Topiramate in Neonates Receiving Whole Body Cooling for Hypoxic Ischemic Encephalopathy: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Topiramate in Neonates Receiving Whole Body Cooling for Hypoxic Ischemic Encephalopathy (NCT01765218), a randomized phase 1, phase 2 study. The primary outcome is Number of Patients With Seizures.

2. Clinical Question

3. Trial Design

The registry describes a randomized, quadruple-masked, parallel-group Phase 1 and Phase 2 trial. Its title is “Topiramate in Neonates Receiving Whole Body Cooling for Hypoxic Ischemic Encephalopathy,” and its official title describes topiramate as an adjuvant to therapeutic hypothermia. The listed condition is hypoxic ischemic encephalopathy.

The recorded enrollment is 34. The posted arm sizes are 17 for placebo and 17 for topiramate. The placebo group received the institution’s standard of care, described as whole body cooling for 72h followed by gradual rewarming. Infants assigned to topiramate were to receive topiramate in addition to cooling, at 5 mg/kg enterally each day for a total of 5 doses. The first dose was to be administered as soon as possible on admission.

The registry lists a start date of 2013-02 and a primary completion date of 2022-01-27. The study status is terminated. The reason for termination and any effects of termination on recruitment, follow-up, or analysis are not reported in the registry record. The registry records one outcome measure and no posted secondary outcomes. It does not report the trial’s original sample-size justification, the intended recruitment target beyond the recorded enrollment, or the trial’s original statistical analysis plan.

Randomization and masking describe design features, not guarantees that the groups were comparable in every respect or that assessment was free of bias. The record does not give details of the randomization procedure, the people included in the quadruple masking, or whether masking was maintained. Those 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.

Overall Study

Reasons for not completing this period:

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Number of Patients With Seizures, measured over At 4 weeks post-natal age or the time of hospital discharge (whichever is earlier).

The primary endpoint is the number of participants with clinical or electrical seizures. The registry defines the observation window as before hospital discharge or before 4 weeks post-natal age, whichever was earlier. The comparison is between the topiramate group and the control group.

This is a binary outcome at the participant level: each participant is counted according to whether a seizure event was recorded during the specified window. The posted counts are 1 of 17 participants in the topiramate group and 3 of 17 in the placebo group. The registry lists the outcome unit as participants with seizures and the measure as a number.

In general, an event count can answer a clinically interpretable question about how many participants experienced the defined outcome, while a comparison of event proportions accounts for the number of participants in each group. The choice of seizures as the primary endpoint is consistent with the trial’s stated goal of evaluating topiramate as an adjunct to cooling. The registry does not report additional rationale for choosing this endpoint, nor does it report a definition distinguishing clinical from electrical seizures or details about how seizure events were assessed.

The endpoint’s time window matters when interpreting the result: it does not describe seizure occurrence beyond the earlier of hospital discharge or 4 weeks post-natal age. The record does not provide results for a longer follow-up period.

7. Results Posted in the Registry

• Topiramate: 5.9% (1/17) versus 17.6% (3/17) on control; risk difference -11.8 percentage points (95% CI -33.1 to 9.5), risk ratio 0.33 (0.04–2.89); unadjusted two-proportion z-test p = 0.287.

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is binary at the participant level: the registry reports counts of participants with seizures. In general, binary-outcome analyses compare the chance or proportion of an event between groups. The posted comparison uses a risk difference, which expresses the topiramate event proportion minus the placebo event proportion. Its negative point estimate corresponds to the lower observed proportion in the topiramate arm. A risk difference of zero would represent equal event proportions; the reported confidence interval includes that value.

What does the risk ratio add?

In general, the risk ratio divides the event proportion in one group by the event proportion in the comparison group. A ratio of one represents equal event proportions; values below one indicate a lower event proportion in the numerator group, and values above one indicate a higher event proportion. The posted risk ratio is 0.333, with a 95% confidence interval from 0.038 to 2.893. Because the interval includes one and spans widely on both sides, the estimate is not precise enough to determine the direction or size of a relative difference reliably.

How is an odds ratio different?

