NEWS FROM THE NCDR
Patterns of Cardiac Marker Surveillance After Elective Percutaneous Coronary Intervention and Implications for the Use of Periprocedural Myocardial Infarction as a Quality Metric: A Report From the National Cardiovascular Data Registry (NCDR)
Tracy Y. Wang, MD, MS*,*,
Eric D. Peterson, MD, MPH*,
David Dai, MS*,
H. Vernon Anderson, MD ,
Sunil V. Rao, MD*,
Ralph G. Brindis, MD, MPH ,
Matthew T. Roe, MD, MHS* on behalf of the National Cardiovascular Data Registry (NCDR)
* Duke Clinical Research Institute, Durham, North Carolina
University of Texas Health Science Center, Houston, Texas
Northern California Kaiser Permanente, Oakland, California.
* Reprint requests and correspondence: Dr. Tracy Y. Wang, Duke Clinical Research Institute, 2400 Pratt Street, Room 0311, Terrace Level, Durham, North Carolina 27705. (Email: wang0085{at}mc.duke.edu).
|
Abbreviations and Acronyms
| | ACC = American College of Cardiology | | AHA = American Heart Association | | CABG = coronary artery bypass grafting | | CK = creatine kinase | | MI = myocardial infarction | | NCDR = National Cardiovascular Database Registry | | PCI = percutaneous coronary intervention | | ULN = upper limit of normal |
|
 |
The Clinical Issue
|
|---|
With recent advances in catheter technology and adjunctive pharmacotherapy, percutaneous coronary intervention (PCI) has achieved high procedural success rates and is a widely used means of myocardial revascularization among patients with coronary artery disease. Outcomes after PCI vary depending on patient selection, angiographic complexity, operator skill, and institutional care practices. Periprocedural myocardial infarction (MI) is detected in up to 30% of patients undergoing PCI, depending on the marker tested, population studied, and threshold for diagnosis based upon the degree of marker elevation (1). In clinical trials of PCI-related therapies, periprocedural MI is commonly adjudicated as an outcome measure and has been shown to be significantly associated with increased long-term mortality (2–5). However, there is no clearly defined threshold at which this increase in risk and definition of "myocardial infarction" converge, nor is there consensus on what treatment should be rendered when elevated markers are detected after a procedure (6). As such, expert panels have varied in their recommendations. Although some recommend routine cardiac marker testing after PCI (1), others recommend measurements driven by evidence of symptomatic ischemia or procedural complications (7,8).
Recent quality improvement initiatives have led to the development of both regional and national databases scrutinizing PCI practice and outcome variations (9–11). Whereas periprocedural MI has been suggested as a metric for assessing the quality of PCI care, in light of the aforementioned differences in opinion regarding the role of periprocedural marker testing, the viability of periprocedural MI as a quality metric should be assessed. Studies to date examining the incidence of periprocedural MI have largely been composed of single-center registries or post hoc clinical trial database analyses in which post-procedure cardiac marker measurements are mandated by study protocol. With the American College of Cardiology National Cardiovascular Database CathPCI Registry (ACC-NCDR), we investigated contemporary patterns of post-PCI cardiac marker testing in a large national database, patient predictors of marker testing, and variability among hospitals in the frequency of marker measurement and periprocedural MI detection.
|
Data From the ACC-NCDR
|
|---|
The ACC-NCDR is a large, ongoing, national registry of diagnostic cardiac catheterization and PCI in the U.S. and, as such, offers a unique opportunity to examine contemporary patterns of post-PCI cardiac marker testing. Patient and hospital characteristics were collected prospectively via a standardized set of data elements and definitions (12).
Between January 1, 2004, and March 30, 2007, data from 708,481 consecutive patients undergoing PCI in the U.S. were entered into the ACC-NCDR database. Patients who presented with acute coronary syndrome (n = 458,658) or with pre-procedure creatine kinase (CK)-MB or troponin levels greater than the upper limit of normal (ULN) (n = 36,428) were excluded, owing to the inability to determine whether post-PCI marker testing was performed as a result of the presenting ACS event and likely overlap of positive cardiac markers from the pre- to post-PCI time periods. The final study population consisted of 213,395 patients who underwent elective PCI at 463 hospitals with no reported evidence of myocardial necrosis before the PCI procedure.
|
Patterns of Cardiac Marker Testing
|
|---|
Across all hospitals, a median of 7% of patients undergoing PCI had post-procedure CK-MB levels measured. As shown in Figure 1, most hospitals did not routinely measure CK-MB levels after elective PCI; 298 of 463 hospitals (64.4%) measured post-procedure CK-MB levels in <20% of patients who underwent PCI. On the other end of the spectrum, 59 of 463 hospitals (12.7%) performed post-procedure marker testing more routinely in  70% of patients undergoing PCI. Hospitals that routinely measured post-procedure cardiac markers (defined as those that measured markers in 70% of patients undergoing PCI) had higher diagnostic catheterization and PCI volumes/year (Table 1).
