Safety and Efficacy Trial to Treat Diastolic Heart Failure Using Ambrisentan: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety and Efficacy Trial to Treat Diastolic Heart Failure Using Ambrisentan (NCT00840463), a randomized phase 4 study. The primary outcome is Change in Pulmonary Vascular Resistance (Wood Units).

2. Clinical Question

The study concerns people with pulmonary hypertension and heart failure with preserved ejection fraction, described in the registry record as diastolic heart failure. It asks whether ambrisentan, compared with placebo, is associated with changes in pulmonary vascular resistance and other measured outcomes, while also monitoring safety.

The stated clinical context is that ambrisentan is approved for pulmonary arterial hypertension but not for congestive heart failure. That context explains why a study in this population included safety monitoring as well as physiological and functional assessments; it does not establish the benefit or risk of treatment in this trial. The record lists six-minute walk distance, functional class and the Short Form-36 physical functioning score as secondary outcomes. It also describes echocardiography and invasive hemodynamic measurements. Results for those outcomes are not reported in the registry record. The central statistical question is therefore narrower than the overall study question: what can be learned from the small posted arm-level pulmonary vascular resistance summaries, and how much uncertainty remains because no formal comparative analysis is reported?

3. Trial Design

The registry classifies this as a Phase 4, randomized, double-masked, parallel-group study. Participants were assigned to ambrisentan or matching placebo, with allocation described as 1:1. The listed enrollment is 4. The arm descriptions are blank, and the record does not report further details about concealment of the allocation sequence, masking procedures, or eligibility criteria. Those aspects cannot be inferred from the design labels alone.

The brief summary describes treatment over 16 weeks, with visits at least monthly. It outlines dose escalation from 2.5 mg daily to 5mg and then 10mg daily if tolerated. Adverse reactions were to be reviewed, with liver function testing and pregnancy testing if applicable. A six-minute walk test and quality-of-life survey were planned at baseline, week 4 and week 16. An echocardiogram, right heart catheterization and left ventricular end diastolic pressure measurement were described for the 16 week visit.

The registry lists a study start of 2009-01 and primary completion of 2013-09, and gives the status as terminated. The reason for termination, whether the intended follow-up was completed, and whether the full set of planned assessments was obtained are not reported in the registry record. The summary's expected completion statement is a projection, not evidence that the study finished as planned.

Although the design is randomized and masked, those features do not by themselves establish how the outcomes were analyzed or whether all participants were included in each analysis. The registry does not report the allocation method in operational detail, the analysis population, a statistical analysis plan, or a participant flow account. The results therefore need to be read as posted summaries rather than as a complete account of the conduct and analysis of the trial.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Change in Pulmonary Vascular Resistance (Wood Units), measured over Baseline and Four months.

The posted primary efficacy outcome is change in pulmonary vascular resistance (PVR), expressed in Wood units. The record describes PVR as calculated from mean pulmonary artery pressure minus wedge pressure, divided by cardiac output. In general, this measure combines pressure information from the pulmonary circulation with cardiac output into a hemodynamic index. It is used in this study as a physiological outcome relevant to pulmonary hypertension.

The stated measurement period is baseline and four months. The brief summary describes a 16-week visit for catheterization and other assessments, but the registry does not explain the relationship between that visit schedule and the four-month timeframe attached to the outcome. The listed primary outcome uses a mean summary and gives a full range for dispersion.

The registry contains an important wording distinction: the brief summary calls safety the primary endpoint, whereas the primary efficacy outcome entry identifies change in PVR. These statements may refer to different aspects of the study, but the record does not clarify their relationship. It is therefore safest to describe PVR as the posted primary efficacy outcome, while noting that the summary also names safety as primary. The reason for choosing PVR, and any prespecified threshold for a clinically meaningful change, are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Several statistical ideas help distinguish what the posted numbers describe from what they cannot establish. The registry gives arm-level summaries for PVR, rather than a formal comparative estimate. An arm-level mean is a descriptive summary of observations in one treatment group. A treatment comparison generally requires estimating how the groups differ, with an uncertainty measure that reflects both the observed data and the analysis method. The record does not report that comparative estimate or its uncertainty.

What does the mean and full range communicate?

The reported mean change describes the average change among the observations contributing to that arm's posted summary. The full range describes the observed minimum and maximum. These answer different questions: the mean gives a central summary, while the range gives the span of observed values. Neither one alone describes how precisely a population-level effect has been estimated. A full range is not a confidence interval and should not be read as one.

Here, the arm summaries are -0.75 Wood units with a range from -0.77 to -0.72 for ambrisentan, and 2.81 Wood units with a range from 2.81 to 2.81 for placebo. The associated counts are n 2 and n 1, respectively. A range based on very few observations can give an unstable picture of variability; in particular, the range for an arm represented by a single observation cannot show variation among multiple observations. This is a general statistical limitation of interpreting ranges from small samples, not evidence that the underlying population has no variability.

Why is a confidence interval useful?

