Physiological Changes With High-Flow Nasal Cannula: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Physiological Changes With High-Flow Nasal Cannula (NCT03700606), a non-randomized na study. The primary outcome is Percent of Unventilated Lung as Assessed by Electrical Impedance Tomography.

2. Clinical Question

3. Trial Design

This completed study enrolled 80 participants and is registered as a non-randomized, unmasked crossover study. Its stated purpose was to measure physiological parameters in extremely low birth weight infants receiving HFNC compared with nasal CPAP. The registry lists respiratory distress syndrome in premature infants as the condition and Sharp HealthCare as the sponsor. The study start date is 2019-03-15, and the primary completion date is 2021-12-20. The phase is listed as NA.

The intended sequence began with a baseline period on nCPAP. Participants then crossed over to HFNC, with measurements described at 30 minutes and at 6 hours, unless failure criteria prompted an earlier return to nCPAP. The record also describes a later measurement after participants returned to nCPAP. HFNC was described at a flow rate of 8 L/min, while baseline nCPAP settings were 5-7 cmH₂O. The record does not report the full operational definition of failure criteria.

A crossover design collects measurements under more than one condition in the same participant. This can make within-participant comparisons informative because each participant contributes measurements under different support conditions. However, interpretation depends on the sequence, timing, stability of the measured condition, and whether earlier support might affect later measurements. The registry record identifies the study as non-randomized and does not report randomized assignment of the order of conditions. It also does not report masking. The treatment conditions therefore are not equivalent to independently randomized groups, and changes across periods may reflect both support condition and time or sequence.

The registry record does not report a detailed statistical analysis plan, the method used to account for repeated measurements, or a numerical comparison of the conditions. It lists no analysis results. Accordingly, the design and scheduled measurements can be described, but the magnitude, uncertainty, and statistical significance of the primary comparison cannot be evaluated from the 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.

Before assignment: Two participants experienced unexpected data loss prior to analysis and were excluded from all study periods. During Period 3, 20 participants were unable to complete the planned 6-hour EIT assessment following transition to HFNC. Per protocol, these participants required immediate transition back to CPAP, precluding collection of the 6-hour time-point data. EIT data were obtained for all participants after re-establishing CPAP support for 60-minutes and included in the results.

Baseline Measurement (CPAP)

Reasons for not completing this period:

30 min post HFNC start

6 hours after initial transition to HFNC

Period 4: Stabilization on CPAP

Analysis population for the primary outcome, as defined in the registry: In Period 3, twenty infants who met predefined HFNC failure criteria were returned immediately to nCPAP and therefore did not complete the scheduled 6-hour HFNC EIT assessment. All remaining participants contributed Period 3 data as available. In Period 4, EIT was collected for nearly all infants, although one participant had missing primary outcome data for percent of unventilated lung.

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 Percent of Unventilated Lung as Assessed by Electrical Impedance Tomography, measured over Primary outcome was measured during each study period: baseline on nCPAP, 30 minutes after starting HFNC, at 6 hours on HFNC or earlier if failure, and 60 minutes after returning to nCPAP..

The primary endpoint is the percentage of lung regions that are not engaged in tidal volume, assessed using electrical impedance tomography (EIT). The record describes this as an assessment of unventilated lung and says atelectasis is calculated from the lung fields not engaged in tidal volume. EIT records changes in electrical impedance across the chest and can be used to characterize the distribution of ventilation without relying only on a single overall respiratory measure.

This endpoint is relevant to a physiological comparison because it focuses on how much of the imaged lung field participates in tidal ventilation under the different respiratory-support conditions. The planned measurements span baseline nCPAP, early HFNC, later HFNC or an earlier failure-related assessment, and a return to nCPAP. Repeated measurements can describe how the endpoint varies over the study sequence, although a change across periods does not by itself establish that the support condition caused the change.

