This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Impact of EMpagliflozin on Cardiac Function and Biomarkers of Heart Failure in Patients With Acute MYocardial Infarction (NCT03087773), a randomized phase 3 study. The primary outcome is Changes of Nt-proBNP (N-terminales Pro Brain Natriuretic Peptide) Levels.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03087773 |
| Phase | Phase 3 |
| Condition | Acute Myocardial Infarction |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Enrollment | 476 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Medical University of Graz |
| Results posted | Yes |
2. Clinical Question
Population
People with acute myocardial infarction, including participants with and without type 2 diabetes mellitus. The study summary describes investigation of heart-failure biomarkers within 6 months after the event. Further eligibility details are not reported in the registry record.
Intervention
Empagliflozin 10 mg, described in the registry record as the study drug given to participants assigned to that group.
Comparator
A placebo oral tablet. The record identifies a randomized, parallel-group comparison and describes quadruple masking, but does not specify which roles were masked.
Primary question
How did NT-proBNP levels change from randomization to week 26 in the empagliflozin and placebo groups, and what evidence was reported for a difference between those changes?
The registry describes the primary outcome as the difference in change in NT-proBNP between treatment groups from randomization to week 26. Its posted results give a median change and interquartile range for each group. They do not include a reported between-group estimate or an inferential analysis, so the posted arm summaries alone do not answer the comparative question with a formal estimate of uncertainty.
The study also lists changes in ejection fraction, left ventricular end-diastolic volume, E/è ratio and left ventricular end-systolic volume, as well as duration of hospital stay, among its secondary outcomes. Their presence indicates that the trial considered outcomes beyond the primary biomarker, but the posted results included for this record do not give results for those secondary measures. Detailed definitions of the population, including eligibility requirements and baseline clinical characteristics, are not reported in the registry record.
3. Trial Design
This was a phase 3, randomized, quadruple-masked, parallel-group trial sponsored by the Medical University of Graz. The registry lists 476 participants as enrolled. One group received empagliflozin 10 mg and the other received a placebo oral tablet. In a parallel-group design, participants are assigned to a group and outcomes are compared across groups; the record does not describe further details of the allocation process or masking procedures.
The registry gives a study start date of 2017-05-11, a primary completion date of 2022-05-03 and a status of completed. These dates describe the recorded study timeline, but they do not establish when each participant was enrolled, how long recruitment took, or whether follow-up was complete for every enrolled participant. Such details are not reported in the registry record.
The primary outcome is the change in NT-proBNP from randomization to week 26. The record describes the measure as a median with interquartile range and gives results for the empagliflozin and placebo groups. The outcome counts shown with those results are 217.0 and 227.0, respectively. These group counts do not sum to the overall enrollment, and the reasons for that difference are not reported in the registry record.
Safety results use at-risk counts of 237 and 239 for the respective groups, which differ from the outcome counts. The record does not explain the relationship between these populations or give a participant flow account. Five secondary outcomes are posted, while the record identifies six outcome measures overall. Further outcome results or analyses are not reported in the registry record.
The trial design permits a randomized comparison, but design features alone do not establish the magnitude or precision of an effect. The posted outcome summaries are descriptive by arm. No formal between-group analysis for the primary outcome is reported in the registry record.
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
| Stage | Empagliflozin | Placebo Oral Tablet |
|---|---|---|
| STARTED | 237 | 239 |
| COMPLETED | 217 | 227 |
| NOT COMPLETED | 20 | 12 |
Reasons for not completing this period:
| Reason | Empagliflozin | Placebo Oral Tablet |
|---|---|---|
| Withdrawal by Subject | 8 | 4 |
| Lost to Follow-up | 12 | 8 |
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.
