A Study Comparing Cardiovascular Effects of Ticagrelor and Clopidogrel in Patients With Peripheral Artery Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Comparing Cardiovascular Effects of Ticagrelor and Clopidogrel in Patients With Peripheral Artery Disease (NCT01732822), a randomized phase 3 study. The primary outcome is Composite of Cardiovascular (CV) Death/MI/Ischemic Stroke.

2. Clinical Question

3. Trial Design

EUCLID was a randomized, quadruple-masked, parallel-group Phase IIIb study comparing ticagrelor with clopidogrel in patients with established peripheral artery disease. The registry lists an enrollment of 13885 and identifies the study as completed. Randomization assigned participants to one of the two oral treatment regimens. The ticagrelor arm used ticagrelor 90 mg twice daily with clopidogrel placebo once daily; the clopidogrel arm used clopidogrel 75 mg once daily with ticagrelor placebo twice daily.

The primary analysis population was the full analysis set, which the registry defines as all randomized patients. The primary outcome was the composite of cardiovascular death, myocardial infarction, or ischemic stroke. The recorded timeframe was from randomization to the primary analysis censoring date, with an average of 2.5 years. The registry describes censoring for participants without an event using the minimum of the primary analysis censoring date, last endpoint assessment date, and non-cardiovascular death date.

The posted arm summaries list 751 participants with the composite among 6930 in the ticagrelor arm, and 740 among 6955 in the clopidogrel arm. The registry also reports a Cox regression hazard ratio and lists secondary outcomes including mortality, acute limb ischemia, and revascularization outcomes. The reported secondary-outcome list contains eight outcomes. The record lists 24 outcome measures and 21 analyses, but further details about the relation between each measure and analysis are not reported in the registry record.

The randomization method beyond the designation of randomized, the method used to conceal allocation, and the trial's original sample-size justification are not reported in the registry record. The date range listed for the study runs from 2012-12-04 to primary completion on 2016-09-26. Those dates describe the recorded study timeline, rather than the follow-up duration of every individual participant.

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.

Before assignment: Screened patients randomised to study drug: 85.5%; n=13885 Patients who were not randomised: 14.5%; n=2352 Patients with inclusion criteria for symptomatic lower extremity PAD failed: n=489 Patients with poor metabolizer status for CYP2C19: n=616 Patients with other reason: n=1374

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The population was the full analysis set, which included all randomized patients.

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 Composite of Cardiovascular (CV) Death/MI/Ischemic Stroke, measured over From randomization to PACD, an average of 2.5 years.

The primary endpoint was the composite of cardiovascular death, myocardial infarction, or ischemic stroke. A composite endpoint counts participants who experience at least one of its specified outcomes, rather than treating each component as interchangeable in clinical importance. Combining outcomes can capture more events for analysis and can represent a broader clinical question, but interpretation still depends on knowing which components contribute to the result. Component-specific primary results are not reported in the posted primary summary.

The registry describes the primary measure as the number of participants with one of the composite events. For participants without a recorded event, censoring was defined at the earliest of the primary analysis censoring date, last endpoint assessment date, and non-cardiovascular death date. The timeframe is given as from randomization to the primary analysis censoring date, with an average of 2.5 years. This definition matters because participants may have different periods during which events can be observed, and censoring indicates that observation for the endpoint ends at a specified point.

The endpoint was used to compare the assigned groups on a clinically focused combination of cardiovascular outcomes. The registry also posts a Cox regression hazard ratio, a time-to-event summary that accounts for the timing of events and censoring under its model. The available record does not report the component-specific event breakdown or further details about endpoint adjudication.

7. Results Posted in the Registry

• Clopidogrel 75 mg od: 10.6% (740/6955) versus 10.8% (751/6930) on control; risk difference -0.2 percentage points (95% CI -1.2 to 0.8), risk ratio 0.98 (0.89–1.08); unadjusted two-proportion z-test p = 0.708.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary at the participant level for the posted count summary: a participant either had a recorded event in the composite or did not. The arm summaries report counts and group sizes, while the recalculation compares the resulting proportions. This is an unadjusted comparison based on the posted summary statistics, not a reconstruction of participant-level data. It does not account for covariates or event timing.

