Comparison Between NICOM-Cheetah and Flotrac-Vigileo for Cardiac Output Monitoring.: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparison Between NICOM-Cheetah and Flotrac-Vigileo for Cardiac Output Monitoring. (NCT03394768), a registered phase not stated study. The primary outcome is Degree of Disagreement, %.

2. Clinical Question

The study is described as observational, not as a randomized comparison. The arm descriptions refer to the same patient population and indicate that both monitors were used: FloTrac was connected to the arterial catheter, while NICOM Cheetah electrodes were placed on the thorax. Thus, the question is about agreement between device readings, rather than whether one assigned treatment improves an outcome compared with another.

The record lists an enrollment of 26. It does not report detailed eligibility criteria, how participants were recruited, or whether every enrolled participant contributed a complete set of readings. It also does not report how the device readings were synchronized or how the primary endpoint would be summarized. These omissions limit how fully the clinical question and measurement process can be reconstructed from the registry record.

3. Trial Design

This was an observational device-comparison study registered as NCT03394768, with Changi General Hospital listed as sponsor. Its official title describes a comparison of NICOM Cheetah and FloTrac-Vigileo for cardiac output monitoring in the surgical intensive care unit. The study record lists 26 participants and gives the condition as Cardiac Output, Low.

The record describes FloTrac as connected to an arterial catheter as part of usual practice in the hospital’s surgical intensive care unit. NICOM Cheetah uses electrodes placed on the anterior thoracic wall. The arm descriptions refer to the same patient population and indicate use of both devices, rather than two separate groups receiving different treatments. No additional intervention was performed, and care followed department protocols. The detailed operational procedures are not reported in the registry record.

The registry lists a start date of 2018-08-10 and a primary completion date of 2021-01-05. Its status is TERMINATED. The reason for termination, whether the intended enrollment was reached, and whether study procedures were completed for all enrolled participants are not reported in the registry record. The record does not explain whether termination affected the planned collection or analysis of measurements.

The primary endpoint was the percentage degree of disagreement between the devices, assessed using Bland-Altman analysis. Data collection was specified as 10 hours for each participant. The registry lists one outcome measure and no analysis records. It does not report random allocation, a distinct control group, a statistical analysis plan, sample-size justification, or details of how readings were selected and analyzed. Consequently, the study is best understood as an observational agreement assessment whose numerical primary result is 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

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 Degree of Disagreement, %, measured over 10 hours of data collected for each participant.

The primary endpoint is titled Degree of Disagreement, %. It concerns how closely cardiac output readings from NICOM Cheetah and FloTrac correspond, with disagreement expressed as a percentage. The registry defines cardiac output as the amount of blood the heart pumps through the circulatory system in a minute. Each participant was to contribute 10 hours of data collection.

Agreement is a different question from whether two measurements are correlated. Two devices can move in the same direction across patients yet still differ systematically in their readings. Bland-Altman analysis is generally used to assess the distribution of differences between measurement methods, rather than merely their association. It can help describe whether one method tends to read higher or lower and how widely individual paired readings differ. Those are general features of the method; the registry record does not state which Bland-Altman summaries were planned or report their values.

The percentage scale is intended to express the size of disagreement in relation to cardiac output measurements. The record does not explain the precise formula, the averaging or pairing procedure, or the threshold that would count as acceptable agreement. Without those specifications and the endpoint estimate, the amount and precision of device disagreement cannot be judged from the registry record. The endpoint is therefore clinically relevant to measurement comparison, but its numerical result and interpretation are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical aim is measurement agreement: assessing whether two devices produce sufficiently similar cardiac output readings when used in the study setting. This differs from testing whether a treatment changes a clinical outcome. The registry names Bland-Altman analysis and expresses the primary endpoint as a percentage degree of disagreement. It does not provide the numeric analysis result or enough procedural detail to reproduce the analysis.

What does Bland-Altman analysis assess?

As a general statistical method, Bland-Altman analysis compares paired measurements from two methods. It focuses on the differences between those measurements and considers how the differences vary with the magnitude of the measured quantity. A typical interpretation distinguishes average disagreement from the spread of individual differences. The average can indicate a systematic tendency for one method to read higher or lower; the spread indicates how much individual paired values vary around that tendency. The registry record does not state which summaries were calculated, how they were presented, or what agreement criterion was selected.

Correlation alone would not answer the same question. A high correlation can occur when two methods track changes together but differ consistently in scale or level. Conversely, an agreement analysis directly examines the size and distribution of paired differences. This is a general distinction between association and agreement, not a claim about the observed NICOM Cheetah and FloTrac readings. Their numerical relationship is not reported in the registry record.

Why does the measurement structure matter?

