Phase II Study of the Safety and Efficacy of the Relay Thoracic Stent-Graft: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Phase II Study of the Safety and Efficacy of the Relay Thoracic Stent-Graft (NCT00435942), a non-randomized na study. The primary outcome is Primary Safety Endpoint: Distribution of Major Adverse Events.

2. Clinical Question

3. Trial Design

This was a Phase II clinical study of the safety and efficacy of the Relay Thoracic Stent-Graft in patients with thoracic aortic pathologies. The registry lists the phase as NA. It describes a non-randomized, parallel-group design with no masking. The planned enrollment is recorded as 180. The study sponsor was Bolton Medical. The registry records a start date of 2007-01, primary completion in 2011-04, and a completed status.

The two groups received different procedures: endovascular treatment with the Relay device or open surgical repair. Because allocation was not randomized, treatment assignment could be related to patient characteristics, clinical circumstances, or other factors. Which factors influenced assignment, and whether the groups were adjusted to account for them, are not reported in the registry record. Consequently, the surgical group is not a randomized control for estimating a causal treatment effect.

The study purpose distinguishes device-related adverse events through 1 year as an efficacy evaluation from major adverse events through 1 year as a safety comparison. Long-term follow-up is described through 5 years. The primary safety endpoint is a participant count: whether a participant experienced at least one event from a defined list within 1 year after the procedure. That endpoint is conceptually different from reporting all adverse events or serious adverse events. The registry record does not provide enough detail to establish whether the separate posted serious-adverse-event counts are the results for the primary composite.

The record lists two outcome measures and one analysis. Detailed procedures for allocation, follow-up, outcome adjudication, and statistical modeling are not reported in the registry record. The available design description therefore supports a cautious account of the study structure, but not a reconstruction of the full protocol or analysis plan.

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 subjects who underwent the Relay implant procedure (Intention to Treat)

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 Primary Safety Endpoint: Distribution of Major Adverse Events, measured over 1 year.

The primary safety endpoint counts participants who experienced at least one major adverse event within 1 year after the procedure. The listed events are aneurysm-related mortality, stroke, paralysis or paraplegia, myocardial infarction, procedural bleeding, respiratory failure, renal failure, and wound-healing complications. The endpoint is recorded in participants, rather than as a count of every event experienced by each participant.

Using a participant-level composite can capture whether a person experienced any event from a clinically important list, even when individual event types are uncommon. It also gives a single summary endpoint for a safety comparison. However, a composite can combine events that differ in type and clinical importance. Its overall count does not, by itself, show which component occurred, how often each component occurred, or whether a participant experienced more than one component.

The registry describes the endpoint population as subjects who underwent endovascular treatment or open surgical repair. Its timeframe is 1 year. The available record does not report the primary endpoint’s group-specific results, an effect estimate, a confidence interval, or a test result. The separately posted serious-adverse-event counts should not automatically be read as this composite endpoint: the record does not establish that they use the same event definition or follow-up window.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The record describes a binary, participant-level safety endpoint: a participant either experienced at least one listed major adverse event within 1 year or did not. For a binary endpoint, an analysis commonly summarizes the number of participants with the event and the number assessed in each group. The registry defines the primary endpoint in this way, but does not report its group-specific results or the statistical procedure used to compare them.

What does a participant count represent?

A participant count is not the same as a total event count. If a participant experiences more than one event in the endpoint list, the composite endpoint still counts that participant once. This can make the endpoint easier to summarize, but it can conceal the distribution of event types and repeated events. The registry’s definition says “at least 1” major adverse event, so the relevant unit is the participant, not each event occurrence.

For a binary outcome, a group’s observed event proportion is generally calculated from the number of participants with the event relative to the number assessed. A comparison can be expressed in several ways, including an absolute difference or a relative measure. These measures answer different questions and have different interpretations. No group-specific primary-endpoint proportions or comparative effect estimates are reported in the registry record, so none can be interpreted for this endpoint.

Why does the design affect interpretation?

Random allocation, in general, is intended to balance measured and unmeasured prognostic factors across groups on average. Here, allocation is recorded as non-randomized. The groups may therefore differ before treatment, and those differences may influence event rates. Even a clear numerical contrast in observed counts would not separate treatment effects from other possible explanations unless the analysis and its assumptions adequately addressed confounding. The registry record does not report the allocation process, covariate adjustment, or a method for addressing confounding.

Masking is recorded as none. In general, masking can reduce the influence of knowledge about treatment assignment on outcome assessment or reporting, depending on who is masked and how outcomes are measured. The registry does not detail outcome adjudication or other safeguards. The listed events include outcomes that may be objectively documented as well as complications whose ascertainment can depend on clinical assessment. The record does not explain how each component was identified or verified.

