Highlights from the American Heart Association Annual Scientific Sessions 2000: November 12 to 15, 2000☆☆☆

Study: 

Earlier and more effective risk stratification in chest pain units using near-patient multimarker testing: results of the Chest Pain Evaluation by CK-MB, Myoglobin, and Troponin I study.

Presenter: 

Dr L. Kristin Newby, Duke Clinical Research Institute, Durham, NC

Background: 

The use of cardiac markers, primarily creatine kinase-MB (CK-MB), a cardiac troponin (I[cTnI] or T[cTnT), and in some cases myoglobin, has become standard to risk-stratify chest pain unit patients undergoing evaluation of symptoms of myocardial ischemia. However, each marker has a different specificity for myocardial injury, the release and clearance characteristics of each differ, and patients are seen for evaluation with varying durations of chest pain.1 The CHECKMATE study was designed to evaluate quantitative bedside measurement of combinations of markers with different time-to-positivity characteristics (myoglobin, CK-MB, and cTnI) for risk-stratifying patients with chest pain without ST-segment elevation. The study had two main goals: first, to evaluate the relation of the results of such multimarker strategies with 30-day clinical outcomes and, second, to examine the effect of bedside multimarker testing on time to detection of positive results.

The study enrolled 1005 patients with possible myocardial ischemia from the emergency department of 6 US hospitals that also had functional CPUs. Patients were excluded if the initial electrocardiogram (ECG) showed ST-segment elevation or left bundle-branch block (LBBB) that led to acute reperfusion therapy or its consideration. Blood samples were obtained at enrollment (baseline), 3 hours, and 6 hours for analysis with the (Deerfield, Ill) Dade-Behring Stratus CS STAT near-patient instrument. Additional samples at 9 to 12 and 16 to 24 hours were collected from patients remaining in the hospital at those times. Two multimarker strategies (MMS) were prospectively defined: MMS-1 (all 3 markers) and MMS-2 (CK-MB and cTnI). Blood also was collected according to the site’s usual CPU “rule-out” protocol and analyzed by the hospital’s laboratory. A single-marker local laboratory strategy (LL) was defined with the site-laboratory result for CK-MB because at all sites it was considered the primary marker necessary to diagnose myocardial infarction (MI).

Results: 

The final analysis included data for 968 patients for LL testing (96%), 953 for MMS-1 (95%), and 955 for MMS-2 (95%). At baseline 18.9% of patients were positive by MMS-1, 14.3% by MMS-2, and 5.2% by LL (P = .001, all comparisons). By serial testing 23.9% of patients were positive by MMS-1, 18.8% by MMS-2, and 8.8% by LL (P = .001, all comparisons). The median time to positivity was slightly but not significantly shorter for MMS-1 (2.5 hours) than MMS-2 (2.8 hours), and both were shorter than with LL testing (3.4 hours, P = .023 vs MMS-1 and P = .026 vs MMS-2). The median age of the CHECKMATE study population was 52 years and nearly one half was female. For all testing strategies, patients who were positive were older, more often male, and more likely to have diabetes, hypertension, prior anginal symptoms or MI, congestive heart failure, and chronic renal insufficiency. There was a strong relationship between baseline MMS status and the rate of death or MI at 30 days. For MMS-1, the rate was 18.8% among positive patients versus 3.0% among negative patients (P = .001). Although slightly fewer patients were positive by MMS-2, the relationship of its status to 30-day death or MI was similar (positive, 21.9% vs negative, 3.2%; P = .001). MMS-1 was a better discriminator for 30-day mortality (2.0% among positive vs 0.0% among negative patients, P = .007) than either MMS-2 (positive, 1.8% vs negative, 0.2%; P = .055) or LL (positive, 0.0% vs negative, 0.5%; P = 1.000) results. Results were similar with serial testing. At baseline, MMS status (MMS-1 or MMS-2) was the strongest independent predictor of death or MI at 30 days in multivariable analyses. Prior MI, female sex, and abnormal ECG changes also were significant independent baseline predictors in both models. The overall C-index values of both models (0.804 for MMS-1 and 0.823 for MMS-2) were >0.8, indicating good ability to predict the outcomes of interest.

Interpretation: 

The use of bedside quantitative multimarker strategies for cardiac marker testing in CPUs provided clear benefits over laboratory-based single-marker testing: they improved 30-day risk stratification and identified high-risk patients earlier among those presenting to CPUs. Significantly more patients were identified as high risk when myoglobin was incorporated in the MMS versus a strategy including only CK-MB and cTnI or single-marker LL testing.

Study: 

The promise of combined low-molecular-weight heparin and platelet glycoprotein IIb/IIIa inhibition—Results from Platelet IIB/IIIA Antagonist for the Reduction of Acute Coronary Syndome Events in a Global Organization Network (PARAGON) B

Presenter: 

Dr Debabrata Mukherjee, Cleveland Clinic Foundation, Cleveland, Ohio

Background: 

Heparin has been shown to be beneficial in acute coronary syndromes. Low-molecular-weight heparin (LMWH) has a more predictable anticoagulant effect than unfractionated heparin, is easier to administer, and does not require monitoring. The Thrombolysis in Myocardial Infarction (TIMI) 11B and Efficacy Safety Subcutaneous Enoxaparin in Non-Q-Wave Coronary Events (ESSENCE) trials have shown that LMWH (enoxaparin) reduces by 20% the risk of death and ischemic events. Similarly, the Fragmin During Instability in Coronary Artery Disease (FRISC) II trial showed that dalteparin statistically significantly reduced death and reinfarction. Glycoprotein IIb/IIIa inhibitors have also been shown to be beneficial among patients with acute coronary syndromes treated with aspirin and unfractionated heparin. Little data exist on the combination of LMWH and glycoprotein IIb/IIIa inhibitors. The objective of this study was to assess the potential beneficial treatment effect of the combination of LMWH and glycoprotein IIb/IIIa inhibitors in the PARAGON B trial. PARAGON B randomized 5184 patients with acute coronary syndromes to receive either lamifiban or placebo in addition to standard therapy. Among the lamifiban-assigned patients, 410 were initially treated with LMWH rather than unfractionated heparin on the basis of physician discretion. Patient characteristics in the LMWH and unfractionated heparin groups were similar with respect to age, sex, hypertension, and diabetes. More patients in the LMWH group had prior MI and heart failure.

