Safety Study of External Counterpulsation as a Treatment for Acute Ischemic Stroke: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety Study of External Counterpulsation as a Treatment for Acute Ischemic Stroke (NCT00983749), a randomized phase 1 study. The primary outcome is Feasibility and Tolerability of External Counterpulsation.

2. Clinical Question

The primary endpoint was framed as a feasibility and tolerability question, not as a definitive comparison of clinical benefit. Tolerance meant no indication to stop the procedure or reduce pressure to a non-therapeutic level. The full-pressure feasibility definition additionally required sustained tolerance, for at least 30 minutes, of a pressure capable of causing a 15% augmentation of mean flow velocity (MFV) in 90% of subjects. For the sham-pressure group, feasibility meant sustained tolerance of sham pressure in all subjects. These different criteria matter when interpreting the posted counts: they describe performance against the trial's stated definitions, rather than a general measure of recovery from stroke.

3. Trial Design

The registry describes NCT00983749 as a randomized, controlled phase 1 study with a parallel-group design and single masking. The sponsor was the University of California, San Diego. The study is marked completed. Its recorded start was 2009-12 and its primary completion was 2013-02. Enrollment is listed as 23. The record does not state who was masked, nor does it provide further details about the randomization procedure.

The two arms received different pressure settings during one-hour ECP treatments. In the full-pressure arm, pressure was increased in a tiered, dose-escalating manner up to 300mmHg while assessments were made. The sham-pressure arm received inactive pressure of 75mmHg. The posted primary outcome counts are 12 participants in the full-pressure arm and 9 in the sham-pressure arm meeting the reported outcome, with arm sizes of 13 and 10, respectively. Those counts should be read with the endpoint definitions in mind; they are not measures of stroke symptom improvement.

This was an early-phase safety and feasibility study, as reflected by its phase 1 designation and the outcomes selected. The registry's study purpose also mentions blood-flow velocity in arteries supplying the brain and stroke symptoms. However, the record lists no posted secondary outcomes and no analyses. The registry record does not report the original sample-size justification, detailed statistical analysis plan, allocation concealment, masking procedures, or reasons for the chosen enrollment. Those details cannot be inferred from the design label or the posted results.

Random allocation can support a fair comparison by reducing systematic differences between groups, but its success cannot be fully evaluated from the brief registry information. With the posted group sizes and primary outcome counts, the comparison is descriptive and imprecise. The trial's design is best understood as an early assessment of whether the procedure could be carried out and tolerated under the specified conditions, with a sham-pressure group for comparison—not as a definitive efficacy trial.

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: Before randomization, all patients underwent duplex ultrasound scanning of the legs to rule out deep vein thrombosis (DVT), along with an assessment of temporal window adequacy for transcranial doppler (TCD); patients meeting these criteria were randomly assigned to one of two treatment arms (all the patients that were enrolled met these criteria).

Acute Phase (Enrollment to Hospital DC)

Follow-up Phase (Hospital DC to 30 Days)

Reasons for not completing this period:

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 Feasibility and Tolerability of External Counterpulsation, measured over During one hour of treatment.

The primary outcome was feasibility and tolerability of ECP during one hour of treatment. Tolerance was defined as having no indication to stop the procedure or lower pressure to a non-therapeutic level. Feasibility had an additional, arm-specific meaning. In the full-pressure group, it required sustained tolerance for at least 30 minutes at any pressure capable of causing a 15% augmentation of MFV in 90% of subjects. In the sham-pressure group, it required sustained tolerance of sham pressure in all subjects.

This endpoint suits an early-phase study because it focuses on whether the procedure can be delivered and continued under the stated conditions. The endpoint is not a direct measure of stroke recovery, disability, or long-term clinical benefit. The registry's description states that the study also sought to assess possible effects on blood-flow velocity and stroke symptoms, but secondary outcome results are not reported in the registry record.

