To PRAMI or not to PRAMI: Should we do PCI of the Non-Infarct-Related Artery in STEMI Patients?

The problem

We activated the cath lab for an inferior ST-elevation myocardial infarction (STEMI) in a 45-year-old woman 2 hours after onset of severe substernal chest pain. The electrocardiogram (ECG) showed ST elevations in leads 2, 3, and AVF (Figure 1). Coronary angiography demonstrated a patent left anterior descending coronary artery (LAD) with luminal irregularities and a totally occluded circumflex artery (Figure 2). The right coronary artery (RCA) had 75-80% proximal lesion with additional luminal irregularities (Figure 3). The culprit circumflex artery was stented with a 3.0 x 23 mm drug-eluting stent (DES) with no residual narrowing and TIMI-3 flow (Figure 4). Although we had a widely patent circumflex (CFX) vessel with good flow in a pain-free patient, there was persistent ST segment elevation. Left ventriculography was then performed. The left ventricular end diastolic pressure (LVEDP) was 12 mmHg. The LV gram showed normal LV wall motion in all segments.

My cardiology fellow said we should now stent the RCA since we were already here and the patient was quite stable. I demurred and said we will be back in a couple weeks unless the patient develops angina. The RCA could be treated after recovery from the acute event.

Given this scenario and the recent publication of the Preventive Angioplasty in Acute Myocardial Infarction by Wald et al¹ suggesting benefit to treating both the culprit and other stenosed arteries, should we now treat the non-infarct related artery (IRA) when finished with the culprit IRA of the STEMI?  

I think we could have stented the RCA without a problem and the patient would have done well.  Nonetheless, I elected not to do the RCA because 1) current recommendations of guidelines, irrespective of the one recent study to the contrary, and 2) my deep suspicion of the reperfusion injury pattern on ECG with normal wall motion, despite the clinical stability of the patient.  

What do the percutaneous coronary intervention (PCI) guidelines tell us?

The guidelines² tell us that we should not do (a class III recommendation) PCI of the non-IRA following the culprit IRA PCI, except when there is hemodynamic compromise (Table 1). This recommendation is supported by 3 studies^3,4,5 (Table 2) and suggests higher mortality for treating both culprit and non-culprit STEMI vessels, but that staging and treating the non-IRA artery < 60 days after the index procedure is associated with lowest mortality in these patients.  We should recall that the supporting data come from retrospective studies and meta-analysis of non-randomized and observational studies.

What did the PRAMI study tell us?

The PRAMI study tested whether treating both the non-IRA and IRA during STEMI was better than PCI of the culprit-only approach. About 460 patients were randomized into 2 groups, each to one of the 2 strategies and after 2 years, the preventative PCI group, (that is, all ‘significant’ lesions were treated during STEMI) had lower death and lower major adverse cardiac event rates. This was a unique, prospective, randomized, albeit small, trial suggesting a change in the current STEMI PCI approach. PRAMI has been criticized in several ways, including concerns about the small sample size and premature cessation of the study (thus, the end points appear more significant that they might be with larger, longer follow-up). Despite these concerns, this study will certainly lead to a larger, prospective, multicenter study to challenge our current STEMI practice. 

What did my colleagues say about this case?

Peter Ver Lee, Maine Medical Center:  I would let her recover from the [acute] MI and bring her back in 48 hours and do the RCA if you think it is > 90%. If it is < 90%, I would use fractional flow reserve (FFR) to guide revascularization. If she has more pain before 48 hours, I would stent the RCA. But I [also] can see bringing her back in a couple weeks, too. I think that once a patient has had an acute coronary syndrome (ACS), they deserve as complete a revascularization as we can give them. FAME 2 showed the superb outcomes we can achieve with stenting severe lesions. The stented arm essentially reverts to the “background noise” of the Cohort B group [the non-ischemic coronary artery disease group treated medically] in terms of MACE [major adverse cardiac events].

Mike Kutcher, Duke University: The RCA could be considered a “co-infarct” vessel. The PRAMI study can be used to support a judgment call to intervene on this or on a “non-infarct” vessel, if the setting is right. If your contrast load and radiation dose was not too high and vascular access and guide position good, you could justify intervening on the RCA. On the other hand, if circumstances were not favorable, the operator could use “judgment” to not intervene on another vessel and cite the current guidelines. The PRAMI study gives some defense, as Teddy Roosevelt would say, to the “man (or woman) in the arena.” 

John Bittl, Ocala, Florida:  Guidelines are not commandments and cannot govern the treatment of every patient in every clinical situation. In formulating the PCI guideline, it was clear that almost all the evidence against performing multivessel PCI at the same time as culprit PCI in the acute setting came from observational studies. Some evidence in favor of multivessel PCI came from small studies. In 2011, some of us on the PCI writing committee lobbied against the Class III (Harm) recommendation and were in favor of a more discretionary classification, similar to that used by the European Society of Cardiology: “With the exception of cardiogenic shock, PCI for STEMI should be limited to the culprit stenosis (Class IIa, Level of Evidence: B).” Unlike a Class III recommendation, the IIa recommendation does not condemn multivessel PCI and in practice, gives the operator the option to exercise the judgment to proceed with multivessel PCI in some STEMI patients who have multiple complex lesions, and do not show clinical or electrocardiographic improvement after culprit PCI.

