Comparing the Laryngoscopy View Using Miller and Macintosh Laryngoscopy Blades in Infants Under Two Years of Age: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparing the Laryngoscopy View Using Miller and Macintosh Laryngoscopy Blades in Infants Under Two Years of Age (NCT01717872), a randomized na study. The primary outcome is The View of the Larynx (Glottis Opening) as Determined by POGO Score (Percentage of Glottic Opening)..

2. Clinical Question

The trial addresses a practical measurement question: whether the laryngeal view differs between two blade designs when used in infants. The registry summary notes that straight Miller blades are commonly used in pediatric practice in some institutions, and describes differing views about how the blade should be used in relation to the epiglottis. The trial was intended to compare views rather than report a clinical outcome such as successful intubation or a health outcome. Such outcomes are not reported in the primary endpoint description.

The protocol descriptions include photographs made under more than one blade-position condition. The primary endpoint description, however, specifies a comparison between Miller lifting the epiglottis and Macintosh lifting the tongue. The registry does not explain how those conditions map onto all the photographs described for each arm. That distinction matters when interpreting the estimate: it is a comparison of the stated viewing conditions, not necessarily a comparison of every possible way either blade might be used.

3. Trial Design

The registry identifies this as a randomized, single-masked, parallel trial. The trial status is completed. Enrollment is listed as 50, with 25 participants in each arm in the posted outcome data. The listed phase is NA. The registry does not report who was masked, although the arm descriptions specifically state that a blinded assessor evaluated the photographs.

The two arms were the Macintosh laryngoscope blade and the Miller laryngoscope blade. The intervention is listed as a device. For each blade, the descriptions refer to photographs of the larynx with the epiglottis lifted and not lifted. The primary endpoint description specifies the comparison of the Miller blade lifting the epiglottis with the Macintosh blade lifting the tongue. Further details about assignment procedures, sequence concealment, or whether and how the photographed conditions were ordered are not reported in the registry record.

The primary completion date is 2013-07 and the start date is 2012-10. The sponsor is State University of New York at Buffalo. The registry lists Percent of Glottic Opening as the condition and gives three outcome measures and three analyses. It identifies one primary outcome and lists two secondary outcomes concerning POGO scores with the Miller and Macintosh blades under specified lifting conditions.

The design allows a randomized comparison of blade groups, while the photographic assessment by a blinded assessor is intended to reduce the influence of knowledge about the assessment on scoring. The record does not give enough detail to determine exactly which other trial personnel were masked or how randomization was implemented. It also does not report a detailed analysis plan, sample-size justification, or rules for handling protocol deviations. Those omissions limit how fully the conduct and analysis can be reconstructed from 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

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 The View of the Larynx (Glottis Opening) as Determined by POGO Score (Percentage of Glottic Opening)., measured over 30 seconds.

The primary endpoint is the view of the larynx, described as glottis opening and measured with the Percent of Glottic Opening (POGO) score. The unit is percentage of Glottic opening, and the endpoint is recorded at 30 seconds. The registry describes the primary comparison as the POGO score with the Miller blade lifting the epiglottis versus the score with the Macintosh blade lifting the tongue.

POGO is a measure of how much of the glottic opening is visible during laryngoscopy. In general, an airway view measure can capture the visibility of an anatomical target under a particular viewing condition. It is not, by itself, a measure of whether a subsequent procedure succeeds or whether a patient experiences a clinical benefit or harm. The registry does not state that the primary endpoint measures intubation success.

The arm descriptions say that laryngeal photographs were assessed by a blinded assessor. This approach makes the outcome an assessment of the recorded view rather than a direct report of the operator’s judgment alone. The record does not explain the scoring instructions, whether multiple assessors were used, how disagreements were resolved, or how measurement reliability was evaluated. It also does not describe how the 30-second time point was implemented or whether other timing details were standardized.

For continuous measurements, a mean summarizes the central level of the scores, while a confidence interval expresses uncertainty around that estimated mean. Those summaries do not, on their own, quantify the uncertainty in the difference between arms. That distinction is important because the clinical question concerns a comparison, whereas the primary arm-level results are reported separately.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports a continuous primary outcome summarized by a mean and a 95% confidence interval in each arm. A mean represents the average of the measured scores in that group. A confidence interval communicates uncertainty in the estimated mean, conditional on the method used to construct it. It is not a range containing every participant’s score, and it is not a direct interval for the difference between the randomized groups.

