Transfusion Strategies in Pediatric Cardiothoracic Surgery: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Transfusion Strategies in Pediatric Cardiothoracic Surgery (NCT00350220), a randomized phase 2 study. The primary outcome is Mean Arterial Lactate Level.

2. Clinical Question

The study addressed uncertainty about the hemoglobin level at which infants and children should receive red-cell transfusions after this surgery. Its central comparison was between a more liberal transfusion threshold and a more restrictive approach that also required clinical indications. The posted primary outcome is a continuous measure of arterial lactate, rather than a direct measure of transfusion use or a patient-centered event. The registry also lists outcomes related to postoperative oxygen use, ventilation, vasoactive-agent administration, blood volume transfused, and mortality before hospital discharge.

3. Trial Design

The registry describes a prospective, randomized, controlled clinical trial with a parallel-group design. It is listed as phase 2, with no masking. The University of Rochester is named as sponsor. The record lists enrollment of 60 and a completed status. The recorded study start is 2006-07 and primary completion is 2009-09.

The two groups followed different transfusion strategies after cavopulmonary connection. The high-hemoglobin group had a goal above 13g/dl, with red-cell transfusion specified for hemoglobin under 13g/dl regardless of clinical indication. The low-hemoglobin group was intended to avoid transfusion unless hemoglobin was under 9.0 g/dl; its description also required clinical indications. Both arm descriptions specify a transfusion volume of 10cc/kg when their stated criteria were met.

The posted primary outcome is mean arterial lactate over the first 48 hours after surgery. The registry lists six secondary outcome descriptions: oxygen utilization during the 8 Hour to 72 Hours Post-operative Period, length of mechanical ventilation, length of oxygen use, length of vasoactive-agent administration, volume of blood transfused, and mortality before hospital discharge. The record lists eight outcome measures overall, but does not report additional descriptions in the outcome information summarized in the record.

Random allocation is reported, but the method used to generate the allocation sequence, conceal allocation, or handle deviations from assigned strategies is not reported in the registry record. Masking is listed as none. Details about who knew assignment, adherence to the thresholds, co-interventions, and follow-up procedures are 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.

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 Mean Arterial Lactate Level, measured over 48 hours.

The primary endpoint is mean arterial lactate level, measured in mmol/L, for the first 48 hours after surgery. Lactate is a continuous laboratory measure, so the registry presents a mean and standard deviation for each group. The comparison is expressed as a difference between group means.

In general, a continuous physiologic measure can help describe how groups differ across a postoperative period without reducing each participant’s result to a yes-or-no event. A group mean gives a summary of the measured values, while the standard deviation describes how dispersed those values are around the mean. Neither alone shows the full distribution of individual results.

The record does not report how often lactate was measured, how measurements across the 48-hour period were combined for each participant, or how missing measurements were handled. It also does not report a threshold at which a lactate difference would be clinically important. Therefore, the numerical comparison can be read as a comparison of recorded group averages, but the registry information does not establish how that difference should be translated into clinical importance.

7. Results Posted in the Registry

• Low Hb (Restrictive) Group: mean 1.40 versus 1.40 on control; difference 0.00 (95% CI -0.23 to 0.23), standardized effect d = 0.00; z-test p = 1.000.

8. Statistical Interpretation

9. Statistical Methods Explained

The registry posts group means, standard deviations, a mean difference, a confidence interval, and an unadjusted p-value for the primary continuous outcome. It does not report enough detail to identify the sponsor’s full statistical analysis plan or the exact test used. The explanations below distinguish the posted statistics from general statistical concepts.

What does the mean difference represent?

The mean difference compares the average arterial lactate in the low-hemoglobin group with the average in the high-hemoglobin group. The posted estimate is 0.0 mmol/L, with the group means each reported as 1.4 mmol/L. The direction of a mean difference depends on the order of subtraction; the registry identifies the comparison as the low-hemoglobin group relative to the high-hemoglobin group.

In general, a mean difference retains the original measurement units. It can therefore be interpreted in mmol/L rather than as a standardized score. The mean is sensitive to the values observed in the participants, and a similar average between groups does not imply that every participant had a similar result. Individual-level distributions are not reported in the registry record.

