Mortality Sky-High After MI at Major Vascular Surgery, But MI Risk Low

December 06, 2019

The risk for peri- or postoperative myocardial infarction (MI) in patients undergoing major vascular surgery is fairly low, but those who experience such MIs have had an outsized risk of dying within a year, despite good postdischarge medical management, investigators conclude based on an observational study.

The MI risk within a month of surgery was 1.6%, but patients having an MI died within a year at more than five times the rate of patients without MI, in an adjusted analysis.

Such mortality, most of which was within 48 to 72 hours of the procedure, was higher among patients who had undergone endovascular aortic aneurysm repair (EVAR) or peripheral bypass procedures than among those who had undergone other vascular surgeries.

The surgical patients with conditions like diabetes or structural heart diseases were most likely to experience an MI, as were those whose procedures consisted of open abdominal aortic aneurysm (AAA) repair or peripheral arterial bypass.

And in a less obvious association, MI risk was significantly elevated in patients who were preoperatively on P2Y12 inhibitors like clopidogrel, in the analysis based on 2012 to 2017 insurance data from more than 26,000 patients in Michigan.

One of the study's big messages is that the MI rate in these surgical patients was low, compared with historical expectations, "but if you had one, mortality was about 37% out to 1 year," Peter K. Henke, MD, vascular surgeon at University of Michigan, Ann Arbor, told theheart.org | Medscape Cardiology.

"Evidence-based guidelines were routinely followed for post-MI care," he pointed out. Overwhelmingly, most patients in the analysis were discharged on statins, beta blockers, antiplatelets, and other guideline-directed meds for secondary prevention.

"Even with that, we still had that high level of mortality," said Henke, who is lead author on the study, published December 4 in JAMA Surgery, with lead author Robert J. Beaulieu, MD, of the same institution.

That the patients were seemingly medically well managed highlights another key message of the study, Henke observed.

The identified risk factors for MI were mostly unmodifiable, and that, along with the low MI rate, suggests "we're probably hitting up against something of a natural limit" in evidence-based measures to mitigate MI risk at vascular surgery. And contemporary guidelines don't offer many recommendations for prolonging survival after periprocedural MI, he said.

"More research is needed to find the best consistent therapies to improve long-term outcomes. A nearly 40% mortality at 1 year is really significant."

Mortality that high "is almost for certain a real signal, and I think other people have shown that," P.J. Devereaux MD, PhD, McMaster University, Hamilton, Ontario, Canada, told theheart.org | Medscape Cardiology.

In general, he said, patients who have peri- or postoperative events "still tend to have suboptimal medical management, although the use of optimal meds tends to be better after vascular surgeries than after other kinds of noncardiac surgery.

Devereaux, who isn't associated with the current analysis, said the findings are important for "drawing attention to what is a big problem. Where we fail these patients is we don't get medical follow up, and we need to change that."

The low rate of procedure-related MI is also noteworthy, observed Kim A. Eagle, MD, also from the University of Michigan, but not a coauthor of the study.

It was less than 2% "likely because they were protected by evidence-based medicine," he told theheart.org | Medscape Cardiology by email.

The high 1-year mortality "has to do with the burden of atherosclerotic cardiovascular disease in those who break through on preventive treatment and have either unusual perioperative demand and/or plaque rupture. The notion that we can somehow prevent all of these in chronically diseased patients is absurd."

Still, it may be possible to improve the risk stratification of these patients before surgery and mark them for intensified care, Henke said. He and his colleagues observed about one-half the risk for peri- or postoperative MI in patients with a normal preoperative electrocardiograms (ECG). But more than half of those without an MI had abnormal ECGs, limiting the predictive value.

Those with MI were also significantly more likely to have an abnormal preoperative stress test result, and significantly less likely to have a normal one. However, 71.7% of the cohort "did not undergo cardiac stress testing prior to their operation, which limited the ability to make meaningful assertions about the role of this potentially useful evaluation," the group writes.

Perhaps more promising, Henke said, would be broader use of biomarkers for preoperative risk stratification; troponins and natriuretic peptides for this purpose are gaining traction in the literature.

