Fresh Versus Old Red Blood Cells for Transfusion: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Fresh Versus Old Red Blood Cells for Transfusion (NCT01319552), a non-randomized na study. The primary outcome is Measure of Non-transferrin-bound Iron.

2. Clinical Question

The study addresses a proposed biological response to transfusion, rather than a clinical outcome such as survival or recovery. Its registry summary describes the investigators’ hypothesis that iron associated with stored red blood cells could be delivered into the circulation and relate to differences between fresher and older blood. The primary endpoint is a measure of non-transferrin-bound iron, assessed after each transfusion.

The comparison is within participants: the same person’s response to the fresh transfusion is compared with that person’s response to the older stored transfusion. This design can account for stable differences between participants when estimating a within-person contrast. However, the registry identifies the allocation as nonrandomized and reports no masking. The order of transfusions and any procedures used to address order or period effects are not reported in the registry record.

The central question is therefore whether the measured iron response differs between the two transfusion conditions in this study. The registry record does not provide the observed endpoint values, the estimated contrast, or an uncertainty interval, so it does not establish the size or precision of any difference.

3. Trial Design

This completed study, NCT01319552, is titled Fresh Versus Old Red Blood Cells for Transfusion. Columbia University is listed as the sponsor. The recorded start was 2008-12, and primary completion was 2010-10. The phase is listed as NA. Enrollment was 14.

The design is recorded as a nonrandomized crossover. The intervention descriptions specify autologous transfusions: one unit stored for 3-7 days for the fresh condition and one unit stored for 40-42 days for the old condition, both under standard conditions. The registry summary says that each participant’s own blood was to be transfused both fresh and after storage, with outcomes compared. This makes the intended comparison a within-participant contrast rather than a comparison of unrelated groups.

Random allocation is not reported; instead, allocation is explicitly listed as NON RANDOMIZED. Masking is listed as NONE. The registry record does not report the order in which conditions were given, the interval between transfusions, the method used to assign condition order, or whether any washout period was used. It also does not report a protocol for handling period effects or carryover. These details matter in a crossover because the response to a later transfusion could be affected by timing or by an earlier intervention.

The record lists one outcome measure and zero analyses. It identifies the primary endpoint and its timeframe, but does not report numerical outcome results or a statistical analysis of that endpoint. A planned comparison in the study description is not the same as a posted estimate: the registry record documents what was to be measured, not the resulting size or precision of the contrast.

The trial’s stated focus is a biological marker after transfusion. Its design does not, by itself, provide a randomized comparison of clinical outcomes between fresh and older stored blood. The nonrandomized allocation and small enrollment also constrain how broadly any findings could be interpreted, even if numerical results were available.

4. Participant Flow and Accounting

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

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Measure of Non-transferrin-bound Iron, measured over four hours after transfusion.

The primary endpoint is the measure of non-transferrin-bound iron, recorded in μM. The registry describes the measure as the comparison of the increase in this marker for each participant between the fresh and old blood transfusions, assessed four hours after transfusion. The summary statistic is a mean, with standard deviation specified as the dispersion measure.

Non-transferrin-bound iron is used in the study as a biological measure relevant to the investigators’ hypothesis about iron associated with stored red blood cells. In general, an endpoint linked to a proposed biological mechanism can help characterize whether a physiological response occurs after an intervention. Such a marker is not automatically a measure of patient benefit or harm, and it does not on its own establish that a biological change causes a clinical outcome.

The within-participant framing is important. The endpoint concerns the change for a participant under each transfusion condition, followed by a comparison of those responses. The registry does not report the numerical means, standard deviations, participant-level changes, or estimated difference between conditions. It also does not specify the detailed measurement procedure or how any missing measurements were handled.

Accordingly, the endpoint definition tells readers what the investigators intended to assess and when. It does not reveal the magnitude, direction, or precision of the observed response. Those results are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry specifies a continuous primary measure summarized by a mean and standard deviation. In general, continuous measurements can take values across a scale, rather than being restricted to event categories. A mean conveys the central average, while a standard deviation describes dispersion among observations. These summaries are useful for describing data but do not, by themselves, show whether two conditions differ or how uncertain a comparison is.

The endpoint is framed as a comparison of each participant’s response under two transfusion conditions. In general, a paired analysis uses the within-person differences rather than treating observations from the same participant as independent. Pairing can account for stable characteristics shared by the person’s two measurements. It does not remove all sources of bias: changes over time, condition order, carryover, measurement error, and nonrandomized assignment may still affect the observed contrast. The registry does not report the specific analysis used, the sequence of transfusions, or any adjustment for these design features.