A total of 52,746 patients (24.7% of the total PCI population) had CK-MB assessment after PCI. Among the 181,990 patients treated at 404 hospitals that do not routinely measure post-PCI markers, there were very few differences in baseline demographic or clinical characteristics between patients with and without CK-MB testing (Table 2). However, patients with post-PCI CK-MB measurement had higher-risk angiographic characteristics such as multivessel disease, bifurcation or type C lesions, and lesions requiring the use of debulking devices. Although periprocedural ischemia was not captured in the data collection form, patients with measured markers had longer procedural times and higher rates of procedural complications such as vessel dissection or acute vessel closure (Table 3).
View this table:
[in this window]
[in a new window]
|
Table 2 Baseline Patient Characteristics Among 181,990 Patients Treated at 404 Hospitals That Measured Post-PCI CK-MB Levels in <70% of Patients
|
|
View this table:
[in this window]
[in a new window]
|
Table 3 Angiographic and Procedural Characteristics Among 181,990 Patients Treated at 404 Hospitals That Measured Post-PCI CK-MB Levels in <70% of Patients
|
|
Multivariable logistic regression was used to determine factors independently associated with cardiac marker testing in patients who underwent PCI at hospitals that do not routinely assess post-procedure markers. Among these independent patient and hospital variables (Table 4), the most significant factor associated with testing was the presence of a periprocedural complication.
View this table:
[in this window]
[in a new window]
|
Table 4 Factors Associated With Post-Procedure Cardiac Marker Testing Among 181,990 Patients Treated at 404 Hospitals That Measured Post-PCI CK-MB Levels in <70% of Patients
|
|
Association of Marker Testing With Periprocedural MI Detection and Outcomes.
Among the 52,746 patients with measured CK-MB levels, 12,728 (24%) had new CK-MB elevations greater than the ULN after procedure. Of these patients, 8,351 (16%) had peak CK-MB levels between 1 and 3 x ULN and 4,377 (8%) had peak CK-MB levels >3 x ULN—a threshold often used to define periprocedural MI in consensus recommendations (6). Among patients with peak CK-MB levels >3 x ULN, only 14.9% had reported periprocedural complications such as acute closure, perforation, or dissection.
Patients undergoing elective PCI at hospitals that routinely measured post-procedure markers had a trend toward lower in-hospital mortality compared with those treated at hospitals that do not perform routine marker testing: odds ratio 0.74, 95% confidence interval 0.53 to 1.02 after adjustment for independent predictors of mortality derived from a validated mortality risk model for patients undergoing PCI (unpublished model from the ACC-NCDR database, c-index = 0.93, including age, body mass index, diabetes, prior heart failure, prior PCI, prior cerebrovascular or peripheral vascular disease, renal dysfunction, cardiogenic shock, ejection fraction, lesion severity, lesion complexity [ACC/American Heart Association (AHA)/Society of Cardiovascular Angiography and Interventions (SCAI) lesion classification type], and pre-procedure Thrombolysis In Myocardial Infarction [TIMI] flow). Patients treated at hospitals that routinely performed post-procedure CK-MB testing were also more likely to be discharged on guideline-recommended secondary prevention therapies.
As shown in Figure 2, periprocedural MI detection (defined as peak CK-MB levels >3 x ULN) correlated with the frequency of CK-MB measurement (p < 0.0001). Hospitals that more routinely measured markers had significantly higher rates of periprocedural MI detection (4.8% vs. 1.6%, p < 0.0001). However, the association was reversed when the denominator was limited to only those patients with measured CK-MB levels (n = 52,746). Hospitals that tested less frequently had higher rates of periprocedural MI detection (11.4% vs. 5.4%, p < 0.0001).