The primary outcome is specified as a continuous measure, with the mean as the summary parameter and standard deviation as the stated measure of dispersion. A mean describes the average measured value; a standard deviation describes the spread of participant measurements around that average. Neither provides the numerical result without the corresponding observed values. The registry record does not report the primary outcome means, standard deviations, estimated contrasts, or confidence intervals. Thus, the endpoint and its measurement schedule are described, but the observed level and precision of the endpoint are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean and standard deviation describe?

The primary measure is continuous, and the registry specifies the mean as its summary parameter and standard deviation as its measure of dispersion. The mean is an average across the observed values. The standard deviation conveys how spread out those values are around the average. These summaries describe the observed distribution; they are not interchangeable with an estimate of the difference between study periods. To understand a comparison, readers would also need a contrast or another suitable estimate, together with an indication of its uncertainty. The registry record does not report these numerical outcome summaries.

Why does the crossover structure matter?

In general, a crossover study measures participants under more than one condition. Since measurements from the same participant are related, an analysis should account for that dependence rather than treating every period’s observations as if they came from unrelated people. A within-participant analysis can reduce the influence of stable differences between participants, but it does not automatically resolve changes over time, order effects, or carryover from an earlier condition. Those issues are relevant when support conditions are applied in sequence. The registry record does not report the statistical model, the method for handling repeated observations, or a specific assessment of carryover.

The record identifies the design as non-randomized and unmasked. In general, random allocation can help balance factors that influence which condition is received first, while masking can reduce some influences on measurement or assessment. The record does not describe either protection. This does not establish that bias occurred, but it limits what can be concluded about how fully sequence and assessment influences were controlled.

How should the measurement schedule be understood?

The primary outcome was scheduled at baseline on nCPAP, 30 minutes after starting HFNC, at 6 hours on HFNC or earlier if failure occurred, and after return to nCPAP. These are not all equivalent observation circumstances: one is baseline, some occur after a transition, and another follows a return to the earlier support. A difference between observations may reflect the support condition, elapsed time, clinical changes, or a combination. In general, a statistical analysis of repeated periods needs to preserve the timing and within-participant structure rather than treating the observations as one undifferentiated set.

The stated primary outcome summary is a mean with standard deviation. No estimate of a period-to-period change, standard error, confidence interval, or p-value is reported in the registry record. A confidence interval, in general, expresses uncertainty around an estimated quantity by identifying a range of values compatible with the data under the analysis assumptions. Without an estimate and its uncertainty, readers cannot judge how precisely a difference has been measured. The lack of a reported test result also means statistical significance cannot be assessed from this record.

What do failure-related measurements mean for analysis?

Twenty infants meeting predefined HFNC failure criteria returned immediately to nCPAP and did not complete the scheduled 6-hour HFNC EIT assessment. The registry says they were analyzed as Period 3 without a 6-hour EIT measurement. A measurement at an earlier failure-related return is not the same timing as a measurement taken after completing the scheduled HFNC period. In general, missingness related to a participant’s condition or ability to remain on a study condition can be informative: the availability of a measurement may be connected to the outcome or clinical course. The record does not report an analysis method for this issue, so its potential effect on a period comparison cannot be quantified.

The record also notes missing primary outcome data for one participant in Period 4, who was analyzed as Period 4 without the 60-minute return-to-nCPAP EIT measurement. The record says EIT was collected for nearly all infants in Period 4, but does not provide a complete numerical account of observations by period or explain how the missing Period 4 value was handled in any model. No imputation approach or sensitivity analysis is reported in the registry record.

What can the planning assumptions tell readers?

The registry describes sample-size planning illustrations based on the observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. Such calculations are planning illustrations, not observed power. In general, a planning calculation estimates the sample size needed to detect a chosen effect under stated assumptions; it does not show that an effect was detected in the completed study. The registry record does not report numerical planning counts or the observed effect value needed to examine those illustrations, and it does not report the trial’s own sample-size justification. These assumptions therefore cannot substitute for primary-outcome results or an analysis plan.

