August 16, 2004; revision received November 1, 2004; accepted November 16, 2004.
Primary percutaneous coronary angioplasty (PCI) is superior to fibrinolytic therapy as a reperfusion strategy in ST-segment–elevation myocardial infarction (STEMI) when performed expeditiously and expertly at high-volume PCI hospitals.1–3 Compared with fibrinolytic therapy, primary PCI results in higher infarcted-artery patency and lower rates of reinfarction, stroke, and death.1,2,4 However, PCI programs currently are available in fewer than 25% of acute-care hospitals in the United States.5 Physicians at hospitals without PCI capability therefore have 2 treatment options for patients with STEMI: rapid patient transfer to a hospital with primary PCI capability or on-site fibrinolytic therapy.
See p 718
Results from 5 randomized, clinical trials suggest that the advantage of primary PCI can be extended to patients transferred from referral hospitals.6–10 A quantitative review of these studies suggested that for every 100 patients treated, interhospital transfer for primary PCI, instead of on-site fibrinolytic therapy, prevented 7 major adverse cardiac events, defined as the combination of death, nonfatal reinfarction, and nonfatal stroke.11 Importantly, average time from randomization at the initial hospital to first balloon inflation at the PCI hospital was relatively short—approximately 90 minutes—in the 4 trials that were performed in European countries with highly organized national emergency medical systems.6,7,9,10 In the only trial to include PCI hospitals in the United States, time from randomization to first balloon inflation was 120 minutes.8
Rapid and complete restoration of coronary artery blood flow and myocardial perfusion is the mechanism by which reperfusion therapy reduces morbidity and mortality in STEMI. Longer times to treatment are inversely related to outcome with both fibrinolytic therapy12,13 and PCI.14–16 Current guidelines for STEMI recommend a door-to-needle time within 30 minutes for fibrinolytic therapy and a door-to-balloon time within 90 minutes for primary PCI as treatment goals.17,18 Door-to-balloon times >1 hour beyond the anticipated door-to-needle time with fibrinolytic therapy may diminish the mortality benefit associated with primary PCI.19 Thus, excessive delays associated with interhospital transfer for primary PCI in the real-world setting may reduce the advantage of PCI noted in randomized, clinical trials, in which patients were carefully selected for enrollment and delays in initiating treatment were minimized.
The purpose of this investigation was to describe patterns of time to treatment in patients undergoing interhospital transfer for primary PCI in the United States and to identify patient and hospital characteristics associated with longer times. In addition, we evaluated whether longer interhospital transfer times were associated with shorter times to treatment at the PCI hospital, which would suggest early mobilization of the cardiac catheterization team. We used data from the National Registry of Myocardial Infarction (NRMI)-3 and -4 cohorts for this analysis.
We included patients in the NRMI-3 and -4 cohorts admitted between January 1, 1999, and December 31, 2002. This registry represents an ongoing observational study of patients with acute myocardial infarction admitted to participating hospitals in the United States. Details of NRMI hospitals, cohorts, and data collection methods have been described elsewhere.20,21 In brief, the NRMI protocol specifies that all consecutive patients with the diagnosis of acute myocardial infarction are enrolled at participating hospitals. Cases are typically screened for the International Classification of Diseases, 9th Revision, Clinical Modification discharge diagnosis code of 410.X1. Myocardial infarction is then confirmed on the basis of a suggestive patient history and one or more of the following criteria: (1) cardiac biomarker (eg, creatine kinase MB or troponin) elevation; (2) ECG evidence; and (3) scintigraphic, echocardiographic, or autopsy evidence.
Although NRMI does not perform on-site monitoring of data collection, routine queries are sent to local research coordinators for clarification of data inconsistencies. The internal validity of prior collections of NRMI data has been established through a comparison with the Cooperative Cardiovascular Project.22 Institutional review board approval was obtained at participating centers if required by the hospital.
We limited our analysis to patients (1) with symptoms of STEMI within 12 hours of presentation, (2) with associated ST-segment elevation in 2 or more leads or left bundle-branch block on their first ECG, and (3) who were transferred from an acute-care hospital and underwent primary PCI. Patients who received any fibrinolytic therapy, either at the initial hospital or en route to the referral hospital, were excluded from this analysis. After excluding 653 patients with missing or inconsistent data on the timeline of events, our study population consisted of 4278 patients at 419 PCI hospitals. We also performed a subgroup analysis for 3641 patients (85.1% of the total cohort) with no contraindication for fibrinolytic therapy.
