Second-Line Uterotonics in Postpartum Hemorrhage: A Randomized Clinical Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Second-Line Uterotonics in Postpartum Hemorrhage: A Randomized Clinical Trial (NCT03584854), a randomized phase 4 study. The primary outcome is Uterine Tone Score.

2. Clinical Question

3. Trial Design

This was a phase 4 randomized, double-masked, parallel-group trial sponsored by Brigham and Women's Hospital. The registry record lists enrollment of 100 participants and a completed status. Participants were assigned to one of two medication sequences: intramuscular carboprost followed by methylergonovine if needed, or intramuscular methylergonovine followed by carboprost if needed.

The design compares the first-drug strategies in a setting where a second uterotonic could be used. Consequently, the clinical contrast is not simply exposure to one medication versus exposure to the other in isolation. It is the comparison of the assigned initial drug and the subsequent treatment sequence as described in the arms. The registry record does not report how often the second drug was administered within each arm as part of the primary analysis.

The primary outcome was uterine tone at 10 minutes following administration of the first study drug. The registry record also lists secondary outcomes, including tone at an earlier assessment, use of additional uterotonic medication, transfusion, additional intervention, quantitative blood loss, hematocrit drop, hospital stay, and severe maternal morbidity. Results for some listed outcomes are not reported in the registry record.

Although the record identifies randomization and double masking, it does not report the allocation procedure, who was masked, or details of the masking implementation. The original sample-size justification and the trial's prespecified statistical analysis plan are not reported in the registry record. Thus the posted design information establishes the broad randomized comparison, but not all operational or analytic details needed to assess how the trial was planned and conducted.

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: In the carboprost group, 50 participants were enrolled and data for 49 participants were analyzed since the primary outcome was not measured in 1 participant. In the methylergonovine group, 50 participants were enrolled and data for 48 were analyzed since 2 participants did not have the primary outcome measured.

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 Uterine Tone Score, measured over at 10 minutes following administration of the first study drug.

The primary endpoint was the obstetrician-assessed uterine tone score at 10 minutes after the first study drug. It used a 0-10 scale: 0 represented no tone and 10 represented perfect or excellent tone. The registry record reports the group mean and standard deviation for this continuous measure.

Tone is relevant to the trial question because the study aimed to compare the immediate uterine response to the two second-line medication strategies. A scored assessment turns that clinical observation into a measure that can be summarized across participants and compared between randomized groups. Because the scale is bounded and based on an observer's assessment, the score should be understood as a rating of tone at a specified time, not as a direct measure of blood loss, transfusion, or later clinical outcomes.

The endpoint's timing is important to interpretation. It captures tone at a single assessment after the first assigned medication, while the treatment strategies allowed the alternate drug to be given if needed. The registry record does not describe how the score was assessed beyond identifying the obstetrician as the assessor, nor does it report further details about training or assessment reliability. The result therefore addresses the recorded tone score at the specified time and should not be read as a complete measure of every aspect of postpartum hemorrhage management.

7. Results Posted in the Registry

• Methylergonovine Maleate: mean 4.40 versus 4.70 on control; difference -0.30 (95% CI -1.04 to 0.44), standardized effect d = -0.16; z-test p = 0.425.

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.

Uterine Tone Score

• 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:
  • Methylergonovine Maleate: mean 6.40 versus 6.80 on control; difference -0.40 (95% CI -1.15 to 0.35), standardized effect d = -0.21; z-test p = 0.295.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Methylergonovine Maleate and the control. The interval gives the range of effects that remain compatible with the data.

Number of Subjects Receiving Additional Uterotonic

• 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:
  • Methylergonovine Maleate: 30.0% (15/50) versus 34.0% (17/50) on control; risk difference -4.0 percentage points (95% CI -22.3 to 14.3), risk ratio 0.88 (0.50–1.57); unadjusted two-proportion z-test p = 0.668.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Methylergonovine Maleate and the control. The interval gives the range of effects that remain compatible with the data.

Number of Subjects Requiring Transfusion

• 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:
  • Methylergonovine Maleate: 6.0% (3/50) versus 4.0% (2/50) on control; risk difference 2.0 percentage points (95% CI -6.5 to 10.5), risk ratio 1.50 (0.26–8.60); unadjusted two-proportion z-test p = 0.646.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Methylergonovine Maleate and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover several related dimensions of postpartum hemorrhage management: an earlier uterine tone assessment, use of additional uterotonic medication, transfusion, and other outcomes listed in the registry record. The posted secondary results span an immediate clinical rating and subsequent treatment or event measures. These outcomes provide context for the primary tone comparison, but they address different questions and need not move in the same direction.

