Stenting vs. Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Stenting vs. Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis (NCT00576693), a randomized phase 3 study. The primary outcome is Any Stroke or Death Within 30 Days of Enrollment or Any Revascularization Procedure OR an Ischemic Stroke in the Territory of the Symptomatic Intracranial Artery Beyond 30 Days After Enrollment..

2. Clinical Question

The trial asked whether adding intracranial angioplasty and stenting to intensive medical management would change the risk of a defined stroke-or-death outcome, compared with intensive medical management alone. The registry describes people with symptomatic stenosis of a major intracranial artery, including high-risk patients with 70%-99% stenosis who had a transient ischemic attack or stroke within 30 days before enrollment. The stated primary hypothesis proposed a reduction in the primary endpoint; the results must be read as a comparison of observed outcomes, not as a treatment recommendation.

The registry record describes the best approach for preventing another stroke or transient ischemic attack in this setting as uncertain at the time the study question was framed. The comparison therefore concerns the additional role of stenting alongside intensive medical care, rather than stenting instead of medical management.

3. Trial Design

This was a randomized, parallel-group phase 3 trial with no masking. The registry lists 451 enrolled participants and records the study as completed. The start date was 2008-10 and primary completion was 2013-04. The sponsor was the Medical University of South Carolina.

Participants were assigned to intensive medical management plus stenting or intensive medical management alone. The posted primary outcome counts were 52 among 224 participants in the stenting-plus-medical-management arm and 34 among 227 participants in the medical-management-alone arm. These are participant counts with an event, rather than event rates calculated for the reader. The registry states that all patients enrolled were included in the primary outcome analysis.

The endpoint combined events occurring during different periods. It included any stroke or death within 30 days after enrollment; any stroke or death within 30 days of a revascularization procedure of the qualifying symptomatic artery during follow-up; or an ischemic stroke in the territory of that artery from day 31 after entry through the end of follow-up. The posted mean length of follow-up was 2.4 years.

Randomization supports a planned comparison between the assigned strategies, while the parallel design means participants were followed in their assigned arm rather than crossing between arms. The record says masking was none. Further details of the randomization procedure, allocation concealment, recruitment locations, adherence, treatment switching, and the original statistical design assumptions are not reported in the registry record. The record also does not identify whether the main analysis followed an intention-to-treat label, although it states that all enrolled participants were included in the primary outcome analysis.

The device description and medical regimens are recorded in the registry, but some arm-description text is incomplete in the available record. Consequently, the full treatment protocols cannot be reconstructed from the registry entry alone.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All patients enrolled in the study were included in the primary outcome analysis.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Any Stroke or Death Within 30 Days of Enrollment or Any Revascularization Procedure OR an Ischemic Stroke in the Territory of the Symptomatic Intracranial Artery Beyond 30 Days After Enrollment., measured over Mean length of follow-up was 2.4 years.

The primary endpoint was a composite measure: any stroke or death within 30 days after enrollment; any stroke or death within 30 days of a qualifying revascularization procedure during follow-up; or an ischemic stroke in the territory of the symptomatic intracranial artery beginning on day 31 after study entry and continuing through follow-up. Stroke included ischemic stroke and specified forms of brain hemorrhage.

A composite endpoint combines more than one clinically defined event into a single outcome for analysis. In general, this can capture a broader range of important outcomes and provide one overall comparison, but the result does not tell the reader how each component contributed unless component-specific results are also reported. Component-specific primary outcome counts are not reported in the registry record.

The timing rules are part of the endpoint definition: some events counted in the first 30 days after enrollment or a revascularization procedure, while later events counted only if they were ischemic strokes in the territory of the symptomatic artery. The registry reports a mean follow-up of 2.4 years. That duration describes follow-up on average; it does not give each participant's follow-up time or the distribution of follow-up.

The outcome is presented as the number of participants with an event. Its purpose is to compare the assigned strategies on the specified combination of early stroke or death and later territory-specific ischemic stroke. The clinical importance of individual components and their separate frequencies cannot be assessed from the primary outcome count alone.

