Citrulline in Severe Sepsis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Citrulline in Severe Sepsis (NCT01474863), a randomized phase 2 study. The primary outcome is Vasopressor Dependency Index.

2. Clinical Question

The study’s stated hypothesis concerned preventing or mitigating the development or progression of acute lung injury in patients with severe sepsis. The posted primary outcome, however, is a measure of vasopressor dependency rather than a direct measure of lung injury. The registry record lists one outcome measure and one analysis. It does not report a posted secondary outcome analysis that would directly answer the broader lung-injury question. The interpretation therefore needs to distinguish the trial’s motivation from the particular outcome and analysis available for statistical interpretation.

3. Trial Design

This was a randomized, placebo-controlled phase 2 study of intravenous L-citrulline in patients with severe sepsis at risk for or with acute lung injury. The registry lists low-dose citrulline, high-dose citrulline, and placebo groups. The title and brief summary describe the trial as double-blind, while the registry’s masking entry says triple. Which people were masked, and how that masking was implemented, are not reported in the registry record.

The registry records enrollment of 72 and gives the study status as completed. The recorded start was 2012-08, and primary completion was 2016-03. These dates describe the trial record; they do not establish the dates on which participants were enrolled or followed individually. The record identifies Vanderbilt University as sponsor and lists severe sepsis and acute lung injury as conditions.

The available outcome result is a continuous measure reported as a mean with standard deviation. The primary outcome is the Vasopressor Dependency Index at day 4, based on the worst value measured hourly through the study infusion. Group summaries are posted for all three groups. The registry’s own inferential analysis is an ANOVA across the groups, with p = 0.86. Separately, the posted summary statistics permit unadjusted re-calculations comparing each citrulline group with placebo. Those pairwise calculations are not a replacement for the sponsor’s overall analysis, because they address different statistical contrasts.

The registry record does not report the original sample-size justification, planned analysis details beyond the posted ANOVA, or a detailed protocol for outcome collection. In particular, the reason for choosing the recorded enrollment, rules for handling missing endpoint observations, and analysis population definitions are not reported in the registry record. The planning figures based on the observed effect are illustrative calculations, not a description of the trial’s original design.

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 Vasopressor Dependency Index, measured over day 4.

The primary endpoint is the Vasopressor Dependency Index at day 4. It is described as the worst value of an index measured hourly through the study infusion. The index is mean arterial blood pressure divided by a catecholamine index. That catecholamine index is dimensionless and combines the doses of dopamine, dobutamine, adrenaline, noradrenaline, and phenylephrine using the weights specified in the registry description. The listed dose units are ug/kg/min. The registry states that a higher index is better.

This endpoint represents a hemodynamic measure involving blood pressure and catecholamine dosing. In general, combining a clinical measure with treatment intensity can summarize more than either component alone, but the meaning depends on the definition, timing, and context of the measure. The registry description supplies the formula and direction but does not report validation information or explain the clinical rationale for selecting this index as the primary endpoint.

The endpoint is continuous, and the posted summaries are means and standard deviations for each group. A mean describes the group’s average value, while the standard deviation describes variation among observations around that average. The registry record does not state whether the analysis used every randomized participant or only those with an observed day 4 value, nor does it report a detailed account of missing endpoint measurements. The result therefore describes the posted summaries and analysis, rather than establishing how unobserved endpoint values were handled.

The trial’s broader aim concerned acute lung injury, but the posted primary endpoint is the Vasopressor Dependency Index. No secondary outcome results are posted in the registry record. As a result, this endpoint alone does not provide a direct summary of every clinical or biochemical question named in the study description.

7. Results Posted in the Registry

• Low Dose Citrulline: mean 50.00 versus 44.40 on control; difference 5.60 (95% CI -11.41 to 22.61), standardized effect d = 0.19; z-test p = 0.519.
• High Dose Citrulline: mean 49.00 versus 44.40 on control; difference 4.60 (95% CI -12.34 to 21.54), standardized effect d = 0.16; z-test p = 0.595.

8. Statistical Interpretation

9. Statistical Methods Explained

Continuous outcomes are often summarized by a mean and a measure of spread. Here, the registry identifies the Vasopressor Dependency Index as continuous and reports means with standard deviations. The mean locates the group average on the scale of the outcome. The standard deviation describes how dispersed individual observations are around that average. A mean difference compares two group averages in the outcome’s original units, while a standardized effect expresses the contrast relative to variability. These summaries answer related but different questions: the original-unit contrast is directly tied to the endpoint scale, while standardization can help describe the contrast relative to the spread of observations.