Results: 

No difference was seen in the primary end points of death, reinfarction, and severe recurrent ischemia for unfractionated heparin compared with LMWH (12.2% vs 10.3%, P = .13); likewise no difference was seen in the combination of death or reinfarction at 30 days (11% vs 9%, P = .11) or 6 months (15.5% vs 12.8%, P = .08). However, there was a statistically significant reduction in revascularization in patients treated with the combination of LMWH plus lamifiban compared with unfractionated heparin plus lamifiban (44.3% vs 33.7%, P = .001). Stroke (1.1% vs 0.7%) and major bleeding (1.6% vs 1.5%) were comparable in the two heparin groups. A propensity analysis was performed to assess the efficacy of the LMWH in all patients and after adjustment for differences in baseline characteristics. There was a statistically significant benefit with LMWH and lamifiban for revascularization at 30 days (P < .001), a trend toward benefit on stroke (P = .07), and no difference for the triple combined end point or for death and reinfarction at 30 days. The limitations of this study are those related to its retrospective nature, the fact that the use of LMWH was not randomized, and the small sample size.

Interpretation: 

In the PARAGON B study, the combination of lamifiban plus LMWH was associated with less revascularization at 30 days and similar risk of bleeding as the combination of unfractionated heparin plus lamifiban. These findings are promising and more data on the combination of LMWH and glycoprotein IIb/IIIa inhibitors will be available soon from ongoing trials.

Study: 

Antithrombotic combination using tirofiban and enoxaparin: The Analysis of Coronary Ultrasound Thrombolysis Endpoints (ACUTE) II study

Presenter: 

Dr Marc Cohen, Hahnemann University, Philadelphia, Pa

Background: 

The combination of the glycoprotein (GP) IIb/IIIa inhibitor tirofiban and unfractionated heparin (UFH) has been shown to reduce death, myocardial infarction, and recurrent ischemia by 32% in the Platelet Receptor Inhibition for Ischemic Syndrome Management in Patients Limited by Unstable Signs and Symptoms (PRISM-PLUS) trial. Likewise in the ESSENCE study, enoxaparin, a low-molecular-weight heparin (LMWH), reduced by 16% the occurrence of the same combined end point. The ACUTE I study showed that enoxaparin plus tirofiban has less variability in inhibiting platelet aggregation than does UFH plus tirofiban. The hypothesis of the ACUTE II trial was that the combination of LMWH enoxaparin with tirofiban added to aspirin is safe and well tolerated. Patients were eligible if they had unstable angina or non-ST-elevation myocardial infarction within 24 hours of chest pain or evidence of myocardial infarction by ECG or cardiac enzymes. A total of 525 patients were randomized in a double-blind manner to receive either tirofiban plus enoxaparin or tirofiban plus UFH in a ratio of 3 to 2. Tirofiban was administered as a loading infusion (0.4 mg/kg/min for 30 minutes) followed by a maintenance infusion of 0.2 mg/kg/min for 47.5 to 107.5 hours. UFH was administered as a 5000-U bolus and maintained initially at 1000 U/h titrated up to an activated partial thromboplastin time of 1.5 to 2.5 times the normal range for a duration of 24 to 96 hours. Coronary angiography followed by percutaneous intervention was permitted after 24 hours. The primary objective was to determine the rates of TIMI bleeding, the need for transfusion, and the incidence of severe thrombocytopenia in each arm. Bleeding incidents were assessed until 24 hours after study drug completion. Clinical outcomes such as death, another myocardial infarction, and revascularization were also assessed through 30 days.

Results: 

Baseline characteristics in both groups were well matched, with approximately half the patients having non-ST-elevation myocardial infarction. Regarding the safety end points, no differences were seen between enoxaparin plus tirofiban and UFH plus tirofiban, respectively, regarding major bleeding (0.6% vs 0.5%), minor bleeding (2.9% vs 3.3%), or blood loss with no evident site of bleeding (1.0% vs 1.4%). Likewise, both groups had similar rates of transfusion (2.2% vs 2.9%) and incidence of thrombocytopenia (0.3% vs 0.5%). Both groups had similar mortality at 30 days (enoxaparin 2.2% vs UFH 2.4%), similar rates of reinfarction (7.0% vs 7.6%), and the combination (9.2% vs 10%). However, there was a trend toward benefit on recurrent ischemia with enoxaparin compared with UFH (11.7% vs 16.7%).

Interpretation: 

This trial provides the first safety data regarding the combination of enoxaparin plus tirofiban in patients with unstable angina and non-ST-elevation myocardial infarction. These findings support the hypothesis that enoxaparin could be used in exchange for UFH with tirofiban with a comparable low rate of bleeding events. Larger trials would be needed to establish the relative efficacy outcomes for these combinations.

Study: 

Treat Angina With Aggrastat and Determine Cost of Therapy With an Invasive or Conservative Strategy (TACTICS)–TIMI 18

Presenter: 

Dr Christopher Cannon, Brigham and Women’s Hospital, Boston, Mass

Background: 

Previous randomized clinical trials have produced mixed results in terms of whether patients benefitted more with an early invasive strategy or a more conservative strategy after presentation with an acute coronary syndrome (ACS) without ST elevation. Advances in therapeutic options including glycoprotein (GP) IIb/IIIa receptor blockers and coronary stents have improved outcomes in both groups of patients. A total of 2220 patients with unstable angina/non-ST-elevation myocardial infarction were treated initially with aspirin, intravenous unfractionated heparin, and tirofiban. Patients were then randomized into one of two treatment arms: (1) early “invasive” strategy consisting of cardiac catheterization within 4 to 48 hours after hospital admission, with medical therapy, angioplasty, or bypass surgery carried out on the basis of the results of the catheterization or (2) early conservative strategy consisting of initial medical management with catheterization reserved for patients with a positive exercise test or recurrent ischemia on medical therapy. The primary end point was a composite of death, myocardial infarction, or rehospitalization for ACS at 6 months.

Results: 

There was a statistically significant reduction in the composite (primary) end point of death/myocardial infarction/rehospitalization for ACS at 6 months for patients randomized to the early invasive strategy. In the group of patients who were troponin positive at presentation the benefit was even more pronounced, with a reduction in the primary end point from 24% to 14.3%, an almost 10% absolute reduction (Table I).

