Delayed Primary PCI Generally Superior to Early Fibrinolysis for STEMI

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


Patients experiencing an ST-segment elevation myocardial infarction (STEMI) who receive primary percutaneous coronary intervention (PCI) within 2 hours of alerting the medical system are less likely to die in both the short and long term than those who are given early fibrinolytic therapy, according to a study published online July 15, 2011, ahead of print in the American Journal of Cardiology. Moreover, even when primary PCI is delayed longer than 3 hours, patients have similar mortality compared with those given fibrinolysis within 2 hours.

The findings also confirm that regardless of the type of reperfusion therapy, treatment delay is associated with incremental increases in mortality.

In an analysis based on the DANAMI-2 (Danish Acute Myocardial Infarction-2) trial, investigators led by Michael Maeng, MD, PhD, of Aarhus University Hospital (Skejby, Denmark), looked at the relative impact of system delay (defined as the interval from first medical contact to start of fibrinolysis or contrast injection) on 1,572 STEMI patients randomized to primary PCI or fibrinolysis.

Fibrinolysis Typically Available Earlier

Patient delay, or the time from symptom onset to first medical contact, was similar for both reperfusion groups. However, the median system delay was shortest for the fibrinolysis group (90 minutes), with longer delays for patients sent directly to primary PCI hospitals (127 minutes) and those who went to a referral hospital first (148 minutes; P for trend < 0.001). Interestingly, the median door-to-balloon time for direct primary PCI patients was only 14 minutes, while door-to-needle time was 50 minutes and door-to-balloon time for transferred primary PCI patients was 127 minutes (P for trend < 0.001).

At 30 days, 113 patients had died, 61 in the fibrinolysis group and 52 in the overall primary PCI group. Prolonged system delay was associated with incremental increases in mortality for both strategies (table 1).

Table 1. Thirty-Day Mortality by System Delay

 

< 1 Houra

1-2 Hours

2-3 Hours

> 3 Hours

P Valueb

Fibrinolysis

5.6%

6.9%

9.5%

11.5%

0.008

Primary PCI

2.6%

7.5%

7.7%

0.02

a Because only 6 PCI patients met this time frame, they were included in the 1-2 hour category.
b For trend.

However, the efficacy of primary PCI was somewhat less sensitive to delay. When invasive treatment was performed within 1 to 2 hours of first medical contact, there was a trend toward lower mortality compared with fibrinolysis administered within 1 hour (2.6% vs. 5.6%; adjusted HR 0.33; 95% CI 0.10-1.19; P = 0.07). In addition, mortality with PCI at 1 to 2 hours was clearly reduced compared with fibrinolysis within 1 to 2 hours (2.6% vs. 6.9%; adjusted HR 0.37; 95% CI 0.14-0.95; P = 0.04). When primary PCI was delayed more than 3 hours, it yielded a mortality rate similar to that of fibrinolysis given within 1 to 2 hours (7.7% vs. 6.9%; adjusted HR 0.97; 95% CI 0.52-1.83; P = 0.94).

At a median follow-up of 7.8 years, 442 patients had died, 233 in the fibrinolysis group and 209 in the primary PCI group. Again, longer system delay was associated with increased mortality in both groups (table 2).

Table 2. Eight-Year Mortality by System Delay

 

< 1 Houra

1-2 Hours

2-3 Hours

> 3 Hours

P Valueb

Fibrinolysis

23%

29%

31%

38%

0.02

Primary PCI

22%

26%

29%

0.04

a Because only 6 PCI patients met this time frame, they were included in the 1-2 hour category.
b For trend.

Primary PCI within 1 to 2 hours produced the best long-term outcome of all the permutations of treatments and system delays, showing numerically lower mortality vs. fibrinolysis within 1 hour (21.8% vs. 23.4%; adjusted HR 0.74; 95% CI 0.49-1.27; P = 0.31) and a trend toward superiority over fibrinolysis within 1 to 2 hours (21.8% vs. 29.4%; adjusted HR 0.74; 95% CI 0.53-1.03; P = 0.07). Again, primary PCI delayed by more than 3 hours showed mortality rates similar to those of fibrinolysis with a delay of 1 to 2 hours (29.3% vs. 29.4%; adjusted HR 0.87; 95% CI 0.63-1.19; P = 0.38).

