Risk Stratification and Interventional Cardiology: Robert L. Frye Lecture

Mayo Clin Proc, December 2003, Vol 78

Risk Stratification and Interventional Cardiology

1507

Special Article

Risk Stratification and Interventional Cardiology: Robert L. Frye Lecture DAVID R. HOLMES, JR, MD straightforward, makes intuitive sense, and is easy and efficient to apply. Mayo Clin Proc. 2003;78:1507-1518

Risk stratification and risk-benefit ratios are extremely important in guiding patient-physician interactions as well as patient and family counseling. Risks associated with percutaneous transluminal coronary angioplasty are (1) compromise of the vessel lumen or vessel integrity, (2) unsuccessful procedure, and (3) restenosis. Predicting mortality risk depends on the specific patient population to be treated and on the specific mortality model used. The most common models are those from New York State, the American College of Cardiology, the Northern New England Cooperative Group, the University of Michigan, and The Cleveland Clinic Foundation. As more data and sophisticated analyses become available, risk stratification will become more accurate as long as the approach used is

CI = confidence interval; GUSTO-I = Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries; GUSTO-III = Global Use of Strategies to Open Occluded Coronary Arteries; NHLBI = National Heart, Lung, and Blood Institute; OR = odds ratio; PCI = percutaneous coronary intervention; PRESTO = Prevention of REStenosis with Tranilast and its Outcomes; PTCA = percutaneous transluminal coronary angioplasty; ROC = receiver operating characteristic

E

tions of the prototype equipment and because of a combination of incredible operator skill by Dr Andreas Gruntzig and a modicum of good fortune, the initial procedure was an unqualified success, with long-lasting, excellent results both angiographically, as assessed by follow-up angiography at 10 and 23 years, and clinically.2,3 The field of interventional cardiology may have turned out differently if the initial procedure had been complicated by dissection, acute occlusion, shock, or the need for emergency surgery. This index case revolutionized modern cardiovascular care. Although the index patient procedure was remarkably successful, subsequent experiences were not always as salutary. At an initial National Heart, Lung, and Blood Institute (NHLBI) PTCA workshop held in 1979,4 Gruntzig described his initial 76 patients, with data accumulated from September 1977 to June 1979. In this cohort, he was able to cross the lesion with the balloon catheter in 63 patients (83%), and the procedure was ultimately successful in 55 (72%). Among the 21 unsuccessful cases, acute closure occurred in 6, urgent surgery was necessary in 8, and elective surgery was required in 5. At the Mayo Clinic in Rochester, Minn, in 1978, the initial experience included 25 consecutive patients with a mean age of 50 years and a mean ejection fraction of 66%; 92% of the patients had single-vessel disease. The procedure was successful in 52% of the 25 patients. Coronary artery bypass graft surgery was performed within 24 hours in 32%, and emergency surgery was necessary in 12% of these patients. In 1980, at a symposium in Switzerland, 29 of the most active centers in the world reported their aggregate experi-

valuation of risk-benefit ratios and risk stratification are essential components of optimizing care for individual patients. This process is becoming increasingly evidence based as scientific information accumulates in the field of interventional cardiology. When any new procedure is developed, there is a learning curve; this process needs to account for diverse pieces of information, including the desired goals for the procedure (from both the patient’s and the physician’s perspectives), technical details, patient selection criteria, complication prevention and management, and alternative treatment strategies. Interventional cardiology amalgamates all these datasets. The first percutaneous transluminal coronary angioplasty (PTCA) was performed on September 16, 1977, in a 38-year-old man with a single discrete 85% stenosis involving the left anterior descending coronary artery, who presented with severe angina pectoris.1 The early equipment used for this procedure was unforgiving and bulky, and it offered limited steering capability. On the basis of these technical considerations, patient selection was limited to those with lesions in proximal coronary arterial segments without substantial tortuosity. Despite the limita-

From the Division of Cardiovascular Diseases and Internal Medicine, Mayo Clinic, Rochester, Minn. This article is based on Dr Holmes’ Robert L. Frye Lecture at Mayo Clinic, Rochester, Minn, July 23, 2003. Address reprint requests and correspondence to David R. Holmes, Jr, MD, Division of Cardiovascular Diseases, Mayo Clinic, 200 First St SW, Rochester, MN 55905. Mayo Clin Proc. 2003;78:1507-1518

1507

© 2003 Mayo Foundation for Medical Education and Research

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

1508

Risk Stratification and Interventional Cardiology

ence in 819 patients; success was achieved in 543 patients (66%). Throughout these early years, abundant data accumulated on the risks of the dilatation procedure. Three major groups of complications of PTCA were identified: (1) compromise of the vessel lumen or vessel integrity, (2) unsuccessful PTCA, and (3) restenosis. COMPROMISE OF THE VESSEL LUMEN OR VESSEL INTEGRITY Compromise of the vessel lumen or vessel integrity was typically the result of dissection or a thrombotic event, which resulted in acute closure and was the subject of multiple observational studies.5-13 These studies documented that acute or threatened closure occurred in 6% to 10% of dilatation procedures. In the NHLBI PTCA registry of 1801 patients, acute or threatened closure occurred in 122 (6.8%).5,14,15 In this and other series, occlusion was the single most important risk factor for in-hospital mortality, myocardial infarction, and emergency bypass surgery. Specific lesion characteristics associated with a marked increase in risk were identified—notably, those lesions containing a thrombus. In a sequential series of studies, Mabin et al,9 Sugrue et al,10 Reeder et al,11 and Singh et al12,13 showed that PTCA for treatment of lesions with a preexisting thrombus was associated with a marked increase in the incidence of acute closure. Even in a recent series of 7184 patients, the presence of a thrombus was an independent predictor of Q-wave infarction (odds ratio [OR], 3.78; 95% confidence interval [CI], 1.8-8.0) and an independent predictor of the composite end point of death, Qwave infarction, and emergency surgery (OR, 2.37; 95% CI, 1.4-4.1).12 UNSUCCESSFUL PTCA As mentioned previously, approximately 40% of the procedures during the early years of intervention were unsuccessful. There were 2 types of unsuccessful procedures: (1) uncomplicated, when the lesion could not be crossed, and (2) complicated, when the lesion was made worse by the procedure. The ability to predict the specific type before the procedure was performed was extremely limited. RESTENOSIS As investigators have learned more about restenosis, they have identified its complexity more fully.16-24 The current belief is that restenosis has 4 components: acute elastic recoil, chronic negative remodeling, neointimal hyperplasia, and excessive matrix formation. The extent to which restenosis would be an issue was not fully appreciated during the early years of intervention. In one of the earliest reports, the NHLBI documented results of follow-up an-