7. Results Posted in the Registry

• Intensive Medical Management Alone: 15.0% (34/227) versus 23.2% (52/224) on control; risk difference -8.2 percentage points (95% CI -15.5 to -1.0), risk ratio 0.65 (0.44–0.95); unadjusted two-proportion z-test p = 0.026.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary result is a binary outcome: each participant either met the composite endpoint during the stated follow-up or did not. The registry reports event counts and the number of participants in each arm. These counts are the basis for the unadjusted recalculation. That recalculation compares the arms using the posted summary data and is separate from the sponsor's time-to-event analysis.

What does the risk difference describe?

A risk difference expresses the contrast between the proportions of participants with an event in two groups. Its sign depends on the order of the groups in the comparison. In this record, the comparison is framed as medical management alone relative to stenting plus medical management; the negative direction therefore corresponds to fewer observed events in the medical-management-alone arm. The recalculated confidence interval does not include zero, which is consistent with a nonzero difference under that unadjusted calculation. This does not establish that the difference is clinically important or explain its cause.

The risk difference is an absolute comparison, unlike a risk ratio or odds ratio, which expresses a relative contrast. The registry includes recalculated relative measures as well, but their numerical estimates are not needed to describe the posted event counts or the direction of the unadjusted risk difference. The specific test procedure used for the unadjusted recalculation is not reported in the registry record.

Why is a log-rank test also reported?

A log-rank test is generally used to compare time-to-event experience between groups. It uses information about when events occur and can account for participants having different lengths of observed follow-up through censoring, where applicable. This differs from a simple comparison of whether each participant had an event by the end of follow-up. The registry identifies the sponsor analysis as log-rank and reports its p-value, but does not provide the test statistic, an effect estimate, a confidence interval, censoring details, or the specific model specification. Those details are not reported in the registry record.

The two approaches can therefore yield p-values that are not identical: they summarize the data differently and need not use the same assumptions. The recalculated p-value is 0.026, while the sponsor's log-rank result is reported as having a p-value below 0.05. The latter p-value is recorded to more decimal places in the registry, but its exact numerical value is not repeated in this explanation. The registry does not report a discrepancy flag, so the posted results do not identify a conflict requiring the sponsor's analysis to take precedence over the recalculation.

How should a p-value be read?

In general, a p-value measures how unusual data at least as inconsistent with the null hypothesis as the observed data would be, assuming the null model and test assumptions hold. It is not the probability that the null hypothesis is true, and it is not an estimate of the size or practical importance of an effect. A p-value also does not show whether the observed difference is likely to persist in a different population or setting.

For this trial, the p-values should be read alongside the event counts, the direction of the comparison, and the endpoint definition. A p-value alone cannot communicate the number of participants affected, distinguish the endpoint components, or show the full range of effect sizes compatible with the data. The recalculated confidence interval provides a range for the unadjusted risk difference, while the sponsor's log-rank result is not accompanied by a posted interval or effect estimate.

What can be said about the analysis plan?

The registry record lists one primary outcome measure and one reported analysis. It describes the population for the primary analysis as all patients enrolled. It does not report the original statistical analysis plan, the planned primary test, interim analyses, stopping rules, covariate adjustment, stratification variables, or procedures for handling missing outcomes. Those elements should not be inferred from the p-values or from the fact that the study was randomized.

The trial is classified as phase 3, randomized, unmasked, and parallel-group. Those design labels describe the broad study structure, not the full statistical methods. In general, randomization is intended to balance prognostic factors on average across assigned groups, but the registry record does not provide baseline balance information or enough detail to assess the implementation of randomization. The posted comparison is therefore best described using the outcomes and analysis details that are actually recorded.

10. Confidence Intervals and Precision

A confidence interval gives a range of effect estimates that are reasonably compatible with the data and the method used to calculate the interval. In general, a narrower interval indicates greater statistical precision, while a wider interval indicates more uncertainty about the size of the effect. An interval should be interpreted together with its scale and direction: the risk-difference interval is an absolute contrast, and its sign reflects the order of the groups in the comparison.