| Characteristic | Unit | Empagliflozin | Placebo Oral Tablet | Total |
|---|---|---|---|---|
| Participants analysed | 237 | 239 | 476 | |
| Age, Continuous | years | 57 (52 to 64) | 57 (52 to 65) | 57 (52 to 64) |
| Sex: Female, Male: Female | Participants | 42 | 42 | 84 |
| Sex: Female, Male: Male | Participants | 195 | 197 | 392 |
| Race and Ethnicity Not Collected | Participants | — | — | 0 |
| Region of Enrollment: Austria | Participants | 237 | 239 | 476 |
| Body Mass Index | kg/m² | 27.7 (25.3 to 30.3) | 27.2 (24.9 to 30.2) | 27.6 (25.1 to 30.3) |
| Systolic blood pressure | mmHg | 125 (116 to 131) | 125 (118 to 131) | 125 (117 to 131) |
| Diastolic blood pressure | mmHg | 78 (74 to 85) | 78 (75 to 85) | 78 (74 to 85) |
| Obesity | Participants | 68 | 70 | 138 |
| Type 2 diabetes | Participants | 30 | 33 | 63 |
| Hypertension | Participants | 92 | 107 | 199 |
| Dyslipidaemia | Participants | 71 | 64 | 135 |
| Smoking active or former | Participants | 171 | 170 | 341 |
| Coronary artery disease | Participants | 28 | 25 | 53 |
| History of stroke | Participants | 5 | 1 | 6 |
6. Primary Endpoint
The registry lists the primary outcome as Changes of Nt-proBNP (N-terminales Pro Brain Natriuretic Peptide) Levels, measured over 26 weeks.
| Element | Registry entry |
|---|---|
| Result type | Median |
| Measure of spread | Inter-Quartile Range |
| Unit | pg/mL |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint concerns the change in N-terminal pro-brain natriuretic peptide, abbreviated NT-proBNP, from randomization to week 26. The registry gives the measurement unit as pg/mL and describes the intended comparison as the difference in change between the empagliflozin and placebo groups. It reports the group results as medians with interquartile ranges.
As a general statistical concept, a change endpoint compares a participant’s value at follow-up with that participant’s value at the starting point. Comparing changes between randomized groups can describe whether their observed trajectories differ over the specified period. The record gives group-level summaries, but it does not report a direct estimate of the difference between groups or a confidence interval for that contrast.
The registry identifies NT-proBNP as a biomarker of heart failure and makes it the primary outcome. A biomarker outcome can provide a measurable study endpoint over a defined follow-up interval. It is not, by itself, the same as a clinical outcome such as hospitalization, nor does a change in a biomarker alone establish a change in patients’ health. The registry record does not report a threshold for interpreting individual values or a rationale for selecting this endpoint.
For empagliflozin, the median change was -84.9 pg/mL, with reported interquartile range endpoints of -91.7 and -72.6. For placebo, the median change was -82.2 pg/mL, with reported endpoints of -89.2 and -69.8. These are separate group summaries. They should not be read as a formal estimate of the between-group difference.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
For the primary outcome, the registry reports a median change in NT-proBNP of -84.9 pg/mL in the empagliflozin group, with interquartile range endpoints of -91.7 and -72.6. In the placebo group, the median change was -82.2 pg/mL, with endpoints of -89.2 and -69.8. The two summaries describe the observed distributions within the groups from randomization to week 26.
The registry’s stated primary question is comparative: it concerns the difference in change between groups. The posted results do not give that contrast as a numerical estimate, do not provide a confidence interval for it, and do not report a p-value or other inferential test. It is therefore not possible to use this record’s posted summaries to quantify the precision of the group difference or assess it using a reported significance test. The arm medians are not substitutes for an adjusted or unadjusted between-group analysis.
As a general statistical concept, a confidence interval describes uncertainty around an estimated quantity, such as a group contrast. The interquartile range instead describes the middle portion of the observations within a group. It is useful for showing the spread of individual outcomes but does not, on its own, quantify uncertainty in the difference between groups. The endpoints reported for each group are interquartile range endpoints, not a confidence interval for the treatment comparison.
The outcome counts reported are 217.0 for empagliflozin and 227.0 for placebo. The registry lists 476 enrolled participants overall, but does not explain why the primary-outcome counts differ from the enrollment total or from each other. The safety at-risk counts are 237 and 239. Because the record does not explain the analysis populations, these counts should be treated as outcome-specific counts rather than assumed to represent the same participants used for every analysis.
Five secondary outcomes are listed as posted, including cardiac-function measures and duration of hospital stay. Results for those outcomes are not reported in the registry record. Although six outcome measures are identified overall, the record reports no analyses. Consequently, the available posted results support a description of the primary outcome by group and the safety counts, but not a broader conclusion about comparative effects across the listed outcomes.
