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 STeroids to REduce Systemic Inflammation After Infant Heart Surgery (NCT03229538), a randomized phase 3 study. The primary outcome is Number of Participants at Each Global Rank Endpoint Based Upon Their Most-severe Outcome.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03229538 |
| Phase | Phase 3 |
| Condition | Congenital Heart Disease in Children, Inflammatory Response |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Quadruple |
| Enrollment | 1263 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Kevin Hill |
| Results posted | Yes |
2. Clinical Question
Population
Infants undergoing heart surgery with cardiopulmonary bypass. The registry record identifies congenital heart disease in children and inflammatory response as the conditions studied. It does not report further eligibility details or a more specific description of the enrolled population.
Intervention
Intravenous methylprednisolone. The registry summary says study drug was planned for administration 8 to 12 hours before the anticipated start of surgery and again in the operating room at initiation of cardiopulmonary bypass.
Comparator
Intravenous isotonic saline as placebo. The registry summary describes the study as placebo-controlled and states that participants were randomized in a 1:1 fashion to methylprednisolone or placebo.
Primary question
How did participants in the two randomized groups fare on a composite global-rank endpoint based on each participant’s most severe outcome? The endpoint combines postoperative length of stay with mortality and major morbidity categories, ordered by severity. The registry record does not report the number of participants in each rank category or a single treatment comparison for this endpoint, so the primary question cannot be answered from the recorded results alone.
3. Trial Design
The registry describes STRESS as a phase 3, randomized, parallel-group trial, with quadruple masking recorded. Its brief summary calls the study prospective, double blind, multicenter, and placebo-controlled. The recorded enrollment is 1263. That enrollment figure should not be interpreted as the number of participants included in every analysis: group-specific analysis populations and exclusions are not reported in the registry record.
Participants were assigned in a 1:1 fashion to intravenous methylprednisolone or intravenous isotonic saline placebo. The planned dosing times were 8 to 12 hours before the anticipated start of surgery and at initiation of cardiopulmonary bypass in the operating room. The registry record describes follow-up for primary and secondary outcomes through the hospitalization. It also states that serious study drug-related adverse events were to be collected for 7 days after the last dose.
The study’s stated objective was to evaluate pharmacokinetics, pharmacodynamics, safety, and efficacy of methylprednisolone in infants undergoing heart surgery with cardiopulmonary bypass. Blood sampling for pharmacokinetic and pharmacodynamic evaluation was planned in a subset of enrolled participants. The registry lists outcomes that include mortality, major complications, prolonged postoperative stay, prolonged mechanical ventilation, low cardiac output syndrome, infectious complications, other postoperative complications, and time to maximum concentration.
The registry status is completed. The start date is 2017-10-18 and the primary completion date is 2022-03-31. These design and timing details describe the registry entry; they do not establish which participants contributed data to each endpoint or analysis. Those analysis-population details are not 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.
Before assignment: 1263 patients consented and randomized; 63 did not receive study drug and are excluded from final analysis.
Overall Study
| Stage | Methylprednisolone Arm | Placebo Arm |
|---|---|---|
| STARTED | 599 | 601 |
| COMPLETED | 599 | 601 |
| NOT COMPLETED | 0 | 0 |
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 | Methylprednisolone Arm | Placebo Arm | Total |
|---|---|---|---|---|
| Participants analysed | 599 | 601 | 1200 | |
| Age, Continuous | days | 126 (14 to 191) | 124 (14 to 182) | 125 (14 to 187) |
| Sex: Female, Male: Female | Participants | 279 | 267 | 546 |
| Sex: Female, Male: Male | Participants | 320 | 334 | 654 |
| Ethnicity (NIH/OMB): Hispanic or Latino | Participants | 80 | 63 | 143 |
| Ethnicity (NIH/OMB): Not Hispanic or Latino | Participants | 519 | 538 | 1057 |
| Ethnicity (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 5 | 4 | 9 |
| Race (NIH/OMB): Asian | Participants | 15 | 12 | 27 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 4 | 0 | 4 |
| Race (NIH/OMB): Black or African American | Participants | 90 | 102 | 192 |
| Race (NIH/OMB): White | Participants | 428 | 425 | 853 |
| Race (NIH/OMB): More than one race | Participants | 13 | 15 | 28 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 30 | 25 | 55 |
| Region of Enrollment: United States | Participants | 599 | 601 | 1200 |
6. Primary Endpoint
The registry lists the primary outcome as Number of Participants at Each Global Rank Endpoint Based Upon Their Most-severe Outcome, measured over Until hospital discharge, up to 4 months.