What does the overall ANOVA test?

In general, analysis of variance, or ANOVA, compares group means through an overall test. For this trial, the registry reports an ANOVA across the low-dose, placebo, and high-dose groups, with p = 0.86. An overall test asks whether the observed pattern is inconsistent with a model in which the group means are equal. It does not automatically provide a separate test of each treatment-versus-placebo contrast, nor does it say which group is higher. The registry record does not give the ANOVA model specification or its estimated group contrasts. Therefore, the sponsor-reported result should be read as an overall test, not as a detailed account of every pairwise comparison.

What do the pairwise re-calculations add?

The unadjusted re-calculations use the posted group means, standard deviations, and sample sizes to estimate mean differences between each citrulline group and placebo. The low-dose comparison has an estimated mean difference of 5.6, and the high-dose comparison has an estimated mean difference of 4.6. Their confidence intervals and p-values are reported alongside those estimates. These calculations are described as recalculated because they are simple unadjusted calculations from the posted summary statistics, rather than the registry’s reported ANOVA result. They are useful for understanding the treatment-versus-placebo contrasts but should not be mistaken for the sponsor’s own pairwise analysis.

For an unadjusted mean comparison, the uncertainty of the estimated difference depends on the variability of observations and the amount of information in each group. Greater variation makes it harder to distinguish a group contrast from sampling noise; more observations can generally improve precision, all else being equal. The posted standard deviations are substantial relative to the estimated contrasts, and the intervals are correspondingly broad. General statistical reasoning explains why a modest estimated contrast combined with appreciable variation may remain uncertain, but it does not turn these estimates into proof of no effect.

Why do confidence intervals matter?

A confidence interval gives a range of values that are compatible with the data and the method used to calculate the interval. In repeated use of a procedure, intervals formed at a stated confidence level would cover the target parameter at that rate under the procedure’s assumptions. That general interpretation is not a probability statement about the particular fixed true value after the interval has been calculated. For these comparisons, the reported intervals extend from negative to positive values and include zero, so they do not identify a single direction of difference. Their breadth also means that the estimated effects are not highly precise.

In general, a p-value measures how unusual data at least as incompatible with a null hypothesis would be if that hypothesis and the analysis assumptions were 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. The pairwise p-values are unadjusted; the registry record does not state whether any multiplicity procedure was planned or applied. The ANOVA p-value and the pairwise p-values should not be treated as interchangeable because they address different hypotheses.

Why can variability affect planning?

Over-dispersion means that outcome variation is greater than a model allows for. In general, when a model underestimates variability, standard errors may be too small and intervals may be too narrow, making evidence appear more precise than it is. Conversely, a model that accounts for variability appropriately is important for reliable inference. The registry record reports standard deviations but does not report diagnostics for distributional assumptions, an assessment of over-dispersion, or a detailed modeling strategy. There is no basis in the record to claim that over-dispersion occurred; it is a general consideration when interpreting and planning analyses.

Mean-based methods can also be sensitive to skewness, extreme values, or unequal variability, depending on the data and analysis. The registry record does not provide participant-level observations, distribution plots, or diagnostics that would allow those features to be assessed. The endpoint is described as the worst value measured during the infusion period, which is a defined summary of repeated measurements; how that measure was distributed across participants is not reported.

What can the posted analysis establish?

The registry supports a description of group means, standard deviations, an overall ANOVA p-value, and unadjusted pairwise re-calculations with confidence intervals. It does not report sufficient detail to reconstruct all aspects of the original inferential analysis. The original statistical analysis plan, analysis population, treatment of missing values, covariate adjustment, and multiplicity strategy are not reported in the registry record. These omissions matter because statistical conclusions depend not only on the outcome values but also on the population analyzed, the model, and the decisions made before or during analysis.

The sample-size illustrations use the observed effects and stated planning assumptions. They describe how participant requirements would change under those assumptions if the effect to detect were a specified fraction of the observed effect. In general, smaller target effects require more participants for the same power when other design assumptions are held fixed. These illustrations do not reveal the trial’s original design rationale, which is not reported in the registry record, and they do not measure the chance that the observed result was correct.

Finally, statistical evidence and clinical meaning are related but separate. A statistically estimated difference on the Vasopressor Dependency Index would still require interpretation in the context of the endpoint’s clinical relevance, study population, safety information, and the broader outcomes of interest. The registry describes a direction in which higher values are better, but does not report a threshold for a clinically important difference. Statistical results alone therefore cannot establish the clinical importance of a contrast.

