Sleep Apnea and Fetal Growth Restriction: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Sleep Apnea and Fetal Growth Restriction (NCT04084990), a randomized phase 3 study. The primary outcome is Birth Weight.

2. Clinical Question

3. Trial Design

The registry describes a Phase 3, randomized, triple-masked, parallel-group study sponsored by Washington University School of Medicine. Enrollment was 3. The study start date is 2019-11-18, and the primary completion date is 2021-07-27. Its status is listed as terminated; the reason for termination is not reported in the registry record.

The study description outlines a screening pathway for pregnant patients diagnosed with FGR: an OSA questionnaire followed by a home sleep monitor. Patients diagnosed with OSA were to be assigned to nightly aPAP or no aPAP, described as standard care. The intervention arm is described as nightly aPAP use through delivery, and the device listed is S9 VPAP Adapt. The record identifies the design as randomized and parallel, but does not report details of sequence generation, allocation concealment, or the practical procedures used to implement masking. The identities of the masked parties are not reported in the registry record.

The registry lists birth weight as the primary outcome, measured in grams at one day, with the description “Weight of child at time of birth.” Gestational age at delivery is listed as a secondary outcome. Although two outcome measures are indicated, the registry does not report a numerical result for gestational age at delivery. It also does not report an analysis for either outcome. The reported birth-weight arm summaries include one recorded participant in each group, although the study description says two mothers were randomized to aPAP and explains why one aPAP-assigned fetus had no birth weight recorded.

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: 32 participants provided informed consent to participate. They underwent an obstructive sleep apnea (OSA) risk assessment questionnaire. Those screening as high risk for OSA underwent home sleep apnea testing (HSAT). Those screening as positive on the HSAT, which was defined as an apnea hypopnea index (AHI) \>= 10 OR an AHI \>= 5 with oxygen desaturation index \>= 5, were randomized to receive aPAP or not. Of the 32 participants, only 3 met criteria for enrollment in the randomization phase.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Of two mothers randomized to aPAP therapy, one required pre-term induction of labor for maternal safety for preeclampsia with severe features. This fetus experienced in-utero fetal demise, so birth weight was not obtained.

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 Birth Weight, measured over 1 day.

Birth weight is the weight of the child at the time of birth, recorded in grams. It is the trial’s primary endpoint and is intended to capture the newborn’s size at delivery. The registry specifies a one-day timeframe. In general, a continuous endpoint such as weight can be described by its location and variability, and a comparison between randomized groups can estimate how their observed distributions differ. The record’s selected summary is the mean, with full range as the dispersion measure.

The posted arm results contain one recorded birth weight in each group: 2320.0 grams for aPAP and 570.0 grams for No aPAP. With one observation recorded in each arm, the listed full range in each group has the same lower and upper value as its reported birth weight. These observations do not characterize the range of outcomes that might occur across a larger group of participants. The registry does not report an inferential comparison or uncertainty interval for the endpoint.

The record explains that, of two mothers randomized to aPAP, one required pre-term induction of labor for maternal safety because of preeclampsia with severe features. This fetus experienced in-utero fetal demise, so birth weight was not obtained. That absence is directly relevant to interpreting the primary endpoint: the recorded aPAP birth weight does not represent all participants assigned to that arm. The registry does not report an imputation or other statistical method for handling this missing birth weight.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies the study as randomized, triple masked, and parallel group. In general, randomization assigns participants to study groups using a chance-based process; it is intended to reduce systematic differences in prognostic characteristics. A parallel design follows separate groups receiving different assigned strategies. Masking is intended to limit the influence of knowledge about assignment on aspects of study conduct or assessment. The registry does not describe the randomization procedure, allocation concealment, or which participants or study personnel were masked, so the specific safeguards used cannot be assessed from this record.

The primary outcome is continuous: birth weight in grams. The registry names the mean as the summary measure and full range as the dispersion measure. In general, the mean describes the arithmetic center of recorded observations, while the range spans the lowest to the highest recorded value. These summaries are most informative when there are enough observed outcomes to describe the group. In the posted results, each arm has one recorded birth weight, so its reported full range runs from that single observation to itself. This does not provide information about within-group variability across participants.

What does an inferential comparison add?

A descriptive summary reports what was observed in the study participants. An inferential comparison uses a statistical model or test to estimate how compatible the observed data are with different possible group effects, while accounting for sampling variability under the stated assumptions. An interval estimate can also express uncertainty around an estimated contrast. The registry reports no analyses, p-values, confidence intervals, or adjusted estimates for birth weight. Consequently, there is no posted inferential result to interpret, and the precision of a between-group effect is not quantified.

In general, very small groups yield imprecise estimates because each observation can have substantial influence on the group summary. Missing outcomes can add another source of uncertainty, particularly when the reason an outcome is absent is connected to clinical events. Here, the record explains that a birth weight was not obtained for one aPAP-assigned fetus after in-utero fetal demise. The record does not state whether any analysis was planned to account for that missing outcome, or whether an analysis population different from the available birth-weight observations was used. No assumption about the missing value can therefore be evaluated.