| Element | Registry entry |
|---|---|
| Result type | Count Of Participants |
| Measure of spread | not applicable |
| Unit | Participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is a composite global-rank measure. Each randomized participant is assigned a rank based on their most severe outcome, with higher ranks representing more severe outcomes in the ordering described by the registry. Ranks 1 through 90 correspond to postoperative length of stay in days. Rank 91 represents postoperative length of stay greater than 90 days. Higher ranks encode specified complications, transplant, and operative mortality, with rank 97 assigned to operative mortality.
This construction combines outcomes that differ in clinical severity and type. A participant’s rank reflects the most severe qualifying outcome rather than a separate score for every event they experienced. A ranked composite can retain information about both severe events and length of stay, rather than reducing every outcome to a simple event/no-event indicator. It also imposes an ordering: interpretation depends on the endpoint’s stated hierarchy.
The registry labels the primary measure as the number of participants at each global rank endpoint. However, the category-specific counts and a summary comparison between randomized groups are not reported in the registry record. The timeframe is until hospital discharge, up to 4 months. Without the distribution across ranks, readers cannot determine from this record how the groups compared on the primary endpoint, or how much the result reflects length of stay versus more severe categories.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a clinically broad primary endpoint, but does not give the number of participants in each rank category or a single summarized comparison between groups. It therefore does not establish whether the randomized groups differed on the primary endpoint, nor the size or precision of any such difference. The record explicitly indicates that the primary outcome has several categories or time points and is not reduced to one comparison. That limitation matters: enrollment and trial design alone cannot substitute for the outcome distribution.
The endpoint’s hierarchy combines postoperative length of stay and outcomes such as cardiac arrest, organ failure, reoperation, mechanical circulatory support, transplant, and operative mortality. A comparison of this endpoint would need to be interpreted in light of how participants were distributed across that hierarchy. A summary statistic, if available, would not by itself show which categories contributed most to the comparison. The registry record does not report a category-level account that would allow that assessment.
The safety figures are descriptive counts. The registry lists 2 participants with serious adverse events in the methylprednisolone arm and 0 in the placebo arm. It lists 155 and 176 participants, respectively, with other adverse events, and 12 and 17 deaths, respectively. These counts should not be converted into a conclusion about comparative risk without the relevant analysis, uncertainty estimates, definitions, and analysis populations. In particular, the record does not report whether the listed adverse events were considered related to study treatment.
Even when two observed counts differ, random variation and the amount of information behind the counts affect interpretation. Confidence intervals can express the range of effects compatible with the data under a stated analysis, while a p-value addresses compatibility with a specified null model. Neither is reported for the primary endpoint in the registry information. No estimate of the primary treatment contrast, confidence interval, or p-value is available in this record.
Accordingly, the defensible reading is limited to the design, endpoint definition, and recorded counts. The trial was randomized and placebo-controlled, but the missing primary outcome distribution prevents a reader from judging the primary comparison from this registry record. The adverse-event figures are useful for describing what was recorded, not for inferring a treatment effect on the primary endpoint or establishing causation.
9. Statistical Methods Explained
The statistical methods that were actually used for the primary endpoint are not reported in the registry record. The endpoint definition does, however, provide a basis for understanding what an analysis would need to represent and what its results would mean. The following explanations are general statistical context, not claims about the trial’s analysis plan.
What does a global rank represent?
A global rank places outcomes on an ordered scale. Here, ranks 1 through 90 correspond to postoperative length of stay in days, while ranks 91 through 97 represent increasingly severe outcome categories as described in the registry. The participant is ranked according to their most severe outcome. This means the endpoint does not simply count whether any one complication occurred, and it does not report each participant’s full history of events.
When outcomes are ranked, an analysis can compare the distribution of outcomes across groups while respecting the stated ordering. Different statistical approaches can do this, and they may answer somewhat different questions. For example, an analysis might compare overall rank distributions or estimate the probability that a randomly selected participant from one group has a more favorable rank than a participant from the other group. These are general examples only; the registry record does not state which approach was used, and no such estimate is reported.
