Statin Neuroprotection and Carotid Endarterectomy: Safety, Feasibility and Outcomes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Statin Neuroprotection and Carotid Endarterectomy: Safety, Feasibility and Outcomes (NCT02850081), a randomized phase 3 study. The primary outcome is Prevalence of eCD.

2. Clinical Question

The investigators also hypothesized that pre-operative statin use could lower early mortality and that maximal doses might be needed for optimal neuroprotection. Early mortality is listed as a secondary outcome. The posted record does not establish whether either hypothesis was supported: it gives counts, not formal tests or adjusted estimates, and the available groups are small. The relevant statistical question is therefore how to describe these observed data and how much uncertainty remains, rather than whether the groups can be ranked as if the study had established a treatment effect.

3. Trial Design

The registry identifies NCT02850081 as a Phase 3, randomized, triple-masked, parallel study sponsored by Columbia University. It lists enrollment of 31 and a status of terminated. The recorded study dates are 2017-06-01 for start and 2020-03-31 for primary completion. The record does not explain why the study terminated.

The study included distinct pathways based on participants’ existing statin use. The maximal-dose group was observational: participants already taking the specified maximal-dose regimens were observed before CEA. The less-than-maximal-dose group was described as randomized to continue the current dose with placebo or increase to a maximal dose. Statin-naive participants were randomized to atorvastatin 10 mg or atorvastatin 80 mg before CEA. The available description of the lower-dose arm ends before fully stating the dose-change regimen, so details beyond that description are not reported in the registry record.

The registry’s overall design labels should be read alongside these arm descriptions. In particular, the label randomized does not make the pre-existing maximal-dose group randomized. Comparisons involving that group may reflect differences in who was already taking each regimen, as well as any effect of treatment. The record does not report the methods used to generate or conceal random assignments, or identify which parties were masked despite the triple-masked classification.

The primary outcome was a count of participants meeting the registry definition of eCD. It was assessed using pre-operative and post-operative neurocognitive evaluations, with the post-operative assessments at Day 1 and Day 30. The registry states that participants who did not undergo CEA were excluded from the primary outcome analysis. Further definitions of the analysis population and the handling of participants without evaluable assessments 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: Participants who did not undergo CEA were not analyzed and therefore not included here.

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 Prevalence of eCD, measured over 30 Days: 1) Pre-op vs. Post-CEA Day 1 (12-25 hrs post-op) and 2) Pre-op vs. Post-CEA Day 30.

The primary endpoint is the prevalence of early cognitive dysfunction, recorded as the number of participants meeting the stated neurocognitive criterion. The registry defines eCD as performance at least 2 SD worse than the reference group in two or more cognitive domains, or at least 1.5 SD worse than the reference group in all cognitive domains. The reference group is part of that definition; further details about its composition are not reported in the registry record.

The outcome compares pre-operative assessment with performance after CEA at two planned time points: Post-CEA Day 1, 12-25 hrs post-op, and Post-CEA Day 30. Including both assessments allows the registry to describe cognitive changes at an early post-operative assessment and at a later follow-up assessment. The rationale for choosing this endpoint is consistent with the investigators’ stated concern about early cognitive dysfunction after CEA and their hypothesis about potential neuroprotection from pre-operative statin use.

The posted outcome counts are descriptive. The record explicitly says early termination and limited sample size meant the study was not adequately powered to support inferential statistical comparisons, so no formal statistical testing was conducted for these data. The endpoint counts therefore report how many participants met the criterion in each listed group; they do not by themselves establish that a regimen caused a change in cognitive performance.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a descriptive count tell the reader?

The primary result is a count of participants meeting the eCD definition in each group. A count is a direct description of the posted observations. It is different from an estimate accompanied by a measure of uncertainty, and it does not by itself answer whether a difference between groups is compatible with chance variation or reflects a systematic effect.

For binary outcomes such as meeting or not meeting an endpoint definition, a formal comparison typically uses both the number with the outcome and the number assessed in each group. The registry provides those counts for the listed primary-outcome groups, but states that formal statistical testing was not performed. The posted numerical pattern must therefore remain descriptive. No inferential p-value or confidence interval for the eCD comparisons is reported in the registry record.

Why does the study design affect interpretation?

Randomization is intended to make groups comparable, on average, with respect to factors that could affect outcomes. That protection applies to comparisons created by random assignment, not automatically to all groups described in a trial. Here, the maximal-dose group is explicitly identified as observational and consists of participants already taking maximal-dose regimens. Its comparison with randomized groups could reflect pre-existing differences, in addition to any effect related to statin dose.

The less-than-maximal-dose and statin-naive arms are described as randomized, but small reported groups limit the precision of any comparison. Random assignment does not guarantee that the realized groups are balanced, particularly when groups are small. The registry does not report baseline characteristics that would allow readers to assess how similar the groups were, nor does it report an adjusted analysis.

What is the role of a confidence interval?

