This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record.
1. Trial at a Glance
COREVALVE was a completed, single-group clinical study evaluating the CoreValve Evolut 34R TAVR system in participants with aortic valve stenosis. The registry describes the intervention as a device and identifies treatment as the primary purpose.
| Feature | COREVALVE |
|---|---|
| Study acronym | COREVALVE |
| Registry identifier | NCT02746809 |
| Study title | Medtronic CoreValve Evolut R U.S. Clinical Study (Evolut 34R Addendum) |
| Condition | Aortic Valve Stenosis |
| Design model | Single group |
| Allocation | NA |
| Masking | None |
| Primary purpose | Treatment |
| Enrollment | 72 |
| Number of arms | 1 |
| Intervention | CoreValve Evolut 34R TAVR system (device) |
| Phase | NA |
| Status | Completed |
| Lead sponsor | Medtronic Cardiovascular |
| Sponsor type | Industry |
2. Clinical Question
The clinical question is centered on the use of the CoreValve Evolut 34R TAVR system in participants with aortic valve stenosis. The registry defines four primary endpoints addressing short-term all-cause mortality, disabling stroke, device success, and early post-procedure prosthetic regurgitation.
Population
Participants in the COREVALVE clinical study with the registry-listed condition of aortic valve stenosis.
Intervention
CoreValve Evolut 34R TAVR system (device).
Comparator
There was no separate comparator arm. The study used a single-group design.
Primary question
How does the study population perform with respect to the prespecified 30-day clinical endpoints and early post-implantation device and prosthetic-valve endpoints?
3. Trial Design
CoreValve Evolut 34R
- CoreValve Evolut 34R TAVR system
- Device intervention
- Aortic valve stenosis
- No separate comparator arm
Early endpoint assessment
- 30-day all-cause mortality
- 30-day disabling stroke
- Device success from 24 hours to 7 days post implantation
- Early prosthetic regurgitation assessment from 24 hours to 7 days post implantation
4. Endpoints
The registry specifies four primary endpoints. Two are 30-day clinical outcomes, while the other two evaluate early device and prosthetic-valve performance between 24 hours and 7 days after implantation.
| Primary endpoint | Time frame | Registry definition |
|---|---|---|
| Percentage of Participants With All-cause Mortality | 30 Days | Percentage of participants with all-cause mortality at 30 days. |
| Percentage of Participants With Stroke (Disabling) | 30 Days | Percentage of participants with disabling stroke (VARC-II definitions) at 30 days. |
| Device Success Rate | 24 hours to 7 days post implantation | Device Success defined as absence of procedural mortality; correct positioning of a single prosthetic heart valve into the proper anatomical location; intended performance of the prosthetic heart valve, defined as absence of patient prosthesis-mismatch and mean gradient < 20 mmHg (or peak velocity < 3 m/sec); and absence of moderate or severe prosthetic valve regurgitation. |
| The Percentage of Subjects With no More Than Mild Prosthetic Regurgitation at Early Post Procedure Echocardiogram | 24 hours to 7 days post implantation | The percentage of subjects with no more than mild prosthetic regurgitation at early post procedure echocardiogram as assessed by echo core laboratory. |
Endpoint structure
These endpoints measure different dimensions of an intervention's early clinical and technical performance. All-cause mortality and disabling stroke are patient-level clinical outcomes. Device success is a composite device-performance endpoint with several required conditions. The prosthetic-regurgitation endpoint focuses specifically on an echocardiographic finding assessed by an echo core laboratory.
Clinical outcomes
All-cause mortality and disabling stroke capture serious patient-level outcomes at 30 days.
Device performance
Device success requires all of the registry's listed conditions to be satisfied rather than merely one favorable measurement.
Valve performance
The prosthetic-regurgitation endpoint assesses whether regurgitation is no more than mild during the early post-procedure echocardiographic assessment.
Different time scales
The clinical endpoints use a 30-day time frame, whereas device success and prosthetic regurgitation use the early 24-hour to 7-day post-implantation window.
