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 A Net Ultrafiltration Challenge to Predict Preload-dependence (UF CHALLENGE) (NCT05214729), a randomized na study. The primary outcome is Diagnostic Performance of the Net Ultrafiltration (UF NET) Challenge to Detect de Novo Preload Dependence.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT05214729 |
| Phase | NA |
| Condition | Acute Kidney Injury |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | None (open label) |
| Enrollment | 20 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Hospices Civils de Lyon |
| Results posted | Yes |
2. Clinical Question
Population
Critically ill patients with acute kidney injury receiving continuous renal replacement therapy. The registry record lists an enrollment of 20. It also says that four patients did not perform the second challenge; the reason is truncated in the record.
Intervention
A fast net ultrafiltration challenge removing 250 ml of ultrafiltrate over 15 minutes, and a slow challenge removing 250 ml over 30 minutes. Each participant was intended to undergo both challenges.
Comparator
The diagnostic reference was a postural maneuver—passive leg raising or Trendelenburg—used to assess preload dependence after a challenge. The order of the fast and slow challenges was randomized, with a 24-hour washout between them.
Primary question
How well does the relative change in calibrated cardiac index during each ultrafiltration challenge distinguish patients classified as preload-dependent by the postural reference assessment? The planned measure of diagnostic discrimination was the area under the receiver-operating-characteristic curve (AUROC).
In general, a diagnostic-performance question asks whether a measurement separates observations classified as positive from those classified as negative by a reference standard. Here, the reference classification used pulse-contour continuous cardiac-index variation during a postural maneuver: positive if the variation was > 10% and negative if it was ≤ 10%. The study’s focus is the change in cardiac index during fluid removal, not a comparison of patient outcomes between treatment groups.
The record describes this as a diagnostic study rather than a trial designed to establish a therapeutic benefit. It does not report the primary AUROC estimate or its confidence interval, so the diagnostic performance observed in this study cannot be assessed from the registry record.
3. Trial Design
This completed study used randomized allocation, no masking, and a crossover model. The registry lists enrollment of 20 and identifies acute kidney injury as the condition. The sponsor was Hospices Civils de Lyon. The study began on 2022-07-14 and its primary completion date was 2025-01-02.
Each participant was intended to undergo both net ultrafiltration challenges: a fast challenge removing 250 ml over 15 minutes and a slow challenge removing 250 ml over 30 minutes. Randomization assigned the order. One sequence began with the fast challenge and then used the slow challenge; the other began with the slow challenge and then used the fast challenge. A washout period of 24 hours separated the challenges. This within-participant arrangement allows the two challenge conditions to be evaluated in the same people, while randomized order helps avoid making one challenge systematically occur first.
The registry description states that all enrolled patients underwent two challenges, but also states that four patients did not perform the second challenge. The explanation is cut off after “presence of pr” and is not reported in full in the registry record. The record does not give a complete account of which observations were available for each analysis.
The primary endpoint was intended to assess diagnostic discrimination using cardiac-index changes measured at baseline and immediately after each challenge, with the postural maneuver serving as the reference assessment. The registry lists three outcome measures and one analysis. It does not report a detailed statistical analysis plan, the method used to calculate confidence intervals, or whether the analysis accounted for the crossover structure. Those details are not reported in the registry record.
The registry status is completed, but completion status alone does not establish that every planned measurement was obtained or that results were posted for every outcome. The primary diagnostic-performance 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.
Before assignment: 20 patients were randomized. Of the 44 not randomized, 44 met inclusion criteria and met exclusion criteria.
First Challenge (15 or 30 Minutes)
| Stage | Slow to Fast Group | Fast to Slow Group |
|---|---|---|
| STARTED | 10 | 10 |
| Received the Intervention | 10 | 10 |
| COMPLETED | 10 | 10 |
| NOT COMPLETED | 0 | 0 |
Wash-out Period (24 to 48h)
| Stage | Slow to Fast Group | Fast to Slow Group |
|---|---|---|
| STARTED | 10 | 10 |
| COMPLETED | 9 | 7 |
| NOT COMPLETED | 1 | 3 |
Reasons for not completing this period:
| Reason | Slow to Fast Group | Fast to Slow Group |
|---|---|---|
| Cardiac index > 4 L/min/m2 | 0 | 2 |
| Preload dependent prior to 2nd challenge | 0 | 1 |
| Expected death in the following 24h | 1 | 0 |
Second Challenge (15 or 30 Minutes)
| Stage | Slow to Fast Group | Fast to Slow Group |
|---|---|---|
| STARTED | 9 | 7 |
| Received the Intervention | 9 | 7 |
| COMPLETED | 9 | 7 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: All enrolled patients underwent two UF NET challenges, one fast and one slow, consisting in the removal of 250 ml of ultrafiltrate in randomized order and in a cross-over design (with a 24h wash-out period between challenged). 4 patients did not perform the second challenge due to the presence of predefined exit criteria during the wash-out period, as defined by the allocations sequence.