Overall, the design and outcome definitions are described, but the analytical methods needed to interpret the repeated-measures comparison are not reported in the registry record. Without those methods and the underlying estimates, the data’s precision and the robustness of any comparison remain unknown.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and four secondary outcome domains: Geometric Center of Ventilation (CoV) - Ventral Dorsal, End-expiratory Lung Impedance, Relative Tidal Stetch, and Oxygenation Ratio. It lists five outcome measures overall. The record does not report a multiplicity strategy, such as a prespecified hierarchy or adjustment for testing multiple outcomes. In general, when several outcomes or comparisons are examined, the chance of at least one apparently unusual result can increase; a stated analysis plan helps readers understand how that issue was addressed. No outcome-specific analysis results are listed, so multiplicity-adjusted or unadjusted findings cannot be evaluated.

Missingness is described for the primary outcome in two ways. Twenty infants meeting HFNC failure criteria returned to nCPAP and did not complete the scheduled 6-hour HFNC EIT assessment. The registry says they were analyzed as Period 3 without that scheduled measurement, while the remaining participants contributed Period 3 data as available. Separately, one participant had missing primary outcome data in Period 4 and was analyzed without the 60-minute return-to-nCPAP EIT measurement. The record does not report the method used to include incomplete observations in a statistical analysis, an imputation method, or sensitivity analyses.

The analysis populations are not fully described. The registry record does not report a formal intention-to-treat or per-protocol population, a detailed account of exclusions, or the number of observations analyzed at each period. It also does not report whether the failure-related measurements were treated differently from scheduled 6-hour measurements in a model. As a result, readers can identify some reasons for unavailable measurements but cannot reconstruct the analysis set or assess the impact of missing data. No detailed statistical analysis plan is reported in the registry record.

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.

Most frequent non-serious adverse event terms

The registry lists a single safety group, “Participants in HFNC Trial,” with 80 at risk in the serious adverse event category and 0 affected. It lists 80 at risk in the other adverse event category and 48 affected. Deaths are listed as 0, and no top serious adverse event terms are listed. These are the registry’s recorded counts; no percentages or comparative safety estimates are reported.

Serious adverse events and other adverse events are separate registry categories. The zero in the serious adverse event count should be read as no serious adverse events recorded in that category in this listing, not as evidence that no adverse events of any kind occurred. The other-adverse-event count records events in its own category. The record does not provide event descriptions or severity details for those 48 events, nor does it show a comparison of event rates between the respiratory-support conditions.

Because the safety figures are presented for the trial group rather than as condition-specific comparisons, they do not establish whether events were associated with HFNC, nCPAP, or another circumstance. The listing also does not provide timing, exposure duration by condition, or a causal assessment. These counts describe the safety information reported in the registry record and should not be interpreted as a randomized safety comparison.

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

The study asks a focused physiological question: how the percentage of lung regions not engaged in tidal volume varies across nCPAP and HFNC periods in extremely low birth weight infants. It uses EIT and a repeated measurement schedule, including baseline, early HFNC, later HFNC or earlier return after failure, and a later nCPAP period. The record also reports that some participants did not complete the scheduled later-HFNC measurement and that one participant lacked the Period 4 primary measurement.

What the record does not establish is the numerical answer. It does not report the mean endpoint values by period, a contrast between conditions, uncertainty around a contrast, or a statistical test. Without those results, it is not possible to say from the registry record whether the endpoint increased or decreased, how large any change was, or how precisely it was estimated. The completed status and enrollment count do not themselves supply evidence about the direction or magnitude of a physiological difference.

The design also shapes interpretation. Crossover measurements can be useful for describing within-participant changes, but this study is recorded as non-randomized and unmasked. Differences across periods may be influenced by time and sequence as well as by the support condition. The record does not describe an analysis method that accounts for repeated measurements or these potential influences. The failure-related early returns also mean that the scheduled later-HFNC assessment was not available for all participants, and the record does not report sensitivity analyses for this issue.

Accordingly, the registry record can be used to understand the study’s question, protocol outline, endpoint definition, and selected safety counts. It cannot support a conclusion about whether either support condition produced a particular physiological result, nor can it establish a treatment effect. This distinction is important: a registered study design and an outcome definition are not substitutes for posted numerical outcome findings.