Clinical Data and Outcome Measures
Patient records included data on demographics (eg, age, sex, and insurance status), cardiovascular risk factors (eg, history of hypertension, diabetes mellitus, hyperlipidemia, and smoking) and other comorbidities (eg, history of myocardial infarction, coronary artery bypass graft surgery or PCI, chronic renal insufficiency, and stroke). Clinical data were also available on presenting characteristics such as chest pain, symptoms of congestive heart failure, systolic blood pressure and pulse, and specific time and day of arrival at the initial hospital. Available data from the first ECG included location of myocardial infarction, the number of leads with ST-segment elevation and depression, nonspecific ST-segment or T-wave changes, and bundle-branch block patterns.
PCI hospital characteristics were also available from NRMI and supplemented with data from a 2000 American Hospital Association survey.5 Hospital information included data on annual primary PCI volume, percentage of patients with reperfusion who received PCI, teaching status, open-heart surgery capabilities, urban or rural location, nonprofit status, and geographic region. Rural facilities were defined as those in counties with a population <50 000.
For included patients, the complete timeline consisted of (1) time of symptom onset, (2) time of arrival at the initial hospital, (3) time of arrival at the referral hospital, and (4) time of first balloon inflation during primary PCI. Our primary end point was “total” door-to-balloon time measured from the time of arrival at the initial hospital to time of the first balloon inflation at the referral hospital. We also evaluated 2 other outcome measures: (1) door-to-door time (ie, time of arrival at the initial hospital to time of arrival at the referral hospital) and (2) PCI hospital door-to-balloon time (ie, time of arrival at the referral hospital to time of the first balloon inflation at the referral hospital).
Bivariate analyses were performed to evaluate differences in baseline patient and hospital characteristics across different time categories for total door-to-balloon time with the following cutoffs: <2 hours, 2 to 4 hours, and >4 hours. These cutoffs were chosen a priori on the basis of a literature review and preliminary data exploration of time-related variables. ANOVA tests were used to evaluate differences in continuous variables, and χ2 tests were used for categorical variables.
Multivariable hierarchical models were used to assess the relationship of total door-to-balloon, door-to-door, and PCI hospital door-to-balloon times with patient and hospital characteristics. Failing to account for clustering of outcomes at the hospital level may overestimate the significance of statistical associations. At the same time, hospitals enter and leave enrollment at different times in NRMI during the 4-year study period, so it is also true that outcomes are correlated within reporting study periods. Although several methods exist to account for clustering of outcomes within hospitals, multivariable hierarchical models allowed us to account for clustering within both hospitals and reporting study periods. Before the multivariable analysis, all time-related variables were transformed logarithmically because of their skewed distributions. Parameter estimates and confidence intervals (CIs) were then retransformed into their natural units before reporting by the use of standard simulation techniques.23
Finally, we sought to evaluate whether PCI hospital door-to-balloon times decreased as door-to-door times increased because of prehospital mobilization of the PCI hospital’s cardiac catheterization laboratory. To examine this issue, we calculated the Pearson correlation coefficient between logarithmic transformations of the 2 components of total door-to-balloon time: (1) door-to-door time and (2) PCI hospital door-to-balloon time. We used SAS version 8.2 (SAS Institute Inc) and Stata SE version 8.0 (Stata Corp) for all analyses.
We identified 4278 transfer patients with STEMI who underwent primary PCI at 419 hospitals. Tables 1 and 2⇓ display baseline characteristics for the transfer patients and PCI hospitals. Chest pain was noted at presentation in >90% of patients, and few patients were in cardiogenic shock. Initial ECGs showed at least 3 leads with ST-segment elevation in more than 80% of patients. Half of the patients presented within 2 hours of symptom onset, and only one third arrived on weekdays between 8 am and 4 pm. Most PCI hospitals were urban, teaching, not-for-profit centers with annual primary PCI volumes of <20 cases.
