Comparison of Sugammadex With Neostigmine During Laparoscopic Cholecystectomy or Appendectomy (P05699): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparison of Sugammadex With Neostigmine During Laparoscopic Cholecystectomy or Appendectomy (P05699) (NCT00724932), a randomized phase 3 study. The primary outcome is Time From Start of Administration of Investigational Medicinal Product (IMP, Sugammadex or Neostigmine) to Recovery of the Fourth Twitch/First Twitch (T4/T1) Ratio to 0.9.

2. Clinical Question

The endpoint is a physiological measure of recovery from neuromuscular blockade, not a direct measure of how a participant felt or of later recovery after leaving the operating room. The trial also set out to compare safety and evaluate operating-room and post-anesthetic-care-unit length of stay, but results for those length-of-stay questions are not reported in the registry record.

The treatment strategies differed both in the reversal medicine and in when it was administered: sugammadex was given at 1-2 PTC, while neostigmine was given at reappearance of T2. The primary comparison therefore addresses these two protocol strategies as conducted, rather than isolating a medicine effect at an identical monitoring point. The registry record does not report a separate comparison holding administration timing constant.

3. Trial Design

This was a phase 3, randomized, parallel-group, active-controlled trial. The registry labels the masking as double, while the official title describes the trial as safety-assessor blinded. It was multicenter and included adults undergoing laparoscopic cholecystectomy or appendectomy under propofol anesthesia. The trial status is completed. The registry lists enrollment of 140, a start date of 2008-07-16, and primary completion on 2009-04-01.

Participants received rocuronium to induce neuromuscular blockade. The sugammadex group received 4.0 mg.kg-1 at 1-2 PTC. The neostigmine group received 50 µg.kg-1 at reappearance of T2, together with atropine in a ratio of 5:1 for neostigmine:atropine. These different administration points are part of the comparison being evaluated. The registry record does not report further details about allocation concealment, how the masking was implemented, or the number and locations of participating centers.

The primary outcome was time from the start of investigational medicinal product administration to recovery of the T4/T1 ratio to 0.9. Its listed unit is minutes, and the summary measure is the geometric mean with a 95% confidence interval. The primary outcome population included randomized participants who received investigational medicinal product and had at least one efficacy measurement. Imputed recovery times were used when times were missing.

Other registered outcomes include recovery times to T4/T1 ratios of 0.7 and 0.8, as well as counts of participants experiencing pre-treatment or post-treatment serious and non-serious adverse events. The registry lists 34 outcome measures and 3 analyses, while 4 secondary outcomes are posted. The record does not report the basis for the sample size, a planned allocation ratio, or a more detailed 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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Intent-To-Treat (ITT) Population consisted of all randomized participants who received IMP and had at least one efficacy measurement. Imputed recovery times were used in cases of missing times.

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: The Baseline Analysis Population consisted of randomized participants who received investigational medicinal product (IMP). Four participants randomized to Sugammadex did not receive IMP and three participants randomized to receive Neostigmine did not receive IMP.

6. Primary Endpoint

The registry lists the primary outcome as Time From Start of Administration of Investigational Medicinal Product (IMP, Sugammadex or Neostigmine) to Recovery of the Fourth Twitch/First Twitch (T4/T1) Ratio to 0.9, measured over From start of IMP administration to recovery of T4/T1 ratio to 0.9 (ranging from ~2 minutes to ~9 minutes).

The primary endpoint is the time from starting the investigational medicinal product to recovery of the T4/T1 ratio to 0.9. The ratio compares the height of the fourth twitch with the height of the first twitch after train-of-four stimulation of the ulnar nerve. In this trial, the nerve was stimulated every 15 seconds and twitch responses were assessed at the adductor pollicis muscle. A ratio closer to 1.0 indicates more complete recovery of the monitored twitch response.

The registry describes 0.9 as the minimum acceptable ratio indicating recovery from neuromuscular blockade. The endpoint is therefore a timed physiological marker: a shorter time means the monitored ratio reached the stated recovery threshold sooner. It is useful for comparing the timing of recovery under the two treatment strategies, but it does not on its own describe operating-room duration, post-anesthetic-care-unit stay, or every aspect of a participant’s recovery.

The trial reports the endpoint as a geometric mean in minutes with a 95% confidence interval. A geometric mean is a summary used for positive measurements whose distribution may be skewed; in general, it gives a central value on a multiplicative rather than an additive scale. The posted timeframe describes recovery as ranging from approximately 2 minutes to approximately 9 minutes. That timeframe is descriptive and does not replace the group-specific estimates or their uncertainty.

The endpoint is tied to a specific measurement protocol and threshold. Its interpretation depends on the train-of-four monitoring procedure, the definition of the threshold, and the handling of missing recovery times. The registry describes the monitoring and imputation use, but does not report the specific imputation method.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the geometric mean describe?

In general, the geometric mean summarizes positive values after taking their logarithms, averaging on that scale, and translating the average back to the original scale. It is often useful when observations are right-skewed or when relative changes are more meaningful than absolute changes. It should not be read as the ordinary arithmetic average. In this trial, the registry explicitly identifies the primary summary as a geometric mean and gives time in minutes.