The registry describes randomization and masking, which are design features intended to support a fair comparison and reduce the influence of knowledge of assignment. Their presence does not fill the gap in reported comparative analysis. The registry record also does not report covariate adjustment, handling of missing outcome data, subgroup findings, or a statistical analysis plan. No causal or clinical interpretation beyond the posted descriptive results can be established from these summaries alone.
9. Statistical Methods Explained
Randomization assigns participants to study groups by chance. As a general statistical principle, this can reduce systematic differences between groups at the start of a trial, although the registry record does not provide baseline characteristics or details of how assignment was generated. It identifies the study as randomized and parallel-group, with empagliflozin compared against placebo. It also describes masking as quadruple, but does not state which participant, clinician, investigator, or outcome-assessment roles were masked.
What does a median summarize?
The primary outcome is reported as a median. In general, the median is the central value after observations are ordered; it is less influenced by unusually high or low observations than the arithmetic mean. This can be a useful way to summarize a distribution that is skewed or contains extreme values. The registry does not state that skewness or outliers motivated the choice of summary, so that rationale should not be assumed.
The empagliflozin median change is -84.9 pg/mL, and the placebo median change is -82.2 pg/mL. Each is a within-group summary of change from randomization to week 26. Neither value directly estimates the difference between groups. A comparative analysis would need to estimate the contrast between groups using an explicitly described method; no such method or estimate is reported in the registry record.
What does the interquartile range describe?
In general, the interquartile range describes the spread from the lower to the upper quartile, covering the middle half of an ordered distribution. It is a descriptive measure of variability among observations. The registry labels the primary outcome dispersion as an interquartile range and reports endpoints of -91.7 and -72.6 for empagliflozin, and -89.2 and -69.8 for placebo. Those values describe within-group distributions. They are not bounds on the uncertainty of a treatment contrast.
Different statistical questions require different uncertainty summaries. A confidence interval for the group contrast would express uncertainty about that estimated comparison. A hypothesis test would assess the data against a specified null hypothesis using a stated method. Neither a confidence interval nor a test result for the primary comparison is reported in the registry record. No inferential procedure should be inferred from the descriptive medians and interquartile ranges.
Why does the analysis population matter?
The registry lists 476 enrolled participants, while the primary results have outcome counts of 217.0 and 227.0. The serious adverse event summaries use at-risk counts of 237 and 239. Counts can differ across outcomes when the populations or available records differ, but the record does not give reasons for these particular counts. It does not report whether the primary analysis included all randomized participants, only participants with observed week 26 measurements, or another defined population.
In general, specifying an analysis population helps readers understand which participants contribute to an estimate and what happens when outcome data are unavailable. Without that information, the reported summaries cannot be confidently characterized as intention-to-treat, per-protocol, or another analysis population. The registry record does not describe imputation, exclusion rules, or other missing-data methods.
What can be inferred from the planning assumptions?
The record describes a planning illustration using a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects defined as fractions of the observed effect. These are assumptions for a sample-size illustration, not evidence that the trial was designed or analyzed using those exact methods. The record does not report the trial’s own sample-size justification, its target effect, or a protocol-based sample-size calculation.
As a general concept, a sample-size calculation estimates how many participants may be needed to detect a specified effect under stated assumptions. Such a calculation is distinct from observed power, which should not be used as a retrospective summary of whether a result was significant. Any sample sizes from an illustration based on observed results would be planning aids, not counts of participants observed in this trial. Numerical planning counts are not reported in the registry record.
The record does not specify the statistical model, whether the primary comparison was adjusted, how repeated or baseline measurements were handled, or whether robust methods were used. Those omissions matter because the analysis method defines the estimand, the assumptions used, and the uncertainty calculation. The registry’s outcome description identifies the measurement and time point, but it is not a complete statistical analysis plan.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists five posted secondary outcomes and six outcome measures overall. The primary outcome is the change in NT-proBNP from randomization to week 26. The record does not report a hierarchy of outcomes, a multiplicity adjustment, or a plan for interpreting several statistical tests together. It also reports no analyses for the listed outcomes. Without those details, the registry does not establish how any multiplicity from testing several outcomes was addressed.