Why does the hierarchy matter?
The rank scheme assigns meaning to the ordering of outcomes. A longer postoperative stay receives a higher rank within the length-of-stay portion, and specified severe complications occupy higher categories. Any interpretation therefore depends on the hierarchy being clinically meaningful and applied consistently. A single overall summary may conceal whether a difference arises mainly from length of stay or from severe events. Category-specific counts and the analysis method would help readers understand that composition, but they are not reported for the primary endpoint in the registry record.
Composite endpoints can improve efficiency by combining several outcomes, but they also require careful interpretation. Components may differ in frequency, severity, and susceptibility to change. A composite result does not automatically show that every component moved in the same direction. In this trial, the registry supplies the ranking definitions but not the category-level primary results needed to inspect the contributions of the components.
What would an estimate and confidence interval add?
An effect estimate summarizes the comparison defined by an analysis. A confidence interval describes the uncertainty around that estimate under the chosen statistical model and repeated-sampling framework. A narrow interval generally indicates greater precision than a wide interval, though precision is not the same as clinical importance. For the primary endpoint, the registry record does not report an effect estimate or confidence interval. It also does not report a p-value or the statistical test used.
General statistical practice is to interpret an estimate together with its uncertainty, endpoint definition, and analysis population, rather than treating a thresholded p-value as the whole result. The necessary primary estimate and uncertainty are unavailable in this registry record, so that fuller interpretation cannot be made.
How do counts and distributional assumptions differ?
The primary measure is described as a number of participants at each rank. Counts by category describe how many participants fall into each outcome level; they are not, by themselves, an analysis of the overall rank distribution. If counts are modeled, the chosen method should reflect the data structure and assumptions. In general, over-dispersion occurs when count variability is greater than a model expects. Ignoring over-dispersion can make standard errors too small and confidence intervals too narrow. This is general context, not a statement that the trial used a count model or experienced over-dispersion.
Because the endpoint is ordered and includes a length-of-stay component, an analysis could also depend on how ties, censoring, and participants without a recorded outcome were handled. The registry record does not describe those methods. It does not report whether any adjustment for baseline characteristics was made, what covariates might have been considered, or how missing outcome information was handled. These details are important for understanding what a treatment estimate targets and how robust it may be.
What can be inferred from the design?
Random allocation supports a comparison of groups formed by the trial assignment, and masking can reduce some sources of bias. Those design features do not supply the missing endpoint results or resolve questions about analysis choices. The record identifies randomization, quadruple masking, and a parallel design, but does not explain the masking roles or report a detailed statistical analysis plan. The sponsor’s analysis, if available in a full report, would take precedence over any simple recalculation from summary data. No recalculated primary comparison is presented in this registry record.
The sample-size assumptions described for planning illustrations are not the trial’s own design justification. Such illustrations use an assumed effect based on an observed effect and specified planning conditions; they are not evidence of observed power. The registry record does not report the trial’s sample-size justification, and no planning illustration should be read as the power achieved by the completed trial.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary global-rank endpoint and multiple secondary outcomes, including mortality, death or major complication, prolonged postoperative stay, prolonged mechanical ventilation, low cardiac output syndrome, infectious complications, other postoperative complications, and a pharmacokinetic measure. Examining multiple outcomes can increase the chance that at least one result appears notable by chance if each is tested independently. Statistical plans may address this through a prespecified hierarchy, adjustment, or a clearly exploratory interpretation. The registry record does not report whether multiplicity was controlled, how outcomes were prioritized analytically, or whether any secondary analyses were designated exploratory.
The primary endpoint itself has multiple rank categories, but a count for each category is not equivalent to a set of independent primary treatment tests. The registry record does not provide a single comparison or the method used to summarize the endpoint. It also does not report whether the analysis plan included subgroup analyses, interim analyses, or stopping rules. Those details are not reported in the registry record and should not be inferred from the trial’s phase or completion status.
Missing-data procedures and analysis populations are likewise not reported in the registry record. The enrollment entry gives 1263, but the record does not describe the number included in the primary analysis, how exclusions were handled, or whether the analysis followed randomized assignment regardless of treatment received. It also does not explain how missing outcome values, withdrawals, or incomplete follow-up were managed. These choices can affect estimates and their interpretation.