The posted estimates are 2.4 minutes for sugammadex and 8.4 minutes for neostigmine. Those are summaries of recovery times, not statements that all participants recovered at those exact times. The confidence intervals convey uncertainty around each estimated group summary. They do not show the full spread of individual participants’ times; the registry does not post an individual-level distribution in the registry record for this analysis.

How should the ratio and p-value be read?

The registry reports an ANOVA with the parameter described as a geometric mean ratio of final values. The ratio is 3.4, with a 95% confidence interval from 2.8 to 4.1, and p < 0.001. In general, a ratio compares groups multiplicatively, while a difference compares them additively. A ratio of 1.0 represents equality on the ratio scale. The record lists the group order as sugammadex and neostigmine but does not specify which group is in the numerator; interpreting the ratio’s direction from the ratio alone would therefore require an assumption. The arm-specific estimates provide the direct description of the posted times.

In general, a p-value is the probability, assuming the null model and other analysis assumptions, of obtaining a result at least as incompatible with that model as the observed result. It is not the probability that the null hypothesis is true, nor is it a measure of clinical importance. The reported p-value indicates strong statistical evidence against equal values under the comparison used, but it does not account for unreported design choices or multiple testing unless those are incorporated in the analysis.

Why use confidence intervals?

In general, a confidence interval gives a range of values compatible with the data and the specified method at its stated confidence level. The interval is not a guarantee that the true value lies inside it, and its width depends on both the information in the data and the analysis assumptions. The registry reports 95% confidence intervals for the arm estimates and the ratio. Those intervals make precision visible alongside the central estimates and help distinguish an estimate from a single fixed value.

The interval for the group ratio is reported as 2.8 to 4.1. Because the record does not state the numerator convention, the ratio should be read with that limitation in mind. The arm-level times give a transparent indication of which group had the lower posted geometric mean time; they should not be converted into a new ratio or a new difference when the task is to describe the posted analysis.

What does ANOVA contribute?

In general, analysis of variance (ANOVA) is a modeling framework for comparing group outcomes. The registry identifies ANOVA as the method for the reported two-group analysis. It does not provide further detail about transformations, covariate adjustment, model diagnostics, or whether any other factors were included. The geometric-mean reporting suggests a multiplicative summary, but the record does not provide enough methodological detail to reconstruct the model specification. No additional modeling choices should be inferred.

General statistical context is important because the validity and interpretation of model-based intervals and p-values rely on assumptions. For time outcomes, distributional shape and the handling of missing values can affect estimates. A skewed distribution can make an arithmetic mean less representative, which is one reason geometric means may be used for positive measurements. Over-dispersion, in general, means that observed variation exceeds what a chosen model expects; when present and unaddressed, it can make uncertainty estimates too small. The registry record does not report whether over-dispersion was assessed or how model assumptions were checked, so no conclusion about those diagnostics can be drawn.

How do missing values affect the endpoint?

The registry defines the primary efficacy population as all randomized participants who received investigational medicinal product and had at least one efficacy measurement. It also states that imputed recovery times were used for missing times. In general, imputation replaces a missing value with an estimate under a specified procedure, and the procedure’s assumptions matter to the resulting estimate. The record does not report the number of missing recovery times, the imputation method, or any sensitivity analysis. Consequently, the effect of missingness on the posted result cannot be evaluated from the registry record alone.

The efficacy counts are 66 for sugammadex and 65 for neostigmine. Those counts should be understood with the stated efficacy population definition; they are not necessarily identical to every other analysis denominator. The safety data use different denominators. The registry record does not explain why the denominators differ, so they should not be treated as interchangeable.

What can planning illustrations tell readers?

In general, a sample-size calculation estimates how many participants a future study might need under stated assumptions about the target effect, significance level, allocation, and statistical method. A planning illustration based on an observed effect is not the same as the original trial’s design justification, and it is not observed power. The registry record does not report the trial’s sample-size justification. Any planning quantities should therefore be treated as illustrations conditional on their assumptions, not as evidence that this trial had a particular probability of detecting an effect.

The sample-size assumptions recorded for illustrations specify a two-sided significance level of 0.05, equal allocation, a normal approximation, and effects expressed as fractions of the observed effect. Those are assumptions for planning calculations, not details that establish the original design. For time-to-event calculations, the record says the total number of events is used; other calculations use participants per arm. The primary endpoint is reported as a recovery time, but the registry’s description of planning assumptions does not establish that it used a time-to-event analysis for the primary result.

More broadly, the registry reports 34 outcome measures, 3 analyses, and 4 posted secondary outcomes. The record does not report a full analysis plan, diagnostic results, or enough detail to independently reproduce the ANOVA. The posted result can be interpreted as reported, while those omissions limit deeper assessment of how sensitive the inference is to analytic choices.

10. Analyses Reported by the Sponsor

The registry reports the primary comparison as an ANOVA, with a geometric mean ratio of 3.4, a 95% confidence interval from 2.8 to 4.1, and p < 0.001. It also posts arm-level geometric means and confidence intervals: 2.4 minutes (2.1 to 2.7) for sugammadex and 8.4 minutes (7.2 to 9.8) for neostigmine. These are complementary summaries: the arm estimates describe recovery times in each group, while the ratio and p-value express the reported group comparison.

