Efficacy and Safety of Intravenous Phenobarbital in Neonatal Seizures: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Intravenous Phenobarbital in Neonatal Seizures (NCT04320940), a randomized phase 3 study. The primary outcome is Neonates Who do Not Require Additional Seizure Treatment After the First Dose of Phenobarbital..

2. Clinical Question

The primary outcome concerns whether additional seizure treatment was needed after the first phenobarbital dose. It is a binary outcome at the participant level: a neonate either did or did not require additional treatment in the stated period. The registry describes the endpoint as a percentage, but the posted results give counts of participants.

The study rationale states that neonatal seizures can have long-term adverse effects, including death, and that placebo-controlled studies are not appropriate for this population. The planned contrast was therefore between a higher and a lower phenobarbital regimen. The stated aim was to demonstrate greater efficacy at a higher dose than at a lower dose by assessing prevention of subsequent seizures.

The result is limited to the posted endpoint counts. It does not by itself describe seizure burden, later outcomes, or whether subsequent treatment was needed after a second dose. Those are addressed by separate secondary outcomes in the registry, and their results are not included in the posted outcome data. The enrollment was 4, so any comparison between the regimens is necessarily based on very sparse participant counts.

3. Trial Design

The registry describes a randomized, double-blind, controlled, parallel-group Phase 3 study of intravenous phenobarbital sodium injection in neonates with electrographic or electroclinical seizures. It lists quadruple masking. Enrollment was 4. The record lists the study start as 2021-03-12 and primary completion as 2023-02-01, and the study status as terminated. The reason for termination is not reported in the registry record.

The parallel groups received different active dosing strategies. The lower-dose group was assigned 20 mg/kg initially, with a further 20 mg/kg if required. The higher-dose group was assigned 40 mg/kg initially, with a further 10 mg/kg if required. This is a comparison of whole regimens, including the possibility of an additional dose, rather than a comparison of isolated doses under otherwise identical treatment.

The planned primary outcome was the number or percentage of neonates who did not require additional seizure treatment after the first dose during the first 24 hours after treatment. Secondary outcomes listed in the registry concern the need for additional treatment after the second dose and seizure burden over 48 hours following initial administration. The registry lists four outcome measures overall, while three secondary outcomes are named in the record.

NEMA Research, Inc. is listed as the sponsor. Details beyond the broad design description—including the randomization procedure, allocation concealment, operational details of masking, sites, and eligibility criteria—are not reported in the registry record. The record also does not report a target enrollment or a sample-size justification. The observed enrollment of 4 should not be taken as evidence of the originally intended sample size or as an explanation for why the trial was terminated.

The record's rationale says a placebo-controlled design was not appropriate for this population and describes the intended dose comparison. That statement reports the study's rationale; it does not establish that the trial's design or observed results answer every question about treatment in neonatal seizures. The small enrollment and the posted counts sharply limit what can be learned from the comparison.

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

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 Neonates Who do Not Require Additional Seizure Treatment After the First Dose of Phenobarbital., measured over 24 hours.

The primary endpoint is the number of neonates who did not require additional seizure treatment after the first phenobarbital dose during the first 24 hours after treatment. The registry describes this outcome as a percentage, with participants as the unit. The posted arm results are counts: 2 participants in the 20mg group and 1 participant in the 40mg group met the endpoint. The registry lists 3 participants in the 20mg group and 1 participant in the 40mg group.

This endpoint captures an immediate treatment-response question that is relevant to the study's stated aim of preventing subsequent seizures. It also has a clear participant-level interpretation: whether or not additional treatment was needed after the initial dose within the specified timeframe. A binary endpoint can be straightforward to communicate, but it reduces a participant's experience to a yes-or-no outcome. It does not describe the number, duration, or severity of seizures, or the timing of additional treatment in greater detail.

The registry lists seizure burden over 48 hours as a secondary outcome, along with outcomes relating to whether additional treatment was needed after the second dose. These are distinct from the primary endpoint and should not be inferred from its counts. Results for those secondary outcomes are not reported in the posted results. The primary endpoint counts alone therefore provide only a narrow view of the clinical course during the stated observation period.

With only 4 participants enrolled, the endpoint counts are highly limited as a basis for comparing regimens. The counts document what was recorded for this outcome; they do not establish a stable estimate of the difference between the groups or the likely result in a broader population.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the primary outcome as binary: participants either did or did not require additional seizure treatment after the first phenobarbital dose during the first 24 hours. The posted values are counts of participants meeting the endpoint in each randomized arm. For a binary outcome, a conventional comparison would use the event and non-event counts in both groups to estimate how outcomes differ between regimens. The registry does not report a test, an effect estimate, or an interval for this comparison.

What does a confidence interval convey?

In general, a confidence interval describes uncertainty around an estimated effect under a specified statistical method. It is not a guarantee that the true effect lies within the interval, and it does not determine whether an effect matters clinically. With sparse data, estimated effects can be especially sensitive to the outcome of each participant. Here, no confidence interval is reported in the registry record, so there is no posted interval from which to assess comparative precision.

Why is a comparison not estimable?

The record marks the binary two-arm comparison as not estimable from the posted values. The enrollment and outcome counts are extremely small, including only one participant in the 40mg group. With so few observations, the data provide little information for a stable estimate of a between-group effect. This is not the same as demonstrating that the regimens have identical outcomes, nor does it establish that a difference exists. It means a comparative result is not available from the posted analysis.

More generally, the method used for a binary endpoint needs to reflect the data structure and the intended estimand. Methods that rely on large-sample approximations can be unreliable when counts are sparse. Exact approaches can be used in some small-sample settings, but the registry record does not say that an exact method was used or planned. The test, model, effect measure, handling of zero cells, and any covariate adjustment are not reported in the registry record. No method should be attributed to the trial without that information.

What does randomization contribute?

Random allocation is intended to make treatment groups comparable on average, including with respect to measured and unmeasured factors. It does not ensure that small groups will be balanced in any particular trial. With enrollment of 4, randomization cannot be relied on to create comparable groups across all prognostic characteristics. Baseline characteristics and the allocation procedure are not reported in the registry record, so the degree of balance cannot be assessed from the information presented.

Masking can reduce the influence of treatment knowledge on aspects of trial conduct and outcome assessment. The record lists quadruple masking, but does not identify the roles masked or explain how masking was implemented. Because the primary endpoint concerns whether additional seizure treatment was required, it is also relevant that operational decisions could influence a recorded endpoint; however, the record does not describe decision criteria or adjudication procedures. These details are not reported in the registry record.

What can sample-size planning illustrate?

In general, sample-size planning asks how many participants would be needed to detect a specified effect under assumptions about the outcome, allocation, and statistical test. The sample-size illustrations associated with the posted results are described as planning illustrations based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. They are not the trial's original design justification and should not be interpreted as observed power. The registry record does not report the trial's own sample-size justification.

The illustrations are particularly difficult to interpret when the observed effect is derived from only 4 participants. A planning calculation based on a sparse observed effect can vary substantially with small changes in those counts and should not be treated as evidence that a study was adequately or inadequately powered. The trial's target sample size and planning assumptions beyond the stated illustration assumptions are not reported in the registry record. No numerical sample-size requirement should be inferred from the enrollment alone.

Over-dispersion is a general issue for outcomes measured as counts: observed variability may exceed what a simple model assumes, which can make uncertainty estimates too narrow if the model does not account for it. That concept is not directly applicable to the posted primary endpoint as a participant-level binary outcome, and no count-model analysis is reported. Seizure burden, listed as a secondary outcome, could require details about how repeated events are defined and analyzed; those details are not reported in the registry record.