The posted counts are 12 of 13 participants in the full-pressure arm and 9 of 10 in the sham-pressure arm for the primary outcome. Because the criteria differ by arm, a group comparison should be understood in the context of those definitions. The endpoint captures a combination of procedural feasibility and tolerability rather than a single biological or patient-reported response. The registry record does not provide participant-level information or further detail about how outcome assessments were adjudicated.

7. Results Posted in the Registry

• Full-pressure ECP: 92.3% (12/13) versus 90.0% (9/10) on control; risk difference 2.3 percentage points (95% CI -21.3 to 25.9), risk ratio 1.03 (0.79–1.33); unadjusted two-proportion z-test p = 0.846.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary comparison is a binary outcome: each participant is counted as meeting or not meeting the feasibility and tolerability endpoint. For this kind of outcome, the observed counts and group denominators give a direct description of how many participants in each arm met the endpoint. The posted counts are 12 of 13 for full-pressure ECP and 9 of 10 for sham-pressure ECP. The record also gives the estimated risk difference and an unadjusted p-value. The risk difference expresses the contrast between the groups on an absolute scale; a value around zero indicates little estimated separation, while values on either side of zero indicate a difference in opposite directions.

What does the risk difference tell a reader?

A risk difference compares the proportion with the outcome in one group with the proportion in the other. It is an absolute measure, so it is usually easier to relate to the event proportions than a relative measure. The registry's comparison is for full-pressure ECP against sham-pressure ECP. The estimated risk difference is 0.023. Its confidence interval crosses zero, indicating that the data do not precisely distinguish a positive difference from a negative one. The interval should be considered alongside the observed counts, rather than interpreting the point estimate as a settled effect.

What is a p-value?

The unadjusted p-value is 0.846. In general, a p-value measures how surprising results at least as incompatible with a specified null hypothesis would be if that hypothesis and the statistical model were true. It is not the probability that the null hypothesis is true, nor does it measure the size or importance of an effect. A large p-value does not prove that there is no difference, and it does not demonstrate equivalence. With limited information and imprecise estimates, a result may fail to distinguish between groups even when meaningful differences remain compatible with the data.

Why is the confidence interval important?

In general, a confidence interval presents a range of effect estimates compatible with the observed data under the method's assumptions. It communicates precision as well as direction. The interval for this comparison is broad and crosses zero. That means the estimated difference is uncertain, not that every value in the interval is equally likely or that the interval guarantees the true effect lies within it. Because the interval includes effects on both sides of zero, the posted result does not support a precise conclusion about which arm has the higher underlying probability of meeting the endpoint.

What is meant by an unadjusted analysis?

In general, an unadjusted comparison uses the group outcome data without accounting for additional covariates. Such a comparison is straightforward, but it does not address possible imbalances in participant characteristics or other factors. The registry labels the p-value as unadjusted and lists no analyses. The registry record does not report whether other adjustments were prespecified or undertaken, nor does it give a detailed statistical analysis plan. No adjusted result should be inferred.

Why does the endpoint definition matter?

The primary measure combines feasibility and tolerability, and the registry gives different feasibility criteria for the full-pressure and sham-pressure groups. For full pressure, feasibility required sustained tolerance of a pressure capable of a specified MFV augmentation in a specified proportion of subjects. For sham pressure, it required sustained tolerance of sham pressure in all subjects. Tolerance itself meant no indication to stop or reduce pressure to a non-therapeutic level. These operational definitions are central to interpreting the counts. They should not be replaced with a broader interpretation such as general treatment acceptability or clinical improvement.

In general, a binary endpoint is easy to summarize, but it can conceal variation in experience and timing. The registry reports counts rather than participant-level observations, and it does not provide additional detail about duration of tolerance, pressure achieved by each participant, or assessment decisions. The outcome therefore supports a limited summary of whether the stated criterion was met, not a detailed account of how each participant responded.

How should the adverse-event counts be treated statistically?

Adverse-event counts are descriptive safety information. The registry records serious adverse events in the two arms and also reports counts for other adverse events. With few participants and rare events, estimates of event frequency are inherently imprecise. In general, observing no serious events in a group does not establish that the underlying risk is zero. Likewise, a recorded event does not by itself establish causality. The registry does not report attribution, event timing, or participant-level overlap in enough detail to make stronger claims.