"It may be that patients who have positive biomarkers, or positive stress tests, need to get better preoperative management. Better titration of blood pressure, for example, or increasing their statin dose may help with plaque stability," he said.

Compared with some forms of imaging, observed Devereaux, "you can do biomarkers at a 20th of the cost, get the result in minutes, and it predicts way better. I think that's the direction people are going to go in, because it's just so much more accurate, so much cheaper, and so much quicker and more convenient."

Preventive oral anticoagulation (OAC) may also protect patients with periprocedural MI and could potentially reduce 1-year mortality. In the MANAGE trial, "we took patients who had myocardial injury for noncardiac surgery, which includes MIs, and showed that with an intermediate-dose anticoagulant, you decrease the risk of major recurrent events," said Devereaux. He was principle investigator of the trial, which randomized the patients to receive dabigatran (Pradaxa, Boehringer Ingelheim) or placebo.

Henke also singled out the MANAGE trial. "That may be an area that needs further study in the postop patient who has an MI. Should that be more routinely adopted? Probably so," he said.

"There's a lot more that can be done for these patients. We need a lot more research on how to improve their outcomes," Devereaux said.

The current analysis looked at 26,231 patients undergoing vascular surgery in a prospective database that was part of a Michigan statewide quality-improvement initiative; their mean age was 69 years and about two-thirds were men.

The surgical procedures included EVAR, open AAA repair, peripheral arterial bypass, carotid endarterectomy, and carotid stenting. Peri- or postoperative MI, which occurred in 410 patients, was defined as MI within 30 days of the procedure by electrocardiography and biomarkers consistent with the universal definition of the time.

Adjusted Odds Ratio (OR) for MI Within 1 Month of Major Vascular Surgery
Predictor OR (95% CI) P Vaue
Open AAA repair 4.53 (2.73–7.52) <.001
Peripheral artery bypass 2.375 (1.818–3.102) <.001
Heart failure 1.519 (1.163–1.983) .002
Diabetes 1.514 (1.201–1.907) <.001
Valvular disease 1.447 (1.024–2.046) .04
Coronary artery disease 1.381 (1.058–1.803) .02
Preoperative P2Y12 inhibitors 1.365 (1.080–1.725) .009
Age 1.032 (1.019–1.045) <.001

Of the patients with MI within 30 days of the procedure, 37.4% had died within a year of surgery, compared with only 5.1% of patients without an MI (P < .001). Among those with MI, 1-year mortality was 45.3% for those who had undergone peripheral bypass surgery, 41.2% for those who had EVAR, and 33.3% for those who had carotid surgery.

Adjusted Odds ratio (OR) for Death in the Year After Major Vascular Surgery
Predictor OR (95% CI) P Value
MI within 1 month of surgery 5.62 (4.04–7.84) <.001
Heart failure 1.93 (1.61–2.3) <.001
Chronic obstructive
pulmonary disease
1.73 (1.48–2.02) <.001
Diabetes 1.35 (1.15–1.60) <.001

That preoperative treatment with P2Y12 inhibitors would worsen MI risk may seem counterintuitive but is consistent with other evidence.

"I suspect that's due to unmeasured confounders, meaning that those patients on Plavix or others in that class of drugs have a greater burden of coronary heart disease," Henke said.

Also, those on an antiplatelet before the surgery had a higher rate of periprocedural bleeding, he observed, "and that was associated with blood transfusion. If you required a blood transfusion, your risk of MI was significantly elevated."

Devereaux agreed on both points. "I suspect it's more the bleeding, or it also could relate to the fact that these are high-risk patients. They're on Plavix for a reason."

Henke and the other authors reported no conflicts. Devereaux has reported that members of his research group "do not accept honoraria or other payments from industry for their own personal financial gain. They do accept honoraria/payments from industry to support research endeavors and costs to participate in meetings"; and that he has received grants from Abbott Diagnostics, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Coviden, Octapharma, Philips Healthcare, Roche Diagnostics, and Stryker; and has participated in an advisory board meeting for GlaxoSmithKline and an expert panel meeting with AstraZeneca and Boehringer Ingelheim.

JAMA Surgery. Published online December 4, 2019. Abstract

Follow Steve Stiles on Twitter: @SteveStiles2. For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....