November 24th, 2009
Expand or Restrict Primary PCI?
L. David Hillis, MD
At last week’s AHA meeting, a late breaking clinical trial reported that STEMI patients in Massachusetts who underwent primary PCI at hospitals without cardiac surgery on site had similar rates of death, MI, and target vessel revascularization at 1 year as those who underwent primary PCI at hospitals with cardiac surgery on site. In contrast, the recently released AHA/ACC updated STEMI guidelines make the development of regional systems of STEMI care “a matter of utmost importance.”
Should primary PCI be limited to regional centers, or should it continue to be performed at hospitals without cardiac surgery on site to maximize access to it?
7 Responses to “Expand or Restrict Primary PCI?”
NEJM Journal Watch — Recent Cardiology Articles- To Lower Stroke Risk in Patients Undergoing Atrial Fibrillation Ablation, Is a Device or a Medication Better?
- Intensive vs. Standard Blood Pressure Lowering in Patients with Diabetes
- Anticoagulation in Patients with Cancer-Associated Pulmonary Embolism: How Long is Long Enough?
- Spironolactone Use After Percutaneous Intervention for Acute Myocardial Infarction
- CRISPR/Cas9 Gene Editing for Transthyretin Cardiomyopathy
Expand or Restrict Primary PCI
During the past year several studies have been reported addressing the safety of PCI without on-site cardiac surgery. These have used data from the NCDR CathPCI Registry, the NRMI database and the NY State PCI database. None of these reports have shown a difference in mortality or the need for urgent surgery among facitities with and without on-site surgery. Interventional cardiologists abroad have been performing both urgent and elective PCIs without on-site surgery for years with excellent results. The British Cardiovascular Intervention Society (http://www.bcis.org.uk/) collects data on all PCIs in the United Kingdom and posts their audit results online annually. About half of all PCI centers in the UK operate without on-site surgery with excellent results. Can primary and elective PCI be performed safely without on-site surgery? I think we have the answer and it is – – YES – – – providing the structure, process and quality recommendations which have been published for such facilities are followed.
However, just because it is safe to do PCI without on-site backup does not justify a PCI centers on every street corner in the US. I think the guideline writers have the right focus by emphasizing the development of regional systems. Of course the “devil is in the details” when it comes to defining exactly what is meant by a “regional center” The whole issue of PCI without on-site surgery in the US was born from a desire to provide the best therapy for STEMI to patients in rural areas who could not receive PCI in a timely fashion by the necessary transfer. Since one cannot sustain a PCI a program on STEMI cases alone, this served as justification for allowing elective PCIs at such facilities. PCI programs were (emphasis on the past tense) profitable for hospitals and unfortunately, this may have been an additional motive driving some expansion. Changes in reimbursement for PCI will put a financial strain on many lower volume PCI programs and they may well close; probably this is a good thing. If the geography or other factors dictate the need for a regional PCI center to deliver timely STEMI care and that facility needs to operate without on-site surgery – OK by me. For the future, we need to focus on developing the most efficient, patient-centered and highest quality system to deliver primary and elective PCI to the appropriate patients.
Competing Interests: For the record, I perform PCIs at facilities with and without on-site cardiac surgery.
Mortality Same at Low and High Volume Primary PCI Sites
Today’s JAMA reports that in-hospital mortality following primary PCI is similar (<4%) at low and high volume centers (< 36 and >70 primary PCI procedures annually, respectively). This study flies in the face of what has previously been reported. Has primary PCI (e.g., stenting) and adjunctive therapy improved to the point that volume is no longer important? Should we abandon volume criteria as a surrogate for quality of care and focus on outcome instead? This would certainly allow us to make “quality” primary PCI more accessible.
Because of the widespread use of stents and more effective anti-platelet therapy, the chance of requiring an emergent trip to the operating room for a failed primary PCI procedure is, thankfully, extremely low. In the uncommon scenario of a patient presenting with STEMI and a concomitant mechanical complication of MI, however, emergent surgery would be needed. If a PCI hospital with surgical back-up also has long door-to-balloon times, then really what good is the surgical back-up anyway? However, hospitals performing PCI without onsite cardiac surgery need to address additional quality of care issues that surgical centers do not such as inservices for the CCU staff to prepare for post primary PCI patients who may be hemodynamically unstable post-procedure, and for the management of intra-aortic balloon pumps. If a hospital without surgical back-up can achieve the benchmarks regarding time of first ECG, activation of the cath lab, door-to-balloon time, appropriate post-cath CCU care, and, importantly, a fail-safe plan to emergently transfer the patient to an operating room at a surgical center if needed, then I think it is reasonable and would provide primary PCI to more patients. If, however, those benchmarks cannot be achieved, I would not support the performance of primary PCI in that hospital. So, in my opinion, it depends on the merits of the individual PCI program, and should be assessed on a hospital by hospital basis.
Lower volume PCI centers
The debate about a volume-outcome relationship is interesting. Is there a cut-off? If a doc does 100 PCIs per year, would they be just as good if they did only 99 or 98 or . . . . . you get the idea. Although we may not know what the absolute low end cut-off number should be, would you want your doc to have only done one PCI in the last year. I doubt it. Interventional cardiology changes rapidly and when I hit a spell that I am not in the lab, I know I feel a bit rusty. I’ll ask for a 24 mm stent and the nurse says, but that stent only comes in a 23 mm length. True, no big deal over 1 mm, but at some point the low volume effect will start to erode your overall clinical sharpness in the lab. The opposite is also true. If you do 1000 PCIs without a single complication, but 950 of them were on 60% stenoses that should have been treated medically, you cannot claim good quality. Quality should trump volume any day, but it becomes very hard to assess quality, appropriateness and complication rates at low volumes.
What Volume
I have previously been a propenent of centers doing a minimum number of primary PCIs annually, but I was never convinced that 75 /yr was the “magic number”. I must admit, though, that I was not prepared for the JAMA article which showed similar mortality results for centers doing < 36 and those doing >70 primary PCI procedures annually. So what is the minimum number or primary PCIs that we should expect a center/hospital to perform annually?
As I noninterventionalist, I wonder how much this is due to the improvements and simplification of the procedures over time. In the past operators had to be experienced and seemed to need some level of natural talent (our version of athletic ability) to work with the equipment available. In my outsider’s view experience, and my view as a program director, it seems that talent and experience matter less than they used to. In fact, I’m often quite surprised at which fellows choose interventional cardiology and how little that correlates with natural aptitude with regard to procedural skills! It’s also amazing to me how quickly they can learn how to do straightforward PCI. At the end of the day, most of them seem to end up doing fine procedurally, which argues that the technical advances have narrowed the difference between the more and less talented operators, and probably has the same effect with regard to procedural experience. What do you all with more gray hair than me think about the simplification of native vessel PCI over time and its effect on the volume relationships?
“Rescue PCI”: Rescuing the Operator
James: I think you’re right. Improvements in radiographic and interventional equipment have a played a huge role in making PCI more accessible and safer…often trumping the modest skills of many interventionalists. In my opinion, the major benefit of stenting is not its lower rate of restenosis compared to POBA, but the fact that few patients need emergent CABG because of acute coronary dissections following stenting; with POBA severe coronary dissections occured in ~5% of patients. “Rescue stenting” is rarely done to salvage the patient…it usually rescues the interventionalist.