Table I. Cardiac events at 6 months

	End point	Conservative (n = 1106)	Invasive (n = 1114)	Odds ratio	Statistical significance
	Death/myocardial infarction/rehospitalization  for ACS (%)	19.4	15.9	0.78	P = .025
	Death/myocardial infarction (%)	9.5	7.3	0.74	P = .048
	Death (%)	3.5	3.3	0.93	P = .74
	Myocardial infarction (%)	6.9	4.8	0.67	P = .029
	Rehospitalization for ACS (%)	13.7	11.0	0.78	P = .054

In patients with ACS without ST elevation and treated with aspirin, heparin, and the GP IIb/IIIa inhibitor tirofiban, an early invasive strategy resulted in a significant reduction in the composite end point of death, myocardial infarction, and rehospitalization for ACS at 6 months compared with a more conservative approach of initial medical therapy. This benefit was amplified in patients with a positive troponin at presentation.

Interpretation: 

Previous studies of aggressive versus conservative management of these patients did not incorporate current state-of-the-art therapies such as GP IIb/IIIa inhibitors and coronary stents. The results of TACTICS-TIMI 18 provide compelling evidence that the clinical management of ACS without ST elevation should include broader application of an early invasive strategy with GP IIb/IIIa inhibitor use.

Study: 

Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering (MIRACL)

Presenters: 

Drs Anders Olsson, Linköping University Hospital, Linköping, Sweden, and Gregory Schwartz, University of Colorado Health Sciences Center, Denver, Colo

Background: 

Lipid lowering with selected HMG CoA reductase inhibitors is effective at improving recurrent clinical events among patients with elevated lipid levels. This reduction in recurrent cardiovascular events was seen in clinically stable patients who were enrolled in the Scandinavian Simvastatin Survival Study (4S), Calcium Antagonist in Reperfusion (CARE), and Long-Term Intervention With Pravastatin in Ischemic Disease (LIPID) trials at least 3 months after the qualifying event. The value of early adjunctive statin use after an acute coronary syndrome (ACS) is unknown. MIRACL was designed to evaluate the effect of early initiation of atorvastatin after a non-ST-segment elevation ACS.

Results: 

MIRACL was an international, randomized, double-blind, placebo-controlled study. Eligible patients had a non-ST-segment elevation ACS and were randomized to receive atorvastatin 80 mg daily or placebo between 24 and 96 hours after hospital admission. No treatment algorithm for the medical management of the non-ST-elevation ACS was stipulated. The principal exclusions for enrollment in MIRACL included planned coronary revascularization and current lipid-lowering therapy. Of note, patients were not required to have hypercholesterolemia. The primary composite end point of the study was time to death or the first occurrence of nonfatal myocardial infarction, resuscitated cardiac arrest, or recurrent symptomatic myocardial ischemia requiring hospitalization in the 16-week period after randomization. MIRACL enrolled 3086 patients between May 1997 and September 1999. The mean total cholesterol level at study entry was 206 mg/dL with a mean low-density lipoprotein level of 124 mg/dL. Assignment to the atorvastatin arm was associated with a decrease in the incidence of the primary composite end point compared with placebo (14.8% vs 17.4%, respectively, RRR 16%, P = .048). Twelve patients were lost to follow-up (8 in the atorvastatin arm, 4 in the placebo arm). The statistical significance of the composite end point was primarily due to a 26% decrease in the rate of worsening angina (P = .02), but death and nonfatal myocardial infarction both trended in favor of atorvastatin. In an analysis that was not prespecified, assignment to atorvastatin was associated with a 50% reduction in the risk of stroke over the 16-week interval (12 vs 24 events, P = .045). No important problems were observed regarding safety and tolerability.

Early intensive lipid-lowering treatment with atorvastatin, initiated during the acute phase of unstable angina or non-Q-wave myocardial infarction, significantly reduced early recurrent ischemic events. A benefit was observed in a population that included low to normal baseline low-density lipoprotein cholesterol levels. Treatment was safe and well tolerated.

Interpretation: 

This was the first trial to evaluate the effects of a statin in the immediate post-ACS period. Potential concerns regarding the safety of initiating aggressive lipid lowering during or immediately after an episode of ACS prompted this evaluation. These concerns were based on the theoretic risk of plaque destabilization resulting from acute alterations in lipid composition of the plaque after initiation of aggressive lipid-lowering therapy. The results of MIRACL are reassuring that in the short to intermediate term after ACS there was no signal for increased instability.

The primary composite end point in MIRACL barely reached statistical significance (P < .049 because of a .001 statistical penalty). Hospitalization for worsening angina with new objective evidence of ischemia was the principal end point that led to a statistically significant composite. However, no difference in coronary revascularization was seen between the two arms as a result of the rehospitalization. This has led many to question the robustness of the end point and thus the broad incorporation of statins after ACS regardless of lipid levels, an issue that carries large economic consequences. Additionally, because of the borderline statistical value in MIRACL, events in the patients lost to follow-up could have had a significant negative impact.

It is unclear whether the patient population enrolled in MIRACL is a true representation of the current non-ST-elevation ACS patient population, especially given the findings of TACTICS. MIRACL excluded patients with planned coronary revascularization procedures. Furthermore, the use of GP IIb/IIIa inhibitors, a proved therapy in non-ST-elevation ACS, was very limited (only 1% of patients received one) in MIRACL.

Several cardiovascular clinical trial thought leaders have emphasized the importance of collecting additional evidence in this patient population to further explore the modest early benefit in recurrent ischemic events seen in MIRACL. Another clinical trial currently evaluating the early aggressive initiation of statin therapy following ACS is the A to Z (Aggrastat to Zocor) trial. There are important differences in the trial design between MIRACL and A to Z. Patients enrolled in A to Z will have an extended follow-up period, more than 1 year compared with 16 weeks in MIRACL. The A to Z trial includes the full spectrum of ACS because it will evaluate patients who are optimally treated for an ST-elevation myocardial infarction, a cohort not evaluated in MIRACL. Patients treated with percutaneous coronary revascularization will be included in the A to Z trial, whereas MIRACL excluded patients with planned coronary revascularization. The medical treatment algorithm is more uniform and optimal in A to Z because it includes the use of the GP IIb/IIIa inhibitor tirofiban, an agent with proved benefit in the non-ST-elevation ACS populations of A to Z and MIRACL. The results of A to Z should provide further clarity regarding the early use of statin therapy after ACS.

