Pediatric Cardiac Output Monitoring Observational Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pediatric Cardiac Output Monitoring Observational Study (NCT04465370), a registered phase not stated study. The primary outcome is Bias of FloTrac Arterial Pressure Cardiac Output (L/Min) Against Swan Ganz Invasive Cardiac Output.

2. Clinical Question

3. Trial Design

This was a prospective, single-arm, nonrandomized, observational study. Its stated purpose was to assess cardiac-output monitoring in pediatric subjects by comparing FloTrac and ClearSight systems with intermittent thermodilution Swan-Ganz measurements. The registry describes the study as completed and lists enrollment of 89. The sponsor is Edwards Lifesciences.

Because the study was observational and nonrandomized, its central question is measurement comparison rather than the effect of assigning participants to competing treatments. The record does not describe allocation to different treatment arms. It also does not establish that the comparison groups were independent or explain how the groups were formed. The entry that describes 60 evaluable pivotal subjects in each comparison group follows application of predetermined exclusions for hemodynamic instability or poor signal quality; how those group counts relate to overall enrollment is not explained in the registry record.

The primary measurement was collected during cardiac catheterization, with the timeframe described as an average of 2 hours. Four outcome measures are listed, but the registry record does not provide their individual results in the information recorded for this analysis. No analysis records are listed. That does not establish that no analysis was performed; it means that analysis details and numerical findings are not available in the registry record.

The record gives a study start date of 2020-09-02 and a primary completion date of 2023-01-05. These dates and the completed status describe the study’s registry timeline, but they do not supply the results needed to judge how closely the monitoring methods agreed.

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: After applying the pre-determined data exclusion criteria for hemodynamic instability or poor signal quality, there were 60 Evaluable Pivotal Subjects in each of the comparison groups.

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 Bias of FloTrac Arterial Pressure Cardiac Output (L/Min) Against Swan Ganz Invasive Cardiac Output, measured over Duration of cardiac catheterization procedure, an average of 2 hours.

The primary endpoint is the mean bias of FloTrac arterial pressure cardiac output (APCO) against Swan-Ganz invasive cardiac output (ICO), expressed in L/min. The record says that bias was calculated using Bland-Altman analysis and reports the endpoint’s dispersion measure as a 95% confidence interval. The measurement period is cardiac catheterization, described as averaging 2 hours.

In general, bias summarizes the average difference between two methods measuring the same quantity. A mean bias near zero would indicate little average difference, but would not by itself show that individual paired readings are close to one another. Bland-Altman methods are used to assess agreement between measurement methods, not simply whether their averages are associated. In general, assessment of individual agreement may also consider the spread of paired differences, often expressed through limits of agreement. Such limits are not stated in the registry record.

A confidence interval describes uncertainty around an estimated quantity. For this endpoint, it would indicate the precision of the estimated mean bias, not the range of differences expected for individual measurements. The registry record does not report the mean bias or its 95% confidence interval. Consequently, neither the average discrepancy nor the precision of that estimate can be interpreted from the endpoint entry alone.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does mean bias describe?

For a continuous measurement, bias is based on the difference between readings from two methods. The mean bias summarizes the average of those differences across the measurements included in an analysis. It answers a question about average systematic separation: whether one method tends, on average, to give readings above or below the other. Here the comparison is FloTrac APCO against Swan-Ganz ICO, and the unit is L/min.

In general, an average difference is only one part of measurement agreement. Readings could have a small average difference while individual paired readings vary considerably in opposite directions. Mean bias alone therefore does not establish that the methods can be used interchangeably for individual measurements. The registry describes a Bland-Altman analysis, a general approach for examining differences between measurement methods. It does not report limits of agreement or another measure of the spread of individual differences.

What does the confidence interval mean?

The endpoint lists a 95% confidence interval as its dispersion measure. In general, a confidence interval conveys uncertainty around an estimate under the assumptions of the analysis. For mean bias, it concerns the precision of the estimated average difference. It is not a range intended to contain a stated proportion of individual paired differences. A narrow interval would indicate greater precision for the average estimate than a wider interval, but neither the interval nor the estimate is reported in the registry record. No conclusion about precision can therefore be made.

The record does not state whether the confidence interval was adjusted for any design features, repeated measurements, or other sources of dependence. It also does not give a confidence-interval method or a test result. Those details are not reported in the registry record and should not be inferred from the endpoint label alone.

Why can exclusions affect the interpretation?