The median time from the initial hospital arrival to the first balloon inflation at the PCI hospital (ie, total door-to-balloon time) was 180 minutes. The median door-to-door time was 120 minutes, and the median PCI hospital door-to-balloon time was 53 minutes. Total door-to-balloon time was <2 hours in 16.2%, between 2 and 4 hours in 55.4%, and >4 hours in 28.4% of patients. Only 4.2% had total door-to-balloon times within 90 minutes. The Figure displays the distribution of total door-to-balloon, door-to-door, and PCI hospital door-to-balloon times within the study cohort.
Several patient characteristics were associated with total door-to-balloon time in unadjusted analyses (Table 3). Compared with those with total door-to-balloon times of <2 hours, patients with times >4 hours were more often older, female, and nonwhite; owned less commercial insurance; and in general had more complex medical conditions (eg, a higher incidence of diabetes, hypertension, and prior bypass surgery). These patients were also less likely to have chest pain or hemodynamic stability on presentation. More often, the initial ECGs showed less lead involvement (ie, only 2 leads with ST-segment elevation) or left bundle-branch block. Patients with total door-to-balloon times >4 hours also presented much later after symptom onset and more often on weekends during off-hours.
PCI hospitals that were nonteaching, located in urban areas, and had a not-for-profit status tended to have a greater percentage of their patients undergo primary PCI with total door-to-balloon times of <2 hours (Table 3). In contrast, facilities with an annual primary PCI volume of <20 or those that used primary PCI as a reperfusion strategy in <20% of their STEMI patients had a greater percentage of cases with total door-to-balloon times >4 hours.
Table 4 displays patient and hospital characteristics associated with substantial variations in total door-to-balloon time after multivariable adjustment. Although comorbid conditions (eg, diabetes mellitus) increased the total door-to-balloon time slightly, the absence of chest pain, prolonged symptom duration before presentation, and an off-hours time of arrival at the initial hospital (ie, 12 midnight to 7:59 am) were all associated with substantial increases in total door-to-balloon times. In contrast, more extensive ECG findings were associated with much shorter total door-to-balloon times. Hospital characteristics associated with total door-to-balloon time included facility location and teaching status. After adjustment for patient and presenting characteristics, for example, the mean door-to-balloon time for an urban, nonteaching hospital was 181 minutes (95% CI, 173 to 188 minutes) in comparison with 249 minutes (95% CI, 207 to 290 minutes) for a rural, teaching hospital.
Similar findings were seen for the multivariable analyses of door-to-door time, with some notable exceptions. Small but significantly longer door-to-door times were noted in female (+6.2 minutes; 95% CI, 2.5 to 10.4 minutes; P=0.001) and nonwhite (+6.3 minutes; 95% CI, 0.8 to 12.1 minutes; P=0.024) patients. Also, prior bypass graft surgery was not significantly associated with door-to-door time (+3.8 minutes; 95% CI, −3.7 to 11.6 minutes; P=0.33). Limiting our study population to those patients with no contraindication to fibrinolytic therapy did not alter our overall results.
Finally, we found a positive correlation between logarithmic transformations of the 2 components of total door-to-balloon time: door-to-door and PCI hospital door-to-balloon times (correlation coefficient, 0.25; P<0.001). This suggests that longer door-to-door times were not associated with shorter PCI hospital door-to-balloon times. Rather, each additional 30-minute increase in door-to-door time was associated with a small but significant increase in PCI hospital door-to-balloon time (+2.7 minutes; 95% CI, 2.1 to 3.0 minutes; P<0.001).
Our analysis found that the median total door-to-balloon time for transfer patients in the United States undergoing primary PCI was 180 minutes, with ≈4% treated within 90 minutes and 15% treated within 120 minutes. These results reveal an important discrepancy between the performance of primary PCI in transfer patients within a real-world setting and the strategy tested in recent randomized, clinical trials. Also, when interpreted within the context of national quality guidelines—which have recently reduced the treatment goal for primary PCI from a door-to-balloon time within 120 minutes to within 90 minutes—these results suggest that transfer for primary PCI in the United States is presently failing to achieve established benchmarks in the majority of STEMI patients.17,18 Hospitals without PCI capability that are planning to use a transfer strategy for STEMI patients need to be aware of their own total door-to-balloon times and to incorporate this information into clinical decision making when selecting between reperfusion strategies.