10. Confidence Intervals and Precision

The recalculated 95% confidence interval for low-dose citrulline versus placebo ranges from -11.41 to 22.61 around a mean difference of 5.6. For high-dose citrulline versus placebo, the interval ranges from -12.34 to 21.54 around a mean difference of 4.6. Each interval includes zero and covers values on both sides of it. Accordingly, the observed estimates do not precisely determine whether the population mean in a citrulline group is higher or lower than the placebo mean.

Confidence intervals are useful because they communicate uncertainty in the estimated contrast, not just its direction or a single point estimate. In general, a wide interval indicates that a range of effect sizes remains compatible with the data and the calculation method. These intervals allow both small differences and larger differences in either direction. That breadth limits how confidently the effect’s magnitude can be described. The point estimates are positive, but the interval limits show that the data do not rule out a contrast in the opposite direction.

The intervals are unadjusted re-calculations from posted summary statistics. They are not the confidence intervals from the sponsor’s reported ANOVA, for which the registry does not post an estimate or interval. The registry record also does not say whether the original analysis used covariate adjustment, a particular missing-data method, or any multiplicity adjustment. Those choices can affect estimated precision and inferential interpretation.

Precision is distinct from whether an estimate crosses a conventional significance threshold. A non-significant result does not show that there is no difference, and a confidence interval is not a guarantee that the true effect lies inside it. Rather, the interval communicates the range supported by the data under the calculation’s assumptions. Given the reported variability and the width of both intervals, the posted pairwise results are uncertain about the size and direction of any underlying contrast. The registry does not report a threshold for a clinically important difference, so the intervals cannot be classified against such a threshold.

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 illustrative calculations based on the observed effect, with equal allocation between the two arms, a two-sided significance level of 0.05, and a normal approximation. Each planning scenario uses the stated fraction of the observed effect as the effect to detect. The trial’s own original sample-size assumptions and justification are not reported in the registry record.

The 80% power column gives the calculated participants per arm under which the stated planning method would have an 80% chance of detecting the target effect if that effect and the assumptions were correct. The 90% power column gives the corresponding planning figure for a 90% chance. Power is a design property under specified assumptions; it does not describe the probability that a result is true or that a study has found an effect.

For the low-dose comparison, the planning illustrations list 444 participants per arm for 80% power and 595 for 90% power when using 100% of the observed effect. For 75% of the observed effect, the corresponding figures are 790 and 1,057; for 50% of the observed effect, they are 1,776 and 2,377. For the high-dose comparison, the figures are 630 and 843 at 100% of the observed effect, 1,119 and 1,498 at 75%, and 2,517 and 3,370 at 50%.

Smaller effects require more participants to detect under the same planning assumptions, which is why the participant figures rise as the target effect moves from 100% to 75% to 50% of the observed effect. These figures are planning illustrations, not observed power. They should not be read as a retrospective rating of the trial or as evidence about the trial’s original design choices.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 5.600; pooled standard deviation σ = 29.777.
2. n = 2 × (1.96 + 0.842)² × 29.777² / 5.600² = 443.8, rounded up to 444 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
delta <- 5.6                      # observed difference in means
sigma <- 29.7770739065272                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1474863)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 444 (80% power), 595 (90% power)
 75% of observed effect: n per arm = 790 (80% power), 1,057 (90% power)
 50% of observed effect: n per arm = 1,776 (80% power), 2,377 (90% power)
power.t.test: 444.8 
Simulated power at n = 444 per arm: 0.798 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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. Analyses Reported by the Sponsor

The sponsor’s posted analysis is an ANOVA across the three groups, with p = 0.86. This is the registry’s own reported inferential result. It is an overall test, whereas the recalculated results compare low-dose citrulline with placebo and high-dose citrulline with placebo separately. The different p-values therefore correspond to different statistical questions and should not be treated as competing versions of the same test.

The recalculated pairwise estimates are based on posted group summaries and are unadjusted. They give a mean difference of 5.6 for low-dose citrulline versus placebo, with a 95% confidence interval from -11.41 to 22.61 and p = 0.519, and a mean difference of 4.6 for high-dose citrulline versus placebo, with a 95% confidence interval from -12.34 to 21.54 and p = 0.595. The registry record does not post sponsor-reported pairwise estimates or confidence intervals for these contrasts.

