A Dose-ranging Safety and Efficacy Study of Ecallantide to Reduce Surgical Blood Loss Volume: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Dose-ranging Safety and Efficacy Study of Ecallantide to Reduce Surgical Blood Loss Volume (NCT00816023), a randomized phase 2 study. The primary outcome is Cumulative Volume of Packed Red Blood Cells Transfused at 12 Hours Post Surgery.

2. Clinical Question

3. Trial Design

This was a Phase 2 randomized, placebo-controlled, parallel-group, dose-ranging study. The registry lists quadruple masking. Its official title identifies the study as CONSERV-1 and describes subjects exposed to cardiopulmonary bypass during primary coronary artery bypass graft surgery. The purpose recorded in the registry was to assess efficacy and identify optimal dose or doses of ecallantide in reducing blood loss in this surgical setting.

The registry lists an enrollment of 276. That is the trial enrollment figure, not the denominator for every posted analysis. The primary outcome used the Modified Intent to Treat population, defined as randomized subjects who received any amount of study drug and were analyzed according to planned treatment assignment. The outcome group denominators are 65 for low-dose ecallantide, 58 for medium-dose ecallantide, 63 for high-dose ecallantide, and 63 for placebo. The registry does not explain the difference between enrollment and the outcome analysis denominators.

The study started in 2009-03 and its primary completion was 2009-12. Its status is completed. The named sponsor is Cubist Pharmaceuticals LLC, a subsidiary of Merck & Co., Inc. (Rahway, New Jersey USA). The registry lists treatment-emergent adverse events as a secondary outcome. Further detail about the randomization sequence, concealment procedures, masking implementation, or rationale for the planned enrollment is not reported in the registry record. The trial design establishes a randomized comparison, but the posted record does not establish the specific operational safeguards used to carry it out.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Modified Intent to Treat population was the basis of this analysis, defined as all randomized subjects who received any amount of study drug and analyzed according to the planned treatment assignment.

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 Cumulative Volume of Packed Red Blood Cells Transfused at 12 Hours Post Surgery, measured over Start of surgery up to 12 hours after the end of surgery.

The primary endpoint was the cumulative volume of packed red blood cells transfused, measured in mL, from the start of surgery through 12 hours after the end of surgery. It is a continuous outcome: each participant’s measure is a volume, and the reported group summaries are means with standard deviations.

This outcome captures transfusion use during a defined period around surgery. In general, a continuous endpoint can preserve information about the magnitude of measured values rather than reducing them to categories. Transfusion volume is a clinically interpretable measure in a study focused on surgical blood loss, though it is not itself a direct measurement of total blood loss. Transfusion decisions may also reflect clinical practice and patient circumstances, so the endpoint should be interpreted as transfused volume rather than as a direct estimate of bleeding alone.

The registry reports the outcome at 12 hours after surgery and specifies the analysis population. It does not give further details about transfusion thresholds, decisions, or measurement procedures. Those details are not reported in the registry record. The endpoint therefore allows a comparison of the posted group means within its stated time window, while leaving uncertainty about how the clinical decisions contributing to transfusion volume were standardized.

7. Results Posted in the Registry

• Ecallantide Low Dose: mean 401.90 versus 377.60 on control; difference 24.30 (95% CI -189.37 to 237.97), standardized effect d = 0.04; z-test p = 0.824.
• Ecallantide Medium Dose: mean 383.10 versus 377.60 on control; difference 5.50 (95% CI -213.80 to 224.80), standardized effect d = 0.01; z-test p = 0.961.
• Ecallantide High Dose: mean 410.30 versus 377.60 on control; difference 32.70 (95% CI -185.28 to 250.68), standardized effect d = 0.05; z-test p = 0.769.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous because it is a measured volume in mL. The registry summarizes each group using a mean and standard deviation. The mean describes the group’s average observed transfusion volume. The standard deviation describes how dispersed individual measurements are around that average. A large standard deviation relative to a mean indicates that the observations vary substantially; it does not indicate uncertainty in the mean estimate by itself.

What does the reported test assess?

The registry reports a Jonckheere-Terpstra analysis across all treatment groups, with p = 0.474. In general, this rank-based method assesses evidence for an ordered pattern across groups when the groups have a prespecified or meaningful order. It uses the ordering of observed values rather than relying solely on differences between group means. The registry does not state the exact alternative hypothesis or describe how the test was implemented, so further detail about its intended direction or assumptions is not reported in the registry record.

A p-value describes how compatible the observed data are with a specified null model, under the assumptions of the test. It is not the probability that the null hypothesis is true, and it is not a measure of clinical importance. The reported p-value is for the across-group test, not for each individual dose-versus-placebo mean difference. A non-small p-value does not prove that effects are absent; interpretation also depends on the range of values reasonably compatible with the data.

