This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Calibration and Validation of Masimo's O3 Regional Oximetry Device in Neonates, Infants and Children (NCT03123354), a na na study. The primary outcome is Accuracy of Sensor by Arms Calculation of Percent rSO2.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03123354 |
| Phase | NA |
| Condition | Surgery |
| Allocation | Not applicable (single group) |
| Design model | Single Group |
| Masking | None (open label) |
| Enrollment | 44 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Masimo Corporation |
| Results posted | Yes |
2. Clinical Question
The registry describes a device-calibration and validation study in neonates, infants and children undergoing cardiac catheterization. The central question is whether regional oxygen saturation measured by the O3 sensor is close to a cerebral oxygen saturation reference derived from arterial and jugular venous blood oxygen saturation.
Population
Neonates, infants and children scheduled for a general cardiac catheterization procedure. The registry lists enrollment of 44, but does not report the analyzed population or participant characteristics.
Intervention
All subjects were assigned to the test group and received an O3 regional oximeter sensor during the procedure. The study is recorded as single-group and unmasked.
Comparator
The endpoint description identifies a blood-based cerebral oxygen saturation reference, constructed from arterial and jugular venous blood oxygen saturation. There is no separate control group in the registry record.
Primary question
How accurately did sensor rSO2 measurements track the blood-based reference, as quantified by the root-mean-squared error of bias? The registry names this endpoint but does not report its numerical result.
This is therefore a measurement-agreement question, not a comparison of clinical outcomes between treatment groups. The data description indicates repeated or continuous sensor measurements during one visit, with a timeframe of up to 4 hours. The record does not report how many readings each participant contributed, how readings were paired with reference samples, or how participant-level measurements were combined for the endpoint.
3. Trial Design
The registry identifies NCT03123354 as a completed, single-group device study sponsored by Masimo Corporation. Its title describes calibration and validation of the O3 regional oximetry device in neonates, infants and children. The study record lists enrollment of 44 and one outcome measure. It records a start date of 2015-05-01 and a primary completion date of 2017-03-13.
All subjects were enrolled in the test group and received an O3 regional oximeter sensor during their scheduled, general cardiac catheterization procedure. Allocation is recorded as NA, masking as none, and the model as single group. Accordingly, the design does not provide a randomized or masked comparison between groups. The registry itself cautions that its arms are not a like-for-like control comparison. The blood-based reference described for the primary endpoint is a measurement comparator, not a separate trial arm.
The stated purpose was to collect cerebral oximetry data for device calibration and validation. The primary measurement was continuous regional oxygen saturation, or rSO2, compared with a reference cerebral oxygen saturation based on arterial and jugular venous blood oxygen saturation. The endpoint timeframe was one visit, up to 4 hours.
The record does not report a detailed protocol, the number or timing of blood samples, the number of sensor readings per participant, or the procedure used to select and pair observations. It also does not report eligibility criteria beyond the population and procedure described in the title and summary. Phase is listed as NA. These details cannot be inferred from the single-group designation or from the study’s stated calibration and validation purpose.
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
| Stage | O3 Regional Oximeter Sensors |
|---|---|
| STARTED | 44 |
| COMPLETED | 38 |
| NOT COMPLETED | 6 |
Reasons for not completing this period:
| Reason | O3 Regional Oximeter Sensors |
|---|---|
| Physician Decision | 6 |
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.
Population: The subjects included for the analysis are within the desired calibration range, with no sensor placement issues in the data collection time period.
| Characteristic | Unit | O3 Regional Oximeter Sensors |
|---|---|---|
| Participants analysed | 33 | |
| Age, Customized: 0-23 months | Participants | 14 |
| Age, Customized: 2-12 years | Participants | 19 |
| Sex: Female, Male: Female | Participants | 15 |
| Sex: Female, Male: Male | Participants | 18 |
| Race/Ethnicity, Customized: Ethnicity – White | Participants | 16 |
| Race/Ethnicity, Customized: Ethnicity – Hispanic/Latino | Participants | 11 |
| Race/Ethnicity, Customized: Ethnicity – Asian | Participants | 4 |
| Race/Ethnicity, Customized: Ethnicity – Not Hispanic | Participants | 2 |
| Region of Enrollment: United States | participants | 33 |
6. Primary Endpoint
The registry lists the primary outcome as Accuracy of Sensor by Arms Calculation of Percent rSO2, measured over One visit; up to 4 hours.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | %rSO2 |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is titled “Accuracy of Sensor by Arms Calculation of Percent rSO2.” It concerns the agreement between the O3 sensor’s regional oxygen saturation readings and a reference cerebral oxygen saturation measured using a combination of arterial and jugular venous blood oxygen saturation. Its unit is %rSO2, and the registry describes the endpoint as a number.