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.
| Characteristic | Unit | Fast to Slow Group | Slow to Fast Group | Total |
|---|---|---|---|---|
| Participants analysed | 10 | 10 | 20 | |
| Age, Continuous | years | 64 (50 to 70) | 62 (54 to 66) | 62 (52 to 68) |
| Sex: Female, Male: Female | Participants | 1 | 2 | 3 |
| Sex: Female, Male: Male | Participants | 9 | 8 | 17 |
| Race and Ethnicity Not Collected | Participants | — | — | 0 |
| Region of Enrollment: France | participants | 10 | 10 | 20 |
| Sepsis-related Organ Failure Assessment (SOFA) | Units on a scale | 12 (11 to 14) | 12 (8 to 13) | 12 (11 to 13) |
| Norepinephrine dose (tartrate formula) | µg/kg/min | 0.22 (0.07 to 0.39) | 0.23 (0.09 to 0.29) | 0.22 (0.06 to 0.31) |
6. Primary Endpoint
The registry lists the primary outcome as Diagnostic Performance of the Net Ultrafiltration (UF NET) Challenge to Detect de Novo Preload Dependence, measured over Baseline and immediately at the end ot each UF NET challenge (approximately 15-30 minutes, 1 day apart).
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | Proportion probability (0 to 1) |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is the diagnostic performance of the net ultrafiltration challenge for detecting de novo preload dependence. The registry specifies the AUROC of the relative variation in calibrated cardiac index, measured from baseline to the end of a challenge. Performance was to be assessed against a postural reference maneuver—passive leg raising or Trendelenburg—using pulse-contour continuous cardiac-index variation to classify preload dependence. A result was positive when variation was > 10% and negative when it was ≤ 10%.
In general, an AUROC describes how well a continuous measurement ranks observations classified as positive relative to those classified as negative by a reference standard. It evaluates discrimination across possible cutoffs rather than reporting the performance of only one chosen cutoff. The confidence interval describes uncertainty around the estimated AUROC; a wider interval indicates less precision, while a narrower interval indicates greater precision. These are general interpretations, not findings about this study.
The endpoint is relevant to the stated diagnostic aim because a cardiac-index response during ultrafiltration is being compared with a separate assessment of preload dependence. The registry lists the endpoint unit as a proportion probability from 0 to 1 and specifies a mean with a 95% confidence interval. However, neither the observed AUROC nor its confidence interval is reported in the registry record. The record therefore gives the endpoint definition but not the result needed to judge the observed discrimination or its precision.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record defines a diagnostic comparison, but it does not report the numerical primary outcome. It gives no AUROC estimate for either challenge, no 95% confidence interval, and no direct statistical comparison of fast versus slow challenge performance. Consequently, the registry record does not establish how well either challenge discriminated preload-dependent from non-preload-dependent observations, nor whether their diagnostic performance differed.
The design has a crossover structure: participants were assigned to perform both challenge conditions in randomized order, separated by a 24-hour washout. In general, crossover studies can make within-person comparisons because each participant receives or experiences both conditions. This can reduce the influence of stable differences between participants. It also requires attention to order, carryover, and incomplete paired observations. The registry record does not report whether the analysis accounted for these features or how observations from participants who did not complete the second challenge were handled.
The record states that four patients did not perform the second challenge, while another sentence says all enrolled patients underwent both challenges. The stated reason for noncompletion is truncated. This inconsistency makes the available analysis population and the completeness of paired measurements unclear. The record does not report how many participants contributed to each challenge-specific diagnostic analysis or how missing measurements were treated.