13. Why This Trial Matters Statistically

This trial is statistically interesting because it combines repeated physiological measurements with sequential respiratory-support conditions. A crossover structure can provide within-participant information, which is valuable when outcomes vary substantially from one infant to another. At the same time, repeated observations require methods that recognize that measurements from the same participant are related. The record identifies the crossover structure but does not report the analytical method used to account for that dependence.

The scheduled periods also make timing central to interpretation. Measurements occur at baseline, after HFNC begins, during later HFNC, and after a return to nCPAP. These periods differ in both support condition and elapsed time. In general, separating those influences is a design and analysis challenge; it cannot be resolved just by comparing period averages. Non-randomized allocation and the absence of masking, as recorded, add further reasons to interpret any potential contrast cautiously.

The failure-related early returns make the data structure especially relevant. Twenty infants returned to nCPAP before completing the scheduled 6-hour HFNC assessment. That observation is part of the study’s clinical course, but it also means the later-HFNC measurement is not available under the same schedule for everyone. Statistical reporting would need to make clear how such observations contribute to estimates and how missing measurements affect interpretation. The registry record does not provide that analysis.

Finally, the absence of posted numerical outcome results is itself consequential. The record states what was to be measured and when, but readers cannot assess effect size or precision. This distinction illustrates why a completed study record, an endpoint definition, and an analysis result are different kinds of evidence. For statistical interpretation, the measured estimates and their uncertainty are essential.

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 registry record does not report numerical results for the primary endpoint, despite identifying the mean and standard deviation as the intended summary. No condition or period estimates, contrasts, confidence intervals, or p-values are listed. It also reports zero analysis results. Consequently, the primary physiological comparison cannot be quantified or assessed for precision from the record.

The study is non-randomized and unmasked. The record does not describe random assignment of the sequence, a masking procedure, or methods to account for possible sequence, time, or carryover effects. Although the design is crossover, the statistical approach to repeated measurements is not reported. The record also does not report a detailed statistical analysis plan, a formal analysis population, or the handling of multiple outcomes.

Measurement availability differs across the scheduled periods. Twenty infants meeting HFNC failure criteria returned to nCPAP and did not complete the scheduled 6-hour HFNC EIT assessment; one participant lacked the 60-minute return-to-nCPAP EIT measurement in Period 4. The record does not report a missing-data method or sensitivity analyses. Failure-related absence of a scheduled measurement may be particularly important when interpreting the later-HFNC period, because measurement availability is tied to whether the participant remained on HFNC.

The record gives limited detail about the primary endpoint’s operational measurement beyond its EIT basis and the planned schedule. It does not report a full definition of HFNC failure criteria, a complete period-by-period observation count, or numerical outcome distributions. The safety listing does not break adverse events down by respiratory-support condition. These limitations restrict quantitative interpretation and prevent a direct assessment of whether observed differences, if any, are robust to analysis choices.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Physiological Changes With High-Flow Nasal Cannula — NCT03700606.
• Registry record last updated 2026-04-17; status last verified 2022-05.
• Results first posted 2026-04-17.
• 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

NCT03700606, “Physiological Changes With High-Flow Nasal Cannula,” is a completed, non-randomized, unmasked crossover study sponsored by Sharp HealthCare. It enrolled 80 participants and lists respiratory distress syndrome in premature infants as the condition. The study compared physiological measurements across periods involving nCPAP and HFNC in extremely low birth weight infants.

The primary outcome was the percentage of unventilated lung regions assessed by EIT, summarized as a mean with standard deviation. Measurements were planned at baseline on nCPAP, after starting HFNC, during later HFNC or earlier if failure occurred, and after return to nCPAP. Twenty infants returned early after meeting HFNC failure criteria, and one participant lacked the Period 4 primary measurement.

The registry record does not report numerical primary-outcome results or analysis results. It lists 0 serious adverse events, 48 other adverse events, and 0 deaths. The statistical analysis plan, methods for repeated measurements and missing data, and multiplicity approach are not reported in the registry record.