The registry lists zero analyses. Therefore, the posted outcome counts should be read descriptively rather than as results from a reported inferential procedure. The analysis population, missing-data approach, prespecified model, covariate strategy, and any interim-analysis plan are not reported in the registry record. These omissions prevent assessment of whether the posted counts follow a prespecified analysis plan or how the investigators intended to handle complications in the data.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Phenobarbital Sodium Injection 20mg", "Phenobarbital Sodium Injection 40mg")
n      <- c(3, 1)
events <- c(2, 1)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Phenobarbital Sodium Injection 20mg  3  2  66.7  9.4  99.2  20.8  93.9  19.4  93.2
ARM  Phenobarbital Sodium Injection 40mg  1  1  100.0  2.5  100.0  20.7  100.0  15.8  98.7

11. Multiplicity, Missing Data and Analysis Populations

The registry lists the primary endpoint and three secondary outcomes: not requiring additional seizure treatment after the first dose, not requiring additional treatment after the second dose, and seizure burden over 48 hours following initial administration. The primary endpoint is distinct from the secondary outcomes, which address later treatment need and seizure burden. The registry also lists four outcome measures overall. The results for the secondary outcomes are not reported in the posted data.

When a study evaluates multiple outcomes, the chance of at least one apparently notable result can increase if each outcome is tested separately without an appropriate plan. In general, multiplicity can be addressed by specifying a hierarchy, adjusting significance thresholds, or treating analyses as exploratory. Which approach was planned, whether any secondary outcomes were formally tested, and whether multiplicity adjustments were intended are not reported in the registry record. No testing hierarchy should be assumed.

Missing-data handling is also not reported in the registry record. The posted primary counts list 3 participants in the 20mg arm and 1 participant in the 40mg arm, consistent with the total enrollment of 4, but the record does not describe analysis-population definitions, outcome availability procedures, or how missing observations would have been handled. It is not possible to infer from these counts whether there were no missing data under a prespecified definition.

The record does not report whether the analysis was intention-to-treat, per-protocol, or based on another population. It also does not report interim analyses or stopping rules. The terminated status does not explain the reason for termination or establish that a particular analysis was conducted. These missing details constrain interpretation of both the primary result and the unreported secondary outcomes.

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: pulmonary hypertension (33.3%).

The serious adverse-event counts are reported separately by arm. In the 20mg group, 3 participants were at risk and 1 was reported as having a serious adverse event. In the 40mg group, 1 participant was at risk and none was reported as having a serious adverse event. The record lists zero deaths in each group. These are counts, not a comparative safety analysis.

The registry lists pulmonary hypertension as a top serious adverse-event term at 33.3%. The denominator and event-level details for that term are not reported in the registry record, so the percentage should not be used to infer more detail than the record states. The record also lists zero other adverse events in each group, distinct from the serious adverse-event counts.

These observations come from a total enrollment of 4. A serious event recorded in a small group does not, by itself, establish that the intervention caused the event; likewise, zero recorded events in a group of one participant cannot rule out risk. The registry does not report a statistical comparison of safety outcomes, event timing, severity details beyond the serious classification, or attribution to study treatment. The counts describe what was posted, with substantial limits on what can be concluded about comparative safety.

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

The posted primary endpoint records that 2 participants in the 20mg group and 1 participant in the 40mg group did not require additional seizure treatment after the first dose. The arm sizes were 3 and 1, respectively. These counts describe the observed participants; they do not establish which dosing regimen is more effective. The registry does not report an inferential comparison, and it marks a two-arm binary comparison as not estimable.

The study's stated question was whether a higher phenobarbital dose would show greater efficacy than a lower dose in preventing subsequent seizures. The posted data do not answer that question with useful precision. The higher-dose group contains one participant, and the full study enrolled 4 participants. Any estimate from these data would be highly dependent on the outcome of a very small number of individuals.

The result does not show that the regimens are equivalent. Equivalence requires a suitable design and analysis with a prespecified margin, neither of which is reported in the registry record. Nor does the result demonstrate that the higher-dose strategy is superior. A difference in raw counts is not itself a statistical test or a dependable estimate of a treatment effect.

The primary outcome is limited to additional treatment after the first dose during the first 24 hours. It does not capture every aspect of seizure control, nor does it provide the results of the listed secondary outcomes on second-dose treatment or seizure burden over 48 hours. Those secondary results are not reported in the registry record. Safety counts are likewise too sparse to establish comparative safety.

In general, results from a small randomized study can be useful descriptively and may identify questions for further study, but their precision depends on the amount and quality of information available. The record reports that this study was terminated, yet the reason is not reported. Without the planned sample size, analysis plan, outcome availability details, and a reported comparison, the posted results should be understood as a limited account of four participants rather than a definitive evaluation of either regimen.

