Perioperative MIs Bring Too Many Patients Back to the Hospital After Noncardiac Surgery

ORLANDO, FL—One in five patients with perioperative myocardial infarction following noncardiac surgery are readmitted to the hospital within 30 days, and one in seven die within the same period, according to a new study.

The findings were presented in a poster session today at the American College of Cardiology 2018 Scientific Session by Nathaniel Smilowitz, MD (New York University School of Medicine, NY), and simultaneously published in Circulation.

“I hope that [these findings] will increase attention to this topic, because people have ignored it,” Christian Mueller, MD, PhD (University Hospital of Basel, Switzerland), who was not involved with the study, told TCTMD. “It's an iatrogenic complication, and of course we hate to talk about iatrogenic complications. And it's bad for business [since] surgeries are an important source of income for hospitals. In fact, I think patients and their caregivers would reconsider undergoing surgery if they would be aware of the true risk/benefit ratio of many operations.”

For the study, Smilowitz and colleagues looked at data from almost 4 million hospitalizations for noncardiac surgery in the 2014 United States Nationwide Readmission Database. Perioperative MI was identified in 8,085 patients.

More than 1,000 patients with perioperative MI (14.0%) died during the index hospital admission compared with only a few (0.3%) in the nonperioperative MI arm (P < 0.001). The rest were almost three times more likely to be readmitted to the hospital within 30 days than their counterparts without perioperative MI (19.1% vs 6.5%; P < 0.001), and these readmissions were most often due to infections (30.0%), cardiovascular complications (25.3%), and bleeding (10.4%). For those with perioperative MI readmitted within 30 days, in-hospital mortality was 11.3%.

Risk of in-hospital death during the index admission was highest among patients with perioperative MI undergoing thoracic surgery (28.4%), general surgery (20.3%), vascular surgery (15.3%), and neurosurgery (12.5%).

By 6 months, 42.4% of patients with perioperative MI who were discharged after their index hospitalization were readmitted and 17.6% of patients had died.

Of note, patients with perioperative MI were generally older, more likely to be male, and had a greater number of cardiovascular risk factors than those without. Coronary revascularization was only performed in 15.8% of patients with perioperative MI—12.5% of NSTEMI and 30.4% of STEMI patients.

Confirming Prior Research

“It is well known to the community that perioperative MI associated with poor outcomes,” Xuming Dai, MD (NewYork-Presbyterian, Flushing, NY), also commenting on the study, told TCTMD. “This data added increasing 30-day readmission rate to the list.”

The fact that those who experienced perioperative MI were older and sicker “certainly contributed to the worse outcomes,” Dai said in an email, adding that a “propensity-score analysis or other statistic adjustment might provide more information on how much of these were actually due to the perioperative MI.”

Dai acknowledged that the low rate of revascularization for patients with perioperative STEMI and NSTEMI as well as their high in-hospital mortality “were consistent with data published by our group in recent years.” It might be interesting to determine if intervening on these MIs would have “any impact on readmission and mortality,” he added.

Mueller also said that the study findings were in line with what his team has concluded in prior research. However, he commented, the incidence of perioperative MI described here is “just the tip of an iceberg.” Because the researchers used administrative data, Mueller explained, “they were only able to capture perioperative myocardial infarctions that were detected by the clinicians and caused enough issues and problems that the clinicians decided to code them. It's fair to assume that the true incidence of relevant perioperative myocardial injury and/or infarction may be 20 times higher.”

The main reason many of these would be MIs go undiscovered, he said, is because “the vast majority of patients that suffer a perioperative myocardial infarction do not experience pain” due to the analgesics they are on. “Thereby we would only detect these events if we apply a sensitive method that is kind of independent of the patient reporting pain.” Both the VISION study and Mueller’s own research have now documented that “only about 10% of patients maximum really suffer chest discomfort within their perioperative myocardial infarction,” he said. “So the problem is even much larger than estimated in their report.”

A Call for More Scrutiny Before Surgery

More work should be done to determine which patients should and should not be offered certain surgeries. Clinicians should also be more fastidious about monitoring specific high-risk patients after their surgeries, Mueller suggested. But even more so, “if we are honest and aware that relying on the patient to report pain detects only one out of 10 [perioperative MIs], then we should be honest enough and smart enough and say, ‘Okay, we have to apply the screening process that helps us detect the other nine.’”

That can “easily be done” by measuring cardiac troponin before and after operations, he continued. A substantial rise indicates that damage has occurred to the patient’s myocardium, and “we can rapidly identify the patient and consider what kind of management consequences this event could have.” Only “very few institutions worldwide” have initiated systematic troponin screening programs to detect perioperative myocardial infarction, Mueller said.

Prevention is another issue entirely. Anesthesiologists can play a large role in this by increasing the mean perioperative blood pressure measurements they deem appropriate, he commented. “Their current standards are that surgical patients maintain a mean blood pressure in the range of 60 mg Hg during the operation, [but] the fact that the incidence of perioperative myocardial infarction is so high I think clearly highlights that this is inaccurate. Likely if you would have the option to monitor sufficient blood supply to the myocardium, it would, even during the operation, allow us to take appropriate measures to in fact prevent this perioperative myocardial infarction to occur.”

Future studies should focus on broader screening programs for patients prior to surgery as well as mechanisms to better identify the different pathophysiologies that underlie perioperative MI, according to Mueller, who said, “In many patients, it's not so easy to identify the true pathophysiology.”

Lastly, “much more attention should be given to . . . particularly the immediate postoperative period [when] patients are often not appropriately monitored,” Mueller concluded.

Dai said that while “cardiologists are trying hard to use existing risk stratification models to assess cardiac risk for noncardiac surgery,” there is currently no single solution for preventing perioperative MIs. “We need more, especially prospective, studies for this,” Dai advised.

He noted that he is participating in a multicenter consortium working to understand the current practice of in-hospital STEMIs, of which 50% of are perioperative, and to improve their care.