Generally, a confidence interval expresses the uncertainty around an estimated effect by giving a range of values reasonably compatible with the data and the method used. A narrow interval indicates greater precision than a wide interval. For rare or infrequent binary outcomes and small groups, uncertainty can be substantial, even when the observed count is zero. The registry does not report confidence intervals for the primary outcome, so the precision of a comparative effect cannot be read directly from the posted counts.

Generally, a p-value evaluates how unusual data would be under a specified null model and analysis plan. It is not the probability that a treatment works, and it does not convey the size or clinical importance of an effect. No formal hypothesis test was reported for the primary outcome. There is therefore no registry p-value to interpret, and one should not infer a test result from the displayed counts.

What do the planning assumptions mean?

Any sample-size values associated with planning are illustrations based on the observed effect, not observed power and not the study’s original design justification. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is defined as a stated fraction of the observed effect. For time-to-event illustrations, the output is a total event count; for other outcomes, it is participants per arm. These assumptions describe a planning exercise and do not remedy the limited information in the completed study.

Observed power is not a useful substitute for an interval estimate or a direct account of uncertainty. The record’s statement that the study was not adequately powered for inferential comparisons is the relevant context. Planning calculations based on the observed effect cannot establish that a future study would reproduce that effect, and they should not be interpreted as evidence that the current analysis had a particular probability of detecting a true effect.

Why does the endpoint definition matter?

The primary outcome classifies participants according to performance relative to a reference group across cognitive domains. It is a threshold-based binary endpoint: a participant either meets the stated criterion or does not. Such a classification can make a complex set of neurocognitive measurements easier to summarize, but it does not show the full score distribution. Different patterns of score change may lead to the same classification, and changes that do not cross the threshold are not counted as eCD under the definition.

The registry specifies comparisons from pre-operative assessment to Post-CEA Day 1 and Post-CEA Day 30. Its posted primary result is described as prevalence, but the available record does not give separate numerical details for each comparison time point. It is not reported in the registry record whether the listed counts refer to one or both assessment windows, or how repeated assessments were handled in producing the posted outcome.

In statistical terms, these results support a limited descriptive reading: the record states the endpoint definition, the assessment schedule, and the observed counts for the listed analyzed groups. It does not provide the inferential analyses needed to quantify comparative uncertainty or establish a treatment effect. The observational nature of one group, small group counts, and incomplete information about analysis populations further constrain interpretation.

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("Observational - Maximal Dose - ARM 1", "Less Than Maximal Dose - ARM 2", "Statin Naive - ARM 3 [Atorvastatin 10mg]", "Statin Naive - ARM 4 [Atorvastatin 80mg]")
n      <- c(12, 6, 2, 2)
events <- c(2, 0, 0, 0)
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  Observational - Maximal Dose - ARM 1  12  2  16.7  2.1  48.4  4.7  44.8  5.0  45.4
ARM  Less Than Maximal Dose - ARM 2  6  0  0.0  0.0  45.9  0.0  39.0  0.4  41.0
ARM  Statin Naive - ARM 3 [Atorvastatin 10mg]  2  0  0.0  0.0  84.2  0.0  65.8  0.8  70.8
ARM  Statin Naive - ARM 4 [Atorvastatin 80mg]  2  0  0.0  0.0  84.2  0.0  65.8  0.8  70.8

11. Multiplicity, Missing Data and Analysis Populations

The registry lists the primary outcome as prevalence of eCD and one secondary outcome, prevalence of early mortality. It records 2 outcome measures and 0 analyses. The record does not report a statistical analysis plan, a hierarchy of hypotheses, or a method for controlling multiplicity across outcomes or assessment times. Because formal statistical testing was not conducted, there are no posted p-values whose interpretation would depend on a multiplicity adjustment. Whether any adjustment was planned is not reported in the registry record.

Missing-data handling is also not described. The registry states that participants who did not undergo CEA were not analyzed and were not included in the primary outcome. It does not explain how participants without a completed neurocognitive assessment, incomplete domain scores, or other unavailable outcome data were handled. The number of such participants and reasons for any missing assessments are not reported in the registry record.

The analysis population is therefore defined only in part: people who did not undergo CEA were excluded from the primary endpoint analysis. The record does not provide a complete participant flow or state whether the remaining counts represent all eligible participants with an assessment at each time point. Nor does it report whether participants were analyzed according to randomized assignment, received treatment, or another population definition. These omissions matter because exclusions and incomplete follow-up can affect how representative the observed endpoint counts are of the enrolled population.

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 0% reporting threshold.

The registry reports serious adverse events, other adverse events, and deaths by listed group. For the maximal-dose group, it records 0 serious adverse events among 12 at risk, 0 other adverse events among 12 at risk, and 0 deaths. For the less-than-maximal-dose group, the corresponding counts are 0 among 6, 0 among 6, and 0 deaths. For each statin-naive group, the registry reports 0 serious adverse events among 2 at risk, 0 other adverse events among 2 at risk, and 0 deaths.

These are observed counts during the registry’s safety reporting, not proof that adverse events cannot occur. In particular, the small groups provide limited information about uncommon events. The record lists no serious adverse-event terms. It does not provide enough detail to assess the duration or completeness of safety follow-up, the criteria used to attribute events, or whether all events were captured. The zero counts should be read as no events recorded in these listed groups, not as a guarantee of safety or a comparative safety conclusion.

