Evaluation of Recombinant Factor VIIa in Patients With Severe Bleeding: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluation of Recombinant Factor VIIa in Patients With Severe Bleeding (NCT00184548), a randomized phase 3 study. The primary outcome is Mortality.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, double-blind, parallel-group, placebo-controlled trial conducted globally. The intervention was eptacog alfa (activated), and the comparator was placebo. The registry record lists enrollment of 554 participants and identifies trauma and acquired bleeding disorder as conditions.

The trial title describes severely injured trauma patients with bleeding refractory to standard treatment. The registry lists blunt- and penetrating-trauma groups. For the primary mortality outcome, however, results are reported only for blunt trauma. The record states that no penetrating-trauma primary analysis was done because of low statistical power. The primary analysis population is described as the blunt-trauma intention-to-treat set, with patients who discontinued, withdrew, or were lost to follow-up before day 30 excluded from the mortality analysis.

The registry status is terminated. Its summary states that this trial and trial F7TRAUMA-1648 (NCT00323570) have been merged. The reason for termination, the circumstances of the merger, and the original design assumptions are not reported in the registry record. The registry also does not report the allocation ratio or the details of the randomization procedure. The listed phase, masking, and parallel-group features describe the broad design, but they do not establish further operational details.

Mortality was the primary outcome, with secondary outcomes covering days alive and free of pulmonary or renal dysfunction requiring medical intervention, time to death, red blood cell transfusion, massive transfusion, and allogeneic transfusion. The record lists 7 outcome measures and 6 analyses. It does not report a complete statistical analysis plan or explain how the outcomes were prioritized beyond identifying mortality as primary.

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: Eligible subjects were those with either blunt and/or penetrating trauma injuries and confirmed clinical indicators for active haemorrhage refractory to standard treatment (including blood component therapy and surgical haemostatic procedures).

Overall Study

Analysis population for the primary outcome, as defined in the registry: Blunt trauma patient population: Intention-to-treat (ITT) analysis set. Patients who discontinued (withdrawn or lost to follow up) before day 30 were excluded from analyses. Penetrating Trauma patient population: No analysis done due to low statistical power.

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 Mortality, measured over from day 0 to 30.

The primary endpoint was the number of participants who died from any cause from day 0 to 30. The outcome is a binary event for each participant: death during the specified follow-up period or no recorded death during that period. It is presented as participant counts in the blunt-trauma groups.

Mortality is used to assess whether the groups differ in survival over the stated follow-up window. Because it counts deaths from all causes, it does not depend on attributing each death to a particular cause. In general, a binary endpoint can be compared by examining event proportions, a difference in those proportions, or a relative measure such as a risk ratio or odds ratio. These measures answer related but not identical questions.

The registry describes the primary analysis as intention-to-treat for the blunt-trauma population, but it also says participants who withdrew or were lost to follow-up before day 30 were excluded from the mortality analysis. The number excluded and the handling of their outcomes are not reported in the registry record. No primary mortality analysis was done in the penetrating-trauma population, which the record attributes to low statistical power. Therefore, the posted primary result addresses the blunt-trauma analysis population rather than providing an estimate for all enrolled trauma patients.

7. Results Posted in the Registry

• rFVIIa, Blunt Trauma: 11.0% (24/218) versus 10.7% (26/242) on control; risk difference 0.3 percentage points (95% CI -5.4 to 6.0), risk ratio 1.02 (0.61–1.73); unadjusted two-proportion z-test p = 0.927.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Number of Days Alive and Free of Pulmonary and/or Renal Dysfunction Requiring Medical Intervention

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • rFVIIa, Blunt Trauma: mean 17.20 versus 16.40 on control; difference 0.80 (95% CI -1.09 to 2.69), standardized effect d = 0.08; z-test p = 0.406.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between rFVIIa, Blunt Trauma and the control. The interval gives the range of effects that remain compatible with the data.

Number of Units of Transfused Red Blood Cells From Time of First Dose

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • rFVIIa, Blunt Trauma: mean 6.90 versus 8.10 on control; difference -1.20 (95% CI -3.13 to 0.73), standardized effect d = -0.11; z-test p = 0.223.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between rFVIIa, Blunt Trauma and the control. The interval gives the range of effects that remain compatible with the data.