Mayo Clin Proc, December 2003, Vol 78

giography in 557 patients who had an initially successful PTCA.16 Those investigators identified restenosis in 33.6% of the patients and further identified associated risk factors. These risk factors included male sex, a large pressure gradient before the procedure or persisting afterward (a surrogate for a suboptimal result), new-onset angina or unstable angina, diabetes mellitus, and treatment of a saphenous vein bypass graft. However, the predictive power of any of these factors was not tested. In response to these issues and complications, multiple classes of devices and improved iterations within each class have been developed.25-31 The combination of new devices, improved operator experience, and the dramatic broadening of criteria for patient selection and type of lesion has led to the marked increase in success rate and procedural volume; thus, percutaneous coronary intervention (PCI) (a change in terminology from “PTCA”) is now the most frequently used revascularization approach for coronary artery disease. In approximately 90% of all procedures, PCI now involves stent placement and more intense antiplatelet strategies, including dual oral antiplatelet drugs and intravenous glycoprotein IIb/IIIa receptor inhibitors.32-38 During the past 25 years, as the field has matured, more scientific data have become available for risk-benefit ratio analysis and risk stratification. Some of the same problems and complications identified early in the field continue to exist; the frequency of some of these has decreased, whereas the frequency of others has remained the same. In addition, with the changes in technology and in criteria for patient selection and type of lesion, some new complications have developed. In the largest and most contemporary interventional cardiology trial focused on restenosis, Prevention of REStenosis with Tranilast and its Outcomes (PRESTO), 11,484 patients were randomly assigned to receive either placebo or 1 of 4 dosages of oral tranilast, a drug that prevented restenosis in small pilot studies.39 In this trial, no drug effect was identified, and angiographic restenosis rates in the 5 groups ranged from 32% to 35%, a range similar to that reported in 1984,16 although the patient populations treated and the devices used in PRESTO were different from those in the 1984 series. Other investigators used this PRESTO dataset to evaluate the ability to predict subsequent restenosis (M. Singh, MD, B. J. Gersh, MB, ChB, DPhil, R. L. McClelland, PhD, et al, unpublished data, 2003). They evaluated 1312 patients who had a single lesion treated successfully by PCI and were enrolled in the angiographic substudy of PRESTO. The authors used clinical and angiographic variables that had been identified in studies available before the procedure and applied them to the PRESTO database. These

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

Mayo Clin Proc, December 2003, Vol 78

variables were forced into a logistic regression model. The authors also used a bootstrap approach to evaluate a model derived solely from the PRESTO database. The multivariate model for factors associated with subsequent restenosis is shown in Table 1. Although these factors had significant associations with the development of subsequent restenosis, the area under the receiver operating characteristic (ROC) curve was 0.63, indicating only a modest ability to discriminate between patients with and without restenosis. By adding factors identified in previous studies, including sex, vessel size, lesion length, prior percutaneous treatment, unstable angina, diabetes mellitus, and lesion type, the area under the ROC curve was still only 0.63, again indicating a modest ability to discriminate between patients with and without restenosis. The ability to predict subsequent restenosis has become increasingly important. Within the past year, there has been intense interest in drug-eluting stents, and one of them has been approved by the US Food and Drug Administration.40-42 These stents have been studied in selected patients with certain types of lesions in randomized, multicenter clinical trials. Typically, in these trials only 1 de novo lesion has been treated per patient; in-stent restenosis rates have been less than 5% and in-segment restenosis rates less than 10%. However, these stents are approximately 3 times as expensive as bare metal stents. Because of current economic and reimbursement issues, it would be helpful to predict which patients would benefit maximally from these drug-eluting stents and which would benefit only minimally. This will become even more important as the stents are used to treat multiple lesions in multiple vessels. At present, the ability to predict which patients will receive maximal benefit from drug-eluting stents is limited. ACUTE CLOSURE As mentioned previously, with conventional PTCA, acute or threatened closure occurs in approximately 6% to 10% of patients. With stent implantation and adjunctive therapy, the incidence is significantly lower, although when it occurs, acute or threatened closure is still associated with nonfatal myocardial infarction or death in up to approximately 50% of patients. Orford et al43 evaluated a singlecenter database and found that stent thrombosis occurred in 0.57% of 4509 consecutive patients treated with a stent and dual antiplatelet therapy. With multivariate analysis, the number of stents placed was the only factor associated with increased frequency of stent thrombosis. In a multicenter experience, Cutlip et al44 evaluated 6 randomized clinical trials and registries involving 6186 patients and 6219 vessels treated with 1 or more stents and dual antiplatelet therapy. Stent thrombosis occurred in 0.9% of the patients. Only 3 factors were associated with increased rates: final

Risk Stratification and Interventional Cardiology

1509

Table 1. Multivariate Model for Factors Associated With Subsequent Restenosis* Factor

OR (95% CI)

P value

Treated diabetes mellitus Nonsmoker Vessel <3 mm vs ≥3 mm Lesion length >20 mm vs <10 mm Ostial location Prior PCI

1.45 (1.06-1.98) 1.34 (1.01-1.86) 1.32 (1.04-1.68) 2.36 (1.49-3.75) 1.82 (1.13-2.9) 1.41 (1.09-1.81)

.02 .04 .02 .003 .13 .008

*CI = confidence interval; OR = odds ratio; PCI = percutaneous coronary intervention.

dissection (OR, 3.8; 95% CI, 1.9-7.7); stent length (OR, 1.3; 95% CI, 1.2-1.5); and final minimal lumen diameter (OR, 0.4; 95% CI, 0.2-0.8). Because the incidence of stent thrombosis is low, the ability to predict this specific event is low. Meticulous attention to optimizing stent implantation results and ensuring adequate antiplatelet therapy is the mainstay to prevent this severe complication. However, 1 group of patients had an increased rate of late occlusion, presumably thrombotic in most patients. These were patients treated with vascular brachytherapy for restenosis who had a new stent placed and who received no long-term dual antiplatelet therapy.45-48 In several series, late occlusion of the target lesion was noted in up to 8% to 9% of these patients. Addressing both issues by minimizing new stent placement and, more importantly, by prolonging dual antiplatelet therapy has resolved this problem. UNSUCCESSFUL PCI Despite changes in selection criteria in patient populations undergoing PCI, the success rates have increased markedly.25,27-29,49 Reports from registries involving large numbers of patients are now available (Table 2); success rates are uniformly excellent with low in-hospital mortality. SurTable 2. Recent Series of Patients Undergoing Percutaneous Coronary Intervention*

Study

Year

No. of patients

NNE30 NCN50 SCAI25 NACI49 NHLBI29 ACC-NCDR31

1999 2000 2000 2000 2000 2002

34,752 76,904 19,510 4279 1559 100,292

Success (%)