The listed serious-event terms include transient ischaemic attack and exacerbated chronic obstructive airways disease, each with a reported frequency of 7.7%. The registry provides no further detail in this record to connect those terms to specific participants, explain their relation to treatment, or establish whether the entries overlap. Death counts are not reported in the registry record, so they should not be interpreted as zero.

These methods are appropriate to describe the posted information without overstating it. The comparison is a simple binary-outcome contrast, accompanied by a confidence interval and an unadjusted p-value. The registry does not report a more extensive analysis plan, missing-data methods, or analysis populations. Those omissions limit the assessment of how the analysis was conducted and how robust the result might be to alternative assumptions.

10. Confidence Intervals and Precision

The risk-difference estimate is 0.023, but the estimate alone does not convey how uncertain the comparison is. Its confidence interval crosses zero and is broad, extending to values on either side of no difference. This means the data do not precisely determine the direction or magnitude of the underlying group difference. The interval is the key precision information for judging how much the posted estimate can be distinguished from other plausible effects under the analysis assumptions.

In general, a confidence interval represents the uncertainty in an estimate using the observed data and a statistical method. A wider interval indicates less precision than a narrower one. It is not a guarantee about the true effect, and its interpretation depends on the method's assumptions. Because the registry's posted interval includes zero, it is consistent with no difference as well as with differences favoring either arm. It does not prove equality or establish that the groups are clinically interchangeable.

The limited group sizes are important context for this uncertainty. The posted counts are 12 of 13 and 9 of 10. When event counts and denominators are limited, estimates can be sensitive to the outcomes of only a small number of participants. The registry record does not report a detailed analysis plan, confidence-interval method, or missing-data handling, which further constrains evaluation of the calculation.

The p-value of 0.846 is not a substitute for the interval. It addresses a different question and does not describe the range of effects compatible with the data. The most defensible reading is that the posted comparison is imprecise: the observed counts are close, while the interval leaves substantial uncertainty about the underlying difference. That uncertainty should remain central when considering the primary result.

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, not total events. They are illustrative calculations based on the observed effect and are not the trial's original design assumptions. The calculation assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For each planning scenario, the effect to detect is a stated fraction of the observed effect.

For planning at 100% of the observed effect, the figures are 2,376 participants per arm for 80% power and 3,180 participants per arm for 90% power. At 75% of the observed effect, the figures are 4,349 participants per arm for 80% power and 5,821 participants per arm for 90% power. At 50% of the observed effect, the figures are 10,063 participants per arm for 80% power and 13,471 participants per arm for 90% power.

In general, power is the chance that a planned study will detect an effect of the specified size under the calculation assumptions if that effect is present. The 80% power column describes a planning target with that chance; the 90% power column uses a higher target and therefore requires more participants under these assumptions. Smaller effects are harder to distinguish from random variation, so planning to detect a smaller fraction of the observed effect requires more participants per arm.

These are planning illustrations, not observed power. The registry record does not report the trial's original sample-size justification or design assumptions. The displayed calculations should not be taken as evidence that the completed trial had a particular probability of detecting an effect.

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 = 0.900; treatment proportion p₁ = 0.923; average p̄ = 0.912.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4016 = 0.7871.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4013 = 0.3377.
4. n = (0.7871 + 0.3377)² / (0.0231)² = 2,375.7, rounded up to 2,376 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 <- 0.9                          # control proportion
p1 <- 0.923076923076923                          # 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(983749)
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 = 2,376 (80% power), 3,180 (90% power)
 75% of observed effect: n per arm = 4,349 (80% power), 5,821 (90% power)
 50% of observed effect: n per arm = 10,063 (80% power), 13,471 (90% power)
power.prop.test: 2375.7 
Simulated power at n = 2,376 per arm: 0.806 (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 a study tests several outcomes, comparisons, time points, or subgroups. In general, conducting many tests can increase the chance that at least one result appears statistically notable by chance. A prespecified plan may address that issue through outcome hierarchies or other adjustments. The registry lists two outcome measures but reports no secondary outcomes posted and no analyses. It does not report whether multiplicity adjustments were planned or used, so none should be assumed.