What do the pairwise mean differences represent?

The posted recalculations subtract the placebo mean from each ecallantide group mean. These are simple, unadjusted comparisons based on the posted summary statistics. Positive estimates indicate that the corresponding ecallantide group’s mean transfusion volume is higher than the placebo mean; negative values would indicate a lower mean. The estimated differences are 24.3 mL for low dose, 5.5 mL for medium dose, and 32.7 mL for high dose. These estimates do not adjust for other variables or for making several comparisons.

The uncertainty intervals around these estimates reflect sampling uncertainty under the recalculation’s assumptions. In general, an interval gives a range of estimates compatible with the data and method; a wider interval indicates less precision. All three posted intervals include zero and cover both negative and positive differences. The endpoints are wide relative to the point estimates, so the results do not distinguish a small mean difference from a range of potentially larger differences in either direction. The confidence level for these recalculated intervals is not reported in the registry record.

Why can the mean and rank-based test tell different parts of the story?

A mean comparison focuses on differences in averages. A rank-based test uses the relative ordering of observations and can respond differently when distributions are skewed, have outliers, or differ in shape. In general, these methods do not test identical hypotheses. The registry’s across-group test is not a substitute for the pairwise recalculations, and the pairwise results do not reproduce the across-group test.

The group standard deviations are large in relation to the means, which signals marked spread in the measured outcome. In general, over-dispersion or substantial variability can make group differences harder to estimate precisely because participants’ outcomes vary widely around the average. The posted summaries do not show the underlying outcome distributions, so their shape, outliers, or concentration at particular values cannot be assessed from the means and standard deviations alone.

What analysis population was used?

The primary analysis population was the Modified Intent to Treat population, defined as all randomized subjects who received any amount of study drug and analyzed according to planned treatment assignment. This definition retains assignment as the basis of comparison for included subjects, while excluding randomized subjects who received no study drug. The registry does not report how many randomized subjects, if any, were excluded for that reason, or whether missing outcome values occurred. Details of imputation, handling of incomplete measurements, and sensitivity analyses are not reported in the registry record.

10. Confidence Intervals and Precision

The recalculated low-dose mean difference versus placebo is 24.3 mL, with an interval from -189.4 to 238.0 mL. For medium dose, the estimate is 5.5 mL, with an interval from -213.8 to 224.8 mL. For high dose, the estimate is 32.7 mL, with an interval from -185.3 to 250.7 mL. Each interval includes zero and spans differences in both directions.

In general, a confidence interval conveys the range of values that are compatible with the observed data and the method used to calculate it. A narrow interval indicates greater precision than a wide interval. The posted intervals are broad compared with their estimates, so the point estimates should not be read as precise measurements of the underlying mean differences. They leave substantial uncertainty about the direction and magnitude of the differences. The confidence level attached to these recalculated intervals is not reported in the registry record.

The group means alone do not describe this uncertainty. The standard deviations are 624.9 mL, 620.6 mL, 639.5 mL, and 608.5 mL for low-dose, medium-dose, high-dose ecallantide, and placebo, respectively. Those values describe variability among measurements, while the intervals for the mean differences describe uncertainty in the comparisons. The registry’s Jonckheere-Terpstra result gives p = 0.474, but it does not supply an interval estimate for the across-group test. Its reported confidence level is 95%, while the interval bounds are not reported.

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 illustrations express the required sample size as participants per arm for 80% power and participants per arm for 90% power. They are not total event counts. The calculations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For each illustration, the effect to detect is the stated fraction of the observed effect. These are planning assumptions for the illustration; the trial’s original design assumptions and sample-size justification are not reported in the registry record.

In plain language, the 80% power column gives a planning estimate for the number of participants per arm needed to have an 80% chance of detecting the specified effect under the stated assumptions. The 90% power column applies a higher target chance of detection and therefore generally requires more participants per arm. The planning calculations use the observed effect as a reference, then consider planning for 75% of the observed effect and 50% of the observed effect. Smaller effects need more participants to detect, all else equal.