General statistical practice distinguishes a planned analysis population from the people who contribute data to a particular outcome. Readers need both the group definitions and the handling of missing information to assess how closely the reported comparison reflects the original randomization. Without those details and the primary outcome results, the registry entry is insufficient for a complete assessment of multiplicity or missing-data risk.
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 |
|---|---|---|---|
| Methylprednisolone Arm | 2 | 599 | 0.3% |
| Placebo Arm | 0 | 601 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Methylprednisolone Arm | 12 of 599 |
| Placebo Arm | 17 of 601 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Methylprednisolone Arm | 155 of 599 |
| Placebo Arm | 176 of 601 |
Most frequent serious adverse event terms
| Event term | Organ system | Methylprednisolone Arm | Placebo Arm |
|---|---|---|---|
| Cardiac dysfunction | Cardiac disorders | 1 of 599 | 0 of 601 |
| Hypotension | Vascular disorders | 1 of 599 | 0 of 601 |
| Enterovirus infection | Infections and infestations | 1 of 599 | 0 of 601 |
| Rhinovirus infection | Infections and infestations | 1 of 599 | 0 of 601 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Methylprednisolone Arm | Placebo Arm |
|---|---|---|---|
| Arrhythmia | Cardiac disorders | 155 of 599 | 176 of 601 |
| Cardiac dysfunction | Cardiac disorders | 133 of 599 | 135 of 601 |
| Postoperative thoracic procedure complication | Injury, poisoning and procedural complications | 125 of 599 | 135 of 601 |
| Respiratory failure | Respiratory, thoracic and mediastinal disorders | 74 of 599 | 83 of 601 |
| Procedural complication | Injury, poisoning and procedural complications | 41 of 599 | 42 of 601 |
| Chylothorax | Respiratory, thoracic and mediastinal disorders | 36 of 599 | 37 of 601 |
| Cardiac Arrest | Cardiac disorders | 31 of 599 | 37 of 601 |
| Cardiac operation | Surgical and medical procedures | 30 of 599 | 35 of 601 |
| Vocal cord dysfunction | Skin and subcutaneous tissue disorders | 27 of 599 | 35 of 601 |
| Ventricular assist device insertion | Surgical and medical procedures | 31 of 599 | 25 of 601 |
Serious events with the highest arm-level rate: cardiac dysfunction (0.2%), hypotension (0.2%), enterovirus infection (0.2%), rhinovirus infection (0.2%).
The registry lists safety figures by randomized arm. In the methylprednisolone arm, 2 participants are recorded with serious adverse events, 155 with other adverse events, and 12 deaths. In the placebo arm, the corresponding figures are 0, 176, and 17. These are counts as recorded; they are not adjusted comparisons and should not be read as rates or percentages.
The registry also lists four top serious adverse-event terms: cardiac dysfunction, hypotension, enterovirus infection, and rhinovirus infection, each shown as 0.2%. The entry does not assign these terms to a specific arm in the information, so they should not be attributed to either group. It does not report the number of events underlying each term or whether these events were considered related to study treatment.
The study summary says serious study drug-related adverse events were to be collected for 7 days after the last dose. That planned collection window does not establish that every listed safety figure uses that same window. The registry record does not report detailed event definitions, exposure-adjusted rates, confidence intervals, or a comparative safety analysis. Differences between raw counts may reflect group size, follow-up, event definitions, or chance; the figures alone do not establish a causal effect.
12. What the Result Does, and Does Not, Mean
The registry record establishes that a randomized, placebo-controlled trial was conducted to assess methylprednisolone in infants undergoing heart surgery with cardiopulmonary bypass. It gives the primary endpoint’s ranking structure and reports some safety counts. It does not report the primary outcome distribution across ranks or a single treatment comparison. Therefore, the record does not support a conclusion about the primary endpoint’s comparative result.
The endpoint is a composite that ranges from postoperative length of stay to severe complications and operative mortality. Its overall meaning depends on how outcomes were distributed and on the analysis used to compare groups. A result for the composite, even if reported, would not necessarily imply that every component differed in the same way. The registry provides no category-level counts or component-specific primary comparison with which to evaluate that issue.