What does a risk difference represent?

A risk difference subtracts one group's event proportion from the other group's proportion. Its sign depends on the order of the groups. The posted recalculation expresses clopidogrel relative to ticagrelor and gives an estimate of -0.002. In this orientation, a negative value means the observed proportion was lower in the clopidogrel group. The confidence interval, from -0.012 to 0.008, includes zero, so the data are compatible with no difference as well as with differences in either direction within the interval. The interval is an estimate of uncertainty under the calculation's assumptions; it is not a range containing a known fixed treatment effect with certainty.

What does the Cox hazard ratio represent?

The registry's sponsor-reported analysis uses Cox regression and reports a hazard ratio. In general, a hazard ratio compares the instantaneous event rates between groups among participants who remain at risk at each point in follow-up. It uses event timing and can accommodate censored observations under the model. The reported value is 1.02, with a 95% confidence interval from 0.92 to 1.13 and p = 0.650. A hazard ratio of 1 represents no difference in hazards; the reported interval includes that value. The registry record does not report whether the proportional-hazards assumption was assessed or whether other model diagnostics were conducted.

The risk difference and hazard ratio are not the same estimand. The risk difference contrasts observed proportions over the recorded outcome period, whereas the hazard ratio summarizes relative event rates over time through a model. Their values and p-values need not match exactly. The posted recalculation is unadjusted and uses the group summaries, while the sponsor-reported result comes from Cox regression. The recalculation is useful as a transparent comparison of the posted counts, but it does not replace the trial's reported model-based analysis.

Why do censoring and follow-up matter?

Censoring occurs when the endpoint observation period ends for a participant without a recorded primary event. The registry specifies that censoring occurs at the minimum of the primary analysis censoring date, last endpoint assessment date, and non-cardiovascular death date. In general, time-to-event methods use the information available up to each participant's censoring time rather than treating every censored participant as if an event occurred. These methods rely on assumptions about the observation and censoring process. Further details on how those assumptions were evaluated are not reported in the registry record.

The registry states that the primary timeframe runs from randomization to the primary analysis censoring date, with an average of 2.5 years. An average does not mean that every participant was followed for that duration. The arm-level counts alone do not describe the timing of events or the amount of person-time observed. Consequently, the raw proportions and the Cox hazard ratio answer related but distinct questions.

How should the p-values be read?

The reported p-value of 0.650 accompanies the Cox regression analysis. The unadjusted recalculation reports p = 0.708 for the risk-difference comparison. A p-value measures how unusual data at least as discrepant as those observed would be under a specified null model and analysis. It is not the probability that the null hypothesis is true, nor does it measure the size or importance of an effect. Both p-values are above the conventional 0.05 threshold used in the planning illustration, but the registry record does not report a comprehensive statistical-analysis plan or specify all inferential thresholds used for trial conclusions. The estimates and confidence intervals are therefore essential to interpretation alongside the p-values.

Randomization supports a comparison between assigned groups, but it does not remove uncertainty from the estimates. The registry identifies the full analysis set as all randomized patients, yet does not report details on any additional analysis populations, missing endpoint handling beyond the stated censoring rule, covariate adjustment, or sensitivity analyses. Those omissions limit how fully the modeling and robustness of the trial's reported result can be assessed from the registry record alone.

10. Confidence Intervals and Precision

The sponsor-reported hazard ratio is 1.02, with a 95% confidence interval from 0.92 to 1.13. A confidence interval describes a range of estimates compatible with the data under the analysis method and its assumptions. For a hazard ratio, 1 is the value corresponding to no difference. Because the interval includes 1, it does not establish a clear difference between the groups under this analysis. The interval also extends to values on both sides of 1, so the estimate is not precise enough to exclude every difference in either direction.