Study: 

Do Tirofiban and ReoPro Give Similar Efficacy Outcomes Trial (TARGET)

Presenter: 

Dr Eric J. Topol, Cleveland Clinic Foundation, Cleveland, Ohio

Background: 

Glycoprotein IIb/IIIa inhibition improves clinical outcomes when administered during percutaneous coronary intervention (PCI). Clinical investigation with abciximab, and more recently with eptifibatide, has demonstrated a statistically significant benefit in randomized, placebo-controlled trials. This benefit during PCI is reflected in current American College of Cardiology/American Heart Association guidelines. In the Randomized Efficacy Study of Tirofiban for Outcomes and Restenosis (RESTORE) trial, tirofiban, a small-molecule glycoprotein IIb/IIIa inhibitor, failed to demonstrate a significant difference in the primary composite end point during PCI compared with placebo at 30 days despite a significant difference at 7 days. However, tirofiban has a favorable price profile compared with abciximab. In light of the proved benefits of abciximab, another placebo group comparison to tirofiban during PCI would be unethical. Therefore TARGET was designed as a noninferiority study to compare the efficacy of tirofiban and abciximab when initiated at the start of coronary stenting procedures in patients with either stable coronary disease or unstable angina.

Results: 

TARGET randomized 4812 patients at 150 sites in 18 countries. The study was conducted in a double-blind, double-dummy fashion. The dose and administration of tirofiban followed the investigational use in RESTORE. Patients received tirofiban as an intravenous bolus (10 μg/kg) followed by an 18- to 24-hour infusion (0.15 μg/kg/min) or abciximab as an intravenous bolus (0.25 μg/kg) followed by a 12-hour infusion (0.125 μg/kg/min to a maximum of 10 μg/min). All patients received standard of care adjunctive PCI therapies including aspirin, heparin, and clopidogrel. The primary composite end point was the incidence of death, nonfatal myocardial infarction, and urgent coronary revascularization at 30 days. The rate of the composite end point was 7.6% with tirofiban compared with 6.0% for abciximab (odds ratio 1.26, P = .037) at 30 days. Secondary end points were the individual events in the composite at 30 days. Death rates were 0.5% for tirofiban and 0.4% for abciximab. The rate of nonfatal myocardial infarction favored abciximab (5.4% vs 6.9% with tirofiban, P = .04). The significant difference in nonfatal myocardial infarction emerged early in the periprocedural period. Urgent coronary revascularization was required in 0.9% of patients receiving tirofiban and in 0.7% receiving abciximab. Results were generally consistent across subgroups with the exception of an even greater abciximab benefit in patients with acute coronary syndromes (ACS), a subgroup analysis that was not prespecified. Both tirofiban and abciximab were generally well tolerated. Statistically significant differences were seen in TIMI minor bleeding (3.5% with tirofiban vs 5.6% with abciximab, P = .001) and thrombocytopenia (<100,000, 0.5% for tirofiban vs 2.4% with abciximab, P < .001). Thus abciximab is superior to tirofiban in coronary stenting procedures with a relative risk reduction in the combined rate of death, nonfatal myocardial infarction, and urgent coronary revascularization of 27%.

Interpretation: 

TARGET is the first comparator trial of glycoprotein IIb/IIIa inhibition during stent PCI, and its results raise more questions than they provide answers. The questions focus on the optimal dose strategy for tirofiban, potential key structural and nonplatelet property differences with abciximab, the significance of periprocedural myocardial infarction, and the ideal glycoprotein IIb/IIIa antagonist setting (ACS or PCI).

One potential explanation for the failure of tirofiban to demonstrate noninferiority may relate to the dose and administration. Tirofiban was dosed identically to RESTORE, a negative PCI trial compared with placebo at the primary 30-day end point. Further studies are needed to clarify the degree of platelet inhibition with the dose of tirofiban in TARGET and RESTORE. Additionally, the potential contribution of the nonplatelet effects of abciximab needs further evaluation.

The incidence of nonfatal myocardial infarction was the principal component of the primary composite end point that led to a statistically significant difference. This difference was noted as early as 8 hours after PCI, at the time of the first postprocedure enzyme collection. The rationale for its inclusion is the associated prognostic significance of periprocedural myocardial infarction. Although this subject is controversial, abciximab clearly reduced the degree of enzyme leak after PCI. Whether this translates to a mortality difference beyond 30 days remains to be seen.

The recently presented Global Utilization of Streptokinase and Tissue-Plasminogen Activator for Occluded Arteries (GUSTO-IV) ACS trial evaluated abciximab in the isolated setting of ACS (without PCI) and failed to demonstrate benefit over heparin alone. However, tirofiban and eptifibatide have data that show benefit in ACS. Now that abciximab has been shown to be superior to tirofiban during PCI, there may be different treatment algorithms depending on patient presentation. To compound the confusion, in an analysis that was not prespecified, abciximab demonstrated even further benefit compared with tirofiban in patients who had ACS. Only eptifibatide has demonstrated benefit in both ACS and PCI patient populations.

Another important question raised by the results of TARGET relates to abciximab and eptifibatide. Should we consider these antagonists comparable during coronary stenting? Certainly it is difficult to ignore the durable mortality benefit seen with abciximab. Six-month eptifibatide PCI data from the Efficacy Safety Prospective Randomized Ibopamine Trial (ESPRIT) will be available in the near future, but the clear answer lies in a head-to-head trial. Perhaps the most important question is, will a comparison trial ever be done?

Study: 

In-hospital and 30-day Economic and Clinical Outcomes with Platelet GP IIb/IIIa ReoPro vs Integrilin Therapy During Percutaneous Coronary Intervention (PRICE) Trial.

Presenter: 

Dr Frank V. Aguirre, Prairie Cardiovascular Consultants, Springfield, Ill

Background: 

Placebo-controlled trials have established the clinical efficacy and safety of the glycoprotein IIb/IIIa inhibitors in patients undergoing percutaneous coronary intervention. Despite their remarkable efficacy, however, these agents are used in about 50% of coronary interventions in the United States and far less in other parts of the world. In a significant proportion of these cases, the therapy is given in a bailout fashion, where it is reserved for only those patients having a complication in the catheterization laboratory. Perhaps the single most important issue constraining the more widespread use of these agents is their cost: abciximab (c7E3 Fab, ReoPro, Centocor, Malvern, Pa, and Eli Lilly, Indianapolis, Ind), a human-murine chimeric monoclonal Fab antibody fragment, costs around $1400 per treatment course. Eptifibatide (Integrilin, COR Therapeutics, South San Francisco, Calif, and Schering-Plough, Kenilworth, NJ), is a cyclic heptapeptide based on the Lys-Gly-Asp (KGD) amino acid sequence. This agent is a highly specific, competitive blocker of the glycoprotein IIb/IIIa complex. At a cost of around $400 per treatment course, it offers an attractive alternative to abciximab therapy. To date, however, the efficacy of these glycoprotein IIb/IIIa receptor blockers has not been directly compared in randomized clinical trials. Although a lower-priced drug is economically more attractive, greater utilization of resources over time would negate the early cost savings of an inferior drug.