Number of Patients Receiving 10 Units or More (Massive Transfusion) of Red Blood Cells From Time of Injury

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • rFVIIa, Blunt Trauma: 50.2% (111/221) versus 54.3% (134/247) on control; risk difference -4.0 percentage points (95% CI -13.1 to 5.0), risk ratio 0.93 (0.78–1.10); unadjusted two-proportion z-test p = 0.384.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between rFVIIa, Blunt Trauma and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several aspects of early clinical course and resource use: days alive without specified pulmonary or renal dysfunction, timing of death, red blood cell transfusion, massive transfusion, and allogeneic transfusion. Five secondary results are posted. Across the available comparisons, the analyses do not form a consistent pattern of clear separation between the blunt-trauma groups. Most results are statistically uncertain, while the registry's Wilcoxon result for red blood cell units has a p-value below the conventional threshold used in many analyses.

That result should be read in context rather than treated as independent confirmation of the primary endpoint. It concerns a different outcome and uses a rank-based method, whereas the primary endpoint is mortality. The primary mortality analyses did not identify a statistically clear difference. A secondary finding cannot, by itself, establish a mortality effect or resolve uncertainty in the primary result.

Several outcomes were examined, and the registry record does not report whether the analyses were adjusted for multiple comparisons. In general, when many outcomes are tested, the chance of obtaining at least one apparently notable result by chance increases. Unadjusted secondary p-values therefore carry less evidential weight than a prespecified primary analysis, especially when the full testing hierarchy and analysis plan are unavailable.

Not all outcomes have posted results for both trauma types. The secondary outcome entries shown are for blunt trauma, while no penetrating-trauma secondary comparisons are posted in these results. The record also lists additional secondary outcomes for which outcome-level numerical results are not presented in the registry entries. The overall pattern should therefore be understood as a description of the posted blunt-trauma secondary analyses, not as a complete assessment of every listed outcome or of all enrolled participants.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Number of Days Alive and Free of Pulmonary and/or Renal Dysfunction Requiring Medical Intervention
• Time to Death From Time of First Dose
• Number of Units of Transfused Red Blood Cells From Time of First Dose
• Number of Patients Receiving 10 Units or More (Massive Transfusion) of Red Blood Cells From Time of Injury
• Number of Units of All Allogeneic Transfusions From Time of First Dose

9. Statistical Interpretation

10. Statistical Methods Explained

For mortality, the recalculation treats each participant as having either experienced the event or not, and compares the unadjusted event proportions in the blunt-trauma groups. The posted group counts are 24 deaths among 218 participants receiving rFVIIa and 26 among 242 receiving placebo. The recalculated unadjusted p-value is 0.927. This approach does not adjust for other participant characteristics or incorporate the timing of death.

The registry reports a logistic regression analysis for the primary endpoint, with p = 0.934, but does not post an estimate or confidence interval for that analysis. Logistic regression models the odds of a binary event. In general, an adjusted regression estimate can differ from an unadjusted comparison because the model may account for variables beyond group assignment. The adjustment variables and full model specification are not reported in the registry record, so the precise relationship between the posted logistic result and an unadjusted estimate cannot be reconstructed from the record.

A second registry analysis uses a Cox proportional hazards model. It reports a hazard ratio of 1.18, a 95% confidence interval from 0.64 to 2.17, and p = 0.589. Unlike a comparison based only on whether death occurred by the end of follow-up, a time-to-event analysis uses the observed time to death and can account for differing follow-up times. In general, the Cox model relies on assumptions about how the relative hazard behaves over time. The registry record does not report diagnostics or other details that would show how those assumptions were assessed.

The secondary continuous outcomes are posted as means and standard deviations, with unadjusted recalculations of mean differences. For days alive and free of pulmonary and/or renal dysfunction requiring medical intervention, the reported means are 17.2 days for rFVIIa and 16.4 days for placebo. The recalculated mean difference is 0.8 days, with a confidence interval from -1.085 to 2.685 and p = 0.406. The registry's ANCOVA analysis reports an LS means estimate of 0.90, a 95% confidence interval from -0.83 to 2.63, and p = 0.308. ANCOVA compares groups after adjustment for model covariates; the covariates are not reported in the registry record.

For red blood cell units transfused from hour 0 to 24, the posted means are 6.9 for rFVIIa and 8.1 for placebo. The recalculated unadjusted mean comparison has p = 0.223. The registry instead reports a Wilcoxon (Mann-Whitney) analysis, with a Hodges-Lehmann shift estimate of 1.0, a 95% confidence interval from 0 to 2, and p = 0.038. The Wilcoxon approach compares the distributions or ranks rather than testing only a difference in arithmetic means. Its reported shift estimate is not interchangeable with the recalculated mean difference. The registry does not explain why this analysis was selected or how the direction of its shift estimate is coded.

