Impact of Surgical Consultation on Outcomes in Hemodynamically Supported High-Risk Percutaneous Coronary Intervention: Insights From PROTECT II Randomized Study

Abstract: Background. In observational studies of patients undergoing percutaneous coronary intervention (PCI), surgical ineligibility is associated with increased mortality. Whether the use of hemodynamic support during PCI can mitigate the adverse prognostic importance of surgical ineligibility is unknown. Methods and Results. We sought to evaluate the association between request for surgical consultation (presumed surgical ineligibility) prior to PCI and clinical outcomes in 427 patients with multivessel coronary artery disease or unprotected left main disease and severely reduced left ventricular systolic function undergoing PCI assisted by hemodynamic support (intraaortic balloon pump or Impella) from the PROTECT II randomized trial. Patients in whom surgical consultation was requested prior to PCI (n = 201) were compared with those in whom surgical consultation was not requested (n = 226). The primary endpoint of this analysis was the composite of 90-day major adverse cardiac and cerebrovascular events (MACCE). Demographic and procedural variables were similar between patients receiving surgical consultation and patients not receiving surgical consultation, with the exception that the prevalence of prior coronary artery bypass graft surgery was significantly higher in patients not receiving surgical consultation (42.0% vs 25.4%; P<.001); these patients also had a higher proportion of lesions within a saphenous vein graft, and a greater prevalence of moderate/severe vessel calcification. MACCE rate at 90 days was similar in patients receiving surgical consultation compared with patients not receiving surgical consultation (23.4% vs 29.0%, respectively; P=.19). Conclusions. In this high-risk cohort of patients undergoing hemodynamically supported PCI, clinical outcome was not associated with an antecedent request for surgical consultation (presumed surgical ineligibility). Whether the use of hemodynamically supported PCI can lessen the risk conferred by surgical ineligibility requires further study.

J INVASIVE CARDIOL 2016;28(5):187-192. Epub 2016 February 15. 

Key words: high-risk PCI, Impella device, intraaortic balloon pump

_________________________________________________

Many clinical, anatomical, and procedural factors are associated with an increased risk of adverse events in patients undergoing PCI (percutaneous coronary intervention).¹ Advanced age, reduced left ventricular function, prior myocardial infarction (MI), cardiogenic shock, and diabetes mellitus are clinical factors that have been shown to contribute to poor outcomes.^2-4 Anatomical factors, including multivessel coronary artery disease, left main stenosis, and high SYNTAX score, also contribute to adverse outcomes during PCI.^5-7 The occurrence of no-reflow, intraprocedural stent thrombosis, and Thrombolysis in Myocardial Infarction (TIMI) flow <3 are well described procedure-related factors that are associated with increased risk.^8,9 More recently, surgical ineligibility has also been found to identify patients who are at a higher risk for PCI.^10,11 However, many of the aforementioned studies evaluating these factors were performed in relatively low-risk patient groups. 

PROTECT II was a trial of patients undergoing high-risk PCI who were randomized to receive support with either intraaortic balloon counterpulsation (IABP) or the Impella device (Abiomed, Inc).¹² Patients were required to have severe left ventricular dysfunction and complex coronary anatomy including unprotected left main disease, three-vessel disease, and/or sole circulation lesions as targets for revascularization. PROTECT II thus represents a truly high-risk and contemporary group of patients undergoing PCI, which yields the opportunity to explore concepts of PCI risk.

In this study, we sought to evaluate the concept of surgical ineligibility on outcomes when applied to a truly high-risk PCI group. We examined the impact of surgical consultation (and presumed surgical ineligibility) on the outcomes of the patients enrolled in PROTECT II.