In-hospital mortality (%)

CABG (%)

91.9 96.8 … 90.4 93.7 94.5

1.1 1.3 0.5 1.6 1.9 1.4

1.3 1.7 0.5 1.7 0.4 1.9

*ACC-NCDR = American College of Cardiology-National Cardiovascular Data Registry; CABG = coronary artery bypass graft; NACI = New Approaches in Coronary Interventions; NCN = National Cardiovascular Network; NHLBI = National Heart, Lung, and Blood Institute; NNE = Northern New England; SCAI = Society for Cardiac Angiography and Interventions.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

Risk Stratification and Interventional Cardiology

1510

Mayo Clin Proc, December 2003, Vol 78

Emergency CABG (%)

2.0

1.5

1.0

0.5

0.0 1992

1994

1996

1998

2000

Year of procedure

Figure 1. Change in frequency of emergency coronary artery bypass graft (CABG) surgery after percutaneous coronary intervention (n=18,593). With the introduction of coronary stents (which came into general use in the mid 1990s), the rates of emergency CABG surgery decreased significantly (P<.001).

gical rates are also low. Emergency surgery rates are particularly low in light of the widespread use of intracoronary stent implantation. Seshadri et al51 evaluated emergency surgery after PCI in a single-center observational study of 18,593 procedures. In this group, 113 (0.61%) required emergency surgery. During the period studied, the rates of emergency surgery decreased significantly (Figure 1), particularly after the introduction of stent implantation. However, when emergency surgery was required, associated adverse sequelae were frequent: mortality, 15%; Q-wave infarction, 12%; stroke, 5%; and respiratory failure, 2%. For risk stratification, it is important to identify patients in whom a stent could not be placed because of severe tortuosity, severe calcification, or severe diffuse disease. In such patients who do not have a “bail-out” stent option, the risks of the procedure are clearly increased, as are the risks of potential emergency surgery. CONTRAST MEDIA–ASSOCIATED RENAL FAILURE Contrast media–associated renal failure has received considerable attention as a result of treating higher risk patients who have more comorbid conditions; performing more extensive, prolonged procedures with larger volumes of contrast media; and recognizing the extent and severity of the problem.52-57 Contrast media–associated renal failure accounts for approximately 12% of all in-hospital cases of renal failure, exceeding the rate of aminoglycoside-induced nephrotoxicity. Rihal et al58 evaluated acute renal failure in a single-center observational study of 7586 patients. In this population, 3.3% (254 patients) had acute renal failure, defined as a serum creatinine concentration that increased more than 0.5 mg/dL from baseline. Among

patients in whom acute renal failure developed, in-hospital mortality was 22%; among patients without this complication, it was 1.4%. Using multivariate analysis, Rihal et al58 identified 4 factors significantly associated with subsequent acute renal failure: baseline creatinine level, treatment of acute myocardial infarction, shock, and the volume of contrast medium administered. Knowledge of this risk has resulted in the development of multiple strategies, including new pharmacological approaches such as the use of fenoldopam, N-acetylcysteine, and theophylline. Controversy remains regarding the efficacy of each of these because the results of small trials and registries have been conflicting. At present, avoiding volume depletion and hypotension and using minimal volumes of contrast media are the fundamental approaches for preventing acute renal failure associated with angiographic procedures. MORTALITY Mortality modeling has attracted the most interest.59-66 Several models have been developed—some for single-center registries and some for large multicenter databases. These models have used many similar variables, although not identical, including patient demographics, clinical presentation, type of procedure, and angiographic anatomical features (Table 3). The most common, well-known models include those from New York State, the American College of Cardiology, the Northern New England Cooperative Group, the University of Michigan, and The Cleveland Clinic Foundation. These models were all evaluated in the NHLBI Dynamic Registry of 4448 patients treated with PCI from 1997 to 1999.66 Stents were placed in 74% of the patients, and glycoprotein IIb/IIIa agents were used in 27%. In-hospital death occurred in 64 patients (1.4%). Three of the models predicted similar mortality: the New York State model, 1.50% (95% CI, 0.89%-1.70%); the Northern New England model, 1.3% (95% CI, 1.0%-1.7%); and the Cleveland Clinic model, 1.7% (95% CI, 1.3%-2.1%). With these 3 models, the difference between observed and predicted in-hospital mortality in high-risk patient subsets was small. The other 2 models yielded different results: the University of Michigan model predicted a smaller proportion of deaths (1.1%), and the American College of Cardiology registry predicted substantially more deaths (13.5%) (Figure 2). Predicting the risk of mortality depends on the specific patient population to be treated and on the specific model used. Taking these into consideration, it is still possible to determine a general risk of death for a patient population, although applying it to an individual patient is problematic. Singh et al65 further refined risk stratification. They evaluated 5463 PCI procedures to identify clinical and

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

Mayo Clin Proc, December 2003, Vol 78

Risk Stratification and Interventional Cardiology

1511

Table 3. Variables Used in Mortality Models*

*CABG = coronary artery bypass graft; IABP = intra-aortic balloon pump; LAD = left anterior descending coronary artery; MI = myocardial infarction; PCI = percutaneous coronary intervention; PTCA = percutaneous transluminal coronary angioplasty; SCAI = Society for Cardiac Angiography and Interventions.

No. of predicted deaths

240

New York State Age Female sex Prior MI and timing Shock Hemodynamic instability Renal failure Peripheral vascular disease Diabetes mellitus Congestive heart failure Prior PTCA Prior CABG surgery Multivessel disease IABP Left ventricular ejection fraction American College of CardiologyNational Cardiovascular Data Registry Acuteness of PCI Shock IABP Age Diabetes mellitus Left ventricular ejection fraction Acute MI within 24 h Lesion classification based on SCAI Left main disease Proximal LAD lesion Renal failure Chronic lung disease Thrombolytic therapy Device other than stent used Northern New England Cooperative Group Age Therapy for acute MI Shock Urgent PCI Emergent PCI Left ventricular ejection fraction Creatinine level ≥2.0 mg/dL Peripheral cerebrovascular disease Congestive heart failure IABP Attempted treatment of class C lesion University of Michigan Consortium MI within 24 h Shock Creatinine level >1.5 mg/dL Cardiac arrest Number of diseased vessels Age Left ventricular ejection fraction Thrombus Peripheral vascular disease Female sex Cleveland Clinic Foundation Multicenter Age Shock Acute MI Lesion complexity Male sex Number of diseased vessels

New York State Northern New England Cleveland Clinic ACC University of Michigan

210 180 150 120 90 60 30 0 0

10

20

30

40

50

No. of observed deaths

Figure 2. In-hospital mortality by risk decile observed in the National Heart, Lung, and Blood Institute Dynamic Registry and predicted by using the New York State, Northern New England, Cleveland Clinic, American College of Cardiology (ACC), and University of Michigan models. From Holmes et al,66 with permission from the American Heart Association.

