Evaluate Carotid Artery Plaque Composition by Magnetic Resonance Imaging in People Receiving Cholesterol Medication: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluate Carotid Artery Plaque Composition by Magnetic Resonance Imaging in People Receiving Cholesterol Medication (NCT00715273), a randomized phase 4 study. The primary outcome is Annualized LRNC Volume Change in Carotid Plaque Composition, as Assessed by MRI.

2. Clinical Question

3. Trial Design

This was a completed, phase 4, randomized, quadruple-masked, parallel-group study sponsored by the University of Washington. It enrolled 217 participants. The registry record lists coronary artery disease, carotid artery diseases, and atherosclerosis as conditions of interest. The stated study purpose was to examine whether combinations of cholesterol-lowering medications were associated with changes in the fat content of atherosclerotic deposits in people with coronary artery disease or carotid artery disease.

Participants were assigned to one of three groups: atorvastatin with placebo niacin and placebo colesevelam; atorvastatin with niacin and placebo colesevelam; or atorvastatin with niacin and colesevelam. These were described as single, double, and triple therapy, respectively. The LDL-C treatment target was 80 mg/dl or lower in the first two groups and 60 mg/dl or lower in the triple-therapy group. The registry record does not describe the randomization procedure, the allocation ratio, or which roles were included in quadruple masking.

The primary outcome was annualized LRNC volume change in carotid plaque assessed by MRI, measured at Years 1, 2, and 3. The endpoint description says MRI measurements were performed at four time points, with image assessment blinded to scan sequence, treatment, lipid levels, and clinical course. The primary endpoint analysis was restricted to participants with detectable baseline LRNC. Arm-level primary endpoint results are reported for 60 participants in the single-therapy group, 61 in the double-therapy group, and 61 in the triple-therapy group. The reasons that these analysis counts differ from total enrollment are not reported in the registry record.

The record lists one secondary outcome: a composite of cardiovascular disease death, non-fatal heart attack, stroke, and worsening ischemia requiring medical interventions. It reports three outcome measures overall, but does not provide further detail about the other measures in the information summarized for these results. The trial's original sample-size justification and other original design assumptions 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

Analysis population for the primary outcome, as defined in the registry: Analysis group was limited from total study population due to the need for detectable lipid-rich necrotic core (LRNC) measurement at study baseline.

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 Annualized LRNC Volume Change in Carotid Plaque Composition, as Assessed by MRI, measured over Measured at Years 1, 2, and 3.

The primary endpoint is annualized change in lipid-rich necrotic core (LRNC) volume within carotid plaque, assessed using MRI. LRNC is a lipid-rich component of plaque, and the registry describes the endpoint as a measure of plaque lipid composition. Results are expressed as mean change in mm^3/year. A negative mean indicates that the measured LRNC volume declined over the annualized measurement period; it describes the direction of the measured change, not its clinical importance or cause.

The endpoint description explains that image analysis used contours around the artery lumen, the outer wall, and plaque features. Volume measurements were based on area and MRI slice thickness. The registry also describes tissue volume as a percentage of wall volume, while the posted primary outcome is the annualized LRNC volume change. The MRI measurements were assessed without knowledge of the time sequence of examinations, treatment, lipid levels, or clinical course, according to the endpoint description.

An imaging measure can be used to characterize changes in plaque composition without waiting for a clinical event to occur. That is a general reason to study such measures; it does not make an imaging change equivalent to a change in heart attack, stroke, or other clinical outcomes. The primary analysis was limited to participants with detectable LRNC at baseline, since the endpoint required a baseline LRNC measurement. The registry record does not explain how the endpoint was handled when later MRI measurements were unavailable or how the annualized measure was calculated for individual participants.

7. Results Posted in the Registry

• 2 - Double Therapy Group: mean -15.10 versus -4.60 on control; difference -10.50 (95% CI -22.13 to 1.13), standardized effect d = -0.32; z-test p = 0.077.
• 3 - Triple Therapy Group: mean -9.40 versus -4.60 on control; difference -4.80 (95% CI -16.23 to 6.63), standardized effect d = -0.15; z-test p = 0.410.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a continuous measure: annualized LRNC volume change in mm^3/year. A continuous outcome can take a range of numeric values, rather than representing a simple count of people with or without an event. The registry reports the mean for each treatment group and its standard error. The mean describes the group average; the standard error describes uncertainty in estimating that average. It is not the same as the spread of individual participants' measurements.

