Safety and Efficacy Study of the Medtronic CoreValve® System in the Treatment of Severe, Symptomatic Aortic Stenosis in Intermediate Risk Subjects Who Need Aortic Valve Replacement (SURTAVI).: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy Study of the Medtronic CoreValve® System in the Treatment of Severe, Symptomatic Aortic Stenosis in Intermediate Risk Subjects Who Need Aortic Valve Replacement (SURTAVI). (NCT01586910), a randomized na study. The primary outcome is All-cause Mortality or Disabling Stroke Rate Expressed as a Posterior Probability.

2. Clinical Question

The central comparison is between a catheter-based valve implantation procedure and surgical valve replacement in the stated intermediate-risk population. The registry reports group-level results for the combined primary outcome and a Bayesian analysis of the difference. The outcome combines deaths from any cause after valve intervention with disabling stroke, so it captures more than cardiovascular death alone and is not a measure of either component in isolation.

The broader trial record also includes secondary measures such as major adverse cardiovascular and cerebrovascular events, adverse events, conduction disturbance, functional status, walking distance, days alive out of hospital, and quality of life. These can describe additional aspects of outcomes, but they do not replace the primary comparison. The clinical question should be interpreted within the population and analysis definition recorded for the primary endpoint; the registry does not report further eligibility details in the record.

3. Trial Design

SURTAVI is recorded as a randomized, parallel-group trial with no masking. The registry lists an enrollment of 1746 and identifies Medtronic Cardiovascular as the sponsor. The recorded start is 2012-04, and primary completion is recorded as 2018-07. The status in the registry is active, not recruiting. The phase is listed as NA.

The two groups with primary outcome results are Medtronic CoreValve® System TAVI and SAVR. The primary analysis population is described as all randomized subjects with an attempted implant procedure. The outcome records include 864 participants for the TAVI group and 796 for the SAVR group. These are the endpoint-specific group sizes; the registry does not provide a separate account in this record of exclusions or missing observations for the primary analysis.

The registry lists the Medtronic CoreValve® Evolut R System TAVI as an intervention as well as the CoreValve® System and SAVR. However, the primary results presented in the registry record compare the CoreValve® System TAVI group with SAVR. The brief summary also contains a description of a single-arm, non-randomized phase. How that description relates to the reported randomized comparison is not explained in the registry record, so it should not be used to infer additional design details about the primary analysis.

Further information about the randomization procedure, allocation ratio, blinding of outcome assessment, protocol deviations, or the detailed conduct of the trial is not reported in the registry record. The parallel randomized design establishes the intended group comparison, but the registry entries do not describe every operational feature needed to assess conduct or analysis fully.

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: Participant Population= Consisted of all randomized subjects with an attempted implant procedure.

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.

Population: mITT

6. Primary Endpoint

The registry lists the primary outcome as All-cause Mortality or Disabling Stroke Rate Expressed as a Posterior Probability, measured over 24 months.

The primary endpoint is the all-cause mortality or disabling stroke rate at 24 months, expressed in the record as a posterior probability and reported as a percentage of participants. All-cause mortality includes deaths from cardiovascular and non-cardiovascular causes after valve intervention. Disabling stroke is defined using a modified Rankin Scale score of 2 or more at 90 days and an increase of at least one category from the participant’s pre-stroke baseline.

Combining death and disabling stroke makes the endpoint a measure of serious outcomes across two clinically important domains. It also means that a group-level result cannot show, by itself, whether any observed difference is more attributable to mortality or to stroke. The registry does not provide separate primary endpoint results for those components in the posted primary analysis.

The record gives the group estimates as medians with 95% confidence intervals and gives a Bayesian posterior median of the difference for the non-inferiority comparison. An interval around each group estimate describes uncertainty in that group’s estimate; the group intervals are not the same thing as an interval for the between-group difference. The registry does not report an interval for that difference in the posted Bayesian analysis.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary analysis is identified as Bayesian and reports a posterior median of the difference for a non-inferiority comparison. The record also gives a median and 95% confidence interval for each group’s primary outcome estimate. These reported quantities have different interpretations: group-specific estimates describe each group, while a difference describes a contrast between groups. The registry does not explain the relationship between its posterior terminology and the group-level confidence intervals in further detail.