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2 Periprocedural MI Rate for 463 Hospitals
Periprocedural myocardial infarction (MI) rate for 463 hospitals when the denominator is (A) total number of patients undergoing percutaneous coronary intervention (PCI) at each hospital and (B) total number of PCI patients at each hospital with post-procedure creatine kinase (CK)-MB testing. ULN = upper limit of normal.
|
|
|
Discussion
|
|---|
This report from the ACC-NCDR shows that post-procedure cardiac markers are assessed in only one-quarter of patients after elective PCI, with wide variations in the pattern of marker surveillance across hospitals in the U.S. Hospitals that routinely performed marker testing had higher rates of periprocedural MI detection despite a trend toward lower mortality and greater adherence to recommended medications that suggest better overall quality of care for PCI patients at these hospitals. Therefore, in the absence of routine cardiac marker surveillance after PCI, the use of periprocedural MI as a quality metric for PCI will be misleading.
Current Patterns of Post-Procedure Cardiac Marker Testing.
In the ACC/AHA/SCAI PCI guidelines, post-procedure CK-MB measurement was designated as a class IB recommendation for patients with suspected ischemia during PCI, whereas routine marker measurement for all patients undergoing PCI was given a class IIC recommendation (8,13). This report shows that most clinicians are following guideline recommendations and primarily performing testing in patients with higher-risk lesions or procedural complications. As a result, cardiac markers are assessed infrequently after PCI; yet the frequency of marker measurement correlated directly with the likelihood of periprocedural MI detection. Although clinically apparent procedural complications such as distal embolization, side branch occlusion, or intimal dissection contribute to periprocedural myocardial injury (14–17), the overall incidence of such complications was low and the majority of patients with CK-MB elevation had no reported periprocedural complication. These observations raise the concern that in the absence of complications meeting the threshold for suspected ischemia during PCI that might instigate post-procedure marker testing, hospitals that do not routinely perform marker testing might significantly underestimate the true incidence of periprocedural MI.
Notably, 8,166 patients (4% of the overall population) had troponin-only measurements after PCI. Troponin has become the preferred cardiac biomarker, owing to its high sensitivity and specificity for myocardial damage (6). Among patients undergoing PCI, the magnitude of post-procedural troponin elevation correlates directly with the extent of myocardial injury (6,18). Yet, although troponin is more sensitive than CK-MB in detecting smaller amounts of periprocedural myonecrosis, the prognostic implications of troponin elevation after PCI are less certain (19–22). In a study by Cavallini et al. (20), troponin elevation was detected in 44% of patients undergoing PCI (vs. 16% CK-MB elevation) but, unlike CK-MB, did not predict 2-year mortality (odds ratio 1.2, 95% confidence interval 0.9 to 1.7). Similarly, a study by Miller et al. (23) showed that the rise in troponin post-procedure was not predictive of long-term outcomes. As such, troponin elevations were not incorporated into the definition of periprocedural MI in this report.
Implications of Detecting Periprocedural Myocardial Injury.
Routine marker measurement after PCI has been controversial, largely owing to its uncertain utility in the absence of clinically apparent ischemia during PCI, particularly in regard to therapeutic decision-making after PCI when an asymptomatic periprocedural MI is detected. Among patients with uncomplicated PCI, CK-MB elevation has been associated with discrete regions of delayed hyperenhancement within the target vessel perfusion territory on magnetic resonance imaging as well as more diffuse plaque burden and thrombus-rich lesions (24–26). As such, practice guidelines recommend that patients with periprocedural CK-MB elevation >3 x ULN should be treated like a standard MI patient with initiation of appropriate secondary prevention measures before discharge (6,8,13). This report showed that patients who underwent routine post-procedure CK-MB testing were more likely to be discharged on aspirin, clopidogrel, beta-blocker drugs, and statin drugs, so routine surveillance of post-PCI cardiac markers seems to correlate with better quality of overall hospital care for patients undergoing PCI.
On an institutional level, routine post-procedure marker measurement could potentially stimulate more rigorous assessment of PCI quality and influence PCI practices (27). Periprocedural MI rates higher than national standards should prompt institutional review of the patient case-mix for PCI procedures, decision-making regarding the choice of percutaneous versus surgical revascularization for individual patients, operator technique, device selection, and the use of adjunctive pharmacotherapies during PCI procedures that affect periprocedural complications.
Should Periprocedural MI Be Used as a Quality Metric?.
Periprocedural MI, an end point frequently used in clinical trials, has been suggested as a quality metric for PCI care (28). However, on the basis of these results, observational databases, in contrast to clinical trials, cannot provide an accurate assessment of the frequency of periprocedural MI, owing to wide variations in the measurement of post-procedure cardiac markers. Although several states have instituted quality improvement data registries that mandate post-procedural marker assessment (10,11), very few hospitals nationally perform systematic post-PCI marker surveillance. Because most periprocedural MIs are clinically silent, routine marker measurement after PCI is necessary to make adequate comparisons of quality of care and procedural outcomes across hospitals.