The values vary markedly by comparison because the observed mean differences vary in relation to the variability in the outcome. For example, planning for 100% of the observed medium-dose effect gives 195,863 participants per arm for 80% power and 262,205 for 90% power. Planning for 100% of the observed high-dose effect gives 5,720 participants per arm for 80% power and 7,658 for 90% power. These are illustrations based on observed effects, not a record of what the trial planned to enroll. This is not observed power.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 24.300; pooled standard deviation σ = 616.889.
2. n = 2 × (1.96 + 0.842)² × 616.889² / 24.300² = 10,116.7, rounded up to 10,117 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 <- 24.3                      # observed difference in means
sigma <- 616.8886269835                     # 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(816023)
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 = 10,117 (80% power), 13,544 (90% power)
 75% of observed effect: n per arm = 17,986 (80% power), 24,078 (90% power)
 50% of observed effect: n per arm = 40,467 (80% power), 54,174 (90% power)
power.t.test: 10117.7 
Simulated power at n = 10,117 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. Analyses Reported by the Sponsor

The registry’s reported analysis is a Jonckheere-Terpstra test across the ecallantide dose groups and placebo, with p = 0.474. It reports a test result rather than a mean difference or an interval for a specific dose-versus-placebo comparison. The recalculated results instead give unadjusted mean differences for each ecallantide dose compared with placebo, along with interval bounds and p-values.

These analyses differ in both method and question. The rank-based test evaluates an ordered pattern across groups, whereas each recalculated comparison estimates the difference between two group means. Consequently, the reported p-value should not be compared as if it were the p-value for any one of the recalculated pairwise comparisons. The registry record does not report sponsor estimates for those pairwise mean differences or specify a multiplicity adjustment for them.

The discrepancy flag is false, so the posted recalculations are not identified as conflicting with the sponsor’s reported analysis. The recalculations remain unadjusted summaries from posted group statistics and should not be mistaken for a replacement of the registry’s across-group analysis. The registry does not describe additional details about the sponsor’s statistical analysis beyond the named method and reported p-value.

14. Multiplicity, Missing Data and Analysis Populations

The registry reports one primary outcome and one across-group analysis using the Jonckheere-Terpstra method. It also posts three recalculated comparisons, each contrasting an ecallantide dose group with placebo. These pairwise recalculations are unadjusted. The registry does not report a multiplicity procedure for controlling the chance of false-positive findings across comparisons, and the recalculated p-values should therefore be read as separate, unadjusted results rather than as a family-wise adjusted analysis.

The registry lists treatment-emergent adverse events as a secondary outcome and reports adverse-event summaries. It does not provide a complete statistical analysis plan or describe whether other outcomes or subgroup analyses were planned. The planned approach to multiplicity across primary, secondary, and safety outcomes is not reported in the registry record. The registry does not report interim analyses or stopping rules; their existence or absence cannot be inferred from the posted information.

The analysis population for the primary outcome is defined as randomized subjects who received any amount of study drug, analyzed according to planned assignment. This definition is informative, but details about missing outcome data are not reported in the registry record. The record does not state how missing values were handled, whether all included subjects had observed outcome measurements, or whether sensitivity analyses were conducted. It also does not explain the difference between the enrollment figure and the reported primary outcome group denominators.

These omissions limit assessment of the full analysis process. In general, a prespecified plan helps readers understand which outcomes and comparisons were prioritized, how multiple tests were handled, and what was done when measurements were incomplete. Without those details, the posted estimates and test result can be described, but the full set of analytic decisions and their consequences cannot be independently evaluated from the registry record.

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: post procedural haemorrhage (9.7%), atrial fibrillation (6.3%), respiratory failure (5.1%), hypotension (5.1%), cerebral infarction (3.4%).

The registry reports serious adverse events among subjects at risk by group: 19 of 62 in the high-dose group, 19 of 59 in the medium-dose group, 15 of 65 in the low-dose group, and 15 of 63 in the placebo group. These are counts with denominators, not an adjusted comparison of safety between groups. The record also lists other adverse events affecting all subjects at risk in each group; the group counts are 62 of 62, 59 of 59, 65 of 65, and 63 of 63, respectively.

The listed serious-event terms and percentages include post procedural haemorrhage at 9.7%, atrial fibrillation at 6.3%, respiratory failure at 5.1%, hypotension at 5.1%, cerebral infarction at 3.4%, and postoperative wound infection at 3.2%. The registry does not identify the denominator or treatment-group distribution for each listed percentage in this summary, so these percentages should not be assigned to a particular group. They are descriptive safety information, not evidence of a causal treatment effect.

Deaths are not reported in the registry record. Further information about event severity, timing, adjudication, or statistical comparisons is not reported in the registry record. The counts should therefore be read as posted safety summaries, not as a complete safety analysis.

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

The registry reports similar-looking group means for packed red blood cell transfusion volume, with standard deviations that indicate substantial variability. The overall Jonckheere-Terpstra test gave p = 0.474. The unadjusted recalculated mean differences versus placebo were 24.3 mL, 5.5 mL, and 32.7 mL for low, medium, and high dose, respectively. Their intervals are broad and include zero, leaving considerable uncertainty about the size and direction of each mean difference.