Safety information is posted separately from the primary diagnostic endpoint. The registry lists serious adverse events for the fast and slow challenge groups, as well as zero other adverse events and zero deaths in those group records. These counts should not be read as evidence about diagnostic accuracy. Nor do they by themselves establish comparative safety: the study is small, the entries relate to challenge-specific records, and no statistical comparison is reported.
The three listed outcome measures include the primary diagnostic-performance outcome and secondary outcomes concerning the respective performance of fast and slow challenges and the prevalence of preload dependence after a challenge. Numerical results for these outcomes are not reported in the registry record. The registry lists one analysis, but does not report its model, assumptions, handling of multiple outcomes, or inferential results. No sample-size illustration for the primary outcome is available because the primary outcome does not have two arms in the form required for that calculation.
Overall, the record supports a description of the study question, design, endpoint definition, and selected adverse-event counts. It does not support a conclusion about the magnitude, precision, or statistical significance of the primary diagnostic result. A completed status indicates the study is marked completed in the registry; it does not fill in the unreported outcome estimates.
9. Statistical Methods Explained
The main statistical method named in the registry is the area under the receiver-operating-characteristic curve (AUROC). A receiver-operating-characteristic curve describes how a continuous measurement classifies observations as the decision threshold changes. In general, the AUROC condenses that discrimination information into one summary: it reflects how often the measurement ranks an observation with the reference-positive classification higher than one with the reference-negative classification. This is a general explanation of the method, not a reported result from this study.
In this trial, the continuous measurement was the relative variation in calibrated cardiac index from baseline to the end of a net ultrafiltration challenge. The reference classification was based on a postural maneuver—passive leg raising or Trendelenburg—using pulse-contour continuous cardiac-index variation. The registry defines a positive reference result as variation > 10% and a negative result as variation ≤ 10%. The AUROC therefore concerns discrimination against that particular reference definition; it does not independently establish that the reference classification is error-free.
The registry describes the endpoint as a mean with a 95% confidence interval, with a proportion-probability unit on a scale from 0 to 1. In general, a confidence interval expresses uncertainty in an estimate under the assumptions of the analysis. It is not a range containing a fixed proportion of individual participants’ measurements. The interval’s width helps readers assess precision, but its interpretation depends on the analysis method and data structure. Neither the estimate nor its interval is reported in the registry record, and the method used to construct the interval is not reported in the registry record.
The randomized crossover design matters for analysis. Each participant was intended to complete a fast challenge and a slow challenge, with order randomized and a 24-hour washout. In general, an analysis of crossover data should recognize that measurements from the same participant are related rather than treating them as independent observations. It may also need to address whether the order of challenges or residual effects from the first challenge could influence measurements during the second. The registry record does not report the model used, whether paired observations were analyzed, or how order and possible carryover were considered.
Missing observations are particularly relevant in a crossover design because a participant without a second challenge may not contribute a complete within-person comparison. The registry states that four participants did not perform the second challenge, but its explanation is truncated. It does not report the number included in each endpoint analysis, whether incomplete cases contributed any available data, or whether any imputation was used. These details are not reported in the registry record.
The registry lists three outcome measures and one analysis. It does not explain whether the primary outcome was analyzed separately for each challenge, whether a direct contrast between challenge durations was planned, or how secondary outcomes were handled. In general, when several outcomes or comparisons are examined, multiplicity can increase the chance of at least one apparently notable result if no adjustment or hierarchy is specified. Whether and how that issue was addressed is not reported in the registry record.
Over-dispersion is a general concern when outcome variability is greater than a statistical model expects. It can make uncertainty estimates too small if the model assumes less variation than the data contain. This concept is not a reported feature of the trial, and the registry does not identify a model for which over-dispersion was assessed. More generally, model diagnostics, distributional assumptions, and sensitivity analyses are not reported in the registry record.