The arm descriptions indicate that the same patient population had FloTrac connected to an arterial catheter and NICOM Cheetah electrodes placed on the thorax. That description is consistent with measurements from both devices in the same participants, but the detailed measurement protocol is not reported. In general, paired measurements are not independent: readings from the same participant can be more alike than readings from different participants. If multiple readings were collected during the specified 10 hours, those repeated observations could also be related within a participant. The registry record does not report the number or timing of readings, whether they were paired at matching times, or how within-participant dependence was handled.

These design details affect precision. Treating multiple readings from one person as if each came from an unrelated person can make an analysis appear more precise than it is. General approaches to repeated measurements can account for clustering within participants, but the registry record does not identify an approach or state whether repeated readings were analyzed individually or summarized first. No particular method should therefore be attributed to this study beyond the stated use of Bland-Altman analysis.

What does a percentage disagreement measure?

Expressing disagreement as a percentage can put differences in relation to the scale of cardiac output rather than leaving them only in the original measurement units. The exact percentage formula matters: different choices for the denominator or for handling direction can produce different interpretations. The registry record gives the endpoint title and unit but does not specify the formula. It also does not report whether the percentage represents a signed difference, an absolute difference, or another summary. Those definitions are necessary for interpreting the direction and magnitude of the endpoint.

In general, measurement agreement should be interpreted against a contextually justified acceptable range. A numerical estimate is not self-interpreting: its relevance depends on the expected use of the measurements and the consequences of disagreement. The registry record does not report an acceptable-agreement threshold or a rationale for one. It also does not provide an uncertainty interval. Without a point estimate, an interval, and a defined criterion, the practical meaning of the endpoint cannot be assessed from the registered results.

How should planning calculations be read?

The planning illustration is described as based on the observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is expressed as a stated fraction of the observed effect. These are planning assumptions, not evidence that the study used those assumptions in its design. The record does not report the trial’s own sample-size justification or protocol rationale.

Such calculations are hypothetical illustrations of the number of participants per arm associated with a specified power under the stated assumptions. They are not the study’s observed power and do not replace the actual analysis. The record describes the calculations as using an observed effect, but it does not report a numerical primary endpoint result in the registry record. The planning figures therefore cannot supply the missing agreement estimate or show whether the observed study data were precise.

The registry lists no analysis records. It does not report model specifications, confidence intervals, hypothesis tests, handling of repeated measurements, or sensitivity analyses. General statistical concepts can clarify what an agreement analysis would address, but they cannot fill in the study-specific methods or results that are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure: percentage degree of disagreement between NICOM Cheetah and FloTrac. It does not list a set of secondary outcomes or report multiple analysis records. A single listed primary outcome can reduce concern about choosing among many competing primary tests, but it does not establish that no additional analyses were performed. The analysis plan and any prespecified hierarchy of outcomes are not reported in the registry record.

Multiplicity can arise when many endpoints, subgroups, time points, or statistical comparisons are examined. As a general matter, the more results considered, the greater the chance that a seemingly notable result arises by chance if no adjustment or prespecified interpretation is used. The registry record does not describe a multiplicity strategy, subgroup plan, or interim analysis. It is therefore not possible to determine how such issues were addressed, if they arose.

Missing-data handling is also not reported. The registry specifies 10 hours of data collection for each participant, but does not state how much data were collected, how many participants had complete measurements, how gaps or failed device readings were handled, or whether any observations were excluded. The enrollment of 26 should not be interpreted as confirmation that 26 participants contributed complete paired data.

The analysis population is not described. The record does not say whether the analysis included all enrolled participants, only participants with usable paired measurements, or another defined group. It does not report a protocol for handling withdrawals or incomplete measurement periods. These omissions leave uncertainty about both the amount of missing information and the population represented by any analysis; no numerical endpoint result 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.

The registry’s safety figures list 26 participants at risk for serious adverse events and 0 affected. For other adverse events, it also lists 26 at risk and 0 affected. The death count is 0. These are the recorded figures for all participants; the record does not provide event terms or a separate safety breakdown by device.

Zero recorded events is a description of the registry entry, not proof that no events occurred in every circumstance or that the monitoring devices have no risks. The record does not explain how adverse events were collected, how long safety follow-up lasted, or whether all participants completed the relevant observation period. The 10-hour endpoint collection timeframe should not be assumed to define the safety follow-up period.

The study is observational and does not describe distinct randomized device groups. The all-participant figures therefore do not provide a comparative estimate of safety between the monitors. They also do not establish that any event, had one been recorded, was caused by either device. The registry reports 0 serious events, 0 other events, and 0 deaths among the listed 26 at-risk participants in each adverse-event category; interpretation beyond those counts is not reported in the registry record.

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

The registry record establishes that an observational study was registered to compare cardiac output readings from NICOM Cheetah and FloTrac in a surgical intensive care unit. It identifies the primary measure as percentage disagreement assessed by Bland-Altman analysis, specifies 10 hours of data collection per participant, and lists an enrollment of 26. These details define the intended question and broad measurement context.

