Mechanistic Inflammatory Sub-study Embedded Within the Albumin vs Balanced Crystalloid Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Mechanistic Inflammatory Sub-study Embedded Within the Albumin vs Balanced Crystalloid Trial (NCT04963569), a registered phase not stated study. The primary outcome is Longitudinal Change Over Time in Cell Types and Pro and Anti Inflammatory Cytokines in Sepsis..

2. Clinical Question

3. Trial Design

This record describes the Mechanistic Inflammatory Sub-study Embedded Within the Albumin vs Balanced Crystalloid Trial, also referred to in its summary as MIS-ABC Sepsis. It is a sub-study of the ABC Sepsis trial, focused on immune responses during the early stages of sepsis and around intravenous fluid treatment. The registry lists Sepsis as the condition and the University of Edinburgh as the sponsor.

The registry record lists the study as completed, with a start date of 2021-12-13 and a primary completion date of 2022-06-08. Enrollment is listed as 26. The summary describes three blood samples around the time participants came to hospital. The primary outcome specifies measurement at presentation, 12 hours and 24 hours. The purpose was exploratory: to map changes in cell types and inflammatory cytokines, and to investigate whether the measurements might help explain differences associated with fluid treatment or identify people who might become sicker despite treatment.

The arm descriptions refer to participants in the main ABC Sepsis trial who were randomized to either 5% Human Albumin Solution or Balanced Crystalloid as the sole resuscitative fluid in the first 6 hours. However, the allocation method is not stated in this sub-study record. The registry specifically cautions that the arms are not a like-for-like control comparison. Therefore, the arm descriptions alone do not establish that results from this record support a randomized treatment-effect comparison.

The record lists 43 outcome measures and 0 analyses. The primary endpoint is described as a participant count for eligibility in an exploratory analysis, rather than as a reported estimate of change in a particular biomarker. The record does not report a detailed statistical analysis plan or the biomarker results.

4. Participant Flow and Accounting

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

Overall Study

Analysis population for the primary outcome, as defined in the registry: Non-numerical outcome; all participants included in analysis.

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 Longitudinal Change Over Time in Cell Types and Pro and Anti Inflammatory Cytokines in Sepsis., measured over At presentation, 12 and 24 hours..

The primary outcome is titled as longitudinal change over time in cell types and pro- and anti-inflammatory cytokines in sepsis. Its stated timeframe is presentation, 12 hours and 24 hours. The description identifies the purpose as exploratory mapping of the inflammatory response over those time points. It lists cytokines including TNF-alpha, IL-6, GM-CSF, IL-10 and IFN-gamma, among others, and says there is no defined limit to the cell types or cytokines measured.

Importantly, the registered outcome measure is described as the number of participants eligible for inclusion in the exploratory analysis and later outcomes. Its unit is participants, and its population description says all participants included in analysis. This is not itself a reported numerical measure of cytokine concentration or cell-type change. The registry record does not report the number eligible or the values of the inflammatory markers at any sampling time.

Repeated measurements can be useful in mechanistic research because they describe how biological measurements evolve during an illness, rather than capturing only one moment. In general, measurements at prespecified times can help distinguish a baseline pattern from later changes. They can also generate hypotheses about biological processes. But an exploratory endpoint needs reported measurements and an appropriate analysis before readers can assess the size, direction or precision of any change. Those results are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The outcome is framed as a longitudinal, exploratory investigation. In general, longitudinal analysis uses repeated observations from the same participant to study change over time. This differs from analyzing each time point as if it came from an unrelated set of people, because measurements from one person may be related to one another. A suitable analysis would need to account for that within-person relationship. The registry record does not state which model or method was used.

The time points named for the primary outcome are presentation, 12 hours and 24 hours. These observations could describe the early trajectory of an inflammatory marker. In general, a trajectory can be represented using time-specific summaries, estimated changes from an initial measurement, or a model that uses all measurements together. Those approaches answer related but not identical questions. The record does not report which summary or contrast was selected, whether time was treated as a set of categories or as a continuous measure, or which comparisons were prespecified.

The endpoint description says the analysis is exploratory and that there is no defined limit to the cell types or cytokines measured. It names a range of inflammatory markers, including TNF-alpha, IL-6, GM-CSF, CXCL10, IL-5, IFN-a2, IL-4, CCL3, IL-10, IL-12p70, CCL20, IFN-gamma, CXCL8 and CCL4. In general, biomarker measurements can have different scales and distributions, and the appropriate summaries and models depend on the measurement properties. The registry record does not state the assay details, data transformations, distributional assumptions, or model used for any marker.

