Review of the ISCHEMIA and ISCHEMIA-CKD trials

Initial invasive or conservative strategy for stable coronary artery disease

N Engl J Med 2020;382:1395-407 - ISCHEMIA

N Engl J Med 2020;382:1608-16 - ISCHEMIA-CKD

Background: The COURAGE trial, published in 2007, represented a major reversal in cardiovascular medicine. In patients with stable CAD an initial strategy of revascularization plus medical therapy did not reduce the chance of dying or having a heart attack compared to an initial strategy of medical therapy alone.

Prior to these results, patients with stable CAD were routinely managed with an initial invasive approach and the field of cardiology was intensely focused on finding coronary blockages and “fixing” them in symptomatic and asymptomatic patients alike. Thus, it’s not surprising that following results from COURAGE, the practice continued to be vigorously defended and applied routinely in the management of patients with stable CAD.

The first major attempt to reverse the results of COURAGE came from the FAME 2 trial, published in 2012, which tested the hypothesis that patients with stable CAD and an abnormal fractional flow reserve (FFR) in the cath lab would do better with an initial invasive strategy compared to medical therapy alone. The trial was stopped early for efficacy but the positive results were driven entirely by revascularization during follow up - not death or heart attack. The trial was criticized for being stopped inappropriately without providing an answer to whether an early invasive strategy improved hard endpoints compared to initial medical therapy alone. The concepts of “faith healing” and “subtraction anxiety” are useful for understanding the results and limitations of the FAME 2 trial.

The ISCHEMIA trial which began enrolling patients in 2012 sought to overcome limitations of COURAGE and FAME. The investigative aim of the study was to test the hypothesis that in patients with stable CAD and moderate to severe ischemia on provocative testing, an initial invasive strategy reduced a composite of major cardiac events compared to initial medical therapy alone.

The ISCHEMIA-CKD trial was performed in conjunction with the ISCHEMIA Research Group to address an important knowledge gap in managing patients with CAD. Patients with advanced chronic kidney disease (CKD) experience a higher rate of cardiac events than their counterparts without CKD; however, they are also at a higher risk of procedural complications. The standard of care at the time was generally to manage a patient with stable CAD and CKD like any other patient with CAD despite the fact that such patients were historically excluded from participation in clinical trials and thus, there was really no data from clinical trials to guide decision making.

The ISCHEMIA-CKD investigators sought to test the hypothesis that in patients with advanced CKD and stable CAD and moderate to severe ischemia on stress testing, an initial invasive strategy reduced death or MI compared to initial medical therapy alone.

Patients: For the ISCHEMIA trial, eligible patients had to be at least 21 years of age or older with at least moderate ischemia on a qualifying stress test based on the following criteria:

1. Nuclear perfusion with SPECT or PET with >/= 10% ischemic myocardium

2. Echocardiography with >/= 3/16 segments with stress-induced severe hypokinesis or akinesis

3. Cardiac MRI with >/= 10% ischemic myocardium on perfusion imaging and/or >/= 3/16 segments with stress-induced severe hypokinesis or akinesis on wall motion assessment

4. Exercise treadmill test without imaging that met all 4 following criteria

   1. clinical history of typical angina or typical angina during the stress test

   2. absence of resting ST depression > 1.0 mm or confounders that render exercise EKG non-interpretable (LBBB, LVH with repolarization, pacemaker, etc.)

   3. exercise-induced horizontal or downsloping ST depression >/= 1.5 mm in 2 leads or >/= 2.0 mm in any lead or ST elevation >/= 1.0 mm in a non-infarct territory

   4. either of the following:

      1. workload at which ST segment criteria are met is NOT to exceed completion of stage 2 of a standard Bruce protocol or 7 METS if a non-Bruce protocol is used

      2. ST segment criteria are met at <75% max predicted HR

The option of exercise treadmill testing without imaging was approved in 2014 to improve recruitment and generalizability of the trial results. This is the predominant form of stress testing used throughout the world.

There were 28 total exclusion criteria applied, key ones include: GFR <30 ml/min, a recent ACS event, unprotected left main stenosis of at least 50%, a LVEF <35%, NYHA class III or IV heart failure, previous PCI within 12 months or ACS within 2 months, stroke within 6 months, unacceptable angina despite use of medical therapy at maximum acceptable doses, high risk of bleeding, history of noncompliance with medical therapy, and life expectancy less than duration of trial due to non-cardiovascular comorbidity.

