Generalized pairwise comparison methods to analyze (non)prioritized composite endpoints.

In the analysis of composite endpoints in a clinical trial, time to first event analysis techniques such as the logrank test and Cox proportional hazard test do not take into account the multiplicity, importance, and the severity of events in the composite endpoint. Several generalized pairwise comparison analysis methods have been described recently that do allow to take these aspects into account. These methods have the additional benefit that all types of outcomes can be included, such as longitudinal quantitative outcomes, to evaluate the full treatment effect. Four of the generalized pairwise comparison methods, ie, the Finkelstein-Schoenfeld, the Buyse, unmatched Pocock, and adapted O'Brien test, are summarized. They are compared to each other and to the logrank test by means of simulations while specifically evaluating the effect of correlation between components of the composite endpoint on the power to detect a treatment difference. These simulations show that prioritized generalized pairwise comparison methods perform very similarly, are sensitive to the priority rank of the components in the composite endpoint, and do not measure the true treatment effect from the second priority-ranked component onward. The nonprioritized pairwise comparison test does not suffer from these limitations and correlation affects only its variance.

One year cost effectiveness of sirolimus eluting stents compared with bare metal stents in the treatment of single native de novo coronary lesions: an analysis from the RAVEL trial.

OBJECTIVE: To assess the balance between costs and effects of the sirolimus eluting stent in the treatment of single native de novo coronary lesions in the RAVEL (randomised study with the sirolimus eluting Bx Velocity balloon expandable stent in the treatment of patients with de novo native coronary artery lesions) study. DESIGN: Multicentre, double blind, randomised trial. SETTING: Percutaneous coronary intervention for single de novo coronary lesions. PATIENTS: 238 patients with stable or unstable angina. INTERVENTIONS: Randomisation to sirolimus eluting stent or bare stent implantation. MAIN OUTCOME MEASURES: Patients were followed up to one year and the treatment effects were expressed as one year survival free of major adverse cardiac events (MACE). Costs were estimated as the product of resource utilisation and Dutch unit costs. RESULTS: At one year, the absolute difference in MACE-free survival was 23% in favour of the sirolimus eluting stent group. At the index procedure, sirolimus eluting stent implantation had an estimated additional procedural cost of 1286. At one year, however, the estimated additional cost difference had decreased to 54 because of the reduction in the need for repeat revascularisations in the sirolimus group (0.8% v 23.6%; p < 0.01). After adjustment of actual results for the consequences of angiographic follow up (correction based on data from the BENESTENT (Belgium Netherlands stent) II study), the difference in MACE-free survival was estimated at 11.1% and the additional one year costs at 166. CONCLUSIONS: The one year data from RAVEL suggest an attractive balance between costs and effects for sirolimus eluting stents in the treatment of single native de novo coronary lesions. The cost effectiveness of drug eluting stents in more complex lesion subsets remains to be determined.

Clinical and angiographic factors associated with asymptomatic restenosis after percutaneous coronary intervention.

BACKGROUND: Angiographic restenosis after percutaneous coronary interventional procedures is more common than recurrent angina. Clinical and angiographic factors associated with asymptomatic versus symptomatic restenosis after percutaneous coronary intervention were compared. METHODS AND RESULTS: All patients with angiographic restenosis from the BENESTENT I, BENESTENT II pilot, BENESTENT II, MUSIC, WEST 1, DUET, FINESS 2, FLARE, SOPHOS, and ROSE studies were analyzed. Multivariate analysis evaluated 46 clinical and angiographic variables, comparing those with and without angina. The 10 studies recruited 2690 patients who underwent percutaneous revascularization and 6-month follow-up angiography (86% of those eligible). Restenosis (>/=50% diameter stenosis) occurred in 607 patients and was clinically silent in 335 (55%). Male sex (P=0.008), absence of antianginal therapy with nitrates (P=0.0002) and calcium channel blockers (P=0.02) at 6 months, greater reference diameter after the procedure (P=0.04), greater reference diameter at follow-up (P=0.004), and lesser lesion severity (percent stenosis) at 6 months (P=0.0004) were univariate predictors of asymptomatic restenosis. By multivariate analysis, only male sex (P=0.04), greater reference diameter at follow-up (P=0.002), and lesser lesion severity at 6 months (P=0.0001) were associated with restenosis without angina. CONCLUSIONS: Approximately half of patients with angiographic restenosis have no symptoms. The only multivariate predictors of silent restenosis at 6 months were male sex, greater reference diameter at follow-up, and lesser lesion severity on follow-up angiography.