angiographic risk factors associated with the major complications of in-hospital death, Q-wave myocardial infarction, emergent or urgent coronary artery bypass graft surgery, and stroke. They used these risk factors to develop a simple integer score. In this group of patients, they identified 5 clinical and 3 angiographic variables that were associated with this composite risk of complications (Table 4). The average ROC curve area was 0.782, which is excellent. The expected and observed procedural complication rates are shown in Figure 3. The Hosmer-Lemeshow goodness-offit-test (6 df) was 1.83 (P=.93). Use of the integer risk score showed the following: 2145 procedures (39.3%) were classified as very low risk, and 1.0% of the patients in this group had a procedural complication (expected range, ≤2%); 2182 procedures (39.9%) were classified as low risk, and 3.0% of the patients in this group had a procedural complication (expected range, >2%-5%); 809 procedures (14.8%) were classified as moderate risk, and 6.2% of the patients in this group had a procedural complication (expected range, >5%-10%); 210 procedures (3.8%) were classified as high risk, and 19.5% of the patients in this group had a procedural complication (expected range, 10%-25%); and 117 procedures (2.1%) were classified as very high risk, and 35% of the patients in this group had a procedural complication (expected range, >25%). With the use of readily observed clinical and angiographic characteristics as risk factors, procedural risk can be estimated more accurately for the composite complications of in-hospital death, Q-wave myocardial infarction, urgent or emergency coronary surgery, and stroke.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

1512

Risk Stratification and Interventional Cardiology

Mayo Clin Proc, December 2003, Vol 78

Table 4. Multivariate Predictors of Procedural Complications After Percutaneous Coronary Intervention* Variable

Integer score

Model coefficient†

Odds estimate

95% CI

P value

Cardiogenic shock Left main coronary artery disease Serum creatinine level >3 mg/dL Urgent or emergent procedure NYHA classification ≥III Thrombus Multivessel disease Age, No. of decades after 30 y Intercept

5 5 3 2 2 2 2 1 NA

1.599 1.467 0.881 0.758 0.745 0.644 0.618 0.313 –5.965

4.95 4.34 2.41 2.13 2.11 1.90 1.86 1.37 NA

3.4-7.2 2.5-7.6 1.4-4.2 1.5-3.1 1.4-3.1 1.4-2.6 1.3-2.6 1.2-1.6 NA

<.001 <.001 .001 <.001 <.001 <.001 <.001 <.001 NA

*CI = confidence interval; NA = not applicable; NYHA = New York Heart Association. †Model χ28 = 293.3, P<.001; mean ± SD for bootstrap receiver operating characteristic curve areas, 0.782±0.018. From Singh et al,65 with permission from the American College of Cardiology.

Risk stratification models have also been developed for selected subsets of patients at very high risk of dying. Hasdai et al67 studied predictors of death among patients with cardiogenic shock. Of the 41,021 patients in the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO-I) trial, they evaluated 2968 patients with cardiogenic shock. The published mortality rates are extremely high for patients in whom cardiogenic shock develops; therefore, identification of patients who have an even greater risk of

0.9 0.8

Estimated risk

0.7 0.6 0.5 0.4 0.3

Integer coefficient Risk factor 6 Age, 90-99 y 5 Preprocedural shock Left main coronary 5 artery disease 5 Age, 80-89 y 4 Age, 70-79 y 3 Renal disease 3 Age, 60-69 y 2 Nonelective procedure 2 Multivessel disease 2 NYHA class III or greater 2 Thrombus 2 Age, 50-59 y 1 Age, 40-49 y 0-25 Risk score, typical

1.0 0.9 0.8

Complication rate

1.0

dying may foster the development of improved treatment strategies.68,69 Hasdai et al67 used logistic regression modeling techniques to evaluate the relationships and associations between demographic, clinical, and hemodynamic characteristics and 30-day mortality among 2968 patients; 995 of these patients underwent right heart catheterization. There were significant differences between patients who had a fatal outcome within 30 days and those who survived 30 days (Table 5). Both clinical and hemodynamic variables were associated with 30-day mortality.67 A prognos-

25%

0.7 0.6

3

0.5 8

0.4

13 14

0.3 11

0.2

0.2

55 24

10%

0.1 2%

0.0 Risk score: 0 At risk, %: 39.3

0.1

5%

29 87 120 207 228 290

0.0 5 39.9

10 14.8

3.8

15 2.1

20

25

0 1 2 3

145 230 215

25

74

4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Integer score

Figure 3. Left, Estimated rates of procedural complications with the integer scoring system. The integers are proportional to the estimated continuous coefficient from the logistic model. Percentages of patients at risk are shown for each of the 5 risk categories: ≤2% is very low risk for complications with coronary angioplasty; >2% to 5%, low risk; >5% to 10%, moderate risk; >10% to 25%, high risk; and >25%, very high risk. NYHA = New York Heart Association. Right, The observed and corresponding predicted procedural complications in the validation-set with the proposed risk score. The x-axis represents the risk score in the validation-set. The y-axis represents the complication rate. The solid line shows the predicted procedural complications derived from the risk score. The bars represent observed procedural complications in the validation-set. Three patients had a score >19; all 3 were free of complications. From Singh et al,65 with permission from the American College of Cardiology.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

Mayo Clin Proc, December 2003, Vol 78

Risk Stratification and Interventional Cardiology

1513

Table 5. Hemodynamic Variables (Including Data Collected Only for Patients With Cardiogenic Shock), Right Heart Catheterization Data, and 30-Day Mortality Rate Data for Patients With Right Heart Catheterization Data*† Characteristic Lowest cardiac output ≤5.1 (eg, 4.5 vs 3.5) >5.1 (eg, 6.5 vs 5.5) Oliguria Mean arterial pressure during shock Highest pulmonary capillary wedge pressure Altered sensorium Ventricular septal defect Heart rate during shock Prior congestive heart failure Cold, clammy skin

Wald χ2

P value

49.78

<.001

28.20 24.02

<.001

18.57 11.28 9.76 9.56

.001

9.43 5.53

Odds ratio

95% CI

0.57‡ 1.13‡ 2.41

0.48-0.66 1.04-1.24 1.74-3.34

1.80 3.83 1.10§

1.28-2.54 1.65-8.89 1.03-1.16

1.87 1.56

1.25-2.80 1.08-2.25

*CI = confidence interval. †Overall model χ216 = 197.79; concordance index = 0.798; validated model = 0.785. ‡For an increase of 1.0 unit. §For an increase of 10 units. From Hasdai et al,67 with permission from Mosby.