The endpoint description says that predetermined criteria were applied to exclude data for hemodynamic instability or poor signal quality. In general, excluding measurements that cannot be reliably obtained may be appropriate for a method-comparison analysis, but the resulting estimate applies to the measurements retained under those criteria. If excluded observations differ systematically from retained ones, the analyzed population may not represent all enrolled participants or all measurement conditions. The record does not state how many observations were excluded for each reason or how exclusions affected the endpoint estimate.

How should planning calculations be read?

Any sample-size figures associated with planning illustrations are not observed power. The stated illustration assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is defined as a stated fraction of the observed effect. These are assumptions for hypothetical planning calculations, not evidence that the trial used that design or that its sample size was justified by those assumptions. The registry record does not report the trial’s own sample-size justification, the observed effect needed for those calculations, or numerical planning results in the endpoint information.

The stated planning convention distinguishes time-to-event calculations, which give a total number of events using the Schoenfeld method, from other calculations, which give participants per arm. That convention describes how planning quantities are expressed; it does not establish that this observational measurement study had two randomized arms. No trial-specific sample-size plan is reported in the registry record.

What can be concluded from the analysis entry?

The registry record states the intended endpoint, unit, method, timeframe, and a description of the evaluable population. It does not give the endpoint estimate, confidence interval, limits of agreement, or a reported statistical test. Accordingly, the available method description identifies how the primary question was framed, while leaving the numerical answer and its precision unavailable.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists four outcome measures, but the record does not describe how the outcomes were prioritized beyond identifying the primary endpoint. It does not report whether any adjustment for multiple comparisons was planned or used. In general, examining multiple outcomes can increase the chance of observing an apparently notable result by chance if each is tested separately. Whether that concern applies, and how it was handled, cannot be determined from the registry record.

No analysis records are listed. The registry record therefore does not provide an analysis plan, test results, or details about how the primary endpoint and other outcomes were analyzed. The absence of those entries does not establish that no analysis occurred; it means that the details are not reported in the registry record.

The endpoint description says that data were excluded under predetermined criteria for hemodynamic instability or poor signal quality. It does not report the number of exclusions, the number of missing measurements, how missingness was defined, or how missing data were handled. It also does not explain whether all eligible observations were included in the primary analysis or whether any additional analysis populations were used.

The record reports 89 enrolled participants and 60 evaluable pivotal subjects in each comparison group after application of the exclusion criteria. It does not explain the relationship between those counts. The number of participants included in each analysis, and the handling of repeated or incomplete measurements, are not reported in the registry record. These omissions limit assessment of how the analyzed observations relate to enrollment and whether missingness or exclusions could affect the result.

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 serious adverse event terms

Serious events with the highest arm-level rate: other (2.2%), admission for atg treatment for rejection (1.1%), transplant rejection amr grade 2 (1.1%), cardiac (1.1%).

For the treatment population, the registry lists 89 participants at risk for serious adverse events and 5 affected. It lists 89 at risk for other adverse events and 0 affected. The recorded number of deaths is 0. These are registry counts; the entry does not provide a comparison group or a basis for attributing events to a monitoring device.

The serious adverse-event terms listed include “Other” at 2.2%, “Admission For Atg Treatment For Rejection” at 1.1%, “Transplant Rejection Amr Grade 2” at 1.1%, and “Cardiac” at 1.1%. The record does not clarify the relationship among these terms, whether categories overlap, or the detailed event narratives. They should therefore be read as the registry’s listed terms and percentages, not as proof of cause or as a complete clinical account.

In general, adverse-event counts describe events recorded within the defined population and observation period. They do not, by themselves, establish that an event was caused by the device or study procedures. The registry entry does not report an attribution assessment or additional detail needed to interpret causality. The counts also do not show how risks compare with another monitoring approach, since the record does not describe a randomized control group.

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

The study addresses a measurement-comparison question in pediatric patients undergoing cardiac catheterization. It focuses on whether FloTrac APCO measurements differ, on average, from invasive Swan-Ganz ICO measurements obtained by intermittent thermodilution. The registry describes a prospective, single-arm, nonrandomized, observational design and identifies mean bias with a 95% confidence interval as the primary endpoint.

The record does not report the endpoint estimate or confidence interval. It therefore does not establish the direction or size of any average difference, the precision of that estimate, or the degree of agreement between individual readings. A stated method and endpoint are not substitutes for the observed result. The listed adverse-event counts likewise do not establish whether events were caused by the monitoring methods.