Hospitals that routinely measure post-procedure CK-MB levels likely represent those that have a strong interest in quality improvement, as suggested by the trend toward lower adjusted in-hospital mortality and greater use of appropriate discharge medications. However, the principle of "the more you look, the more you find" is demonstrated here. Hospitals that routinely screen for periprocedural MI might in fact be penalized for practicing a "quality process" by demonstrating worse outcomes (higher rates of periprocedural MI) when benchmarked against hospitals that measure post-procedure markers less frequently. Thus, at this time, periprocedural MI cannot be a quality metric for PCI care unless routine post-procedure marker measurement is implemented universally.
Several limitations of the data were present. First, among patients with post-procedural marker measurement, the incidence of periprocedural MI might be underestimated if the patient was discharged before enough time elapsed for post-PCI markers to reach a true peak. Second, detailed information regarding periprocedural ischemic symptoms and electrocardiographic changes was not collected in this registry. Finally, although we adjusted for a broad range of clinical and hospital variables in the mortality analysis, unmeasured confounders might have influenced the results.
|
Conclusions
|
|---|
Although periprocedural MI has been advocated as a quality indicator of PCI care, current data show that the majority of hospitals do not systematically assess cardiac markers after elective PCI; thus, the incidence of periprocedural MI might be significantly underestimated in current clinical practice. Therefore, the use of periprocedural MI as a quality metric for PCI procedures will remain misleading until efforts to standardize marker measurement practices are successful.
|
References
|
|---|
1. Califf RM, Abdelmeguid AE, Kuntz RE, et al. Myonecrosis after revascularization procedures J Am Coll Cardiol 1998;31:241-251.[Abstract/Free Full Text]2. Abdelmeguid AE, Topol EJ, Whitlow PL, Sapp SK, Ellis SG. Significance of mild transient release of creatine kinase-MB fraction after percutaneous coronary interventions Circulation 1996;94:1528-1536.[Abstract/Free Full Text] 3. Ioannidis JP, Karvouni E, Katritsis DG. Mortality risk conferred by small elevations of creatine kinase-MB isoenzyme after percutaneous coronary intervention J Am Coll Cardiol 2003;42:1406-1411.[Abstract/Free Full Text] 4. Brener SJ, Ellis SG, Schneider J, Topol EJ. Frequency and long-term impact of myonecrosis after coronary stenting Eur Heart J 2002;23:869-876.[Abstract/Free Full Text] 5. Kong TQ, Davidson CJ, Meyers SN, Tauke JT, Parker MA, Bonow RO. Prognostic implication of creatine kinase elevation following elective coronary artery interventions JAMA 1997;277:461-466.[Abstract/Free Full Text] 6. Thygesen K, Alpert JS, White HD, et al. Universal definition of myocardial infarction Circulation 2007;116:2634-2653.[Free Full Text] 7. Levine GN, Kern MJ, Berger PB, et al. Management of patients undergoing percutaneous coronary revascularization Ann Intern Med 2003;139:123-136.[Abstract/Free Full Text] 8. Smith Jr. SC, Feldman TE, Hirshfeld Jr. JW, et al. ACC/AHA/SCAI 2005 guideline update for percutaneous coronary intervention-summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/SCAI Writing Committee to Update the 2001 Guidelines for Percutaneous Coronary Intervention) J Am Coll Cardiol 2006;47:216-235.[Free Full Text] 9. Brindis RG, Fitzgerald S, Anderson HV, Shaw RE, Weintraub WS, Williams JF. The American College of Cardiology-National Cardiovascular Data Registry (ACC-NCDR): building a national clinical data repository J Am Coll Cardiol 2001;37:2240-2245.[Free Full Text] 10. Department of Health Care Policy. Harvard Medical School http://www.massdac.com 2001Accessed May 5, 2008. 11. Hannan EL, Arani DT, Johnson LW, Kemp Jr. HG, Lukacik G. Percutaneous transluminal coronary angioplasty in New York State. Risk factors and outcomes. JAMA 1992;268:3092-3097.[Abstract/Free Full Text] 12. ACC National Cardiovascular Data Registry, Cardiac Catheterization Module v3.02, Data Definitions http://www.accncdr.com/WebNCDR/NCDRDocuments/datadictdefsonlyv30.pdf 1992Accessed March 23, 2008. 