Surgeons Argue: STS Data Again Show ISCHEMIA Isn’t Definitive for CABG

The STS registry results bring yet another argument against the downgrade of CABG in the 2021 revascularization guidelines.

Surgeons Argue: STS Data Again Show ISCHEMIA Isn’t Definitive for CABG

SAN DIEGO, CA—A new analysis that compares patients in the Society of Thoracic Surgeons (STS) Adult Cardiac Surgery database against those of the ISCHEMIA trial has many surgeons once again calling for a reevaluation of the most recent US coronary revascularization guidelines. The observational study was the second to spark discussion during the Society of Thoracic Surgeons (STS) annual meeting last weekend.

As previously reported, the 2021 revascularization guidelines from the American Heart Association (AHA), American College of Cardiology (ACC), and Society for Cardiovascular Angiography and Interventions (SCAI) downgraded CABG to a 2b recommendation for patients with stable multivessel disease. Neither the STS nor the American Association for Thoracic Surgery (AATS) ultimately endorsed the guidelines.

Earlier in the meeting, an observational analysis of Medicare data sought to reopen the debate about whether this deemphasis of surgery was warranted.

Joseph Sabik III, MD (University Hospitals Cleveland Medical Center, OH), who presented the findings from the STS database, told TCTMD he believes that using ISCHEMIA to justify the downgrade was an “overreach.”

For one, ISCHEMIA was not powered to specifically look at CABG outcomes but rather compared optimal medical therapy alone or with the addition of invasive coronary angiography, followed by an intervention—PCI or CABG—if needed. In the trial, 20% of patients randomized to the invasive strategy didn’t undergo revascularization, and of those who did, just 26% were treated with CABG. Additionally, the patients who went on to have surgery had fewer comorbidities and complications than the patients enrolled in the STS database between July 2020 and June 2022. For these reasons, Sabik argued that “to write guidelines based on that is a little bit worrisome.”

Sabik had served as the STS’ representative on the guideline-writing committee but stepped off before publication due to differences of opinion. In his presentation, he called for “an urgent reevaluation [of the guidelines] to align coronary surgery recommendations with contemporary evidence-based clinical practice.”

Commenting on the findings for TCTMD, primary ISCHEMIA investigator David J. Maron, MD (Stanford University School of Medicine, CA), said while the data Sabik presented “make perfect sense,” the guideline writers did not make the decision to downgrade CABG based solely on the ISCHEMIA findings. “But I do think that the publication of our results led the committee to reevaluate the totality of evidence,” he added. “And when you look at meta-analyses of randomized controlled trials, with or without ISCHEMIA, there is not a mortality benefit from CABG in stable patients with a normal ejection fraction.”

Despite the current 2b recommendation, the new guidelines haven’t changed his referral patterns, he commented, adding that it has “made for interesting conversations.”

“I'm happy to continue to recommend revascularization for patients who are clearly impacted in terms of their quality of life, and that is an easy decision,” said Maron.


In his presentation, Sabik looked at all 299,478 enrolled patients in the STS Adult Cardiac Surgery Database who underwent isolated CABG between July 2020 and June 2022 and had complete CAD data. After applying the five main exclusion criteria from ISCHEMIA to the cohort—LVEF < 35%, left main stenosis ≥ 50%, NYHA class III/IV heart failure, GFR < 30 mL/min, and severe ischemia—only about one-third of the original cohort remained (n = 105,698).

Generally, the ISCHEMIA participants were younger and less likely to have hypertension, diabetes, previous stroke, peripheral vascular disease, renal dysfunction, and prior MI compared with those in the STS database. Also, ISCHEMIA patients had better LV function and less extensive coronary artery disease.

In comparing the expected outcomes between the STS patients who would and would not have been eligible for ISCHEMIA, the former group had a lower predicted risk of morbidity and mortality than the latter (8% vs 12%).

Most of the ISCHEMIA patients were stratified into Duke Prognostic Groups 4 (29.5%), 5 (35.5%), and 6 (26.2%), meaning they had moderate or severe two-vessel CAD or moderate three-vessel CAD. However, Sabik noted the STS population tended to be stratified into categories 5 (20.3%), 6 (40.3%), and 7 (32.2%). Even after excluding all patients with left main disease in category 7 and also only comparing ISCHEMIA eligible patients, those in the STS database still had more complex CAD.

“Therefore, the ISCHEMIA study should not have been used to decrease the survival class of recommendation for coronary artery bypass surgery versus medical therapy in patients with stable ischemic heart disease and triple-vessel CAD,” Sabik concluded.