Gregg Stone, Columbia University, New York City: The first decision point (in the non-shock patient) is whether there are ongoing symptoms (or ischemia) arising from the second lesion. This would certainly warrant treatment. In this case, the continued ST elevation is likely due to distal embolization, vasoconstriction, and/or capillary destruction, and not subtotal flow from the RCA (especially with normal wall motion in the inferior wall). The second reason to possibly treat [the non-IRA] in the acute phase would be if the lesion appears unstable (ruptured plaque, thrombus, 95% severe, TIMI flow <3). This does not appear to be the case here (given the limitations of the single view we have been shown). The third reason would be to prevent future events (and perhaps reduce resource consumption and costs for separate caths/admissions).

Unfortunately, in PRAMI, the types of lesions randomized were not described, and there was no core lab [analysis]. Were these non-culprit lesions ruptured 90% stenoses, or stable-appearing 60% lesions? Potential reasons not to routinely treat acutely are complications should the second vessel go down or embolize (especially if abnormal wall motion in the infarct zone has not had a chance to improve, or at least “firm up”); ADP antagonists not yet being absorbed (this concern to be mitigated by cangrelor in the future), increasing procedural risk in the setting of platelet activation with ACS; systemic vasoconstriction in STEMI that can make non-culprit lesions appear more severe than they are; and lack of validation of FFR in non-culprit vessels in the STEMI setting to help assess these lesions. 

Thus, while I am very intrigued by PRAMI, I don’t believe we should change practice on the basis of one modest-sized study with potential selection bias (465 patients enrolled over 5 years), not understanding the lesion characteristics that were randomized. I do believe, however, that based on this study, the guidelines should go from class III to IIb. In this case, I would have done as you did, and deferred, probably for a stress test in a 1-2 months (absent spontaneous symptoms), or if you really believed it was a 90% stenosis, cath/PCI in 2-4 weeks.

Lloyd W. Klein, Chicago: PRAMI is a very intriguing study and multivessel PCI at the time of STEMI can be useful in some circumstances, mostly in ischemia at a distance and when there are additional culprit lesions. The guidelines were always too tightly worded for these circumstances, and perhaps that is going to be on the table now for discussion. The concern that a “cowboy” can use the data to justify unnecessary and possibly dangerous procedures is one that we as a group will have to try to “corral.”

I did a patient last week with an acute inferior MI who also had what seemed to be a tight 90% LAD lesion. After a similar discussion to yours, I brought the patient back after 48 hours to do that stenosis, and amazingly, it was now like 60% and FFR was 0.92. Of course, there are cases where FFR after STEMI can be fallaciously normal; but I can’t explain the improved angiographic appearance. In your case, the RCA is not the culprit, and it is significant. I would not stent it at the acute setting, but I would in the next few weeks.

Sam Butman, Arizona:  I wonder what would have happened if you did do the RCA at the same sitting and there was some no reflow (bradycardia, hypotension, and operator diaphoresis). Do you think you would be more prone to wait for more definitive trials that might separate which patients should and which should not undergo ad hoc acute multivessel PCI?  For what it is worth, I would have done as you did, i.e., wait and reassess in hospital or post hospitalization.

Mort Kern, Long Beach: I rarely interrupt the flow of chatter, but wanted to put 2 more thoughts into the pot about a couple of issues that seem to be recurrent and problematic. 

1. In many patients, PCI of the non-IRA ‘severe’ stenosis may be as early as 2 days. The rationale for this approach in an asymptomatic patient does not exist (outside of PRAMI). Those who advocate early return, where is justification? Why not just wait a couple weeks to let the patient stabilize, allow for return of renal function, and institute medical therapy for ischemia?
2. Lloyd [Klein] told us about a 90% non-IRA artery stenosis that was 60% a few days later and had negative FFR [i.e., non-ischemic]. In the acute setting with high catecholamines, etc., vasomotor tone could be increased to produce this result. To his credit, the oculo-stenotic reflex was overcome with good sense, and the lesion was objectively non-ischemic by FFR. This leads us to the PRAMI patients, with some patients having stents in non-IRA that may be non-significant lesions and hence good outcomes by stenting (mostly). [Stenting of non-significant lesions is wrong, but it doesn’t always show up in the outcome studies, except in the FAME design, because of the use of FFR. STEMI patients with and without non-culprit stents might have the same outcome in PRAMI if the lesions were unimportant, but got stented anyway.] 

Peter Ver Lee, Maine Medical Center:  The decision to treat is one of 4 choices: 1) do it [PCI of the non-IRA] right away (Class III), 2) do it before they leave hospital, usually in 24-48 hours [early staged PCI], 3) send home and bring back in 2 weeks to have it done [late staged PCI], or 4) don’t do it at all unless they have symptoms or positive stress test. 