In general, a confidence interval expresses the uncertainty around an estimated quantity under a specified statistical method and its assumptions. For a randomized comparison, an interval around a between-arm contrast would help readers assess which effect sizes remain compatible with the data and how imprecise the estimate is. The registry record does not report a confidence interval for a between-arm PVR contrast. The listed full ranges do not substitute for one, because they describe observed values within arms rather than uncertainty around a comparative estimate.

Why does the analysis population matter?

Trial analyses can use different participant sets, depending on their stated rules and the availability of outcome measurements. For example, a protocol may define an analysis population based on random assignment, treatment received, or outcome completion. Those choices can affect what the estimate represents. The registry record does not report the analysis population, how missing measurements were handled, or why the PVR summary counts differ from the listed enrollment. It is therefore not possible to determine whether the summaries represent all randomized participants or a selected subset.

What can randomization and masking contribute?

In general, random allocation is intended to reduce systematic differences between groups at the time of assignment, while masking is intended to limit effects of knowing the assigned treatment on care, measurement or assessment. The registry labels the design randomized and double-masked and describes a 1:1 allocation. These design features are relevant, but they do not guarantee balance in a very small trial, confirm that masking was maintained, or replace a reported analysis. The registry does not provide operational details that would allow those aspects to be evaluated.

Why is over-dispersion relevant?

For outcomes such as event counts, over-dispersion means that the observed variability is greater than a simple model assumes. In general, failing to allow for over-dispersion can make uncertainty estimates too small and tests too optimistic. That issue is most directly relevant to count outcomes or models for repeated events; it does not change the meaning of the PVR means and ranges. The registry does not report a model for adverse events or other outcomes, so there is no basis for saying whether an over-dispersion adjustment was considered or needed.

What do recalculated results mean?

Any recalculated result based only on posted summary statistics would be a simple, unadjusted calculation, not the sponsor's full analysis. In this record, no recalculated comparative estimate is included in the registry entries. The sponsor's own analysis takes precedence if a discrepancy is identified, but the discrepancy indicator is false. That does not establish that a sponsor analysis exists in the registry; a formal sponsor analysis is not reported in the registry record.

Can sample-size planning be inferred?

General sample-size illustrations use assumptions about the target effect, variability, allocation, significance level and desired power. The record describes the assumptions for planning illustrations based on a fraction of the observed effect, equal allocation, a two-sided significance level of 0.05 and a normal approximation. Such planning calculations are illustrations, not observed power and not the trial's original design justification. The registry does not report the trial's sample-size justification, and no planning sample-size values are given in the registry record. They cannot be used to infer what the investigators planned or whether the study had adequate power.

10. Multiplicity, Missing Data and Analysis Populations

The record lists three secondary outcomes: change in six-minute walk distance, change in functional class and change in Short Form-36 physical functioning. The brief summary also describes echocardiographic and catheterization assessments. Testing multiple outcomes can raise the chance that at least one result appears notable by chance when each is assessed separately. In general, multiplicity procedures address this issue by adjusting interpretation or analysis across a family of comparisons. The registry record does not report a multiplicity strategy, a hierarchy among outcomes, or which results were intended to be confirmatory.

The primary-outcome descriptions are also not fully aligned: the brief summary calls safety primary, while the outcome entry calls PVR change the primary efficacy outcome. The record does not clarify whether safety and PVR had distinct primary roles, or how they were prioritized in the analysis. No multiplicity adjustment or testing plan is reported.

Missing-data handling is not reported in the registry record. The PVR summaries include n 2 for ambrisentan and n 1 for placebo, while the enrollment entry lists 4. No reason is given for the outcome counts or for any unavailable measurements. The record does not describe imputation, sensitivity analyses, or rules for excluding observations. It is therefore not possible to assess whether missingness could have influenced the PVR summaries.

The analysis populations are also not reported. The registry does not state whether analyses followed randomized assignment, treatment received, outcome availability, or another definition. Without that information, readers cannot determine precisely which participants the estimates represent. These omissions affect interpretation of both efficacy and safety findings and prevent a full assessment of the statistical plan.

11. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: admission for heart failure (33.3%).

The safety counts are reported separately for serious adverse events, other adverse events and deaths. For serious adverse events, the registry lists 1 affected among 3 at risk in the ambrisentan group and 0 affected among 1 at risk in the placebo group. For other adverse events, it lists 3 affected among 3 at risk with ambrisentan and 1 affected among 1 at risk with placebo. Deaths are listed as 0 in each group. These are counts as recorded, not estimates adjusted for exposure time or differences in follow-up.

The listed serious-event term is “Admission for heart failure,” with 33.3% reported. The record does not provide further detail about event timing, recurrence, severity beyond the serious-event classification, or causal attribution to treatment. The percentage should be read as a registry-reported figure for that term, not as a comparative risk estimate.

With such small at-risk counts, each observation has substantial influence on the displayed proportions. The counts do not support a stable estimate of how often events would occur in a broader population, and they do not establish that an event was caused by either assigned treatment. The record does not report exposure-adjusted rates or a statistical comparison of adverse events. No interpretation of relative safety can be based on the counts alone.