The recalculated interval for the risk difference lies on the side corresponding to fewer observed events in the medical-management-alone arm and excludes no difference. Its endpoints are not repeated because the registry's displayed values have more than three decimal places. The interval still allows a range of possible differences, so the point estimate should not be treated as the uniquely established effect. The calculation is unadjusted and is based on the posted counts rather than individual participant data.

The sponsor's reported log-rank result has no accompanying effect estimate or confidence interval in the registry record. Its p-value alone does not show the range of time-to-event effects compatible with the data, or how much the event timing differed between groups. A confidence interval for that analysis, along with the underlying time-to-event estimate, would provide more information about precision.

Precision is distinct from statistical significance. A finding can have a p-value below a chosen threshold while the effect estimate remains uncertain, and an interval's width helps describe that uncertainty. In this trial, the posted count-based interval and the sponsor's time-to-event result answer related but not identical statistical questions. Neither should be interpreted without the endpoint's composite definition, the follow-up context, and the limitations of the registry reporting.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. The primary outcome is binary, so the illustrations estimate the number of participants needed in each of two equally allocated arms. They are not event-count targets and are not the trial's observed enrollment. The original sample-size justification and design assumptions for the trial are not reported in the registry record.

The planning illustration uses a two-sided significance level of 0.05, equal allocation between the arms, and a normal approximation. It uses the observed effect as its reference and considers planning for 100%, 75%, or 50% of that observed effect. The figures describe the number of participants per arm for 80% power and the number of participants per arm for 90% power under those assumptions.

In plain language, the 80% power column gives the estimated per-arm enrollment for a design that would have an 80% chance of detecting the specified effect if that effect were the true one and the assumptions held. The 90% power column gives the corresponding estimate for a 90% chance. For planning for 100% of the observed effect, the illustrations are 357 participants per arm for 80% power and 477 participants per arm for 90% power. For 75% of the observed effect, they are 661 and 884 participants per arm, respectively. For 50% of the observed effect, they are 1,543 and 2,065 participants per arm, respectively.

The pattern illustrates a general planning principle: smaller effects are harder to distinguish from random variation, so they require more participants under otherwise similar assumptions. These figures are planning illustrations based on the observed effect and assumptions stated for the illustration; they are not observed power and do not establish the power of the completed trial. The trial's own original design assumptions are not reported in the registry record.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.232; treatment proportion p₁ = 0.150; average p̄ = 0.191.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5559 = 1.0895.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5528 = 0.4653.
4. n = (1.0895 + 0.4653)² / (-0.0824)² = 356.3, rounded up to 357 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.232142857142857                          # control proportion
p1 <- 0.149779735682819                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(576693)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 357 (80% power), 477 (90% power)
 75% of observed effect: n per arm = 661 (80% power), 884 (90% power)
 50% of observed effect: n per arm = 1,543 (80% power), 2,065 (90% power)
power.prop.test: 356.3 
Simulated power at n = 357 per arm: 0.808 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Analyses Reported by the Sponsor

The sponsor's reported analysis is a log-rank test comparing the two randomized groups. The registry gives a p-value for this analysis but does not report a corresponding effect estimate or confidence interval. The unadjusted recalculation instead uses the posted event counts to compare the proportions of participants with a primary endpoint event; its p-value is 0.026, and its confidence interval excludes no difference.

These analyses use different summaries of the outcome. The recalculation treats the outcome as a binary event status, whereas the log-rank test compares time-to-event experience. Differences between their p-values do not, by themselves, indicate a contradiction. The registry does not flag a discrepancy between the posted results and the recalculation. The sponsor's result is therefore reported as a separate analysis, not as a replacement for the count-based estimate.