As a general statistical concept, multiplicity arises when several hypotheses or outcomes are assessed. If each is interpreted as a separate opportunity for a positive finding, the chance of at least one false-positive conclusion can increase. Prespecified testing sequences or adjustment methods can address this issue, but neither is described in the registry record. This is especially relevant when considering secondary outcomes, which are not reported with comparative estimates or tests.
Missing-data handling is not reported in the registry record. The primary outcome counts are 217.0 and 227.0, compared with 476 enrolled overall; safety at-risk counts are 237 and 239. The record does not explain these differences, state how many measurements were unavailable, or describe whether missing values were imputed, excluded, or handled by another approach. The analysis population for the primary outcome is also not specified.
Baseline values, participant flow, exclusions after randomization, and sensitivity analyses are not reported in the registry record. Without them, readers cannot assess whether the posted summaries represent all randomized participants or how sensitive findings might be to missing outcomes or analysis choices.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Empagliflozin | 31 | 237 | 13.1% |
| Placebo Oral Tablet | 32 | 239 | 13.4% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Empagliflozin | 3 of 237 |
| Placebo Oral Tablet | 0 of 239 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 3% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Empagliflozin | 18 of 237 |
| Placebo Oral Tablet | 9 of 239 |
Most frequent serious adverse event terms
| Event term | Organ system | Empagliflozin | Placebo Oral Tablet |
|---|---|---|---|
| Any hospitalisation | Cardiac disorders | 31 of 237 | 32 of 239 |
| Hospitalisation due to heart failure | Cardiac disorders | 3 of 237 | 4 of 239 |
| Hospitalisation due to cardiovascular event | Cardiac disorders | 2 of 237 | 5 of 239 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Empagliflozin | Placebo Oral Tablet |
|---|---|---|---|
| Urinary tract infection | Infections and infestations | 11 of 237 | 7 of 239 |
| Genital fungal infection | Infections and infestations | 7 of 237 | 2 of 239 |
Serious events with the highest arm-level rate: any hospitalisation (13.4%), hospitalisation due to cardiovascular event (2.1%), hospitalisation due to heart failure (1.7%).
The registry reports serious adverse event counts for the empagliflozin and placebo groups using at-risk counts of 237 and 239, respectively. Serious adverse events affected 31 participants in the empagliflozin group and 32 in the placebo group. The registry also reports 3 deaths in the empagliflozin group and 0 in the placebo group. These are counts; the record does not provide an analysis comparing the groups or a confidence interval for the difference.
Other adverse events affected 18 participants in the empagliflozin group and 9 in the placebo group. The record does not provide further detail about the definition or severity represented by this category. The most frequently listed serious terms are Any hospitalisation at 13.4%, Hospitalisation due to cardiovascular event at 2.1%, and Hospitalisation due to heart failure at 1.7%. The registry does not identify these term percentages by treatment arm in the posted information, so they should not be attributed to either group.
As a general statistical principle, adverse-event counts need context, including the number of participants at risk, the observation period, event definitions, and whether one participant could contribute more than one event. The registry supplies at-risk counts but does not report the event-collection window, detailed definitions, or a comparative safety analysis. The counts describe reported events; by themselves, they do not establish whether a difference between groups is meaningful or attributable to treatment.
12. What the Result Does, and Does Not, Mean
The posted primary results show group-level median changes in NT-proBNP from randomization to week 26: -84.9 pg/mL for empagliflozin and -82.2 pg/mL for placebo. The registry also reports interquartile range endpoints for each group. These summaries describe observed outcomes within each arm; they are not a direct estimate of the difference between arms.
The registry’s primary question concerns the difference in change between treatment groups. A between-group estimate, its uncertainty, and an inferential test are not reported in the registry record. Therefore, the posted summaries do not establish the size or precision of a comparative effect. They also do not support a conclusion that one treatment is superior, equivalent, or different from the other on the primary outcome.
The endpoint is a biomarker, and the registry also lists cardiac imaging measures and hospital stay as secondary outcomes. Results for those secondary outcomes are not reported in the registry record. The listed safety counts are descriptive and are not accompanied by a comparative analysis. This means the posted record gives some outcome and safety information, but does not provide a full analysis of clinical outcomes or a comprehensive benefit-risk assessment.