What does a Bayesian posterior median convey?

In general, Bayesian analysis combines a statistical model with prior distributions to produce a posterior distribution for quantities of interest. A posterior median is the central value of that distribution, with half the posterior probability on either side under the fitted model. It is not, by itself, a measure of precision. An interval or a probability tied to a decision threshold is needed to convey uncertainty or support a non-inferiority judgment. The prior distributions, model specification, posterior interval, and decision threshold used for this trial are not reported in the registry record.

How should the confidence intervals be read?

In general, a confidence interval reflects uncertainty in an estimate under a specified statistical procedure. The primary group estimates have 95% confidence intervals. Those intervals describe uncertainty around the estimates for each group; they do not directly provide the uncertainty for the difference. To assess a between-group contrast, the interval for the contrast itself is particularly informative. The registry’s reported Bayesian comparison gives a posterior median difference of -1.4 but no accompanying interval in the record. As a result, the precision of that reported difference cannot be judged from the difference estimate alone.

The recalculated secondary comparisons are simple, unadjusted comparisons based on the posted summary statistics. For the binary neurological injury outcome, the recalculation reports a risk difference, confidence interval, and unadjusted p-value. For length of stay, it reports a mean difference, confidence interval, and unadjusted p-value. In general, a risk difference expresses the contrast between the outcome proportions, while a mean difference contrasts group averages. A p-value describes how compatible the observed data are with a specified null model; it does not measure the size or clinical importance of an effect, and it does not account for multiplicity unless the analysis explicitly does so.

Why does variability matter for length of stay?

The length-of-stay outcome is reported as a mean with a standard deviation in each group. The TAVI group has a mean of 5.7 days and standard deviation of 4.9; the SAVR group has a mean of 9.8 days and standard deviation of 8.0. In general, a standard deviation describes the spread of observations around a mean. When observations are highly variable, the mean difference can be less precisely estimated than it would be with less variability, all else being equal. Length-of-stay data can also have a skewed distribution, with some stays much longer than others. The registry does not report a distributional assessment or an alternative analysis for this outcome.

Over-dispersion is a general concern when the observed variability is greater than a statistical model assumes. If a model understates variability, uncertainty estimates can be too narrow and tests can appear more decisive than warranted. The posted recalculation uses the reported means and standard deviations; it does not establish whether the original analysis examined or addressed over-dispersion or skewness.

What do the planning illustrations mean?

The sample-size calculations are planning illustrations based on observed effects, not the trial’s own design justification and not observed power. The stated assumptions are a two-sided significance level of 0.05, equal allocation, and a normal approximation. The calculations consider planning for 100%, 75%, and 50% of the observed effect. For neurological injury, the listed participants per arm for 80% power are 213, 340, and 678, respectively; for 90% power they are 284, 455, and 907. For length of stay, the corresponding participants per arm for 80% power are 41, 72, and 162; for 90% power they are 54, 96, and 216. These values illustrate how required sample size changes when the effect to detect is smaller or the target power is higher. They do not describe the actual enrollment rationale, and the trial’s sample-size justification is not reported in the registry record.

In general, a power calculation depends on the assumed effect, outcome variability or event rate, significance level, allocation, and analysis method. A calculation based on an effect observed after the trial may be optimistic or otherwise unrepresentative of the effect anticipated during planning. It should therefore be read as a scenario for future planning under stated assumptions, not as evidence that the completed study had a particular probability of detecting an effect.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a Bayesian posterior median of the difference of -1.4 for a non-inferiority analysis. That is the registry’s reported analysis and should be distinguished from the ancillary unadjusted recalculations for the secondary outcomes. The recalculations use posted summary statistics and do not reproduce the sponsor’s Bayesian primary analysis.

The sponsor’s reported result does not include a posterior interval, a probability statement, or the non-inferiority margin in the registry record. Those are important for understanding the precision of the estimate and how a non-inferiority conclusion would be assessed. The group-level primary estimates and their 95% confidence intervals give additional descriptive information, but they do not supply those missing features of the Bayesian comparison.