The exact sponsor p-value is recorded in the registry to more than three decimal places; it is described only as below 0.05. The registry does not provide details such as censoring rules, test statistic, or an effect estimate for the log-rank analysis. The sponsor's analysis should consequently be interpreted within the limits of the reported information. No claim about the relative importance of the analyses or the statistical plan can be made from details that are not reported in the registry record.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure and one reported analysis, and it lists no posted secondary outcome results. This limited set of reported analyses does not establish that only one analysis was conducted. The statistical analysis plan, any additional planned analyses, and any adjustment for multiple comparisons are not reported in the registry record. It is therefore not possible to determine whether multiplicity was anticipated or how it was handled.

The primary analysis population is described as all patients enrolled in the study. That is useful information about who was included, but the record does not name the population using a formal intention-to-treat or other analysis-set label. Details of exclusions after enrollment, adherence, crossovers, and treatment received are not reported in the registry record.

Missing outcome information can affect a trial analysis if participants do not have complete follow-up or if event status cannot be determined. The registry does not report the number of participants with missing outcome data, the completeness of follow-up, how censoring was handled, or any imputation or sensitivity analyses. The mean follow-up of 2.4 years does not supply those details; an average does not show how much follow-up varied among participants.

The record also does not report interim analyses, early stopping criteria, subgroup analyses, covariate adjustment, or an account of protocol deviations. These omissions do not demonstrate that such procedures were or were not used. They mean the posted information cannot support a detailed assessment of multiplicity, missing-data assumptions, or the robustness of the analysis to alternative methods. Interpretation should remain anchored to the stated primary endpoint, the all-enrolled analysis description, and the results that are reported.

15. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: ischemic stroke in the territory of qualifying symptomatic artery (17.9%), transient ischemic attack / cerebral infarct with temporary signs (13.4%), neurology (11.6%), cardiac general (7.9%), systemic hemorrhage (6.7%).

For serious adverse events, the registry records 146 participants affected among 224 at risk in the intensive-medical-management-plus-stenting group, and 121 affected among 227 at risk in the intensive-medical-management-alone group. For other adverse events, it records 117 affected among 224 at risk in the stenting-plus-medical-management group and 118 among 227 at risk in the medical-management-alone group. These are counts of participants recorded as affected, not a count of every individual event occurrence.

The registry lists serious adverse event terms including ischemic stroke in the territory of the qualifying symptomatic artery (17.9%), transient ischemic attack / cerebral infarct with temporary signs (13.4%), Neurology (11.6%), Cardiac General (7.9%), Systemic Hemorrhage (6.7%), and Metabolic / Laaboratory (6.2%). The record does not identify the denominators or whether these listed term percentages are arm-specific, so they should not be used to construct an arm-to-arm comparison.

Death counts are not reported in the adverse-event record. The adverse-event counts alone do not establish that an event was caused by an assigned intervention, and they do not describe event severity, timing, duration, or recurrence beyond the registry's classification. They should be read as reported safety counts, not as a complete causal assessment or a treatment recommendation.

16. What the Result Does, and Does Not, Mean

The posted primary outcome counts are higher in the stenting-plus-medical-management arm than in the medical-management-alone arm: 52 of 224 participants compared with 34 of 227. The unadjusted recalculation points toward fewer events in the medical-management-alone arm, with a p-value of 0.026 and an interval that excludes no difference. The sponsor's log-rank analysis is also reported with a p-value below 0.05, but without a time-to-event effect estimate or confidence interval.

These results describe the trial's primary composite outcome in the enrolled population and under the analyses reported in the registry. They do not establish the separate effects on each component of the composite, because component-specific primary endpoint counts are not reported. They also do not show whether outcomes differed across subgroups, since subgroup results are not reported in the registry record.

A statistical result is not the same as a clinical judgment. The p-values do not measure the importance of the observed difference, and the counts cannot tell readers how the result would apply to people outside the enrolled study population. The record does not provide enough information to assess adherence, missing outcomes, the full distribution of follow-up, or all relevant analysis choices. The endpoint and mean follow-up provide context, but not a complete account of participant-level outcomes.

