Safety of Sildenafil in Premature Infants With Severe Bronchopulmonary Dysplasia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety of Sildenafil in Premature Infants With Severe Bronchopulmonary Dysplasia (NCT04447989), a randomized phase 2 study. The primary outcome is Safety Based Upon Number of Participants With Hypotension.

2. Clinical Question

3. Trial Design

The registry describes a randomized, placebo-controlled, parallel-group phase 2 study in premature infants with severe bronchopulmonary dysplasia. The brief summary characterizes it as multicenter, sequential dose-escalating, and conducted with infants in neonatal intensive care units. The design information lists quadruple masking, while the brief summary uses the term double-masked. The record does not explain the difference between those descriptions, so the masking procedures cannot be characterized more specifically from the registry record.

The dose plan had three sildenafil cohorts. Cohort 1 received sildenafil 0.5 mg/kg intravenously or 1 mg/kg enterally every 8 hours for 28 days; Cohort 2 received 1 mg/kg intravenously or 2 mg/kg enterally on the same schedule; and Cohort 3 received 2 mg/kg intravenously or 4 mg/kg enterally on the same schedule. Placebo was administered intravenously or enterally every 8 hours for 28 days. The registry describes the study as sequential dose-escalating but does not report the rules or criteria for moving between dose cohorts.

The enrollment figure is 125. The posted primary outcome results list 31 participants in Cohort 1, 29 in Cohort 2, 32 in Cohort 3, and 30 in placebo. The study status is completed. Its start date is 2021-05-27, and its primary completion date is 2024-12-12. The registry record does not report further details about randomization procedures, allocation concealment, the timing of dose-escalation decisions, or whether the cohorts were enrolled concurrently or consecutively.

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: Safety population

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.

Population: Intent to Treat (ITT) population

6. Primary Endpoint

The registry lists the primary outcome as Safety Based Upon Number of Participants With Hypotension, measured over 28 days post last dose of study drug, up to 9 weeks.

The primary endpoint counts participants who experienced hypotension. It is a binary safety outcome at the participant level: each participant either met the event definition during the stated follow-up or did not. The endpoint timeframe is 28 days after the last dose of study drug, with follow-up described as up to 9 weeks. The registry identifies the analysis population as the safety population.

The event definition is based on clinical significance and response, rather than on a numeric blood pressure threshold. A low blood pressure event counted when the treating physician considered it clinically significant and judged it to require a fluid bolus, or initiation or escalation of inotropic, vasopressor, or systemic steroid therapy with the specific intent of raising blood pressure. The registry record does not report a numeric threshold for hypotension or further adjudication procedures.

Counting participants with an event is a direct way to describe how often a defined safety event occurred in each group. It does not by itself describe the event’s duration, severity beyond the definition, or relationship to study treatment. The posted primary outcome records one participant with hypotension in Cohort 1 and zero in Cohort 2, Cohort 3, and placebo. The endpoint therefore describes event counts during the specified follow-up; interpretation also depends on the small group sizes and the uncertainty around the estimated between-group difference.

7. Results Posted in the Registry

• Cohort 1, Sildenafil: 3.2% (1/31) versus 0.0% (0/30) on control; risk difference 3.2 percentage points (95% CI -3.0 to 9.4); unadjusted two-proportion z-test p = 0.321.
• Cohort 2, Sildenafil: 0.0% (0/29) versus 0.0% (0/30) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0); unadjusted two-proportion z-test p = n/a.
• Cohort 3, Sildenafil: 0.0% (0/32) versus 0.0% (0/30) on control; risk difference 0.0 percentage points (95% CI 0.0 to 0.0); unadjusted two-proportion z-test p = n/a.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a binary outcome because it records whether each participant experienced the defined event during follow-up. For a binary outcome, the number of participants with an event and the number assessed give the basic group-level description. A risk difference expresses the difference between the event proportions in two groups, with positive values indicating a higher observed proportion in the sildenafil group than in placebo. The reported differences are unadjusted comparisons, not estimates adjusted for other participant characteristics.

What does the risk difference express?