In addition, our analysis identified several patient and hospital characteristics that were associated with prolongations in times to treatment in transfer patients. As expected, several of these factors are similar to those previously associated with longer times to reperfusion in nontransfer STEMI patients receiving fibrinolytic therapy or primary PCI.24,25 Such characteristics represent a number of organizational barriers (ie, delays in triage, evaluation and diagnosis; limited staffing during off-hours) that continue to hinder efforts to minimize times to treatment in all STEMI patients.
We also found that patients transferred to teaching hospitals in rural areas had substantially longer total door-to-balloon times—in fact, >1 hour longer—when compared with patients transferred to urban, nonteaching hospitals. It may be that in this latter situation, physical barriers such as geographic distance make it difficult for rural facilities to achieve rapid times to treatment without a substantial investment of resources. On-site fibrinolytic therapy may be a very reasonable, next-best option for reperfusion therapy under these circumstances.
Although our analysis challenges the overall generalizability of results from recent randomized, clinical trials, it should be noted that ≈15% of transfer patients in NRMI did have a total door-to-balloon time within 2 hours. Recently, the Minneapolis Heart Institute has initiated a hospital transfer network for STEMI, drawing on the experiences from European hospitals and the trauma center network that exists in the United States.26 These investigators have reported median total door-to-balloon times between 90 and 100 minutes and door-to-door times of 70 minutes for transfer patients with STEMI. These results suggest that attention to logistical details and reengineering of process-of-care systems could make transfer for primary PCI a realistic strategy for some regions of the United States.
In addition to reducing door-to-door times, Larson and colleagues26 have shown the importance of lowering PCI hospital door-to-balloon times by implementing strategies such as early activation of the PCI team and direct transfer of patients to the catheterization laboratory (as opposed to emergency department–to–emergency department transfers). If these latter strategies were more widely used, it would be expected that longer door-to-door times would be associated with shorter PCI hospital door-to-balloon times. However, this finding was not seen in our analysis and suggests a potentially important area for quality improvement in transfer patients with STEMI.
Finally, it is important to note that under certain circumstances, transfer may remain the best option for reperfusion even when prolonged times to treatment are anticipated. Primary PCI is recommended for patients who are ineligible for fibrinolytic therapy or are in cardiogenic shock.17,27,28 Regardless of physician treatment preferences or hospital location, it is becoming increasingly evident that regional and national strategies for triaging patients with STEMI need to be established in the United States. This is an especially important public health priority, given the gaps that currently exist for both pharmacological and PCI evidence-based treatment strategies.29–32 These issues must be balanced by the fact that fibrinolytic therapy remains a reasonable option for many patients with STEMI. In fact, the benefits associated with transfer in the randomized, clinical trials were primarily related to reductions in recurrent myocardial infarction. These benefits are likely to be reduced in a real-word setting, where high rates of early cardiac catheterization are to be expected in patients with STEMI after fibrinolytic therapy. Additional trials that are currently underway are evaluating “pharamcoinvasive therapy” as a strategy for combining fibrinolytic therapy and PCI in STEMI, and their results may have direct implications for transfer patients.
Our analysis should be interpreted in the context of the following study design issues. First, this was an observational study of voluntarily reported data from participating hospitals. Although NRMI hospitals do include a wide variety of acute-care hospitals (1 in 4 hospitals in the United States contributed to NRMI-4), these centers generally tend to be larger and more specialized than nonparticipating hospitals,21 but very few performed large volumes of primary PCI, with only 17.4% reporting >40 cases per year to NRMI. Longer door-to-balloon times may be due to the inclusion of these relatively inexperienced centers. Second, transfer patients in NRMI represent a select group of patients who were chosen for transfer and may not be representative of all patients eligible for reperfusion therapy or transfer. Third, we did not report on the association between total door-to-balloon time and in-hospital death, because NRMI collects minimal baseline data on patients from the first hospital. Nevertheless, there is strong and consistent evidence linking increases in door-to-balloon time to higher mortality with primary PCI.15,16 Finally, door-to-balloon and door-to-door times aggregate several distinct decision-making points in the care of patients with STEMI: presentation to ECG, ECG to treatment decision, treatment decision to transfer initiation, and transport time. We were unable to assess the significance of these individual components in our analysis.
