REstrictive Versus LIbEral Fluid Therapy in Major Abdominal Surgery: RELIEF Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for REstrictive Versus LIbEral Fluid Therapy in Major Abdominal Surgery: RELIEF Study (NCT01424150), a randomized na study. The primary outcome is Disability-free Survival.

2. Clinical Question

The study also described questions about complications, goal-directed therapy, colloids, and a composite of septic complications or death. The fluid regimen was randomly assigned; goal-directed therapy and colloid use were described as measured covariates associated with local practices and beliefs. The registry does not report a posted analysis of those possible effect modifiers.

The primary outcome combines survival with a definition of disability based on health status. The record defines disability as a persistent reduction in health status, assessed with a 12-item version of the World Health Organisation Disability Assessment Schedule. The stated threshold is 24 points, reflecting a disability level of at least 25%. The outcome was assessed up to 1 year after surgery.

3. Trial Design

RELIEF is a completed, randomized, parallel-group trial comparing restrictive with liberal fluid therapy in major abdominal surgery. The registry records single masking, but who was masked is not reported in the registry record. The phase is listed as not applicable. The sponsor is Bayside Health. Enrollment is recorded as 3000.

The recorded study start is 2013-07, and primary completion is 2017-09. These dates describe the registry record; they do not establish when every participant was enrolled or followed. The primary outcome timeframe is 1 year postoperative.

Random assignment applied to the fluid strategy. The study description says goal-directed therapy and colloid use would be measured rather than randomly assigned. It proposed examining whether goal-directed therapy modified the effect of fluid restriction, but the registry record does not report an interaction analysis. The record also does not provide an analysis of colloid use as an effect modifier.

The primary outcome results list 1493 participants in the liberal arm and 1490 in the restrictive arm. Those are the denominators reported for that outcome, not a replacement for the registry's enrollment figure. The registry does not report why the outcome denominators differ from the enrollment total, or whether this reflects exclusions, missing outcome data, or another reason.

The record lists secondary outcomes including death, a composite septic outcome or death, sepsis, surgical site infection, pneumonia, acute kidney injury, pulmonary oedema, and duration of time in ICU or HDU. Although 14 secondary outcomes are recorded as posted, detailed results in the record include death, the composite outcome, and sepsis. Results for the other secondary outcomes 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.

Population: 2983 (99.4%) met the inclusion criteria for the modified intention-to treat population (1490 in the restrictive fluid group and 1493 in the liberal fluid group)

6. Primary Endpoint

The registry lists the primary outcome as Disability-free Survival, measured over 1 year postoperative.

The primary endpoint is disability-free survival up to 1 year postoperative. It combines survival with freedom from a persistent reduction in health status. The registry defines disability as a persistent reduction lasting at least 6 months, measured with a 12-item version of the World Health Organisation Disability Assessment Schedule. A score of 24 points is the stated threshold, reflecting a disability level of at least 25% and the boundary between disabled and not disabled under the described WHO guidelines. Participants were to assess disability themselves when able; a proxy report was to be used when they could not. The date of onset of new disability was to be recorded.

This composite outcome captures more than survival alone: a participant must be alive and meet the stated disability criterion to be classified according to the endpoint definition. That makes it relevant to the study's question about survival with functional status, rather than mortality in isolation. The registry refers to further details in a Procedures Manual and Statistical Analysis Plan, but those details are not reported in the registry record.

The posted outcome is represented by participant counts and a binary comparison. The record lists 165 of 1493 participants in the liberal arm and 172 of 1490 in the restrictive arm. Since counts alone do not explain how the registry's result coding maps to the positive outcome label, the statistical contrast should be read as the posted coded endpoint proportion rather than as an independently reconstructed measure of successful disability-free survival.

7. Results Posted in the Registry

• Restrictive: 11.5% (172/1490) versus 11.1% (165/1493) on control; risk difference 0.5 percentage points (95% CI -1.8 to 2.8), risk ratio 1.04 (0.85–1.28); unadjusted two-proportion z-test p = 0.671.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Death

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Restrictive: 6.4% (95/1490) versus 6.4% (96/1493) on control; risk difference -0.1 percentage points (95% CI -1.8 to 1.7), risk ratio 0.99 (0.75–1.30); unadjusted two-proportion z-test p = 0.952.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Restrictive and the control. The interval gives the range of effects that remain compatible with the data.