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

The result documents the primary endpoint counts recorded for the analyzed groups: eCD was reported in 2 participants in the observational maximal-dose group and in 0 participants in each of the other listed groups. It also documents zero serious adverse events, zero other adverse events, and zero deaths in the listed safety groups. These observations are useful as a record of what was reported, but their meaning is constrained by the design and the small group counts.

The findings do not establish that a higher statin dose prevents eCD, that one regimen is superior, or that the regimens are equivalent. The registry says that early termination and limited sample size meant the study was not adequately powered for inferential statistical comparisons. The results were therefore presented descriptively without formal statistical testing. No treatment-effect estimate or uncertainty interval is reported for the primary outcome.

The maximal-dose group was observational, so a difference involving that group cannot be attributed solely to dose. The randomized comparisons are also based on small listed groups, and zero observed events do not rule out events that might be seen with more complete or larger-scale observation. The record does not provide a complete account of missing outcomes or the analysis population beyond excluding participants who did not undergo CEA.

The endpoint records whether a participant crossed a threshold based on performance across cognitive domains relative to a reference group. It does not, by itself, communicate the size or clinical significance of individual changes. The registry also does not give enough detail to determine how the listed counts map separately to both post-operative assessment times. Accordingly, these data are best understood as limited descriptive evidence, not a definitive assessment of efficacy or safety.

14. Why This Trial Matters Statistically

This study addresses a question where statistical design is especially important: whether pre-operative statin dose is associated with cognitive outcomes after CEA. The registry describes a combination of an observational maximal-dose group and randomized groups for lower-dose and statin-naive participants. Recognizing that distinction helps prevent an observational contrast from being mistaken for a randomized treatment comparison.

The record also illustrates how a clearly stated endpoint can coexist with substantial uncertainty about its interpretation. eCD is defined using performance thresholds across cognitive domains, and assessments were planned both soon after surgery and at Day 30. Yet the posted counts are small, the study was terminated, and the registry explicitly says that formal inferential comparisons were not supported. A numerical difference in counts is not, on its own, evidence of a reliable effect.

For clinical researchers and biostatisticians, the important statistical lesson is to separate description from inference. Counts show what was observed in the analyzed groups. Formal comparisons would require an appropriate analysis and a clear account of uncertainty, the analysis population, missing outcomes, and how comparisons across outcomes or assessment times were handled. Those details are not reported in the registry record.

The study also underscores why a zero count should not be overread. In small groups, the absence of recorded events is not equivalent to proof that the event is absent in the broader population. The safety counts and cognitive outcome counts are informative as descriptive results, while remaining insufficient for definitive comparative conclusions.

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 specific to this study constrain interpretation. The registry status is terminated, and the record states that early termination and limited sample size prevented adequate inferential statistical comparisons. The reason for termination is not reported in the registry record. The overall enrollment is listed as 31, but the primary outcome groups have smaller denominators; the complete participant flow and reasons for the difference are not reported.

The maximal-dose group was observational, unlike the randomized groups described for lower-dose and statin-naive participants. This makes comparisons that include the observational group vulnerable to differences associated with pre-existing statin use. The registry does not report baseline characteristics or adjusted analyses that could help assess or address such differences.

The primary endpoint relies on thresholds relative to a reference group, but details of that reference group are not reported in the registry record. The record also does not explain how the posted counts relate separately to both post-operative time points, or how incomplete or missing neurocognitive assessments were handled. The analysis population is only partly described: participants who did not undergo CEA were excluded, but other eligibility and evaluability rules are not stated.

Finally, no formal statistical testing, confidence intervals, or comparative effect estimates are reported. The record lists one secondary outcome but does not report a multiplicity strategy or detailed analysis plan. Safety counts are also sparse, and the duration and completeness of adverse-event follow-up are not described. These limitations mean the posted results should be treated as descriptive and uncertain.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Statin Neuroprotection and Carotid Endarterectomy: Safety, Feasibility and Outcomes — NCT02850081.
• Registry record last updated 2026-05-07; status last verified 2026-04.
• Results first posted 2026-05-07.
• 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

NCT02850081, titled Statin Neuroprotection and Carotid Endarterectomy: Safety, Feasibility and Outcomes, is recorded as a Phase 3, randomized, triple-masked, parallel study sponsored by Columbia University. Enrollment is listed as 31; the study status is terminated. The recorded start date is 2017-06-01 and primary completion date is 2020-03-31.

The primary outcome was prevalence of eCD after CEA, assessed against pre-operative performance at Post-CEA Day 1 and Day 30. The registry reports 2 cases in the observational maximal-dose group and 0 in each of the other listed groups. It states that participants who did not undergo CEA were excluded and that early termination and limited sample size precluded formal inferential comparisons. The secondary outcome is prevalence of early mortality. No serious adverse events, other adverse events, or deaths are recorded in the listed safety groups. The registry record does not report a formal analysis plan or detailed handling of missing outcomes.