How should planning calculations be understood?

In general, a sample-size calculation describes the number of participants needed under stated assumptions about the effect to detect, variability, significance level, allocation, and analysis. Such calculations are planning tools, not evidence that a study has adequate precision after it is completed. The registry does not report a sample-size justification or the assumptions used to set the planned enrollment. The record describes an illustrative calculation framework based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. However, it does not report numerical planning results in the posted material available for this analysis. Any such illustration should be understood as planning rather than as observed power or as a feature of the trial’s original design.

The word “power” in a planning context refers to the probability that a specified statistical procedure would detect a particular effect if that effect were present and the model assumptions held. It is not a measure of how likely the observed result is to be true. A post hoc calculation based on the observed effect is especially difficult to interpret as a measure of evidential strength; it does not replace a confidence interval or a prespecified analysis. No observed-power claim can be drawn from the registry record.

For the primary endpoint, the missing birth weight also matters to any hypothetical planning calculation based on the observed effect. The two recorded values are not a complete outcome record for the aPAP-assigned group as described in the study narrative. The registry does not provide a variance estimate suitable for characterizing a broader population, and does not report an analysis that incorporates the missing outcome. Accordingly, a calculation based on the observed summaries would have substantial limitations and would not repair the absence of an inferential analysis.

Over-dispersion is a general concern when outcome variability is greater than a statistical model assumes. It can make standard errors too small and confidence intervals too narrow if the model does not accommodate the extra variation. The record does not report a dispersion assessment or a fitted model, and the single recorded birth weight in each arm cannot characterize the degree of variation within either group. No conclusion about over-dispersion is possible from these summaries.

What can be concluded from the posted methods?

The registry provides design labels, outcome definitions, arm summaries, and a brief explanation for a missing primary outcome. It does not provide a statistical analysis plan, model specification, covariate adjustment strategy, interim-analysis description, or multiplicity procedure. These details are not reported in the registry record. The safest interpretation is therefore descriptive: the recorded birth weights are available, but a comparative estimate and its uncertainty are not.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists birth weight as the primary outcome and gestational age at delivery as a secondary outcome. It indicates two outcome measures, but only the birth-weight arm values are reported in the outcome data. A numerical result for gestational age at delivery is not reported in the registry record. The record lists 0 analyses, so it does not provide inferential tests for either outcome.

When a study examines multiple outcomes or makes multiple comparisons, the chance of at least one apparently notable result can increase. In general, a multiplicity plan specifies whether and how statistical inference accounts for that issue. No multiplicity procedure or testing hierarchy is reported in the registry record. Because there are no posted inferential analyses, no multiplicity-adjusted or unadjusted test result can be interpreted.

Missing data are particularly consequential for the primary endpoint. The registry states that one of two mothers randomized to aPAP had a fetus that experienced in-utero fetal demise after pre-term induction of labor for maternal safety related to preeclampsia with severe features; birth weight was not obtained. The aPAP birth-weight summary therefore represents one recorded observation, not both aPAP assignments described in the narrative. The registry does not report imputation, a missing-data model, a sensitivity analysis, or another method for addressing this outcome.

The record also does not report named analysis populations, such as whether all randomized participants were included in a designated primary analysis regardless of adherence or outcome availability. It is therefore not possible to establish from the registry which analysis population was intended or used. The reason for termination and the statistical plan are not reported in the registry record. These omissions limit assessment of how the available outcome values relate to the original analysis intentions.

11. 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.

The registry reports serious adverse events for 0 of 2 participants at risk in the aPAP group and 0 of 1 participant at risk in the No aPAP group. For other adverse events, it likewise reports 0 of 2 at risk in the aPAP group and 0 of 1 at risk in the No aPAP group. The reported death count is 0 in each group. These are the event counts and at-risk denominators recorded in the registry.

In general, a zero count means that no affected participants were recorded for that category in the specified group; it does not establish that an event is impossible or that a strategy has no safety risks. The denominators are very small, so the counts offer limited information about how frequently adverse events might occur in a larger population. The registry does not list serious adverse-event terms, and its posted counts do not include a statistical comparison between groups.

The narrative about preeclampsia with severe features and in-utero fetal demise is separately relevant to interpretation of the missing birth weight. The safety counts should not be used to erase or reinterpret that narrative. The registry does not report further detail about adverse-event collection, follow-up completeness, or the relationship between the narrative event and the tabulated adverse-event categories.

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

The posted results document two recorded birth weights: 2320.0 grams in the aPAP arm and 570.0 grams in the No aPAP arm. Each arm’s reported birth-weight summary contains one observation. The registry also documents why a birth weight was unavailable for one aPAP-assigned fetus. These registry entries describe the available outcome record, but they do not establish a causal effect of aPAP on birth weight.

A randomized design can support a causal comparison when the study is adequately conducted and analyzed, but the design label alone does not supply an estimate or demonstrate its precision. Here, the registry reports no analysis, confidence interval, or p-value. The available arm values cannot show whether an apparent difference is compatible with chance variation, whether it would persist in a larger group, or how uncertain a group-level estimate would be. The record also does not report a planned enrollment justification or the reason the study was terminated.