Composite Septic Outcome or Death

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Restrictive: 21.7% (323/1490) versus 19.8% (295/1493) on control; risk difference 1.9 percentage points (95% CI -1.0 to 4.8), risk ratio 1.10 (0.95–1.26); unadjusted two-proportion z-test p = 0.196.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Restrictive and the control. The interval gives the range of effects that remain compatible with the data.

Sepsis

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Restrictive: 10.6% (157/1481) versus 8.7% (129/1487) on control; risk difference 1.9 percentage points (95% CI -0.2 to 4.0), risk ratio 1.22 (0.98–1.52); unadjusted two-proportion z-test p = 0.075.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Restrictive and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes recorded in the registry cover mortality, septic outcomes, infections, pneumonia, acute kidney injury, pulmonary oedema, and time in ICU or HDU. Detailed result comparisons in the registry record are available for death, the composite septic outcome or death, and sepsis; results for the other listed secondary outcomes are not reported in the registry record. Across the three posted comparisons, the estimated risk differences are close to zero or positive for restrictive minus liberal, and each interval includes zero. The sepsis estimate is closest to excluding zero, but its interval still includes zero.

These results do not independently overturn or confirm the primary finding. They address different outcomes and timeframes, and they are secondary to the prespecified primary question as described in the registry. A secondary endpoint can add context, but it cannot be treated as if it were the primary endpoint, especially when multiple outcomes have been examined.

General statistical explanation: When many outcomes are tested, the chance of at least one apparently notable result arising by chance increases if each is judged separately without adjustment. Multiplicity procedures can control that risk, but the registry does not report a multiplicity plan or adjusted secondary analyses. Unadjusted secondary p-values therefore carry less weight as confirmatory evidence than a single primary analysis with a prespecified plan. They are better treated as exploratory signals requiring cautious interpretation.

The record lists 14 secondary outcomes as posted but contains detailed results for three. That difference limits the ability to assess the full pattern across the secondary outcomes or to determine how the posted comparisons were selected. The number of outcomes alone does not reveal the analysis hierarchy, and that hierarchy is not reported in the registry record.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Death
• Composite Septic Outcome or Death
• Sepsis
• Surgical Site Infection
• Pneumonia
• Acute Kidney Injury
• Pulmonary Oedema
• Total Duration of Time Spend in the ICU or HDU (in Days)

9. Statistical Interpretation

10. Statistical Methods Explained

What does a risk difference represent?

General statistical explanation: A risk difference is the event proportion in one group minus the event proportion in the comparison group. A value of zero represents equal proportions in the two groups. Positive and negative values indicate the direction of the contrast according to the subtraction order. In the posted comparisons, the contrast is restrictive minus liberal. The risk difference is expressed as a proportion, so it describes an absolute difference rather than a relative change.

The primary outcome comparison reports an estimated risk difference of 0.005. The registry also reports the arm proportions as 0.111 and 0.115. These values are posted estimates; they should not be treated as adjusted for baseline characteristics or other covariates, because adjusted estimates are not reported in the registry record. The primary outcome counts and proportions must also be interpreted in light of the endpoint coding described in the registry.

What does the p-value say?

General statistical explanation: A p-value measures how unusual data at least as incompatible with a specified null hypothesis would be, assuming that hypothesis and the statistical model are correct. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. A large p-value does not prove that groups are identical. A small p-value does not by itself show that an effect is important or free from bias.

The posted p-values are labelled unadjusted. They describe each comparison without an adjustment for the number of outcomes or comparisons being examined. The primary p-value is 0.671. For the posted secondary outcomes, the p-values are 0.952 for death, 0.196 for the composite septic outcome or death, and 0.075 for sepsis. These values should be read alongside the estimates and confidence intervals, not in isolation.

Why use a confidence interval?

General statistical explanation: A confidence interval gives a range of estimates compatible with the observed data under the assumptions of the method. Its width conveys precision: narrower intervals indicate more precision, while wider intervals indicate greater uncertainty. A confidence interval that includes zero for a risk difference is compatible with no difference, as well as with some positive and negative differences. It does not assign a probability to the true value lying in that particular interval.

The primary interval runs from -0.018 to 0.028. It includes zero, and it extends from a possible negative difference to a possible positive difference. The interval therefore does not narrowly pin down the direction or size of the contrast. The registry does not describe whether the interval is adjusted for covariates or multiplicity; the posted comparisons are unadjusted.

What can randomization support?