As a general statistical concept, a result’s interpretation depends on the estimand, the analysis population, the precision of the estimate, and the assumptions behind the method. The registry record does not report the primary analysis method, missing-data approach, or analysis population. It also does not explain the difference between enrollment and the outcome-specific counts. Those gaps limit what can be concluded from the posted summaries.
The study is described as randomized, quadruple-masked, and parallel-group, but these design features do not replace a reported analysis. They describe how the comparison was structured, while the posted results determine what can be said about its measured outcomes. The available information should therefore be read as descriptive arm-level reporting rather than as proof of a comparative treatment effect.
13. Why This Trial Matters Statistically
This trial is statistically notable because its registry description sets out a randomized comparison and a defined primary biomarker outcome, while its posted primary results are descriptive summaries by group. That distinction is important: reporting an outcome in each arm is not the same as reporting the randomized comparison that the primary question calls for. A reader needs the between-group estimate and its uncertainty to understand the size and precision of the contrast.
The primary outcome is summarized using medians and interquartile ranges. These measures describe central tendency and spread within each group. They do not provide uncertainty around the group contrast. Recognizing the difference prevents misreading the interquartile range endpoints as confidence limits for a comparative estimate.
There is also a difference between overall enrollment, primary-outcome counts, and safety at-risk counts. The record gives 476 enrolled participants, outcome counts of 217.0 and 227.0, and safety at-risk counts of 237 and 239. Without an account of participant flow and analysis populations, it is not possible to determine from the registry record why the counts differ or whether missing measurements influenced the reported summaries.
Finally, the registry lists secondary outcomes but does not report their results or analyses. It does not describe multiplicity handling, missing-data methods, or the statistical analysis plan. These are not merely technical details: they shape how estimates should be interpreted, how several outcomes should be considered together, and how robust the findings may be to incomplete data. The record supports careful description of what was posted, while leaving these analytical questions unresolved.
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 does not report a formal estimate of the primary between-group contrast, its confidence interval, or a p-value. Although it defines the primary outcome as a difference in change between groups, the posted values are separate arm medians. The descriptive summaries cannot establish the precision or statistical significance of the treatment comparison.
The record does not explain why the primary outcome counts are 217.0 and 227.0 when total enrollment is 476, or why safety at-risk counts are 237 and 239. A participant-flow account, reasons for missing outcome data, and the analysis population are not reported in the registry record. Missing-data methods and sensitivity analyses are also not reported.
Five secondary outcomes are posted, but their results are not reported in the registry record. The record does not explain how multiple outcomes were prioritized or whether any adjustment for multiplicity was planned. Although it includes sample-size planning assumptions, the trial’s own sample-size justification and target effect are not reported.
The record does not report baseline characteristics, detailed eligibility criteria, allocation-generation procedures, or which roles were masked. It also does not describe the primary statistical model, covariate adjustment, or a complete analysis plan. These omissions limit assessment of the comparability of the groups, the handling of incomplete data, and the appropriateness of the analysis choices.
The safety information is also limited to counts and selected serious-event terms. The record does not report an arm-specific breakdown for the listed term percentages, a comparative safety analysis, or further event definitions and follow-up details. These limitations constrain interpretation of both the efficacy summaries and safety counts.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Impact of EMpagliflozin on Cardiac Function and Biomarkers of Heart Failure in Patients With Acute MYocardial Infarction — NCT03087773.
- Registry record last updated 2024-08-22; status last verified 2024-08.
- Results first posted 2024-08-22.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The registry describes a completed, phase 3, randomized, quadruple-masked, parallel-group trial sponsored by the Medical University of Graz, with 476 participants enrolled. Empagliflozin 10 mg was compared with a placebo oral tablet in people with acute myocardial infarction, with or without type 2 diabetes mellitus.
The primary outcome was change in NT-proBNP from randomization to week 26. Median changes were -84.9 pg/mL for empagliflozin and -82.2 pg/mL for placebo, with interquartile range endpoints reported for each group. A between-group estimate and inferential analysis are not reported in the registry record. Five secondary outcomes are posted, but their results are not reported. Serious adverse events affected 31 participants in the empagliflozin group and 32 in the placebo group; the registry also reports 3 and 0 deaths, respectively. Analysis populations, missing-data methods, and the trial’s statistical analysis plan are not reported in the registry record.