Methods

Patient population. Patients enrolled in the prospective, randomized PROTECT II trial were included.¹² Briefly, PROTECT II was conducted at 112 clinical sites in the United States, Canada, and Europe and included 452 enrolled patients. All clinical sites had experience with high-risk PCI and hemodynamic support. The major inclusion criteria included age ≥18 years, scheduled to undergo non-emergent PCI on an unprotected left main or last patent coronary vessel with a left ventricular ejection fraction (LVEF) ≤35%, or patients with three-vessel disease and LVEF ≤30%. Patients with recent MI, persistently elevated cardiac biomarkers, left ventricular thrombus, platelet count ≤75, 000/mm³, creatinine ≥4 mg/dL, and severe peripheral vascular disease that precluded passage of the Impella device or IABP were excluded. Additional inclusion and exclusion criteria are described by O’Neill et al.¹² Patients were randomized in a 1:1 fashion through an automated interactive voice response system to the Impella 2.5 device (n = 226) or a commercially available IABP (n = 226). 

Notably, the study protocol encouraged all investigators to obtain a surgical consultation on patients considered eligible for study participation. Of note, 9.9% of the patients assessed for eligibility and not enrolled in the PROTECT II study underwent coronary artery bypass graft (CABG) surgery. Investigators were required to document in the case report form if a surgical consultation was considered, requested, and/or performed, as well as the reasons behind this decision. For patients receiving surgical consultation, reasons for not performing CABG were classified as “subject refused CABG surgery” or “subject was not a candidate for CABG based upon coexisting medical conditions.” For patients not receiving surgical consultation, it was presumed that the treating cardiologist deemed their operative risk to be excessive and therefore elected to forego surgical consultation or patient refusal for CABG surgery. All patients underwent objective risk scoring, including additive European System for Cardiac Operative Risk Evaluation score (EuroSCORE), logistic EuroSCORE, Society of Thoracic Surgeons (STS) mortality score, Mayo PCI score, New York PCI score, and the Synergy between PCI with Taxus and cardiac surgery (SYNTAX) score.^13-16 

All patients provided written informed consent and the study was approved by the local investigational review board at each clinical site. The study was conducted in accordance with the Declaration of Helsinki.

Device description and study procedures. The Impella 2.5 device is a 12 Fr, axial flow, rotary blood pump that is placed percutaneously via the common femoral artery and advanced retrogradely across the aortic valve into the left ventricle. For patients randomized to an IABP, a commercially available device was placed percutaneously via the common femoral artery. Complete revascularization during a single procedure was recommended per study protocol. Choice of anticoagulant (heparin, bivalirudin), antiplatelet therapy, glycoprotein IIb/IIIa inhibitor, type of coronary stent (bare metal, drug eluting), and additional treatment devices including rotational atherectomy and distal protection devices were left to the discretion of the treating physician. Upon the completion of the index PCI procedure, patients were weaned from hemodynamic support. For patients who could not be weaned from hemodynamic support following completion of the index PCI, the duration of support was recorded. Hemostasis was obtained by manual hemostasis or use of the preclose technique. 

Endpoints. The primary endpoint for this analysis was the composite major cardiac and cerebrovascular event (MACCE) rate at 90 days. The composite primary endpoint included the components of all-cause death, Q-wave or non-Q wave MI, stroke, transient ischemic attack (TIA), and any repeat revascularization procedure (PCI or CABG surgery). Non-Q wave MI was defined as CPK-MB elevation ≥8x upper limit of normal within 72 hours for periprocedural MI and ≥2x upper limit of normal for spontaneous. The definitions of all primary endpoint components and composites endpoints were described elsewhere.^12,17 

Data management. An independent academic clinical research organization (Harvard Clinical Research Institute, Boston, Massachusetts) provided data collection, management, and monitoring, as well as event adjudication. Angiographic analysis was provided by an independent core laboratory (Beth Israel Deaconess, Boston, Massachusetts). An independent Clinical Events Committee adjudicated all MACCE and clinical endpoints and an independent data safety monitoring board evaluated safety endpoints. 