5. Results
The registry indicates that results were posted, but no formal statistical analyses were posted to ClinicalTrials.gov. The numerical information available here therefore consists of the reported serious adverse-event count by study group; the registry-defined primary endpoints are described separately because their numerical estimates, confidence intervals, and p-values are not reported in the available record.
Safety: Serious Adverse Events
Reported serious adverse events
EVOLUTR 34R: 34 affected participants out of 60 at risk
| Study group | Affected | At risk | Reported measure |
|---|---|---|---|
| EVOLUTR 34R | 34 | 60 | Serious adverse events |
The reported safety measure is a count of participants affected among those at risk. It is not a comparison between randomized treatment groups because COREVALVE has a single-group design. It should also be distinguished from an incidence rate, event rate per person-time, or risk ratio.
The value 34/60 means that 34 of the 60 participants represented in the reported at-risk denominator were affected by a serious adverse event. Because there is no separate comparator arm, this quantity describes the observed study group rather than an estimated treatment-versus-control effect.
The denominator is important: the reported safety measure uses 60 at-risk participants, whereas total enrollment was 72. The two denominators therefore answer different questions and should not be substituted for one another.
No confidence interval or p-value is reported for this safety measure. Consequently, there is no registry-posted precision interval or formal hypothesis test to interpret for this result.
6. Planned Analysis
The registry-defined endpoints are naturally analyzed as proportions of participants meeting each binary outcome definition within the specified follow-up window. For a single-group study, the fundamental descriptive quantity is the observed proportion:
where x is the number of participants meeting the endpoint definition and n is the relevant analysis denominator.
All-cause mortality at 30 days
The registry defines this endpoint as the percentage of participants with all-cause mortality at 30 days. The appropriate primary descriptive analysis would report the number of deaths, the analysis denominator, and the resulting percentage, accompanied by an appropriate confidence interval for a single proportion.
Disabling stroke at 30 days
The disabling-stroke endpoint is also binary and occurs within a fixed 30-day window. The registry specifies VARC-II definitions. A single-group proportion with a confidence interval is therefore the natural statistical framework for summarizing this endpoint.
Device success
Device success is a composite binary endpoint because success requires every listed condition to be satisfied: absence of procedural mortality, correct positioning of a single prosthetic heart valve, intended prosthetic-valve performance, and absence of moderate or severe prosthetic valve regurgitation. A participant failing any required component would not meet the overall device-success definition.
Prosthetic regurgitation
The early echocardiographic endpoint asks for the percentage of subjects with no more than mild prosthetic regurgitation. For a single-group study, this is again naturally summarized as a proportion with an appropriate confidence interval.
7. Statistical Methodology
Single-group proportion analysis
For each binary primary endpoint, the basic statistical object is the proportion of participants satisfying the endpoint definition. Unlike a randomized two-arm trial, COREVALVE does not provide a treatment-control contrast within the study design.
The numerator counts participants meeting the endpoint definition and the denominator identifies the population used for that endpoint.
Confidence intervals for a proportion
A confidence interval quantifies statistical uncertainty around the observed proportion. For a binary endpoint, an exact binomial interval or an appropriate score-based interval can be preferable to a simple normal approximation, particularly when event counts are small or proportions are near 0 or 1.
The important distinction is that a confidence interval around a single-group proportion describes uncertainty in the estimated proportion. It does not establish comparative effectiveness against an external treatment unless a valid external comparator and analysis framework are specified.
Composite endpoint analysis
Device success is defined through multiple simultaneous requirements. Statistically, such a composite can be represented as a binary outcome in which success occurs only when every required component is satisfied. This makes the overall success rate sensitive to the least frequently satisfied component.
For COREVALVE, the registry's definition requires absence of procedural mortality, correct positioning, intended prosthetic-valve performance, and absence of moderate or severe prosthetic regurgitation.
Time-windowed binary outcomes
The 30-day mortality and disabling-stroke endpoints are defined over a fixed follow-up period. This is different from a time-to-event endpoint in which the exact event time and censoring process are central to the analysis. When the registry asks for the percentage of participants with an event at a fixed time point, the primary summary is a binary proportion at that landmark.