14. Why This Trial Matters Statistically

This trial matters statistically because it illustrates the gap between a clear research question and the amount of information needed to answer it reliably. The question is clinically direct: whether a higher-dose phenobarbital regimen prevents the need for additional seizure treatment more effectively than a lower-dose regimen. The posted primary outcome also has an understandable participant-level interpretation. Yet enrollment was 4, with only 1 participant in the higher-dose group.

Randomization and masking are important design features, but they do not compensate for sparse outcome data. Randomization supports fair allocation; it cannot ensure balance in very small groups. Masking can limit some forms of bias, although operational details are not reported. Neither feature turns the observed counts into a precise estimate or removes the uncertainty created by the small enrollment.

The record also demonstrates why descriptive counts and inferential results should be distinguished. The counts document participants meeting the endpoint, while the registry reports no analyses and no estimable two-arm comparison. A numerical contrast can look straightforward but still provide very little information about the likely difference between treatment strategies. Precision and analysis matter alongside the direction of the observed counts.

Finally, the terminated status and absence of a reported reason, sample-size justification, and statistical analysis plan leave important questions unanswered. The record cannot establish whether the enrollment reflects an intended design, a premature stop, or another circumstance. These limitations are not grounds to invent an explanation; they are reasons to interpret the posted outcomes narrowly. The trial is a useful example of why transparent reporting of planning, conduct, and analysis is essential to interpret clinical-trial results.

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 most immediate limitation is the small enrollment: 4 participants in total, with 3 in the 20mg group and 1 in the 40mg group. The very small higher-dose group makes its observed result particularly sensitive to the outcome of its single participant. Baseline characteristics are not reported in the registry record, so group comparability on relevant characteristics cannot be assessed.

The study status is terminated, but the reason for termination is not reported in the registry record. The planned enrollment, whether recruitment met expectations, and whether the trial stopped under a prespecified rule are also not reported. The primary completion date is listed as 2023-02-01, but the record does not explain the circumstances surrounding completion or termination.

The registry reports primary endpoint counts but no inferential analysis. It marks the binary two-arm comparison as not estimable. A p-value, confidence interval, effect estimate, and analysis method are not reported in the registry record. The analysis population and handling of missing outcome data are likewise not reported, limiting the ability to assess how the posted counts were produced.

The intervention comparison involves two different dosing regimens, including different initial doses and different additional doses if required. The primary endpoint focuses on need for further treatment after the first dose, while the listed secondary endpoints address treatment after a second dose and seizure burden over 48 hours. Results for those secondary measures are not reported in the posted results. Details of outcome definitions, assessment procedures, and treatment decisions are not reported in the registry record.

Safety observations are also limited by the small number of participants. The record does not report a comparative safety analysis or detailed event-level information. Taken together, these trial-specific limitations mean the posted results are descriptive and cannot provide a precise estimate of comparative efficacy or safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Efficacy and Safety of Intravenous Phenobarbital in Neonatal Seizures — NCT04320940.
• Registry record last updated 2023-10-26; status last verified 2023-10.
• Results first posted 2023-10-26.
• 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

The registry record describes NCT04320940, titled “Efficacy and Safety of Intravenous Phenobarbital in Neonatal Seizures.” It is listed as a randomized, quadruple-masked, parallel-group Phase 3 study sponsored by NEMA Research, Inc., with a status of terminated. The recorded enrollment is 4. The study compared an initial 20 mg/kg phenobarbital regimen, with a further 20 mg/kg if required, against an initial 40 mg/kg regimen, with a further 10 mg/kg if required.

The primary outcome concerns participants who did not require additional seizure treatment after the first dose during the first 24 hours. The posted counts are 2 of 3 in the lower-dose group and 1 of 1 in the higher-dose group. The record lists three secondary outcomes, reports zero analyses, and marks the two-arm binary comparison as not estimable. The reason for termination, statistical analysis plan, sample-size justification, and missing-data approach are not reported in the registry record.