Nasal HFOV Versus Nasal CPAP to Reduce Post-extubation pCO2: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Nasal HFOV Versus Nasal CPAP to Reduce Post-extubation pCO2 (NCT02340299), a randomized na study. The primary outcome is paCO2 at 72 h After Extubation.

2. Clinical Question

The clinical question is a comparison of two forms of non-invasive respiratory support after extubation. The outcome is a physiological measurement rather than a patient-reported outcome: arterial paCO2 indicates the pressure of carbon dioxide in arterial blood. The registry describes the objective as investigating whether nHFOV reduces paCO2 relative to nCPAP at the post-extubation assessment.

The result available for the primary endpoint does not form a two-group comparison. The endpoint could be assessed for 2 participants in the nHFOV group, with a median of 65.5 mmHg and a full range from 52.0 to 79.0 mmHg. In the nCPAP group, all patients had to be reintubated within 72 hours, so paCO2 at the primary assessment could not be assessed for that group. This means the observed primary-endpoint values cannot establish a between-group difference. The registry record does not report a comparison estimate or inferential test for the endpoint.

3. Trial Design

The registry describes a randomized, parallel-group trial with no masking. Its title is “Nasal HFOV Versus Nasal CPAP to Reduce Post-extubation pCO2,” and the official title specifies very low birth weight infants. The sponsor is Charite University, Berlin, Germany. The record lists the study start as 2015-01-01, primary completion as 2017-12-31, and status as terminated. The phase is listed as NA.

The two interventions were devices used immediately after extubation. The nHFOV description specifies binasal prongs, a frequency of 10 Hz, an I:E ratio of 33:66, amplitude of 20 cm H2O, Pmean of 8 cm H2O, and flow of 7 l/min. FiO2 was to be set to maintain SpO2 at 90-94%. The nCPAP description specifies binasal prongs, a CPAP level of 8 cm H2O, and flow of 7 l/min, with FiO2 set to maintain SpO2 at 90-94%. The registry states that weaning was left to the attending physician’s discretion.

Enrollment is listed as 6. The posted primary outcome is paCO2 at 72 hours after extubation, with a measurement window from 64 h to 80 h. The outcome is reported as a median and full range. Although the study was randomized, the primary outcome could not be evaluated in the nCPAP group because all patients in that group were reintubated within 72 hours. Random assignment therefore does not yield an assessable primary-outcome comparison in the posted data. Details about the sequence-generation procedure, allocation concealment, and other implementation aspects 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

Analysis population for the primary outcome, as defined in the registry: In patients that were reintubated within 72 h, the primary outcome, paCO2 72 h after extubation, could not be assessed. In the nCPAP group, all patients had to be reintubated within 72 h. The primary outcome, paCO2 72 h after extubation, could therefore only be assessed in the nHFOV group.

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 paCO2 at 72 h After Extubation, measured over 64 h to 80 h.

The primary endpoint is arterial partial pressure of carbon dioxide (paCO2), measured between 64 and 80 hours after extubation and described as assessed, on average, 72 hours after extubation. Its unit is mmHg. The registry specifies the median and full range as the summary measures. This is a physiological endpoint: it quantifies carbon dioxide pressure in arterial blood at a defined post-extubation time window.

In general, a prespecified physiological measure can provide a direct, consistently defined way to compare groups on a biological characteristic relevant to the study question. A fixed assessment window also makes clear when the measurement is intended to represent participants’ status. These are general reasons such outcomes are used; they do not establish that a difference in paCO2 necessarily corresponds to a particular patient-important benefit or harm.

For this trial, the ability to collect the endpoint depended on participants remaining eligible for the assessment. The registry states that patients reintubated within 72 h could not have the primary outcome assessed. All patients in the nCPAP group were reintubated within 72 h, leaving no paCO2 measurements for that group at the primary time point. The registry reports measurements for 2 nHFOV participants: the median was 65.5 mmHg, with a full range from 52.0 to 79.0 mmHg. Because the comparator endpoint is unavailable, the posted primary endpoint does not support a numerical contrast between the assigned groups.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the reported median describe?

The primary endpoint is summarized by a median and full range. In general, the median is the central value after observations are ordered; it is less influenced by extreme values than the arithmetic mean. The full range identifies the smallest and largest observed values. For the nHFOV group, the registry reports a median of 65.5 mmHg and a full range from 52.0 to 79.0 mmHg. These statistics describe the observations available for that group. They do not, on their own, estimate a difference from nCPAP or provide a measure of uncertainty around such a difference.

With very few observations, descriptive summaries can change substantially when additional observations are included. A range also does not show how values are distributed between its endpoints. It is therefore important to keep the summary tied to its scope: it characterizes the assessable nHFOV outcomes, rather than the randomized groups as a whole. The nCPAP outcome has no reported value because the primary assessment was unavailable after reintubation.

Why is there no direct group contrast?

A between-group analysis requires outcome information for the groups being compared, or a clearly specified method for handling unavailable outcomes. The record reports paCO2 measurements for 2 nHFOV participants and none for the nCPAP group. Since all patients in the nCPAP group were reintubated within 72 hours, the primary endpoint could not be assessed there. There is thus no pair of observed group summaries from which to describe the primary endpoint contrast.

