Prevention of Acute Kidney Injury in Cardiac Surgery Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Prevention of Acute Kidney Injury in Cardiac Surgery Patients (NCT00921518), a randomized phase 2 study. The primary outcome is Number of Participants With Acute Kidney Injury.

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group phase 2 trial with triple masking. The record lists 92 participants as enrolled and marks the study completed. The study start is recorded as 2009-01, with primary completion recorded as 2012-01. The sponsor is Community Medical Center, Scranton, PA.

Participants were assigned to Normal Saline or Sodium Bicarbonate. The posted outcome counts are 20 participants with AKI among 48 in the saline arm and 14 among 44 in the bicarbonate arm. These are the observed group counts for the primary outcome; they should not be confused with estimates of what would occur in a larger population.

The saline regimen is described as beginning before surgery and continuing through cardiopulmonary bypass and for six hours following bypass. The bicarbonate description specifies the solution and the initial infusion rate, then says the rate is reduced during bypass; its final wording is truncated in the registry record. The brief description identifies intravenous hydration with isotonic saline or sodium bicarbonate 150 mEq/L as the comparison of interest.

Randomization can help make treatment groups comparable on average, while masking can reduce the influence of treatment knowledge on care or outcome assessment. The registry identifies the trial as randomized and triple-masked but does not report details about the randomization procedure, who was masked, or how masking was maintained. It also does not report the trial's original sample-size justification or its full statistical analysis plan.

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 Number of Participants With Acute Kidney Injury, measured over 5 days.

The primary endpoint was the number of participants with acute kidney injury, with the assessment window recorded as 5 days. The outcome description says that AKI was defined using the AKIN criteria. This is a binary participant-level endpoint: each participant is classified as having or not having the event under the stated criteria during the assessment period. The registry record does not provide further detail on how the criteria were operationalized or how the assessment was conducted.

A binary endpoint is useful when the clinical question concerns whether an event occurred, rather than how much a continuous measurement changed. It allows the event counts in the randomized groups to be compared directly and supports estimates of the difference in event risk between groups. The count does not describe the severity, duration, or consequences of AKI, and the registry record does not report those dimensions as primary outcomes.

The background description notes that acute kidney injury has no uniform criteria and mentions a commonly used definition based on a rise in serum creatinine. However, the posted primary endpoint specifically identifies AKIN criteria. The broader background definition should therefore not be substituted for the endpoint definition used in the registry record.

7. Results Posted in the Registry

• Sodium Bicarbonate: 31.8% (14/44) versus 41.7% (20/48) on control; risk difference -9.8 percentage points (95% CI -29.4 to 9.7), risk ratio 0.76 (0.44–1.32); unadjusted two-proportion z-test p = 0.328.

8. Statistical Interpretation

9. Statistical Methods Explained

The posted comparison is a risk-difference analysis for a binary outcome. A risk is the proportion of participants in a group who experience the event over the stated assessment period. The risk difference expresses the absolute difference between the two group risks, with the comparison oriented as Sodium Bicarbonate relative to Normal Saline. A negative estimate in this orientation points toward a lower observed event risk in the bicarbonate group; interpretation still depends on the uncertainty around that estimate.

What does the confidence interval communicate?

A confidence interval gives a range of effect estimates that are reasonably compatible with the observed data under the method's assumptions. It is not a range containing a fixed proportion of individual participant outcomes, and it is not a probability statement that the true effect lies in the particular calculated interval. For this comparison, the registry interval includes the no-difference value and also extends into estimates favoring either group. That breadth signals uncertainty about the size and direction of the underlying effect.

What does the p-value add?

The registry reports an unadjusted p-value of 0.328. A p-value evaluates the compatibility of the observed data with a specified null model, conditional on the model and analysis assumptions. It does not measure the probability that a treatment effect exists, and it does not show whether an effect is clinically important. An unadjusted p-value also does not account for multiplicity from multiple outcomes or analyses. The registry reports one primary outcome measure and no secondary outcomes, but the complete analysis plan and any additional analyses are not reported in the registry record.