The PRICE trial was a small, single-center, prospective, randomized, double-blind pilot study of 320 patients undergoing percutaneous coronary intervention (89% of the cases received stents) between April 1999 and January 2000. The purpose of the trial was to examine the in-hospital and 30-day clinical and economic outcomes among patients undergoing elective percutaneous coronary intervention, randomly assigned to receive abciximab or eptifibatide. The degree of platelet inhibition achieved with each agent was compared by the rapid platelet function assay RPFA (Ultegra). Patients were randomized to receive abciximab (n = 163) (bolus dose of 0.25 mg/kg and 12-hour infusion at 0.125 μg/kg/min) or eptifibatide (n = 157) (bolus dose 180 μg/kg followed by an infusion at 2 μg/kg/min for 18 to 24 hours). Patients with acute myocardial infarction within 48 hours or with unstable angina with electrocardiographic changes within 12 hours before the intervention were excluded from the trial, as were patients with old saphenous vein graft lesions.

Results: 

The primary end point of the study, total in-hospital costs based on an intention-to-treat analysis, was significantly lower for the eptifibatide-treated patients, $7207 vs $8268, P = .009. The difference in cost was accounted for entirely by the reduced pharmacy costs of eptifibatide compared with abciximab, $511 and $1561, respectively. At 6 months the mean cumulative cardiovascular costs were also reduced with eptifibatide ($9961 vs $11,075, P = .027). Clinical outcomes were similar in both groups, the in-hospital composite of death, myocardial infarction, and need for urgent revascularization occurring in 5.1% of the eptifibatide-treated patients and in 4.9% of those treated with abciximab (P = .84). The 30-day and 6-month rates for this composite were 6.3% vs 5.6% (P = .95) and 15.9% vs 15.3% (P = .99), respectively. The median percent platelet inhibition compared with baseline was 84% (78%, 90% [25th, 75th percentiles]) and 94% (90%, 99%) at drug discontinuation for abciximab- and eptifibatide-treated patients, respectively.

Interpretation: 

When used at doses similar to those tested in the Platelet IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy (PURSUIT) trial (180 μg/kg bolus followed by a 2 μg/kg/min infusion) and in patients undergoing nonurgent percutaneous coronary intervention, predominantly with stent implantation, eptifibatide was associated with lower in-hospital costs with no apparent differences in clinical outcomes. It should be pointed out, however, that the study was not adequately powered to detect a difference in the efficacy of these agents. Further studies, including long-term follow-up, will be required to confirm these findings.

Study: 

Comparison of long-term losartan versus captopril therapy on heart failure–related clinical outcomes and improvement in symptoms and quality of life: Evaluation of Losartan in the Elderly (ELITE) II

Presenter: 

Dr Marvin A. Konstam, New England Medical Center, Boston, Mass

Results: 

The nonsignificant difference in mortality among patients randomized to losartan compared with patients who received captopril in ELITE II extends to changes in New York Heart Association (NYHA) class and quality of life (QOL). Of the 1882 patients eligible for the QOL substudy, data on baseline and 1-year follow-up were available for 1540 (82%) patients; 804 patients (52%) received losartan and 736 (48%) received captopril. QOL was measured with the EQ5-D visual analog scale, with 0 being the worst and 100 being the best health state. At 1 year, 700 of 804 (87%) losartan patients and 643 of 736 (86%) captopril patients had survived. Patients who died were assigned a QOL score of 0. Although NYHA class and QOL among both groups of patients showed improvement at 1 year compared with baseline, there were no significant differences between the two patient groups.

Interpretation: 

At 1 year, there was no difference in NYHA class or QOL between patients who received losartan and those who received captopril. Although these results are not dramatic, they serve to increase anticipation for the QOL data from the recently reported Valsartan–Heart Failure (Val-HeFT) trial, which used another agent (valsartan) in the same class as losartan.

Study: 

Caution needed in interpreting the impact of dosing errors: A case study from Assessment of the Safety and Efficacy of a New Thrombolytic Agent (ASSENT) 2

Presenter: 

Dr Christopher B. Granger, Duke University Medical Center, Durham, NC

Background: 

Medication errors have recently been recognized as a major public health problem across the United States. Thrombolysis, with its narrow therapeutic window for dose, is a therapy where errors have the potential to lead to serious adverse consequences. This issue was first examined in the GUSTO-I trial. Medication errors occurred in 13.5% of patients treated with streptokinase and in 11.5% of those treated with tissue plasminogen activator (tPA). Importantly, mortality was significantly higher in patients with a medication error than in patients who received the correct dose. More recently, analysis of the Intravenous nPA for Treatment of Infarcting Myocardium Early (InTIME)-II study, which compared accelerated tPA with single-bolus lanoteplase, demonstrated a higher rate of dosing errors among the tPA group than the nPA group (7.3% vs 5.7%, P < .001). Again, mortality was higher among patients with medication errors than among those who received the correct dose (12.5% vs 5.9%, P < .001).

Critical to any discussion of this subject is the issue of cause and effect—the association between worse outcomes and incorrect dosing could be due to a direct effect of giving the wrong drug dose or to confounding factors, such as sicker patients being more likely to receive the wrong dose. This study sought to address this issue by analyzing the relationship between misdosing and adverse outcomes for active drug (tPA in this case) and for placebo in the ASSENT 2 trial: if the wrong dose of placebo was also associated with poor outcome, this would support the argument that incorrect dosing was a marker, rather than the cause, of poor outcome.

In ASSENT-2, almost 17,000 patients with an acute (within 6 hours of symptom onset) ST-segment-elevation myocardial infarction were randomized to either single-bolus weight-adjusted tenecteplase (TNK, 30 to 50 mg) with tPA placebo or to 90-minute weight-adjusted tPA infusion (<100 mg) with TNK placebo. Dosing errors were defined as giving more or less than the dose calculated on the basis of body weight.

Results: 

More patients assigned to TNK received the correct dose (96.4% vs 95.1% for tPA). The patients who received the incorrect doses of tPA or tPA placebo were older (67 vs 61 years), lighter (69 vs 78 kg), and more likely to be female (42% vs 22%) and Killip class IV (0.7% vs 0.3%). As expected, 30-day mortality rates were higher in patients who were underdosed (19.8% vs 5.4% for the correct dose) and overdosed (9.8% vs 5.4% for the correct dose) with tPA. The main finding of this analysis was that the wrong dose of placebo was associated with nearly identical increase in risk (23.8% for underdosed patients vs 5.4% for the correctly dosed patients and 10% for the overdosed patients vs 5.4% for the correctly dosed patients). Furthermore, rates of intracranial hemorrhage were increased in those receiving incorrect doses of study drug, with the highest rate occurring in patients who were underdosed (3.1% and 2.0%, tPA and tPA placebo patients, respectively, vs 0.9% in the correctly dosed patients). Dosing errors with TNK/TNK-placebo were not significantly associated with worse outcome.