Statistical analyses. For the current analyses, the per-protocol definition for study inclusion was applied, which included 427 patients. The per-protocol definition included all consenting, randomly assigned patients who met all protocol eligibility criteria and excluded 21 patients enrolled in the intention-to-treat cohort because of the following reasons: 2 patients withdrew consent; 4 patients had LVEF ≥35%; 4 patients did not qualify based upon coronary anatomy; 3 patients had recent MI; 1 patient had severe peripheral vascular disease; 2 patients had severe peripheral vascular disease or aortic stenosis; 2 patients had platelet count <70,000; 1 patient had creatinine >4 mg/dL; 1 patient was lost to follow-up after hospital discharge; and 1 patient had PCI performed following study termination. Categorical variables are presented as numbers and percentages. Continuous variables are presented as mean ± standard deviation. Student’s t-test was used to compare categorical variables and chi-square or Fisher’s exact test was used to compare continuous variables. Kaplan-Meier estimates of the cumulative incidence of MACCE through 90 days was constructed and a log-rank test was used to compare the curves between the surgical consultation requested and no surgical consultation requested groups. A P-value <.05 was considered statistically significant. 

Results

Baseline characteristics. Surgical consultation was requested in 201 patients (47%). The main reasons these patients were not felt to be candidates for surgical bypass were associated medical conditions (83%) and patient refusal (17%). Baseline demographics, including age, gender, prior MI, history of angina, presence of congestive heart failure, New York Heart Association (NYHA) class, cardiomyopathy, prior PCI, peripheral vascular disease, diabetes mellitus, hypertension, chronic obstructive pulmonary disease, renal insufficiency, and hyperlipidemia, were similar in both groups (Table 1). Prior PCI and CABG surgery were more prevalent in patients not receiving surgical consultation compared with patients receiving surgical consultation (44.0% vs 34.5% [P=.046] and 42.0% vs 25.4% [P<.001], respectively). History of prior stroke was more prevalent in patients receiving surgical consultation, compared with patients not receiving surgical consultation (18.9% vs 10.6%, respectively; P=.02). Insulin-requiring diabetes mellitus was more prevalent in patients receiving surgical consultation compared with patients not receiving surgical consultation (55.8% vs 38.1%, respectively; P=.01). 

LVEF was severely reduced in all patients (23.7 ± 6.3%) and was similar between both groups. For the entire cohort, the logistic EuroSCORE and STS score were 18.6 ± 17.6% and 6.0 ± 6.6%, respectively, and were similar between patients receiving surgical consultation and patients not receiving surgical consultation. The Mayo PCI score was similar between patients receiving surgical consultation and patients not receiving surgical consultation (9.0 ± 3.4% and 8.4 ± 3.5%, respectively; P=.13). The New York PCI score was also similar between patients receiving surgical consultation and patients not receiving surgical consultation, (10.8 ± 3.3% and 11.0±3.4%, respectively; P=.13). There were no differences regarding the year of study enrollment between patients receiving surgical consultation and patients not receiving surgical consultation.

Baseline angiographic findings. Vessel location for PCI was not associated with surgical consultation, with the exception of lesions located within a saphenous vein graft (Table 2). PCI on a saphenous vein graft lesion occurred in 4.0% of patients receiving surgical consultation vs 7.6% of patients not receiving surgical consultation (P=.01). Lesion characteristics, including lesion length, presence of thrombus, presence of a total occlusion, bifurcation lesion, or American College of Cardiology/American Heart Association lesion classification C, were not associated with receiving surgical consultation. The presence of moderate/severe lesion calcification was more prevalent in patients receiving surgical consultation compared with patients not receiving surgical consultation (42.2% vs 34.1%, respectively; P=.01). Duration of the index PCI procedure was significantly longer in patients receiving surgical consultation compared with patients not receiving surgical consultation (1.14 ± 0.76 hours vs 0.95 ± 0.59 hours, respectively; P=.01). The number of patients receiving device support for ≥3 hours was significantly higher in patients receiving surgical consultation compared with patients not receiving surgical consultation (21.1% vs 12.6%, respectively; P=.02).

Clinical events. The composite endpoint of MACCE at 30 days was similar in patients receiving surgical consultation compared with patients not receiving surgical consultation (23.4% vs 21.7%, respectively; P=.67) (Table 3 and Figure 1). Individual major components of MACCE, including death, MI, stroke/TIA, and repeat revascularization, were also similar between the groups. The composite endpoint of MACCE at 90 days was similar in patients receiving surgical consultation compared with patients not receiving surgical consultation (23.4% vs 29.0%, respectively; P=.19) (Table 3 and Figure 1). Individual major components of MACCE, including death, myocardial infarction, stroke/TIA, and repeat revascularization, were also similar between the two groups. 