Acute and 6-month clinical and angiographic outcome after implantation of the ACS Duet stent for single-vessel coronary artery disease: final results of the European and US ACS Multi-link Duet Registry.

The aim of the study was to determine the safety and efficacy of the second-generation ACS Multi-Link Duet coronary stent system for the treatment of single, symptomatic, de novo, native coronary artery lesions. Between February and June 1998, 427 patients (69.3% male, 51.5% class 3 or 4 angina, 20.1% diabetic, 43.6% hyperlipidemia) were included at 38 centers in this prospective observational study. All patients received ticlopidine 500 mg/day for 1 month and aspirin > or =100 mg/day. The Duet stent was available in 8, 18, and 28 mm length and 3.0, 3.5, and 4.0 mm diameter. After adequate predilatation, stents were successfully implanted, at up to 16 atm, in 99.3% of patients. Mean vessel diameter by core laboratory quantitative coronary angiography was 3.0 +/- 0.53 mm and postprocedural minimum luminal diameter was 2.79 +/- 0.43 mm (12% +/- 9.3% diameter stenosis). At 30 days, 96.7% of patients were event-free and at 6 months 88.1% remained free of major adverse cardiac events. The restenosis rate was 18.1%. The ACS Duet stent was safely implanted in >99% of target lesions by a diverse group of international investigators. With late outcomes at least comparable to the best published results, this stent platform provides safe and effective percutaneous treatment of obstructive coronary artery disease. Cathet Cardiovasc Intervent 2001;54:25-33.

Clinical and quantitative coronary angiographic predictors of coronary restenosis: a comparative analysis from the balloon-to-stent era.

OBJECTIVES: We sought to assess whether coronary stents have modified the predictive value of demographic, clinical and quantitative coronary angiographic (QCA) predictors of coronary restenosis. BACKGROUND: A systematic analysis in a large cohort of registries and randomized trials of the percutaneous transluminal coronary angioplasty (PTCA) and stent era has never been performed. METHODS: A total of 9,120 treated lesions in 8,156 patients included in nine randomized trials and 10 registries, with baseline, post-procedural and six-month follow-up QCA analyses, were included in this study. Predictors of restenosis were identified with univariate and multivariate logistic regression analyses. Interaction terms were introduced in the regression equation to evaluate whether the predictors of restenosis were common to both eras or specific for either one of the revascularization techniques. RESULTS: The restenosis rate was 35% after PTCA and 19% after angioplasty with additional stenting. In the univariate analysis, favorable predictors were previous coronary artery bypass graft surgery (CABG), stent use, stent length and a large pre-procedural minimal lumen diameter (pre-MLD); unfavorable predictors were weight, body mass index, diabetes mellitus, multi-vessel disease, lesion length and a high residual post-procedural diameter stenosis (post-DS). Predictors specific for the PTCA population were a large post-procedural MLD (post-MLD) as favorable and a severe pre-procedural DS (pre-DS) as unfavorable. Favorable predictors specific for the stent population were a large post-MLD and a large pre-procedural reference diameter (pre-RD). In the multivariate analysis, the best model included the following favorable predictors: stent use, a large post-MLD, previous CABG and the interaction term between stent use and a large post-MLD; unfavorable predictors were lesion length and diabetes mellitus. CONCLUSIONS: There are no major differences in demographic and clinical predictors of coronary restenosis between PTCA and stent populations. In the modern (stent) era, a severe pre-DS is no longer an unfavorable predictor of restenosis. Still important, but more so in the stent population, is a large post-MLD (optimal result). Finally, a larger pre-RD became a favorable predictor with the advent of stenting.