The Cohort 1 comparison has an estimated risk difference of 0.032. Its confidence interval runs from -0.030 to 0.095, and its unadjusted p-value is 0.321. The estimate is based on one event in Cohort 1 and none in placebo. A risk difference is on the absolute scale, so it describes the difference in event proportions rather than a relative change. The reported estimate and interval should be read in light of the sparse event count. The Cohort 2 and Cohort 3 comparisons show no events in either group and report a risk difference of 0.0. Equal observed proportions can occur even when the underlying risks are uncertain.

How should a confidence interval be read?

General statistical explanation: a confidence interval describes the range of values that are reasonably compatible with the data under the method used to calculate it. It is not a guarantee that the true value lies in the interval, nor does it give the probability that a particular treatment effect is correct. For Cohort 1, the interval spans values below and above zero, so it does not rule out either direction of difference. The registry record does not report the method used to calculate the posted confidence interval. The zero-width intervals shown for comparisons with no events in either group require particular caution: a displayed interval of 0.0 to 0.0 should not be interpreted as precise evidence of zero uncertainty.

What does a p-value contribute?

General statistical explanation: a p-value describes how unusual data at least as incompatible with a specified null hypothesis would be if that hypothesis and the statistical model were correct. It is not the probability that the null hypothesis is true. The registry posts an unadjusted p-value of 0.321 for the Cohort 1 comparison. It does not post p-values for the other two cohort comparisons. A p-value alone does not describe effect size, clinical importance, or precision, and it does not replace the event counts and confidence interval.

What can be learned from zero events?

General statistical explanation: an event not being observed in a group does not show that its underlying probability is zero. With sparse outcomes, estimates may be highly sensitive to a small number of events, and conventional approximations can behave poorly at the boundary where observed counts are zero. The reported comparisons for Cohort 2 and Cohort 3 have no observed hypotension events in either study group. Their identical observed counts are informative as a record of what happened in these participants, but do not establish equal risks. The registry record does not specify the statistical method used for those estimates.

What is the role of the planning illustrations?

General statistical explanation: sample-size calculations can illustrate how many participants per arm might be needed to detect a chosen effect under stated assumptions. They are planning aids, not evidence about what the completed study had power to detect after the outcomes were observed. The illustrations accompanying the results use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect defined as a stated fraction of the observed effect. These are planning assumptions, not a report of the trial’s original design justification. The registry record does not report the trial’s own sample-size justification. The listed assumptions do not establish that the trial was designed or powered to detect the observed hypotension difference.

The registry identifies the primary population as the safety population. The record does not describe the statistical analysis plan, the handling of missing outcome data, any covariate adjustment, or the method used for confidence intervals. Those details should not be inferred from the displayed point estimates. The registry reports zero analyses, so the posted comparisons should be understood as descriptive, unadjusted recalculations from the reported group summaries rather than as a reported sponsor model.

10. Confidence Intervals and Precision

Precision is limited by the small number of recorded hypotension events. The Cohort 1 result rests on one event, while no events were recorded in the other sildenafil cohorts or placebo. A single event can have a noticeable effect on an estimated proportion when group sizes are limited. The point estimate for Cohort 1 compared with placebo is 0.032, and its confidence interval extends from -0.030 to 0.095. This interval includes zero and allows for uncertainty in both directions.

The interval is more informative than the point estimate alone because it makes the range of compatible differences visible. It does not indicate that every value in that range is equally plausible, and it does not establish the probability that the true difference falls within the interval. The method used to calculate it is not reported in the registry record. The unadjusted p-value of 0.321 is also not a substitute for an interval: it does not show the range of plausible effects or how precisely the effect was estimated.

For Cohort 2 and Cohort 3, the reported risk differences and interval limits are all 0.0 because there were no events in either comparison group. These displayed intervals should not be taken as evidence of perfect precision. General statistical explanation: methods that rely on approximations can be uninformative when all observed counts are zero, because the data contain little information about how often an event might occur in a wider population. The registry record does not report the interval method or any alternative uncertainty estimates for those comparisons.

The planning illustrations use the observed effect as a reference and specify a two-sided significance level of 0.05, equal allocation, and a normal approximation. Such calculations depend on their assumptions and are illustrations for planning, not observed power and not a guarantee that a future study would achieve a particular result. The registry record does not report the trial’s original sample-size justification.