Discussion

In the current study, we evaluated the association between request for surgical consultation and clinical outcome in patients undergoing hemodynamically supported PCI in the PROTECT II trial. We found some baseline demographic, angiographic, and procedural differences between patients receiving surgical consultation compared with patients not receiving surgical consultation. However, clinical outcome at 90 days was similar, suggesting that request for surgical consultation was not associated with worse outcome in this high-risk PCI cohort. Our analysis was applied to patients receiving per-protocol treatment, although similar results were obtained in the intention-to-treat cohort.  

Several prior studies have evaluated the impact of surgical consultation and/or surgical eligibility on patients with complex coronary artery disease undergoing revascularization.^10,11 Waldo et al evaluated 1013 patients with unprotected left main or multivessel coronary artery disease undergoing revascularization.¹¹ Twenty-two percent of patients were deemed ineligible for bypass surgery. These patients had a mean age of 72 ± 12 years, only 14% had “severe systolic dysfunction” (LVEF not reported), 27% had cerebrovascular disease, and 41% had three-vessel disease. The most common reasons for surgical ineligibility were poor surgical targets, advanced age, and renal insufficiency. Surgical ineligibility was found to be a potent independent predictor of both in-hospital and long-term mortality with PCI, and the authors thus advocated adopting surgical ineligibility as a risk factor for PCI patients. 

McNulty et al evaluated 101 consecutive patients with unprotected left main disease undergoing non-emergent PCI.¹⁰ Fifty-four percent of patients were classified as surgically ineligible, with the most common reasons cited as advanced age, severe lung disease, poor surgical targets, and severe systolic dysfunction. Surgical ineligibility was independently associated with worse clinical outcome over a median follow-up of 18 months.  

PROTECT II represents a strictly high-risk and contemporary cohort of patients undergoing complex PCI. Advanced age, severely reduced left ventricular systolic function, and a high prevalence of insulin-requiring diabetes mellitus, renal insufficiency, and prior bypass surgery characterize this patient cohort. The PCI registry of the SYNTAX trial includes patients with unprotected left main or multivessel disease who were excluded from randomization in SYNTAX because they were not considered to be candidates for bypass surgery.¹⁸ These patients all received PCI and outcomes were closely assessed, thus providing a contemporary study for comparison. Objective risk scores, including the additive and logistic EuroSCORE, the prevalence of prior CABG, and the number of patients with an LVEF <30%, were all significantly higher in PROTECT II. These observations suggest that patients enrolled in PROTECT II were at higher risk for PCI and therefore assertions about risk factors from the SYNTAX PCI registry may not apply to higher-risk groups. 

The decision to request surgical consultation is often complex and multifactorial and may not be solely based upon medical evidence. In prior clinical studies, refusal for surgery was thought to be a high-risk marker because the comparator group was often patients who were eligible for surgery.^10,11 In contrast, in the PROTECT II cohort, most patients were felt by the treating cardiologist to not be a candidate for surgery a priori to receiving formal surgical consultation. As such, one could argue that the comparator group of those who refused surgery may, in some respects, have an overall lower risk. In a recent study by Hannan et al, among patients with indications for either PCI or CABG surgery, only 5% were recommended by their treating cardiologist for CABG surgery.¹⁹ Geographic and interhospital variation also plays a role in the mode of revascularization.^20,21 Baig et al evaluated practice patterns in England and found significant geographic variation throughout the country that was not explained by procedural volume.²² Patient preference also plays a key factor when considering CABG surgery.²³ Use of the heart team for decision-making has been suggested as one method to reduce this variation and provide clear and consistent recommendations for care.²⁴ Although PROTECT II was designed and initiated prior to the concept of the heart team approach, we feel the current findings of the lack of an association between request for surgical consultation and clinical outcome are relevant to contemporary practice. 