Repeated-measure data can also be incomplete. In general, missing measurements reduce the information available about a participant's trajectory. The consequences depend on how much data are missing and why they are missing. An analysis might use all available observations under stated assumptions, or it might define a complete-case population; these choices can lead to different interpretations. The registry record does not report the number with samples at each time point, the reasons for missing samples, a missing-data method, or whether sensitivity analyses were conducted. The listed enrollment should therefore not be treated as a count of complete longitudinal records.

Group comparisons require careful interpretation. The arm descriptions identify participants assigned in the main ABC Sepsis trial to Albumin or Balanced Crystalloid, but the allocation method is not stated in this sub-study record. The record also cautions that the arms are not a like-for-like control comparison. In general, a randomized allocation can support causal comparisons when the trial design and analysis preserve that allocation and other assumptions hold. Without an allocation statement and a reported analysis plan for this sub-study, readers cannot assess whether a contrast between these groups should be interpreted causally. The registry record does not report adjustment variables or a model for between-group comparisons.

The registry lists 43 outcome measures and 0 analyses. An exploratory program with multiple outcomes can generate many possible statistical comparisons: across markers, time points and groups. In general, the more results examined, the greater the chance that some patterns may appear notable by chance. Statistical methods may address this through prespecification, multiplicity adjustment or emphasis on estimation and replication. The record does not report how multiplicity was handled, whether any particular markers were prioritized, or whether the analyses were considered hypothesis-generating only beyond the description of the primary analysis as exploratory.

The record describes planning illustrations based on a two-sided significance level of 0.05, equal allocation, a normal approximation and a fraction of the observed effect. These assumptions describe illustrations, not the study's own design or an observed measure of power. In general, a sample-size calculation depends on the effect size targeted, variability, allocation and chosen error rates. If an effect is estimated after data collection, planning around that observed effect does not establish that a study had adequate power in advance. Numerical planning estimates and a sample-size justification are not reported in the registry record.

A confidence interval, in general, expresses the uncertainty around an estimated quantity under the assumptions of the method used. A p value addresses compatibility between data and a specified null model; it does not describe the size or practical importance of an effect. Neither type of result can be interpreted for this biomarker question because estimates, intervals and p values are not reported. The registry also does not state a statistical analysis plan, interim analyses, or the specific analysis population beyond describing the primary outcome as a count of participants eligible for exploratory analysis.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Albumin", "Crystalloid")
n      <- c(14, 12)
events <- c(14, 12)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Albumin  14  14  100.0  76.8  100.0  78.5  100.0  78.2  99.8
ARM  Crystalloid  12  12  100.0  73.5  100.0  75.8  100.0  75.3  99.8

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 43 outcome measures and 0 analyses. The primary biomarker outcome is explicitly exploratory, and its description allows for an undefined range of cell types and cytokines. This creates a broad analytic setting: in general, examining many markers and time points increases the chance of finding some apparently unusual patterns by chance. The record does not report a multiplicity adjustment, a hierarchy of outcomes, or a list of markers prioritized for confirmatory testing. It therefore does not establish how any nominal statistical findings should be interpreted across the full set of measures.

The primary outcome description says it counts participants eligible for inclusion in the exploratory analysis and later outcomes. It also describes the population as all participants included in analysis. However, the eligibility count itself is not reported. The registry does not give sample counts at presentation, 12 hours or 24 hours, or indicate how many participants had complete measurements across time. Enrollment of 26 should not be read as proof that all enrolled participants contributed every outcome.

Missing-data procedures, reasons for missing samples, analysis-population definitions and sensitivity analyses are not reported in the registry record. In general, these details matter because the available measurements may not represent all enrolled participants equally, particularly when the likelihood of a missing sample is related to illness or follow-up. The record also does not state whether the analysis population was defined before results were examined. These omissions prevent an assessment of how missing observations and outcome multiplicity affected the reported findings; the longitudinal findings themselves are not reported.

12. 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 lists serious adverse events by arm. In Albumin, 14 participants are listed as at risk and 0 as affected by a serious adverse event. In Balanced Crystalloid, 12 participants are listed as at risk and 0 as affected. Other adverse events are also listed as 0 in both arms, with 14 and 12 at risk, respectively. No serious adverse-event terms are listed.

Deaths are recorded separately from the serious adverse-event counts: 1 in Albumin and 0 in Balanced Crystalloid. The record does not report a causal assessment linking the death to treatment, and the separate death count should not be folded into the adverse-event count without further information. These are descriptive registry counts. They do not establish that the groups had equivalent safety, and they do not estimate the frequency of events beyond the stated at-risk counts. The registry record does not provide further detail about event definitions, follow-up duration or severity grading.

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

The study's scientific aim is to describe immune changes during early sepsis and around fluid treatment. Repeated blood sampling at presentation, 12 hours and 24 hours could, in principle, help researchers characterize changes in cell types and inflammatory cytokines over time. Such measurements may support mechanistic hypotheses about how illness and treatment relate to immune response. The registry summary also describes an interest in whether measurements might help identify people who become sicker despite treatment.