The record identifies feasibility and tolerability as the primary outcome and describes interest in blood-flow velocity and stroke symptoms. However, results for those secondary interests are not reported in the registry record. Their absence means the posted primary result should not be treated as a complete account of all outcomes originally contemplated. The registry also does not report whether additional outcomes were measured or analyzed.

Missing-data methods are not reported in the registry record. The displayed primary outcome counts use the arm denominators listed in the results, but the record does not explain whether any participants had incomplete assessments, how such cases were handled, or whether any data were excluded. It also does not provide a participant flow account that would allow those details to be assessed. The registry's enrollment entry is 23, while the listed arm sizes are 13 and 10; no further explanation of their relationship is given.

The analysis populations are not reported in the registry record. It is therefore not possible to determine whether the posted comparison follows an intention-to-treat approach, a per-protocol approach, or another population definition. In general, the chosen analysis population can affect interpretation, especially when adherence, exclusions, or missing outcomes occur. No assumptions about those matters should be added to the posted counts.

The record lists zero analyses and does not include a detailed statistical analysis plan. It also does not report interim analyses, multiplicity control, prespecified subgroup analyses, or sensitivity analyses. This lack of detail does not establish that such procedures did or did not occur outside the registry record; it means they cannot be evaluated from the information recorded. Accordingly, the primary comparison is best read as the posted unadjusted contrast, with uncertainty about analysis choices that are not documented.

14. 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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: transient ischaemic attack (7.7%), chronic obstructive airways disease exacerbated (7.7%).

The registry reports serious adverse events in 1 participant in the full-pressure ECP group and 0 participants in the sham-pressure ECP group. Other adverse events are reported in 12 participants in the full-pressure group and 8 in the sham-pressure group. These are counts, not estimates that establish a treatment-related risk. The record lists transient ischaemic attack and exacerbated chronic obstructive airways disease among the serious adverse-event terms, each reported as 7.7%.

Adverse-event counts need context. In general, a recorded event may be related to treatment, the underlying condition, or another cause; the count alone cannot determine attribution. The registry record does not report event timing, detailed severity assessments, participant-level overlap, or causal attribution. It also does not provide enough information to determine whether the terms listed correspond to the same or different participants.

Death counts are not reported in the registry record; a missing entry should not be read as zero deaths. With the limited group sizes and few serious events, the counts are descriptive and imprecise. They do not by themselves establish comparative safety or rule out uncommon harms.

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

The posted primary results show that 12 of 13 participants in the full-pressure arm and 9 of 10 in the sham-pressure arm met the feasibility and tolerability outcome as defined. The estimated risk difference is 0.023, with an unadjusted p-value of 0.846. The confidence interval crosses zero and is broad, so the comparison does not precisely establish the direction or magnitude of a difference between the arms.

These findings concern the ability to tolerate treatment and meet the study's specific feasibility criteria during one hour. They are not evidence about recovery, disability, or longer-term stroke outcomes. The registry describes blood-flow velocity and stroke symptoms as study interests, but results for secondary outcomes are not reported in the registry record. No conclusion about those outcomes can be drawn from the primary endpoint counts.

The counts also do not establish equivalence. A failure to distinguish the groups statistically is not proof that their effects are the same. The uncertainty reflected in the confidence interval leaves room for differences in either direction. Nor does the posted result show that ECP has or lacks a clinically meaningful effect; the endpoint was designed around feasibility and tolerability, not a definitive clinical-benefit comparison.

Safety information is similarly limited. Serious adverse events were recorded in the full-pressure group and not in the sham-pressure group, while other adverse events were recorded in both. The registry does not report enough detail to determine causality or fully characterize those events. Death counts are not reported in the registry record. These results therefore provide descriptive safety observations rather than a complete safety assessment.