The investigators observe that “[u]sing the same recommended [time] limits for fibrinolysis and [primary PCI] . . . seems an incorrect simplification. Even when fibrinolysis can be initiated within 1 hour of first medical contact, primary PCI remains the better strategy provided it can be performed within 2 hours, a finding that is in accord with the European STEMI guidelines. Furthermore, even when primary PCI is subjected to a system delay of more than 3 hours, it does not increase mortality, they add.

The goal of consistently initiating primary PCI within 2 hours is achievable for most American patients, the authors write, because 80% of adults in the United States live within 60 minutes of a hospital that offers primary PCI.

A Broader Perspective on Delay

“This paper is a good reminder that for both forms of reperfusion, system delay can incur additional risks,” said Harlan M. Krumholz, MD, SM, of Yale University School of Medicine (New Haven, CT), in a telephone interview with TCTMD.

“We’ve done very well in improving the timeliness of treatment at the point the patient comes into the [PCI] hospital,” he added. “Now this study says we have to take a broader view and try to [minimize] any delay from the time the patient activates the healthcare system.”

Sorin J. Brener, MD, of Weill Cornell Medical College (New York, NY), made a similar point. “I find this analysis much more relevant than looking at door-to-balloon or door-to-needle times,” he told TCTMD in a telephone interview, noting that the total ischemic time is far more important for outcome than any particular component of delay.

The main message of the study is that even late primary PCI is as good as early fibrinolysis, said Dr. Brener, adding that the data suggest that although the efficacy of both reperfusion methods declines over time, PCI may be somewhat less time dependent than fibrinolysis.

He cautioned, however, that the study assumes the quality of PCI and the type of lytics used are the same at all hospitals, when in fact they are not.

Patient Characteristics May Play a Role in Treatment Choice

Dr. Krumholz said the study cannot answer the question of exactly when primary PCI is no longer worth the prolonged delay it might entail, although the data are consistent with the idea that when PCI is delayed an extra hour or 2 beyond the time when lytics could be given—typically at a referral hospital—a patient may benefit more from early pharmacological therapy. But that crossover point remains unsettled, he said, and importantly patient characteristics may tip a decision toward one treatment or the other.

Dr. Brener likewise stressed that both the excess time to PCI and multiple factors such as the patient’s age and MI location need to be taken into account. “For example, a 40-year-old who presents with a large anterior wall MI within an hour of symptom onset will probably benefit more from lytics than delayed angioplasty,” he said. “[The contrary is true for] an 80-year-old who presents with an inferior wall MI 4 hours from symptom onset where lytics are dangerous because of the risk of bleeding and the overall myocardium at risk is not that large.”

Dr. Brener added that even when lytics are given, patients should still be taken to a PCI-capable hospital in case rescue angioplasty is needed.

He noted that the current findings are broadly in tune with US guidelines, which recommend that primary PCI be performed if the patient presents within 3 hours of symptom onset and PCI can be done within 90 minutes of first medical contact; otherwise, fibrinolysis within 30 minutes is preferred.

Key to reducing system delay, Dr. Krumholz noted, is coordination between referral and PCI-capable hospitals, in particular having a plan in place to quickly diagnose patients and transfer them as soon as possible. Just as important, Dr. Brener added, is having EMS personnel interpret an ECG onsite and, except in rural settings involving long distances, transport patients directly to a PCI-capable hospital.

Study Details

Patients assigned to fibrinolysis received 300 mg of aspirin orally, an IV beta blocker, and tissue plasminogen activator (alteplase; given as a 15-mg bolus and an infusion of 0.75 mg per kilogram of body weight over 30 minutes, followed by an infusion of 0.5 mg per kilogram for a period of 60 minutes), and an IV bolus of unfractionated heparin (5,000 U) followed by a 48-hour infusion of unfractionated heparin.

 


Source:
Nielsen PH, Terkelsen CJ, Nielsen TT, et al. System delay and timing of intervention in acute myocardial infarction (from the Danish Acute Myocardial Infarction-2 [DANAMI-2] trial). Am J Cardiol. 2011;Epub ahead of print.

 

 

Related Stories:

Disclosures
  • The study contains no statement regarding conflicts of interest.
  • Drs. Brener and Krumholz report no relevant conflicts of interest.

We Recommend

Comments