Postoperative Delirium: Brain Vulnerability and Recovery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Postoperative Delirium: Brain Vulnerability and Recovery (NCT03110185), a non-randomized na study. The primary outcome is EEG During Delirium and After Recovery and in Patient Controls..

2. Clinical Question

3. Trial Design

NCT03110185, titled Postoperative Delirium: Brain Vulnerability and Recovery, is listed as completed. The registry identifies Washington University School of Medicine as the sponsor, records an enrollment of 91, and classifies the study as non-randomized, unmasked, and parallel. Its phase is listed as NA. The recorded start date is 2015-11-25 and the primary completion date is 2019-03-08.

The design grouped cardiothoracic ICU patients according to whether postoperative delirium was diagnosed. The registry describes a delirium arm and a no-delirium arm, rather than treatment groups assigned by chance. In the delirium arm, patients were to wear a diffuse optical tomography device in addition to normal monitors and undergo non-contrast functional MRI. The no-delirium arm was to be monitored in the same way. EEG, fcDOT, and fcMRI are listed as device interventions.

The study’s stated purpose was to use EEG and diffuse optical tomography to examine brain function during delirium and recovery, and to compare brain function in patients who recovered from delirium with patients who did not have delirium using DOT and fMRI. The registry also describes ongoing review of electronic medical records to coordinate study procedures and collect surgery, recovery, and mental-status information.

Non-randomized allocation matters to interpretation. Delirium status defined the groups, so any difference between them could reflect differences associated with delirium, other patient or clinical characteristics, or both. The record does not establish that the groups were comparable on other factors. The registry record does not report a protocol-based sample-size justification, detailed statistical analysis plan, or interim-analysis plan.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Participants were not assigned to groups, as baseline delirium status determined which group the participants fell into (delirium vs. no delirium).

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Due to time constraints and limited availability of staff and the EEG technology, no participants completed both the pre-operative and post-operative EEG. Therefore, these data were not analyzed.

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: 9 participants who completed study procedures were excluded from analyses due to poor data quality.

6. Primary Endpoint

The registry lists the primary outcome as EEG During Delirium and After Recovery and in Patient Controls., measured over Data acquisition will occur on 4 separate days between postoperative days 1-8..

The primary outcome is titled EEG During Delirium and After Recovery and in Patient Controls. It concerns delta (0.5-4 Hz), theta (4-8 Hz), and alpha (8-12 Hz) EEG waveforms. The stated acquisition schedule was 4 separate days between postoperative days 1-8. These frequency bands describe different components of EEG activity; examining them can characterize brain electrical patterns at the times and in the patient groups of interest.

As general statistical context, repeated EEG observations can be useful for studying how a measure changes over time and how patterns differ across clinical groups. Such comparisons require the relevant observations to be available and a clear analysis plan for defining the time points, groups, and outcome summaries. The registry record says that no participants completed both the pre-operative and post-operative EEG, and that these data were therefore not analyzed.

Thus, the endpoint description identifies what was intended to be measured, but the record does not provide an analyzed primary EEG result. It reports no group estimate, measure of variability, confidence interval, or test result for this primary endpoint. The endpoint’s planned measurement schedule should not be mistaken for evidence that the planned comparison was completed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does non-randomized allocation mean? In this study, patients were grouped by postoperative delirium status rather than assigned to groups at random. As general statistical context, random allocation can help balance measured and unmeasured characteristics between groups on average. Without it, the groups may differ for reasons besides the status being studied. That makes a direct group contrast descriptive of the observed groups, not automatically an estimate of a causal effect. The registry record specifically cautions that these arms are not a like-for-like control comparison.

What does a parallel design describe? A parallel design generally means that groups are followed as separate groups, rather than participants crossing from one assigned treatment to another. Here, the groups are described by whether patients in the cardiothoracic ICU were diagnosed with postoperative delirium. The registry lists masking as none. That identifies the recorded design classification; the record does not describe additional procedures for blinded outcome assessment.

Why is the primary endpoint not interpretable as a comparative result? The primary outcome describes delta (0.5-4 Hz), theta (4-8 Hz), and alpha (8-12 Hz) EEG waveforms, with acquisition planned on 4 separate days between postoperative days 1-8. The registry says no participants completed both pre-operative and post-operative EEG, so these data were not analyzed. An outcome description or planned acquisition schedule does not supply the missing analysis. There is no posted primary estimate, standard error, confidence interval, or test result with which to quantify a difference or its uncertainty.

What would an estimate and confidence interval ordinarily contribute? As general statistical explanation, an estimate summarizes the observed contrast or outcome, while a confidence interval gives a range of values compatible with the data and the model at the stated confidence level. The interval’s width helps convey precision. A small or large p-value alone would not describe the size of an effect, and a confidence interval would not remove bias caused by non-randomized group formation. No primary EEG estimate or interval is reported in this registry record.