The massive-transfusion endpoint is binary: it counts participants receiving 10 units or more of red blood cells from hour 0 to 24. The recalculation compares the event proportions without adjustment; the registry reports a logistic regression odds ratio of 1.175, a 95% confidence interval from 0.817 to 1.690, and p = 0.384. In general, an odds ratio describes the relative odds of the event, not the difference in event proportions. The registry record does not report adjustment variables for this model.

Confidence intervals describe a range of estimates compatible with the data and the statistical model. In general, wider intervals indicate less precision. P-values describe how unusual data at least as incompatible with a null hypothesis would be under that hypothesis and model; they are not the probability that the null hypothesis is true. No p-value alone establishes clinical importance, equivalence, or causation.

11. Confidence Intervals and Precision

Precision is best judged from the confidence intervals, not from the point estimates alone. For the registry's primary time-to-event analysis, the hazard ratio is 1.18, with a 95% confidence interval from 0.64 to 2.17. That interval is broad and includes the null value for a hazard ratio. The data are therefore compatible with different directions and magnitudes of the relative hazard, and the estimate does not pin down the size of a mortality difference precisely.

The registry's logistic regression result for mortality is reported only as p = 0.934. Without a corresponding estimate and confidence interval, its precision cannot be assessed directly from that analysis entry. The unadjusted recalculation reports a p-value but its risk-difference interval is not repeated in the narrative. In general, a p-value does not communicate the full range of plausible effect sizes and should not substitute for an interval.

For the secondary outcomes, the registry reports confidence intervals for the adjusted ANCOVA analysis of days alive and free of dysfunction, the Wilcoxon analysis of red blood cell units, and the logistic regression analysis of massive transfusion. These intervals vary in their width and in whether they include their respective null values. In general, an interval spanning the null does not prove no effect; it indicates that the data do not rule out the null under the analysis model. A narrow interval can still encompass effects with different practical importance, depending on the endpoint.

Precision is also shaped by the number of analyzable participants and events, the variability of the outcome, and exclusions or missing observations. The mortality analysis excludes participants who withdrew or were lost to follow-up before day 30, but the number excluded is not reported in the registry record. The penetrating-trauma primary analysis was not done because of low statistical power. These features limit how confidently the posted primary results can be generalized beyond the blunt-trauma analysis population.

12. 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 are illustrations based on observed effects, not a report of the trial's original sample-size design. For the binary and continuous endpoints, the figures are participants per arm; for time-to-event outcomes, the stated convention is total events. The assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The target effect is the stated fraction of the observed effect.

The 80% power column gives the estimated participants per arm required, under those assumptions, for an 80% chance of detecting the specified effect if that effect were the true one. The 90% power column gives the corresponding planning count for a 90% chance. These are planning probabilities under an assumed effect and model, not guarantees about what a future trial would find.

Smaller target effects require more participants because they are harder to distinguish from random variation. For example, the mortality planning illustration lists 216,075 participants per arm for 80% power when planning for 100% of the observed effect, and 859,670 participants per arm when planning for 50% of the observed effect. The much larger figures reflect how small the observed mortality contrast is in relation to the uncertainty and event frequency in the posted data.

These calculated counts are not the trial's original design requirements. The original sample-size justification and assumptions are not reported in the registry record. This is not observed power: it does not measure whether the completed trial had a sufficient chance of detecting an effect, but illustrates planning counts under stated assumptions and effect scenarios.

13. 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 = 0.107; treatment proportion p₁ = 0.110; average p̄ = 0.109.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4403 = 0.8630.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4403 = 0.3706.
4. n = (0.8630 + 0.3706)² / (0.0027)² = 216,074.7, rounded up to 216,075 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 <- 0.107438016528926                          # control proportion
p1 <- 0.110091743119266                          # 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(184548)
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 = 216,075 (80% power), 289,263 (90% power)
 75% of observed effect: n per arm = 383,105 (80% power), 512,869 (90% power)
 50% of observed effect: n per arm = 859,670 (80% power), 1,150,854 (90% power)
power.prop.test: 216074.7 
Simulated power at n = 216,075 per arm: 0.799 (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.