What do the group means describe?

The reported means are -4.6 mm^3/year for single therapy, -15.1 mm^3/year for double therapy, and -9.4 mm^3/year for triple therapy. The registry's endpoint description defines the measure as annualized volume change. Thus, a negative value represents a decline in measured volume over the annualized period. A group mean does not show how much participants varied around that average, nor does it show the individual measurement pattern. The standard errors reported for the groups are 5.0, 3.2, and 3.0, respectively.

What does a mean difference express?

A mean difference compares the average outcome in one group with the average in a reference group. The posted double-therapy comparison is -10.5 mm^3/year relative to single therapy; the posted triple-therapy comparison is -4.8 mm^3/year relative to single therapy. These are recalculated, unadjusted comparisons from the posted summary statistics. A negative difference means the first named group had a lower mean annualized change than the single-therapy group. It does not by itself establish that the group assignment caused the difference or that the difference is clinically important.

The comparisons also report pooled standard deviations of 32.54 and 31.95, and standardized effect sizes of -0.323 and -0.150, respectively. A pooled standard deviation describes the scale of variation used to standardize the difference. A standardized effect size expresses a mean difference in standard-deviation units, which can help compare the size of differences measured on different scales. These standardized estimates depend on the quantities used in the recalculation and are not a separate clinical outcome. The registry record does not describe these recalculated statistics as the trial's prespecified analysis.

Why do confidence intervals matter?

A confidence interval is a range of values compatible with the data under the method used to construct it. In general, it conveys both the estimated direction and the precision of an estimate. A narrow interval indicates greater precision than a wide interval, while an interval that includes zero leaves open the possibility of no mean difference under the comparison being considered. The posted intervals are -22.13 to 1.135 for double therapy versus single therapy and -16.23 to 6.628 for triple therapy versus single therapy. The confidence level and the trial's original interval-construction method are not reported in the registry record, so those details should not be inferred from the displayed endpoints alone.

The unadjusted p values are 0.077 for double therapy versus single therapy and 0.41 for triple therapy versus single therapy. A p value is a measure of how unusual data at least as incompatible with a specified null hypothesis would be under that hypothesis and the assumed analysis model. It is not the probability that the null hypothesis is true, nor a measure of the clinical value of a treatment. Because these are unadjusted recalculations from summary statistics, they do not account for any multiplicity adjustment or other modeling choices that may have been made in the trial's original analysis. The sponsor's own analysis does not take precedence on the basis of a reported discrepancy; the registry record does not flag a discrepancy.

What can standard errors and over-dispersion tell us?

Standard errors quantify uncertainty in estimated means, not the full variability among participants. For a continuous endpoint, the observed spread of individual outcomes is relevant to the precision of group comparisons. In general, greater variability relative to the difference being studied makes it harder to estimate a difference precisely. Over-dispersion is a term often used when observed variation exceeds what a chosen model expects. It can make model-based standard errors too small if the model does not allow enough variability. The registry record does not report a dispersion assessment or a model-based method for the primary analysis; the pooled standard deviations are included in the recalculated comparisons.

The endpoint has a longitudinal measurement description: scans were assessed at four time points, and the primary outcome is described as measured at Years 1, 2, and 3. Repeated MRI observations from the same participant are not independent in the way observations from different people may be. In general, an analysis of repeated measures may need to account for within-participant correlation and the timing of observations. The registry record does not report the exact model, covariance structure, or participant-level calculation used for the primary analysis. The simple unadjusted recalculations from arm summaries do not substitute for such details.

What does the recalculation represent?

The posted comparisons use the single-therapy group as the reference and compare each combination-therapy group with it. The recalculated differences use the posted arm means, and the comparisons include the arm sample sizes and standard errors. They are not adjusted for other participant characteristics or for multiple comparisons. No adjusted result or complete analysis specification is reported in the registry record. Accordingly, these estimates are useful for describing the scale and uncertainty of the reported group contrasts, but they cannot reveal whether the original analysis used a different model, population, or handling of repeated measurements.