The unadjusted risk-difference recalculation is -0.002, with a 95% confidence interval from -0.012 to 0.008. For a risk difference, zero is the no-difference value. This interval includes zero and allows for a range of group contrasts, including estimates favoring either direction. It is therefore important not to interpret the small point estimate as proof that the underlying difference is exactly zero.

These intervals arise from different methods and should be read on their own scales. The hazard ratio is a relative time-to-event measure; the risk difference is an absolute contrast between posted proportions. Neither interval describes uncertainty in every possible outcome or subgroup. The record does not report confidence intervals for the individual primary components in the posted primary summary, so precision for cardiovascular death, myocardial infarction, and ischemic stroke separately cannot be assessed from these results.

Confidence intervals are not guarantees that the true effect lies within the stated limits. They quantify sampling uncertainty under the chosen method and its assumptions. The broad interpretation is that the posted estimates are centered near no difference, while the intervals retain uncertainty about the magnitude and direction of a possible contrast.

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 values are planning illustrations based on the observed effect, not the trial's original design calculation. For this binary outcome, the figures refer to participants per arm, not total events. The illustration assumes a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. It considers detecting 100%, 75%, or 50% of the observed effect. The trial's own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that the planning calculation estimates would be needed to have an 80% chance of detecting the selected effect, if that effect were present and the assumptions held. The 90% power column uses a higher target chance of detection and therefore lists larger required group sizes. For planning at 100% of the observed effect, the entries are 387,265 participants per arm for 80% power and 518,437 participants per arm for 90% power. For planning at 75% of the observed effect, they are 689,861 and 923,528 participants per arm. For planning at 50% of the observed effect, they are 1,555,312 and 2,082,121 participants per arm.

The pattern reflects a general statistical principle: smaller effects are harder to distinguish from random variation, so detecting them with the same target power generally requires more participants. These values are illustrative and depend on the stated assumptions and the observed effect used as the reference. They are not observed power, and they should not be interpreted as a retrospective measure of whether the completed trial had enough information to answer its question. The curve should be read as a visual representation of the same planning relationship, not as additional evidence about the trial outcome.

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 = 0.108; treatment proportion p₁ = 0.106; average p̄ = 0.107.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4378 = 0.8582.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4378 = 0.3685.
4. n = (0.8582 + 0.3685)² / (-0.0020)² = 387,264.6, rounded up to 387,265 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 <- 0.108369408369408                          # control proportion
p1 <- 0.106398274622574                          # 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(1732822)
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 = 387,265 (80% power), 518,437 (90% power)
 75% of observed effect: n per arm = 689,861 (80% power), 923,528 (90% power)
 50% of observed effect: n per arm = 1,555,312 (80% power), 2,082,121 (90% power)
power.prop.test: 387264.6 
Simulated power at n = 387,265 per arm: 0.801 (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. Analyses Reported by the Sponsor

The sponsor-reported primary analysis is a Cox regression with a hazard ratio of 1.02, a 95% confidence interval from 0.92 to 1.13, and p = 0.650. This is the trial's reported model-based comparison and uses time-to-event information. It is distinct from the simple unadjusted recalculation of the arm-level event proportions, which reports a risk difference of -0.002, a 95% confidence interval from -0.012 to 0.008, and p = 0.708.

The two analyses are not expected to produce identical estimates or p-values because they summarize the comparison differently. The risk-difference calculation uses posted participant counts and group sizes; the Cox analysis models event timing and censoring. Both estimates are close to their respective no-difference values, and both confidence intervals include the no-difference value for their measure. The recalculation is a supplemental check of the posted summaries, not a replacement for the sponsor's reported Cox analysis. The registry record does not provide participant-level data or sufficient additional model details to reproduce the sponsor's analysis independently.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists eight secondary outcomes, including the composite of cardiovascular death, myocardial infarction, ischemic stroke, and acute limb ischemia; cardiovascular death; myocardial infarction; all-cause mortality; a composite involving cardiovascular death, myocardial infarction, and all-cause stroke; acute limb ischemia; lower extremity revascularization; and any revascularization across coronary, peripheral, and other listed sites. It also records 24 outcome measures and 21 analyses. The registry entries do not provide results for these secondary outcomes in this summary.