Interpretation: 

In ASSENT 2, tPA dosing errors occurred in 5% of patients and were associated with considerably increased mortality rates. The finding that the relationship between worse outcome and incorrect dose was similar for patients treated with active drug and placebo suggests that the worse outcome of these patients was explained by confounding. The patients who were incorrectly dosed had higher risk features of older age, lighter body weight, and more often female gender. In ASSENT 2 the misdosing on average was only 5% to 10% off the correct dose, and therefore the direct impact of more serious errors in real practice may be much greater. These results highlight the need for the same methodologic rigor in analysis of the impact of dosing errors as in other research.

Study: 

Quality of life after balloon angioplasty or stenting for acute myocardial infarction: One-year results from the Stent Primary Angioplasty in Myocardial Infarction (STENT-PAMI) Trial

Presenter: 

Dr Stephane Rinfret, Beth Israel Deaconess Medical Center, Boston, Mass

Background: 

The 1-year results from the STENT-PAMI trial showed that patients with a myocardial infarction (MI) who were randomized to coronary stenting (CS) had a slightly higher mortality rate (5.8% vs 3.3%, P = .07) and a significantly lower target-vessel revascularization rate (10.6% vs 21%, P < .01) than patients randomized to balloon angioplasty (BA). A substudy, restricted to the 554 North American patients, was conducted to compare 1-year quality of life (QOL) after CS versus BA. QOL data were available for 509 (92%) patients. Nonrespondents had lower ejection fractions and higher rates of stroke/transient ischemic attack and were more likely to be discharged to a rehabilitation facility. The Short Form-36 and the Seattle Angina Questionnaire were used to measure QOL at 1 month, 6 months, and 1 year.

Results: 

Patients who underwent CS (n = 254) tended to be older, whereas BA patients (n = 255) had higher rates of chronic obstructive pulmonary disease. There were no differences in mortality or second myocardial infarction rates between stent and BA patients at 6 months; however, BA patients had a significantly higher rate of repeat revascularization (21.6% vs 14.6%, P = .05). Although this trend continued to 1 year, the difference was no longer statistically significant (37.7% vs 31.5%, P = .16). With respect to QOL, there were no differences between the two groups at 1 month. At 6 months stent patients reported reduced angina frequency, better disease perception, less pain, and better general health. These differences persisted even after adjustment for baseline characteristics. There were no statistically significant differences in QOL among the two groups at 1 year. Comparisons of 6-month QOL measures by repeat revascularization status revealed that patients who underwent repeat procedures had significantly lower disease and general health perceptions than those who did not undergo additional revascularization procedures.

Interpretation: 

Primary stenting after myocardial infarction appears to be associated with improved QOL at 6 months. The lower QOL among BA patients may be accounted for by the increased need for repeat revascularization and higher frequency of angina. This difference in QOL is not, however, sustained beyond 6 months.

Study: 

Obesity and survival in chronic heart failure

Presenter: 

Dr Stefan D. Anker, Charité Berlin/MDC Berlin-Buch, Germany

Background: 

A weight-loss phenomenon among patients with congestive heart failure (CHF), termed “cardiac cachexia,” has previously been associated with poor long-term prognosis. The influence of body weight (particularly obesity) on survival in CHF, independently of the presence of cardiac cachexia, remains poorly understood.

Results: 

This study included 589 consecutive CHF patients at a single clinical center. Cardiac cachexia (weight loss >7.5%) was identified in 62 patients and these patients were excluded from subsequent analyses. Among the remaining 527 patients, cumulative mortality was 13% after 1 year and 27% after 3 years. Patients were divided into quintiles on the basis of body mass index (BMI). There were no significant differences between the quintiles for age, New York Heart Association (NYHA) class, peak VO2, and ejection fraction. In univariate analyses high BMI was significantly associated with improved survival. Overall, the best survival was seen in the 4th quintile, which was statistically significantly better compared with quintiles 1 to 3 (Table II).

Table II. Quintiles of BMI and survival in 589 patients with CHF

		BMI Quintiles
		1	2	3	4	5
	BMI (kg · m–2)	<24	24-26	26-28	28-32	>32
	3-Year survival (%)	65.4	70.3	71.6	81.5	75.2
	Statistical significance*	P = .002	P = .04	P = .02	NA	P = .15
	*Versus group 4.

NA, Not available.

Among CHF patients without cardiac cachexia, increased BMI in the overweight/obese range is associated with better survival. High fat stores may indicate preserved metabolic efficiency or higher energetic reserves.

Interpretation: 

In the current presentation and in prior publications, these investigators have put forth an attractive hypothesis of biologic plausibility to explain the favorable effects of elevated BMI and higher fat stores, which are known to be directly related to increased total lipoprotein levels.1, 2, 3 Briefly, they propose that higher levels of cholesterol and lipoproteins may effectively modulate the inflammatory consequences of CHF decompensation (by binding and detoxifying bacterial endotoxin). Additionally, they propose that the daily physical work of heavier patients (“passive training”) results in a slightly improved metabolic capacity as seen in CHF exercise training studies.

An alternative explanation for these novel and intriguing observations is that the analyses, post hoc from an observational registry, are confounded. This explanation is based on the fundamental principle that in nature quantitative differences are common (that is, something may be a little better or a little worse among subsets of patients) but qualitative differences are rare (that is, something favorably influences outcomes in one subset and adversely in another). It is probable that BMI in this population is simply a marker of other predictors of improved outcome not ascertained in the database or controlled for in the current analyses. For example, one might imagine that heavier patients would have relatively higher blood pressure, allowing more aggressive titration of life-saving therapies such as angiotensin-converting enzyme inhibitors, β-blockers, and spironolactone; that there was more tobacco abuse among the patients with lower BMI; etc. The observations of this study should be confirmed among other populations, and the intriguing hypotheses should be tested prospectively in clinical trials.