How should uncertainty be considered?

A confidence interval, in general, conveys sampling uncertainty around an estimate under a specified statistical method. A narrow interval indicates greater precision than a wide interval, though neither interval width nor statistical significance resolves bias from non-randomized treatment assignment. The registry record does not report a confidence interval or a test result for the primary endpoint. Precision and the compatibility of the data with different effect sizes therefore cannot be evaluated from the record’s endpoint description alone.

Over-dispersion is a general issue when counts vary more than a simple statistical model expects. It is especially relevant to some count or rate outcomes, where variability may exceed a model’s assumptions. The primary endpoint is described as a participant count for a binary event, not as a repeated-event rate. The record does not report a model or dispersion assessment. It would be inappropriate to infer that an over-dispersion adjustment was or was not used.

What does the planning illustration mean?

The registry describes a sample-size planning illustration based on the observed effect, using a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It describes the effect to detect as a stated fraction of the observed effect. These are planning assumptions, not evidence that the trial was designed prospectively around them. The illustration does not establish the original design rationale, which is not reported in the registry record.

In general, a planning calculation estimates the number of participants needed to detect a specified effect under stated assumptions. Its output depends on those assumptions and should not be confused with the precision or strength of evidence in the completed study. Any sample-size values in such an illustration are planning quantities, not observed power. The record does not report a sample-size justification from the original protocol, nor does it provide enough detail to assess whether the assumptions correspond to the primary safety endpoint.

The summary of methods is therefore limited: the endpoint definition and broad design are available, but the statistical model, handling of confounding, uncertainty estimates, and detailed analysis procedures are not reported in the registry record. General statistical concepts can clarify what would be needed for interpretation, but they cannot fill those record-specific gaps.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Relay Device Group", "Surgical Control Group")
n      <- c(120, 60)
events <- c(27, 51)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Relay Device Group  120  27  22.5  15.4  31.0  15.9  30.8  16.0  30.8
ARM  Surgical Control Group  60  51  85.0  73.4  92.9  73.9  91.9  73.8  91.8

11. Multiplicity, Missing Data and Analysis Populations

The registry lists two outcome measures and one analysis, but the record does not describe a multiplicity adjustment or explain how the analyses were prioritized. Multiplicity refers, in general, to the increased chance of a false-positive finding when several outcomes or comparisons are tested. Whether an adjustment is needed depends on the analysis plan and the claims being made. The record does not report a statistical plan that would establish how this issue was handled.

Missing-data handling is not reported in the registry record. The available safety summaries give group counts and participants at risk, but do not explain whether all participants had complete follow-up, how incomplete observations were treated, or whether participants lost to follow-up were included in any analysis. Without that information, the impact of missing outcome data cannot be assessed.

The primary endpoint population is described as subjects who underwent the endovascular procedure or open surgical repair. The registry does not state whether the analysis used an intention-to-treat approach, a per-protocol population, or another analysis population, nor does it explain exclusions after treatment assignment. This matters because the population analyzed can affect the interpretation of a comparison, particularly in a non-randomized study.

The record does not report a protocol-defined hierarchy of endpoints, interim analyses, or sensitivity analyses. These details should not be inferred from the number of outcomes or analyses listed. The absence of these details limits assessment of selective reporting, robustness to missing data, and the statistical interpretation of any comparative result.

12. 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 0.8% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: bleeding events (55.0%), pulmonary complications (33.3%), cardiac complications (20.0%), neurological complications (15.8%), digestive/gastrointestinal complications (13.3%).

The registry’s serious-adverse-event summaries report 68 of 120 affected in the Relay Device Group and 45 of 60 affected in the Surgical Control Group. For other adverse events, the corresponding entries are 97 of 120 and 58 of 60. These are the posted group counts and denominators; they should be kept distinct from the primary safety endpoint unless the record confirms that they represent the same endpoint definition and timeframe.

The registry also lists serious adverse-event terms with percentages: Bleeding events, 55.0%; Pulmonary Complications, 33.3%; Cardiac complications, 20.0%; Neurological complications, 15.8%; Digestive/gastrointestinal complications, 13.3%; and General, 12.5%. Denominator details and whether categories overlap are not reported alongside these entries. These percentages should therefore not be interpreted as directly comparable risks between the treatment groups.

Serious adverse events and the primary composite are not automatically interchangeable. The composite is defined by a specific set of major events within 1 year after the procedure. The registry does not explicitly connect the serious-event summaries to that definition. The counts describe reported safety data, but do not establish why events occurred or provide a like-for-like causal comparison across the non-randomized groups.

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

The registry record establishes the broad study question, the non-randomized parallel-group design, the one-year primary safety endpoint definition, and group-level serious and other adverse-event counts. It also states that long-term follow-up was conducted through 5 years. These details help readers understand what the study set out to evaluate and what safety information was entered in the registry.

