PROTECTED TAVR: Stroke PROTECTion With SEntinel During Transcatheter Aortic Valve Replacement: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for PROTECTED TAVR: Stroke PROTECTion With SEntinel During Transcatheter Aortic Valve Replacement (NCT04149535), a randomized na study. The primary outcome is The Rate of Stroke Through 72 Hours Post TAVR Procedure or Discharge (Whichever Comes First).

2. Clinical Question

3. Trial Design

PROTECTED TAVR (NCT04149535) is recorded as a completed, randomized, parallel-group trial sponsored by Boston Scientific Corporation. The registry lists enrollment of 3000. The recorded start date is 2020-02-07, and the primary completion date is 2022-02-01. The phase is listed as NA.

Participants were assigned to TAVR with the Sentinel Cerebral Protection System or TAVR without Sentinel. The posted arm counts are 1501 in the Sentinel group and 1499 in the group without Sentinel. The registry records no masking. The posted primary outcome population is intent-to-treat, which refers to analyzing participants according to their assigned group rather than reclassifying them according to treatment received. Details of the trial’s implementation of that principle are not reported in the registry record.

The study’s stated aim was to demonstrate that Sentinel significantly reduces the risk of peri-procedural stroke after TAVR. The primary outcome covers stroke through 72 hours after the procedure or discharge, whichever came first. Events were adjudicated by an independent Clinical Events Committee using Neurologic Academic Research Consortium definitions. The registry posts one outcome measure and no secondary outcome measures. The original protocol, the allocation process, and additional details about conduct are not reported in the registry record.

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: Intent-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 The Rate of Stroke Through 72 Hours Post TAVR Procedure or Discharge (Whichever Comes First), measured over <=72 hours.

The primary endpoint counts participants who experienced any stroke through 72 hours after TAVR or discharge, whichever came first. It includes hemorrhagic, ischemic, and strokes of undetermined status, and includes both disabling and nondisabling events. The registry specifies the intent-to-treat population and reports the outcome as a count of participants.

Stroke is relevant to the stated study question because the trial aimed to assess peri-procedural stroke in the setting of TAVR. A defined observation window makes clear which events qualify for this endpoint; the discharge rule means the endpoint window may end at discharge if that occurs before 72 hours. An independent Clinical Events Committee adjudicated the events using NeuroARC definitions. Such adjudication applies stated criteria to event classification, rather than relying only on an unreviewed report of an event.

The endpoint is binary at the participant level: a participant either had a qualifying stroke within the specified window or did not. The reported group counts are 34 participants with stroke among 1501 assigned to TAVR with Sentinel and 43 among 1499 assigned to TAVR without Sentinel. This outcome does not describe stroke severity beyond the listed inclusion of disabling and nondisabling events, nor does it measure events outside the specified window.

7. Results Posted in the Registry

• TAVR Without Sentinel: 2.9% (43/1499) versus 2.3% (34/1501) on control; risk difference 0.6 percentage points (95% CI -0.5 to 1.7), risk ratio 1.27 (0.81–1.97); unadjusted two-proportion z-test p = 0.296.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary at the participant level: each person is counted as having experienced a qualifying stroke or not having experienced one during the endpoint window. The registry reports counts and proportions for the two assigned groups, along with comparative measures. The detailed statistical methods used for the sponsor’s primary analysis are not reported in the registry record.

What does a risk difference express?

A risk difference subtracts the event proportion in one group from the event proportion in the other. The registry’s comparison is TAVR without Sentinel relative to TAVR with Sentinel. Its reported risk difference is 0.006. A value of zero represents equal event proportions; positive and negative values indicate the direction of the difference according to the comparison order. The sign is meaningful only when the order of groups is clear.

The reported group proportions are 0.023 for TAVR with Sentinel and 0.029 for TAVR without Sentinel. The registry’s risk-difference estimate is rounded and is reported as 0.006. No new value should be derived by subtracting the displayed proportions: the posted estimate and displayed arm summaries may have different rounding, and the analysis method is not specified in the registry record.

How do relative measures differ?