Angiographic and clinical one-year follow-up of the Cordis tantalum coil stent in a multicenter international study demonstrating improved restenosis rates when compared to pooled PTCA and BENESTENT-I data: the European Antiplatelet Stent Investigation (EASI).

The Cordis tantalum coil stent was assessed in a nonrandomized multicenter trial: 275 patients with stable or unstable angina were entered. Clinical follow-up was for 1 year, with repeat angiography at 6 months. The major adverse cardiac event rates (MACE) were 3%, 14%, and 17% at 1, 7, and 13 months, respectively. The procedural success rate was 96% and the subacute occlusion rate 1.5%, in a group of patients over 60% of whom had ACC/AHA type B2 or C lesions. The binary restenosis rate at 6 months was 17.3%. Minimum lumen diameter increased from 1.07 +/- 0.28 mm preprocedure to 2.93 +/- 0.34 mm poststenting and at 6 months was 1.99 +/- 0.69 mm. These results demonstrate that the Cordis tantalum stent can be used to treat complex lesions with good procedural success and low rates of subacute thrombosis and restenosis at 6 months.

Long-term treatment with a platelet glycoprotein-receptor antagonist after percutaneous coronary revascularization. EXCITE Trial Investigators. Evaluation of Oral Xemilofiban in Controlling Thrombotic Events.

BACKGROUND: When administered intravenously at the time of percutaneous coronary revascularization, glycoprotein IIb/IIIa receptor antagonists decrease the incidence of death and nonfatal myocardial infarction and the need for urgent revascularization. We hypothesized that long-term administration of oral glycoprotein IIb/IIIa antagonists, which block the aggregation of platelets, might stabilize intravascular plaque and prevent additional ischemic cardiac events. METHODS: We conducted a prospective, double-blind trial in which 7232 patients were randomly assigned to receive 20 mg of oral xemilofiban or placebo 30 to 90 minutes before undergoing percutaneous coronary revascularization, with maintenance doses of 10 or 20 mg of xemilofiban or placebo administered three times daily for up to 182 days. There were two primary composite end points: one was death, nonfatal myocardial infarction, or urgent revascularization at 182 days, and the other was death or nonfatal myocardial infarction at 182 days. RESULTS: Death, myocardial infarction, or urgent revascularization occurred within 182 days in 324 patients who received placebo (Kaplan-Meier cumulative event rate, 13.5 percent), 332 who received 10 mg of xemilofiban (13.9 percent, P=0.82 for the comparison with placebo), and 306 who received 20 mg of xemilofiban (12.7 percent, P=0.36 for the comparison with placebo). The incidence of death or myocardial infarction was also similar in all three groups. Clinically significant hemorrhagic complications and thrombocytopenia were infrequent. CONCLUSIONS: The administration of the glycoprotein IIb/IIIa antagonist xemilofiban before percutaneous coronary revascularization and for up to six months thereafter does not significantly reduce the incidence of important clinical end points.

Randomized comparison of primary stenting and provisional balloon angioplasty guided by flow velocity measurement. Doppler Endpoints Balloon Angioplasty Trial Europe (DEBATE) II Study Group.

BACKGROUND: Coronary stenting improves outcomes compared with balloon angioplasty, but it is costly and may have other disadvantages. Limiting stent use to patients with a suboptimal result after angioplasty (provisional angioplasty) may be as effective and less expensive. METHODS AND RESULTS: To analyze the cost-effectiveness of provisional angioplasty, patients scheduled for single-vessel angioplasty were first randomized to receive primary stenting (97 patients) or balloon angioplasty guided by Doppler flow velocity and angiography (523 patients). Patients in the latter group were further randomized after optimization to either additional stenting or termination of the procedure to further investigate what is "optimal." An optimal result was defined as a flow reserve >2.5 and a diameter stenosis <36%. Bailout stenting was needed in 129 patients (25%) who were randomized to balloon angioplasty, and an optimal result was obtained in 184 of the 523 patients (35%). There was no significant difference in event-free survival at 1 year between primary stenting (86.6%) and provisional angioplasty (85.6%). Costs after 1 year were significantly higher for provisional angioplasty (EUR 6573 versus EUR 5885; P:=0.014). Results after the second randomization showed that stenting was also more effective after optimal balloon angioplasty (1-year event free survival, 93.5% versus 84.1%; P:=0. 066). CONCLUSIONS: After 1 year of follow-up, provisional angioplasty was more expensive and without clinical benefit. The beneficial value of stenting is not limited to patients with a suboptimal result after balloon angioplasty.

Continuous ST-segment monitoring associated with infarct size and left ventricular function in the GUSTO-I trial.

BACKGROUND: The aim of this study was to evaluate whether in patients with myocardial infarction, the intensity and duration of myocardial ischemia as measured by continuous ST monitoring are associated with infarct size and residual left ventricular function. METHODS AND RESULTS: The analyses included patients with myocardial infarction, receiving thrombolytic therapy, who were enrolled in the electrocardiographic substudy of GUSTO-I, monitored by a vector-derived 12-lead electrocardiographic recording system, and in whom either infarct size (defined as cumulative release of alpha-hydroxybutyrate dehydrogenase activity per liter of plasma over a 72-hour period [Q(72)]) or left ventricular ejection fraction (LVEF) was determined. With the use of linear regression analysis, we investigated the association of various ST-trend characteristics with Q(72) (206 patients) and with LVEF (180 patients). A higher area under the ST trend since thrombolysis until 50% ST recovery and a higher area under recurrent ischemic episodes (ST reelevations) were significantly associated with a higher Q(72), whereas only a higher area under recurrent ischemic episodes was significantly associated with a lower LVEF. These associations remained after adjusting for other patient characteristics such as age, sex, infarct location, and time to treatment. CONCLUSIONS: These findings support the physiologic hypothesis that both the intensity and duration of myocardial ischemia (both reflected by the estimated areas under the ST-trend curve) determine myocardial damage and thus are associated with infarct size and ejection fraction in patients with acute myocardial infarction who receive thrombolytic therapy.

Redundancy of single diagnostic test evaluation.

Diagnostic research and diagnostic practice frequently do not cohere. Studies commonly evaluate whether a single test discriminates between disease presence and absence, whereas in practice a test is always judged in the context of other information. This study illustrates drawbacks of single-test evaluation and discusses principles of diagnostic research. We used data on 140 patients suspected of pulmonary embolism who had an inconclusive ventilation-perfusion lung scan. We evaluated three tests: partial pressure of oxygen in arterial blood (PaO2), x-ray film of the thorax, and leg ultrasound. On the basis of single-test evaluations, ultrasound was most informative. Given a prior probability of 0.27, it had a much better combination of positive and negative predictive value (0.71 and 0.21, respectively) relative to thorax x-ray (0.33 and 0.11) and PaO2 (0.35 and 0.27). The combination of positive and negative likelihood ratio was also more promising for ultrasound (7.3 and 0.7) than for thorax x-ray (1.3 and 0.3) and PaO2 (1.3 and 0.9). As the tests are always performed after the history and physical, we judged their added value using multivariable logistic modeling with receiver operating characteristic (ROC) analyses. The ROC areas of the model, including history and physical, with additional PaO2, thorax x-ray, or ultrasound, were 0.75, 0.77, 0.81, and 0.81, respectively, which indicated similar added value of thorax x-ray and ultrasound. Application of the models to patient subgroups also yielded added predictive value for thorax x-ray film. Thus, the results of single-test evaluations may be very misleading. As no diagnosis is based on one test, single-test evaluations have limited value in diagnostic research and only have relevance in the context of screening and the initial phase of test development. Diagnostic research should always apply an approach of constructing, extending, and validating diagnostic models in agreement with routine clinical work-up using logistic regression analyses.

A randomized placebo-controlled trial of fluvastatin for prevention of restenosis after successful coronary balloon angioplasty; final results of the fluvastatin angiographic restenosis (FLARE) trial.

BACKGROUND: The 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) reductase inhibitors competitively inhibit biosynthesis of mevalonate, a precursor of non-sterol compounds involved in cell proliferation. Experimental evidence suggests that fluvastatin may, independent of any lipid lowering action, exert a greater direct inhibitory effect on proliferating vascular myocytes than other statins. The FLARE (Fluvastatin Angioplasty Restenosis) Trial was conceived to evaluate the ability of fluvastatin 40 mg twice daily to reduce restenosis after successful coronary balloon angioplasty (PTCA). METHODS: Patients were randomized to either placebo or fluvastatin 40 mg twice daily beginning 2-4 weeks prior to planned PTCA and continuing after a successful PTCA (without the use of a stent), to follow-up angiography at 26+/-2 weeks. Clinical follow-up was completed at 40 weeks. The primary end-point was angiographic restenosis, measured by quantitative coronary angiography at a core laboratory, as the loss in minimal luminal diameter during follow-up. Clinical end-points were death, myocardial infarction, coronary artery bypass graft surgery or re-intervention, up to 40 weeks after PTCA. RESULTS: Of 1054 patients randomized, 526 were allocated to fluvastatin and 528 to placebo. Among these, 409 in the fluvastatin group and 427 in the placebo group were included in the intention-to-treat analysis, having undergone a successful PTCA after a minimum of 2 weeks of pre-treatment. At the time of PTCA, fluvastatin had reduced LDL cholesterol by 37% and this was maintained at 33% at 26 weeks. There was no difference in the primary end-point between the treatment groups (fluvastatin 0.23+/-0.49 mm vs placebo 0.23+/-0.52 mm, P=0.95) or in the angiographic restenosis rate (fluvastatin 28%, placebo 31%, chi-square P=0.42), or in the incidence of the composite clinical end-point at 40 weeks (22.4% vs 23.3%; logrank P=0.74). However, a significantly lower incidence of total death and myocardial infarction was observed in six patients (1.4%) in the fluvastatin group and 17 (4.0%) in the placebo group (log rank P=0.025). CONCLUSION: Treatment with fluvastatin 80 mg daily did not affect the process of restenosis and is therefore not indicated for this purpose. However, the observed reduction in mortality and myocardial infarction 40 weeks after PTCA in the fluvastatin treated group has not been previously reported with statin therapy. Accordingly, a priori investigation of this finding is indicated and a new clinical trial with this intention is already underway.

The ARTS study (Arterial Revascularization Therapies Study).

The rising costs of health care have forced policy makers to make choices, and new treatments are increasingly assessed in terms of the balance between additional costs and additional effects. The recent recognition that stenting has a major and long-lasting effect enhancing balloon PTCA procedure has made it imperative to compare in patients with multivessel disease the standard surgical procedure with multiple stenting in a large scale multinational and multicentre approach (19 countries, 68 sites). Selection and inclusion of patients is based on a consensus of the cardiac surgeon and interventional cardiologist on equal 'treatability' of patients by both techniques with analysis of clinical follow-up (event-free survival) on the short (30 day), medium (1 year), and long-term (3 and 5 year) with analysis of cost-effectiveness and quality of life (EuroQol and SF-36). Of the entire trial, the primary null hypothesis which needs to be rejected is that there will be no difference in event-free survival or effectiveness (E), at 1 year and also that the direct and indirect costs (C) per event-free year are not different between surgery or stenting. For this to become significant with a power of 90% one needs 1200 patients. Between April 97 and June 98, 1205 patients have been randomized with a monthly recruitment of 83 patients. Expected costs, effects and cost-effectiveness ratio (CE ratio) are: Stent high costs 2 VDStent high costs 3 VDStent low costs 2 VDStent low costs 3 VDCABG costs (C)$19.297$24.566$16.638$20.456$21.350 effects (E)81%81%81%81%88% CE ratio$23.876$30.397$20.586$25.322$24.348 Clinically, stenting is not expected to be more effective than CABG, but should be cost effective in both the 2- and 3-VD group when using the lower cost estimate and in the 2 VD group when using the higher cost assumptions.

Intravascular ultrasound-guided optimized stent deployment. Immediate and 6 months clinical and angiographic results from the Multicenter Ultrasound Stenting in Coronaries Study (MUSIC Study)

OBJECTIVES: A study was set up to validate the safety and feasibility of intravascular ultrasound-guided stenting without subsequent anticoagulation, and its impact on the 6 months restenosis rate. METHODS: The study was designed to be multicentred, prospective, and observational. RESULTS: One hundred and sixty-one patients with stable angina and a de novo coronary artery lesion were enrolled. In four patients, the implantation of a Palmaz-Schatz (with spiral bridge) stent had failed. One of these four patients died 3 days following bypass surgery. In two other patients, intravascular ultrasound assessment was not performed. One hundred and twenty-five of the remaining 155 patients (81%) were treated with aspirin (100 mg x day(-1)), because all three criteria for optimized stent expansion were met. Twenty-two of the remaining 38 patients (25%), in whom at least one criterion was not met were treated with aspirin and acenocoumarol (3 months, INR 2.5-3.5), while 16 patients only received aspirin. Stent thrombosis was documented in two patients (1.3%) for which repeat angioplasty was performed. During the hospital stay, there were no deaths or Q-wave myocardial infarctions. Five patients (3.2%) sustained a non-Q-wave myocardial infarction. During the follow-up period (198+/-38 days, complete for all patients, except one), one patient (0.6%) sustained a Q-wave myocardial infarction, one (0.6%) underwent bypass surgery, and repeat angioplasty was performed in nine patients (5.7%). In two of the nine patients, repeat angioplasty involved another lesion. Therefore, the target lesion revascularization rate during follow-up was 4.5% (seven patients). At quantitative coronary angiography, the minimal lumen diameter (mean+/-SD) increased from 1.12+/-0.34 mm before to 2.89+/-0.35 mm after stenting. Repeat angiography at 6 months was performed in 144 patients (92%). The minimal lumen diameter at follow-up was 2.12+/-0.67 mm. Restenosis (diameter stenosis of 50% or more) was documented in 12 patients or 8.3%. When the two patients with documented stent thrombosis are included, the restenosis rate amounts to 97%. CONCLUSIONS: These data confirm that, in selected patients, stents can safely be implanted without the use of systemic anticoagulation, provided optimal stent expansion is achieved. The exact role of intravascular ultrasound in the achievement of these results needs to be established by appropriately designed studies. In the meantime, intravascular ultrasound coupled with the Palmaz-Schatz stent incorporating a spiral bridge, may have contributed considerably to the immediate angiographic outcome, which in turn may explain the favourable clinical and angiographic outcome at 6 months.

Effect of acadesine on myocardial ischaemia in patients with coronary artery disease.

Acadesine, an adenosine regulating agent, attenuates the adverse effects of ischaemia on ventricular function in animals. This study examined its influence on pacing-induced ischaemia in 47 patients undergoing coronary angiography. After 15 min of recovery from control pacing, an infusion of acadesine (5, 10, 20, 50 mg/kg i.v.) was commenced and after a further 15 min the protocol was repeated with the infusion continued. At higher doses, minor beneficial effects on ejection fraction and myocardial lactate metabolism were observed. Haemodynamics were unaffected. Systemic lactate rose in relation to acadesine, up to 60% (P < 0.001 versus placebo). The data may indicate that acadesine stimulates anaerobic glycolysis in man.

Application of treatment thresholds to diagnostic-test evaluation: an alternative to the comparison of areas under receiver operating characteristic curves.