Hasdai et al70 evaluated factors associated with the development of shock after thrombolytic therapy for acute infarction. Such attempts may be helpful; if patients are identified as being at high risk for the development of shock, they may be transferred for more aggressive treatment. These authors evaluated baseline variables in the GUSTO-I trial and developed a scoring system for predicting the risk of shock. They then assessed the validity of this model in the GUSTO-III study of 14,960 patients; in 643 patients, shock developed after enrollment. In GUSTO-I, 4 variables provided more than 85% of the information

tic algorithm based on point scores was developed for patients with no right heart catheterization data (Tables 68) and for patients with right heart catheterization data (Tables 9-11). With these scores, the estimated probability of 30-day mortality ranged from 10% to 90%. Studies such as those by Singh et al65 and Hasdai et al67 will form the basis for developing new treatment strategies, which may be used to further improve outcomes or to triage patients who may not survive, irrespective of treatment rendered. These studies will also form the basis for useful family counseling and planning.

Table 6. Scoring System for Continuous Variables for Predicting Mortality From Cardiogenic Shock Among Patients With No Right Heart Catheterization Data* Baseline Age

Height

Systolic blood pressure

Heart rate

Time to thrombolytic treatment

Year

Points

cm

Points

Beats/min

Points

mm Hg

Points

Hour

Points

20 30 40 50 60 70 80 90 100

0 3 5 8 12 21 40 60 80

140 150 160 170 180 190 200 210

66 46 30 37 30 23 15 8

40 60 80 100 120 140 160 180 200

8 12 17 21 25 29 33 37 42

20 40 60 80 100 160 180 200 220 240

36 28 20 12 5 1 2 3 5 6

0 2 4 6 8 10 12 14

10 4 10 16 15 13 11 9

*Scoring system for categorical variables is shown in Table 7. From Hasdai et al,67 with permission from Mosby.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

1514

Risk Stratification and Interventional Cardiology

Mayo Clin Proc, December 2003, Vol 78

Table 8. Scores Corresponding to Risks of 30-Day Mortality From Cardiogenic Shock Among Patients With No Right Heart Catheterization Data*

Table 7. Scoring System for Categorical Variables for Predicting Mortality From Cardiogenic Shock Among Patients With No Right Heart Catheterization Data* Risk factor

Points

Prior infarction Prior angina Infarction location Anterior Inferior Other Killip class I II III IV Diabetes Smoking status Current Former Never No extramyocardial factors corrected Altered sensorium Cold, clammy skin Oliguria Ventricular septal defect Ventricular rupture Arrhythmia

13 8

Points

Probability of 30-day mortality (%)

103 126 141 154 165 176 189 204 227

10 20 30 40 50 60 70 80 90

13 0 10 0 11 20 1 13

*Scoring system is shown in Tables 6 and 7. From Hasdai et al,67 with permission from Mosby.

0 1 8 8 18 18 29 34 74 10

the determination of the probability of developing in-hospital shock, ranging from 1% to 50%.

*Scoring system for continuous variables is shown in Table 6. From Hasdai et al,67 with permission from Mosby.

needed to predict the occurrence of shock. These included age, systolic blood pressure, heart rate, and Killip class on presentation (Table 12). When these same 4 variables were assessed in GUSTO-III, they accounted for more than 95% of the predictive information, with a concordance index of 0.796. From their data, Hasdai et al also developed an algorithm based on point scores (Table 13). This allowed

SUMMARY AND CONCLUSIONS Risk stratification and risk-benefit ratios form an increasingly important part of patient-physician interactions and patient and family counseling. As more data and more sophisticated analyses become available, this approach will become more accurate, provided that it is straightforward, makes intuitive sense, and is easy and efficient to apply. Such strategies will foster a more rational approach to optimizing care and recommending the most appropriate treatment strategy. However, of importance, statistics and models are based on populations of patients; for patients and families with unique situations, these models may be used to provide general strategies rather than specific solutions.

Table 9. Scoring System for Continuous Variables for Predicting Mortality From Cardiogenic Shock Among Patients With Right Heart Catheterization Data* During shock Mean arterial pressure

Age

Lowest cardiac output

Heart rate

Highest pulmonary capillary wedge pressure

Year

Points

mm Hg

Points

Beats/min

Points

L/min

Points

mm Hg

Points

20 30 40 50 60 70 80 90 100

0 11 22 33 44 56 67 78 89

20 40 60 80 100 120 140 160 180

20 38 35 17 19 25 32 38 44

20 40 60 80 100 120 140 160 180

5 11 16 22 27 33 38 43 49

2 4 6 8 10 12 14 16 18

42 13 0 7 13 20 27 33 40

10 20 30 40 50 60

31 0 24 26 25 24

*Scoring system for categorical variables is shown in Table 10. From Hasdai et al,67 with permission from Mosby.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

Mayo Clin Proc, December 2003, Vol 78

Risk Stratification and Interventional Cardiology

Table 11. Scores Corresponding to Risks of 30-Day Mortality From Cardiogenic Shock Among Patients With Right Heart Catheterization Data*

Table 10. Scoring System for Categorical Variables for Predicting Mortality From Cardiogenic Shock Among Patients With Right Heart Catheterization Data* Risk factor

Points

Killip class I II III IV Prior infarction Altered sensorium Cold, clammy skin Oliguria Ventricular septal defect

7 26 25 0 15 15 15 23 38

1515

*Scoring system for continuous variables is shown in Table 9. From Hasdai et al,67 with permission from Mosby.

Points

Probability of 30-day mortality (%)

138 160 175 188 199 210 223 238 260

10 20 30 40 50 60 70 80 90

*Scoring system is shown in Tables 9 and 10. From Hasdai et al,67 with permission from Mosby.

Table 12. Baseline Predictors of Developing Cardiogenic Shock for 38,942 Patients With 1889 Events* Characteristic Age Systolic BP Heart rate Killip class II vs I III vs I MI location Anterior vs other Inferior vs other United States Treatment SK-IV vs tPA Combo vs tPA SK-SQ vs tPA Previous MI Previous CABG surgery Weight Female Hypertension Previous PTCA Diastolic BP

Wald χ2

df

P value

285.14 279.55 225.28 161.35

1 2 3 2

<.001 <.001 <.001 <.001

77.05 43.92 36.87

2

Hazard ratio

95% CI

1.47†

1.40-1.53

1.70 2.95

1.52-1.90 2.39-3.63

1.62 1.07 1.39

1.21-2.15 0.80-1.43 1.26-1.53 1.22-1.59 1.07-1.40 1.28-1.66 1.20-1.50 1.21-1.76 0.91-0.97 1.09-1.35 1.05-1.26 0.54-0.91 1.01-1.11

<.001

1 3

<.001 <.001

25.61

1

<.001

1.39 1.22 1.46 1.34

15.38 13.65 12.69 8.42 7.31 5.73

1 1 1 1 1 2

<.001 <.001 <.001 .004 .007 .02

1.46 0.94† 1.22 1.15 0.70 1.06†

*Overall model χ221 = 1789; concordance index = 0.761; validated model = 0.758. BP = blood pressure; CABG = coronary artery bypass graft; CI = confidence interval; Combo = combination of tissue-type plasminogen activator (tPA) and streptokinase (SK); IV = intravenous; MI = myocardial infarction; PTCA = percutaneous transluminal coronary angioplasty; SQ = subcutaneous. †For an increase of 10 units. From Hasdai et al,70 with permission from the American College of Cardiology.