The recalculated secondary comparisons are not a competing analysis of the primary endpoint. They address other outcomes and use straightforward, unadjusted methods. No discrepancy between a posted sponsor result and a recalculated result is flagged in the registry record. Even so, the sponsor’s reported primary analysis remains the relevant source for the trial’s stated Bayesian non-inferiority comparison; a separate recalculation from summary values would not replace its model or decision procedure.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists many outcome measures and analyses, but a multiplicity strategy is not reported in the registry record. It is not stated whether outcomes were organized in a testing hierarchy, whether secondary comparisons were adjusted, or whether particular secondary outcomes were designated as confirmatory. Without that information, unadjusted secondary p-values should be treated as exploratory evidence rather than as a set of independent confirmations. In general, examining multiple outcomes increases the opportunity for chance findings, and the interpretation of an isolated p-value depends partly on the analysis plan.

Handling of missing data is also not described in the record. There is no reported account of missing endpoint observations, exclusions from analysis, imputation, censoring, or sensitivity analyses for missingness. These details matter because missing outcomes can affect both the estimated event rates and the uncertainty around comparisons. The analysis population is described as all randomized subjects with an attempted implant procedure, but the registry does not explain how participants without an attempted implant or those with incomplete outcome information were handled.

The record does not report an interim analysis plan, stopping rules, or whether interim looks at outcome data were conducted. It also does not report a detailed statistical analysis plan, the non-inferiority margin, or the model and prior distributions for the Bayesian primary analysis. These omissions prevent a full assessment of how the primary decision was prespecified and how multiplicity, missing information, and interim monitoring were addressed.

The planning illustrations use stated assumptions for hypothetical future studies. They are not evidence of the original sample-size rationale, which is not reported in the registry record. No conclusion about the adequacy of the original sample size or the trial’s planned power can be drawn from those illustrations alone.

12. Safety Results

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

Deaths

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: anaemia (25.5%), atrial fibrillation (21.2%), atrioventricular block complete (18.2%), pleural effusion (12.8%), hypotension (10.9%).

The registry’s serious adverse-event records list 718 affected participants among 864 at risk in the Medtronic CoreValve® System TAVI group and 694 among 796 at risk in the SAVR group. For other adverse events, the record lists 803 among 864 at risk for TAVI and 754 among 796 at risk for SAVR. These are recorded counts, not adjusted comparisons. The registry does not provide a comparison test or a detailed explanation of the collection window in the record.

The listed serious adverse-event terms include anaemia at 25.5%, atrial fibrillation at 21.2%, complete atrioventricular block at 18.2%, pleural effusion at 12.8%, hypotension at 10.9%, and haemorrhagic anaemia at 10.6%. The denominator and whether these terms are ranked by participants or event reports are not reported in the registry record, so the percentages should not be reinterpreted as comparative rates between groups.

The registry also records 98 deaths in the TAVI group and 80 in the SAVR group. These counts do not, by themselves, indicate comparative risk or causation. Details such as timing, cause-specific counts, and attribution to the intervention are not reported. Serious adverse events and deaths are important safety information, but the posted summaries do not provide enough detail for a complete comparative safety assessment.

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

The posted primary results describe the estimated frequency of a combined outcome in each group and give a Bayesian median difference. They provide information about the trial’s recorded comparison, but the registry entry does not report enough of the Bayesian analysis to establish from this record how non-inferiority was formally judged. In particular, the non-inferiority margin and uncertainty interval for the reported difference are not reported in the registry record.

The outcome combines all-cause mortality with disabling stroke, so it does not tell readers which component contributed to the group estimates. Nor does a group-level rate describe every participant’s experience or establish why an outcome occurred. The endpoint definition and analysis population apply to the reported primary result; they do not imply that every participant had complete outcome follow-up or that all randomized participants entered the endpoint analysis.

The secondary estimates provide additional, outcome-specific information, but they are unadjusted recalculations and should not be treated as proof of an overall advantage or disadvantage. Their p-values do not account for the many listed outcomes. The safety counts likewise do not establish causation or allow a full comparison without further details about timing, follow-up, event definitions, and analysis.