What does the reported test compare?

The reported test is the Wilcoxon (Mann–Whitney) test. In general, this is a rank-based method for comparing observations between two independent groups. It orders the observed values and uses their relative ranks, rather than relying solely on a comparison of group means. Depending on the shapes of the group distributions, its result can reflect differences in location or broader distributional patterns. It should not automatically be described as a test of the difference between means.

The registry gives the test result as p > 0.05, but does not report a test statistic, an exact p-value, or a confidence interval for a rank-based effect. It also does not give the full score distributions. Therefore, the reader cannot use the posted summary to assess whether the score distributions had similar shapes or to determine precisely which distributional feature drove the test result. The reported arm means remain useful descriptions, but they do not replace an effect estimate aligned with the reported test.

Why does an effect estimate matter?

An effect estimate describes the size and direction of a comparison on an interpretable scale. A confidence interval around that estimate indicates how precisely it is known. General: a p-value alone does not tell readers how large an effect might be or which values remain plausible. Here, the registry reports arm-specific means and their confidence intervals but no between-arm estimate and no interval for that estimate. The registry’s missing or non-estimable comparison values mean a simple unadjusted recalculation is not available for this continuous two-arm comparison.

General: overlap or non-overlap between separate arm-level confidence intervals is not a substitute for a confidence interval or test for the contrast. The uncertainty in a difference depends on the uncertainty in both groups and on the comparison method. In addition, a mean-based contrast and a rank-based test address related but not identical statistical summaries. The appropriate interpretation should preserve that distinction rather than treating the reported p-value as if it tested the difference between the reported means.

How should the p-value be read?

The p-value is reported only as greater than the stated threshold, not as a more precise value. It does not state the probability that the observed result occurred by chance alone, and it does not establish equivalence. General: a failure to reject a statistical null may arise when an effect is small, when the data are imprecise, or when the test is not sensitive to the particular pattern of differences. The registry does not give enough information to distinguish among those explanations in this trial.

The registry does not report a prespecified margin for equivalence or non-inferiority. Therefore, the result cannot be interpreted as showing that the two blade conditions are sufficiently similar within a clinically acceptable range. Equivalence claims require a specified margin and an analysis designed to assess whether the uncertainty excludes effects outside that margin; neither is reported.

What does the design contribute?

Random allocation is intended, in general, to make baseline characteristics comparable between groups on average and to support an unbiased comparison when the trial is conducted and analyzed appropriately. The registry identifies the allocation as randomized, but does not describe the sequence-generation or concealment procedures. The parallel model means the record describes separate randomized arms, rather than a crossover comparison. The record does not provide further detail about whether there were repeated measurements within participants or how the multiple photographic conditions related to the primary analysis.

The registry labels masking as single and separately says that the assessor was blinded. It does not identify every masked role or provide enough detail to evaluate masking of the operators or other personnel. For an image-assessed endpoint, assessor blinding can reduce the chance that knowledge of the blade influences scoring, but it does not by itself resolve possible differences in image acquisition or other sources of measurement variation. Procedures for image standardization and score reliability are not reported.

General: over-dispersion describes variation greater than a statistical model expects. It can make standard errors too small if the model does not account for that extra variability. The posted results do not report a model-based mean comparison or enough distributional detail to evaluate over-dispersion in a model. The rank-based test avoids some distributional assumptions associated with a conventional mean comparison, but it has its own interpretation and does not provide the missing mean contrast.

What can the sample-size illustrations tell readers?

Any sample-size illustration based on an observed effect is a planning exercise, not observed power and not evidence that the original design had a particular probability of detecting an effect. The stated illustration assumptions describe a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect expressed as a fraction of the observed effect. The registry does not report a design justification or say that these assumptions were used to plan the trial. Because the observed between-arm effect is not reported or estimable in the registry’s comparison data, the planning calculations for this continuous comparison are marked as missing or not estimable. They should not be read as evidence about the adequacy of the trial’s original planning.

10. Analyses Reported by the Sponsor

The registry’s reported analysis is a Wilcoxon (Mann–Whitney) comparison with p > 0.05. The registry does not provide an estimated between-arm effect, an interval for that effect, or numerical comparison values from which an unadjusted recalculation is available. The recalculation is therefore not estimable for this continuous two-arm endpoint.