REFERENCES 1. 2. 3.

Gruntzig A. Transluminal dilatation of coronary-artery stenosis [letter]. Lancet. 1978;1:263. King SB III, Schlumpf M. Ten-year completed follow-up of percutaneous transluminal coronary angioplasty: the early Zurich experience. J Am Coll Cardiol. 1993;22:353-360. Meier B. The first patient to undergo coronary angioplasty—23year follow-up [letter]. N Engl J Med. 2001;344:144-145.

4.

5.

Willman VL, Bulkley B, Gruntzig A, et al. Proceedings of the Workshop on Percutaneous Transluminal Coronary Angioplasty: June 15-16, 1979. Bethesda, Md: US Dept of Health, Education, and Welfare, Public Health Service, National Institutes of Health; 1980:9-15. DHEW publication no. (NIH) 80-2030. Holmes DR Jr, Holubkov R, Vlietstra RE, et al, Co-Investigators of the National Heart, Lung, and Blood Institute Percutaneous Transluminal Coronary Angioplasty Registry. Comparison of com-

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

1516

Risk Stratification and Interventional Cardiology

Mayo Clin Proc, December 2003, Vol 78

Table 13. Algorithm for Determining the Probability of In-hospital Cardiogenic Shock* Step 1. Find the value that most closely matches the patient’s risk factors and circle the points for each predictive factor. BP Age Year Points

Heart rate

Systolic

Diastolic

Beats/ min Points

mm Hg Points

mm Hg Points

Weight

Thrombolytic therapy

Killip class

kg Points

Therapy Points

Class Points

Miscellaneous risk factors

MI site Site

Points

Factor

Points

Previous MI Previous CABG surgery No previous PTCA Female Hypertension United States

5

20

6

40

3

80

59

40

4

40

19

tPA

0

I

0

Anterior

8

30

12

60

0

100

49

60

5

60

17

SK-IV

5

II

9

Inferior

1

40

19

80

8

120

39

80

7

80

15

Combo

3

III

17

Other

0

50 60

25 31

100 120

14 17

140 160

32 27

100 120

9 11

100 120

12 10

SK-SQ

6

70

37

140

19

180

23

140

13

140

8

80 90

43 49

160 180 200 220 240 260

22 24 27 29 32 34

200 220 240 260 280

18 14 9 5 0

160 180 200

15 16 18

160 180 200 220

6 4 2 0

6 6 3 2 5

Step 2. Sum the circled points of all predictive factors. Step 3. Determine the predicted survival from cardiogenic shock corresponding to the total number of points. For example, a 71-year-old (37 points), 60kg (17 points) woman (3 points) from the United States (5 points) with a history of hypertension (2 points) who presents with systolic BP of 126 mm Hg (39 points), diastolic BP of 64 mm Hg (5 points), and heart rate of 123 beats/min (17 points), is in Killip class III (17 points), and has had an anterior MI (8 points) treated with SK-IV (5 points) would have a total score of 155. This score corresponds to a predicted probability of 40% for cardiogenic shock occurring after thrombolytic therapy.

Points

Probability of 30-day survival from in-hospital cardiogenic shock (%)

92 103 110 114 118 130 137 142 146 149 152 155 158 160

1 2 3 4 5 10 15 20 25 30 35 40 45 50

*BP = blood pressure; CABG = coronary artery bypass graft; Combo = combination of tissue-type plasminogen activator (tPA) and streptokinase (SK); IV = intravenous; MI = myocardial infarction; PTCA = percutaneous transluminal coronary angioplasty; SQ = subcutaneous. From Hasdai et al,70 with permission from the American College of Cardiology.

6.

plications during percutaneous transluminal coronary angioplasty from 1977 to 1981 and from 1985 to 1986: the National Heart, Lung, and Blood Institute Percutaneous Transluminal Coronary Angioplasty Registry. J Am Coll Cardiol. 1988;12:11491155. Sinclair IN, McCabe CH, Sipperly ME, Baim DS. Predictors, therapeutic options and long-term outcome of abrupt reclosure. Am J Cardiol. 1988;61:61G-66G.

7.

8.

Simpfendorfer C, Belardi J, Bellamy G, Galan K, Franco I, Hollman J. Frequency, management and follow-up of patients with acute coronary occlusions after percutaneous transluminal coronary angioplasty. Am J Cardiol. 1987;59:267-269. Kuntz RE, Piana R, Pomerantz RM, et al. Changing incidence and management of abrupt closure following coronary intervention in the new device era. Cathet Cardiovasc Diagn. 1992;27:183190.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

Mayo Clin Proc, December 2003, Vol 78

9. 10.

11. 12. 13.

14.

15.

16.

17. 18. 19.

20. 21.

22. 23.

24. 25.

26. 27.