Early post-implantation endpoints
Device success and prosthetic regurgitation use an early assessment window of 24 hours to 7 days post implantation. The analysis must therefore preserve the registry's specified window rather than silently treating these outcomes as 30-day outcomes.
8. Statistical Methods Explained
Why is a single-group proportion appropriate here?
The study is explicitly classified as a single-group design with one arm. There is therefore no randomized comparator against which to calculate a treatment effect. For a binary endpoint, the natural descriptive measure is the proportion of participants meeting the endpoint definition.
What does a proportion tell us?
A proportion describes how common an outcome was within the defined analysis population and time frame. It does not by itself explain why the outcome occurred, establish causality, or determine how the outcome would compare with another treatment or historical population.
Why is the denominator important?
The denominator identifies the population to which the reported proportion applies. This is especially important for safety reporting because the registry reports 34 affected out of 60 at risk, while the overall study enrollment was 72. Those denominators should not be treated as interchangeable.
Why use confidence intervals?
An observed proportion from a finite sample is an estimate rather than a perfectly known population parameter. A confidence interval communicates the statistical uncertainty surrounding that estimate. Wider intervals generally reflect less information, while narrower intervals indicate greater statistical precision under the chosen interval method.
How is device success different from a single measurement?
Device success is a composite definition. A participant must satisfy all listed requirements. A favorable result on one component therefore does not automatically imply overall device success.
Why should the 24-hour to 7-day window be preserved?
The time frame is part of the endpoint definition. An early post-implantation assessment answers a different question from a 30-day assessment. Mixing the two windows would change the endpoint being analyzed.
9. Interpreting the Four Primary Endpoints Together
The four primary endpoints provide complementary information rather than four interchangeable measurements.
| Endpoint | What it measures | Statistical perspective |
|---|---|---|
| All-cause mortality | Death from any cause at 30 days | Binary 30-day clinical outcome |
| Disabling stroke | Disabling stroke using VARC-II definitions at 30 days | Binary 30-day clinical outcome |
| Device success | Simultaneous satisfaction of the registry's device-success criteria | Composite binary device-performance outcome |
| No more than mild prosthetic regurgitation | Early echocardiographic prosthetic-regurgitation status | Binary early post-procedure imaging outcome |
A useful statistical reading keeps these endpoints conceptually separate. Mortality and disabling stroke concern patient outcomes. Device success summarizes several technical and physiological requirements. Prosthetic regurgitation focuses on a specific valve-performance finding. A single favorable endpoint therefore cannot substitute for the others.
10. The Statistical Meaning of a Single-Group Study
Randomized controlled trials derive much of their inferential strength from comparing contemporaneous treatment groups created by randomization. A single-group study has a different statistical structure. It can characterize outcomes observed after an intervention, but it does not contain a randomized internal control group.
What the design supports
Descriptive estimation of clinical and device outcomes within the studied population and specified follow-up windows.
What it does not provide
A randomized within-study estimate of the difference between the CoreValve Evolut 34R system and another treatment.
Why external comparisons are difficult
Differences in eligibility, baseline characteristics, follow-up, endpoint definitions, and clinical practice can create differences unrelated to treatment.
Why endpoint definitions matter
Comparisons are meaningful only when populations, definitions, denominators, and time frames are sufficiently aligned.
11. Safety Interpretation
The registry reports serious adverse events for the EVOLUTR 34R study group as 34 affected participants out of 60 at risk. This is the principal numerical safety result available in the record described here.
| Safety measure | Study group | Reported value |
|---|---|---|
| Serious adverse events | EVOLUTR 34R | 34/60 |
Because the study has one arm, this safety result is descriptive. It cannot be expressed as a randomized relative risk, odds ratio, hazard ratio, or between-group risk difference using the COREVALVE design alone.
12. Limitations
- No randomized comparator: the single-group design does not provide an internal randomized control group for estimating a treatment effect.
- Different endpoint windows: mortality and disabling stroke are defined at 30 days, whereas device success and prosthetic regurgitation are assessed from 24 hours to 7 days post implantation.