In general, odds are the chance of an event relative to the chance of no event. An odds ratio compares these odds between groups. The posted odds ratio is 0.292, with a 95% confidence interval from 0.027 to 3.133. A value of one represents equal odds. The interval includes one and is wide, so the estimate is compatible with a range of relative associations. When events are uncommon, an odds ratio and a risk ratio may be numerically similar, but they remain different measures and should not be treated as identical.

What does the p-value describe?

The registry posts an unadjusted p-value of 0.287. In general, a p-value describes how unusual data at least as discrepant as those observed would be under a specified null hypothesis and analysis model. It is not the probability that the null hypothesis is true, and it does not measure the probability that the finding is due to chance. A p-value alone does not convey the size or precision of an effect. Here, the registry does not report the test used, its assumptions, or a fuller analysis plan, so the p-value should be interpreted alongside the event counts and confidence intervals rather than in isolation.

Why does the small number of events matter?

In general, estimates from binary outcomes depend on the observed event and non-event counts. When the event counts are small, a change in a small number of observations can materially shift a proportion or relative estimate. Confidence intervals reflect this uncertainty, although the precise method used to construct the posted intervals is not reported in the registry record. The registry gives counts of 3 and 1 participants with seizures in the two groups, so the interval estimates are appropriately central to interpretation: the point estimates alone would give an incomplete picture.

Randomization is intended to support a fair comparison by assigning participants to groups in a way that reduces systematic differences at baseline. Masking can reduce some sources of bias in care and outcome assessment. These are general design principles; the registry record does not describe the randomization procedure, the masking roles, or how masking was maintained. The posted results also do not report baseline characteristics, so the degree of observed group similarity cannot be assessed from the registry information.

The record labels the trial randomized, quadruple-masked, and parallel-group, and lists the comparison as topiramate versus placebo. It does not report an adjusted effect estimate, a model specification, or an analysis population such as intention-to-treat or per-protocol. It also does not report how missing outcome data were handled. These omissions do not establish that no such methods were used; they mean those methods are not reported in the registry record.

Multiplicity refers, in general, to the challenge of interpreting results when many outcomes, groups, time points, or analyses are tested. More testing can increase the chance of at least one apparently unusual result, so an analysis plan may specify how to address that issue. The registry lists one outcome measure and no posted secondary outcomes, but does not report whether additional outcomes or analyses were planned, whether interim analyses occurred, or whether any multiplicity adjustment was used. Conclusions about these matters cannot be drawn from the posted record.

The posted effect estimates and p-value are unadjusted according to the record. No sponsor analysis that takes precedence is identified, and the registry does not report further analytical details. General statistical interpretation therefore focuses on the recorded counts, the posted estimates, and their uncertainty without assuming an unreported test, model, or data-handling procedure.

10. Confidence Intervals and Precision

The risk difference estimate is -0.118, with a 95% confidence interval from -0.331 to 0.095. In general, a confidence interval describes a range of estimates compatible with the data and the method used to construct it. It is not a guarantee that the true effect lies in the interval, nor does it assign a probability to each possible value after the data have been observed.

This interval includes zero, the value corresponding to no difference in event proportions. It also extends to negative and positive values, so the data are compatible with a lower observed event proportion in the topiramate group, little or no difference, or a higher event proportion. Its breadth indicates substantial uncertainty about the magnitude and direction of the underlying difference. The point estimate alone should not be treated as a precise estimate of what would be expected in another setting.

The relative estimates have similarly broad intervals. The risk ratio was 0.333, with a 95% confidence interval from 0.038 to 2.893. The odds ratio was 0.292, with a 95% confidence interval from 0.027 to 3.133. For either ratio, one represents no relative difference, and each interval includes one. Their breadth signals that the recorded event counts do not tightly determine the relative association.

In general, precision tends to improve when more informative outcome data are available, while sparse events can leave estimates uncertain. The registry does not report the interval-construction method, so the exact procedure behind these bounds is not available. The posted intervals should be read as uncertainty around the reported comparisons, not as proof of benefit, harm, or equivalence.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. They are illustrations based on the observed effect, not the trial’s original design assumptions. The registry’s planning description specifies a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For each row, the effect to detect is the stated fraction of the observed effect.

For planning for 100% of the observed effect, the estimates are 117 participants per arm for 80% power and 156 participants per arm for 90% power. For planning for 75% of the observed effect, they are 231 participants per arm for 80% power and 308 participants per arm for 90% power. For planning for 50% of the observed effect, they are 568 participants per arm for 80% power and 760 participants per arm for 90% power.