General statistical explanation: Random assignment is intended to make treatment groups comparable, on average, with respect to measured and unmeasured factors at assignment. It supports a causal comparison of assignment strategies when conduct, follow-up, outcome measurement, and analysis preserve that comparison. Randomization does not ensure that every potential source of bias is absent, and it does not resolve missing outcomes or incomplete reporting.

The registry identifies the trial as randomized and parallel. It does not report sequence generation details, allocation concealment, the identity of the masked parties, adherence summaries, crossover information, or the analysis population. Those details are not reported in the registry record, so the strength of the protections around randomization and analysis cannot be assessed from the record alone.

Why do binary outcomes need careful interpretation?

General statistical explanation: A binary outcome classifies each participant into one of two categories, such as an event or no event, under the study's definition. The meaning of the event depends on how the outcome is coded. A participant count is not self-explanatory when a label could refer to a favourable state or an adverse state. For this primary endpoint, the title is disability-free survival, while the posted counts and risk proportions are relatively low; the record does not explicitly explain the coding of the result values. Reporting both the endpoint definition and the result coding would make the interpretation more transparent.

Risk differences are directly interpretable on an absolute scale but do not show how an effect might vary across participant subgroups. Relative measures, such as risk ratios or odds ratios, can express a different scale of comparison, but they do not remove uncertainty or resolve endpoint coding. The registry posts relative estimates as well, but the central interpretation uses the risk difference and its confidence interval because both communicate the absolute contrast and its uncertainty.

What is not established by the record?

The registry record does not report a full sponsor statistical analysis, an analysis plan, a sample-size justification, or interim analyses. It does not describe how missing outcomes were handled, whether analyses followed intention-to-treat principles, or whether adjustments were made for baseline covariates. The record also does not report the proposed interaction analysis for goal-directed therapy. These omissions prevent a full assessment of analysis choices and their consequences. They do not invalidate the posted numbers, but they limit the conclusions that can be drawn from them.

11. Confidence Intervals and Precision

The primary risk difference is 0.005, with a 95% confidence interval from -0.018 to 0.028. The interval crosses zero, so the posted data are compatible with no difference as well as with differences in either direction. Its width indicates that the estimate is not precise enough to determine a narrow range of possible contrasts. The point estimate alone gives a less complete account than the interval because it does not convey this uncertainty.

General statistical explanation: A confidence interval reflects sampling uncertainty under the statistical method and its assumptions. It is not a guarantee that the true effect lies inside the interval, nor does it show the probability that a particular treatment effect is correct. A confidence interval that includes zero does not prove no effect; it indicates that the data do not distinguish the estimated contrast from zero at the corresponding confidence level under the method used.

The secondary estimates also have intervals that include zero. The death risk difference is -0.001, with an interval from -0.018 to 0.017. The composite septic outcome or death risk difference is 0.019, with an interval from -0.010 to 0.048. The sepsis risk difference is 0.019, with an interval from -0.002 to 0.041. These ranges vary in width and in the positive or negative differences they accommodate. Their uncertainty should be considered alongside the fact that they are unadjusted secondary comparisons.

Precision is not the same as validity. A narrow interval could still be misleading if an analysis were biased or based on an unsuitable outcome definition. Conversely, an interval can be wide even when a trial is randomized and carefully conducted. The registry record does not provide enough detail to assess all analysis assumptions or the handling of missing results.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are participants per arm, not total events. They are illustrations calculated from the observed effect, not a report of the trial's own design calculation. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario uses the stated fraction of the observed effect. The original trial's sample-size assumptions are not reported in the registry record.

The 80% power column gives the estimated participants per arm for a planning scenario that would have an 80% probability of detecting the assumed effect at the stated significance level, if the assumptions and model are correct. The 90% power column gives the corresponding per-arm planning figure for a higher probability of detection. These are planning concepts, not guarantees about what a future study would find.

For the primary endpoint, planning for 100% of the observed effect gives 64,973 participants per arm for 80% power and 86,980 participants per arm for 90% power. For 75% of the observed effect, the figures are 114,960 and 153,898 participants per arm; for 50%, they are 257,422 and 344,615. Smaller effects require more participants to distinguish them from random variation with the same planning assumptions. The figures are not observed power, and they do not describe the actual trial enrollment plan.