- Composite device-success definition: overall device success depends on simultaneously satisfying multiple requirements, so its interpretation requires attention to every component of the definition.
- Denominator differences: enrollment is 72, while the reported serious-adverse-event denominator is 60. Analysis populations should not be assumed to be identical.
- Single-group inference: an observed proportion describes the study population but does not by itself establish comparative effectiveness.
- Uncertainty: point estimates alone can obscure sampling variability. Confidence intervals are important for interpreting the precision of binary endpoint estimates.
- External comparisons: comparisons with other studies require alignment of endpoint definitions, time frames, analysis populations, and clinical context.
- Registry reporting: the record does not provide formal statistical analyses with endpoint estimates, confidence intervals, and p-values for the four primary endpoints.
13. Why This Trial Matters Statistically
COREVALVE is useful as a teaching example because its single-group structure highlights a statistical situation that differs fundamentally from the randomized comparative trial. The registry combines fixed-time clinical outcomes with early technical and echocardiographic endpoints, requiring careful attention to endpoint definitions and denominators.
| Concept | How it appears in COREVALVE |
|---|---|
| Single-group design | One study arm and no separate comparator. |
| Binary endpoint | Mortality, disabling stroke, device success, and prosthetic regurgitation are expressed as participant-level outcome classifications. |
| Single-proportion estimation | The primary descriptive framework for a binary endpoint in a single-group study. |
| Confidence intervals | Useful for quantifying uncertainty around an observed proportion. |
| Composite endpoint | Device success requires all listed success conditions to be satisfied. |
| Fixed follow-up | All-cause mortality and disabling stroke are evaluated at 30 days. |
| Early assessment window | Device success and prosthetic regurgitation are assessed from 24 hours to 7 days post implantation. |
| Analysis denominator | The serious-adverse-event result uses 60 at-risk participants rather than total enrollment of 72. |
| Endpoint definition | The statistical meaning of each result depends on its precise registry definition and time frame. |
14. Clinical Interpretation vs Statistical Interpretation
Statistical interpretation
The study is structured to describe outcomes in a single group. Binary primary endpoints are naturally summarized as proportions, with confidence intervals used to characterize statistical precision.
Clinical interpretation
The four primary endpoints address distinct clinical and device-performance domains: 30-day mortality, 30-day disabling stroke, early device success, and early prosthetic regurgitation.
Keeping these interpretations separate is particularly important in a single-group device study. A descriptive outcome rate is an observation in the studied population; it is not automatically a causal treatment effect.
15. Endpoint-Specific Statistical Reasoning
All-cause mortality at 30 days
For this endpoint, each participant can be classified according to whether death from any cause occurred by 30 days. The primary summary is therefore a 30-day proportion. A confidence interval would describe uncertainty around that estimated proportion.
Disabling stroke at 30 days
The same general binary-outcome framework applies, but the endpoint classification follows the registry's specified VARC-II definitions. The definition is therefore part of the statistical measurement itself rather than merely a clinical footnote.
Device success
The endpoint is more complex because it is defined through multiple required conditions. This means that a participant cannot be counted as a device success if any required element fails. Statistically, the composite endpoint can still be summarized as a single binary proportion, but interpretation benefits from understanding its component structure.
No more than mild prosthetic regurgitation
This endpoint is an early echocardiographic binary classification assessed by an echo core laboratory. Its interpretation should remain tied to the 24-hour to 7-day post-implantation time frame and the specified threshold of no more than mild regurgitation.
16. Statistical Questions Raised by the Design
Can the study estimate a randomized treatment effect? No. With one study group, there is no randomized internal comparator. The primary inferential quantities are therefore within-group outcome estimates rather than between-treatment effects.
Why would a confidence interval be important? A point estimate gives one estimate of the underlying proportion. A confidence interval shows how much statistical uncertainty surrounds that estimate under the selected sampling framework.
Why does the definition of device success matter so much? Because the endpoint requires all listed conditions simultaneously. The overall success rate therefore reflects the combined performance of several separate criteria.
Why cannot the endpoint windows be combined? A 30-day mortality endpoint and an early 24-hour to 7-day device-performance endpoint measure different periods after implantation. Changing the window changes the statistical estimand.