11. Multiplicity, Missing Data and Analysis Populations

The study reports a primary safety outcome across three sildenafil dose cohorts compared with placebo. It also lists five secondary pharmacokinetic outcomes: central volume of distribution, peripheral volume of distribution, clearance, half-life, and peak plasma concentration. The registry reports seven outcome measures overall, but does not provide the names of all outcome measures in the posted outcome information. It reports zero analyses. A statistical plan describing how the primary and secondary outcomes were analyzed is not reported in the registry record.

When multiple outcomes or comparisons are examined, the chance of finding an apparently notable result by chance can increase. General statistical explanation: multiplicity procedures can control that risk, but they may affect the interpretation of p-values and confidence intervals. The registry record does not report whether any adjustment for multiple comparisons was planned or used. It also does not report a prespecified hierarchy among the dose-cohort comparisons or describe how the five named pharmacokinetic outcomes were to be interpreted alongside the primary outcome. The unadjusted p-value for Cohort 1 should therefore not be treated as a multiplicity-adjusted result.

The primary outcome population is identified as the safety population. The record does not report how that population was defined, whether all randomized participants were included, or whether the analysis followed an intention-to-treat principle. Missing outcome data, withdrawals, exclusions, and methods for handling missingness are not described in the posted primary results. The registry record does not report whether missingness differed by group or whether sensitivity analyses were performed.

These omissions matter because analysis populations and missing-data methods can affect which participants contribute to a comparison. Without those details, the posted counts describe the reported safety groups, but do not establish how participants without an observed outcome were handled. No assumptions about randomization, analysis exclusions, imputation, or multiplicity adjustment should be added to the reported findings.

12. 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 serious adverse event terms

Serious events with the highest arm-level rate: retinopathy of prematurity (6.5%), meningitis bacterial (3.3%), streptococcal sepsis (3.3%), urinary tract infection (3.3%), urinary tract infection enterococcal (3.3%).

The registry reports serious adverse events affecting 3 participants in Cohort 1, 0 in Cohort 2, 4 in Cohort 3, and 4 in placebo. It reports deaths of 0, 2, 1, and 0 participants in those same groups. The serious adverse event figures and deaths are distinct summaries; neither should be substituted for the primary hypotension outcome. The registry also lists serious adverse event terms with percentages: retinopathy of prematurity, 6.5%; bacterial meningitis, 3.3%; streptococcal sepsis, 3.3%; urinary tract infection, 3.3%; enterococcal urinary tract infection, 3.3%; and weaning failure, 3.3%.

These counts do not indicate whether events were judged related to sildenafil or placebo. The registry record does not report event-level attribution, timing, severity details beyond the serious-event classification, or the number of events per affected participant. The listed percentages should be read as registry-reported figures; their denominators are not specified in the posted information. Differences among these small groups are descriptive and do not by themselves establish a treatment-related pattern.

Zero participants in a group with a reported event does not show that the event cannot occur. Likewise, a higher count in a group does not establish that the intervention caused the events. Interpretation requires the study context, event definitions, follow-up, and uncertainty, some of which are not reported in the registry record.

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

The posted results document the frequency of hypotension, as defined by the study, in the reported safety groups. One participant in Cohort 1 experienced the event; no participants in Cohort 2, Cohort 3, or placebo did. The unadjusted Cohort 1 risk difference is 0.032, with a confidence interval from -0.030 to 0.095 and an unadjusted p-value of 0.321. These statistics describe a comparison with considerable uncertainty and do not establish that the underlying risks are the same or different.

The zero-event results in Cohort 2 and Cohort 3 are not proof that hypotension cannot occur at those doses. They mean that no participant in those groups met the recorded event definition during the specified follow-up. The displayed zero-width intervals do not provide a sound basis for claiming exact equality or complete certainty. The registry record does not report the method used to derive those intervals.

The findings do not establish whether sildenafil caused or prevented any reported event. The study description identifies a randomized, placebo-controlled, sequential dose-escalating safety study, but the posted figures alone do not show a causal explanation for between-group differences. Serious adverse events and deaths were reported separately, and their relationship to treatment is not reported in the registry record.

These data also do not establish safety for all premature infants with severe bronchopulmonary dysplasia, for other settings, or for use beyond the regimens and follow-up described. The results relate to the participants and group summaries in this record. The registry does not provide the original sample-size justification, detailed analysis plan, or missing-data procedures needed to judge all aspects of the study’s inferential design.