What do the standard deviations describe?

The registry reports standard deviations of 0.4 mmol/L in the high-hemoglobin group and 0.5 mmol/L in the low-hemoglobin group. In general, a standard deviation describes spread around a group mean. It is not a confidence interval for the mean and is not the standard error of the comparison. A larger standard deviation indicates greater dispersion in the recorded values, but group summaries alone do not show whether the data are symmetric, contain outliers, or have other distributional features.

The record does not report a distributional assessment, data transformation, or robust method. It also does not report whether the analysis accounted for repeated lactate measurements over time or used a single summary value for each participant. Those details matter because multiple measurements from one participant are generally related rather than independent. The outcome description identifies a mean over the first 48 hours, but the exact construction of that summary is not reported in the registry record.

How should the confidence interval be read?

In general, a confidence interval describes uncertainty around an estimated quantity under a statistical procedure and its assumptions. For this comparison, the reported interval crosses zero. That means the interval includes no mean difference and also includes values on both sides of zero. Its width gives information about the precision of the estimate: a narrower interval indicates less uncertainty than a wider interval on the same measurement scale.

A confidence interval is not a guarantee that the true difference lies within the particular interval, and it is not a range containing most individual participants’ results. It pertains to the estimated group difference. The exact method used to construct the interval is not reported in the registry record, so its coverage assumptions and any small-sample adjustment cannot be checked from the posted information.

What does the p-value add?

The registry reports an unadjusted p-value of 1.0. In general, a p-value summarizes how often data at least as incompatible with a null model would arise if that model and the analysis assumptions held. It does not say whether a result is clinically meaningful, and it does not give the probability that the null model is correct. The registry does not identify the test statistic, whether the test was one- or two-sided, or whether adjustment was made for covariates. The reported value is explicitly unadjusted, so it should not be described as a multiplicity-adjusted result.

The confidence interval and p-value address related questions but do not replace judgment about the size and importance of a difference. The interval communicates a range of estimates compatible with the analysis procedure, while the p-value summarizes evidence against a null hypothesis. Neither provides information about the clinical threshold that would matter for this endpoint; that threshold is not reported in the registry record.

What can be said about planning assumptions?

The record describes a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. It also states that the effect to detect is a stated fraction of the observed effect. The numerical planning results are not included in the registry information, and no numerical sample-size recommendation can be stated from it.

In general, planning calculations ask how many participants would be needed under specified assumptions to detect a chosen effect with a selected power and significance level. Such figures are prospective illustrations, not evidence that a completed trial had a particular power after observing its results. The registry does not report the trial’s own sample-size justification, target effect, or design power. The planning illustration should therefore not be mistaken for the trial’s original design rationale or for observed power.

More broadly, the registry record does not report the analysis population, handling of missing outcome measurements, adjustment variables, or protocol deviations. These omissions limit independent assessment of how the posted group summaries were derived. The randomized design is relevant context, but it does not fill in analysis details that are absent from the record.

10. Confidence Intervals and Precision

The estimated mean difference is 0.0 mmol/L, and the reported confidence interval extends to both sides of zero. Thus, the posted estimate is centered on no difference, but the interval reflects uncertainty around that estimate. Because the interval includes values in either direction, the result does not isolate a single direction of effect.

In general, precision is assessed by the range of estimates compatible with the data under the interval method. A narrower confidence interval gives a more precise estimate than a wider interval for the same outcome and scale. The registry interval is expressed in the outcome’s units, mmol/L, so its bounds describe possible differences in mean lactate rather than variation among individual lactate values. The interval is not a statement about where most participants’ measurements fall.

The arm-specific standard deviations, 0.4 mmol/L and 0.5 mmol/L, describe observed spread. They should not be substituted for the confidence interval when judging precision of the group comparison. The registry also reports different standard errors for the two group means, but the method used to form the comparison interval is not reported in the registry record.

No clinically important difference or equivalence margin is reported. Consequently, the interval cannot be classified from the registry information as excluding all clinically meaningful differences. The confidence interval communicates statistical uncertainty, while clinical importance requires a threshold grounded in the endpoint’s interpretation. That threshold is unavailable in the record.

