Postpone Noncardiac Surgery at Least a Year in Patients with Stents of Any Type: Registry Analysis

Placement of a coronary stent—regardless of type—within the last year more than doubles the risk that patients might have major adverse cardiac and cerebrovascular events (MACCE) after noncardiac surgery, a registry study suggests. However, beyond 1 year, stented patients appear to be at no excess risk. 

Implications: Postpone Noncardiac Surgery at Least a Year in Patients with Stents of Any Type: Registry Analysis

The data add to accumulating evidence challenging the recommendations regarding the timing of noncardiac surgery, especially in relation to BMS, which have long been considered the safer choice in patients who will need surgery soon after a stent.

In the study, published online ahead of print in the March 8, 2016, issue of the Journal of the American College of Cardiology, Karim D. Mahmoud, MD, PhD (Mayo Clinic; Rochester, MN), and colleagues looked at 24,313 patients who had general or specialty surgery, most under general anesthesia, at their institution between 2006 and 2011. Of those, 1,120 (4.6%) patients had a previous DES or BMS.

Early, Persistent Risk Seen

The rate MACCE at 30 days post-surgery (primary endpoint) was more than double in patients with vs without stents. Perioperative death and MI both were significantly elevated, while cardiac arrest and stroke were not.

Outcomes at 30 days - stent non cardiac

Analysis based on timing of surgery showed a strong relationship between perioperative risk of MACCE and prior stent implantation in the past year. Again, this was true for death and MI, but not for cardiac arrest or stroke.

The highest risk was seen in patients who had undergone stent implantation within the prior 6 months. While this risk declined with longer time intervals, an elevated risk of MI remained—at about 7 times that of controls—out to 1 year. Beyond 1 year, however, patients with stents had similar rates of MACCE as those without stents. The risk of bleeding in patients with stents was slightly higher than in those with no stents and also was found to be time dependent. Those who experienced bleeding also were at higher risk for MACCE and MI.

Surprisingly, when patients with a recent MI were excluded from analysis, the difference in of MACCE at 6-12 months after stenting was significant, suggesting that even those patients considered to have more stable CAD have a higher risk of adverse events from surgery than nonstented patients.

Propensity-adjusted analysis that separated patients who had DES (n = 402) versus BMS (n = 375) showed that the excess perioperative risk was not restricted to those with DES alone. In fact, in patients with a prior BMS, the risk of MACCE at 6-12 months compared with unstented controls was actually much higher than in the DES group (adjusted OR 4.21 vs 1.03, respectively).

Are the Guidelines Wrong?

The study authors say their findings challenge the US and European guidelines, both of which recommend that noncardiac surgery be delayed only 1 month after BMS.

“This recommendation seems to assume that stent thrombosis is the primary mechanism of perioperative myocardial infarction and that re-endothelialization provides protection against it,” Mahmoud and colleagues say. Instead, stent thrombosis occurrence in the study was uncommon, with most perioperative MIs shown to be the result of new lesions or a myocardial oxygen supply/demand imbalance.

The study authors say their data, in combination with a recent report from the Veterans Affairs database and another large registry report, seem to suggest that there is a “vulnerable period of up to 1 year in which a patient undergoing stenting is particularly susceptible to the prothrombotic state and hemodynamic variations that surgery may produce.” For those reasons, they say, the guidelines should be reevaluated. 

Commenting on the study for TCTMD, Emmanouil S. Brilakis, MD, PhD, of the Dallas VA Medical Center (Dallas, TX), said the assumption that BMS are the better option in patients with comorbidities and known need for future surgery appears to be changing.

“In a way, the bare-metal safety may have been overvalued and actually may not be as safe as we thought,” he observed, adding that he is in agreement with the study authors that the guidelines should be reevaluated. Despite the limitations of the study, which include lack of detailed stent-related data and incomplete angiographic assessment in the entire cohort, he noted that these numbers and those from the VA database, “are the best data we have, and they do show, if anything, worse outcomes for bare-metal stent patients.”

Additionally, the lack of stent thrombosis and the observation of new lesions and myocardial oxygen supply/demand imbalance as the cause of perioperative MIs are novel and important findings of the study, Brilakis said. Taken together what these show, he added, is that other aspects of the myocardium beyond the stent area are involved.

“Patients who have stents likely have more severe coronary artery disease [than patients without stents], including other potentially unstable plaques or other intermediate or severe lesions not treated at the time of stenting and those other lesions can cause problems in the increased pressure environment of surgery,” he said.

Brilakis said the data help in the decision making for patients who need surgery by confirming that stenting should be avoided and that if a stent is absolutely necessary, a second-generation everolimus-eluting DES should be the first choice, not a BMS. For patients with an existing stent, any truly elective surgery should be postponed at least a year regardless of stent type, he advised.

But in an editorial accompanying the study, Prashant Vaishnava, MD (Mount Sinai Heart; New York, NY) and Kim A. Eagle, MD (University of Michigan Health System; Ann Arbor, MI), say it is too soon to rethink guideline recommendations based on observational registries.

“We would urge caution about this suggestion and rather encourage more scrutiny into the circumstances that may have motivated a patient to receive a BMS,” they write. For example, since the reasons why BMS may have been chosen over DES were not known to the study authors, it is possible that those patients may have had comorbidities that increased their risk for perioperative MI, Vaishnava and Eagle note.

One thing the study does not address but that will be important going forward, Brilakis added, is whether anything can be done to prevent or minimize the risk of MACCE in a patient with a stent who must undergo surgery within 1 year.

1. Mahmoud KD, Sanon S, Habermann EB, et al. Perioperative cardiovascular risk of prior coronary stent implantation among patients undergoing noncardiac surgery. J Am Coll Cardiol. 2016;67:1038-1049.
2. Vaishnava P, Eagle KA. Coronary stents and risk for noncardiac surgery much ado about something, nothing, or DAPT? J Am Coll Cardiol. 2016;67:1050-1052. 


  • Mahmoud, Vaishnava, and Eagle report no relevant conflicts of interest. 
  • Brilakis reports consulting/speaker honoraria from Abbott Vascular, Asahi, Boston Scientific, Elsevier, Somahlution, St Jude Medical, and Terumo; research support from InfraRedx and Boston Scientific; and that his spouse is an employee of Medtronic. 

Related Stories: