The VIAMI-trial was conducted to investigate the differences in clinical outcome between an invasive and a conservative strategy in patients with demonstrated viability in the infarct-area. The expected event rate in the viability positive group was estimated to be 35 percent. To demonstrate with a power of 80% (α = 0.05, two-sided) that PCI leads to a 50% reduction in event rate in the invasive group compared to the conservative group, 200 patients would be needed in each group.
As a formal stopping rule for the study the following was used: if one of the treatment strategies appeared significantly superior at interim analysis (p ≤ 0.01), the study would be stopped. Interim analysis was performed each time another 100 patients were included.
Baseline descriptive data are presented as a mean ± standard deviations (SD). Differences in clinical and echocardiographic variables are assessed by unpaired Student's t-test. Differences between proportions are assessed by chi-square analysis; a Fisher's exact test is used when appropriate. Event-free survival curves are computed with the Kaplan-Meier method, and the differences between these curves are tested with a log-rank test. The Cox proportional hazards regression analysis was used to estimate the treatment effect as hazard ratio (HR) with 95% confidence intervals. Besides the "crude" effects, adjustments were made for DM, hypertension, hypercholesterolemia, current smoking, family history of CAD (model a), clinical history (angina, myocardial infarction, PCI or CABG) and medication use at baseline (aspirin, beta-blocker, Ca-inhibitor, statins, ACE-I and AT II antagonist) (model b) and for all covariates (model c).
All analyses were performed on an intention-to-treat basis. Outcome per-protocol was also evaluated, since this would reflect the true influence of PCI on clinical outcome. Because after randomization there was a median waiting-time of two days before a revascularization procedure was performed inevitably some events occurred. In the per-protocol analysis these events are excluded from analysis, because they occurred before the by protocol demanded intervention. To make a fair comparison between the two groups in the per-protocol analysis we also excluded the events in the conservative group occurring during the first two days after randomization. All analyses were performed with the use of SPSS software, version 16.0 (SPSS, Inc., Chigago, Illinois).
As a formal stopping rule for the study the following was used: if one of the treatment strategies appeared significantly superior at interim analysis (p ≤ 0.01), the study would be stopped. Interim analysis was performed each time another 100 patients were included.
Baseline descriptive data are presented as a mean ± standard deviations (SD). Differences in clinical and echocardiographic variables are assessed by unpaired Student's t-test. Differences between proportions are assessed by chi-square analysis; a Fisher's exact test is used when appropriate. Event-free survival curves are computed with the Kaplan-Meier method, and the differences between these curves are tested with a log-rank test. The Cox proportional hazards regression analysis was used to estimate the treatment effect as hazard ratio (HR) with 95% confidence intervals. Besides the "crude" effects, adjustments were made for DM, hypertension, hypercholesterolemia, current smoking, family history of CAD (model a), clinical history (angina, myocardial infarction, PCI or CABG) and medication use at baseline (aspirin, beta-blocker, Ca-inhibitor, statins, ACE-I and AT II antagonist) (model b) and for all covariates (model c).
All analyses were performed on an intention-to-treat basis. Outcome per-protocol was also evaluated, since this would reflect the true influence of PCI on clinical outcome. Because after randomization there was a median waiting-time of two days before a revascularization procedure was performed inevitably some events occurred. In the per-protocol analysis these events are excluded from analysis, because they occurred before the by protocol demanded intervention. To make a fair comparison between the two groups in the per-protocol analysis we also excluded the events in the conservative group occurring during the first two days after randomization. All analyses were performed with the use of SPSS software, version 16.0 (SPSS, Inc., Chigago, Illinois).
Free full text: Click here