Why can repeated observations be challenging? General statistical methods for repeated measures account for the fact that observations from the same person are related, rather than treating every measurement as independent. The choice of time points and handling of incomplete records can affect an analysis. The registry specifies an intended EEG schedule but does not report a model for repeated measures, the availability of measurements by time point, or a method for handling incomplete EEG data. Since the primary EEG data were not analyzed, no such method can be inferred from a result.

What do the planning assumptions mean? The record describes sample-size illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The target effect is described as a stated fraction of the observed effect. For time-to-event illustrations, the stated quantity is the total number of events; for other illustrations, it is participants per arm. These are planning assumptions, not the trial’s own design justification and not observed power. The record does not report the trial’s sample-size justification. Because the primary EEG data were not analyzed, the record also gives no primary EEG result from which to assess precision or an observed effect.

Why does over-dispersion matter in general? For count outcomes, over-dispersion means that variability exceeds what a simpler count model assumes. Ignoring it can make standard errors too small and uncertainty appear narrower than it is. This is general statistical context only: the registry entries do not identify a count-outcome model or report that over-dispersion was assessed in this study.

How should the safety numbers be treated statistically? The adverse-event figures are counts within groups and their recorded at-risk populations. Descriptive counts do not by themselves show whether the groups have different underlying risks. A comparative analysis would need to account for the study design and relevant uncertainty, and a non-randomized comparison would remain vulnerable to group differences. The registry record does not report a statistical safety comparison or an adjusted estimate.

What analysis details remain unknown? The record lists 1 analysis but does not describe its statistical model, covariates, contrast, handling of repeated measurements, or uncertainty estimates in the registry record recorded for this trial. It does not report an interim-analysis plan. These details cannot be reconstructed from the outcome title, device list, or safety counts.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 3 outcome measures and 1 analysis, but the record does not report a multiplicity strategy. Multiplicity arises when several outcomes, time points, groups, or contrasts are examined; as general statistical context, repeated testing can increase the chance of at least one apparently unusual result even when no real difference exists. Whether an adjustment is needed depends on the prespecified questions and analysis plan. The registry does not state which comparisons were treated as primary or how any multiplicity was addressed.

For the primary EEG outcome, missingness is not a minor detail: the registry states that no participants completed both the pre-operative and post-operative EEG and that the data were not analyzed. It attributes this to time constraints and limited availability of staff and EEG technology. The record does not report whether some EEG observations were available at individual time points, how many measurements were incomplete, or whether any imputation or other missing-data method was used. No method should be inferred.

The registry describes groups based on postoperative delirium status, but does not provide a detailed analysis-population definition for the primary EEG endpoint. It also does not state which patients contributed to the listed analysis or whether all enrolled participants were included in each outcome analysis. The recorded enrollment of 91 does not resolve that question. The analysis populations and reasons for any exclusions beyond the stated lack of paired EEG are not reported in the registry record.

The record does not report a detailed statistical analysis plan, any interim-analysis plan, or a sample-size justification. Therefore, it is not possible to determine from the registry what analyses were prespecified, how incomplete data were to be handled, or whether outcome selection or analysis decisions were made before results were known.

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 reports serious adverse events among 17 at risk in the delirium group and 33 at risk in the no-delirium group: 0 serious events are recorded in each group. Other adverse events are also recorded as 0 among 17 at risk in the delirium group and 0 among 33 at risk in the no-delirium group. The registry separately reports 0 deaths in the delirium group and 2 in the no-delirium group.

These figures are descriptive counts, not a randomized safety comparison. The number at risk is not the same as the total enrollment recorded for the study, and the registry record does not explain the difference between those figures. The record also does not state the relationship, if any, between the deaths and the adverse-event categories, so they should not be assumed to overlap or be independent.

Zero reported events does not prove that an event could not occur; it means no event is recorded for that category and group in the registry figures. With no reported comparative analysis or uncertainty estimates, the counts do not establish a difference in event risk between groups. Because delirium status defined the groups, the counts also cannot show that delirium status caused or prevented the events.

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

The registry record establishes that NCT03110185 was a completed, non-randomized study of brain measurements in patients with and without postoperative delirium in a cardiothoracic ICU. It describes EEG, diffuse optical tomography, and functional MRI as study measures and outlines comparisons involving delirium, recovery, and patient controls. It also records that electronic medical records were monitored to coordinate procedures and collect clinical and mental-status information.