Most enrolled trial participants underwent coronary computed tomography angiography (CCTA) to exclude left main disease or nonobstructive disease unless they had renal dysfunction that would preclude such testing or known coronary anatomy.

All participating sites met prespecified quality metrics based on the volume and outcomes of PCI and CABG procedures performed. While individual sites determined whether stress testing results met eligibility, all stress tests were reviewed by independent core laboratories.

Enrollment in ISCHEMIA-CKD began 2 years after initiation of ISCHEMIA and the 2 trials ran in parallel and were conducted at most of the same sites. Eligible patients had to have advanced CKD, defined as an estimated GFR of <30 ml/min, and moderate to severe ischemia using the criteria outlined above. However, in contrast with ISCHEMIA, the use of CCTA was not recommended because of the risk of acute kidney injury and core lab review of stress tests was not performed.

Baseline characteristics: A total of 8,518 patients were enrolled in the ISCHEMIA trial and 5,179 underwent randomization at 320 sites in 37 countries. The median age of patients was 64 years, 77% were men, and 66% were white. Most patients (73%) had hypertension, 42% had diabetes and 10% used insulin. The median LVEF was 60%. The frequency of major morbidities included 4% with heart failure, 4% atrial fibrillation, 3% stroke, 7% cerebrovascular disease, and 4% peripheral artery disease. The estimated GFR was not provided in summary table.

Approximately 90% had a history of angina and approximately 65% had it within 4 weeks of enrollment. The Seattle Angina Frequency score was 81 (SAQ scores range from 0 to 100, with higher scores indicating less frequent angina, better function, and greater quality of life). Angina that was new onset or became more frequent within 3 months of enrollment occurred in 26% of the patients.

For ISCHEMIA-CKD, a total of 802 patients were enrolled and 777 underwent randomization at 118 sites in 30 countries. The median age of patients was 63 years, 69% were men, and 64% were white. Nearly all patients (92%) had hypertension and 57% had diabetes (insulin use not documented in summary table). The frequency of major morbidities included 17% with heart failure, 9% stroke, and 6% peripheral artery disease (atrial fibrillation and cerebrovascular disease not documented in summary table). The median LVEF was 58%. The median GFR was 23 ml/min and 14% of patients had GFR <15 ml/min. The median SAQ score was 79. Less information on pre-enrollment angina was provided compared to ISCHEMIA.

Procedures: In both trials, eligible patients were randomly assigned in a 1:1 ratio to an initial invasive strategy of medical therapy, angiography, and revascularization when feasible or to an initial conservative strategy of medical therapy alone, with angiography reserved for failure of medical therapy.

Those assigned to an invasive strategy underwent angiography within 30 days after randomization and complete revascularization of all ischemic territories if feasible. *Sites were provided prespecified guidelines for performing revascularization, including the use of FFR measurements when available and appropriate. Decisions about the type of revascularization, PCI or CABG, were deferred to the local heart team.

*In the ISCHEMIA-CKD trial, strategies to reduce risk of AKI included customized hydration protocol and a contrast-volume threshold provided to the site on the basis of the patient’s GFR and body weight. There were also protocols for PCI techniques involving the use of ultralow contrast volume or no contrast agent.

Medical therapy consisted of intensive secondary prevention with lifestyle and pharmacologic therapy applied equally in both groups with the use of prespecified treat-to-target algorithms.

Patients were followed at 1.5, 3, 6, and 12 months after randomization and every 6 months thereafter.

Endpoints: The primary endpoint of the ISCHEMIA trial was a composite of death from cardiovascular causes, MI, or hospitalization for unstable angina, heart failure, or resuscitated cardiac arrest.

The original trial design specified that 8,000 patients would undergo randomization with 4 years of follow-up for the primary composite endpoint; however, slow recruitment and lower than expected event rates led to modifications being made. In 2015 (enrollment began in 2012), new power calculations determined that 5,000 patients would have at least 83% power to detect an 18% relative reduction in the 4 year rate of the primary outcomes, based on an assumed rate of 20% in the conservative group (16.4% vs 20%).

*Power was re-estimated in 2018 based on a lower than expected event rate. Thus, the final sample size of the trial (5,178 patients) would confer at least 83% power to detect an 18.5% relative reduction in the primary endpoint, assuming an average follow-up of 3 years and an aggregative 4-year cumulative incidence of 14%. (something in ballpark of 13.25% vs 15%).