The exact interval bounds are not repeated in this explanation. The important statistical reading is that the posted interval crosses zero and allows differences on either side, while the estimate itself is 0.0 mmol/L. The unadjusted p-value of 1.0 is consistent with the absence of evidence against a zero-difference null under the reported analysis, but it does not prove equivalence or establish that the true difference is exactly zero.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists six secondary outcome descriptions in addition to the primary mean arterial lactate outcome. It also lists eight outcome measures overall. The record does not report results for those secondary outcomes in the information summarized for this analysis. It lists zero analyses, and does not provide an analysis plan describing the order, hypotheses, or statistical methods for outcomes.

Multiplicity arises when several outcomes or statistical comparisons are examined. In general, making many comparisons can increase the chance that at least one result appears notable by chance if each is assessed without adjustment. Whether and how to adjust depends on the scientific questions and analysis plan. Since the registry record does not report a multiplicity strategy, the unadjusted p-value for the primary outcome should not be represented as adjusted for other outcomes or comparisons. It is also not possible to reconstruct a family-wise interpretation from the record.

The registry does not report missing-data counts for the primary endpoint, the method used to handle missing measurements, or whether all randomized participants were included in the analysis. The analysis population is not reported in the registry record. Therefore, the posted group summaries cannot be assumed to represent an intention-to-treat, per-protocol, or other specified population without additional information.

In general, missing outcome data can affect an estimate when availability is related to a participant’s condition, treatment assignment, or measured outcome. Statistical approaches make assumptions about missingness, and those assumptions should be described to assess their implications. The registry record does not report whether missingness occurred or what assumptions were used. It also does not report adherence, crossovers, or exclusions after randomization.

These absences do not establish that missing data or multiplicity problems occurred. They mean that the registry record does not contain enough information to evaluate those issues. The randomized allocation and posted summaries can be described as recorded, while conclusions about analysis populations, missing-data handling, and adjustments must remain limited.

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 0% reporting threshold.

The posted adverse-event figures show 0 serious adverse events and 0 other adverse events in each group. Each group has 30 participants listed as at risk for both serious and other adverse events. These are event counts as recorded in the registry; they do not establish that every possible event was actively sought or captured in the same way.

Death counts are not reported in the registry record. Although mortality before hospital discharge is listed as a secondary outcome, the corresponding death figures are unavailable in the adverse-event information. The listed outcome should not be read as a reported result.

In general, an observed count of zero is not proof that an event cannot occur. The interpretation depends on how events were defined, monitored, attributed, and followed over time. The registry record does not report those details, nor does it provide event terms for serious adverse events. Therefore, the adverse-event figures support only a narrow statement about the counts posted, not a broad conclusion about comparative safety.

The absence of reported event counts for deaths must be kept distinct from the zero counts recorded for serious and other adverse events. They represent different reporting states and should not be combined into a claim that no deaths occurred.

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

The primary result is a close match between the posted group means: mean arterial lactate is 1.4 mmol/L in each group, with a mean difference of 0.0 mmol/L. The confidence interval crosses zero, and the unadjusted p-value is 1.0. These statistics describe the posted comparison for lactate during the first 48 hours after surgery.

The result does not prove that the transfusion strategies are equivalent. Equivalence requires a defined margin and an analysis designed to assess whether the interval excludes differences beyond that margin. No equivalence margin is reported in the registry record. The result also does not prove that there is no difference in the underlying population: an estimate of zero can coexist with uncertainty about values on either side of zero.

The result does not establish that one strategy is preferable for all outcomes. The registry identifies secondary outcomes involving oxygen utilization, mechanical ventilation, oxygen use, vasoactive-agent administration, blood volume transfused, and mortality before hospital discharge, but their results are not reported in the registry record for this analysis. The primary lactate comparison cannot answer those distinct questions.

Nor can the posted summaries establish how the strategies affected individual patients. Group means and standard deviations do not describe each person’s experience, and details of measurement timing, missing observations, adherence, and analysis population are not reported. The adverse-event record lists zero serious and other adverse events in each group, while death counts are not reported; this is not enough to characterize all safety outcomes.