14. Analyses Reported by the Sponsor

The recalculated primary comparison is an unadjusted comparison of the blunt-trauma mortality counts and gives p = 0.927. The registry's reported logistic regression analysis gives p = 0.934, while its Cox proportional hazards analysis reports a hazard ratio of 1.18, a 95% confidence interval from 0.64 to 2.17, and p = 0.589. These entries use different statistical methods, so their estimates and p-values need not match exactly. The registry does not post an effect estimate or confidence interval for its logistic regression result.

For the primary endpoint, the posted analyses do not point to a statistically clear difference between the groups. The recalculated risk-difference estimate is not a replacement for the registry's own analyses; it is an unadjusted calculation from the posted summary counts. The record does not report enough details about the sponsor's regression model to reproduce or fully assess its adjustment.

In secondary analyses, the registry's adjusted ANCOVA result for days alive and free of dysfunction differs from the unadjusted recalculation, as expected when methods and model specifications differ. For red blood cell transfusion units, the registry reports a Wilcoxon analysis rather than an analysis of the arithmetic means. That distinction matters: the reported rank-based result and the recalculated mean comparison answer different statistical questions. The registry analysis should be interpreted according to its stated method, without treating the unadjusted recalculation as a correction to the sponsor's method.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies mortality as the primary outcome and lists multiple secondary outcomes and analyses. It does not report a multiplicity strategy, a testing hierarchy, or whether any adjustment was made for examining multiple endpoints. This omission matters because repeated testing can increase the chance that at least one result appears statistically notable even when there is no underlying difference. Without the prespecified analysis plan, the interpretation of isolated secondary p-values is uncertain.

The primary mortality population is described as intention-to-treat for blunt trauma, but the record also says that participants who discontinued, withdrew, or were lost to follow-up before day 30 were excluded from the mortality analysis. The number excluded, reasons for exclusion, and missing-outcome assumptions are not reported in the registry record. This means the posted analysis is not a complete account of how all randomized participants contributed to the primary endpoint, despite the intention-to-treat label.

The registry reports no primary analysis for penetrating trauma because of low statistical power. The number of participants in the primary outcome groups for penetrating trauma is listed as zero, while safety denominators are posted for that trauma type. The posted primary estimate therefore cannot be generalized to penetrating trauma on the basis of these results.

Other important analysis details are not reported in the registry record, including the original sample-size justification, the full statistical analysis plan, the covariates in regression and ANCOVA models, and any handling of protocol deviations. These gaps do not invalidate the posted results, but they limit independent assessment of prespecification, model choices, missing data, and the extent to which reported analyses followed the original plan.

16. 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: sepsis (12.0%), pneumonia (10.4%), acute respiratory distress syndrome (7.5%), multi-organ failure (6.8%), deep vein thrombosis (6.2%).

The registry lists serious adverse event counts by trauma type and treatment group. Among blunt-trauma participants, serious adverse events affected 147 of 224 in the rFVIIa group and 177 of 250 in the placebo group. For penetrating trauma, the corresponding counts are 18 of 46 in the rFVIIa group and 20 of 40 in the placebo group. These are descriptive group counts, and the record does not provide a comparative statistical analysis of these figures.

The most frequently listed serious adverse event terms include sepsis at 12.0%, pneumonia at 10.4%, acute respiratory distress syndrome at 7.5%, multi-organ failure at 6.8%, deep vein thrombosis at 6.2%, and small intestinal obstruction at 5.0%. The registry entry does not specify in the registry record how the percentages are assigned across groups or whether participants could contribute more than one event term. These percentages should therefore be read as reported event-term figures, not as a treatment-effect comparison.

The safety entries do not establish that an event was caused by the study treatment. The registry does not report deaths in the safety summary. Group-level adverse event counts also should not be directly substituted for the primary mortality analysis, which has its own endpoint definition and analysis population.

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

The posted primary results describe mortality from day 0 to 30 among blunt-trauma participants included in the analysis. The event counts are 24 of 218 for rFVIIa and 26 of 242 for placebo. The recalculated unadjusted comparison and the registry's reported regression analyses do not show a statistically clear difference between these groups. The Cox estimate is accompanied by a broad confidence interval, so the magnitude and direction of a possible difference remain uncertain.

These findings do not demonstrate that the treatments are equivalent. Equivalence requires a specific design and margin, neither of which is reported in the registry record. A result that is not statistically significant also does not prove that there is no effect; it indicates that the posted data and analysis did not distinguish the groups clearly under the method used.