The posted numbers do not establish that one procedure caused more or fewer events than the other. Treatment was not randomly allocated, and the record does not report the factors that influenced treatment choice or a sufficiently detailed adjusted comparison. Differences between groups could reflect pre-existing differences or other aspects of care as well as treatment. The counts are descriptive unless supported by an analysis that addresses these issues.

The counts also do not provide a complete result for the primary safety endpoint. The endpoint is the occurrence of at least one specified major adverse event within 1 year, while the posted summaries are labeled as serious adverse events and other adverse events. The registry does not establish that their definitions and timeframes match the primary composite. Nor does it provide a group-specific primary endpoint estimate or its uncertainty.

In general, statistical evidence is most informative when the outcome definition, analysis population, effect estimate, and measure of uncertainty are all clear. The available record leaves important details about the statistical plan, missing data, and outcome assessment unreported. This limits the precision of any conclusion about comparative safety. The findings should be understood as registry descriptions rather than a definitive causal comparison or a complete account of the primary endpoint’s result.

The planning illustration described in the record is not a substitute for the trial’s original design documentation or for results-based uncertainty. It reflects stated assumptions for a planning exercise, not observed power. The original sample-size justification and the details needed to assess the comparative analysis are not reported in the registry record.

14. Why This Trial Matters Statistically

This study is statistically instructive because it places a device procedure and open surgery in parallel groups without random assignment. Such studies can document clinical experience and reported event counts, but treatment groups may differ for reasons that also affect outcomes. That distinction between description and causal comparison is central when reading non-randomized evidence.

The primary endpoint is also a composite. Combining several major adverse events into a single participant-level outcome can provide a concise safety measure, but it can obscure which event types contribute to the result. Interpretation therefore depends on clear reporting of both the composite and its components. The record defines the component list, but does not report the primary endpoint’s group-specific results or connect the separate serious-adverse-event summaries to that composite.

The study also illustrates why outcome counts need their denominator, time window, and population definition. The record provides group counts and at-risk denominators for serious and other adverse events, while its primary safety endpoint has a stated one-year timeframe. The relationship between these distinct summaries is not made clear. Without aligned definitions, apparent similarities or contrasts can be misleading.

Finally, a planning illustration based on an observed effect should not be confused with an original prospective sample-size justification or with the strength of evidence in the completed study. Clear separation of planning assumptions, endpoint results, and uncertainty estimates supports more transparent interpretation. The missing analysis details in this record mark the limits of what the reported information can establish.

15. 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 main limitation for interpreting this trial is its non-randomized allocation. The groups cannot be assumed to have comparable baseline risk, and the registry record does not describe how participants were assigned to endovascular treatment or open repair. It also does not report an adjusted comparison or enough information to assess whether confounding was addressed.

The primary endpoint is defined, but its group-specific results and comparative estimate are not reported in the available record. The serious-adverse-event summaries cannot be presumed to be the primary composite because the record does not confirm matching event definitions or follow-up windows. The term-specific percentages are reported without denominator details, and it is not stated whether event categories overlap.

Further limitations include unreported details about outcome assessment, follow-up completeness, missing-data handling, analysis populations, and the statistical model. The registry does not describe a multiplicity strategy, interim analyses, or sensitivity analyses. These omissions prevent a full assessment of precision, robustness, and potential selective reporting.

The record states that long-term follow-up was conducted through 5 years, but the results available in the registry record do not describe long-term outcomes. It also gives a planning illustration based on the observed effect without reporting the original protocol’s sample-size rationale. Consequently, the planning exercise cannot be treated as evidence that the completed study had a particular level of power for its primary endpoint.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Phase II Study of the Safety and Efficacy of the Relay Thoracic Stent-Graft — NCT00435942.
• Registry record last updated 2018-01-16; status last verified 2017-08.
• Results first posted 2017-09-05.
• 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.

19. Record Summary

NCT00435942 is a completed, non-randomized, parallel-group Phase II clinical study of the Relay Thoracic Stent-Graft and open repair for thoracic aortic pathologies. The registry lists no masking, an enrollment of 180, a start date of 2007-01, and primary completion in 2011-04. The sponsor was Bolton Medical.

The primary safety endpoint is the number of participants experiencing at least one listed major adverse event within 1 year after the procedure. The record separately reports serious and other adverse-event counts for the Relay Device Group and Surgical Control Group, but does not establish that these are results for the primary composite. Long-term follow-up is described through 5 years.

Because treatment allocation was non-randomized and the record does not report a detailed comparative analysis, the group counts are descriptive rather than a like-for-like causal comparison. Primary endpoint estimates, confidence intervals, test results, and several key analysis details are not reported in the registry record.