General statistical explanation: absence of strong evidence for a difference is not evidence that two treatments are equivalent. Demonstrating equivalence or noninferiority ordinarily requires a prespecified margin and an analysis designed for that question. Such a margin or analysis is not reported in the registry record. The results should therefore be read as a limited description of recorded events and their uncertain comparison, rather than as a definitive conclusion about treatment effects.

14. Why This Trial Matters Statistically

This study illustrates a recurring statistical challenge in safety research: clinically important events may be uncommon, while the groups available for comparison are limited in size. When few events occur, estimates can change appreciably with only a small change in the event count. The Cohort 1 comparison records one hypotension event against none in placebo, and its confidence interval spans values on both sides of zero. That uncertainty is central to interpretation, not a technical detail to set aside.

The three dose cohorts also make the design distinctive. The registry describes sequential dose escalation and reports separate results for three sildenafil regimens. Separate cohort summaries can help readers see the recorded outcomes at each dose, but the registry does not report the escalation rules, a formal dose-response analysis, or a statistical approach for comparing all cohorts. Without those details, a pattern across the cohorts should not be inferred from the sparse primary events.

The record also shows why outcome definitions matter. Hypotension was not defined by a numeric blood pressure cutoff; it depended on clinical significance and the physician’s judgment that an intervention to raise blood pressure was needed. This makes the endpoint clinically framed, but the registry record does not report further adjudication or consistency procedures. Readers need to distinguish the recorded endpoint from broader concepts of low blood pressure or overall safety.

Finally, the combination of event counts, uncertainty intervals, secondary outcomes, and serious-event summaries underscores the importance of a prespecified statistical plan. The registry does not report multiplicity procedures, missing-data methods, or the original sample-size justification. Those absences limit what can be concluded from the posted comparisons and make clear why statistical interpretation must keep the scope of the available evidence in view.

15. 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 has very few recorded events, limiting the precision of between-group estimates. The Cohort 1 comparison is based on one hypotension event in the sildenafil group and none in placebo; the other two sildenafil comparisons also have no events in either group. The zero-width intervals reported for those latter comparisons should not be interpreted as showing exact certainty. The registry record does not report the methods used to calculate the intervals.

The event definition relies on a treating physician’s judgment that low blood pressure was clinically significant and required intervention with the intent of raising blood pressure. The record does not give a numeric blood pressure threshold, explain whether events were independently adjudicated, or describe procedures for ensuring consistent assessment across centers. The registry describes the study as multicenter but does not name the number or location of centers.

Important design and analysis details are not reported in the registry record. These include the randomization procedure, allocation concealment, dose-escalation criteria, the original sample-size justification, the analysis plan, the handling of missing data, and any adjustment for multiple comparisons. The record identifies the safety population but does not explain how it was defined. It also reports zero analyses.

The registry’s design information lists quadruple masking, whereas its brief summary calls the study double-masked. The record does not clarify the differing descriptions. Serious adverse event counts and deaths are reported, but their relationship to study treatment is not. The listed serious-event percentages have no specified denominator in the posted information. These limitations constrain causal interpretation and generalization beyond the participants, regimens, and follow-up described.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Safety of Sildenafil in Premature Infants With Severe Bronchopulmonary Dysplasia — NCT04447989.
• Registry record last updated 2026-01-26; status last verified 2026-01.
• Results first posted 2026-01-26.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

This completed phase 2 study evaluated sildenafil safety in premature infants with severe bronchopulmonary dysplasia. The registry describes a randomized, placebo-controlled, parallel-group, sequential dose-escalating study. Sildenafil was administered in three dose cohorts, with placebo given on the same every-8-hours schedule for 28 days. Enrollment was 125, and the primary completion date was 2024-12-12.

The primary outcome counted participants with hypotension through 28 days after the last dose, up to 9 weeks. The posted results record one event in Cohort 1 and none in Cohort 2, Cohort 3, or placebo. The Cohort 1 risk difference compared with placebo is 0.032, with a confidence interval from -0.030 to 0.095 and an unadjusted p-value of 0.321. Serious adverse events and deaths are reported separately.

The sparse event counts limit precision. The registry record does not report the original sample-size justification, detailed statistical analysis plan, missing-data methods, or multiplicity procedures.