Study Suggests Noncardiac Surgery Can Be Safe as Early as 3 Months After PCI

Download this article's Factoid (PDF & PPT for Gold Subscribers)

Use of drug-eluting stents (DES) compared with bare-metal stents (BMS) before noncardiac surgery does not increase the risk of adverse events, reports study a published online July 24, 2014, ahead of print in Catheterization and Cardiovascular Interventions. However, outcomes tend to be better when the gap between stenting and surgery exceeds 90 days, suggesting that patients may not need to wait a full year after percutaneous coronary intervention (PCI). 

Sripal Bangalore, MD, MHA, of the New York University School of Medicine (New York, NY), and colleagues looked at all patients undergoing noncardiac surgery within 1 year of PCI in Massachusetts from April 2004 to September 2007. Of 8,415 patients, 1,838 treated with BMS were propensity matched with 3,565 who received DES.

In the DES group, the 30-day rate of the primary net clinical outcome (death, MI, or bleeding) decreased with longer time from PCI to noncardiac surgery, with the lowest rates if the subsequent procedure was performed more than 90 days after PCI. Yet the event rate with BMS was uniformly high regardless of time (table 1).

Table 1. Rates of Death, MI, or Bleeding: Time from PCI to Noncardiac Surgery


1-30 Days

31-90 Days

91-180 Days

181-365 Days

P Value for Trend














Overall, there was no difference between the DES and BMS arms with regard to the 30-day primary net clinical outcome (6.64% vs 7.89%; P = .10), although DES were associated with a 26% reduction in death or MI vs BMS (OR 0.74; 95% CI 0.58-0.94) driven by fewer events when noncardiac surgery was completed more than 90 days after PCI. Bleeding was numerically higher in the DES group (1.90% vs 1.36%; P = .14)

When treatment groups were further stratified by surgical risk, there were no differences in the 30-day primary net clinical outcome between DES and BMS in the low-risk (P = .40) or high-risk (P = .52) groups. However, the rate was lower for DES vs BMS in the intermediate-risk cohort (3.64% vs 7.01%; P = .01).

Perioperative MI Data Missing  

For most PCI patients, it is hard to know whether they will undergo noncardiac surgery within a year, Dr. Bangalore told TCTMD in a telephone interview. Referencing the fact that the surgeries were evenly spread throughout the year after PCI, he said “most patients who receive a stent are also the patients who need some kind of noncardiac surgery.”

Although the current AHA/ACC guidelines recommend waiting at least a year before performing noncardiac surgery after PCI, Dr. Bangalore said, the results of his team’s study suggest an alternative. 

When performed after 90 days, “the risk of any events after noncardiac surgery is pretty low,” he explained. “One thing to take away from this analysis is potentially we don't have to wait for a year. Maybe 90 days would be good enough.” 

However, W. Scott Beattie, MD, PhD, of the University of Toronto (Toronto, Canada), told TCTMD in a telephone interview that he will not change his practice based on this study. In addition to the limitations of a retrospective design, he said, the study could not properly assess perioperative MI.

“If you don't measure troponin on everybody, you are going to miss about two-thirds of [those events] because most perioperative MIs are painless and non-ST-elevations,” he explained. 

Moreover, Dr. Beattie said, the ratio of death to MI in the study is about 5 to 1, but in the recently published POISE-2 trial in which troponin was measured in all patients after surgery, the ratio was flipped. “This is a higher-risk population so you'd expect more MIs…. So all of [the current study’s] conclusions are based on a problematic outcome,” he said. 

Lastly, the administrative codes used for bleeding do not assess blood transfusion, according to Dr. Beattie, “which is probably one of the best markers for perioperative bleeding.”

Choose the Stent Best for the Patient—Not for the Surgery

Dr. Beattie also criticized the study for not having a matched, high-risk non-PCI control arm, which Dr. Bangalore said was done purposefully to reduce confounding. 

Ultimately, “people who get stents have symptomatic coronary disease and they need to be treated…as well as possible,” Dr. Beattie said, adding that the choice between stent types should not depend on whether the patient may need future surgery. “Management [of these patients] is a multifactorial process of care; timing is just one factor,” he concluded.


Bangalore S, Silbaugh TS, Normand S-LT, et al. Drug-eluting stents versus bare metal stents prior to noncardiac surgery. Catheter Cardiovasc Interv. 2014;Epub ahead of print.

Related Stories

  • Drs. Bangalore and Beattie report no relevant conflicts of interest.