In general, power is the chance that a specified analysis would reject a null hypothesis when a particular effect is present, under the planning assumptions. The 80% and 90% columns describe different target probabilities of detecting the assumed effect if it were present. Higher target power generally requires more participants for the same assumed effect. Smaller effects are harder to distinguish from random variation, so planning for a smaller fraction of the observed effect requires more participants per arm in these illustrations.

These figures are not observed power. They are planning illustrations based on this trial’s observed effect and the stated assumptions, not evidence about the trial’s original design or what its investigators planned. The registry does not report the original sample-size justification or original design assumptions.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.176; treatment proportion p₁ = 0.059; average p̄ = 0.118.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4556 = 0.8930.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4480 = 0.3770.
4. n = (0.8930 + 0.3770)² / (-0.1176)² = 116.5, rounded up to 117 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.176470588235294                          # control proportion
p1 <- 0.0588235294117647                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1765218)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 117 (80% power), 156 (90% power)
 75% of observed effect: n per arm = 231 (80% power), 308 (90% power)
 50% of observed effect: n per arm = 568 (80% power), 760 (90% power)
power.prop.test: 116.5 
Simulated power at n = 117 per arm: 0.822 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure: participants with seizures before hospital discharge or before 4 weeks post-natal age, whichever came earlier. It lists no posted secondary outcomes and no analysis entries. Those registry entries describe what is posted; they do not establish that no other outcomes or analyses were planned, collected, or conducted.

Multiplicity is a general concern when several outcomes, time points, subgroup comparisons, or statistical tests are examined. If many tests are conducted, the chance of at least one result appearing noteworthy can increase. A prespecified analysis plan can define primary comparisons and any adjustment or interpretation strategy. The registry record does not report a multiplicity strategy, planned subgroups, interim analyses, or whether any adjustment was made.

The record also does not describe missing data or the number of participants with complete primary outcome information. It does not report how participants who withdrew, had incomplete follow-up, or had an unavailable seizure assessment were handled. The recorded arm sizes and outcome counts do not, by themselves, establish whether every randomized participant was included in the outcome analysis.

Analysis populations are also not specified. In general, an intention-to-treat analysis retains participants in their randomized groups, while a per-protocol analysis may restrict attention to participants meeting specified treatment or follow-up criteria. The registry does not say which approach was used, whether another population was analyzed, or how departures from assigned treatment were handled.

These reporting gaps limit assessment of the analysis process and the completeness of the evidence. They should not be read as proof that the trial lacked a statistical plan or did not address missing data or multiplicity. Those details are simply not reported in the registry record.

14. 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

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: sepsis (5.9%), necrotizing enterocolitis (0.0%), chronic lung disease (0.0%).

The registry reports serious adverse events among 17 participants at risk in each group. The recorded serious adverse event count was 1 in the placebo group and 0 in the topiramate group. Other adverse event counts were 2 in the placebo group and 4 in the topiramate group. The reported death counts were 1 and 0, respectively. These are descriptive counts; the record does not provide an analysis establishing whether events were caused by study treatment.

The registry lists Sepsis at 5.9%, Necrotizing Enterocolitis at 0.0%, and Chronic Lung Disease at 0.0% among the leading serious-event terms. The record does not provide further context for these term-specific percentages in the listed information. They should not be interpreted as comparative treatment effects without details about event definitions, timing, attribution, and the method used to summarize them.

In general, safety counts need context, including how events were defined, how long participants were monitored, whether events were considered treatment-related, and how missing safety information was handled. These details are not reported in the registry record. Small counts are descriptive and cannot, on their own, establish that one group had a safer experience than the other.

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

The posted results describe a randomized comparison of topiramate added to whole body cooling versus placebo alongside standard care in infants with hypoxic ischemic encephalopathy. For the primary seizure outcome, the registry records 1 participant with seizures in the topiramate group and 3 in the placebo group during the stated observation window. The estimated risk difference is -0.118, with a 95% confidence interval from -0.331 to 0.095, and the unadjusted p-value is 0.287.