The analysis population matters as well. The primary analysis was limited to participants with detectable LRNC at baseline. This is a defined subset of the enrolled population, but the registry record does not report additional analysis-population rules or a complete account of missing MRI measurements. A result from this subset should not automatically be generalized to participants without detectable baseline LRNC. The analysis population and missing-data approach are part of the meaning of the estimate, not merely technical details.

10. Confidence Intervals and Precision

The double-therapy comparison with single therapy has a recalculated mean difference of -10.5 mm^3/year, with an interval from -22.13 to 1.135. The triple-therapy comparison with single therapy has a recalculated mean difference of -4.8 mm^3/year, with an interval from -16.23 to 6.628. Both intervals include zero and extend to values on either side of zero. This means the posted comparisons do not pin down a single direction of the underlying mean difference with precision.

In general, a confidence interval gives a range of estimates compatible with the observed data and the method used to calculate the interval. Its width is an important guide to precision: a wide interval leaves more uncertainty about the magnitude and direction of an effect than a narrow one. Here, the intervals span both negative and positive values, so the estimates are not precise enough to rule out no mean difference under the recalculated comparisons. The registry record does not report the confidence level or the full method used to construct these intervals.

The arm standard errors were 5.0 for single therapy, 3.2 for double therapy, and 3.0 for triple therapy. Standard errors describe uncertainty in the estimated means, not the range of individual outcomes. The recalculated pooled standard deviations, 32.54 for the double-therapy comparison and 31.95 for the triple-therapy comparison, describe a different aspect of variation used in the standardized comparisons. These quantities should not be interchanged.

The unadjusted p values, 0.077 and 0.41, are consistent with the uncertainty reflected in the intervals, but they do not state the probability that a treatment difference exists. They are unadjusted and come from simple recalculations of posted summary statistics. The registry record does not report whether the original analysis used a different model or adjustment. The estimates and intervals should therefore be interpreted as descriptive comparisons with appreciable uncertainty, rather than as definitive estimates of the full trial analysis.

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 sample-size figures are planning illustrations for the continuous primary endpoint, and the numbers are participants per arm, not total event counts. They compare scenarios based on detecting 100%, 75%, or 50% of the observed effect. The calculations assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. They are based on the observed effect and are not the trial's original design calculations; the trial's own sample-size assumptions are not reported in the registry record.

For double therapy compared with single therapy, planning for 100% of the observed effect gives 151 participants per arm for 80% power and 202 participants per arm for 90% power. Planning for 75% of the observed effect gives 268 and 359 participants per arm, respectively. Planning for 50% of the observed effect gives 603 and 808 participants per arm. For triple therapy compared with single therapy, the corresponding figures are 696 and 931 participants per arm for 100% of the observed effect, 1,237 and 1,655 for 75%, and 2,782 and 3,724 for 50%.

In plain terms, the 80% power column gives the planned number per arm for a study designed to have an 80% chance of detecting the specified effect under the listed assumptions, if that effect is present. The 90% power column uses a higher chance of detection and therefore requires more participants per arm in these illustrations. Planning for a smaller effect requires more participants because a smaller difference is harder to distinguish from variation in the outcome. These figures are not observed power; they are prospective planning illustrations based on the observed comparisons.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -10.500; pooled standard deviation σ = 32.537.
2. n = 2 × (1.96 + 0.842)² × 32.537² / -10.500² = 150.7, rounded up to 151 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
delta <- -10.5                      # observed difference in means
sigma <- 32.5367794473688                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(715273)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 151 (80% power), 202 (90% power)
 75% of observed effect: n per arm = 268 (80% power), 359 (90% power)
 50% of observed effect: n per arm = 603 (80% power), 808 (90% power)
power.t.test: 151.7 
Simulated power at n = 151 per arm: 0.797 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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

The registry record reports two primary comparisons against the single-therapy group: double therapy versus single therapy, and triple therapy versus single therapy. The posted p values are unadjusted. The record does not report a multiplicity procedure, such as an adjustment across comparisons or across outcomes, nor does it state whether a testing hierarchy was used. When several comparisons are considered, multiplicity is a general statistical concern because the chance of at least one apparently unusual result can increase. Whether and how that concern was addressed in the trial's original analysis is not reported in the registry record.