The appropriate interpretation is therefore bounded: the recorded average lactate values were the same to the reported precision, and the estimated mean difference was 0.0 mmol/L. The interval indicates uncertainty around that estimate. The registry record does not provide the additional design, analysis, or outcome information needed to extend this observation into a general claim about clinical effectiveness or safety.

14. Why This Trial Matters Statistically

This trial is statistically instructive because the primary outcome is continuous and the posted estimate is centered at zero. It illustrates why an observed mean difference, its confidence interval, and a p-value answer related but distinct questions. The estimate describes the observed contrast, the interval conveys uncertainty, and the p-value summarizes compatibility with a null model under the analysis assumptions.

The trial also highlights why a non-significant or zero-centered result should not automatically be called equivalence. A conclusion of equivalence depends on an explicitly defined margin and an analysis suited to that objective. The registry record does not report such a margin. Without it, the statistical result cannot distinguish a sufficiently small difference from a potentially important difference that remains uncertain.

Another useful feature is the contrast between the endpoint and the intervention. The randomized groups followed different hemoglobin thresholds, while the primary outcome was mean arterial lactate over a postoperative period. These are different kinds of measures: one describes the treatment strategy, and the other summarizes a physiologic outcome. Additional outcomes listed in the registry address other aspects of postoperative care, but their results are not reported in the registry record for this analysis.

Finally, the record makes clear that trial registration details and full statistical reporting are not interchangeable. It identifies allocation, masking, arms, outcomes, and primary summaries, but does not report key analysis details such as missing-data handling, analysis population, or the test underlying the p-value. Recognizing what is and is not recorded is part of sound statistical interpretation, especially when a concise numerical result may otherwise invite broader conclusions.

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 registry record leaves several trial-specific details unreported. Although allocation is listed as randomized, the sequence-generation and concealment methods are not reported. Masking is listed as none, but the record does not specify which personnel or participants knew group assignment. It also does not describe adherence to the assigned transfusion thresholds, departures from strategy, or crossover between groups.

The primary endpoint is described as mean arterial lactate for the first 48 hours after surgery, but the measurement schedule and method for combining observations into a participant-level value are not reported. The analysis method, assumptions, and handling of repeated measurements are also not reported. Missing-data counts and methods, the analysis population, and covariate adjustment are not reported in the registry record.

The registry lists six secondary outcome descriptions and eight outcome measures overall, but the secondary results are not reported in the registry record for this analysis. Mortality before hospital discharge is listed as an outcome, while death counts are not reported. The adverse-event record lists zero serious and other events per group, but does not include event terms or monitoring details.

The record does not report the trial’s sample-size justification, target effect, or design power. A planning illustration is described using the observed effect and stated assumptions, but numerical planning results are not included in the record information used. This illustration cannot be treated as the trial’s original design rationale or as observed power.

Finally, no clinically important lactate difference, equivalence margin, or non-inferiority margin is reported. These omissions limit interpretation of whether the interval excludes differences that would matter clinically. The posted estimate remains interpretable as a group comparison, but the registry information does not support stronger conclusions about equivalence, the full outcome profile, or the trial’s analysis robustness.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Transfusion Strategies in Pediatric Cardiothoracic Surgery — NCT00350220.
• Registry record last updated 2015-06-26; status last verified 2015-05.
• Results first posted 2010-02-25.
• 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, phase 2, randomized, parallel-group trial with no masking, sponsored by the University of Rochester. It lists enrollment of 60 and compares a high-hemoglobin strategy with a low-hemoglobin strategy after pediatric cavopulmonary connection.

The primary outcome is mean arterial lactate during the first 48 hours after surgery. The group means are 1.4 mmol/L in both arms, and the posted mean difference is 0.0 mmol/L, with an unadjusted p-value of 1.0. The reported confidence interval crosses zero.

The registry lists secondary outcomes, but their results are not reported in the information summarized for this analysis. Serious and other adverse-event counts are 0 in each group; death counts are not reported. Details of the analysis population, missing-data handling, and statistical test are not reported in the registry record.