Primary PCI for transfer patients in the United States is frequently associated with total door-to-balloon times of 90 minutes or more, which are far longer than the times to treatment reported in recent randomized, clinical trials. This delay may affect the overall efficacy of primary PCI when compared with on-site fibrinolytic therapy and should be considered in the clinical decision-making process when selecting between the 2 reperfusion strategies. As the momentum for primary PCI grows in the United States, improved process-of-care systems are urgently needed to minimize times to treatment in transfer patients.
This project was supported by the National Heart, Lung, and Blood Institute (R01HS10407-01). Dr Nallamothu is supported as a clinical scholar under a K12 grant from the National Institutes of Health (RR017607-01).
Guest Editor for this article was Michael A. Fifer, MD.
Weaver WD, Simes RJ, Betriu A, Grines CL, Zijlstra F, Garcia E, Grinfeld L, Gibbons RJ, Ribeiro EE, DeWood MA, Ribichini F. Comparison of primary coronary angioplasty and intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review. JAMA. 1997; 278: 2093–2098.
Canto JG, Every NR, Magid DJ, Rogers WJ, Malmgren JA, Frederick PD, French WJ, Tiefenbrunn AJ, Misra VK, Kiefe CI, Barron HV. The volume of primary angioplasty procedures and survival after acute myocardial infarction; National Registry of Myocardial Infarction 2 Investigators. N Engl J Med. 2000; 342: 1573–1580.
The Global Use of Strategies to Open Occluded Coronary Arteries in Acute Coronary Syndromes (GUSTO IIb) Angioplasty Substudy Investigators. A clinical trial comparing primary coronary angioplasty with tissue plasminogen activator for acute myocardial infarction. N Engl J Med. 1997; 336: 1621–1628.
American Hospital Association. The Annual Survey of Hospitals Database: Documentation for 2000 Data. Chicago, Ill: American Hospital Association; 2000.
Vermeer F, Oude Ophuis AJ, van den Berg EJ, Brunninkhuis LG, Werter CJ, Boehmer AG, Lousberg AH, Dassen WR, Bar FW. Prospective randomised comparison between thrombolysis, rescue PTCA, and primary PTCA in patients with extensive myocardial infarction admitted to a hospital without PTCA facilities: a safety and feasibility study. Heart. 1999; 82: 426–431.
Widimsky P, Groch L, Zelizko M, Aschermann M, Bednar F, Suryapranata H. Multicentre randomized trial comparing transport to primary angioplasty vs immediate thrombolysis vs combined strategy for patients with acute myocardial infarction presenting to a community hospital without a catheterization laboratory: the PRAGUE study. Eur Heart J. 2000; 21: 823–831.
Grines CL, Westerhausen DR Jr, Grines LL, Hanlon JT, Logemann TL, Niemela M, Weaver WD, Graham M, Boura J, O’Neill WW, Balestrini C. A randomized trial of transfer for primary angioplasty versus on-site thrombolysis in patients with high-risk myocardial infarction: the Air Primary Angioplasty in Myocardial Infarction study. J Am Coll Cardiol. 2002; 39: 1713–1719.
Widimsky P, Budesinsky T, Vorac D, Groch L, Zelizko M, Aschermann M, Branny M, St’asek J, Formanek P. Long distance transport for primary angioplasty vs immediate thrombolysis in acute myocardial infarction: final results of the randomized national multicentre trial—PRAGUE-2. Eur Heart J. 2003; 24: 94–104.
Andersen HR, Nielsen TT, Rasmussen K, Thuesen L, Kelbaek H, Thayssen P, Abildgaard U, Pedersen F, Madsen JK, Grande P, Villadsen AB, Krusell LR, Haghfelt T, Lomhold P, Husted SE, Vigholt E, Kjaergard HK, Mortensen LS. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med. 2003; 349: 733–742.
Zijlstra F. Angioplasty vs thrombolysis for acute myocardial infarction: a quantitative overview of the effects of interhospital transportation. Eur Heart J. 2003; 24: 21–23.
Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet. 1994; 343: 311–322.
Berger PB, Ellis SG, Holmes DR Jr, Granger CB, Criger DA, Betriu A, Topol EJ, Califf RM. Relationship between delay in performing direct coronary angioplasty and early clinical outcome in patients with acute myocardial infarction: results from the global use of strategies to open occluded arteries in Acute Coronary Syndromes (GUSTO-IIb) trial. Circulation. 1999; 100: 14–20.
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.
De Luca G, Suryapranata H, Ottervanger JP, Antman EM. Time delay to treatment and mortality in primary angioplasty for acute myocardial infarction: every minute of delay counts. Circulation. 2004; 109: 1223–1225.
Antman EM, Anbe DT, Armstrong PW, Bates ER, Green LA, Hand M, Hochman JS, Krumholz HM, Kushner FG, Lamas GA, Mullany CJ, Ornato JP, Pearle DL, Sloan MA, Smith SC Jr, Alpert JS, Anderson JL, Faxon DP, Fuster V, Gibbons RJ, Gregoratos G, Halperin JL, Hiratzka LF, Hunt SA, Jacobs AK. ACC/AHA guidelines for the management of patients with acute myocardial infarction—executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to revise the 1999 guidelines for the management of patients with acute myocardial infarction). Circulation. 2004; 110: 588–636.
Van de Werf F, Ardissino D, Betriu A, Cokkinos DV, Falk E, Fox KA, Julian D, Lengyel M, Neumann FJ, Ruzyllo W, Thygesen C, Underwood SR, Vahanian A, Verheugt FW, Wijns W. Management of acute myocardial infarction in patients presenting with ST-segment elevation; the Task Force on the Management of Acute Myocardial Infarction of the European Society of Cardiology. Eur Heart J. 2003; 24: 28–66.
Rogers WJ, Canto JG, Lambrew CT, Tiefenbrunn AJ, Kinkaid B, Shoultz DA, Frederick PD, Every N. Temporal trends in the treatment of over 1.5 million patients with myocardial infarction in the US from 1990 through 1999: the National Registry of Myocardial Infarction 1, 2 and 3. J Am Coll Cardiol. 2000; 36: 2056–2063.
Peterson ED, Pollack CV Jr, Roe MT, Parsons LS, Littrell KA, Canto JG, Barron HV. Early use of glycoprotein IIb/IIIa inhibitors in non-ST-elevation acute myocardial infarction: observations from the National Registry of Myocardial Infarction 4. J Am Coll Cardiol. 2003; 42: 45–53.
King G, Tomz M, Wittenberg J. Making the most of statistical analyses: improving interpretation and presentation. Am J Pol Sci. 1998; 44: 341–355.
Larson DM, Sharkey SW, Unger BT, Mooney MR, Madison JD, Henry TH. Is rapid transfer of ST-elevation myocardial infarction patients for primary angioplasty feasible in the United States? Am J Cardiol. 2003; 92: 152L–153L.
Grzybowski M, Clements EA, Parsons L, Welch R, Tintinalli AT, Ross MA, Zalenski RJ. Mortality benefit of immediate revascularization of acute ST-segment elevation myocardial infarction in patients with contraindications to thrombolytic therapy: a propensity analysis. JAMA. 2003; 290: 1891–1898.
Hochman JS, Sleeper LA, Webb JG, Sanborn TA, White HD, Talley JD, Buller CE, Jacobs AK, Slater JN, Col J, McKinlay SM, LeJemtel TH. Early revascularization in acute myocardial infarction complicated by cardiogenic shock; SHOCK Investigators (Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock). N Engl J Med. 1999; 341: 625–634.
Topol EJ, Kereiakes DJ. Regionalization of care for acute ischemic heart disease: a call for specialized centers. Circulation. 2003; 107: 1463–1466.
Weaver WD. All hospitals are not equal for treatment of patients with acute myocardial infarction. Circulation. 2003; 108: 1768–1771.
Antman EM, Van de Werf F. Pharmacoinvasive therapy: the future of treatment for ST-elevation myocardial infarction. Circulation. 2004; 109: 2480–2486.