In general, statistical results quantify patterns and uncertainty under the methods used. They do not by themselves resolve how a result should be applied to an individual, and they are not medical advice. The record supports describing the reported estimates and limitations; broader conclusions require the complete protocol, statistical analysis plan, and fuller trial reporting, which are not reported in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically informative because it compares two treatment strategies through a randomized, parallel-group design and reports a primary endpoint that combines mortality and disabling stroke. Random allocation is intended to support a fair comparison between groups, although the registry record does not give enough detail to assess the randomization procedure, departures from assignment, or every aspect of analysis conduct.

The primary result also illustrates why an effect estimate should be read alongside its uncertainty and decision rule. The registry gives group estimates with 95% confidence intervals and a Bayesian posterior median difference, but it does not report the interval or non-inferiority margin needed to fully interpret the stated non-inferiority comparison. The point estimate alone cannot show how robust the conclusion is to uncertainty or where it falls relative to a decision threshold.

The secondary outcomes demonstrate that trials often address several different questions at once. A peri-procedural event measure and hospital stay provide distinct information from a longer-term combined endpoint. Their unadjusted comparisons can be useful descriptions, but they have less confirmatory weight when the multiplicity plan is not reported. This distinction is central to avoiding overinterpretation of whichever outcome appears most striking.

The record also highlights the importance of transparent reporting. Details about missing data, analysis populations, model assumptions, interim analyses, and safety-event follow-up help readers judge whether estimates are precise and whether the analysis matches the trial’s design. Where those details are absent, careful interpretation means separating what is reported from what remains uncertain rather than filling gaps with assumptions.

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.

Several limitations are specific to the information recorded for this trial. The primary endpoint is described as a posterior probability, while the reported group values are medians with 95% confidence intervals and the between-group result is a posterior median difference. The registry does not explain these statistical quantities together in detail, and it does not report a posterior interval or the non-inferiority margin. This limits interpretation of the primary comparison as a non-inferiority analysis.

The primary population is described as randomized participants with an attempted implant procedure. The record does not explain how this population differs from all randomized participants or how participants without an attempted implant were handled. It also does not report missingness, follow-up completeness, imputation, or sensitivity analyses. These gaps make it difficult to assess potential effects of exclusions or incomplete outcome information.

The brief summary includes both a randomized comparison and a description of a non-randomized single-arm phase, while the posted primary results compare two randomized groups. The registry record does not clarify how the single-arm description relates to those results. The intervention list also names Evolut R, but the reported primary comparison names CoreValve® System TAVI and SAVR; the relationship between the listed device and analyzed groups is not explained.

For secondary outcomes, detailed results are reported for only some of the listed measures in this record. The multiplicity plan and the methods for outcomes with skewed or variable distributions are not reported. Safety summaries provide counts and selected serious adverse-event terms, but the record does not give the event-collection window or further comparative analysis. These limitations constrain how far the posted numerical summaries can support conclusions about overall efficacy or safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Safety and Efficacy Study of the Medtronic CoreValve® System in the Treatment of Severe, Symptomatic Aortic Stenosis in Intermediate Risk Subjects Who Need Aortic Valve Replacement (SURTAVI). — NCT01586910.
• Registry record last updated 2026-06-24; status last verified 2026-05.
• Results first posted 2019-09-17.
• 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

SURTAVI is recorded as a randomized, parallel-group, unmasked trial sponsored by Medtronic Cardiovascular. Its primary comparison is Medtronic CoreValve® System TAVI versus SAVR in participants with severe, symptomatic aortic stenosis at intermediate surgical risk. The registry lists an enrollment of 1746, a start of 2012-04, primary completion of 2018-07, and an active, not recruiting status.

The primary endpoint is all-cause mortality or disabling stroke at 24 months. The registry reports group medians with 95% confidence intervals and a Bayesian posterior median difference for a non-inferiority analysis, but the non-inferiority margin and uncertainty interval for the difference are not reported in the registry record. Unadjusted recalculations for two secondary outcomes provide additional descriptive comparisons, while the multiplicity plan, missing-data handling, and several other analysis details are not reported.

The posted safety summaries include serious and other adverse-event counts, selected serious-event terms, and deaths. These figures require cautious interpretation because event timing, attribution, and further comparative methods are not reported in the record.