A risk ratio compares the event proportions by division. A value of 1 represents equal proportions, while values on either side represent a relative difference in the direction determined by the group order. The reported risk ratio is 1.266, with a 95% confidence interval from 0.812 to 1.974.

An odds ratio compares the odds of an event between groups rather than comparing event proportions directly. A value of 1 represents equal odds. The reported odds ratio is 1.274, with a 95% confidence interval from 0.808 to 2.01. Risk ratios and odds ratios are related but are not interchangeable measures. In this trial, both are supplementary ways of expressing the group comparison; the registry’s primary outcome is a count of participants with stroke.

What does the p-value tell readers?

The reported unadjusted p-value is 0.296. In general, a p-value evaluates how compatible the observed data are with a null hypothesis under a specified statistical model. A larger p-value does not prove that groups are equivalent, and a smaller p-value would not, by itself, establish clinical importance. A p-value is not the probability that the null hypothesis is true. The record does not report the test used, its detailed assumptions, or any adjusted analysis for the primary comparison.

Why does randomization matter?

Random assignment is intended to make treatment groups comparable, on average, with respect to measured and unmeasured characteristics at the time of assignment. That design supports a comparison by assigned group. It does not ensure that every characteristic is identical between groups, and it does not remove the need to consider missing outcome data, implementation, or the analysis plan. The registry identifies the primary population as intent-to-treat, but further details of how participants were handled in the analysis are not reported in the registry record.

Why is event count important?

For a binary endpoint, the amount of information in a comparison depends not only on the total number of participants but also on how many outcome events occur. When events are uncommon, estimates can remain imprecise even in a trial with substantial enrollment. Confidence intervals communicate this uncertainty by giving a range of values compatible with the data and analysis assumptions. General statistical interpretation does not replace the trial’s own specified analysis, which is not reported in full in the registry record.

10. Confidence Intervals and Precision

The risk-difference estimate is 0.006, and its reported 95% confidence interval ranges from -0.005 to 0.017. A confidence interval is a measure of statistical precision: it describes a range of values compatible with the observed data under the method and assumptions used to construct it. It is not a guarantee that the true effect lies in that range, and it does not represent a probability assigned to a particular effect value.

For this difference, zero denotes no difference between the groups. Because the interval extends below and above zero, the results are compatible with no difference as well as with effects in either direction. The interval also indicates that the point estimate should not be read as a highly precise estimate of the size or direction of any difference. The construction method and any covariate adjustments are not reported in the registry record.

The relative estimates are also uncertain. The risk-ratio interval is 0.812 to 1.974, and the odds-ratio interval is 0.808 to 2.01. Each includes its no-difference value of 1. These intervals reinforce the need to interpret the point estimates alongside their uncertainty rather than treating a single estimate as conclusive.

Precision is distinct from clinical importance. The interval does not specify what size of difference would matter to patients or clinicians, and the registry record does not report a threshold for clinical importance. Nor does an interval that includes the null establish equivalence; an equivalence conclusion requires a prespecified margin and analysis, which are not reported in the registry record.

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 describe participants per arm for 80% power and participants per arm for 90% power. They are planning illustrations based on the observed effect, not the number of participants enrolled in PROTECTED TAVR. The assumptions listed are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. These are planning assumptions for the illustration; the trial’s original design assumptions are not reported in the registry record.

For planning for 100% of the observed effect, the figures are 10,781 participants per arm for 80% power and 14,433 participants per arm for 90% power. For planning for 75% of the observed effect, they are 18,619 and 24,924 participants per arm, respectively. For planning for 50% of the observed effect, they are 40,656 and 54,426 participants per arm, respectively.

In general, power is the chance that a study will detect an effect of a specified size under the assumptions used for planning. The 80% and 90% columns therefore represent different target probabilities of detecting the assumed effect if it is present; they are not measures of the probability that the trial result is correct. A smaller effect is harder to distinguish from random variation, so planning for a smaller fraction of the observed effect requires more participants. The displayed figures are not observed power.