Mabin TA, Holmes DR Jr, Smith HC, et al. Intracoronary thrombus: role in coronary occlusion complicating percutaneous transluminal coronary angioplasty. J Am Coll Cardiol. 1985;5:198-202. Sugrue DD, Holmes DR Jr, Smith HC, et al. Coronary artery thrombus as a risk factor for acute vessel occlusion during percutaneous transluminal coronary angioplasty: improving results. Br Heart J. 1986;56:62-66. Reeder GS, Bryant SC, Suman VJ, Holmes DR Jr. Intracoronary thrombus: still a risk factor for PTCA failure? Cathet Cardiovasc Diagn. 1995;34:191-195. Singh M, Berger PB, Ting HH, et al. Influence of coronary thrombus on outcome of percutaneous coronary angioplasty in the current era (the Mayo Clinic experience). Am J Cardiol. 2001;88:1091-1096. Singh M, Reeder GS, Ohman EM, et al. Does the presence of thrombus seen on a coronary angiogram affect the outcome after percutaneous coronary angioplasty? an Angiographic Trials Pool data experience. J Am Coll Cardiol. 2001;38:624-630. Detre KM, Holmes DR Jr, Holubkov R, et al, Coinvestigators of the National Heart, Lung, and Blood Institute’s Percutaneous Transluminal Coronary Angioplasty Registry. Incidence and consequences of periprocedural occlusion: the 1985-1986 National Heart, Lung, and Blood Institute Percutaneous Transluminal Coronary Angioplasty Registry. Circulation. 1990;82:739-750. de Feyter PJ, van den Brand M, Laarman GJ, et al. Acute coronary artery occlusion during and after percutaneous transluminal coronary angioplasty: frequency, prediction, clinical course, management, and follow-up. Circulation. 1991;83:927-936. Holmes DR Jr, Vlietstra RE, Smith HC, et al. Restenosis after percutaneous transluminal coronary angioplasty (PTCA): a report from the PTCA Registry of the National Heart, Lung, and Blood Institute. Am J Cardiol. 1984;53:77C-81C. Forrester JS, Fishbein M, Helfant R, Fagin J. A paradigm for restenosis based on cell biology: clues for the development of new preventive therapies. J Am Coll Cardiol. 1991;17:758-769. Faxon DP, ed. Restenosis: A Guide to Therapy. London, England: Martin Dunitz; 2001. Hermans WR, Rensing BJ, Foley DP, et al, Multicenter European Research trial with Cilazapril after Angioplasty to prevent Transluminal coronary Obstruction and Restenosis (MERCATOR) study group. Patient, lesion, and procedural variables as risk factors for luminal re-narrowing after successful coronary angioplasty: a quantitative analysis in 653 patients with 778 lesions. J Cardiovasc Pharmacol. 1993;22(suppl 4):S45-S57. Weintraub WS, Kosinski AS, Brown CL III, King SB III. Can restenosis after coronary angioplasty be predicted from clinical variables? J Am Coll Cardiol. 1993;21:6-14. Mercado N, Boersma E, Wijns W, et al. Clinical and quantitative coronary angiographic predictors of coronary restenosis: a comparative analysis from the balloon-to-stent era. J Am Coll Cardiol. 2001;38:645-652. Kastrati A, Schomig A, Elezi S, et al. Predictive factors of restenosis after coronary stent placement. J Am Coll Cardiol. 1997; 30:1428-1436. Hausleiter J, Kastrati A, Mehilli J, et al. Predictive factors for early cardiac events and angiographic restenosis after coronary stent placement in small coronary arteries. J Am Coll Cardiol. 2002;40: 882-889. Cutlip DE, Chauhan MS, Baim DS, et al. Clinical restenosis after coronary stenting: perspectives from multicenter clinical trials. J Am Coll Cardiol. 2002;40:2082-2089. Laskey WK, Kimmel S, Krone RJ. Contemporary trends in coronary intervention: a report from the Registry of the Society for Cardiac Angiography and Interventions. Catheter Cardiovasc Interv. 2000;49:19-22. Al Suwaidi J, Berger PB, Holmes DR Jr. Coronary artery stents. JAMA. 2000;284:1828-1836. Rankin JM, Spinelli JJ, Carere RG, et al. Improved clinical outcome after widespread use of coronary-artery stenting in Canada. N Engl J Med. 1999;341:1957-1965.

Risk Stratification and Interventional Cardiology

28.

29.

30.

31.

32.

33.

34.

35. 36.

37.

38.

39. 40.

41.

42.

43. 44.

1517

Kimmel SE, Localio AR, Krone RJ, Laskey WK, Registry Committee of the Society for Cardiac Angiography and Interventions. The effects of contemporary use of coronary stents on in-hospital mortality. J Am Coll Cardiol. 2001;37:499-504. Williams DO, Holubkov R, Yeh W, et al, Coinvestigators. Percutaneous coronary intervention in the current era compared with 19851986: the National Heart, Lung, and Blood Institute Registries. Circulation. 2000;102:2945-2951. McGrath PD, Malenka DJ, Wennberg DE, et al, Northern New England Cardiovascular Disease Study Group. Changing outcomes in percutaneous coronary interventions: a study of 34,752 procedures in northern New England, 1990 to 1997. J Am Coll Cardiol. 1999;34:674-680. Anderson HV, Shaw RE, Brindis RG, et al, ACC-NCDR. 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. Steinhubl SR, Berger PB, Mann JT III, et al, CREDO Investigators. Early and sustained dual oral antiplatelet therapy following percutaneous coronary intervention: a randomized controlled trial [published correction appears in JAMA. 2003;289:987]. JAMA. 2002; 288:2411-2420. Bertrand ME, Legrand V, Boland J, et al. Randomized multicenter comparison of conventional anticoagulation versus antiplatelet therapy in unplanned and elective coronary stenting: the Full Anticoagulation Versus Aspirin and Ticlopidine (FANTASTIC) Study. Circulation. 1998;98:1597-1603. Leon MB, Baim DS, Popma JJ, et al, Stent Anticoagulation Restenosis Study Investigators. A clinical trial comparing three antithrombotic-drug regimens after coronary-artery stenting. N Engl J Med. 1998;339:1665-1671. Schomig A, Neumann FJ, Kastrati A, et al. A randomized comparison of antiplatelet and anticoagulant therapy after the placement of coronary-artery stents. N Engl J Med. 1996;334:1084-1089. Topol EJ, Moliterno DJ, Herrmann HC, et al, TARGET Investigators. Comparison of two platelet glycoprotein IIb/IIIa inhibitors, tirofiban and abciximab, for the prevention of ischemic events with percutaneous coronary revascularization. N Engl J Med. 2001;344: 1888-1894. EPISTENT Investigators. Randomised placebo-controlled and balloon-angioplasty-controlled trial to assess safety of coronary stenting with use of platelet glycoprotein-IIb/IIIa blockade. Lancet. 1998;352:87-92. ESPRIT Investigators. Novel dosing regimen of eptifibatide in planned coronary stent implantation (ESPRIT): a randomised, placebo-controlled trial [published correction appears in Lancet. 2001;357:1370]. Lancet. 2000;356:2037-2044. Holmes DR Jr, Savage M, LaBlanche JM, et al. Results of Prevention of REStenosis with Tranilast and its Outcomes (PRESTO) trial. Circulation. 2002;106:1243-1250. Regar E, Serruys PW, Bode C, et al, RAVEL Study Group. Angiographic findings of the multicenter Randomized Study With the Sirolimus-Eluting Bx Velocity Balloon-Expandable Stent (RAVEL): sirolimus-eluting stents inhibit restenosis irrespective of the vessel size. Circulation. 2002;106:1949-1956. Morice MC, Serruys PW, Sousa JE, et al, RAVEL Study Group. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med. 2002;346:17731780. Leon MB, Moses JW, Popma JJ, Kuntz RE, Fitzgerald P, SIRIUS Investigators. SIRIUS: a US, multicenter, randomized, double blind study of the sirolimus-eluting stent in de novo native coronary lesions. N Engl J Med. In press. Orford JL, Lennon R, Melby S, et al. Frequency and correlates of coronary stent thrombosis in the modern era: analysis of a single center registry. J Am Coll Cardiol. 2002;40:1567-1572. Cutlip DE, Baim DS, Ho KK, et al. Stent thrombosis in the modern era: a pooled analysis of multicenter coronary stent clinical trials. Circulation. 2001;103:1967-1971.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.