The primary endpoint was measured over time, but the registry does not describe the primary analysis model, the handling of repeated MRI observations, or the method used to derive each participant's annualized change. It also does not report a complete plan for missing MRI measurements, reasons for missingness, imputation methods, or sensitivity analyses. These omissions make it difficult to assess how incomplete follow-up might affect the posted endpoint estimates.

The primary endpoint analysis was limited to participants with detectable baseline LRNC. That criterion defines an analysis population smaller than the overall enrollment, but the registry record does not give further details about other inclusion or exclusion rules for the analysis. The reasons that some enrolled participants do not appear in the arm-level primary endpoint counts are not reported in the registry record. The posted safety denominators are different from the primary endpoint analysis counts; the record does not explain those differences in the information summarized for these results.

Although the trial is described as randomized and quadruple-masked, the registry record does not report allocation concealment details, the exact roles masked, or a complete statistical analysis plan. It lists a cardiovascular composite as a secondary outcome but does not provide a posted result for that composite in the summarized results. The registry record also reports zero posted analysis records, while providing arm-level outcome summaries and results that permit simple unadjusted recalculations. The absence of a registry analysis record does not itself establish what analyses were or were not conducted; it means the full analytic details are not reported in 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: percutaneous coronary intervention (14.1%), hospitalized for worsening ischemia (9.9%), myocardial infarction (5.5%), cancer (5.5%), death - other/multiple/unknown (4.2%).

The registry reports serious adverse events and other adverse events by treatment group, with separate numbers of participants at risk and participants affected. Serious adverse events affected 26 of 71 participants in the single-therapy group, 20 of 73 in the double-therapy group, and 20 of 73 in the triple-therapy group. Other adverse events affected 34 of 71, 40 of 73, and 32 of 73 participants, respectively. The record also reports 7 deaths in the single-therapy group, 7 in the double-therapy group, and 3 in the triple-therapy group.

The listed serious-event terms include percutaneous coronary intervention at 14.1%, hospitalized for worsening ischemia at 9.9%, myocardial infarction at 5.5%, cancer at 5.5%, death - other/multiple/unknown at 4.2%, and peripheral artery revascularization at 4.1%. These figures describe the event terms as listed in the registry. The record does not provide, in the information summarized for these results, a comparative statistical analysis of safety, details about event timing, or a causal assessment of individual events. The counts alone do not establish that differences between groups are attributable to treatment.

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

The primary results describe average annualized changes in MRI-measured carotid plaque LRNC volume among participants with detectable LRNC at baseline. The group means were negative, indicating average declines in measured volume. The recalculated differences were also negative for double therapy and triple therapy relative to single therapy, but both intervals include zero and extend to positive values. The posted comparisons therefore remain uncertain about the size and direction of the between-group mean differences.

These results do not establish whether one regimen produces a clinically important change in plaque composition. The endpoint is an imaging measure, not a cardiovascular event. An imaging difference, even if estimated precisely, would not automatically show a difference in heart attacks, strokes, cardiovascular death, or other clinical outcomes. The registry lists a cardiovascular composite as a secondary outcome, but a result for that composite is not reported in the primary results.

The estimates also apply to the primary analysis population, which was limited to participants with detectable baseline LRNC. The registry record does not provide enough information to determine how missing MRI data, other analysis-population rules, repeated measurements, or adjustment for other factors affected the original analysis. The recalculated comparisons are unadjusted and should not be mistaken for a complete account of the trial's planned analysis.

The safety counts provide a descriptive account of adverse events and deaths by group, but they do not by themselves establish causal relationships or comparative safety. More generally, the posted results should be interpreted as a record of the reported group-level measurements and the uncertainty around simple comparisons. They do not support medical advice or a conclusion about treatment choice. The registry record's incomplete reporting of analysis methods and missing-data handling limits what can be inferred beyond those descriptive estimates.