The record does not flag a discrepancy between the sponsor’s analysis and the summary-statistic re-calculations. Even so, the sponsor’s ANOVA remains the reported trial analysis; the recalculations are complementary, unadjusted descriptions of two specific contrasts. The record does not provide enough detail about the ANOVA model or analysis plan to reproduce the sponsor’s result independently.

14. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when an analysis considers multiple outcomes, comparisons, time points, or other opportunities to find a difference. In general, making many inferential tests without an appropriate strategy can increase the chance of at least one apparently noteworthy result by chance. Possible strategies depend on the study question and design; the registry record does not report whether a multiplicity adjustment or hierarchy was planned or used.

The registry lists one outcome measure and one analysis. The reported analysis is an overall ANOVA across the three groups. The additional treatment-versus-placebo calculations are unadjusted re-calculations from summary statistics, not sponsor-reported pairwise analyses. Their p-values should therefore be interpreted as unadjusted and in the context of the broader set of analyses that may have been considered. The registry does not give a complete analysis plan that would establish the intended family of tests.

Missing-data handling is also not reported in the registry record. The posted group summaries give sample sizes of 26.0 for low-dose citrulline, 22.0 for placebo, and 24.0 for high-dose citrulline, but the record does not explain whether these are all randomized participants, endpoint-observed participants, or an analysis population defined in another way. It does not report the number of missing day 4 measurements, reasons for missingness, imputation methods, or sensitivity analyses. The impact of missing endpoint observations therefore cannot be assessed from the posted results alone.

Analysis populations are not described in the registry record. It is not reported whether the primary analysis followed randomized assignment regardless of treatment received, included only participants who completed treatment, or used another population definition. The record also does not report protocol deviations or exclusions for the primary outcome analysis. These details matter because the group summaries and estimated effects can depend on which participants are included and how missing observations are treated.

The registry record does not provide a detailed statistical analysis plan, interim-analysis information, or rules for selecting among models. Those details should not be inferred from the phase or the posted outcome. Consequently, the reported ANOVA can be described as the sponsor’s posted overall analysis, while broader claims about prespecification, multiplicity control, missing-data robustness, or population choice are not supported by the 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 0% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: death (22.7%).

The serious adverse event summaries list 5 affected participants in the low-dose citrulline group, 5 in the placebo group, and 3 in the high-dose citrulline group. The listed group counts at risk are 26, 22, and 24, respectively. The registry also lists deaths as 5, 5, and 3 for those groups, and other adverse events as affecting 0 in each group.

The registry’s top serious adverse event term is Death, with 22.7% listed. The record does not explain the denominator or calculation behind that percentage, so it should be quoted as the registry’s entry rather than recomputed or assigned to a particular group. The record also does not give event timing, causes, severity detail beyond the serious-event classification, or whether participants with deaths are included among those counted for serious adverse events. The counts should not be combined or treated as independent categories without that information.

These figures are descriptive safety summaries. Without further detail about ascertainment, follow-up, event attribution, and the relationship between the counts and analysis populations, they do not by themselves establish comparative safety. They also do not support conclusions about cause or treatment effect.

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

The posted primary outcome results show that the mean Vasopressor Dependency Index at day 4 was 50.0 in the low-dose group, 44.4 in placebo, and 49.0 in the high-dose group. The registry states that higher values are better. The sponsor’s overall ANOVA has p = 0.86, and the unadjusted pairwise re-calculations have confidence intervals that include zero. Together, these findings do not provide clear statistical evidence of a group difference on this endpoint in the posted analysis.

That interpretation is not the same as concluding that the groups are equivalent. The pairwise intervals are broad and include effects in both directions, so the results leave substantial uncertainty about the size and direction of a population contrast. The registry does not report a clinically important difference threshold, so the estimated values cannot be judged against a stated threshold for clinical relevance. A non-significant result does not prove the absence of an effect.

The trial’s stated hypothesis concerned development or progression of acute lung injury among patients with severe sepsis. The posted primary endpoint is the Vasopressor Dependency Index, and the registry lists no secondary outcome results. Therefore, the results do not directly settle the broader acute lung injury question. Nor do they establish effects on all biochemical, clinical, or safety outcomes mentioned in the study summary.

The safety counts provide additional descriptive information, but the registry record does not give the detail needed to attribute events, assess timing, or infer causality. Similarly, the record does not provide missing-data methods, a full analysis population definition, or a detailed analysis plan. Those gaps constrain the conclusions that can be drawn from the posted results.