The trial compares two strategies that both included intensive medical management; the contrast is the addition of stenting to that management. The result should not be restated as a comparison of stenting alone against no treatment. Nor should the count-based calculation be treated as equivalent to the sponsor's time-to-event analysis. The available statistics support a description of the observed trial comparison, while broader conclusions require information beyond the posted results.

17. Why This Trial Matters Statistically

This trial is statistically notable because it combines a randomized, parallel-group comparison with a composite endpoint that includes events at different times and of different types. That structure makes it especially important to distinguish an event-count comparison from a time-to-event analysis. Both can be informative, but they answer related rather than identical questions.

The registry provides enough information to reproduce a simple unadjusted comparison from the primary event counts, while also recording a sponsor-reported log-rank result. Seeing both makes clear why an analysis should be interpreted in the context of its method, not only by its p-value. The risk difference describes an absolute contrast in participants with an event; the log-rank test compares event timing over follow-up. The registry does not provide a sponsor effect estimate or interval, so the time-to-event result is less fully quantified in the posted record.

The trial also illustrates the difference between an observed effect and a planning assumption. The sample-size illustrations use the observed effect as their reference and show how planning for smaller effects leads to larger participant requirements under the stated assumptions. They do not reconstruct the trial's original design or establish its observed power.

Finally, the record identifies an all-enrolled primary analysis population but leaves several methods details unreported, including missing-data handling and multiplicity procedures. These gaps matter because they limit how confidently a reader can assess robustness and generalizability. Statistical interpretation is strongest when the effect estimate, uncertainty, endpoint components, and analysis choices are all available together.

18. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record does not provide component-specific counts for the composite primary endpoint. Readers therefore cannot tell how many events were early strokes or deaths, events after revascularization procedures, or later territory-specific ischemic strokes. That limits interpretation of which parts of the composite contributed to the observed group counts.

The sponsor's log-rank result is reported without an effect estimate, confidence interval, test statistic, or details of censoring. The unadjusted recalculation uses posted summary counts and cannot reproduce analyses that require participant-level event times or covariates. The record does not provide the original statistical analysis plan, protocol deviations, interim analysis information, or multiplicity procedures.

Although all enrolled patients were included in the primary outcome analysis according to the registry, the record does not report missing outcome counts or how missing data and follow-up were handled. The mean follow-up of 2.4 years does not describe the range or completeness of individual follow-up. Treatment adherence, crossovers, and post-enrollment exclusions are also not reported.

The trial was unmasked, but the registry does not provide further details about outcome assessment or whether endpoint events were adjudicated. Baseline characteristics, randomization implementation, recruitment settings, and subgroup results are not available in the posted results. The record identifies a specific symptomatic intracranial stenosis population; it does not by itself establish how well the result applies to populations outside that study context.

Finally, the arm descriptions are incomplete in the registry record, and the available adverse-event listing does not provide enough detail to compare event types by arm or assess causation. These limitations do not negate the posted results; they define what can and cannot be concluded from the recorded information.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Stenting vs. Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis — NCT00576693.
• Registry record last updated 2018-05-30; status last verified 2018-04.
• Results first posted 2014-07-11.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

NCT00576693 was a completed, randomized, unmasked, parallel-group phase 3 trial sponsored by the Medical University of South Carolina. It enrolled 451 participants and compared intensive medical management plus intracranial stenting with intensive medical management alone. The primary outcome was a composite of specified stroke or death events and later ischemic stroke in the symptomatic artery's territory; mean follow-up was 2.4 years.

The primary event counts were 52 among 224 participants in the stenting-plus-medical-management arm and 34 among 227 in the medical-management-alone arm. The registry reports a sponsor log-rank analysis with a p-value below 0.05; an unadjusted recalculation from the counts has a p-value of 0.026. The registry states that all enrolled participants were included in the primary outcome analysis. It does not report secondary outcome results, the original statistical analysis plan, or several details needed to assess missing data and analysis choices.