The sample-size calculations use the observed effect as their starting point. An observed effect can be unstable, especially when the estimate is uncertain, so these illustrations should not be treated as a recommended target or as evidence that the original study was adequately or inadequately sized. The original design justification is not reported in the registry record.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.111; treatment proportion p₁ = 0.115; average p̄ = 0.113.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4477 = 0.8775.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4477 = 0.3768.
4. n = (0.8775 + 0.3768)² / (0.0049)² = 64,973.0, rounded up to 64,973 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.110515740120563                          # control proportion
p1 <- 0.115436241610738                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1424150)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 64,973 (80% power), 86,980 (90% power)
 75% of observed effect: n per arm = 114,960 (80% power), 153,898 (90% power)
 50% of observed effect: n per arm = 257,422 (80% power), 344,615 (90% power)
power.prop.test: 64973 
Simulated power at n = 64,973 per arm: 0.799 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Multiplicity, Missing Data and Analysis Populations

General statistical explanation: Multiplicity arises when several hypotheses or outcomes are tested. If each test is interpreted at the same threshold without adjustment, the overall chance of a false-positive result across the set can be higher than for a single test. A prespecified hierarchy or adjustment method can help address this issue. The registry record does not report a multiplicity strategy, outcome hierarchy, or adjusted secondary analyses. The secondary p-values should therefore be interpreted as unadjusted and not as definitive evidence across the entire set of outcomes.

The registry says the study would assess whether the effect of fluid restriction was similar with or without goal-directed therapy, using a statistical test of interaction. An interaction asks whether the treatment contrast differs across levels of another factor. The record does not report the interaction estimate, its uncertainty, or the analysis. The presence of a stated hypothesis is not evidence that the analysis was conducted or that it supported a particular conclusion. Colloid use was also described as a measured covariate, but an associated result is not reported in the registry record.

Missing data are an important issue for disability-free survival because the outcome involves participant-reported health status, with proxy reporting when a participant could not assess disability. The registry describes that measurement approach but does not report how many outcomes were missing, how missing observations were handled, or whether sensitivity analyses were conducted. It also does not report whether all randomized participants were included in each analysis or how exclusions were addressed.

The record does not report a full statistical analysis plan, sponsor-reported analyses, interim analyses, or the original sample-size justification. It also does not report whether baseline adjustment was used. These absences limit assessment of prespecification, analysis populations, and robustness. They should not be filled with assumptions about common practice. The posted summary comparisons can be read on their stated terms, while questions about missingness and analysis choices remain unresolved in the registry record.

15. 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 1% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: surgical site infection (21.7%), sepsis (10.5%), pneumonia (3.8%), anastomotic leak (3.3%).

The registry lists serious adverse event figures for both arms. Serious events affected 295 of 1493 participants in the liberal arm and 323 of 1490 in the restrictive arm. It also lists other adverse event figures of 78 in the liberal arm and 79 in the restrictive arm, with 1493 and 1490 participants at risk, respectively. Death counts recorded with the adverse event data are 96 in the liberal arm and 95 in the restrictive arm.

The registry lists surgical site infection, sepsis, pneumonia, and anastomotic leak among the leading serious event terms, with percentages of 21.7%, 10.5%, 3.8%, and 3.3%, respectively. The record does not explain the denominator or calculation basis for these percentages, so they should be read as registry-listed percentages rather than converted into participant-level risks.

General statistical explanation: Adverse event summaries describe events observed during study follow-up; they do not by themselves establish that the assigned intervention caused an event. Comparisons also depend on how events were defined, collected, and attributed. The registry record does not provide detailed event definitions, severity criteria, attribution procedures, or a comparative safety analysis. These summaries therefore describe the posted counts without supporting a causal conclusion.

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

The posted primary comparison is a small estimated risk difference, 0.005, with a confidence interval from -0.018 to 0.028 and an unadjusted p-value of 0.671. The interval includes zero and allows contrasts in either direction. The record therefore does not provide clear statistical evidence of a difference in the coded primary endpoint between assigned fluid strategies. It also does not establish equivalence or rule out effects of a particular size.

The endpoint is described as disability-free survival, an outcome combining survival and freedom from persistent disability. However, the recorded counts and proportions do not make the coding of the result values explicit. They should not be read as a direct count of participants who achieved the positive endpoint without clarification of that coding. The registry record does not report further analysis details that would resolve the issue.

The secondary results add context but remain uncertain. The posted comparisons for death, the composite septic outcome or death, and sepsis have confidence intervals that include zero, and the analyses are unadjusted. They do not establish that the fluid strategies have identical effects, nor do they prove that any observed difference is caused by the assigned regimen. The other listed secondary outcomes do not have detailed results reported in the registry record.