The record does not report the biomarker trajectories or their statistical estimates. It therefore cannot establish which markers changed, the direction or size of any change, or how uncertain those estimates were. Nor does it show whether marker patterns differed between Albumin and Balanced Crystalloid. The primary outcome as described is a count of participants eligible for exploratory analysis, and that count is not reported. Consequently, the stated mechanistic aims should not be confused with findings that answer those aims.

The arm labels refer to fluid assignments in the main ABC Sepsis trial, but this sub-study record does not state its allocation method and warns that the arms are not a like-for-like control comparison. Any apparent differences between groups, if reported elsewhere, would need to be assessed in the context of the relevant design and analysis. The information in this record does not support a causal conclusion about fluid treatment.

The safety counts are limited descriptive information: 0 serious adverse events are recorded in each arm, while deaths are listed as 1 and 0, respectively. These counts do not determine whether treatment caused or prevented an outcome. The registry record is useful for understanding the planned mechanistic focus and the limits of what is reported, but it does not provide the results needed to evaluate the inflammatory research question.

14. Why This Trial Matters Statistically

Mechanistic sub-studies address a different statistical question from trials that focus only on clinical outcomes. They seek to characterize processes that may accompany illness or treatment, often through repeated measurements of biological markers. In general, a time course can be more informative about the development of a response than a single measurement, while also requiring methods that account for repeated observations and incomplete follow-up.

This record makes that distinction visible. Its primary outcome concerns eligibility for an exploratory analysis of cell types and cytokines over time, rather than a reported clinical treatment effect. The planned time points are presentation, 12 hours and 24 hours, and the outcome description covers a broad range of inflammatory measurements. That breadth can be valuable for hypothesis generation, but it also makes transparent reporting of methods, outcome selection and multiplicity especially important.

The registry also illustrates why design details matter when interpreting group differences. Arm descriptions alone do not establish the basis for a valid causal contrast: the allocation method is not stated in this sub-study record, and the record cautions that the groups are not a like-for-like control comparison. Statistical interpretation therefore depends not just on the names of the groups, but on how participants entered them and how the analysis handled the study design.

Finally, a completed status and a stated scientific aim do not substitute for outcome estimates. The record lists 43 outcome measures and 0 analyses, and does not report the longitudinal biomarker results. This makes clear the difference between a registered research question and an answer supported by reported data. It also underscores the value of reporting denominators, missingness, analysis methods and uncertainty alongside exploratory findings.

15. 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 most direct limitation is that the registry record does not report the primary longitudinal biomarker findings. It gives the planned time points and names the broad outcome domain, but no cell-type or cytokine values, changes, estimates, confidence intervals or p values are reported. The participant count eligible for the exploratory analysis is also not reported.

Enrollment is listed as 26, but the registry does not state the number with measurements at each time point or the completeness of follow-up. The safety denominators of 14 and 12 are not a substitute for biomarker analysis denominators. Missing-data reasons and methods are not reported, so the completeness and representativeness of the longitudinal data cannot be assessed.

The outcome is described as exploratory, with no defined limit to cell types or cytokines. The record lists 43 outcome measures but does not report how outcomes were prioritized or how multiplicity was handled. It also lists 0 analyses, leaving no reported estimates with which to judge the statistical evidence.

The allocation method is not stated for this sub-study, and the registry cautions that the arms are not a like-for-like control comparison. This restricts interpretation of any group contrast as a treatment effect. The record also does not report a detailed statistical analysis plan, sample-size justification, interim analyses, assay details or a causal assessment of the separately listed death. These gaps limit both reproducibility and the strength of conclusions that can be drawn from the registry information.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Mechanistic Inflammatory Sub-study Embedded Within the Albumin vs Balanced Crystalloid Trial — NCT04963569.
• Registry record last updated 2025-04-15; status last verified 2025-04.
• Results first posted 2025-04-02.
• 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.

19. Record Summary

The registry record describes a completed mechanistic inflammatory sub-study in people with sepsis enrolled in the ABC Sepsis trial. It lists enrollment of 26 and planned sampling at presentation, 12 hours and 24 hours to explore changes in cell types and inflammatory cytokines around fluid treatment. The arms are described as Albumin and Balanced Crystalloid, but the allocation method is not stated in this record, which cautions that the arms are not a like-for-like control comparison.

The primary outcome describes the number eligible for exploratory analysis, but that value and the longitudinal biomarker results are not reported. The record lists 43 outcome measures and 0 analyses. Safety counts list 0 serious adverse events in each arm and deaths of 1 in Albumin and 0 in Balanced Crystalloid. The available registry information supports description of the study aims and reported counts, but not an assessment of biomarker changes or a causal fluid comparison.