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 = 0.023; treatment proportion p₁ = 0.029; average p̄ = 0.026.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.2237 = 0.4383.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.2236 = 0.1882.
4. n = (0.4383 + 0.1882)² / (0.0060)² = 10,781.0, rounded up to 10,781 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 <- 0.0226515656229181                          # control proportion
p1 <- 0.028685790527018                          # 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(4149535)
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 = 10,781 (80% power), 14,433 (90% power)
 75% of observed effect: n per arm = 18,619 (80% power), 24,924 (90% power)
 50% of observed effect: n per arm = 40,656 (80% power), 54,426 (90% power)
power.prop.test: 10781 
Simulated power at n = 10,781 per arm: 0.802 (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. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when several outcomes, comparisons, or analyses are tested, because repeated testing can increase the chance of at least one apparently notable result under a null model. The registry lists one outcome measure, identifies the stroke measure as primary, and lists no secondary outcome measures. It does not report a multiplicity adjustment or a detailed plan for handling multiple analyses. No additional analysis records are posted, so the registry does not provide a full analysis set from which to assess whether other comparisons were planned or conducted.

The primary endpoint is specified for the intent-to-treat population. In general, this population retains participants in their randomized groups for analysis, helping preserve the comparison created by randomization. The registry does not report participant flow, exclusions from the primary analysis, adherence, crossovers, or how deviations from assigned treatment were handled. The group counts in the outcome summary are 1501 and 1499; further detail about the data contributing to those summaries is not reported in the registry record.

Missing outcome data can affect estimates if outcome status is unavailable for some participants, particularly when the reasons for missingness relate to treatment or outcome. The registry does not report the amount or handling of missing primary outcome data, or any sensitivity analyses. It is therefore not possible to determine from the record whether missingness affected the comparison.

The record also does not report the full statistical analysis plan, the exact test used for the primary comparison, prespecified subgroup analyses, interim analyses, or a missing-data strategy. These absences limit assessment of whether the posted results follow a prespecified plan and how robust the estimates are to alternative assumptions. They do not, by themselves, establish that a particular analysis or safeguard was or was not used outside the registry record.

14. 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 0% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: atrioventricular block complete (6.1%), bundle branch block left (2.7%), cerebrovascular accident (1.2%), vascular procedure complication (1.0%), atrial fibrillation (1.0%).

The registry reports serious adverse events among 285 participants in the TAVR-with-Sentinel group, with 1501 at risk, and among 315 participants in the TAVR-without-Sentinel group, with 1499 at risk. These are participant counts, not a causal comparison. The registry also records 11 deaths in the Sentinel group and 7 in the group without Sentinel. The other-adverse-event count is recorded as 0 in each group.

Serious adverse events are broader than the primary stroke endpoint. The listed serious-event terms include atrioventricular block complete (6.1%), bundle branch block left (2.7%), cerebrovascular accident (1.2%), vascular procedure complication (1.0%), atrial fibrillation (1.0%), and ischaemic stroke (0.9%). These percentages are reproduced as listed; the registry summary does not specify their denominators in the registry record.

Adverse-event counts should be interpreted with attention to the event definitions, observation periods, participant exposure, and reporting procedures. Those details are not fully reported in the registry record. The group counts alone do not establish that an event was caused by the device or procedure, and they do not replace the primary endpoint comparison.

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

The primary outcome comparison records fewer participants with stroke in the group assigned to TAVR with Sentinel than in the group assigned to TAVR without Sentinel: 34 of 1501 versus 43 of 1499. The reported risk difference is 0.006 for the comparison of no Sentinel relative to Sentinel. Its 95% confidence interval, -0.005 to 0.017, includes zero, and the unadjusted p-value is 0.296. The results therefore show a point estimate in one direction, alongside uncertainty that includes no difference and effects in either direction.

This is not evidence that the groups are equivalent. Equivalence requires a prespecified margin and an analysis designed to assess whether any difference is sufficiently small; those details are not reported in the registry record. Conversely, the point estimate alone does not establish a benefit or harm. The magnitude and precision of the estimate, the endpoint definition, and the methods used all matter when interpreting a randomized comparison.

The endpoint is limited to stroke through 72 hours after TAVR or discharge, whichever came first. It includes several stroke classifications and levels of disability, but the posted comparison does not separately estimate each subtype or severity category. It also does not report effects beyond the endpoint window. The findings should be understood as applying to this defined outcome and comparison, rather than to all possible outcomes associated with the procedure or device.