The registry record gives no reported primary estimate, p-value, or confidence interval. It also does not provide an analysis population definition or a detailed plan for missing data, multiplicity, or crossover-specific effects. Those omissions prevent a full appraisal of the statistical execution. The outcome definition allows readers to understand what the primary analysis was intended to measure, but not to reproduce or assess the analysis itself.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists three outcome measures: the primary diagnostic-performance outcome, the respective diagnostic performance of the fast and slow challenges, and preload-dependence prevalence after a challenge. It lists one analysis. It does not state whether the primary endpoint had priority over secondary outcomes, whether comparisons were prespecified in a hierarchy, or whether any adjustment for multiple testing was used. These details are not reported in the registry record. In general, examining multiple outcomes can make isolated findings harder to interpret unless the analysis plan explains how multiplicity is handled.
The record’s description of completion is internally unclear. It says all enrolled patients underwent both challenges, but also says four patients did not perform the second challenge. The stated reason is incomplete in the record. The number of observations contributing to each challenge-specific endpoint, the definition of the analysis population, and the handling of incomplete crossover pairs are not reported in the registry record. It also does not report whether missing endpoint values were excluded, imputed, or addressed through another method.
The registry gives an enrollment of 20 but does not state how many participants were included in each primary or secondary analysis. The primary outcome is described as a crossover diagnostic-performance assessment rather than a comparison with two separate arms. The sample-size illustration is not available for the primary outcome because the required two-arm structure is absent. The trial’s own sample-size justification and any interim-analysis plan are not reported in the registry record.
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 |
|---|---|---|---|
| Fast UF NET Challenge | 11 | 19 | 57.9% |
| Slow UF NET Challenge | 13 | 17 | 76.5% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Fast UF NET Challenge | 0 of 10 |
| Slow UF NET Challenge | 0 of 10 |
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 |
|---|---|
| Fast UF NET Challenge | 0 of 19 |
| Slow UF NET Challenge | 0 of 17 |
Most frequent serious adverse event terms
| Event term | Organ system | Fast UF NET Challenge | Slow UF NET Challenge |
|---|---|---|---|
| Hemodynamic instability episode during the 8h follow-up period after the challenge | Cardiac disorders | 10 of 19 | 12 of 17 |
| Norepinephrine increase during the UF NET challenge | Cardiac disorders | 0 of 19 | 1 of 17 |
| Change in sedation dose | Nervous system disorders | 1 of 19 | 0 of 17 |
Serious events with the highest arm-level rate: hemodynamic instability episode during the 8h follow-up period after the challenge (70.6%), norepinephrine increase during the uf net challenge (5.9%), change in sedation dose (5.3%).
The registry lists serious adverse-event information separately for the fast and slow net ultrafiltration challenge records. For the fast challenge, 19 were listed as at risk and 11 as affected by a serious event. For the slow challenge, 17 were listed as at risk and 13 as affected. The entries list zero other adverse events and zero deaths for each challenge record. These are registry counts, not estimates of comparative risk.
The listed serious-event terms include “Hemodynamic instability episode during the 8h follow-up period after the challenge” at 70.6%, “Norepinephrine increase during the UF NET challenge” at 5.9%, and “Change in sedation dose” at 5.3%. The registry does not provide enough detail in these entries to establish how the terms relate to the affected counts, whether categories overlap, or how the percentages were calculated. They should be read as listed event information rather than added together or treated as independent categories.
Because the study used a crossover design, challenge-specific records may include observations from participants who experienced both challenges; the counts should not be interpreted as independent treatment groups. The registry does not report a statistical safety comparison or detailed event narratives. Zero recorded deaths and zero other adverse events in these entries do not establish the absence of all possible harms.
12. What the Result Does, and Does Not, Mean
The registry record describes a study designed to evaluate whether cardiac-index changes during net ultrafiltration can identify preload dependence in critically ill patients receiving continuous renal replacement therapy. It specifies two challenge durations, randomized order, a crossover structure, and a postural reference assessment. It also defines the primary diagnostic-performance measure as AUROC with a 95% confidence interval.
Those design details explain what the investigators intended to measure; they do not reveal the observed diagnostic performance. The primary AUROC estimates, confidence intervals, and numerical results for the listed secondary outcomes are not reported in the registry record. As a result, the record cannot show how accurately either challenge classified preload dependence, how precise any estimate was, or whether fast and slow challenges differed.
In general, diagnostic discrimination is only one aspect of evaluating a test. An AUROC does not by itself establish whether a test changes clinical decisions or improves patient outcomes. Interpretation also depends on the reference standard, the population studied, and the measurement procedures. In this study, the reference classification was based on a specified postural maneuver and cardiac-index variation threshold. The registry record does not report the primary result needed to assess discrimination against that reference.