When a study examines multiple outcomes, the chance of obtaining at least one apparently unusual result by chance can increase if each is tested independently at a conventional threshold. Statistical approaches to multiplicity can address this issue, but the specific approach depends on the prespecified testing strategy and the intended interpretation of the outcomes. Whether multiplicity adjustments were planned or applied is not reported in the registry record. The record also does not report a hierarchical testing sequence or which outcomes were considered confirmatory versus exploratory. As a result, conclusions about any unreported secondary analysis cannot be drawn from the outcome list alone.

The primary analysis population is identified as the full analysis set, including all randomized patients. The record does not describe additional analysis populations or provide detail on exclusions from individual analyses. It specifies a censoring rule for participants without a primary event, but other missing-data procedures, imputation methods, or sensitivity analyses are not reported in the registry record. The handling of missing data for secondary outcomes is likewise not reported.

Further details of the statistical analysis plan are also not reported, including model covariates, model checks, and any interim analyses. These are not reasons to assume that particular methods were or were not used; they mark limits on what can be evaluated from the registry record. The stated primary endpoint definition, full analysis set, and Cox regression result provide important information, but they do not constitute a complete account of analysis decisions.

15. Safety Results

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

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: pneumonia (2.4%), cardiac failure congestive (1.3%), chronic obstructive pulmonary disease (1.3%), atrial fibrillation (1.0%), carotid artery stenosis (0.9%).

The registry lists serious adverse events by treatment group. In the clopidogrel group, 1788 participants were recorded as affected among 6932 at risk. In the ticagrelor group, 1716 were recorded as affected among 6910 at risk. The registry separately lists other adverse events: 989 affected among 6932 at risk in the clopidogrel group and 1682 among 6910 at risk in the ticagrelor group. These are descriptive counts, and the serious and other adverse-event categories should be read separately.

The listed top serious adverse-event terms are pneumonia at 2.4%, cardiac failure congestive at 1.3%, chronic obstructive pulmonary disease at 1.3%, atrial fibrillation at 1.0%, carotid artery stenosis at 0.9%, and acute kidney injury at 0.7%. The registry record does not specify an arm-specific breakdown for these term percentages in the registry entries, so they should not be assigned to one treatment group. Death counts are not reported in the adverse-event summary.

These figures do not by themselves establish comparative safety. They do not show event timing, duration, severity beyond the serious classification, or the statistical uncertainty around between-group differences. The adverse-event at-risk counts also differ from the group sizes in the primary-outcome summary; the registry record does not explain that difference in the registry entries. The counts are descriptive registry entries and are not a substitute for a fuller safety analysis.

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

The primary endpoint combines cardiovascular death, myocardial infarction, and ischemic stroke in patients with established peripheral artery disease. The posted results show similar observed proportions in the two randomized groups: 0.108 in the ticagrelor group and 0.106 in the clopidogrel group. The unadjusted risk-difference estimate is close to zero, and its confidence interval includes zero. The sponsor-reported Cox hazard ratio is also close to 1, and its confidence interval includes 1. These findings do not show a clear group difference on the primary endpoint in the reported analyses.

They do not prove that the treatments have exactly the same effect. The confidence intervals allow for uncertainty about the magnitude and direction of the contrast, and a failure to establish a difference is not the same as establishing equivalence. The registry record does not report an equivalence or non-inferiority design or criteria. Interpretation should therefore remain tied to the superiority-or-other analysis designation and the estimates that are actually posted.

The composite result also does not establish that each component had the same pattern. Cardiovascular death, myocardial infarction, and ischemic stroke may contribute differently to a composite, and component-specific primary results are not reported in the posted primary summary. Nor does the composite result answer every clinical question about other outcomes such as acute limb ischemia, mortality, or revascularization. Although those appear in the secondary-outcome list, their results are not reported in the registry record for this analysis.

The results apply to the trial's defined population, endpoint, observation process, and analyses. They should not be extended automatically to other populations or outcomes. They are statistical findings from a randomized comparison, not individual treatment guidance. For clinical decisions, factors beyond these posted estimates would be relevant; no treatment recommendation follows from the registry summary alone.