The registry’s aim was to demonstrate a reduction in peri-procedural stroke. The posted primary comparison does not meet that aim by establishing a statistically clear difference in the reported estimate. This statement describes the uncertainty in the posted data; it is not a recommendation about treatment. The complete protocol and analysis plan, details of missing data, and results for secondary outcomes are not reported in the registry record.

16. Why This Trial Matters Statistically

PROTECTED TAVR illustrates why a randomized comparison should be read through both its estimate and its precision. The two groups had different observed stroke counts, and the point estimate of the risk difference favors a lower reported proportion in the Sentinel group. Yet the confidence interval spans zero. Looking only at the direction of the point estimate would miss the uncertainty about the size and direction of the underlying difference.

The study also makes the endpoint definition central to interpretation. Stroke was adjudicated by an independent Clinical Events Committee using NeuroARC definitions and included events through 72 hours after the procedure or discharge, whichever came first. The result therefore answers a time-bounded question about a specified outcome, rather than a broader question about all neurological events or longer-term outcomes.

Rare binary outcomes can produce limited precision even when enrollment is substantial, because the number of events—not enrollment alone—contributes to the information available for estimating a difference. The posted planning illustrations show how required sample size changes under different assumed fractions of the observed effect. They are useful for understanding design sensitivity, but they do not replace the trial’s original design rationale, which is not reported in the registry record.

Finally, a registry result is easiest to evaluate when the protocol, analysis plan, missing-data handling, and multiplicity strategy are fully documented alongside the outcome estimates. The record provides a randomized design, an intent-to-treat endpoint, and comparative estimates, while leaving several analytical details unreported. Recognizing both the useful information and these limits is part of sound statistical interpretation.

17. 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 report the full protocol or detailed statistical analysis plan. In particular, the method used for the primary confidence interval and p-value, any covariate adjustment, and the prespecified definition of the primary hypothesis are not available in the posted record. The trial’s original sample-size rationale and design assumptions are also not reported, so the separate planning illustrations should not be mistaken for the trial’s own design calculation.

The primary outcome is limited to stroke through 72 hours after TAVR or discharge, whichever came first. The record does not provide longer-term stroke results or separate estimates by stroke subtype or disability status. It reports one outcome measure and no secondary outcome measures, limiting the range of posted clinical outcomes available for interpretation.

Details about participant flow, missing primary outcome data, treatment adherence, crossovers, and exclusions from analysis are not reported in the registry record. The intent-to-treat population is named, but the record does not explain how every participant’s outcome status or protocol deviations were handled. The amount and handling of missing data therefore cannot be assessed from the posted information.

The adverse-event summary gives counts and selected serious-event terms, but the registry record does not report all contextual details needed to interpret those events, such as full event definitions and relevant observation periods. These limitations concern the completeness of the posted record; they do not establish that particular procedures were absent from the trial conduct.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: PROTECTED TAVR: Stroke PROTECTion With SEntinel During Transcatheter Aortic Valve Replacement — NCT04149535.
• Registry record last updated 2023-02-02; status last verified 2023-01.
• Results first posted 2023-02-02.
• 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.

21. Record Summary

PROTECTED TAVR (NCT04149535) is a completed randomized, parallel-group trial comparing TAVR with the Sentinel Cerebral Protection System against TAVR without Sentinel. The registry lists enrollment of 3000, no masking, and an intent-to-treat primary outcome population. The primary endpoint is stroke through 72 hours after TAVR or discharge, whichever came first.

The registry reports 34 stroke events among 1501 participants assigned to Sentinel and 43 among 1499 assigned to no Sentinel. The risk difference for no Sentinel relative to Sentinel is 0.006, with a 95% confidence interval from -0.005 to 0.017 and an unadjusted p-value of 0.296. The interval includes zero, so the estimate is uncertain about whether the groups differ.

The registry posts one outcome measure and no secondary outcome measures. Further details about the original design assumptions, complete statistical plan, and missing-data handling are not reported in the registry record.