These results do not establish that the treatment groups have identical outcomes. A test that does not detect an ordered pattern is not proof of equivalence, and the pairwise intervals do not exclude all potentially meaningful differences. Conversely, the observed point estimates alone do not establish a treatment effect. The trial record does not report a prespecified equivalence or non-inferiority margin, and the posted comparisons do not demonstrate equivalence.

The endpoint is transfused packed red blood cell volume over a specified perioperative period, not a direct measurement of all surgical blood loss. The registry’s data can be interpreted as group summaries of transfusion volume within that time window, while clinical decisions and other aspects of the treatment context are not described in detail. The results do not address outcomes beyond those reported in the registry, and they do not give a complete account of how transfusion decisions were made.

In general, interpretation should bring together the estimated differences, their precision, the analysis method, and the population included. Here, the point estimates are small relative to the widths of their intervals, while the group-level standard deviations are large. The posted information therefore supports a cautious description of the measured results, not a definitive claim that ecallantide does or does not reduce blood loss or transfusion requirements.

17. Why This Trial Matters Statistically

This trial is statistically instructive because it pairs a dose-ranging design with a continuous clinical outcome that is highly variable. The group means are close in scale, while the standard deviations are large. That combination helps explain why a mean difference can be estimated with broad uncertainty even when group summaries are available. It also illustrates why the magnitude of an estimate and its precision must be considered together.

The registry’s across-group Jonckheere-Terpstra test and the pairwise mean-difference recalculations represent distinct analytical questions. The test evaluates an ordered pattern over the dose groups and placebo; the recalculations estimate each dose’s mean difference against placebo. Recognizing that distinction prevents a single p-value from being treated as a complete account of all comparisons. The pairwise recalculations are unadjusted, and the registry does not report a multiplicity strategy.

The planning illustrations also show how sample-size requirements depend on the effect used for planning and on variability. Because the observed differences are small relative to the outcome’s spread, the resulting participant-per-arm illustrations can be very large. They are useful as hypothetical planning calculations, not as evidence about the trial’s original design or achieved power. The original design assumptions are not reported in the registry record.

Finally, the record makes the analysis population and endpoint window explicit, but leaves important details about missing data, analysis planning, and outcome measurement unreported. This illustrates a practical lesson in trial interpretation: statistical results are most informative when the estimand, analysis population, methods, and uncertainty are all considered together. Registry summaries allow useful appraisal, while gaps in reporting define the limits of that appraisal.

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.

The posted summary does not include participant-level outcome data or distribution plots, so the shape of transfusion volumes, the presence of outliers, and any concentration of values cannot be assessed from the reported means and standard deviations. This matters because a mean and standard deviation do not show whether the outcome distribution is symmetric or whether a small number of high values influence the average. The registry does not report additional distributional summaries or diagnostics.

The outcome is cumulative packed red blood cell transfusion volume during the stated period, rather than a direct measurement of total surgical blood loss. Transfusion practice may depend on clinical decisions, but thresholds and decision procedures are not reported in the registry record. The record also does not explain how the endpoint was measured beyond its unit and timeframe, or how incomplete measurements were handled.

The primary analysis population excludes randomized subjects who received no study drug, according to its definition. The number of such subjects and the reasons for any exclusions are not reported in the registry record. The record does not state whether outcome data were missing or describe imputation or sensitivity analyses. It also does not report the original sample-size justification, detailed statistical plan, or multiplicity handling.

The recalculated pairwise comparisons are unadjusted and should be interpreted accordingly. Their confidence level is not reported in the registry record. The sponsor’s reported test gives a p-value without an interval estimate or detailed implementation information. Safety summaries are descriptive, and the denominators for the listed serious-event percentages are not specified in the posted summary. These limitations restrict conclusions about precision, analysis choices, and safety comparisons.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: A Dose-ranging Safety and Efficacy Study of Ecallantide to Reduce Surgical Blood Loss Volume — NCT00816023.
• Registry record last updated 2015-08-11; status last verified 2015-07.
• Results first posted 2011-01-12.
• 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

NCT00816023 is a completed Phase 2 randomized, quadruple-masked, parallel-group, placebo-controlled dose-ranging study of ecallantide during coronary artery bypass graft surgery with cardiopulmonary bypass. Enrollment was 276. The primary outcome was cumulative packed red blood cell transfusion volume from the start of surgery through 12 hours after the end of surgery, analyzed in the Modified Intent to Treat population.

The reported Jonckheere-Terpstra analysis across treatment groups gave p = 0.474. Unadjusted recalculated comparisons with placebo gave mean differences of 24.3 mL for low dose, 5.5 mL for medium dose, and 32.7 mL for high dose, each with broad intervals that include zero. The registry also reports treatment-emergent adverse events as a secondary outcome. Details on missing data, the original sample-size justification, and multiplicity procedures are not reported in the registry record.