A confidence interval is, in general, a range of values reflecting uncertainty around an estimated quantity under a specified statistical method. It helps readers assess both the direction and precision of an estimate. A narrow interval indicates greater precision than a wide interval, all else equal; neither width alone nor a significance test establishes clinical importance. The registry record does not report a confidence interval for the primary endpoint.

A p-value, in general, describes how unusual data at least as incompatible with a specified null model would be if that model and the analysis assumptions were correct. It is not the probability that the null hypothesis is true, nor does it describe the size or clinical relevance of an effect. No p-value or inferential analysis for the primary endpoint is reported in the registry record. The record therefore does not support a statement that the endpoint difference was statistically significant or nonsignificant.

The record describes a planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. These are assumptions for planning illustrations, not reported characteristics of the study’s original design. The record does not report the trial’s own sample-size justification. Because the primary observed effect is not reported, numerical planning estimates cannot be interpreted from the registry information alone. Such illustrations are not observed power and do not show the probability that this completed study detected an effect.

The planning description specifies participants per arm for non-time-to-event outcomes; this endpoint is continuous. That generic allocation description should not be mistaken for the study’s crossover structure. In general, calculations for a paired crossover comparison depend on the variability of within-participant differences and assumptions about design features. The registry does not report those inputs or a trial-specific calculation. The stated equal-allocation assumption in the planning description does not document how participants were assigned or sequenced in the trial.

Over-dispersion is a general concern when data vary more than a statistical model assumes. If variability is understated, standard errors and uncertainty intervals may be too small, and inferential conclusions can become overconfident. For a continuous paired endpoint, the relevant variability includes the spread of within-participant differences. The registry gives no observed dispersion values, so it is not possible to assess this feature for the study.

Missing data also affect interpretation. In general, a paired comparison requires usable measurements under both conditions for a participant, unless a method for incomplete pairs is used. Missingness can reduce the information available and can bias an estimate if the reason for missing measurements is related to the response. The registry does not report endpoint completion counts, reasons for missingness, or a missing-data method.

Finally, statistical interpretation depends on the analysis population and prespecified analysis choices. In general, readers need to know which participants and measurements were included, how the endpoint was constructed, and whether the analysis followed a prespecified plan. Those details are not reported in the registry record. The registry identifies the intended endpoint, but does not document enough analysis detail to reproduce an estimate or assess model assumptions.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Fresh Transfusion", "Old Transfusion")
n   <- c(14, 14)
m   <- c(0.16, 3.17)
sp  <- c(0.99, 3.26)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Fresh Transfusion  14  0.160  0.990  -0.412  0.732
CARM  Old Transfusion  14  3.170  3.260  1.288  5.052

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and zero analyses. It identifies a primary endpoint, but does not report numerical results for that endpoint or any analysis. No secondary outcome measures are listed in the record. The statistical plan, including any prespecified testing hierarchy or adjustment for multiple comparisons, is not reported in the registry record.

Multiplicity is, in general, the issue of interpreting results when several outcomes, time points, subgroups, or statistical tests are considered. Examining multiple results can increase the chance of a seemingly unusual finding even when no real effect exists. A prespecified analysis strategy can clarify which comparisons are primary and how any additional testing is handled. Because the registry does not report such a strategy, it is not possible to assess whether multiplicity was addressed or whether unreported analyses were performed.

The record specifies the primary measurement at four hours after transfusion. It does not report whether measurements at other times were collected, how endpoint data were checked, or how participants with incomplete measurements were handled. In general, missingness matters especially in a crossover because a participant may contribute a complete pair or only one condition’s measurement. The registry does not give the number of complete pairs or reasons for any missing data.

The analysis population is also not reported in the registry record. The enrollment is 14, but the record does not state how many participants contributed data to the primary endpoint or whether any were excluded from an analysis. The statistical methods used for the endpoint, including any covariate adjustment or sensitivity analyses, are not reported. As a result, readers cannot evaluate whether the analyzed population matched the intended population or how robust any estimate might be.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports a Transfusion group with 14 at risk for serious adverse events and zero affected. For other adverse events, 14 were at risk and three were affected. The record does not list any serious adverse-event terms. Deaths are not reported.

These are counts in the registry’s safety record, not a detailed account of event type, timing, duration, or relationship to a transfusion condition. The entry of three other adverse events does not establish that three different participants experienced events; the record does not clarify whether the count refers to events or unique people. Nor does it describe whether events occurred after the fresh condition, the old condition, or both.

Zero reported serious adverse events is not proof that serious events cannot occur. In general, safety counts from a small study provide limited information about uncommon events, and incomplete event descriptions limit interpretation. The registry does not report further details that would allow a comparison of safety between the two transfusion conditions.

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

The study asks a specific biological question: whether the increase in non-transferrin-bound iron four hours after transfusion differs when participants receive their own red blood cells stored for 3-7 days versus 40-42 days. The registry describes a nonrandomized crossover design, with enrollment of 14, and identifies the intended endpoint.