Study: 

Less treatment and impaired prognosis after acute myocardial infarction—the impact of diabetes mellitus on acute treatment, hospital and long-term mortality: Results of the Maximal Individual Therapy in Acute Myocardial Infarction (MITRA) and Myocardial Infarction (MIR) registries

Presenter: 

Dr Anselm K. Gitt, Heart Center Ludwigshafen, Ludwigshafen, Germany

Background: 

Diabetes mellitus has been associated with impaired short- and long-term prognosis after acute myocardial infarction in a number of studies. However, less is known regarding the influence of diabetes on the use of evidence-based therapies and on clinical outcomes after acute myocardial infarction (AMI) in contemporary “usual” clinical practice.

Results: 

Data were analyzed from the MIR and MITRA registries. Between 1994 and 1998, 21,416 consecutive patients with AMI from 270 participating hospitals in Germany were included in these two registries. Diabetes was present in 5086 (24%) patients. Diabetic patients were older and had worse cardiovascular risk profiles, and all analyses were multivariate to adjust for these imbalances. Diabetic patients were less likely to be treated with reperfusion therapy (percutaneous transluminal coronary angioplasty/thrombolysis) (38.5% vs 53.1%) and β-blockers (49% vs 58%) but more likely to receive angiotensin-converting enzyme (ACE) inhibitors (61% vs 54%). Diabetes was associated with 30% to 55% increased odds for short- and long-term mortality and for the development of CHF.

After AMI, and adjustment for age and other risk factors, diabetes was associated with significantly worse clinical outcomes compared with nondiabetic patients. The underuse of evidence-based therapies was especially evident among the diabetic population and may account for some of the worse prognoses associated with diabetes.

Interpretation: 

The data on clinical outcomes are consistent with previously published results and extend prior observations to contemporary “routine” clinical practice. This study suggests that despite advances in cardiovascular therapeutics over the past decade diabetes continues to be associated with worse clinical outcomes.

A novel but not surprising observation is the relative and absolute underuse of reperfusion therapies and β-blockers among this high-risk population. Historic concerns regarding the use of antithrombotic/fibrinolytic therapies (risk with retinopathy) and β-blockers (impaired glucose and lipid metabolism, attenuation of hypoglycemia symptoms) among patients with diabetes likely contributed to the underuse of these therapies. However, a wealth of clinical trial evidence has demonstrated the safety and efficacy of both these therapies among patients with diabetes. Similarly, clinical trials have demonstrated the remarkable benefits of ACE inhibitors among diabetic patients, and although patients with diabetes were more likely to get these drugs compared with nondiabetic patients, ACE inhibitors were underused in both subsets.

Continued efforts are needed to optimize the translation of currently available clinical trial evidence into routine clinical practice and to identify features unique to diabetes in an effort to develop novel therapeutic approaches aimed at improving clinical outcomes in this high-risk population.

Study: 

Physician-performed point-of-care echocardiography with a laptop platform compared with physical examination in the cardiovascular patient.

Presenter: 

Dr Kirk T. Spencer, University of Chicago, Chicago, Ill

Results: 

It is well established that the auscultatory skills of physicians are declining, both in cardiologists and in noncardiologists. Echocardiography has become a mainstay in the diagnostic evaluation of cardiovascular patients, but point-of-care echocardiography has been limited by cost and the size of echocardiographic imaging systems. New portable echocardiography devices now make it possible for the widespread use of imaging technology at the point of care. Dr Spencer and colleagues sought to compare the diagnostic accuracy of a physical examination with that of physician-performed bedside echocardiography with a novel laptop-based imaging system.

Four board-certified cardiologists, each with level II echo training, evaluated 40 patients with a range of cardiovascular diagnoses. The patients were chosen on the basis of their known cardiovascular diagnoses and physical examination findings. The cardiologists were given a brief description of each patient’s history. The cardiologists then performed a physical examination without asking for additional historic information. After recording their diagnoses, the physicians performed a 2-dimensional and color Doppler echocardiographic examination with a portable echocardiographic imaging system (Agilent Technologies, Andover, Mass) and separately recorded the echocardiographic findings. For comparison purposes, all patients then underwent a standard echocardiographic evaluation by an experienced sonographer blinded to the results of the physical examination and portable echo.

In the 40 patients, there were 79 cardiovascular findings, of which 34 were considered to be of major clinical significance. Physical examination detected 41% (32/79) of the total findings and 56% (19/34) of the major clinical findings. With the addition of the portable echo device, 71% (56/79) of the total cardiovascular findings and 78% (26/34) of the major clinical findings were detected.

The authors concluded that physician-performed point-of-care echocardiography dramatically improved the diagnostic yield of cardiovascular findings compared with physical examination alone.

Interpretation: 

As technology improves, bringing sophisticated diagnostic testing modalities to the bedside for real-time evaluations will become commonplace. The study presented by Dr Spencer and colleagues is one of the first to report the utility of small point-of-care echocardiographic imaging systems. Their study is unique in that they sought to determine the differences between a standard diagnostic strategy of physical examination with a strategy of physical examination plus point-of-care echocardiography. They found that point-of-care echo increased the diagnostic yield by 30% for all findings and 22% for findings of major clinical importance.

This study highlights a number of important findings. First, the physical examination, even in those with subspecialty cardiology training, is poor at detecting cardiovascular findings. Importantly, the physical examination detected only 56% of the major clinical findings. Point-of-care echo improved the diagnostic yield substantially. Notably, physical examination plus point-of-care echocardiography still missed more than 20% of clinically important findings.

The patients in this study were selected for their cardiovascular findings and the participants were highly trained cardiologists with particular expertise in echocardiography. We do not yet know the impact of this type of technology in a larger screening population or whether those without subspecialty training in cardiology would have similar improvements in diagnostic yield. It is clear that technology is here to stay, but we do not yet know how to apply this technology in the best way to augment our diagnostic abilities in a way that will ultimately benefit the care of patients.

Study: 

Vegetation characteristics by transesophageal echocardiography to predict embolic events in infective endocarditis

Presenter: 

Dr Giovanni Di Salvo and Dr Valeria Pergola, La Timone Hospital, Marseilles, France

Results: 

Echocardiography is a mainstay in the diagnostic evaluation of patients with endocarditis. The prognostic ability of echocardiography to predict adverse outcomes, such as embolic events, is less well established. Embolic events in endocarditis can have devastating consequences such as stroke, splenic abscess, and renal failure. Drs Di Salvo and Pergola, and colleagues sought to determine the echocardiographic predictors of embolic events in a large group of patients with definite endocarditis.