16. Why This Trial Matters Statistically

This trial illustrates how a randomized clinical study can report both an imaging endpoint and a clinical-event outcome. The primary measure is continuous and is expressed as annualized change in LRNC volume, while the listed secondary composite concerns events such as cardiovascular disease death, non-fatal heart attack, stroke, and worsening ischemia requiring medical interventions. The two kinds of outcomes address different statistical and clinical questions and should not be treated as interchangeable.

The registry also highlights the value of distinguishing an arm mean from a between-group contrast. Each group had a negative average annualized change, but the key comparative estimates are the mean differences against single therapy. Their intervals extend across zero, emphasizing the uncertainty in the contrasts rather than relying on the direction of the arm means alone. Reporting both estimates and intervals gives a more informative picture than focusing on a single p value.

The primary analysis population is another important statistical feature. It was limited to participants with detectable LRNC at baseline, so the endpoint required a measurement that was not necessarily available or detectable for everyone enrolled. The relationship between the overall enrollment and the endpoint analysis counts, along with the handling of missing data, is not fully described in the registry record. Such details affect how readers understand the population to which the results apply.

Finally, the sample-size illustrations show how the assumed effect size affects planning. Smaller target effects require larger per-arm sample sizes under the stated assumptions. These illustrations are not the trial's original design justification and are not observed power. Together, the endpoint definition, analysis-population restriction, interval estimates, and missing methodological details demonstrate why statistical interpretation depends on more than a point estimate alone.

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 primary outcome analysis was restricted to participants with detectable LRNC at baseline. The registry record does not report further details about the selection of this analysis population, the number of participants without a usable baseline measurement, or the reasons that the primary endpoint analysis counts differ from total enrollment. This limits assessment of how representative the analyzed participants were of all enrolled participants.

The endpoint description refers to MRI examinations at four time points and an outcome measured at Years 1, 2, and 3, but the registry record does not report the participant-level calculation of annualized change or the statistical model used for repeated measurements. It also does not report how missing or unusable follow-up scans were handled, whether sensitivity analyses were conducted, or whether the original analysis adjusted for other characteristics. Those methods could affect interpretation of the posted summary results.

The comparisons discussed are unadjusted recalculations from arm-level summary statistics. The registry record does not report a confidence level for the intervals, a full statistical analysis plan, or a multiplicity procedure. No direct primary-result comparison between double and triple therapy is reported. The registry also does not give the full results for the listed secondary cardiovascular composite in the results summarized for the primary endpoint.

The study is described as quadruple-masked, but the record does not specify which roles were masked or provide details of the randomization and allocation procedures. The available safety figures are counts and listed event-term percentages; comparative safety methods, event timing, and causal assessments are not reported in the summarized information. These reporting gaps do not change the reported numbers, but they constrain conclusions about precision, generalizability, and the trial's full analysis.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Evaluate Carotid Artery Plaque Composition by Magnetic Resonance Imaging in People Receiving Cholesterol Medication — NCT00715273.
• Registry record last updated 2022-06-07; status last verified 2022-05.
• Results first posted 2022-06-07.
• 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.

21. Record Summary

NCT00715273 is a completed phase 4, randomized, quadruple-masked, parallel-group study sponsored by the University of Washington, with enrollment of 217 participants. It compared atorvastatin alone with atorvastatin plus niacin and with atorvastatin plus niacin and colesevelam. The primary endpoint was annualized MRI-measured change in carotid plaque LRNC volume, with the analysis limited to participants who had detectable LRNC at baseline.

The posted mean changes were -4.6 mm^3/year for single therapy, -15.1 mm^3/year for double therapy, and -9.4 mm^3/year for triple therapy. The recalculated unadjusted mean differences versus single therapy were -10.5 and -4.8, and both reported intervals include zero. The registry record lists a cardiovascular composite as a secondary outcome, but the results summarized for the primary endpoint do not provide that outcome's estimate. Important details about the original statistical plan and missing-data handling are not reported in the registry record.