Accuracy is quantified with Arms, defined in the record as the square root of the sum of the squares of mean bias and the estimated standard deviation of bias. Bias is the difference between the rSO2 measurement and its blood reference. In general, this form of root-mean-squared error combines systematic offset and variation in the measurement differences: a persistent shift contributes through mean bias, while inconsistency contributes through the standard deviation of bias. It therefore summarizes more than whether the average sensor reading is high or low.
Such a measure is used in device validation because the size and spread of measurement errors both matter when judging agreement with a reference. The registry names the endpoint and its calculation but does not report the Arms estimate, its uncertainty, or the underlying bias and standard deviation. The reported record therefore does not establish how close the sensor readings were to the reference in this study.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The key statistical limitation is that the registry record names the primary accuracy endpoint but does not include its numerical result. No Arms estimate, mean bias, estimated standard deviation of bias, confidence interval, or analysis result is reported. As a result, the posted information does not quantify sensor-reference agreement, indicate the direction of any average bias, or show how much the measurement differences varied.
The endpoint definition is informative about the intended analysis. It evaluates differences between the sensor’s rSO2 values and a blood-based cerebral oxygen saturation reference, then combines the average bias and estimated spread of bias into an Arms error. In general, the same overall error can arise from different combinations of average offset and variability. Without the component estimates, an Arms value alone would also be less informative about the pattern of disagreement; in this record, neither the combined estimate nor its components are reported.
The design is single-group, with all subjects receiving the sensor during cardiac catheterization. This is suitable in principle for collecting paired sensor and reference measurements, but it does not create a randomized control comparison. The study’s measurement comparator should not be mistaken for an independent control arm. The registry lists allocation as NA and masking as none, so it does not document randomized allocation or masking. It does not report whether reference measurements were interpreted without knowledge of sensor readings, or vice versa.
The record lists enrollment of 44, but does not state how many participants contributed usable paired measurements or how many observations entered the endpoint calculation. It also does not report participant-level or observation-level uncertainty, confidence intervals, or a statistical analysis result. Consequently, the precision of any estimated accuracy cannot be assessed from the posted record.
The safety entries are distinct from the accuracy endpoint. For the O3 Regional Oximeter Sensors group, the registry lists 0 participants with serious adverse events among 44 at risk, and 2 with other adverse events among 44 at risk. Those counts do not supply information about sensor-reference agreement and should not be treated as evidence about measurement performance. The registry does not list serious adverse event terms.
Overall, the posted record establishes the study’s purpose, single-group structure, intended reference, endpoint definition, and adverse event counts. It does not provide the primary result needed to judge the magnitude or precision of measurement accuracy. No conclusion about the direction, size, or clinical importance of the sensor’s measurement error follows from the endpoint description alone.
9. Statistical Methods Explained
The statistical target is measurement agreement. The sensor produces regional oxygen saturation readings, denoted rSO2 in the endpoint description, and those readings are compared with a cerebral oxygen saturation reference based on arterial and jugular venous blood oxygen saturation. General explanation: agreement analysis asks how far paired measurements differ, rather than whether two treatment groups differ in an outcome.
What does bias represent?
The registry defines bias as the difference between rSO2 and its blood reference. A positive or negative bias would indicate a consistent direction of difference under a specified subtraction convention, but the record does not state a numerical bias or its direction. It also does not provide the observed distribution of paired differences. Thus, the endpoint definition alone cannot show whether the sensor tended to read higher or lower than the reference.
In general, mean bias describes the average signed difference across measurements. It reflects systematic offset, but does not by itself describe the variability of individual differences. A small average can coexist with a wide spread if positive and negative differences offset one another. Conversely, a stable offset can produce a nonzero average with relatively little variation. These are general properties of agreement measures, not findings reported for this study.
Why combine bias and variability?
The registry says Arms is the square root of the sum of the squares of mean bias and estimated standard deviation of bias. This is a root-mean-squared-error construction. Squaring the components makes both systematic offset and spread contribute as nonnegative quantities; taking the square root returns the summary to the scale of the original measurement. The endpoint’s unit is %rSO2.
In general, an error measure that includes both components is useful because agreement depends on more than the average difference. If the sensor’s readings vary widely around the reference, a near-zero mean difference would not describe that variability. If the readings are consistently shifted, a measure that only describes spread would miss the offset. Combining the two addresses both features in one summary. It does not, by itself, show the shape of the differences or whether performance varies across participants or measurement levels.
The record does not report the numerical Arms result, its uncertainty, or the component estimates. It also does not specify whether the analysis treated each reading as independent, first summarized readings within participants, or used another method to account for repeated measurements. Because multiple readings from one participant may be related, the choice of analysis unit can affect uncertainty estimates. This is a general methodological consideration; the registry does not report enough detail to determine how it was handled.