The appropriate interpretation is bounded by the trial's phase 1 purpose, the posted group sizes, the endpoint definitions, and the missing analysis detail. The registry record reports an early comparison of a full-pressure procedure with sham pressure, but it does not resolve questions about clinical effects, long-term outcomes, or safety beyond the observed record.

16. Why This Trial Matters Statistically

This trial is statistically important because it illustrates the difference between an early feasibility question and a definitive clinical-effect question. The primary endpoint asks whether the procedure could be tolerated and sustained under explicit, arm-specific criteria. Such an endpoint can inform whether a procedure is deliverable in a study setting, but it does not answer whether it changes stroke symptoms or patient outcomes.

The comparison also shows why effect estimates should be read with their uncertainty. The posted risk-difference estimate is 0.023, and the confidence interval crosses zero. The p-value is 0.846. Those statistics do not prove that the groups are identical. The interval indicates that the available data leave uncertainty about both the direction and magnitude of the difference. A point estimate alone can make a result seem more decisive than the underlying precision warrants.

Safety summaries require similar restraint. Serious and other adverse-event counts are useful descriptive information, but counts from a limited early-phase study cannot fully characterize uncommon events or establish causation. The registry does not report all details needed to assess timing, attribution, or participant-level overlap, and it does not report death counts.

The registry also highlights the value of complete reporting. Secondary outcome results, analysis populations, missing-data procedures, and detailed statistical methods are not reported in the registry record. Without these details, readers cannot determine how the posted analysis relates to a full prespecified plan or how sensitive it might be to analytic choices. The planning illustrations can help explain sample-size demands, but they are not the original design justification. Together, these issues make careful separation of observed counts, uncertainty, and unanswered questions central to 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.

This is a phase 1 trial with limited numbers in each arm, so the primary comparison is imprecise and adverse-event counts are descriptive. The broad confidence interval crossing zero does not establish equivalence or exclude potentially meaningful differences. The registry record does not report the original sample-size justification, so it is not possible to assess the intended precision or target effect from the study's own design assumptions.

The outcome definitions are specific and differ between the full-pressure and sham-pressure arms. The full-pressure criterion refers to sustained tolerance at a pressure capable of a specified MFV augmentation in a specified proportion of subjects, while the sham criterion refers to sustained tolerance of sham pressure in all subjects. This makes the endpoint closely tied to the trial's operational feasibility question and limits broader interpretation as general tolerability or clinical response.

Further limitations are the absence of posted secondary results and the lack of registry detail on analysis populations, missing-data handling, masking procedures, randomization methods, and the statistical analysis plan. The record lists no analyses. It therefore does not allow evaluation of whether additional adjustments or sensitivity analyses were conducted. Death counts are not reported in the registry record, and adverse-event attribution and participant-level overlap are not described.

The registry also does not provide results for blood-flow velocity or stroke symptoms, despite describing these as study interests. Consequently, the posted record cannot support conclusions about those outcomes. These limitations apply to what can be inferred from the registry record; they should not be filled in by assumption.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Safety Study of External Counterpulsation as a Treatment for Acute Ischemic Stroke — NCT00983749.
• Registry record last updated 2016-06-28; status last verified 2016-05.
• Results first posted 2016-06-28.
• 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

NCT00983749 is a completed, randomized, controlled phase 1 parallel-group study sponsored by the University of California, San Diego. The record lists single masking, enrollment of 23, a start of 2009-12, and primary completion in 2013-02. It compared one-hour full-pressure ECP, escalated in a tiered manner up to 300mmHg, with one-hour sham-pressure ECP at 75mmHg.

The primary outcome was feasibility and tolerability during treatment. The posted counts meeting that outcome are 12 of 13 in the full-pressure arm and 9 of 10 in the sham-pressure arm. The comparison has a reported risk difference of 0.023 and an unadjusted p-value of 0.846; its confidence interval crosses zero. Serious adverse events were reported in 1 participant in the full-pressure arm and 0 in the sham-pressure arm. Secondary outcome results and further analysis details are not reported in the registry record.