In general, missingness can affect interpretation when the reason that a measurement is absent is related to participants’ clinical course or outcome. Here, the registry explicitly links the unavailability of the primary endpoint to reintubation within 72 h. That makes it especially important not to assume that the unobserved paCO2 values would resemble the available nHFOV measurements. The registry record does not report an imputation method, a model for the unobserved outcomes, or a sensitivity analysis.

What would a confidence interval add?

In general, a confidence interval for a group contrast describes the uncertainty around an estimated difference under the assumptions of the statistical method. A narrow interval would indicate greater precision than a wide interval, while an interval spanning values on either side of no difference would be compatible with a range of possible effects. Those interpretations apply only when a contrast and its interval have actually been calculated and the analysis assumptions are appropriate.

No primary-outcome contrast or confidence interval is reported in the registry record. The observed median and range cannot be converted into a valid comparative confidence interval without additional data and an analysis approach. The record also does not report an inferential test for the primary endpoint. It would therefore be inappropriate to infer statistical significance or precision from the descriptive nHFOV values alone.

What does the analysis record say?

The registry lists 19 outcome measures and 13 secondary outcomes posted. It lists zero analyses. The record does not describe the statistical analysis plan, the model intended for the primary outcome, or the population used for an efficacy analysis. It does not report whether an analysis was planned to follow randomized assignment regardless of treatment received, or how participants with missing measurements were to be handled.

In general, the analysis population and missing-data method matter because they determine which participants contribute to an estimate and how unavailable outcomes are treated. An intention-to-treat approach, for example, is a general analysis principle that retains participants in their randomized groups, but it does not by itself resolve an unobserved endpoint. Any method for addressing such missingness requires assumptions that should be stated and examined. Those details are not reported in the registry record.

How should planning calculations be understood?

In general, sample-size calculations are planning tools for a future or planned study. They depend on assumptions about the effect to detect, variability, allocation, significance level, and the statistical method. They do not measure the power achieved after a trial has ended. The registry describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation; it says the target effect is a stated fraction of the observed effect. No numerical planning estimates are included in the record summarized for this trial.

Because the primary outcome is unavailable for nCPAP, the posted primary endpoint does not furnish an observed comparative effect on which to base a meaningful efficacy interpretation. Any sample-size illustration should be understood only as planning, not as observed power or evidence that the trial had a particular chance of detecting an effect. The registry record does not report the trial’s own sample-size justification.

10. Multiplicity, Missing Data and Analysis Populations

The record lists 19 outcome measures and 13 secondary outcomes posted. These include pH, paO2, paCO2, and base excess at specified post-extubation assessments, as well as successful extubation. When multiple outcomes are considered, the chance of observing an apparently notable result among them can increase. In general, a prespecified multiplicity strategy helps clarify which comparisons are primary and how secondary findings should be interpreted. The registry record does not report a multiplicity adjustment or an overall strategy for interpreting multiple outcomes.

The primary outcome is missing for the nCPAP group because all patients in that group were reintubated within 72 hours, and the registry says paCO2 could not be assessed for patients reintubated within 72 h. The primary outcome is reported for 2 nHFOV participants. This is not a routine situation in which values are absent for an unspecified reason: the record ties unavailability to a post-extubation clinical event. No imputation procedure, missing-data model, or sensitivity analysis is reported.

The registry describes the trial as randomized, but it does not report the analysis population or how the randomized groups were handled in the analysis. Details about whether all randomized participants were included in each outcome analysis, and how post-randomization events affected inclusion, are not reported in the registry record. Without those details and a method addressing the missing comparator endpoint, the primary outcome cannot yield a complete randomized-group comparison. The record also does not report interim analyses or a statistical analysis plan.

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 5% reporting threshold.

Most frequent non-serious adverse event terms

The posted safety counts distinguish serious adverse events, other adverse events, and deaths. For nHFOV, serious adverse events were reported for 0 of 4 participants at risk; other adverse events were reported for 2 of 4 at risk; deaths were 0. For nCPAP, serious adverse events were 0 of 2 participants at risk; other adverse events were 0 of 2 at risk; deaths were 0. These are counts in the registry record, not comparative risk estimates.

The denominators differ between the safety groups, and the counts are small. In general, a zero count in a small group does not establish that an event cannot occur or that the underlying risk is zero. Likewise, a difference in observed counts does not by itself establish a causal difference between interventions. The record does not provide a statistical comparison of adverse events or uncertainty intervals around the counts.

The record lists no top serious adverse-event terms. That should be read alongside the reported serious-event counts, rather than as evidence that all possible harms were assessed in the same way. Safety reporting depends on the definitions, observation period, and completeness of recording; details about those aspects are not reported in the registry record. The reported safety counts are descriptive and do not support a broad conclusion about comparative safety.