Why can the estimate be uncertain?

Binary outcome estimates depend on the number of events as well as the number of participants observed. When event counts are limited, the estimated group risks and their difference can vary substantially across samples. Random allocation supports a fair comparison in expectation, but it cannot eliminate sampling variation. A confidence interval reflects this uncertainty rather than resolving it.

The interval and p-value are tied to a statistical model and its assumptions. The registry record does not identify the exact test or method used for the sponsor's primary comparison, nor does it report whether the comparison adjusted for any baseline variables. The posted calculation is therefore best understood as the recorded unadjusted comparison, without claiming additional model details.

Why distinguish relative and absolute measures?

The risk difference is an absolute measure: it describes how far apart the groups are on the risk scale. Relative measures such as a risk ratio or odds ratio express the comparison differently and can give a different impression of magnitude, particularly when event risks are not small. The registry record includes relative estimates as well as the risk-difference comparison, but their interval limits are not repeated in this explanation. Each measure addresses the same group contrast using a different scale; none removes the uncertainty in the observed data.

For a binary outcome, an odds ratio compares the odds of an event rather than the event risks themselves. A risk ratio compares risks directly. Neither should be described as an absolute reduction. The risk difference is often more directly interpretable as a difference in event frequency, while the relative measures may be useful for other comparisons. Interpretation should remain anchored to the observed group counts, the endpoint definition, and the uncertainty interval.

What can the reported analysis support?

The registry supports a descriptive statement about the event counts and a statistical comparison between randomized groups. It does not establish causation beyond the strengths and limitations of the trial, and it does not report enough analytic detail to evaluate every modeling choice. The absence of a reported analysis plan, missing-data approach, or analysis-population definition limits the ability to determine whether the posted comparison follows a prespecified strategy. These details are not reported in the registry record.

10. Confidence Intervals and Precision

The confidence interval for the risk difference is broad and crosses the value representing no group difference. It therefore includes possibilities in which the event risk is lower with Sodium Bicarbonate and possibilities in which it is higher. The observed count pattern favors bicarbonate, but the interval indicates that the data do not pin down the direction or magnitude of the population difference with precision.

Precision depends in part on how much information the trial contains. The analysis includes 48 participants in the Normal Saline group and 44 in the Sodium Bicarbonate group, with 20 and 14 AKI events, respectively. Those counts provide a limited basis for estimating a treatment contrast. A larger study could produce a narrower interval if it accrued more information, although the actual width would also depend on the event pattern and the analysis method.

The p-value of 0.328 is consistent with the uncertainty conveyed by the interval. Neither statistic proves that there is no difference. Conversely, the direction of the observed counts does not show that a real benefit exists. The data leave substantial uncertainty about the effect that might be seen in the target population.

Confidence intervals describe statistical uncertainty under specified assumptions; they do not account automatically for possible bias, deviations from the planned conduct, or limitations in the endpoint definition. The registry record does not report sufficient detail to assess these issues fully. Precision should therefore be considered alongside the study's design, outcome ascertainment, and the limited information available about the analysis.

11. 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 based on the observed effect and the stated planning assumptions, rather than the trial's original sample-size calculation. The assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set as a stated fraction of the observed effect.

For planning at 100% of the observed effect, the illustration lists 375 participants per arm for 80% power and 502 participants per arm for 90% power. For planning at 75% of the observed effect, it lists 677 participants per arm for 80% power and 906 participants per arm for 90% power. For planning at 50% of the observed effect, it lists 1,542 participants per arm for 80% power and 2,064 participants per arm for 90% power.

The 80% power column describes the planned chance of detecting the specified effect under the assumptions if that effect were the true one; the 90% column uses a higher target chance of detection. These are planning scenarios, not observed power. Smaller effects require more participants because they are harder to distinguish from random variation. The curve should be read as a visual representation of the same planning relationship, not as evidence about what the trial actually had power to detect.