The earlier tone measure is related to the primary endpoint because both assess uterine tone, but the assessment times differ. Additional medication use reflects treatment escalation, while transfusion is a separate clinical event. Agreement across such measures could add context, but none should be treated as a direct substitute for the primary endpoint or as a definitive explanation of its result. Other listed outcomes, including blood loss, hematocrit change, additional intervention, length of stay, and severe maternal morbidity, do not have results reported in the registry record.

The posted secondary comparisons are unadjusted. When several outcomes are considered, chance alone can produce an apparently notable comparison, especially when findings are selected after looking across outcomes. The registry record does not report a multiplicity strategy or an outcome hierarchy that would identify which secondary comparisons were confirmatory. For that reason, secondary findings carry less evidential weight than a result from a clearly prespecified primary analysis, and they are best viewed as supportive or exploratory descriptions.

Overall, the secondary information broadens the clinical picture but does not remove the uncertainty in the primary comparison. It also cannot establish a consistent treatment advantage across outcomes when some listed results are absent and the reported comparisons are not adjusted for multiple testing.

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

• Uterine Tone Score
• Number of Subjects Receiving Additional Uterotonic
• Number of Subjects Requiring Transfusion
• Number of Subjects Requiring Additional Intervention
• Quantitative Blood Loss (QBL)
• Hematocrit Drop
• Length of Hospital Stay
• Number of Subjects Experiencing Severe Maternal Morbidity

9. Statistical Interpretation

10. Statistical Methods Explained

The registry record reports group means and standard deviations for the primary continuous outcome, along with a mean difference, confidence interval, and unadjusted p value. The posted comparisons are recalculated unadjusted comparisons from the summary statistics; the registry record reports no analyses. The estimates are useful for describing the data, but they are not a substitute for a full prespecified analysis with participant-level information and documented decisions about missing values and multiplicity.

What does a mean difference represent?

A mean is the arithmetic average of a measured quantity within a group. Here, the quantity is the uterine tone score. The mean difference compares the group averages. The posted primary estimate is defined as the methylergonovine-first mean minus the carboprost-first mean, so a negative estimate indicates a lower observed average in the methylergonovine-first group. This direction describes the contrast; it does not by itself establish that the difference is important or reliable.

The standard deviation describes how dispersed individual scores are around their group mean. It is not the same as the uncertainty in the group mean or in the difference between means. A group can have individual scores spread around its average while the estimated average difference is more or less precise depending on the amount of information available and the variability of scores.

How should the confidence interval be read?

A confidence interval gives a range of values compatible with the observed data under the assumptions of the calculation. It conveys uncertainty around an estimate in a way that the point estimate alone cannot. When an interval for a difference includes zero, the data are compatible with no difference as well as with differences in either direction within the interval. This is not proof that the true effect is zero. The interval's width is also informative: a wider range means less precision about the possible size of the difference.

For the primary comparison, the posted interval includes zero. That pattern aligns with the unadjusted p value of 0.425: the data do not provide strong evidence against a no-difference value under the calculation used. A p value is not the probability that the null hypothesis is true, nor is it a measure of clinical importance. It is a measure of how unusual data at least as incompatible with the null would be under the null model and its assumptions.

What does unadjusted mean?

An unadjusted comparison does not account for multiplicity across the several outcomes or comparisons. When many outcomes are examined, the chance of at least one apparently notable result can increase, even when there is no underlying effect. Adjustment methods or a prespecified hierarchy can address that issue, but the registry record does not report such a plan or adjustment. The results should therefore be read as individual comparisons, with particular caution for secondary outcomes.

The record does not report a model specification, covariate adjustment, or detailed assumptions for the primary analysis. It also does not report an analysis population such as intention-to-treat or per-protocol. The posted comparison should not be attributed to an unreported model or analysis strategy. In randomized trials, preserving the assigned groups in analysis is often important for estimating the effect of assignment, but whether and how that was done beyond the posted outcome counts is not reported in the registry record.

Why can over-dispersion matter?

Over-dispersion is a general statistical issue in count or binary data: the observed variability may be greater than a simple model assumes. If a model understates variability, standard errors can be too small and intervals too narrow. The primary endpoint is a score analyzed as a continuous measure, and the posted summary does not describe a model for over-dispersion. This concept is more relevant to some count or event outcomes than to interpreting the reported mean and standard deviation directly. The registry record does not report model diagnostics for the secondary outcomes.

For binary outcomes, a count of participants with an event is commonly compared with the number at risk in each group. Differences in event frequency can be expressed in several ways, such as a risk difference or a relative measure. These measures answer different questions and can behave differently when events are sparse. The posted secondary comparisons include recalculated risk differences, but no sponsor-reported analysis or detailed modeling approach is reported in the registry record.

Finally, the registry record's lack of reported analyses means that readers cannot verify whether the displayed comparisons match a prespecified primary analysis. The recalculated results are transparent summaries of the posted group statistics, but interpretation should remain tied to those statistics and to the limits of what the record documents.