Study limitations. There are several limitations to the present analysis. We were unable to account for confounders not available within the PROTECT II database. Surgical consultation was recommended per study protocol, but was not mandated. At the time of study enrollment, the concept of the heart team approach had not been introduced into clinical practice. We presume that enrollment in the study with subsequent surgical consultation implied refusal for bypass surgery. Patients in whom surgical consultation was not requested were likely deemed by the treating cardiologist to be too ill to be considered for CABG surgery or otherwise inappropriate for surgery, such as patients with established care directives and those refusing surgery. Specific reasons for not obtaining surgical consultation were not collected. The current analysis represents a subgroup analysis from the per-protocol study cohort. The results regarding the association between surgical consultation and clinical outcome should therefore be interpreted in this regard and considered hypothesis generating. Similar results were seen in the intention-to-treat cohort and in those receiving either IABP or Impella device support (data not shown).

Conclusion

In summary, patients enrolled in PROTECT II who had surgical consultation requested appeared to have similar baseline risk compared with those in whom surgical consultation was not requested. Clinical outcomes of both groups were similar. Although surgical ineligibility had previously been demonstrated to be associated with adverse PCI outcomes, these findings were in studies of patients at lower risk than the PROTECT II cohort. Our findings suggest that in a truly high-risk patient cohort receiving hemodynamically supported PCI, surgical ineligibility may not be a significant risk factor for poor outcome.  

Acknowledgments. The names of the investigators, institutions, and research organizations participating in the PROTECT II trial were previously reported.¹² 

References

1.    Block PC, Peterson ED, Krone R, et al. Identification of variables needed to risk adjust outcomes of coronary interventions: evidence-based guidelines for efficient data collection. J Am Coll Cardiol. 1998;32:275-282.

2.    Mazhari R, Kapur N. Increased risk and increased reward in coronary intervention in older patients with acute coronary syndrome. Heart. 2014;100:1483-1484.

3.    Brennan JM, Curtis JP, Dai D, et al. Enhanced mortality risk prediction with a focus on high-risk percutaneous coronary intervention: results from 1,208,137 procedures in the NCDR (National Cardiovascular Data Registry). JACC Cardiovasc Interv. 2013;6:790-799.

4.    Cohen HA, Williams DO, Holmes DR Jr, et al. Impact of age on procedural and 1-year outcome in percutaneous transluminal coronary angioplasty: a report from the NHLBI Dynamic Registry. Am Heart J. 2003;146:513-519.

5.    Ellis SG, Vandormael MG, Cowley MJ, et al. Coronary morphologic and clinical determinants of procedural outcome with angioplasty for multivessel coronary disease. Implications for patient selection. Multivessel Angioplasty Prognosis Study Group. Circulation. 1990;82:1193-1202.

6.    Feldman T. The SYNTAX score in practice: an aid for patient selection for complex PCI. Catheter Cardiovasc Interv. 2009;73:618-619.

7.    Sianos G, Morel MA, Kappetein AP, et al. The SYNTAX score: an angiographic tool grading the complexity of coronary artery disease. EuroIntervention. 2005;1:219-227.

8.    Anderson HV, Shaw RE, Brindis RG, et al. A contemporary overview of percutaneous coronary interventions. The American College of Cardiology-National Cardiovascular Data Registry (ACC-NCDR). J Am Coll Cardiol. 2002;39:1096-1103.

9.    King SB III, Smith SC Jr, Hirshfeld JW Jr, et al. 2007 focused update of the ACC/AHA/SCAI 2005 guideline update for percutaneous coronary intervention: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines: 2007 Writing Group to Review New Evidence and Update the ACC/AHA/SCAI 2005 Guideline Update for Percutaneous Coronary Intervention, Writing on Behalf of the 2005 Writing Committee. Circulation. 2008;117:261-295.

10.    McNulty EJ, Ng W, Spertus JA, et al. Surgical candidacy and selection biases in nonemergent left main stenting: implications for observational studies. JACC Cardiovasc Interv. 2011;4:1020-1027.

11.    Waldo SW, Secemsky EA, O’Brien C, et al. Surgical ineligibility and mortality among patients with unprotected left main or multivessel coronary artery disease undergoing percutaneous coronary intervention. Circulation. 2014;130:2295-2301.

12.    O’Neill WW, Kleiman NS, Moses J, et al. A prospective, randomized clinical trial of hemodynamic support with Impella 2.5 versus intra-aortic balloon pump in patients undergoing high-risk percutaneous coronary intervention: the PROTECT II study. Circulation. 2012;126:1717-1727.