17. Why This Trial Matters Statistically

This trial illustrates why the result of a clinical comparison should not be reduced to a single p-value. The registry reports participant counts and proportions for a binary composite, as well as a sponsor-reported Cox hazard ratio that incorporates event timing. These measures address related but different features of the data. Reading both alongside their confidence intervals makes the distinction between an estimated effect and its precision more visible.

It also demonstrates the importance of the no-difference value for each measure. Zero is the reference value for a risk difference, while 1 is the reference value for a hazard ratio. Both posted intervals include the relevant reference value. That supports a conclusion of no clear difference in the reported analyses, but does not establish exact equality. This distinction is fundamental in interpreting non-significant trial results.

The composite endpoint raises a further statistical issue: a combined outcome can summarize several events, but the overall result does not show the contribution or precision of each component. The registry's secondary-outcome list provides context about the breadth of outcomes considered, while the absence of their results and of multiplicity details limits interpretation beyond the primary comparison.

Finally, the planning illustration helps distinguish sample-size planning from the completed trial's observed evidence. Its large per-arm figures follow from planning around the small observed effect and the stated assumptions. They should not be confused with observed power or with the trial's original design justification. Together, these features show why transparent reporting of estimands, analysis methods, uncertainty, and multiplicity is central to statistical interpretation.

18. 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 trial-specific details needed for a fuller assessment are not reported in the registry record. The record does not provide the original sample-size justification or the assumptions used to design the completed trial. The planning values associated with the observed effect are separate illustrations and cannot fill that gap. Details of the randomization procedure and allocation concealment are also not reported.

The primary summary gives composite event counts and the sponsor-reported Cox result, but it does not provide the primary component outcomes separately. It therefore cannot show whether cardiovascular death, myocardial infarction, and ischemic stroke contributed similarly to the composite estimate. The record also does not provide participant-level data, event dates, or model diagnostics needed to independently reproduce or assess the Cox regression. Whether proportional hazards was assessed is not reported in the registry record.

The record names secondary outcomes and gives counts of outcome measures and analyses, but the relevant secondary results and multiplicity strategy are not reported in the registry record for this analysis. The handling of missing data beyond the specified primary censoring rule, adherence, crossover, and sensitivity analyses are likewise not reported. Their absence from the registry record should not be interpreted as evidence that such procedures were not conducted.

The adverse-event summary is descriptive and does not give uncertainty estimates for the group comparisons. The serious-event term percentages are not broken down by arm in the registry entries, and death counts are not reported. The adverse-event at-risk counts differ from the primary-outcome group sizes, with no explanation provided in the registry record. These limits restrict what can be concluded about comparative safety from the posted summary.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Study Comparing Cardiovascular Effects of Ticagrelor and Clopidogrel in Patients With Peripheral Artery Disease — NCT01732822.
• Registry record last updated 2017-10-30; status last verified 2017-09.
• Results first posted 2017-10-30.
• 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.

22. Record Summary

The registry record describes EUCLID, a randomized, quadruple-masked, parallel-group Phase IIIb study comparing oral ticagrelor 90 mg twice daily with oral clopidogrel 75 mg once daily in patients with peripheral artery disease. The primary analysis population was the full analysis set of randomized patients. The primary endpoint was cardiovascular death, myocardial infarction, or ischemic stroke, assessed from randomization to the primary analysis censoring date, with an average of 2.5 years.

The posted primary counts were 751 among 6930 for ticagrelor and 740 among 6955 for clopidogrel. The sponsor-reported Cox analysis gave a hazard ratio of 1.02, with a 95% confidence interval from 0.92 to 1.13 and p = 0.650. The unadjusted risk-difference recalculation was -0.002, with a 95% confidence interval from -0.012 to 0.008 and p = 0.708. Both intervals include their respective no-difference values; the results do not establish a clear difference or exact equivalence. Further details on secondary results, multiplicity, and several analysis procedures are not reported in the registry record.