The safety counts describe events recorded in each arm, not the causal effect of methylprednisolone. Without an analysis that accounts for the study design and its uncertainty, raw counts cannot establish that one treatment increased or decreased risk. The registry record also does not report whether the adverse events were judged related to study treatment or give confidence intervals for these figures.
Randomization and masking are important design features, but they do not replace a complete report of results. Readers would need the primary estimate, uncertainty, analysis population, missing-data approach, and relevant event definitions to assess the evidence more fully. Those details are not reported in the registry record. The appropriate conclusion is therefore about what is documented, not about whether methylprednisolone is effective or ineffective.
13. Why This Trial Matters Statistically
This trial illustrates a statistical challenge common to clinical research: how to summarize outcomes that differ in both type and severity. A global-rank endpoint can represent a hierarchy spanning length of stay and serious postoperative outcomes. That can be informative, but it makes the ranking rules and the resulting distribution central to interpretation. Readers need more than the endpoint name to understand what happened across treatment groups.
The trial also shows why a composite result should be accompanied by component-level context. A single summary can combine a change in length of stay with changes in severe complications, and these components may not contribute equally. Without counts by rank or a treatment estimate, the registry record does not let readers assess how the endpoint’s parts shaped the comparison. This is a limitation of the recorded result, not a reason to assume a particular pattern.
Safety reporting raises a related distinction between descriptive counts and comparative inference. The listed counts are useful for identifying what the registry records, but they do not include uncertainty or a stated comparative analysis. Event attribution and follow-up details also matter when interpreting adverse-event summaries. The registry record supplies a planned collection window for serious study drug-related events, but does not clarify the window used for each listed count.
Finally, a completed randomized trial and an enrollment figure do not by themselves establish the size, precision, or clinical meaning of an effect. Transparent reporting of estimates, confidence intervals, analysis populations, and missing-data methods is needed to connect design to conclusions. The unavailable primary endpoint comparison makes this record a useful example of why outcome definitions and complete results must be read together.
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 main limitation for interpreting this trial from the registry record is that category-specific primary results and a single primary treatment comparison are not reported. The endpoint definition is detailed, but the absence of its observed distribution prevents assessment of the direction, magnitude, and precision of the primary result. The registry notes that the primary outcome has several categories or time points and is not reduced to one comparison.
The record does not report the statistical analysis method, effect estimate, confidence interval, or p-value for the primary endpoint. It also does not report the sample-size justification, any interim analyses, or a detailed statistical plan. The planning assumptions associated with sample-size illustrations are not a substitute for the trial’s own design justification and do not represent observed power.
Analysis populations, exclusions, and missing-data handling are not reported in the registry record. Nor does it describe how ties or incomplete endpoint information were treated. These omissions make it difficult to evaluate whether the analyzed participants and methods align with the randomized design.
The adverse-event counts lack the detail needed for a full comparative safety assessment. The record does not specify whether the listed events were judged treatment-related, provide uncertainty estimates, or clearly connect the listed serious-event terms to an arm. The study summary describes a collection window for serious study drug-related adverse events, but the timing basis for each posted count is not specified in the available record.
Finally, the primary endpoint combines outcomes with distinct clinical meanings. Even a reported overall comparison would need component-level context to show which parts of the hierarchy contributed to it. Such context is not reported in the registry record.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: STeroids to REduce Systemic Inflammation After Infant Heart Surgery — NCT03229538.
- Registry record last updated 2026-01-05; status last verified 2023-05.
- Results first posted 2023-05-10.
- 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 STRESS as a completed phase 3 randomized, parallel-group, quadruple-masked trial of intravenous methylprednisolone versus intravenous isotonic saline placebo in infants undergoing heart surgery with cardiopulmonary bypass. Recorded enrollment is 1263. The primary endpoint assigns each participant a rank based on their most severe outcome, combining postoperative length of stay with specified complications, transplant, and operative mortality.
The registry record does not report the number of participants in each primary rank category or a single comparison between groups. It lists serious adverse events in 2 participants in the methylprednisolone arm and 0 in the placebo arm, other adverse events in 155 and 176 participants, and deaths in 12 and 17 participants, respectively. These are descriptive counts, not a complete comparative analysis. The methods used for the primary comparison, analysis populations, missing-data handling, and uncertainty estimates are not reported in the registry record.