Calls for Reevaluation of the Guidelines

Following the presentation, session co-moderator and STS President Thomas MacGillivray, MD (MedStar Washington Hospital Center, Washington, DC), commented that Sabik’s data, in conjunction with the previous Medicare analysis, had helped further clarify the benefits of CABG.

“If I heard you clearly, the message is that the ISCHEMIA trial showed that patients who wouldn't necessarily be considered for [CABG] don't need CABG, which is not the same as CABG is not an effective therapy for people who would benefit from CABG,” said MacGillivray.

Sabik agreed. “The other thing that is very interesting to me, particularly when you really dive into the coronary anatomy and disease, [is] there are so many patients with just moderate disease in ISCHEMIA who weren't getting the benefit from PCI,” he added. “But you're right. Essentially, we took a group of patients who would very have likely been referred for CABG and then applied the outcomes of those patients to determine whether a whole different population of patients should have CABG.”

We've have had several meetings with ACC and AHA leadership. They've been difficult, to say the least. Joseph Sabik III

Regardless of the ultimate driver for the downgrade in CABG for three-vessel disease in the latest revascularization guidelines, MacGillivray asked what the consequences would be for those who only do a “top-line read” of the guidelines and dismiss CABG for all stable coronary disease.

Sabik said MacGillivray’s take on this issue was prescient, because “patients [who are] presenting with acute coronary syndrome, once they are stabilized, are being sent home because they ‘don't need surgery’ according to the guidelines, and we know that's incorrect.”

Audience member Joseph Bavaria, MD (Penn Medicine, Philadelphia, PA), commented that he “used to think that most of the randomized trials were the gold standard, but [now I think] that huge, big data sets like this may be actually better.” He asked if there was any hope for an upcoming reevaluation of the guidelines.

“We've have had several meetings with ACC and AHA leadership,” Sabik reported. “They've been difficult, to say the least. One of the things that has really surprised me is they feel, ‘We had our procedures, we followed our procedures, and this is the outcome. And so therefore we have got to push it forward.’ And in all honestly, you get into these meetings, and you don't get the feeling or the impression that the ACC and AHA leaders believe these [guidelines] are correct. We are still working on that, and it has been difficult.”

As for the comparison between randomized and observational data, Sabik called it a “balance” since “randomized studies have bias at the point of entry [and] observational studies have bias at the point of treatment.” When randomized trials are applied to situations inappropriately, as he argued was done with the guidelines and ISCHEMIA, then “our best available evidence is our observational studies,” said Sabik.

Maron disagreed. “The guidelines should be based on the best available evidence, and the best available evidence is from randomized controlled trials,” he said. “If randomized controlled trials show an improvement in survival, then it should be a class 1 recommendation.”

What will “get a little trickier” is when the extended follow-up of ISCHEMIA are released, Maron continued. Current interim data found a reduction in cardiovascular deaths in patients who were randomized to the invasive strategy, including those who underwent CABG, he reported. “So I think it will get a little more nuanced in discussing improved cardiovascular survival, as opposed to overall survival.”

If those results pan out, Maron said he would be in support of the guidelines classifying CABG as class 1 for improving cardiovascular survival in these patients.

If randomized controlled trials show an improvement in survival, then it should be a class 1 recommendation. David J. Maron

Calling the data “outstanding,” audience member Patrick Myers, MD (Lausanne University Hospital, Switzerland), secretary general of the European Association for Cardio-Thoracic Surgery (EACTS), said it is imperative they be published as soon as possible. “The guidelines committee can only work with published data,” he stressed. “This is the kind of manuscript that we need out today or even yesterday to help us.”

The current challenge for surgeons, according to MacGillivray, “is that the signal-to-noise ratio of the conversation among cardiologists is very much favoring the noise and not the signal. I think that being armed with this data when we go to our heart team meeting [will enable us to] consistently deliver this message of the value of CABG in the improvement of patients. That responsibility rests on all of us to do that and participate in the heart team meetings and deliver that signal.”

  • Sabik JF. The ISCHEMIA study is not representative of patients undergoing coronary surgery: an STS Adult Cardiac Surgery Database analysis. Presented at: STS 2023. January 23, 2023. San Diego, CA.

  • Myers, MacGillivray, and Maron report no relevant conflicts of interest.
  • Sabik reports serving on the advisory board and as a speaker for Medtronic.