13. King III SB, Smith Jr. SC, Hirshfeld Jr. JW, 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 2005 ACC/AHA/SCAI Guideline Update for Percutaneous Coronary Intervention) J Am Coll Cardiol 2008;51:172-209.[Free Full Text] 14. Klein LW, Kramer BL, Howard E, Lesch M. Incidence and clinical significance of transient creatine kinase elevations and the diagnosis of non-Q wave myocardial infarction associated with coronary angioplasty J Am Coll Cardiol 1991;17:621-626.[Abstract] 15. Oh JK, Shub C, Ilstrup DM, Reeder GS. Creatine kinase release after successful percutaneous transluminal coronary angioplasty Am Heart J 1985;109:1225-1231.[CrossRef][Web of Science][Medline] 16. Tardiff BE, Califf RM, Tcheng JE, et al. IMPACT-II Investigators Clinical outcomes after detection of elevated cardiac enzymes in patients undergoing percutaneous intervention. Integrilin (eptifibatide) to Minimize Platelet Aggregation and Coronary Thrombosis-II. J Am Coll Cardiol 1999;33:88-96.[Abstract/Free Full Text] 17. Topol EJ, Yadav JS. Recognition of the importance of embolization in atherosclerotic vascular disease Circulation 2000;101:570-580.[Free Full Text] 18. Selvanayagam JB, Porto I, Channon K, et al. Troponin elevation after percutaneous coronary intervention directly represents the extent of irreversible myocardial injury: insights from cardiovascular magnetic resonance imaging Circulation 2005;111:1027-1032.[Abstract/Free Full Text] 19. Cantor WJ, Newby LK, Christenson RH, et al. Prognostic significance of elevated troponin I after percutaneous coronary intervention J Am Coll Cardiol 2002;39:1738-1744.[Abstract/Free Full Text] 20. Cavallini C, Savonitto S, Violini R, et al. Impact of the elevation of biochemical markers of myocardial damage on long-term mortality after percutaneous coronary intervention: results of the CK-MB and PCI study Eur Heart J 2005;26:1494-1498.[Abstract/Free Full Text] 21. Kini AS, Lee P, Marmur JD, et al. Correlation of postpercutaneous coronary intervention creatine kinase-MB and troponin I elevation in predicting mid-term mortality Am J Cardiol 2004;93:18-23.[Web of Science][Medline] 22. Varani E, Balducelli M, Vecchi G, Gatti C, Lucchi GR, Maresta A. Occurrence of non-Q wave myocardial infarction following percutaneous coronary intervention in the stent era: systematic monitoring of the three markers of myocardial necrosis J Interv Cardiol 2005;18:243-248.[CrossRef][Medline] 23. Miller WL, Garratt KN, Burritt MF, Lennon RJ, Reeder GS, Jaffe AS. Baseline troponin level: key to understanding the importance of post-PCI troponin elevations Eur Heart J 2006;27:1061-1069.[Abstract/Free Full Text] 24. Kanaparti PK, Brown DL. Relation between coronary atherosclerotic plaque burden and cardiac enzyme elevation following percutaneous coronary intervention Am J Cardiol 2000;86:619-622.[CrossRef][Web of Science][Medline] 25. Mehran R, Dangas G, Mintz GS, et al. Atherosclerotic plaque burden and CK-MB enzyme elevation after coronary interventions: intravascular ultrasound study of 2256 patients Circulation 2000;101:604-610.[Abstract/Free Full Text] 26. Ricciardi MJ, Wu E, Davidson CJ, et al. Visualization of discrete microinfarction after percutaneous coronary intervention associated with mild creatine kinase-MB elevation Circulation 2001;103:2780-2783.[Abstract/Free Full Text] 27. Moscucci M, Eagle KA, Share D, et al. Public reporting and case selection for percutaneous coronary interventions: an analysis from two large multicenter percutaneous coronary intervention databases J Am Coll Cardiol 2005;45:1759-1765.[Abstract/Free Full Text] 28. Moscucci M, Rogers EK, Montoye C, et al. Association of a continuous quality improvement initiative with practice and outcome variations of contemporary percutaneous coronary interventions Circulation 2006;113:814-822.[Abstract/Free Full Text] 29. 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.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
S. R. Dixon, C. L. Grines, and W. W. O'Neill
The year in interventional cardiology.
J. Am. Coll. Cardiol.,
June 2, 2009;
53(22):
2080 - 2097.
[Full Text]
[PDF]
|
|
|
|
Top
The Clinical Issue
Data From the ACC-NCDR