In general, measurement agreement is different from association and from clinical benefit. Two methods can move together while still differing systematically, and a small average bias would not necessarily mean that every paired measurement is close. A fuller assessment would require the numerical mean bias and confidence interval, along with information about the spread of individual differences. Those numerical details are not reported in the registry record.

The endpoint population also requires caution. The record states that 60 evaluable pivotal subjects were in each comparison group after exclusions, while overall enrollment was 89. It does not explain how the group counts relate to enrollment. The stated exclusions for instability or poor signal quality further mean that the endpoint is defined for retained evaluable data, but the number excluded and the implications for representativeness are not reported.

Accordingly, the registry record supports describing the research question, design, and recorded safety counts. It does not support a conclusion that one monitoring method is better, worse, interchangeable, or clinically beneficial. Those conclusions require results and analysis details that are not reported in the registry record.

13. Why This Trial Matters Statistically

Comparing a monitoring method with an invasive reference measurement is a distinct statistical problem from comparing treatments. The question is not merely whether readings are related, but whether the two methods give sufficiently similar values for the intended measurement context. The study’s use of mean bias and Bland-Altman analysis reflects that focus on differences between methods.

The clinical setting makes the definition of the analyzed population important. The endpoint description restricts the evaluable data through exclusions for hemodynamic instability or poor signal quality. Such criteria affect which readings contribute to the estimate and, in general, can affect how broadly an agreement result applies. The record does not report how many observations were excluded or how retained and excluded data differed.

The study is also a reminder that a method-comparison endpoint needs more than an average difference to characterize agreement fully. Mean bias describes average separation; the spread of individual differences is also relevant to understanding how readings compare from one observation to another. The registry record does not report limits of agreement or the endpoint’s numerical results, so the intended statistical question cannot be answered from the recorded endpoint entry.

Finally, the study’s observational, single-arm design should be kept distinct from a randomized treatment comparison. Its stated objective concerns monitoring measurements in pediatric subjects, not a randomized estimate of treatment effect. The registry record identifies a relevant measurement question and the planned summary, but the absence of numerical findings and analysis details limits what can be concluded about its answer.

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.

Several limitations specific to the registry record constrain interpretation. First, it describes a single-arm, nonrandomized, observational study and does not state an allocation scheme. This design is suited to the stated measurement-comparison objective, but it does not provide a randomized treatment comparison.

Second, the endpoint description applies exclusions for hemodynamic instability or poor signal quality. The criteria are named, but the number of affected observations, the details of their application, and the consequences for the analysis are not reported. The resulting endpoint population may therefore be difficult to compare with the full enrolled population.

Third, enrollment is listed as 89, while the primary endpoint description states that 60 evaluable pivotal subjects were in each comparison group. The registry record does not explain the relationship between those counts. It also does not specify whether the comparison groups consist of separate participants or how repeated measurements were handled.

Fourth, the registry record does not report the primary endpoint estimate, confidence interval, limits of agreement, or statistical analysis details. The study’s stated measurement timeframe is cardiac catheterization, averaging 2 hours; results beyond that context are not described. The population is pediatric patients with Pediatric ALL, and the stated age range is 12 to 18 years, which defines the population addressed by the study summary rather than establishing applicability to other groups.

The record names Edwards Lifesciences as sponsor. It does not provide further information that would support an assessment of how sponsorship affected study conduct or reporting.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Pediatric Cardiac Output Monitoring Observational Study — NCT04465370.
• Registry record last updated 2024-11-18; status last verified 2024-09.
• Results first posted 2024-11-18.
• 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

NCT04465370 is listed as a completed, prospective, single-arm, nonrandomized, observational study of cardiac-output monitoring in pediatric patients. The registry lists Pediatric ALL as the condition, an age range of 12 to 18 years in the study summary, enrollment of 89, and Edwards Lifesciences as sponsor. The recorded start date is 2020-09-02, and the primary completion date is 2023-01-05.

The primary endpoint is the mean bias of FloTrac APCO against Swan-Ganz ICO, in L/min, calculated using Bland-Altman analysis with a 95% confidence interval. Its timeframe is cardiac catheterization, described as averaging 2 hours. After exclusions for hemodynamic instability or poor signal quality, the record describes 60 evaluable pivotal subjects in each comparison group. The numerical primary result and analysis details are not reported in the registry record.