These observations show a lower event count in the topiramate arm, but the estimates are imprecise. The confidence interval includes no difference and extends in both directions. The relative estimates also have wide intervals that include no relative difference. Thus, the posted data do not establish a precise effect or demonstrate that the groups are equivalent.

The result applies to the defined seizure outcome and its time window: before hospital discharge or before 4 weeks post-natal age, whichever was earlier. It does not describe outcomes beyond that window. The registry does not report details about the completeness of outcome follow-up, analysis populations, missing data, or the reason the study was terminated. These unknowns limit how fully the posted result can be evaluated.

In general, a nonsignificant p-value is not proof that there is no effect, just as a point estimate in one direction is not proof of a treatment effect. The interval estimates and the limited event counts are important to the interpretation. The data should be understood as an uncertain comparison in the registry record, not as a definitive finding about treatment efficacy or safety.

16. Why This Trial Matters Statistically

This trial presents a useful statistical example of why event counts, effect estimates, and uncertainty need to be considered together. The observed seizure counts were small, and the point estimate favored a lower observed event proportion in the topiramate group. Yet the risk-difference interval extends across zero, and the relative-measure intervals include the value corresponding to no difference. A directional point estimate is not enough to establish a precise treatment effect.

The absolute and relative measures also answer different questions. A risk difference describes how far apart the event proportions are on an absolute scale, while risk ratios and odds ratios describe comparisons on relative scales. When the recorded event counts are limited, each measure can be uncertain, and the apparent size of an estimate should be read alongside its interval.

The registry’s planning illustrations highlight the importance of specifying a target effect. Planning for smaller fractions of the observed effect requires more participants per arm under the stated assumptions. These are planning calculations based on the observed effect, not a report of the original design. The registry does not report the original sample-size justification, so the planning illustrations cannot establish whether the trial was designed to detect a particular effect.

The terminated status and the absence of reported detail about missing data, analysis populations, and the reason for termination make transparent interpretation especially important. The record supports a description of the posted counts and estimates, but leaves important questions about analysis and completeness unanswered. Distinguishing what the data show from what the record does not report is central to sound biostatistical reading.

17. 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 outcome is based on a small number of recorded seizure events, and the confidence intervals are wide. This limits the precision of both the absolute and relative comparisons. The registry does not report additional outcome details such as how clinical and electrical seizures were defined or assessed, so the measurement process cannot be evaluated from the record.

The study status is terminated. The reason for termination is not reported in the registry record, nor is the effect, if any, on recruitment, follow-up, or analysis. The record lists enrollment of 34 and arm sizes of 17 each, but does not provide a participant-flow account or explain whether every randomized participant contributed complete primary outcome data.

The registry does not report the original sample-size justification, analysis plan, analysis population, missing-data procedures, or method used to construct the confidence intervals. It also does not report baseline characteristics, details of randomization, or how masking was implemented. These omissions limit assessment of comparability, analytical choices, and possible sources of bias; they do not establish that the trial did not address these issues.

The safety information is also limited to recorded counts and listed serious-event terms. Event definitions, follow-up duration for safety, attribution, and the basis for the term-specific percentages are not reported in the registry record. The posted information therefore supports descriptive reporting, but not a detailed comparative safety assessment.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Topiramate in Neonates Receiving Whole Body Cooling for Hypoxic Ischemic Encephalopathy — NCT01765218.
• Registry record last updated 2024-06-11; status last verified 2024-05.
• Results first posted 2024-06-11.
• 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.

21. Record Summary

The registry record describes NCT01765218, a randomized, quadruple-masked, parallel-group Phase 1 and Phase 2 trial of topiramate as an adjunct to whole body cooling for infants with hypoxic ischemic encephalopathy. The listed comparator is placebo alongside standard care. The recorded enrollment is 34, and the posted arm sizes are 17 for placebo and 17 for topiramate. The study status is terminated.

The primary outcome is the number of participants with seizures before hospital discharge or before 4 weeks post-natal age, whichever was earlier. The posted counts were 3 in the placebo group and 1 in the topiramate group. The risk difference was -0.118, with a 95% confidence interval from -0.331 to 0.095; the unadjusted p-value was 0.287. The interval indicates considerable uncertainty about the comparison.

The registry lists no posted secondary outcomes or analysis entries. The reason for termination and further details about the statistical plan, missing data, and analysis populations are not reported in the registry record.