The most defensible reading is narrow: the trial record reports group summaries for a day 4 hemodynamic endpoint, an overall ANOVA that does not indicate a clear difference, and pairwise recalculations with considerable uncertainty. These results characterize the available registry evidence. They do not establish that intravenous citrulline works or does not work, that the treatments are equivalent, or that the study answers every question motivating the trial.

17. Why This Trial Matters Statistically

This trial illustrates why a statistical interpretation should separate the study’s motivating hypothesis, its prespecified or posted endpoint, and the analysis actually reported. The study description concerns prevention or mitigation of acute lung injury in severe sepsis, while the posted primary result is a day 4 Vasopressor Dependency Index. That distinction matters: a result on a hemodynamic measure cannot automatically be treated as a direct result on lung injury or every other clinical outcome.

It also illustrates the difference between an overall test and treatment-versus-control contrasts. The registry’s ANOVA asks an overall question about the group means, while the recalculations estimate each citrulline group’s mean difference from placebo. These analyses are complementary but not interchangeable. Reporting both the estimate and its uncertainty is more informative than relying on a p-value alone.

The confidence intervals emphasize the limits of precision. Their breadth means the posted data leave room for meaningfully different possible contrasts, despite positive point estimates. This is a useful reminder that the absence of a statistically clear difference does not demonstrate equivalence. It also shows why planning around smaller effects can require substantially larger participant numbers under fixed assumptions.

Finally, the record demonstrates the importance of transparent reporting. Information about missing observations, analysis populations, the original sample-size rationale, multiplicity, and model details is not reported in the registry record. Without those details, readers can interpret the posted summaries and analyses but cannot fully assess robustness or reconstruct all design and analysis decisions. These distinctions are central to evaluating what a clinical trial can support statistically.

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.

The available endpoint is a measure of vasopressor dependency at day 4, while the stated study hypothesis concerns acute lung injury. The registry record does not post secondary outcome results that would directly describe the broader lung-injury question. This limits the scope of conclusions that can be drawn from the available endpoint.

The posted summaries and analyses do not include a detailed account of the analysis population, missing endpoint observations, or the handling of missing data. The registry record does not report whether the analysis followed randomized assignment, which participants contributed to the primary endpoint, or whether sensitivity analyses were performed. These omissions make it difficult to assess the possible effect of exclusions or missing measurements.

The record reports the sponsor’s ANOVA p-value but does not provide the model specification, test statistic, estimated overall contrasts, or confidence interval. The pairwise re-calculations are unadjusted and based on summary statistics. The registry record does not report whether those comparisons were planned, whether multiplicity was addressed in the original analysis, or how the reported analysis related to a full statistical analysis plan.

The endpoint is described as the worst value measured hourly through the study infusion, but the record does not report further details about the measurement schedule, endpoint validation, or the clinical importance of different index values. It states that higher is better, but provides no threshold for a clinically important difference. The specific high and low dose amounts are not reported in the posted results.

The sample-size illustrations are based on the observed effects and stated planning assumptions; they are not the original design rationale. That rationale is not reported in the registry record. The safety summaries also lack information about event timing, causes, attribution, and overlap among categories. Together, these trial-specific reporting limits argue for interpreting the results as descriptive evidence about the posted endpoint and analyses rather than as a complete account of all design, clinical, or safety questions.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Citrulline in Severe Sepsis — NCT01474863.
• Registry record last updated 2017-06-05; status last verified 2017-06.
• Results first posted 2017-06-05.
• 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

NCT01474863 is a completed phase 2 randomized, placebo-controlled trial titled “Citrulline in Severe Sepsis.” The record identifies Vanderbilt University as sponsor, gives enrollment as 72, and records a start of 2012-08 and primary completion of 2016-03. The listed groups are low-dose citrulline, placebo, and high-dose citrulline.

The primary outcome is the Vasopressor Dependency Index at day 4, reported as a mean with standard deviation. The registry’s overall ANOVA has p = 0.86. The unadjusted recalculated comparisons with placebo have mean differences of 5.6 for low-dose citrulline and 4.6 for high-dose citrulline, with confidence intervals that include zero. Serious adverse event counts and deaths are posted for each group, but further event detail is not reported in the registry record.

The record does not report the original sample-size justification, detailed missing-data methods, analysis populations, or multiplicity strategy. Its results support a cautious interpretation focused on the posted endpoint and the uncertainty of the estimated contrasts.