For ISCHEMIA-CKD, the primary endpoint was a composite of all-cause death or MI. The original planned sample size of 1,000 patients was revised to 650 patients because of slow recruitment. Power calculation performed in 2015 determined that enrollment of 500 patients and a mean follow-up of 3 years would provide a power of >81% to detect a reduction of 23-27% in the invasive strategy group based on a 4 year event rate of 60-75% in the conservative group.

*Power was re-estimated in 2018 based on a lower than expected event rate. The final sample size of 777 patients conferred a power of 80% to detect a relative reduction in the incidence of the primary endpoint by 22-24% in the invasive strategy group, assuming an aggregate 4-year event rate of 41-48% in the conservative group and an accrual of 240 to 270 events.

Results: The ISCHEMIA trial enrolled 8,518 patients and randomized 5,179. Data on screening to enrollment is provided in the supplemental files. Among the 3,339 (39%) who were excluded following enrollment, the main reasons for exclusion were lack of moderate to severe ischemia, according to stress core laboratory (1,350 [40% of those excluded and 16% of those enrolled]), did not have obstructive CAD (1,218 [36% of those excluded and 14% of those enrolled]), and had unprotected left main disease (434 [13% of those excluded and 5% of those enrolled]).

There were 2,588 patients in the invasive strategy group and 2,591 in the conservative strategy group. The median follow-up time was 3.2 years and approximately 2.0% withdrew or were lost to follow-up in both groups.

Compared to the conservative strategy, the invasive strategy did not significantly reduce the primary composite endpoint (16.4% vs 18.2%; HR 0.93; 95% CI 0.80-1.08) or secondary endpoints. All-cause death was numerically higher in the invasive strategy group but this was not statistically significant (9.0% vs 8.3%; HR 1.05; 95% CI 0.83-1.32).

No statistically significant subgroup interactions were reported for the primary composite endpoint. In the subgroup of patients with new or more frequent angina within 3 months prior to enrollment, who composed 26% of the overall cohort, the invasive strategy did numerically worse compared to the conservative strategy (17.6% vs 15.9%; p=NS). We only point this out based on our anecdotal experience as cardiologists that, to this day, patients with new angina or recent worsening of existing angina, often undergo revascularization without attempting to maximize medical therapy first.

It is also noteworthy to point out the subgroup of patients who did not meet eligibility criteria according to core laboratories, who accounted for 14% of all participants. Such patients appeared to do much worse with the invasive strategy (24.4% vs 16.3%; p=NS) but the CI is very wide for this group, reflecting its small size. We highlight this finding to emphasize the lack of evidence to support revascularization in patients with mildly abnormal stress tests who make up a significant percentage of abnormal stress tests in real life.

The ISCHEMIA-CKD trial enrolled 802 patients and randomized 777 of them. There were 388 in the invasive strategy group and 389 in the conservative strategy group. The median duration of follow-up was 2.2 years and 1.5% withdrew or were lost to follow-up. In the invasive strategy group, 50.2% underwent revascularization over a 3 year period compared to 19.6% in the conservative strategy group.

Compared to the conservative strategy, the invasive strategy did not reduce the primary endpoint of death or MI (36.4% vs 36.7%; HR 1.01; 95% CI 0.79-1.29) nor did it reduce the 5 component composite endpoint used in the ISCEMIA trial (38.5% vs 39.7%; HR 1.01; 95% CI: 0.79-1.29). Furthermore there was no difference in death (27.2% vs 27.8%) or MI (15.0% vs 15.9%). The invasive strategy did increase stroke (6.4% vs 1.6%; HR 3.76; 95% CI 1.52-9.32). The invasive strategy also increased the composite safety endpoint of death from any cause or initiation of dialysis in patients who were not receiving dialysis at baseline (44.8% vs 42.4%; HR 1.48; 95% CI 1.04-2.11). There were no statistically significant subgroup interactions reported.

Conclusion: In patients with stable CAD and moderate to severe ischemia on stress testing, an invasive strategy involving up-front revascularization and optimal medical therapy does not significantly reduce cardiac events or death compared to optimal medical therapy alone. The same is true in patients with advanced CKD; however, there are significant harms to an invasive strategy including a 3.5 times increased risk of stroke. In such patients who are not already on dialysis, an invasive strategy increased risk of death or initiation of dialysis.

Neither of the ISCHEMIA trials, involving approximately 6,000 randomized patients, provide evidence for an invasive strategy in patients with stable CAD and moderate to severe ischemia on stress testing. Instead, they join the lineage of negative trials testing up-front revascularization in patients with stable disease performed in the modern era.