Diagnostic tests are often evaluated by comparison of the areas under receiver operating characteristic (ROC) curves. In this study the authors compared this approach with a more direct method that takes into account consequences of a diagnosis. Data from a prospective study of diagnosis of pulmonary embolism were used for a motivating example. Using multivariable logistic regression analysis, three diagnostic models were built and compared based on their ROC curves. Although model 1 (0.706) and model 2 (0.702) had the same ROC-curve area, they performed differently when risks and benefits of subsequent decisions were considered by applying the treatment probability threshold. Models 1 and 3 (0.611) had substantially different ROC-curve areas but performed similarly taking into account the therapeutic consequences. This demonstrates that comparison of diagnostic tests using the areas under the ROC curves may lead to erroneous conclusions about therapeutic usefulness. To correspond to daily practice, it would be more appropriate to also consider the clinical implications in evaluating diagnostic tests. This is made feasible by explicit definition and application of a treatment threshold.

Efficiency optimization of the selection period in therapeutic trials.

To determine eligibility for a (randomized) clinical trial, measuring the inclusion and exclusion criteria can be extended over a period of time. During this period, known as the selection period, a patient is repeatedly examined at certain time intervals. This study describes an approach for optimizing the efficiency of the selection period. Efficiency is defined as the costs of randomizing one patient. The objective is to construct prediction models based on data obtained early in the selection period to predict subsequent exclusions. A prediction model increases the efficiency if after its application the costs per randomization are lower. The approach is illustrated using data from the selection period of the Rotterdam Cardiovascular Risk Intervention (ROCARI) trial which was composed of five consecutive patient visits. At each visit, data to determine eligibility was obtained. We found that logistic regression models based on data of the first and second visit could predict exclusions during the third visit. Application of the prediction models suggested that in this particular trial the costs per randomization would decrease by $52. As the initial costs per randomization were $1444, there would be a 3.6% (52/1444) savings in recruitment costs under the prediction models, accounting for a savings of more than $450,000. We conclude that the use of data obtained early in a selection period can predict subsequent exclusions, and therefore could increase the efficiency of such a period. The approach could be applied to data obtained in a pilot study as well as data obtained in the beginning of a prolonged intake period.

Limitations of sensitivity, specificity, likelihood ratio, and bayes' theorem in assessing diagnostic probabilities: a clinical example.

We evaluated the extent to which the sensitivity, specificity, and likelihood ratio of the exercise test to diagnose coronary artery disease vary across subgroups of a certain patient population. Among 295 patients suspected of coronary artery disease, as independently determined by coronary angiography, we assessed variation in sensitivity and specificity according to patient history, physical examination, exercise test results, and disease severity in 207 patients with and 88 patients without coronary artery disease, respectively. The sensitivity varied substantially according to sex (women 30% and men 64%), systolic blood pressure at baseline (53% to 65%), expected workload (50% to 64%), systolic blood pressure at peak exercise (50% to 67%), relative workload (33% to 68%), and number of diseased vessels (39% to 77%). The specificity varied across subgroups of sex (men 89% and women 97%) and relative workload (85% to 98%). The likelihood ratio varied (3.8 to 17.0) across the same patient subgroups, as did the sensitivity. As each population tends to be heterogeneous with respect to patient characteristics, no single level of these parameters can be given that is adequate for all subgroups. Use of these parameters as a basis for calculating diagnostic probabilities in individual patients using Bayes' theorem has serious limitations.

Cost effectiveness of stent implantation versus PTCA: the BENESTENT experience.

The aim of this paper is to assess the costs and effects of stent implantation versus percutaneous transluminal coronary angioplasty (PTCA). Data have been taken from both the BENESTENT-I and BENESTENT-II pilot study. Effects are expressed in terms of event-free survival and costs include those of the initial hospitalization and those during follow-up. The costs per additional event-free survivor after 7 months are estimated at Dfl 88,315, Dfl 28,127 and Dfl 6747 using respectively the results from the BENSTENT-I study, the BENESTENT-II pilot study and phase IV of the BENESTENT-II pilot study. Significant decreases in complications and ischaemic events have made stent implantation not only the most favourable in terms of event-free survival but also in terms of cost effectiveness.