11. Confidence Intervals and Precision

The primary mean difference is -0.3, with a confidence interval that includes zero and extends on both sides of it. This interval communicates that the estimate is not precise enough to determine the direction of a possible difference with confidence. It also shows why the point estimate should not be treated as the only plausible value: values away from the estimate remain compatible with the data under the interval calculation.

Precision is different from whether a result crosses a significance threshold. A wide interval can include zero and still include effects that might matter; a narrow interval can exclude zero while representing a small difference. The registry record does not state what size of change in the tone score would be clinically important. Without that context, the interval can describe statistical uncertainty but cannot by itself determine whether the remaining possibilities are clinically meaningful.

The outcome was not measured for some enrolled participants in each arm, and the available primary measurements were used in the posted comparison. Missing observations reduce the information available and may affect the result if the reasons for missingness relate to the score or treatment. The record does not report analyses that quantify this possibility. Further, the primary result is one outcome at one assessment time; it does not provide precision for unreported outcomes or later events.

Overall, the interval and p value support a cautious reading: the observed data do not resolve whether a difference exists or how large it might be. They should not be interpreted as proof of equivalence or as evidence that all potentially important differences have been excluded.

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 participants per arm for 80% power and participants per arm for 90% power. These are not total trial enrollment figures and do not describe the number enrolled in this trial. The displayed outcomes are not time-to-event outcomes, so the planning quantities are participants per arm rather than total events.

The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect used for planning is a stated fraction of the observed effect. The displayed rows therefore illustrate planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. The original design assumptions used by the trial investigators are not reported in the registry record.

In plain language, 80% power is the planning probability of detecting the specified effect under the stated assumptions if that effect is the true one; 90% power uses a higher target probability under those same assumptions. These columns are planning targets, not guarantees about a future study. They also do not tell whether this completed trial had a particular probability of finding a result. This is not observed power.

The figures increase as the effect to detect becomes smaller: a subtler difference is harder to distinguish from ordinary variation, so the planning illustration calls for more participants per arm. The numbers should be read as conditional illustrations based on the observed effect and stated planning assumptions, not as a retrospective justification of the trial's enrollment or as evidence that a particular sample size would be adequate for every outcome.

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

Worked calculation for this trial

1. Observed difference Δ = -0.300; pooled standard deviation σ = 1.851.
2. n = 2 × (1.96 + 0.842)² × 1.851² / -0.300² = 597.7, rounded up to 598 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 <- -0.3                      # observed difference in means
sigma <- 1.85120160123663                     # 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(3584854)
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 = 598 (80% power), 801 (90% power)
 75% of observed effect: n per arm = 1,063 (80% power), 1,423 (90% power)
 50% of observed effect: n per arm = 2,391 (80% power), 3,201 (90% power)
power.t.test: 598.7 
Simulated power at n = 598 per arm: 0.802 (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.

14. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary outcome and multiple secondary outcomes, but does not report a multiplicity adjustment, a hierarchical testing procedure, or a prespecified rule for interpreting families of outcomes. When multiple statistical comparisons are made, the chance of observing at least one apparently unusual result can rise. Unadjusted secondary p values should therefore not be treated as if each were the sole planned test. The original statistical analysis plan and the trial's approach to multiplicity are not reported in the registry record.

There is also missing primary outcome information. The record states that the primary outcome was not measured for one participant in the carboprost group and two participants in the methylergonovine group. The displayed primary outcome data use 49 and 48 participants, respectively. The reasons the measurements were unavailable beyond the stated nonmeasurement are not reported in the registry record. No imputation, missing-data sensitivity analysis, or comparison of participants with and without measured outcomes is reported.

Missing data can matter if the chance of an outcome being unmeasured is related to the outcome itself or to treatment assignment. The available-case comparison is straightforward to describe, but without further information it cannot establish that missingness had no effect on the estimate. The registry record does not specify how missing observations were handled for every secondary outcome.

The record also does not report whether the primary analysis followed intention-to-treat, per-protocol, or another defined analysis population. Randomization describes assignment, but analysis-population definitions determine which participants contribute to a particular estimate. The posted denominators indicate the observations used for the primary endpoint, not the full analytic policy. These reporting gaps limit assessment of the analysis's prespecification, missing-data handling, and control of false-positive findings across outcomes.

15. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry record reports no serious adverse events among the 50 participants at risk in either group, and no deaths in either group. It also reports other, non-serious adverse events affecting 14 participants in the carboprost-first group and 13 participants in the methylergonovine-first group, with 50 participants at risk in each group. The registry record does not provide serious adverse event terms.