The missing birth weight is not an ordinary unrecorded measurement with an explained administrative cause: the registry connects it to fetal demise after pre-term induction for maternal safety related to preeclampsia with severe features. That information warrants careful separation of the recorded birth-weight summaries from the full set of aPAP assignments. No statistical handling of the missing outcome is described, so a complete-case summary should not be read as a complete account of all randomized participants.

Nor do the posted results establish equivalence between aPAP and standard care. Equivalence requires an appropriate prespecified margin and an analysis precise enough to assess whether plausible effects fall within that margin. Neither an equivalence margin nor such an analysis is reported in the registry record. The record’s zero adverse-event counts are likewise limited observations, not proof that adverse events cannot occur.

The results are best read as a sparse descriptive record from a terminated study with enrollment of 3, one recorded birth weight per arm, a primary outcome missing for an aPAP-assigned fetus, and no reported inferential analyses. They identify what was recorded and an important reason for missingness, while leaving the comparative effect and its precision unresolved.

13. Why This Trial Matters Statistically

This study illustrates why the distinction between a reported outcome value and an estimated treatment effect matters. A numerical summary describes the observations that were recorded. A treatment-effect estimate requires a defined comparison and statistical analysis, while an assessment of precision requires uncertainty information such as a confidence interval. The registry reports two birth-weight values but no analysis result or uncertainty interval. The observations therefore do not provide a precise estimate of the comparison between aPAP and no aPAP.

It also illustrates how the denominator for an endpoint can differ from the number assigned to a group. The study narrative reports two mothers randomized to aPAP, while the birth-weight arm summary contains one recorded observation because the other fetus experienced in-utero fetal demise and birth weight was not obtained. Recognizing that distinction is essential: an outcome summary based on observed measurements does not automatically represent every randomized assignment.

The very small enrollment, termination status, and absence of a reported statistical analysis plan make transparent reporting especially important. Details about missing-data handling, analysis populations, and the reason for termination would help readers assess what was intended and how the recorded results relate to that plan. Those details are not reported in the registry record.

For biostatistical interpretation, the central lesson is restraint. Randomization is a design feature, not a substitute for an outcome analysis. A conspicuous difference between two single observations should not be promoted into a population-level conclusion. Conversely, the lack of a reported test does not show that the strategies are the same. The registry supports a description of the available outcomes and their limitations, but not a firm estimate of the comparative effect.

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 primary endpoint has only one recorded birth weight in each arm. The registry says two mothers were randomized to aPAP, but birth weight was not obtained for one aPAP-assigned fetus after in-utero fetal demise. This leaves the aPAP outcome summary incomplete relative to the assignments described in the narrative. The registry does not report an imputation approach, a sensitivity analysis, or another method for addressing the missing endpoint.

Enrollment was 3, and the study is listed as terminated. The reason for termination and the intended enrollment target are not reported in the registry record. With so few recorded outcomes, the arm summaries cannot characterize the distribution of birth weights or support a stable estimate of between-group variability. No inferential analysis, confidence interval, or p-value is posted, so the comparative effect and its precision cannot be assessed.

Gestational age at delivery is listed as a secondary outcome, but its numerical result is not reported in the registry record. A statistical analysis plan, analysis populations, multiplicity procedure, covariate adjustment strategy, and interim-analysis information are also not reported. The registry identifies the study as triple masked but does not identify who was masked or describe the masking procedures. It also does not report the randomization-generation or allocation-concealment procedures.

The adverse-event entries show no affected participants in the small reported risk groups, but these counts do not characterize uncommon events in a larger population. The registry does not report further details about adverse-event ascertainment or follow-up completeness. Together, the limited outcomes and missing methodological details constrain interpretation to a description of the posted record, rather than a definitive comparative assessment.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Sleep Apnea and Fetal Growth Restriction — NCT04084990.
• Registry record last updated 2022-03-09; status last verified 2022-03.
• Results first posted 2022-03-09.
• 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.

18. Record Summary

NCT04084990, titled “Sleep Apnea and Fetal Growth Restriction,” is listed as a Phase 3 randomized, triple-masked, parallel-group study sponsored by Washington University School of Medicine. Enrollment was 3. The study began on 2019-11-18, reached primary completion on 2021-07-27, and is listed as terminated. The reason for termination is not reported in the registry record.

The intervention was nightly aPAP through delivery, using the listed S9 VPAP Adapt device; the comparator was no aPAP, described as standard of care. Birth weight in grams was the primary outcome, and gestational age at delivery was listed as secondary. The registry reports one birth weight per arm: 2320.0 grams for aPAP and 570.0 grams for No aPAP. It explains that one of two mothers randomized to aPAP had a fetus that experienced in-utero fetal demise, so birth weight was not obtained. No analyses are reported. Serious and other adverse-event counts were 0 in each group, with 0 deaths reported.