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

The posted results establish that the registry contains arm-level PVR summaries and adverse-event counts for this terminated randomized study. The ambrisentan PVR summary is a mean change of -0.75 Wood units, with full range -0.77 to -0.72 and n 2; the placebo summary is 2.81 Wood units, with full range 2.81 to 2.81 and n 1. The record also lists serious and other adverse-event counts, as well as deaths.

These numbers do not provide a formal estimate of the difference between treatments. There is no reported test result, confidence interval for a comparative effect, or adjusted analysis. The descriptive direction of the arm summaries is not enough to establish that the groups differ reliably. The small counts and absent analysis details leave substantial uncertainty about what the results might imply beyond the observations recorded.

The record does not report results for its listed secondary outcomes, and it does not give the reason for termination or explain why the PVR counts are smaller than the enrollment entry. It also does not clarify whether the safety endpoint or PVR was the primary outcome for the analysis. These gaps mean the record cannot answer the study's broader questions about functional outcomes, imaging, hemodynamics beyond PVR, or the planned safety assessments.

In general, a randomized, masked design can strengthen a treatment comparison when it is implemented and analyzed as planned. But a design label is not a substitute for adequate outcome data, transparent analysis methods, and uncertainty estimates. The findings should be understood as limited registry summaries rather than as definitive evidence for or against ambrisentan in this population. They do not establish clinical benefit, lack of benefit, or comparative safety.

13. Why This Trial Matters Statistically

This trial is statistically notable because its design labels suggest a randomized, double-masked comparison, while the posted outcome summaries are based on very small arm counts. That contrast illustrates why study design and statistical information must be considered together. Randomization can support causal comparisons in principle, but a reader still needs the analysis population, the comparative estimate and its uncertainty to judge what the data say.

The record also illustrates how outcome definitions and priorities matter. Safety is identified as primary in the brief summary, while PVR change is named as the primary efficacy outcome. Distinguishing a safety objective from an efficacy outcome is useful, but the registry does not explain how both were treated in the analysis. The listed secondary outcomes add further questions about multiplicity, yet their results and analysis plan are not reported.

Finally, the available PVR results are arm-level mean and range summaries, not a formal treatment contrast. Readers may be tempted to compare the numerical values directly, but without an uncertainty estimate and analysis details, the apparent separation cannot be interpreted as a reliable treatment effect. The record's enrollment entry and smaller outcome counts also show why participant flow and missing-data explanations are important parts of a transparent report.

As a statistical example, the study underscores that concise registry results can be informative about what was measured and what was recorded, while still leaving major questions unanswered. Completeness of reporting is essential for evaluating precision, potential bias, safety, and the strength of any comparative claim.

14. 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 registry record gives limited information about eligibility, baseline characteristics, participant flow and the reasons for termination. The reason the study ended is not reported in the registry record, nor is it stated whether termination occurred before all planned assessments were completed. The listed enrollment and the smaller PVR counts are not reconciled.

The primary outcome description is internally ambiguous: the brief summary calls safety the primary endpoint, while the primary efficacy outcome entry identifies change in PVR. The record does not explain the relationship between those descriptions. It also does not report the statistical analysis plan, the intended analysis population, the handling of missing data, or any method for addressing multiplicity.

The PVR summaries provide an arm mean and full range but no standard error, confidence interval for a between-arm contrast, or hypothesis-test result. The counts contributing to those summaries are very small. The record does not provide the corresponding results for the listed secondary outcomes, despite describing planned assessments of walking distance, functional class, quality of life, echocardiography and catheterization.

Safety information is also limited. The registry lists counts and a serious-event term, but does not report event timing, exposure-adjusted rates, causal assessment, or a statistical safety comparison. The small at-risk counts make the posted figures highly sensitive to individual observations. Taken together, these limitations restrict interpretation to a description of the registry entries; they prevent a precise assessment of comparative efficacy or safety.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Safety and Efficacy Trial to Treat Diastolic Heart Failure Using Ambrisentan — NCT00840463.
• Registry record last updated 2020-05-19; status last verified 2020-05.
• Results first posted 2019-05-22.
• 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.

18. Record Summary

This Phase 4 study is listed as randomized, double-masked and parallel-group, comparing ambrisentan with placebo in pulmonary hypertension associated with heart failure with preserved ejection fraction. The registry lists enrollment as 4, a study start of 2009-01, primary completion of 2013-09, and status as terminated.

The posted primary efficacy outcome is change in PVR in Wood units from baseline to four months. Arm summaries are -0.75 with full range -0.77 to -0.72 and n 2 for ambrisentan, and 2.81 with full range 2.81 to 2.81 and n 1 for placebo. No formal between-arm analysis is reported. The record also lists adverse-event counts and deaths, but does not report results for the listed secondary outcomes.

The registry does not report the reason for termination, missing-data methods, analysis populations, multiplicity procedures, or a statistical analysis plan. It gives inconsistent descriptions of whether safety or PVR is primary. These limitations leave the posted summaries descriptive and imprecise.