These are recorded counts during the reported safety assessment, not proof that serious events cannot occur or that the strategies have the same safety profile. A zero count in a trial of this size does not establish absence of risk. The record does not give detailed descriptions of the other adverse events, their severity, timing, or relationship to study medication, so those counts cannot explain the types of events observed. Safety interpretation should remain limited to the adverse-event information recorded.

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

The central statistical finding is that the posted primary comparison does not clearly distinguish the two randomized treatment sequences on the obstetrician-assessed tone score at 10 minutes. The estimated mean difference is -0.3, and its confidence interval includes zero. The result is compatible with uncertainty about both the direction and size of a possible difference.

This is not the same as demonstrating equivalence. Equivalence requires a prespecified margin defining how small a difference must be to count as clinically unimportant, along with an analysis designed to evaluate that margin. Such a margin and an equivalence analysis are not reported in the registry record. A nonsignificant comparison alone cannot show that the treatments have the same effect.

Nor does the result show that either strategy is ineffective. The trial assessed a particular outcome at a particular time and reported an uncertain between-group estimate. It did not establish all consequences of treatment for blood loss, transfusion, additional intervention, maternal morbidity, or other outcomes. Some secondary outcomes are listed without results in the registry record.

The randomized design is relevant because it aims to compare assigned strategies between groups, but the posted result remains limited by the information available: primary measurements were missing for some participants, the analysis plan and analysis population are not reported, and the comparison is unadjusted. These features do not erase the observed data; they shape how strongly conclusions can be drawn from them.

Accordingly, the result supports a restrained statement about the record: the observed primary scores do not provide clear evidence of a difference between the assigned strategies, while the interval leaves uncertainty about the true difference. It should not be generalized beyond the study population, treatment sequences, outcomes, and timing documented in the registry record.

17. Why This Trial Matters Statistically

This trial illustrates why reporting an effect estimate with its uncertainty is more informative than relying on a thresholded p value alone. The point estimate gives the direction and scale of the observed mean contrast, while the confidence interval shows that the data remain compatible with values on either side of no difference. That combination discourages both an exaggerated claim of superiority and an unwarranted conclusion of equivalence.

It also demonstrates how endpoint choice shapes interpretation. Uterine tone is a scored clinical assessment, while use of additional medication and transfusion represent different aspects of care. A result on one measure does not automatically predict results on the others. Distinguishing the outcome, assessment time, and treatment sequence is essential when interpreting a randomized comparison.

Finally, the record highlights the value of complete statistical reporting. Details about the original sample-size rationale, analysis plan, multiplicity, missing-data handling, and analysis population would help readers determine how the posted comparisons relate to the trial's prespecified aims. The planning illustrations can show how precision depends on assumed effect size, but they cannot replace information about the original design or convert an uncertain estimate into a definitive conclusion.

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 primary endpoint is an obstetrician-assessed score. Although the scale's anchors are described, the registry record does not report assessment training, reliability, or procedures for resolving disagreement. The score may capture an important clinical observation, but it is not a direct measure of blood loss or a complete measure of clinical course.

The treatment arms are sequences that permit the alternate medication if needed. The primary comparison therefore concerns the assigned first-drug strategy in that treatment context. The registry record does not report how often participants received the follow-on medication in each group, which limits interpretation of how much the observed comparison reflects the initial medication alone versus subsequent treatment.

Primary outcome measurements were unavailable for some enrolled participants, and the record does not report an approach to imputation or sensitivity analysis. The original sample-size justification, detailed statistical plan, and analysis population are also not reported. These omissions make it difficult to assess how missing outcomes and analytic choices could affect the estimate.

The primary result is for one outcome at one time point. The registry record lists additional clinical outcomes, but results for several are not reported. The secondary comparisons are unadjusted, and a multiplicity strategy is not reported. Together, these limitations mean that the record supports a focused description of the available primary comparison, rather than broad conclusions across all outcomes or settings.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Second-Line Uterotonics in Postpartum Hemorrhage: A Randomized Clinical Trial — NCT03584854.
• Registry record last updated 2025-10-16; status last verified 2025-09.
• Results first posted 2025-10-16.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

22. Record Summary

The registry record describes a completed phase 4, randomized, double-masked, parallel-group comparison of two second-line uterotonic sequences for atonic postpartum hemorrhage. Enrollment was 100. The primary outcome was the obstetrician-assessed uterine tone score 10 minutes after the first study drug.

The posted primary means were 4.7 for carboprost-first and 4.4 for methylergonovine-first. The recalculated mean difference was -0.3, with an unadjusted p value of 0.425 and a confidence interval that included zero. Primary outcome data were analyzed for 49 and 48 participants, respectively. The registry record does not report the original sample-size rationale, detailed statistical analysis plan, multiplicity approach, or analysis population. The result is uncertain and does not establish either a difference or equivalence between the treatment strategies.