Readers wishing to see the figures/tables for this paper should consult the print version.

In recent years, there has been a significant move away from the practice of noninvasive imaging of patients after acute myocardial infarction (AMI) to a routine invasive approach with subsequent coronary revascularization. This has occurred for several reasons, including the recognized benefits of acute percutaneous coronary intervention (PCI) in patients with ST-segment elevation AMI and in those who are clinically unstable. These groups represent a small fraction of all AMI patients.¹ Current practice guidelines have also been strongly influenced by 2 recent trials that showed superiority of a routine invasive approach compared with one of noninvasive treadmill testing with selective angiography in patients hospitalized for acute coronary syndrome (ACS).^2,3 However, treadmill testing alone is known to be suboptimal for detecting myocardial ischemia⁴ and evaluating risk⁵ compared with other techniques, such as stress myocardial perfusion imaging, and cannot be performed very early after admission because of safety concerns.

We conducted a prospective, multicenter study, the Adenosine Sestamibi SPECT Post-Infarction Evaluation (INSPIRE) trial, to define the role of quantitative adenosine myocardial perfusion imaging as an initial testing method for assessing risk early after hospital admission and to help guide therapeutic decision making in stable survivors of AMI.^6-8

Subjects and methods

A total of 728 stabilized subjects with either ST-segment elevation (60%) or non–ST-segment elevation (40%) AMI from 16 sites worldwide were included in the study.^6,7 All subjects underwent adenosine technetium-99m sestamibi single-photon computed tomography (SPECT) imaging within 10 days of infarction and prior to coronary angiography.

As shown in Figure 1, subject risk and subsequent treatment were prospectively defined based on the perfusion results. Subjects with a small (< 20%) quantified left ventricular (LV) perfusion defect size (PDS) were considered low risk and targeted for medical therapy and early hospital discharge. Subjects with large (≥ 20%) but predominantly nonischemic defects were considered intermediate risk and received catheterization based on the investigator�fs decision. Subjects with a large total (≥ 20%) and ischemic (≥ 10%) LV PDS and an LV ejection fraction (EF) < 35% were considered high risk and encouraged to have coronary angiography with the intent to undergo revascularization. Subjects with an LVEF ≥ 35% were randomly assigned to receive either intensive anti-ischemic medical therapy or coronary revascularization.^6,8 The results from the randomized group have been reported previously.⁸

The INSPIRE study design also incorporated general treatment guidelines to promote optimal patient care. All subjects routinely received aspirin 325 mg/day and medical therapy for hyperlipidemia to reduce low-density lipoprotein cholesterol to < 100 mg/dL. Subjects with an LVEF < 40% or anterior AMI were encouraged to take an angiotensin-converting enzyme inhibitor and a β blocker. Combination antiischemic medical therapy at high doses was used in subjects who had residual angina or ischemia, or both. Subjects in all INSPIRE risk groups were encouraged to have coronary angiography if they developed recurrent refractory angina or other clinical evidence of instability.⁶ For subjects with significant left main or 3-vessel coronary artery disease and in those with diabetes and multivessel disease, coronary artery bypass graft surgery was recommended over PCI.

Cardiac events were defined as cardiac death, reinfarction, or readmission for ACS or severe congestive heart failure. The primary end point was time to first cardiac event, with a secondary end point of time to death or reinfarction. Follow-up was complete at 1 year in 98% of subjects.

Results

The INSPIRE study included a typical cohort of subjects who have experienced AMI, with a mean age of 63 ± 12 years, of whom 69% were male, 28% had prior known coronary artery disease, and 17% had prior AMI. Of the subjects with ST-segment elevation AMI, 58% received thrombolytic therapy, and 32% had anterior infarction.

Adenosine SPECT was performed very early after admission and particularly at US sites where over half of subjects were studied within 2 (mean = 2.7 ± 1.9) days. This afforded a very early risk assessment and appropriate triage to predefined therapeutic strategies.

During the follow-up period (mean = 11 ± 3 months), 95 events occurred in 88 subjects (12.1%), with 12 instances of severe congestive heart failure, 24 readmissions for ACS, 34 reinfarctions, and 18 cardiac deaths. The scintigraphic findings accurately predicted subsequent events. Regardless of age, sex, or infarct location, there was a significant increase in death/infarction (P = .009) and overall cardiac (P = .007) event rates across all 3 risk groups, from low (1.8% and 5.4%, respectively), to intermediate (9.2% and 14.0%, respectively), to high (11.6% and 18.6%, respectively; Figure 2). Event rates were consistently low in all subject subgroups that were identified as low risk by adenosine SPECT. This was observed despite the fact that few of the low-risk subjects had initial (11%) or late (16%) coronary revascularization compared with the intermediate-risk (17% and 21%, respectively) and high-risk (50% and 64%, respectively) INSPIRE groups (Table). Of note, revascularization significantly decreased event rates only in the high-risk INSPIRE group, which was comprised of subjects with large ischemic defects who also had significant LV dysfunction (Table).

The scintigraphic variables that defined the INSPIRE risk groups also predicted outcome. The total and death/reinfarction event rates significantly increased with larger total and ischemic PDS so that each 10% absolute increment in either variable increased the relative risk for any event by 37% and 64%, respectively (Figure 3). The LVEF calculated by gated SPECT also predicted outcome, with a low 1-year event rate (5%) in subjects with an LVEF ≥ 50%, which increased to 27% in the 3% of subjects with an LVEF < 20%. On multivariate analysis, total PDS was the most important independent risk predictor (P < .001), followed by ischemic PDS (P < .005) and LVEF (P = .05).