13.    Nashef SA, Roques F, Michel P, Gauducheau E, Lemeshow S, Salamon R. European system for cardiac operative risk evaluation (EuroSCORE). Eur J Cardiothorac Surg. 1999;16:9-13.

14.    Roques F, Michel P, Goldstone AR, Nashef SA. The logistic EuroSCORE. Eur Heart J. 2003;24:881-882.

15.    Hannan EL, Farrell LS, Walford G, et al. The New York State risk score for predicting in-hospital/30-day mortality following percutaneous coronary intervention. JACC Cardiovasc Interv. 2013;6:614-622.

16.    Singh M, Gersh BJ, Li S, et al. Mayo Clinic risk score for percutaneous coronary intervention predicts in-hospital mortality in patients undergoing coronary artery bypass graft surgery. Circulation. 2008;117:356-362.

17.    Dangas GD, Kini AS, Sharma SK, et al. Impact of hemodynamic support with Impella 2.5 versus intra-aortic balloon pump on prognostically important clinical outcomes in patients undergoing high-risk percutaneous coronary intervention (from the PROTECT II randomized trial). Am J Cardiol. 2014;113:222-228.

18.    Head SJ, Holmes DR Jr, Mack MJ, et al. Risk profile and 3-year outcomes from the SYNTAX percutaneous coronary intervention and coronary artery bypass grafting nested registries. JACC Cardiovasc Interv. 2012;5:618-625.

19.    Hannan EL, Racz MJ, Gold J, et al. Adherence of catheterization laboratory cardiologists to American College of Cardiology/American Heart Association guidelines for percutaneous coronary interventions and coronary artery bypass graft surgery: what happens in actual practice? Circulation. 2010;121:267-275.

20.    Winslow CM, Kosecoff JB, Chassin M, Kanouse DE, Brook RH. The appropriateness of performing coronary artery bypass surgery. JAMA. 1988;260:505-509.

21.    Chan PS, Patel MR, Klein LW, et al. Appropriateness of percutaneous coronary intervention. JAMA. 2011;306:53-61.

22.    Baig SS, Altman DG, Taggart DP. Major geographical variations in elective coronary revascularization by stents or surgery in England. Eur J Cardiothorac Surg. 2015;47:855-859.

23.    Blankenship JC, Miller SC. Risky business: when patient preferences seem irrational. Catheter Cardiovasc Interv. 2013;82:219-220.

24.    Head SJ, Kaul S, Mack MJ, et al. The rationale for heart team decision-making for patients with stable, complex coronary artery disease. Eur Heart J. 2013;34:2510-2518.

___________________________________________________

From the ¹Division of Cardiovascular Medicine, University of Southern California, Los Angeles, California; ²Columbia University Medical Center/New York-Presbyterian Hospital and the Cardiovascular Research Foundation, New York, New York; ³Detroit Medical Center, Cardiovascular Institute, Detroit, Michigan; ⁴Henry Ford Hospital, Detroit, Michigan; and ⁵Beth Israel Deaconess Medical Center, Boston, Massachusetts.

Funding: The PROTECT II trial was funded by Abiomed. 

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Shavelle reports grant funds from St. Jude Medical, National Institutes of Health (NIH), Zoll Medical, Abbott Vascular, and Abiomed; consultant fees from St. Jude Medical. Dr Matthews reports grant and consultant fees from Medtronic; personal fees from Abiomed. Dr Popma reports personal fees from Abiomed. Dr Kirtane reports grant funds from Boston Scientific, Medtronic, Abbott Vascular, Abiomed, St. Jude Medical, Vascular Dynamics, and Eli Lilly. Dr Moses reports grant funds and personal fees from Abiomed. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript submitted September 18, 2015, provisional acceptance given October 5, 2015, final version accepted November 2, 2015.

Address for correspondence: David M. Shavelle, MD, Division of Cardiovascular Medicine, University of Southern California, 1510 San Pablo Street, Suite 322, Los Angeles, CA 90033. Email: shavelle@usc.edu