I think of a STEMI with a second high grade stenosis as a higher risk patient than the guy with only a single vessel STEMI or two high-grade stenoses without a STEMI. But isn’t it appropriate to stent two severe lesions with an intermediate to high risk stress test? You wouldn’t just stent one lesion and see how they do with medical therapy. So if the RCA has an FFR of 0.75 or lower, and the patient is going to be on clopidogrel anyway, why not send them home completely revascularized? The dust has usually settled by 48 hours and I think an FFR-guided revascularization strategy for a proximal vessel is appropriate. I wouldn’t do it at the same time for reasons others have mentioned, and PRAMI is only one study. But just because a patient’s STEMI symptoms have improved after stenting one vessel, doesn’t mean they are “asymptomatic.” They are lying in a hospital bed, doing nothing. How does anyone know if they are “asymptomatic”? 

Habib Samady, Emory University: As I see it, the 3 obvious yet fundamental questions: 

1. What is the natural history of 50-99% non-culprit lesions in STEMI patients (who have already declared themselves as “high systemic risk”)?  How often will these lesions subsequently cause death or MI on optimal medical therapy (OMT)?  
2. What is the risk of PCI at the time of STEMI, in the very early post PCI period, compared to, say, 6-12 weeks later when the inflammation has settled, the myocardium has recovered, and risk factor modification and OMT initiated?
3. How do we identify those 50-99% non-culprit lesions in which the risk benefit ratio of PCI in addition to OMT is favorable and defines the optimal timing for such an intervention? If we can answer these questions, it could help inform us of the best care of these high-risk patients.

Justin Davies, Imperial College London, United Kingdom:  To me, one of the biggest omissions from PRAMI, even in the appendix supplement, was the distribution of stenosis severities the investigators left untreated. This key question, which hopefully will be addressed, is, did the events all occur in 1) 95% proximal LAD lesions, or in 2) 60% mid RCA lesions?  If it is found to be the former, leaving such patients untreated significantly deviates from routine clinical practice and undermines the value of the study. If it is found to be the latter, it raises interesting questions about the way we practice and guide interventional revascularization as a whole.  

The bottom line

Table 3 provides my summary of the differences between the PCI approach to STEMI as suggested by guidelines and PRAMI. Being a conservative interventionalist (is this an oxymoron?), I decided to do culprit-only STEMI and continue with optimal medical therapy followed by reevaluation in 4 weeks or sooner if symptoms appear. Of course, for STEMI patients in shock, we would have done PCI of all significant lesions, as appropriate and recommended. For the time being, we will be following the PRAMI story and applying best clinical judgment for the particular patient at the time of their presentation. We are sure to hear more about this issue in the coming year.

References

1. Wald DS, Morris JK, Wald NJ, Chase AJ, Edwards RJ, Hughes LO, Berry C, Oldroyd KG; PRAMI Investigators. Randomized trial of preventive angioplasty in myocardial infarction. N Engl J Med. 2013 Sep 19;369(12):1115-23. doi: 10.1056/NEJMoa1305520.
2. Levine GN, Bates ER, Blankenship JC, Bailey SR, Bittl JA, Cercek B, Chambers CE, Ellis SG, Guyton RA, Hollenberg SM, Khot UN, Lange RA, Mauri L, Mehran R, Moussa ID, Mukherjee D, Nallamothu BK, Ting HH; American College of Cardiology Foundation; American Heart Association Task Force on Practice Guidelines; Society for Cardiovascular Angiography and Interventions. 2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention. A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Society for Cardiovascular Angiography and Interventions. J Am Coll Cardiol. 2011 Dec 6;58(24):e44-122. doi: 10.1016/j.jacc.2011.08.007.
3. Hannan EL, Samadashvili Z, Walford G, Holmes DR Jr, Jacobs AK, Stamato NJ, Venditti FJ, Sharma S, King SB 3rd. Culprit vessel percutaneous coronary intervention versus multivessel and staged percutaneous coronary intervention for ST-segment elevation myocardial infarction patients with multivessel disease. JACC Cardiovasc Interv. 2010 Jan; 3(1): 22-31. doi: 10.1016/j.jcin.2009.10.017.
4. Toma M, Buller CE, Westerhout CM, Fu Y, O’Neill WW, Holmes DR Jr, Hamm CW, Granger CB, Armstrong PW; APEX-AMI Investigators. Non-culprit coronary artery percutaneous coronary intervention during acute ST-segment elevation myocardial infarction: insights from the APEX-AMI trial. Eur Heart J. 2010 Jul;31(14):1701-7. doi: 10.1093/eurheartj/ehq129.
5. Vlaar PJ, Mahmoud KD, Holmes DR Jr, van Valkenhoef G, Hillege HL, van der Horst IC, Zijlstra F, de Smet BJ. Culprit vessel only versus multivessel and staged percutaneous coronary intervention for multivessel disease in patients presenting with ST-segment elevation myocardial infarction: a pairwise and network meta-analysis. J Am Coll Cardiol. 2011 Aug 9;58(7):692-703. doi: 10.1016/j.jacc.2011.03.046.