The original trial design assumptions, including any sponsor sample-size justification, are not reported in the registry record. The planning illustrations should not be interpreted as evidence that the trial was designed around the observed effect or that its results prove or disprove a treatment effect.

12. 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 = 0.417; treatment proportion p₁ = 0.318; average p̄ = 0.367.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6818 = 1.3363.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6782 = 0.5708.
4. n = (1.3363 + 0.5708)² / (-0.0985)² = 375.0, rounded up to 375 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 <- 0.416666666666667                          # control proportion
p1 <- 0.318181818181818                          # 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(921518)
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 = 375 (80% power), 502 (90% power)
 75% of observed effect: n per arm = 677 (80% power), 906 (90% power)
 50% of observed effect: n per arm = 1,542 (80% power), 2,064 (90% power)
power.prop.test: 375 
Simulated power at n = 375 per arm: 0.803 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure: the number of participants with acute kidney injury. It reports no secondary outcome results. This limits concerns about interpreting a large collection of posted endpoint tests, but the record does not establish that no other outcomes or analyses were considered during the conduct of the study. The full prespecified outcome hierarchy and analysis plan are not reported in the registry record.

Multiplicity arises when several hypotheses are tested, since repeated testing can increase the chance of a nominally unusual result occurring by chance. The posted p-value is identified as unadjusted. The record does not state whether any multiplicity adjustment was planned, whether interim analyses were conducted, or whether the reported analysis was selected from a broader set of analyses. Those details are not reported in the registry record, so the p-value should not be assumed to account for unreported testing.

The record also does not report a missing-data strategy, the number of participants with unavailable primary outcomes, or how any such cases were handled. It does not define an analysis population, such as whether all randomized participants were included according to assignment. Without these details, it is not possible to assess the influence of missing outcome data or departures from randomization on the posted estimate.

Details of allocation concealment, randomization implementation, outcome adjudication, and masking procedures are also not reported. The registry identifies the design as randomized and triple-masked, but does not specify the roles or groups masked. These omissions do not by themselves show that bias occurred; they limit what can be judged from the registry record.

14. 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.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: electrolyte abnormalities (0.0%).

The posted serious adverse event figures are zero in both groups: 0 among 48 participants at risk in Normal Saline and 0 among 44 participants at risk in Sodium Bicarbonate. The registry also reports zero other adverse events in each group. These entries describe the figures posted for the recorded safety categories; they do not establish that no adverse experiences of any kind occurred.

Electrolyte Abnormalities is the listed serious adverse event term, with a reported value of 0.0%. The registry record does not provide additional detail about event definitions, the period of safety follow-up, or the method used to collect adverse events. Death counts are not reported in the registry record.

Zero recorded events in a study of this size should be read as an observation in the enrolled groups, not as proof that an event cannot occur. The safety figures also do not provide a comparative statistical analysis or a precise estimate of rare-event risk. They should be interpreted within the limited scope of what the registry reports.

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

The posted results show fewer observed AKI events in the Sodium Bicarbonate arm than in the Normal Saline arm: 14 of 44 participants compared with 20 of 48. The risk-difference comparison has a confidence interval that includes no difference and estimates on either side of it, and the unadjusted p-value is 0.328. Taken together, these results indicate uncertainty about the direction and size of the underlying between-group effect.

The result does not establish that bicarbonate reduces postoperative AKI, and it does not establish that the treatments have the same effect. A non-small p-value is not evidence of equivalence, and a difference in observed counts is not by itself evidence of a true difference. The study's sample and event counts leave room for a range of effects compatible with the data.

The primary endpoint concerns AKIN-defined AKI within 5 days. The results do not directly address longer-term kidney outcomes, other complications, or outcomes that are not reported. They also do not automatically apply to patients outside the stated clinical context of cardiac surgery, cardiopulmonary bypass, and preoperative kidney insufficiency. The registry's incomplete reporting of eligibility details and participant characteristics further limits assessment of generalizability.