The study is marked completed, and the record includes selected serious adverse-event counts. Completion status and safety entries do not substitute for missing efficacy or diagnostic-performance estimates. The stated enrollment and the report of four participants not completing the second challenge also leave uncertainty about the analysis population and the number of complete within-participant comparisons.
Accordingly, the registry record supports a description of the research question and design, but not a conclusion that either challenge performs well or poorly, or that one duration is preferable. It also does not establish that either challenge improves outcomes or is safe for a particular patient. Such conclusions would require results and analyses not reported in the registry record.
13. Why This Trial Matters Statistically
This trial raises a useful statistical question about evaluating a diagnostic test within a randomized crossover design. The challenge is not simply whether cardiac index changes after ultrafiltration; it is whether that change discriminates between observations classified as preload-dependent and not preload-dependent by a reference maneuver. That distinction makes the reference definition, measurement timing, and method of comparing the continuous index with the classification central to interpretation.
The crossover structure also illustrates why design and analysis must align. Randomizing challenge order can help balance order effects, and having each participant undergo both conditions permits within-person comparisons. At the same time, incomplete second challenges can reduce the availability of paired data, and a valid analysis needs to account for within-participant dependence. The registry does not describe how these features were handled, so the design’s statistical advantages cannot be fully evaluated from the record.
Another important issue is precision. An AUROC estimate without a confidence interval gives limited information about uncertainty; the registry specifies a 95% confidence interval but does not post the estimate or interval. The enrollment is listed as 20, and four participants are reported not to have completed the second challenge, but the analysis denominators are not reported. This makes it difficult to judge how much information supports any potential diagnostic estimate.
Finally, diagnostic-performance results and safety summaries answer different questions. The posted serious-event counts do not establish whether the cardiac-index measure discriminates preload dependence, while an AUROC alone would not describe all relevant safety or clinical consequences. The record’s missing results therefore limit both statistical appraisal and the ability to connect the intended diagnostic question to broader outcomes.
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 main limitation for interpretation is that the registry record does not report the primary AUROC estimates or their 95% confidence intervals. It therefore does not give the numerical diagnostic-performance results, their precision, or a direct comparison of the fast and slow challenges. The secondary outcome results are also not reported in the registry record.
The crossover completion information is unclear: the record says all enrolled patients underwent both challenges and also says four did not perform the second challenge. The reason is truncated, and the analysis denominators are not reported. This leaves the number of complete paired observations and the handling of incomplete data uncertain. The record also does not state whether the statistical analysis accounted for within-participant correlation, randomized sequence, or possible carryover between challenges.
The reference assessment could be passive leg raising or Trendelenburg, and its classification uses pulse-contour cardiac-index variation with the stated threshold. The registry does not report further information about consistency of reference assessments or how results from the alternative maneuvers were handled. No detailed statistical analysis plan, multiplicity approach, model diagnostics, or sensitivity analyses are reported in the registry record.
The enrollment is listed as 20, but a sample-size justification is not reported. A primary-outcome planning illustration is not available because the endpoint is not expressed as a two-arm comparison for that calculation. These limitations concern what can be assessed from the registry record; they do not establish that the study’s full protocol or analysis lacked these elements.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: A Net Ultrafiltration Challenge to Predict Preload-dependence (UF CHALLENGE) — NCT05214729.
- Registry record last updated 2026-02-13; status last verified 2026-01.
- Results first posted 2026-02-13.
- 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
NCT05214729, titled “A Net Ultrafiltration Challenge to Predict Preload-dependence (UF CHALLENGE),” is a completed randomized crossover study with no masking. It lists acute kidney injury as the condition and enrollment of 20. The sponsor is Hospices Civils de Lyon. The study dates recorded are 2022-07-14 for start and 2025-01-02 for primary completion.
The study compared fast and slow net ultrafiltration challenges, each removing 250 ml over different durations, in randomized order with a 24-hour washout. The primary endpoint was AUROC for cardiac-index variation against a postural reference assessment of preload dependence. The registry lists three outcome measures and one analysis. It reports some serious adverse-event counts, but the primary diagnostic-performance estimates and confidence intervals are not reported in the registry record.