General statistical explanation: A trial result is not a universal answer for every setting or participant. Interpretation depends on the population studied, the interventions as delivered, outcome definitions, follow-up, and analysis. The registry identifies adults undergoing major abdominal surgery as the relevant population, but does not provide enough detail in this record to assess all of those factors. Statistical uncertainty and reporting gaps should remain part of the interpretation rather than being replaced by assumptions.

The result does not establish the mechanism of any difference, resolve the role of goal-directed therapy or colloids, or answer every complication question posed in the study description. It describes posted summaries for a randomized comparison, with limited analysis information in the registry record.

17. Why This Trial Matters Statistically

This trial raises a statistically important question because a treatment strategy can affect more than a single clinical event. The primary endpoint combines survival with persistent disability, while secondary outcomes include death and postoperative complications. Such outcomes do not necessarily move in the same direction, and a composite or functional endpoint requires careful definition and transparent coding.

The randomized comparison is useful for studying assigned fluid strategies, but interpretation depends on more than randomization alone. The record identifies single masking without saying who was masked, and it does not report details of allocation procedures, analysis populations, missing outcomes, or adherence. Those details affect how confidently a reader can connect the reported comparison to the causal effect of assignment.

The primary estimate is close to zero, yet its confidence interval spans differences on both sides of zero. That distinction illustrates why a point estimate and a p-value are not enough to describe a result. The interval communicates the precision limitations, while the p-value does not establish equivalence or exclude a meaningful effect.

The trial description also distinguishes randomized treatment from measured practices, including goal-directed therapy and colloid use. That distinction matters: random assignment can support a treatment comparison, while associations involving measured covariates can be affected by differences between participants or clinical settings. The registry does not report the proposed interaction analysis, so the effect-modification question remains unresolved in this record.

Finally, the posted secondary results are unadjusted, and the record does not describe how multiplicity was handled. The combination of a defined primary question, uncertainty around the primary estimate, and incomplete reporting of secondary and analysis details makes this a useful example of separating observed summaries from stronger claims that the record cannot support.

18. 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.

Several trial-specific limitations of interpretation arise from the registry record. The primary outcome is titled disability-free survival, but the reported counts and proportions do not clearly state whether the counted category represents achievement of the positive endpoint or another coding of the outcome. That ambiguity matters when interpreting the direction of the risk difference. The record gives a detailed disability definition but refers to a Procedures Manual and Statistical Analysis Plan whose additional details are not reported in the registry record.

The registry reports enrollment of 3000, while the primary outcome denominators are 1493 and 1490. It does not explain the difference or describe the analysis population. Denominators also vary among secondary outcomes. The record does not report missing-outcome counts, exclusions, or methods for handling missing data, so the reason for these differences cannot be established from the posted information.

The masking field says single, but the identity of the masked parties is not reported. Details of sequence generation, allocation concealment, adherence, crossover, and outcome adjudication are also not reported in the registry record. The registry lists intervention descriptions, but the full arm descriptions are not available in this record.

Only three secondary result comparisons are detailed, despite 14 secondary outcomes being recorded as posted. The record does not report multiplicity adjustment, an outcome hierarchy, or the proposed interaction analysis for goal-directed therapy. It also does not provide a complete sponsor analysis, interim analysis information, or the original sample-size justification. These gaps restrict assessment of prespecification and robustness; they do not justify guessing what the protocol or analysis plan contained.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: REstrictive Versus LIbEral Fluid Therapy in Major Abdominal Surgery: RELIEF Study — NCT01424150.
• Registry record last updated 2026-01-20; status last verified 2017-08.
• Results first posted 2025-01-24.
• 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.

22. Record Summary

The registry describes RELIEF as a completed, randomized, parallel-group trial of restrictive versus liberal fluid therapy for major abdominal surgery, with enrollment recorded as 3000 and single masking. The primary endpoint was disability-free survival up to 1 year postoperative. The posted primary risk difference is 0.005, with a 95% confidence interval from -0.018 to 0.028 and an unadjusted p-value of 0.671.

Detailed secondary comparisons are reported for death, a 30-day composite septic outcome or death, and 30-day sepsis. Their intervals include zero. The registry record does not report detailed results for the other listed secondary outcomes, the proposed interaction analysis, a full statistical analysis plan, missing-data methods, or the original sample-size justification. The endpoint coding and analysis details should therefore be considered when interpreting the posted counts and contrasts.