The safety record reports no serious or other adverse events in the posted group figures, while death counts are not reported in the registry record. Those entries are not sufficient to rule out uncommon harms or to characterize safety beyond the recorded categories and follow-up.

Overall, the registry provides a small randomized comparison with an observed count pattern favoring bicarbonate, alongside substantial statistical uncertainty. The record does not contain enough information to make broader claims about treatment effects, clinical importance, or safety.

16. Why This Trial Matters Statistically

This trial illustrates a common challenge in clinical biostatistics: the observed direction of an effect and the precision of its estimate answer different questions. The event counts are lower in the bicarbonate group, but the interval around the risk-difference estimate spans no difference and effects in either direction. Reporting the counts without the interval could overstate certainty; reporting only the p-value could obscure the range of plausible effects.

The endpoint is binary, and the number of events contributes directly to the information available for estimating a difference between groups. The trial enrolled 92 participants, with a limited number of observed AKI events. This makes the distinction between an estimate and a conclusive result especially important. Randomization supports the comparison, but it does not prevent uncertainty when the study is small.

The record also illustrates why registry transparency matters. Information about the sample-size rationale, statistical analysis plan, missing-data handling, analysis populations, and any interim analyses is not reported. Those details help readers assess whether an analysis was prespecified and how robust the result may be to analytic decisions. Their absence limits evaluation without allowing conclusions about how the study was actually conducted.

Finally, planning illustrations based on the observed effect are not a substitute for the original design rationale and are not observed power. They demonstrate how assumptions about the effect to detect and desired power influence planning, while the original trial assumptions remain unreported. Separating descriptive results, inferential uncertainty, and planning scenarios leads to a more careful reading of a clinical trial record.

17. 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 offers limited detail for evaluating the trial beyond its broad design and posted outcome. It does not report the full eligibility criteria, baseline characteristics, allocation method, allocation concealment, or the practical details of triple masking. The reported group counts can be described, but the record does not show whether groups were balanced on important baseline features.

The endpoint is identified as AKIN-defined acute kidney injury within 5 days, but the record does not provide the operational criteria, assessment schedule, adjudication process, or information about who assessed the outcome. The brief description gives a general creatinine-based background definition, which is not a substitute for detail about the specific AKIN-based outcome measure.

The posted results include one primary measure and no secondary outcomes. There is no reported analysis plan, missing-data method, analysis-population definition, interim-analysis information, or sponsor sample-size justification. Consequently, the extent of prespecification and the handling of any unavailable outcomes cannot be assessed from the registry record.

The sample comprised 92 enrolled participants, and the primary endpoint counts were limited. The confidence interval indicates substantial imprecision. The safety information is also restricted: serious and other adverse-event figures are posted as zero in each group, but follow-up details and death counts are not reported in the registry record.

The bicarbonate arm description is incomplete at its ending, and the record does not provide further clinical outcome detail beyond the primary endpoint. These reporting limitations constrain interpretation; they do not establish that the trial lacked procedures or analyses that are simply not described in the registry record.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Prevention of Acute Kidney Injury in Cardiac Surgery Patients — NCT00921518.
• Registry record last updated 2014-04-15; status last verified 2014-03.
• Results first posted 2014-03-11.
• 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.

21. Record Summary

The registry record describes a completed randomized, triple-masked, parallel-group phase 2 trial of intravenous sodium bicarbonate versus isotonic saline in cardiac-surgery patients with kidney insufficiency. It lists 92 enrolled participants. The primary outcome was the number of participants with AKIN-defined acute kidney injury within 5 days.

AKI was reported in 20 of 48 participants assigned to Normal Saline and 14 of 44 assigned to Sodium Bicarbonate. The unadjusted p-value was 0.328, and the confidence interval for the risk difference includes no difference. The findings are uncertain and do not establish that the groups differ or are equivalent. The registry reports zero serious and other adverse events in both groups; deaths are not reported in the registry record.