The result is specific to the blunt-trauma primary analysis population and its follow-up definition. No primary penetrating-trauma analysis was done because of low statistical power. Participants who discontinued or were lost to follow-up before day 30 were excluded from the mortality analysis, and the number excluded is not reported. The primary result therefore does not answer every question about all enrolled participants or all trauma types.

Secondary results address different outcomes and include a statistically notable rank-based comparison for transfusion units. Because several outcomes were considered and a multiplicity adjustment is not reported, that result should not outweigh the primary endpoint or be treated as definitive evidence of a broader effect. The record supports a description of the posted analyses, not medical advice or a conclusion about how any individual should be treated.

18. Why This Trial Matters Statistically

This trial illustrates why a p-value alone is not an adequate summary of a clinical comparison. The primary mortality p-values are not statistically notable, but that does not prove no difference. The Cox confidence interval spans a broad range, making precision central to interpretation. Reporting the event counts, effect estimates, confidence intervals, and analysis methods together helps distinguish an inconclusive result from evidence of equivalence.

It also shows how the statistical question changes with the method. The registry reports logistic regression and Cox analyses for mortality, while the simple recalculation compares unadjusted event proportions. For a secondary transfusion outcome, a Wilcoxon analysis yields a different inference from the unadjusted mean comparison because rank-based and mean-based methods summarize data differently. Clear reporting of these distinctions is essential for readers comparing results.

The population definition matters as much as the model. The primary analysis is restricted to blunt trauma, excludes participants lost to follow-up or withdrawn before day 30, and has no penetrating-trauma primary analysis. Those choices define what the reported estimate can address. The registry record's missing details about exclusions, model adjustments, multiplicity, and the original sample-size rationale further limit independent evaluation.

Finally, planning calculations based on observed effects are not evidence of the trial's original design or its achieved ability to detect an effect. They illustrate how sample requirements depend on the target effect and assumptions. Keeping those planning illustrations distinct from observed results helps prevent overinterpretation of an underpowered or imprecise comparison.

19. 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 main trial-specific limitation is that the posted primary mortality analysis applies only to blunt trauma. The registry says the penetrating-trauma primary analysis was not done because of low statistical power. The primary estimate therefore does not represent a combined estimate for all trauma groups.

The mortality analysis excludes participants who withdrew or were lost to follow-up before day 30, despite being described as intention-to-treat. The number excluded and the consequences of those exclusions are not reported in the registry record. This makes it difficult to assess how missing follow-up could affect the observed comparison.

The registry status is terminated, and its summary says this trial was merged with another trial. The reason for termination, the timing and process of the merger, and the way the merger affected the posted analyses are not reported in the registry record. The record lists enrollment of 554, while outcome analysis denominators vary by endpoint; the reasons for those differences are not reported in the registry record.

Statistical reporting is incomplete for independent assessment. The full statistical analysis plan, original sample-size justification, regression covariates, multiplicity handling, and model diagnostics are not reported. The logistic mortality entry lacks an effect estimate and confidence interval. Several secondary outcomes are listed, but not all have numerical results in the registry entries. These limitations affect precision, reproducibility, and the ability to judge which analyses were prespecified.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Evaluation of Recombinant Factor VIIa in Patients With Severe Bleeding — NCT00184548.
• Registry record last updated 2014-06-25; status last verified 2014-06.
• Results first posted 2009-11-26.
• 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.

23. Record Summary

NCT00184548 is a Phase 3 randomized, double-blind, parallel-group, placebo-controlled trial of eptacog alfa (activated) in severely injured trauma patients with bleeding refractory to standard treatment. The registry lists enrollment of 554 and a terminated status. Its summary notes that the trial was merged with F7TRAUMA-1648 (NCT00323570).

The primary outcome was all-cause mortality from day 0 to 30. Posted primary results concern blunt trauma: 24 deaths among 218 participants in the rFVIIa group and 26 among 242 in the placebo group. No primary analysis was done for penetrating trauma because of low statistical power. The registry reports logistic regression and Cox analyses; the time-to-event estimate is a hazard ratio of 1.18 with a 95% confidence interval from 0.64 to 2.17.

Secondary outcomes cover organ dysfunction-free days, death timing, transfusion, and massive transfusion. Several registry and recalculated analyses are available, but their methods differ. Multiplicity handling, the original sample-size rationale, and key analysis details are not reported in the registry record.