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

The registry record establishes that this was a randomized comparison of nHFOV and nCPAP immediately after extubation in very low birth weight infants. It also documents the primary endpoint and the outcome values that could be assessed. For nHFOV, paCO2 at the primary assessment had a median of 65.5 mmHg and a full range from 52.0 to 79.0 mmHg among 2 participants. For nCPAP, the endpoint could not be assessed because all patients in that group were reintubated within 72 hours.

These registry entries do not establish whether one support strategy produced lower paCO2 than the other at the primary assessment. The comparator group has no observed endpoint values, and the record does not report an analysis that estimates a group contrast despite this missingness. The nHFOV summary is therefore not evidence of a treatment effect relative to nCPAP. Nor does the absence of nCPAP paCO2 values mean that those values were normal, abnormal, or equal to any particular value; they were not assessed as specified.

The record also does not establish that the observed reintubations were caused by the assigned support strategy. It states that all nCPAP patients had to be reintubated within 72 hours, but the statistical and clinical context needed to interpret this event comparatively is not reported. Randomization is an important design feature, yet the small enrollment and unavailable primary endpoint limit what can be concluded from the posted results.

In general, a trial result is most informative when the endpoint, analysis population, treatment contrast, uncertainty, and missing-data approach are all clear. Here, the primary endpoint is specified, but a two-group estimate and its precision are absent. The record therefore documents trial conduct and limited outcome data; it does not support a definitive comparative conclusion about post-extubation paCO2 or a general conclusion about clinical benefit or harm.

13. Why This Trial Matters Statistically

This trial illustrates why outcome availability is part of the statistical question, not merely a reporting detail. The intended primary comparison concerned paCO2 at a defined time after extubation, but the endpoint could not be assessed for the nCPAP group because all patients in that group were reintubated within 72 hours. The issue is not simply that an outcome value was omitted from a summary: the primary measure was unavailable for an entire randomized group under the stated assessment conditions.

It also demonstrates a distinction between randomization and successful estimation. Random allocation can help make groups comparable at baseline, but it cannot by itself recover an outcome that was not measured. A valid analysis would need to specify how the unavailable endpoint is treated and what assumptions support that treatment. The registry record does not report such an approach or a comparative primary-outcome analysis.

The outcome is reported with a median and full range, which appropriately conveys the form of the recorded summary but does not overcome the missing comparator data. The available nHFOV values are descriptive. They cannot answer the randomized comparison without further assumptions and an analysis that the record does not report.

The safety counts offer a related lesson about scale and uncertainty. Counts from small groups are useful for describing what was recorded, but they are not precise comparative estimates. Reading the counts as proof of equivalent or different safety would go beyond the information in the record.

More broadly, the record shows why interpretation should keep design, measurement, analysis, and uncertainty distinct. Randomized design, a clearly named endpoint, and posted observations are important, but they do not replace a reported treatment contrast with an appropriate account of missing outcomes. This trial is statistically informative as an example of how endpoint availability can constrain interpretation, while its registry results do not resolve the comparative primary question.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The main limitation for interpretation is that the primary endpoint is available only for the nHFOV group. The registry says the endpoint could not be assessed in reintubated patients and that all nCPAP patients were reintubated within 72 hours. The reported nHFOV median and range consequently cannot be compared with a corresponding nCPAP summary. The record does not report an alternative estimand, a method for handling these missing outcomes, or sensitivity analyses under different assumptions.

Enrollment is listed as 6, and the primary endpoint is reported for 2 nHFOV participants. Such a limited set of measured outcomes provides little information about the distribution of paCO2 beyond the reported descriptive summary. The record does not report a confidence interval or inferential test for a primary-outcome contrast. It also does not report the trial’s sample-size justification, so the intended precision or target effect cannot be assessed from the registry record.

The registry lists 19 outcome measures and 13 secondary outcomes posted, but does not report a multiplicity strategy or a complete statistical analysis plan. The analysis population and handling of missing outcomes are not reported. These omissions make it difficult to judge how outcomes were selected, how analyses were planned, or how secondary findings should be interpreted together.

The record identifies the trial as randomized and unmasked, but details of sequence generation and allocation concealment are not reported. Weaning was left to the attending physician’s discretion, as stated in the intervention descriptions. These features are relevant context, but their impact cannot be quantified from the posted information. The trial status is terminated; the record does not explain the reason for termination.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Nasal HFOV Versus Nasal CPAP to Reduce Post-extubation pCO2 — NCT02340299.
• Registry record last updated 2020-08-05; status last verified 2020-07.
• Results first posted 2020-08-05.
• 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

NCT02340299 is a randomized, parallel-group, unmasked trial comparing nHFOV with nCPAP immediately after extubation in very low birth weight infants. The registry lists enrollment of 6, status as terminated, and primary completion on 2017-12-31. The primary endpoint is paCO2 assessed between 64 h and 80 h after extubation, summarized by median and full range.

The nHFOV group had a reported median of 65.5 mmHg and full range of 52.0 to 79.0 mmHg among 2 participants. The primary endpoint was not assessable in the nCPAP group because all patients in that group were reintubated within 72 hours. No comparative primary-outcome analysis is reported. Safety counts are descriptive, and the record does not report a statistical analysis plan, multiplicity approach, or method for addressing the missing comparator endpoint.