The sponsor’s reported test is the only posted inferential comparison that can be described. The arm-level results are means with 95% confidence intervals: 84.0 (76.0 to 92.0) for the Macintosh blade and 78.0 (70.0 to 87.0) for the Miller blade. They should not be presented as a recalculated contrast, since they summarize each arm separately. The reported rank-based test also should not be described as a direct test of the difference between those means.

The registry does not report enough detail to reproduce the sponsor’s analysis, including the underlying observations, test statistic, exact p-value, or a confidence interval for a comparison effect. No discrepancy between a reported estimate and a recalculation is identified, because there is no recalculated comparison estimate to assess. The sponsor result can be reported as posted, with the limitation that its precision and effect magnitude cannot be determined from the registry record.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and two secondary outcomes, for three outcome measures. It also lists three analyses, but the posted information describes a Wilcoxon (Mann–Whitney) comparison for the primary outcome. The registry does not provide results or analysis details for the secondary outcomes in the registry record for this record. It is therefore not possible to determine how those outcomes were analyzed or what their results were.

When a trial examines multiple outcomes or makes multiple statistical comparisons, the chance of obtaining at least one apparently noteworthy result can increase. General: multiplicity procedures can be used to control that risk, or analyses can be organized into a prespecified hierarchy. The registry record does not report whether any such procedure or hierarchy was used. It also does not describe whether the reported test was prespecified, whether there were other unreported analyses, or whether any interim analyses took place.

The registry does not report how missing endpoint observations were handled, whether any participants were excluded from the analysis, or which analysis population was used. The arm descriptions and posted group sizes do not establish whether every randomized participant contributed an evaluable score. Missing-data methods, reasons for missingness, protocol deviations, and the handling of measurements that could not be scored are not reported in the registry record.

These omissions affect interpretation because the analysis population and missing-data approach can influence a trial estimate. Without participant-level accounting, readers cannot determine whether the outcome summaries include all randomized participants or a subset, nor can they evaluate whether missing observations might be related to blade condition or score. The registry’s stated total enrollment and per-arm results do not fill in those details.

The record also does not report a statistical analysis plan, prespecified clinically important difference, or adjustment for baseline characteristics. No inference should be made about whether these methods were or were not used outside the information recorded in the registry. The available evidence supports reporting the listed primary result and noting that the multiplicity, missing-data, and analysis-population procedures are not reported.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

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

The registry reports zero serious adverse events among the 25 participants at risk in the Macintosh arm and zero among the 25 participants at risk in the Miller arm. It also reports zero other adverse events among the participants at risk in each arm. These are the adverse-event counts recorded in the registry; they are not estimates of the risk of every possible harm.

The deaths field is not populated for either arm, so death counts are not reported in the registry record. The record does not specify adverse-event definitions, collection procedures, follow-up duration, or whether events were independently assessed. It also does not provide event-level descriptions because no serious adverse-event terms are listed.

General: an observed count of zero does not demonstrate that an event is impossible or that there is no safety risk. Interpretation depends on how events were sought, the period of observation, and the completeness of reporting. Those details are not reported in the registry record. The figures should therefore be read narrowly as the adverse-event counts posted for the recorded trial arms, not as proof of absence of harm or as a comparison establishing a safety difference.

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

The result describes a randomized comparison of laryngoscopy views measured by POGO score in infants under two years of age. The registry reports arm-level means and confidence intervals, alongside a Wilcoxon (Mann–Whitney) test reported as p > 0.05. In statistical terms, the posted comparison does not meet the reported significance threshold. The registry does not provide a between-arm effect estimate or an interval that would show the precision of such an estimate.

This result does not establish that the blade conditions produce identical views. A non-significant test is not an equivalence result, and no equivalence margin is reported. It also does not establish that one blade is superior, nor does the registry provide the effect estimate needed to quantify the magnitude of a contrast. The separate arm means are descriptive; the test is rank-based and should not be presented as a test of the difference between those means.

The endpoint measures the visible glottic opening under the conditions described in the record. It does not, as reported, establish differences in intubation success, complications, patient outcomes, or the broader performance of the devices in other settings. The primary comparison is specifically described as Miller lifting the epiglottis versus Macintosh lifting the tongue. The record’s broader arm descriptions mention additional photographic conditions, but do not explain how each entered the primary analysis.