What can be learned from the design?
The study is recorded as single-group, with all subjects in the test group receiving the sensor during a scheduled cardiac catheterization. The endpoint compares sensor readings with a blood-based reference, not with readings from an untreated or alternative-device control group. This is a paired measurement-validation framework as described in the outcome definition, but details of the pairing process are not reported. General explanation: paired measurement data can control for differences between people when each sensor reading is compared with a reference from the same measurement context, but interpretation still depends on timing, sampling, and analysis procedures.
Masking is recorded as none. The registry does not explain whether the sensor operator, reference-measurement personnel, or analysts could see the other measurement. It also does not describe any calibration set separate from a validation set. Although the study title and summary use calibration and validation language, the record does not specify how calibration was performed, whether data were divided for calibration and validation, or how independent validation was assessed.
What do the planning assumptions mean?
The record’s planning assumptions describe an illustration based on an observed effect, using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect set to a stated fraction of the observed effect. The assumptions also distinguish total events for time-to-event illustrations from participants per arm for other designs. These are general planning conventions, not evidence that this study used a two-arm design: the trial itself is recorded as single-group, and the planning illustration is explicitly not the trial’s own design. No planning sample-size values or observed primary effect are reported in this record. Such illustrations must not be interpreted as observed power or as a justification for the study’s enrollment.
For this record, the central methods gap is not the definition of the error summary but the absence of the estimate and supporting analysis details. The endpoint description explains what the study intended to calculate. Without the reported result, its uncertainty, and information about the observations included, readers cannot evaluate the study’s empirical accuracy estimate or its precision.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists one outcome measure, identified as the accuracy of the sensor by an Arms calculation of percent rSO2. It lists 0 analyses. The record does not report a multiplicity adjustment, a hierarchy of outcomes, a prespecified testing strategy, or whether additional outcomes were assessed outside the posted record. With no numerical endpoint analysis reported, it is not possible to assess the role of multiple testing in the results.
Missing-data handling is not reported in the registry record. The record does not say whether any sensor readings, blood reference measurements, or paired observations were missing, excluded, or unusable. It does not describe imputation, complete-case analysis, rules for incomplete measurements, or sensitivity analyses. Enrollment of 44 does not establish that all 44 contributed to the primary endpoint.
The analysis population is also not reported. No participant flow or count of participants with evaluable paired measurements is included. The record does not specify whether all enrolled subjects, only those with successful sensor and reference readings, or another defined group entered the endpoint analysis.
These omissions matter because measurement studies depend on valid pairing and clear accounting of the observations included. General explanation: excluding readings can affect an accuracy estimate if the likelihood of exclusion is related to measurement difficulty or error. The registry does not provide information to determine whether that occurred in this study.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| O3 Regional Oximeter Sensors | 0 | 44 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| O3 Regional Oximeter Sensors | 0 of 44 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| O3 Regional Oximeter Sensors | 2 of 44 |
Most frequent non-serious adverse event terms
| Event term | Organ system | O3 Regional Oximeter Sensors |
|---|---|---|
| Mild rash, localized to sensor location | Skin and subcutaneous tissue disorders | 2 of 44 |
For the O3 Regional Oximeter Sensors group, the registry lists 44 participants at risk for serious adverse events and 0 participants affected. It lists 44 at risk for other adverse events and 2 participants affected. No deaths are listed, and no serious adverse event terms are listed.
These are participant counts, not event rates calculated for a particular duration of observation. The record does not describe the nature of the 2 other adverse events, their timing, severity, relationship to the sensor, or whether one participant experienced more than one event. The “other” category should not be assumed to mean that the events were mild or related to the device; those details are not reported.
The safety counts are descriptive entries for a single-group study. There is no separate control group in the registry record for a comparative safety analysis. The listed zero serious adverse events is a report of the posted count, not proof that the risk is absent. The registry also does not describe completeness of adverse event collection or the period covered beyond the study’s stated one-visit timeframe for the primary endpoint. These safety entries do not answer the primary measurement-accuracy question.
12. What the Result Does, and Does Not, Mean
The record describes a study intended to calibrate and validate the O3 regional oximetry device in neonates, infants and children undergoing cardiac catheterization. It identifies a relevant measurement question: how closely sensor rSO2 readings match a cerebral oxygen saturation reference derived from arterial and jugular venous blood oxygen saturation. It also defines an accuracy summary that incorporates both average bias and the estimated spread of bias.
What the record does not establish is the observed accuracy. The numerical Arms estimate is not reported, and neither are its components, uncertainty, or the number of paired observations analyzed. Without those results, readers cannot determine the size or direction of measurement differences, how consistent the readings were, or how precise the estimate may have been.