Why is 34/60 not automatically 34/72? The registry reports serious adverse events among 60 participants at risk. The reported analysis denominator must be preserved rather than replaced by total enrollment.
Can an outside study be used as a control? An external comparison can be descriptive, but it does not recreate randomization. Differences between studies can reflect population, endpoint, follow-up, or practice differences rather than treatment effects.
17. What the Registry Results Can and Cannot Establish
| Question | Statistical interpretation |
|---|---|
| What proportion of the studied population experienced a defined outcome? | A single-group proportion can address this question. |
| How precise is that estimated proportion? | A confidence interval can quantify statistical uncertainty. |
| Was the intervention better than a randomized control? | The single-group design cannot answer this through an internal randomized comparison. |
| Did one endpoint improve relative to another? | The endpoints measure different constructs and should not be treated as interchangeable outcomes. |
| Did every component of device success perform favorably? | The overall device-success definition alone requires all components to be satisfied; component-level results would be needed to assess each requirement separately. |
This distinction between description and comparison is central to interpreting single-group clinical studies. The statistical analysis should match the design rather than importing the methods of a randomized two-arm trial.
18. Record Timeline
Study start
The COREVALVE study began in June 2016.
Primary completion
The registry lists November 2016 as the primary completion date.
Study status
The registry lists the study status as completed, with enrollment of 72 participants and one study arm.
19. Understanding the Serious-Adverse-Event Result
The registry reports 34/60 serious adverse events for the EVOLUTR 34R group. The numerator is the number of affected participants and the denominator is the number at risk for the reported measure.
This value is not a hazard ratio, risk ratio, odds ratio, or treatment-versus-control difference. It also should not be interpreted as a proportion of all 72 enrolled participants because the registry specifies 60 as the at-risk denominator for this safety measure.
Clinical-trial analyses routinely use analysis populations that differ from total enrollment. Correct interpretation therefore requires identifying the denominator attached to the reported result rather than automatically using the enrollment count.
20. Statistical Concepts in This Trial
The COREVALVE study illustrates several foundational statistical ideas that are important when interpreting single-group clinical studies and device endpoints.
21. Important Interpretation Issues
- Design determines the estimand: a single-group study primarily estimates outcome frequencies within the studied population rather than randomized treatment effects.
- Endpoint definitions are statistical specifications: the exact clinical and technical criteria determine who is counted as having an event or achieving success.
- Time frames are part of the estimand: 30-day outcomes should not be conflated with outcomes assessed during the 24-hour to 7-day post-implantation window.
- Composite success requires all components: device success is not equivalent to success on any one individual component.
- Denominators must be preserved: the reported safety denominator of 60 should not be replaced with total enrollment of 72.
- Statistical precision matters: point estimates without uncertainty intervals provide an incomplete picture of sampling variability.
- External comparisons require caution: a nonrandomized comparison with another study cannot reproduce the balance produced by random allocation.
22. Sources
- ClinicalTrials.gov: NCT02746809.
Continue through the Clinical Biostats clinical-trial library
Explore additional clinical trial analyses and statistical explanations covering study design, endpoints, statistical inference, and interpretation.
23. Record Summary
COREVALVE is a completed single-group clinical study of the CoreValve Evolut 34R TAVR system in aortic valve stenosis, with enrollment of 72 participants and one study arm. Its four registry-defined primary endpoints cover 30-day all-cause mortality, 30-day disabling stroke using VARC-II definitions, device success from 24 hours to 7 days post implantation, and no more than mild prosthetic regurgitation during the early post-procedure echocardiographic assessment. The reported serious-adverse-event result is 34/60 for the EVOLUTR 34R group.
Statistically, the most important feature of the study is its single-group structure. The appropriate framework centers on estimating and interpreting within-group proportions, defining the analysis denominator correctly, preserving the specified follow-up windows, and using confidence intervals to communicate uncertainty. Without a randomized comparator, the observed outcomes should be interpreted as descriptive evidence from the studied population rather than as randomized estimates of comparative treatment effect.