A cost analysis and assessment of hospital length of stay was determined from US sites. Subjects defined as low risk spent fewer days in the coronary care unit (2.1 ± 1.5 days) and were discharged earlier (5.5 ± 1.8 days) compared with those in the intermediate-risk (3.5 ± 3.1 and 8.0 ± 5.4 days, respectively) and high-risk (3.6 ± 2.0 and 13.9 ± 13.3 days, respectively) groups (P < .001). This translated into significantly lower hospital costs for the low-risk group ($5609 ± $3632) compared with the intermediate-risk ($9967 ± $7344) and high-risk ($13 269 ± $6432) groups (P < .001), in whom invasive diagnostic and interventional procedures were more frequently used.

Discussion

The INSPIRE trial demonstrated that adenosine SPECT imaging performed early after AMI can safely and accurately stratify risk and thereby guide further therapeutic decision making among stable patients who are at varying clinical risk for subsequent events. Unlike other studies,^2,3 which have formed the basis for current practice guidelines, INSPIRE did not use exercise treadmill testing to define risk in subjects randomly assigned to a conservative strategy but instead used more accurate quantitative stress myocardial perfusion imaging. The Veterans Affairs Non–QWave Infarction Strategies in Hospital (VANQWISH) study is the only other prospective trial to use myocardial perfusion scintigraphy as the noninvasive modality to assess risk in patients with AMI.⁹ This trial showed that a conservative approach guided by imaging resulted in a lower subsequent event rate (19% vs 24%) and was more cost effective¹⁰ than a routine invasive approach.

The use of adenosine as the stressor modality allowed very early imaging to safely identify high-risk subjects. This was a critical study design feature, which recognized that cardiac events tend to recur early after both ST-segment elevation¹¹ and non–ST-segment elevation¹² AMI. In earlier trials, such as the Treat Angina With Aggrastat and Determine Cost of Therapy With Invasive or Conservative Strategy (TACTICS)² and the Fragmin and Fast Revascularization During Instability in CAD (FRISC)³ studies, exercise treadmill testing in the conservative arm led to an unavoidable delay in identifying ischemia. The significantly lower event rate in the invasive vs the conservative arms of TACTICS was primarily attributable to the fact that more than 40% of all events in the conservative arm occurred prior to the index treadmill test. In INSPIRE, more than half of all subjects at US sites were safely imaged with adenosine within 2 days of hospital admission to facilitate subsequent patient care and avoid the pitfalls of these earlier trials. No cardiac events occurred in INSPIRE subjects prior to noninvasive testing, in part, because of the expedited imaging protocol.

An important aspect of early risk stratification is to identify low-risk patients for whom medical therapy and early hospital discharge is appropriate. The INSPIRE low-risk group represented one third of all enrolled subjects who ultimately had the lowest rate of coronary revascularization, the shortest hospital stay, the lowest hospital-related costs, and the lowest overall event rate. It is unlikely that coronary revascularization would have further improved outcome in this group, yet many such patients still currently undergo coronary angiography with subsequent PCI with that intent. INSPIRE demonstrates that such patients can be managed with a quite acceptable low event rate and at a cost savings approaching 65%.¹³

Risk stratification must identify high-risk patients who might benefit from coronary revascularization but importantly also those who will not. The prospectively defined high-risk INSPIRE group appropriately utilized the greatest extent of hospital resources but represented only 9% of all patients. Subjects in this group had the highest 1-year event rate (18%), especially when they did not undergo revascularization (32%). Previous studies have also inferred that patients with a low LVEF and extensive ischemia exhibit a survival benefit with coronary revascularization over medical therapy alone.¹⁴ Conversely, the intermediate-risk INSPIRE group represented another 29% of subjects who were at higher clinical risk but were unlikely to benefit from coronary revascularization because they had minimal or no residual ischemia.¹⁵ In a report on the recent Occluded Artery Trial (OAT), subjects with minimal or no ischemia following AMI were randomly assigned to receive either medical therapy or PCI and stenting of an occluded infarct-related artery.¹⁵ Subjects randomly assigned to the invasive strategy did not show the anticipated prognostic advantage over medical therapy alone and, in fact, a disturbing trend toward a higher reinfarction rate was observed in the invasive arm. The OAT results fully support the INSPIRE imaging strategy of identifying patients without significant ischemia.

Conclusions

The INSPIRE results provide strong support for the initial use of adenosine SPECT in stable AMI patients and are consistent with over 2 decades of smaller clinical studies using nuclear imaging to evaluate a broad spectrum of AMI patients¹⁶ The implications from INSPIRE are that state-of-the-art imaging performed early after AMI in appropriate patients can safely and reliably identify a large patient cohort unlikely to benefit from an invasive approach either because of their low scintigraphic risk profile or lack of inducible ischemia. Conversely, adenosine SPECT can also identify the subset of high-risk patients with extensive ischemia in whom intensive anti-ischemic medical and/or interventional therapies are appropriate. The INSPIRE^6-8 and other recent trials^15,17 support the use of noninvasive imaging as an initial strategy for categorizing risk and triaging subsequent care in stable survivors of ST-segment elevation AMI who have not undergone acute coronary angiography and in the emerging large group of patients with non-ST–segment elevation AMI.

References