General: statistical evidence is most informative when readers can assess both the estimated effect and its uncertainty, alongside the study design and measurement process. The absent between-arm estimate, missing analysis details, and unreported handling of missing data limit that assessment. The posted outcome should therefore be interpreted as a report of arm summaries and a non-significant rank-based comparison, rather than a definitive demonstration of similarity or a quantified treatment effect.

14. Why This Trial Matters Statistically

This trial raises a useful statistical issue: a study can report detailed summaries for each arm while leaving the central comparative quantity unavailable. For a randomized trial, the main scientific question is generally about the contrast between assigned groups. Arm means and their confidence intervals describe each group, but they do not show the uncertainty around the contrast. Reporting an effect estimate with an interval would make the direction, magnitude, and precision of the comparison easier to assess.

The choice of method also matters. The registry reports a Wilcoxon (Mann–Whitney) test for a continuous score but gives means as the arm summaries. Those methods answer related but distinct questions. Clear reporting would explain why the rank-based test was selected, how its result should be interpreted alongside the means, and whether an effect estimate aligned with that test was calculated. The record does not provide those explanations or the numerical comparison result needed for independent interpretation.

The endpoint is an assessor-rated view from photographs, and the descriptions emphasize assessor blinding. This makes measurement procedures important: consistent image acquisition, scoring instructions, and information on scoring reliability would help readers understand how much of the observed variation reflects the underlying view versus measurement. These details are not reported in the registry record.

The case also illustrates why a non-significant p-value should not be treated as evidence of equivalence. Without a contrast estimate, precision interval, and prespecified equivalence margin, readers cannot judge whether effects of meaningful size remain compatible with the findings. The trial is statistically informative as a registry record of a randomized comparison, but the sparse comparative reporting constrains the conclusions that can be drawn from the result.

15. Limitations

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

The principal limitation for statistical interpretation is that the registry does not report a between-arm effect estimate or confidence interval for the primary comparison. The arm means and their 95% confidence intervals do not quantify the contrast directly. The reported p-value is given only as p > 0.05, without a test statistic or exact value. The registry also does not provide underlying observations or distributional summaries that would allow an independent assessment of the rank-based result.

The primary endpoint description and arm descriptions are not fully aligned in their wording. The primary comparison specifies Miller lifting the epiglottis versus Macintosh lifting the tongue, while each arm description refers to photographs with the epiglottis lifted and not lifted. The registry does not explain how the photographic conditions were incorporated into the primary comparison. This leaves uncertainty about the precise operational definition of the analyzed contrast.

Further details needed to assess conduct and analysis are not reported in the registry record. These include randomization procedures, who was masked beyond the blinded assessor, the analysis population, missing-data handling, protocol deviations, and a statistical analysis plan. The record does not report a sample-size justification, interim analyses, a clinically important difference, or multiplicity handling. The listed secondary outcomes do not have accompanying results in the information recorded.

Measurement limitations are also relevant. The record does not describe score training, image standardization, assessor reliability, or resolution of differing assessments. Safety interpretation is limited by absent details on adverse-event definitions, ascertainment, and follow-up, and death counts are not reported. The findings should consequently be read as the specific posted summaries and test result, with substantial uncertainty about comparative magnitude and analysis details.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Comparing the Laryngoscopy View Using Miller and Macintosh Laryngoscopy Blades in Infants Under Two Years of Age — NCT01717872.
• Registry record last updated 2017-03-20; status last verified 2017-02.
• Results first posted 2016-04-11.
• 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.

19. Record Summary

This completed randomized, single-masked, parallel trial compared Macintosh and Miller laryngoscope blades in infants from one day to under two years of age. The registry lists enrollment of 50 and reports 25 participants per arm. The primary endpoint was POGO score at 30 seconds, described as a comparison of Miller lifting the epiglottis with Macintosh lifting the tongue. Photographs were assessed by a blinded assessor.

The posted mean POGO score was 84.0 (95% confidence interval, 76.0 to 92.0) for the Macintosh arm and 78.0 (95% confidence interval, 70.0 to 87.0) for the Miller arm. The reported Wilcoxon (Mann–Whitney) comparison was p > 0.05. A between-arm effect estimate and confidence interval are not reported, and recalculation is not estimable from the registry information. The record lists two secondary outcomes but does not provide their results.

The registry reports zero serious and zero other adverse events in each arm; deaths are not reported. Details on missing data, analysis populations, multiplicity, and the statistical plan are not reported in the registry record.