What the registry record does not provide is the primary endpoint result. It contains no numerical mean, contrast between conditions, uncertainty interval, or inferential analysis for the iron measure. The direction and magnitude of the observed response, and the precision of any estimate, are therefore not reported in the registry record. The endpoint description alone cannot establish whether the responses differed.

Even with an estimate, a biological marker would not automatically answer whether either transfusion condition changes patient outcomes. The study summary presents a hypothesis about iron and possible differences in outcomes, but the listed primary endpoint is a laboratory measure. A marker can inform a proposed mechanism without demonstrating a causal pathway or a clinical effect.

The design also limits what could be inferred about condition effects. A crossover comparison can use within-person information, but this study is recorded as nonrandomized and unmasked. The registry does not report condition order or how timing and carryover were addressed. The small enrollment does not, by itself, show that a result would be absent or present; without an estimate and its uncertainty, precision cannot be assessed.

Thus, the available record establishes the question, the intended comparison, and some safety counts. It does not establish the result of the primary comparison, the statistical significance of any difference, or a clinical benefit or harm. Those conclusions require outcome and analysis information that is not reported in the registry record.

14. Why This Trial Matters Statistically

This trial matters statistically because it illustrates the distinction between a clear research question and a reported result. The registry gives a concrete endpoint, a measurement time, two defined storage conditions, and a within-participant comparison. Yet it does not provide numerical results for that endpoint. Without the estimate and its uncertainty, readers cannot tell what the observed data indicate or how precisely the response was measured.

The crossover structure also highlights why design details matter. Using each participant as their own reference can reduce the influence of stable differences between participants. However, nonrandomized allocation and absent masking leave questions about sequence, timing, and measurement that can affect interpretation. The registry does not report how those issues were handled, so the intended pairing cannot be treated as a complete answer to all sources of bias.

The record also distinguishes a biological endpoint from a clinical outcome. A measure related to the proposed mechanism may be scientifically informative, but it is not a substitute for evidence about patient outcomes. The study summary offers a rationale for examining iron after transfusion; the endpoint definition does not demonstrate that the marker predicts or causes differences in clinical outcomes.

Finally, planning assumptions and observed evidence serve different purposes. A planning illustration based on an observed effect is not the study’s original sample-size justification, and it is not observed power. The registry does not report a trial-specific sample-size justification or the endpoint estimate needed to interpret an observed-effect illustration. Clear separation of these concepts helps readers avoid treating design calculations as 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 main limitation for interpreting this record is the absence of numerical primary outcome results. The registry describes non-transferrin-bound iron and its measurement time, but does not report the observed values, the within-participant contrast, or its uncertainty. No primary endpoint analysis is listed.

The study enrolled 14 participants. The registry does not report the number with complete measurements under both conditions, so the primary analysis population cannot be determined from the record. It also does not describe missing measurements, exclusions, or how incomplete pairs were handled.

The crossover is recorded as nonrandomized and unmasked. The condition order, assignment process, interval between transfusions, and any approach to period or carryover effects are not reported in the registry record. Without those details, readers cannot assess whether the measured response might be influenced by timing or sequence.

The record specifies a mean and standard deviation for a continuous endpoint but does not provide either summary for the observed data. It does not report a confidence interval, p-value, model, or sensitivity analysis. The trial’s own sample-size justification and detailed statistical plan are also not reported. Planning illustrations based on an observed effect cannot fill these gaps or establish the precision of the completed study.

The outcome is a biological measure assessed four hours after transfusion. The registry does not report clinical outcome measures or establish that the marker represents patient benefit or harm. Safety information is limited to aggregate counts; event details, allocation by transfusion condition, and deaths are not reported. These limitations restrict both statistical and clinical interpretation.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Fresh Versus Old Red Blood Cells for Transfusion — NCT01319552.
• Registry record last updated 2015-06-02; status last verified 2015-05.
• Results first posted 2015-06-02.
• 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.

19. Record Summary

NCT01319552 is a completed, nonrandomized, unmasked crossover study sponsored by Columbia University. Enrollment was 14. The registry describes autologous transfusions using red blood cells stored for 3-7 days and for 40-42 days under standard conditions.

The primary measure is non-transferrin-bound iron, in μM, compared within participants four hours after transfusion. It is specified as a mean with standard deviation. The registry record does not report numerical primary endpoint results or analyses. It lists zero serious adverse events among 14 at risk, three other adverse events among 14 at risk, and does not report deaths.

The record defines the research question and intended endpoint, but does not provide the result or its precision. The transfusion order, missing-data handling, analysis population, and detailed statistical plan are not reported in the registry record.