For the primary endpoint, however, the registry does not provide an analyzed result. It says no participants completed both pre-operative and post-operative EEG because of time constraints and limited availability of staff and EEG technology, and that the data were not analyzed. Accordingly, there is no posted primary EEG estimate from which to assess direction, size, or precision. This is not evidence that the EEG patterns were the same, and it is not evidence that they differed.

The distinction between a planned comparison and an analyzed result is important. The endpoint title and frequency bands explain what investigators intended to measure; they do not establish that the intended comparison yielded a result. The registry also describes DOT and fMRI aims, but detailed analyzed findings for those measures are not included in the registry record recorded for this trial.

The study design further limits causal interpretation. Since patients were grouped according to delirium status rather than randomized, any observed group differences could reflect other characteristics as well. The registry itself cautions that the arms are not a like-for-like control comparison. No adjustment strategy or analysis details are available to clarify whether such differences were addressed.

The safety counts document the events entered in the registry for the listed at-risk groups, including the deaths recorded in the no-delirium group. They do not establish comparative safety or a causal relationship with delirium. In sum, the record describes a relevant measurement question and study procedures, but does not provide an analyzed primary EEG comparison or enough statistical detail to infer the size or cause of brain-function differences.

13. Why This Trial Matters Statistically

This trial is statistically notable because the registry makes the gap between a planned endpoint and an analyzed endpoint explicit. The primary question concerned EEG patterns during delirium, after recovery, and in patient controls, with measurements planned across postoperative days. Yet the registry states that no participants completed both pre-operative and post-operative EEG, so those data were not analyzed. This is a clear example of why an endpoint description and acquisition schedule should not be read as a result.

It also illustrates a design issue common to observational clinical research. Grouping patients by a clinical condition can make a clinically meaningful comparison possible, but it does not create the balance expected from random allocation. As general statistical context, observed differences may reflect the condition of interest, other group differences, or a combination. Clear qualification of the comparison is therefore central to interpretation, and the registry does caution that the arms are not like-for-like controls.

The study further shows why operational feasibility is part of statistical evidence. The registry attributes the absence of paired EEG to time constraints and limited staff and technology availability. Without the planned data, the primary analysis could not be performed as described in the record. A carefully specified method cannot compensate for measurements that were not collected in the form needed for the intended comparison.

Finally, the record separates descriptive safety counts from comparative evidence. Counts can document events by group, but without randomization, uncertainty estimates, and a reported analysis, they do not establish a safety difference or causal link. Readers can assess what was recorded while recognizing the limits of what these figures resolve.

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 primary limitation documented by the registry is that the planned pre-operative and post-operative EEG pairing was not achieved. The record says no participants completed both measurements, attributing this to time constraints and limited availability of staff and EEG technology; the data were not analyzed. This prevents interpretation of the primary EEG comparison as an estimated result.

The study was non-randomized and unmasked. Patients were assigned to the described arms according to postoperative delirium status, and the registry cautions that the arms are not a like-for-like control comparison. Other differences between groups could influence an observed contrast. The record does not describe an adjustment method that would address such differences.

The record identifies EEG, fcDOT, and fcMRI, and describes intended comparisons, but the registry entries recorded for this trial do not include detailed analyzed results for the brain-imaging measures. It also does not report the specific statistical model, outcome summaries, uncertainty estimates, or methods for repeated measurements. A single analysis is listed, but its details are not reported in the registry record.

The stated safety counts have different at-risk group sizes, and the record does not explain the relationship between those group figures and the recorded enrollment. It also does not clarify whether deaths overlap with the adverse-event categories. No comparative safety analysis is reported. Finally, the registry does not report a detailed sample-size justification, multiplicity approach, interim-analysis plan, or broader missing-data strategy. These gaps limit interpretation beyond the stated primary EEG non-analysis.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Postoperative Delirium: Brain Vulnerability and Recovery — NCT03110185.
• Registry record last updated 2020-06-02; status last verified 2020-05.
• Results first posted 2020-06-02.
• 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

NCT03110185, Postoperative Delirium: Brain Vulnerability and Recovery, is listed as completed and sponsored by Washington University School of Medicine. The registry records enrollment of 91, non-randomized allocation, no masking, a parallel model, and phase NA. Its recorded start date is 2015-11-25 and primary completion date is 2019-03-08.

The study grouped cardiothoracic ICU patients by postoperative delirium status and listed EEG, fcDOT, and fcMRI. The primary outcome concerned delta, theta, and alpha EEG waveforms, with acquisition planned on 4 separate days between postoperative days 1-8. The registry says no participants completed both pre-operative and post-operative EEG, so those data were not analyzed.

Safety figures record 0 serious and 0 other adverse events in each listed at-risk group; deaths are recorded as 0 in the delirium group and 2 in the no-delirium group. The registry does not provide a primary EEG estimate or detailed analysis methods for interpreting a comparative effect.