1518

45. 46. 47.

48. 49.

50.

51.

52.

53.

54.

55.

56.

57.

Risk Stratification and Interventional Cardiology

Bhatt DL, Bertrand ME, Berger PB, et al. Meta-analysis of randomized and registry comparisons of ticlopidine with clopidogrel after stenting. J Am Coll Cardiol. 2002;39:9-14. Berglund U, Richter A. Clopidogrel treatment before percutaneous coronary intervention reduces adverse cardiac events. J Invasive Cardiol. 2002;14:243-246. Seyfarth HJ, Koksch M, Roethig G, et al. Effect of 300- and 450mg clopidogrel loading doses on membrane and soluble P-selectin in patients undergoing coronary stent implantation. Am Heart J. 2002;143:118-123. Leon MB, Teirstein PS, Moses JW, et al. Localized intracoronary gamma-radiation therapy to inhibit the recurrence of restenosis after stenting. N Engl J Med. 2001;344:250-256. Marks DS, Mensah GA, Kennard ED, Detre K, Holmes DR Jr. Race, baseline characteristics, and clinical outcomes after coronary intervention: the New Approaches in Coronary Interventions (NACI) registry. Am Heart J. 2000;140:162-169. Peterson ED, Lansky AJ, Anstrom KJ, et al, National Cardiovascular Network. Evolving trends in interventional device use and outcomes: results from the National Cardiovascular Network Database. Am Heart J. 2000;139:198-207. Seshadri N, Whitlow PL, Acharya N, Houghtaling P, Blackstone EH, Ellis SG. Emergency coronary artery bypass surgery in the contemporary percutaneous coronary intervention era. Circulation. 2002;106:2346-2350. Freeman RV, O’Donnell M, Share D, et al, Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2). Nephropathy requiring dialysis after percutaneous coronary intervention and the critical role of an adjusted contrast dose. Am J Cardiol. 2002;90: 1068-1073. Solomon R, Werner C, Mann D, D’Elia J, Silva P. Effects of saline, mannitol, and furosemide to prevent acute decreases in renal function induced by radiocontrast agents. N Engl J Med. 1994;331: 1416-1420. Stevens MA, McCullough PA, Tobin KJ, et al. A prospective randomized trial of prevention measures in patients at high risk for contrast nephropathy: results of the P.R.I.N.C.E. Study. J Am Coll Cardiol. 1999;33:403-411. Tepel M, van der Giet M, Schwarzfeld C, Laufer U, Liermann D, Zidek W. Prevention of radiographic-contrast-agent-induced reductions in renal function by acetylcysteine. N Engl J Med. 2000; 343:180-184. Abizaid AS, Clark CE, Mintz GS, et al. Effects of dopamine and aminophylline on contrast-induced acute renal failure after coronary angioplasty in patients with preexisting renal insufficiency. Am J Cardiol. 1999;83:260-263. Rich MW, Crecelius CA. Incidence, risk factors, and clinical course of acute renal insufficiency after cardiac catheterization in

Mayo Clin Proc, December 2003, Vol 78

58. 59.

60. 61.

62.

63. 64.

65.

66.

67. 68. 69. 70.

patients 70 years of age or older: a prospective study. Arch Intern Med. 1990;150:1237-1242. Rihal CS, Textor SC, Grill DE, et al. Incidence and prognostic importance of acute renal failure after percutaneous coronary intervention. Circulation. 2002;105:2259-2264. Ellis SG, Weintraub W, Holmes D, Shaw R, Block PC, King SB III. Relation of operator volume and experience to procedural outcome of percutaneous coronary revascularization at hospitals with high interventional volumes. Circulation. 1997;95:2479-2484. Holmes DR Jr, Berger PB, Garratt KN, et al. Application of the New York State PTCA mortality model in patients undergoing stent implantation. Circulation. 2000;102:517-522. Moscucci M, Kline-Rogers E, Share D, et al, Blue Cross Blue Shield of Michigan Cardiovascular Consortium. Simple bedside additive tool for prediction of in-hospital mortality after percutaneous coronary interventions. Circulation. 2001;104:263-268. O’Connor GT, Malenka DJ, Quinton H, et al, Northern New England Cardiovascular Disease Study Group. Multivariate prediction of in-hospital mortality after percutaneous coronary interventions in 1994-1996. J Am Coll Cardiol. 1999;34:681-691. Rihal CS, Grill DE, Bell MR, Berger PB, Garratt KN, Holmes DR Jr. Prediction of death after percutaneous coronary interventional procedures. Am Heart J. 2000;139:1032-1038. Shaw RE, Anderson HV, Brindis RG, et al, ACC-NCDR. Development of a risk adjustment mortality model using the American College of Cardiology-National Cardiovascular Data Registry (ACC-NCDR) experience: 1998-2000. J Am Coll Cardiol. 2002; 39:1104-1112. Singh M, Lennon RJ, Holmes DR Jr, Bell MR, Rihal CS. Correlates of procedural complications and a simple integer risk score for percutaneous coronary intervention. J Am Coll Cardiol. 2002;40: 387-393. Holmes DR, Selzer F, Johnston JM, et al. Modeling and risk prediction in the current era of interventional cardiology: a report from the National Heart, Lung, and Blood Institute Dynamic Registry. Circulation. 2003;107:1871-1876. Hasdai D, Holmes DR Jr, Califf RM, et al, GUSTO-I Investigators. Cardiogenic shock complicating acute myocardial infarction: predictors of death. Am Heart J. 1999;138:21-31. Holmes DR Jr, Bates ER, Kleiman NS, et al, GUSTO-I Investigators. Contemporary reperfusion therapy for cardiogenic shock: the GUSTO-I trial experience. J Am Coll Cardiol. 1995;26:668-674. Hochman JS, Sleeper LA, Webb JG, et al, SHOCK Investigators. Early revascularization in acute myocardial infarction complicated by cardiogenic shock. N Engl J Med. 1999;341:625-634. Hasdai D, Califf RM, Thompson TD, et al. Predictors of cardiogenic shock after thrombolytic therapy for acute myocardial infarction. J Am Coll Cardiol. 2000;35:136-143.

For personal use. Mass reproduce only with permission from Mayo Clinic Proceedings.