The single-group design means that the study is not a randomized comparison of the sensor against a control group or another device. Its blood-based reference is part of the measurement comparison, not a control arm. The registry does not report enough procedural and analysis detail to evaluate how sensor and reference measurements were paired or whether readings were handled as repeated observations within participants.
The safety entries report 0 participants with serious adverse events and 2 with other adverse events in the sensor group, among 44 at risk for each category. The event descriptions and relationship to the device are not reported. These counts should be read separately from the accuracy endpoint and do not supply a comparative safety assessment.
In general, a device-validation result is most informative when the error estimate is accompanied by uncertainty, component summaries, and a clear account of the observations and analysis population. Those elements are not reported in the registry record. The posted information therefore supports a description of the study question and design, but not a conclusion about how accurate the sensor was in this study or how broadly the result applies.
13. Why This Trial Matters Statistically
This trial matters statistically because it illustrates the distinction between defining an accuracy endpoint and reporting an accuracy result. The registry gives a concrete measurement target: sensor rSO2 compared with a blood-based cerebral oxygen saturation reference. It also defines Arms as a summary combining mean bias and estimated standard deviation of bias. That definition makes clear that accuracy is not simply a question of whether the average sensor reading matches the reference.
For measurement validation, both systematic offset and variability are important. General explanation: a summary that captures both can provide a compact description of error, while component estimates and uncertainty help readers understand what contributes to that summary. In this record, the endpoint formula is stated, but the estimate and components are not reported. This gap prevents the statistical description from becoming an assessment of observed measurement performance.
The single-group structure also clarifies what the design can address. The study is organized around collecting sensor and reference data during a procedure, rather than assigning participants to competing treatment groups. That is a different inferential aim from estimating a treatment effect through randomization. The registry’s allocation and masking entries reinforce the need to interpret the study as a measurement-validation effort and not as a controlled clinical comparison.
Finally, the record emphasizes why transparent reporting matters. Enrollment is listed as 44, yet the number contributing paired measurements, the handling of repeated readings, the analysis population, and the uncertainty around the primary estimate are not reported. Those details are central to understanding precision and reproducibility in a measurement study. The entry therefore offers useful information about the intended question and endpoint while leaving the central numerical answer unavailable.
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 this record constrain interpretation. Most importantly, the primary endpoint result is not reported. The record names the Arms calculation but gives no estimate, confidence interval, mean bias, or estimated standard deviation of bias. The study’s empirical measurement accuracy and precision therefore cannot be assessed from the posted results.
The registry lists enrollment of 44, but does not report the number of participants or paired measurements included in the primary analysis. It does not describe missing or unusable readings, participant flow, or the analysis population. The frequency of sensor measurements and reference sampling is also not reported, so the amount and structure of measurement data are unclear.
The record does not explain how sensor readings were paired with arterial and jugular venous blood oxygen saturation, how the reference was constructed in practice, or how repeated measurements within a participant were handled statistically. It also does not report whether a separate calibration and validation process was used, despite the study’s stated purpose. Masking is recorded as none, while the roles of measurement personnel and analysts are not described.
The study has a single group and no randomized control arm. This limits the types of comparative conclusions available, though the stated endpoint is a comparison of measurements against a reference rather than a between-group treatment contrast. Generalizability is also difficult to judge because detailed participant characteristics and eligibility criteria are not reported. These limitations describe what is absent from the registry record; they do not establish that the procedures were not conducted.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Calibration and Validation of Masimo's O3 Regional Oximetry Device in Neonates, Infants and Children — NCT03123354.
- Registry record last updated 2018-06-13; status last verified 2018-05.
- Results first posted 2018-06-13.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT03123354 is a completed, single-group device study sponsored by Masimo Corporation. It concerns calibration and validation of the O3 regional oximetry device in neonates, infants and children undergoing cardiac catheterization. The record lists enrollment of 44, a start date of 2015-05-01, and a primary completion date of 2017-03-13.
All subjects were in the test group and received an O3 regional oximeter sensor during a scheduled, general cardiac catheterization procedure. The primary endpoint compares continuous sensor rSO2 measurements with a blood-based cerebral oxygen saturation reference and defines accuracy using Arms, based on mean bias and estimated standard deviation of bias. The endpoint timeframe is one visit, up to 4 hours.
The numerical primary endpoint result and its uncertainty are not reported. The record lists 0 participants with serious adverse events and 2 with other adverse events among 44 at risk in the sensor group; event descriptions are not reported. The number of participants with evaluable paired measurements, analysis-population details, missing-data methods, and multiplicity procedures are not reported in the registry record.