In this study, 176 patients with definite endocarditis (as defined by the Duke criteria), were studied with transesophageal echocardiography (TEE). Vegetation lengths were measured and vegetation surface was calculated. In addition, vegetation mobility was assessed by a semiquantitative grading scale (1-4). Cerebral, thoracic, and abdominal computed tomographic (CT) scans were performed in 167 patients to evaluate for peripheral embolic events.

Vegetations were detected in 75% of those with definite endocarditis. Embolic events, as detected by CT scans, were found in 37% of patients. Vegetation length and surface were both larger in those with embolic events. Specifically, in those with vegetations longer than 15 mm there were 32% more embolic events than in those with vegetations less than 15 mm. Vegetation mobility was also predictive of embolic events. Among those with highly mobile vegetations, 83% had an embolic event, which was significantly higher than patients with less mobile vegetations. In a multivariable analysis including clinical, microbiologic, and echocardiographic findings, only vegetation length and mobility were independently predictive of embolic events.

Interpretation: 

Although echocardiography is a critical step in the diagnostic evaluation of patients with endocarditis, it has been less clear whether echocardiographic findings provide specific prognostic information. This study is unique in a number of ways. First, it represents a large number of patients who were evaluated with TEE. In addition, 95% of patients had complete CT scans to determine the rate of embolic events.

This study is important in that it establishes a rate of embolic events, as determined by CT scans, in the modern era of endocarditis patients. It is not clear from this work how many embolic events were clinically apparent and would have been detected without the compulsory CT scans. In addition, the difference in vegetation characteristics between those with clinically apparent embolic events compared with those with silent events was not presented.

This study underscores the importance of echocardiography in the care of patients with endocarditis. Echocardiographic findings provide important diagnostic information and clearly provide prognostic information as well. To move the care of endocarditis patients forward, more work is needed to assess whether early intervention, based on prognostic information, decreases morbidity and mortality in this complex disease.

Study: 

Vascular Endothelial Growth Factor (VEGF) in Ischemia for Vascular Angiogenesis (VIVA) study results at 1 year

Presenter: 

Dr Timothy D. Henry, Hennepin County Medical Center, Minneapolis, Minn

Background: 

Vascular endothelial growth factor (VEGF) has been investigated as an angiogenic agent in a number of animal and human studies. Excitement regarding the potential benefit of this therapy has been tempered with concern over possible side effects, including increased cancer risk. The VIVA study was the first large placebo-controlled randomized study of therapeutic angiogenesis to be conducted in humans. The original VIVA trial included 178 patients with stable angina and viable but ischemic myocardium deemed nonrevascularizable. Patients were randomly assigned in a double-blind fashion to receive either placebo or low-dose (17 mg/kg/min) or high-dose (50 mg/kg/min) human recombinant VEGF protein. The primary end point was exercise time at 60 days with all 3 treatment groups showing significant improvement from baseline with no significant differences among the groups (reported at the American College of Cardiology, 1999). To study long-term safety and efficacy in the treatment group, as well as the response to placebo, 106 patients from the VIVA study were enrolled in a separate trial and followed up for 1 year in a double-blind fashion. Rates of Canadian class angina (CCA), death, myocardial infarction, revascularization, and cancer were collected for each of the three groups.

Results: 

The results are presented in Tables III and IV.

Table III. Mean angina class

	Group	Baseline	60 days	120 days	365 days
	Placebo	2.8 ± 0.6	2.0 ± 0.8	2.2 ± 1.1	2.4 ± 1.6
	17 mg/kg	2.9 ± 0.6	2.0 ± 0.8	1.7 ± 0.9	2.3 ± 1.4
	50 mg/kg	2.7 ± 0.8	1.9 ± 1.0	1.6 ± 0.8	1.9 ± 1.3 (P < .03)

Table IV. One-year clinical events (%)

	Event	Placebo (n = 37)	17 mg/kg (n= 34)	50 mg/kg (n = 35)
	Death	5.3	3.1	0
	Myocardial infarction	2.7	0	5.9
	Revascularization	18.9	9.7	8.8
	Cancer	10.5	2.9	0*
	Any event	31.6	16.1	11.8*
	*P < .04 high versus placebo.

Interpretation: 

The most important finding from this study was the fact that there does not appear to be an increased incidence of cancer associated with VEGF therapy at 1 year. Rates of cancer were actually significantly lower in the high-dose VEGF group. In addition, mean angina class and overall clinical events were significantly lower in the high-dose group compared with placebo. These findings are encouraging and should promote continued investigation in this field.

Study: 

Disparities in ICD use by race and sex

Presenter: 

Dr Judy K. Battle, Duke Clinical Research Institute, Durham, NC

Background: 

Sudden cardiac death affects more than 350,000 individuals in the United States, and ventricular tachyarrhythmias represent the most common cause of sudden cardiac death. Recent studies have proved that the ICD is superior to antiarrhythmic drugs for increasing overall survival.

Although racial and gender differences exist in the use of catheterization, angioplasty, and bypass surgery for the evaluation and treatment of ischemic syndromes, no published studies have examined variation in the use of ICDs by race and sex among survivors of malignant ventricular arrhythmias. The purpose of this study was to examine the use of ICDs by race and sex among Medicare patients between 1991 and 1992.

Results: 

The patient population consisted of 64,936 patients identified in the Medicare National Claims History file with a diagnosis of a life-threatening arrhythmia, including the international classification of disease for ventricular tachycardia, ventricular fibrillation, and cardiac arrest. These were initial admissions and included only those patients admitted to hospitals that offered ICDs. There were 4329 African American patients, and when compared with white patients they were similar in age (median 75 years), more often female (50% vs 35%), and more likely to have cardiac arrest (22% vs 18%), congestive heart failure (50% vs 41%), and diabetes (18% vs 12%). African American patients had higher 1-year mortality rates than whites (41% vs 35%). Consistent with prior investigations, African American patients were less likely to undergo cardiac catheterization, angioplasty, and surgery, and in this study African American and female patients were less than half as likely as white male patients to receive cardiac catheterization (African American female/white male odds ratio [OR] 0.32, African American male/white male OR 0.36, and white female/white male OR 0.47). All these results were statistically significant with narrow confidence intervals.

Interpretation: 

Overall, survivors of sustained ventricular tachycardia and ventricular fibrillation are at extremely high risk for sudden cardiac death and reportedly have a greater benefit from ICD than from medical therapy. However, in this study, Medicare patients with malignant ventricular arrhythmias admitted to hospitals that offered ICDs were less likely to receive the procedure if they were African American or female. Although there are limitations with retrospective analyses of claims data, including a lack of data on medical therapy and patient refusal rates, clearly more data are needed in this area to explain these disturbing results.