The record does not report the primary endpoint estimate, its uncertainty, or a statistical test. It therefore does not show whether the devices produced similar readings, the size or direction of any disagreement, or how precisely agreement was estimated. The listed absence of analysis records reinforces that the registry entry does not provide a numerical answer to its primary question.

The study’s observational design and arm descriptions matter for interpretation. Both monitors are described in relation to the same patient population, rather than as separate randomized treatment groups. The result of interest would be agreement between device measurements, not evidence that one device improves health outcomes. The record does not report detailed measurement pairing or the statistical handling of repeated readings.

The status is TERMINATED, but no reason is given. The status alone does not establish whether data collection stopped early, whether the endpoint was analyzed, or why the record has no numerical result. The registry also does not report the intended sample-size justification, missing-data procedures, or analysis population.

Accordingly, the record supports a description of the study objective and design, but not a conclusion about the observed performance of either monitor. The zero adverse-event counts are descriptive registry figures and do not establish comparative safety. Any planning illustrations are hypothetical calculations under stated assumptions, not observed power and not substitutes for reported endpoint results.

13. Why This Trial Matters Statistically

Agreement between measurement devices is a statistical question with direct relevance to how readings can be interpreted. For this study, the central issue is not whether NICOM Cheetah and FloTrac are associated across observations, but how much their cardiac output readings differ. Bland-Altman analysis is named in the registry for that purpose, making the distinction between correlation and agreement especially important.

The observational setting also highlights the role of measurement design. The arm descriptions refer to both monitors being used in the same patient population, while FloTrac is identified as the standard-of-care reference in the surgical intensive care unit. Paired measurements can support a direct assessment of device differences, but the timing, frequency, and handling of repeated readings influence how such comparisons should be analyzed. Those details are not reported in the registry record.

The record illustrates why a clearly stated endpoint is not enough to interpret a study statistically. Readers also need the estimate, its precision, details about the observations contributing to it, and an explanation of how missing or repeated measurements were handled. Here, the registry identifies the intended endpoint and lists enrollment of 26, but it does not report the numerical result or analysis records.

The terminated status further makes transparent reporting of study completion and analysis important. The reason for termination and its relationship to data collection are not reported. Without those details, the registered question remains recognizable, but the degree to which the collected data answer it cannot be assessed. This is a useful example of how registry design information and posted outcome results serve different purposes: the former describes what was intended, while the latter is needed to evaluate what the data show.

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 principal limitation is that the registry record does not report a numerical result for the primary endpoint. Without the degree-of-disagreement estimate or its uncertainty, the central measurement-agreement question cannot be answered from the record. The record also does not specify an acceptable-agreement threshold, so even a numerical estimate would require context to interpret.

The analysis methods are sparsely described. Bland-Altman analysis is named, but the percentage formula, specific summaries, confidence intervals, handling of repeated readings, and any sensitivity analyses are not reported in the registry record. Although 10 hours of data collection are specified for each participant, the number and timing of readings and how paired readings were matched are not stated.

Enrollment is listed as 26, but the number included in any endpoint analysis and the completeness of measurement data are not reported. The analysis population, exclusion rules, and missing-data procedures are also not reported in the registry record. This limits assessment of how representative the analyzed observations would be of the enrolled participants.

The study is observational, with no randomized allocation stated. Its descriptions indicate use of both devices in the same patient population, but detailed procedures are not available, and the registry explicitly cautions that the arms are not a like-for-like control comparison. The TERMINATED status is listed without a reason, and the record does not explain whether termination affected data collection or analysis. The sample-size justification and analysis plan are likewise not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Comparison Between NICOM-Cheetah and Flotrac-Vigileo for Cardiac Output Monitoring. — NCT03394768.
• Registry record last updated 2021-05-11; status last verified 2021-05.
• Results first posted 2021-03-03.
• 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

NCT03394768 is an observational study sponsored by Changi General Hospital, titled as a comparison of NICOM Cheetah and FloTrac-Vigileo for cardiac output monitoring in the surgical intensive care unit. The registry lists 26 participants, a start date of 2018-08-10, a primary completion date of 2021-01-05, and status TERMINATED.

The primary outcome is degree of disagreement, expressed as a percentage, between the devices, assessed using Bland-Altman analysis. Data collection was specified for 10 hours per participant. The arm descriptions indicate use of both monitors in the same patient population, with FloTrac described as the reference monitor and standard of care in the unit.

The registry lists no analysis records and does not report a numerical primary outcome result. It reports 0 serious adverse events, 0 other adverse events, and 0 deaths among the listed 26 participants at risk. The reason for termination, detailed analysis procedures, and handling of incomplete measurements are not reported in the registry record.