The registry record does not report a separate recalculated estimate or enough model detail to reproduce the sponsor’s ANOVA independently. No discrepancy is recorded between the posted result and a recalculation. Where a sponsor analysis and a simple unadjusted recalculation disagree, the sponsor’s own analysis takes precedence; the record does not indicate such a disagreement for this result. The ratio’s numerator convention is not stated, so the arm-level estimates are the clearest basis for describing the direction of the observed times.

The posted inferential result should be read in the context of the primary endpoint definition, the stated efficacy population, and the use of imputed recovery times. The registry record does not provide the imputation method, the number of missing values, or a complete model specification. Those omissions prevent a full independent audit of the analysis, but they do not change the values reported in the registry.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 34 outcome measures and 3 analyses, while 4 secondary outcomes are posted. The secondary outcomes include recovery time to T4/T1 ratios of 0.7 and 0.8, and counts of participants with pre-treatment or post-treatment serious and non-serious adverse events. The number of registered measures and the smaller number of posted secondary outcomes make it important to distinguish the primary result from the broader set of measured outcomes.

The registry record does not report whether the primary endpoint was protected by a prespecified testing hierarchy, whether multiplicity adjustments were planned or used, or how the 3 analyses relate to the full set of outcome measures. It also does not report interim analyses. Without those details, it is not possible to determine from the registry record how the reported p-value should be adjusted, if at all, for other analyses or outcomes. The primary comparison should therefore be interpreted as the posted analysis, without assuming an unreported multiplicity strategy.

For the primary efficacy analysis, the registry defines the Intent-To-Treat population as randomized participants who received investigational medicinal product and had at least one efficacy measurement. It states that imputed recovery times were used when times were missing. This definition is more specific than simply all randomized participants, because receipt of treatment and at least one efficacy measurement are required. The record does not report how many randomized participants were excluded from this population or why.

The record does not describe the imputation method, the amount or pattern of missingness, or sensitivity analyses using alternative assumptions. In general, the consequences of missing outcome data depend on why values are missing and how the analysis handles them. Those reasons and assumptions are not reported in the registry record, so the robustness of the primary estimate to missing data cannot be assessed from these registry entries.

There are also different arm denominators across efficacy and safety reporting. Primary efficacy counts are 66 and 65; serious and other adverse-event risk counts are 66 and 67. The registry record does not explain the differences. They should be reported as analysis-specific denominators, not reconciled by assuming a common population. The reason the posted safety denominators differ from the efficacy denominators is not reported in the registry record.

12. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: colitis (1.5%), post procedural complication (1.5%), procedural nausea (1.5%), procedural vomiting (1.5%), muscle rigidity (1.5%).

The registry reports serious adverse events among 5 of 66 participants at risk in the sugammadex group and 6 of 67 participants at risk in the neostigmine group. For other, non-serious adverse events, it reports 64 of 66 participants in the sugammadex group and 63 of 67 in the neostigmine group. These are participant counts with the listed denominators; they do not describe how many separate events occurred, whether participants experienced more than one event, or the timing and duration of events.

The registry lists the following top serious adverse-event terms, each at 1.5%: colitis, post procedural complication, procedural nausea, procedural vomiting, muscle rigidity, and vascular calcification. The record does not assign these listed terms to a particular treatment group in the information, so they should not be attributed to either group. Death counts are not reported in the registry record.

Safety counts are descriptive. In general, small differences in event counts do not by themselves establish a comparative safety effect; interpretation also depends on event definitions, follow-up, severity, exposure, and uncertainty. The registry record does not provide comparative statistical tests for these safety figures, nor enough detail to evaluate event timing or relatedness. The different denominators for the safety and primary efficacy results also mean the populations should not be assumed to match.

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

The primary endpoint records the time from starting the reversal medicine to reaching a specified train-of-four recovery threshold. The posted geometric mean times are 2.4 minutes for sugammadex and 8.4 minutes for neostigmine. The registry’s ANOVA comparison reports a geometric mean ratio of 3.4, with a 95% confidence interval from 2.8 to 4.1 and p < 0.001. Together, these results describe a clear difference in the measured recovery times under the trial’s two treatment strategies.

The result applies to the specific setting and procedures described: adults undergoing laparoscopic cholecystectomy or appendectomy under propofol anesthesia, with reversal at different neuromuscular monitoring points in the two groups. It does not isolate the medicine from the timing strategy, because both the medicine and the specified point of administration differed between groups. The registry record does not report an additional comparison that holds timing constant.

The finding is about a monitored physiological endpoint. It does not establish the effect on operating-room or post-anesthetic-care-unit length of stay; although those outcomes were part of the trial’s stated aims, their results are not reported in the registry record. It also does not supply a complete picture of participant experience, longer-term outcomes, or every aspect of safety. The adverse-event counts are descriptive, and the record does not provide a comparative safety analysis in the information reported.

A statistical difference is not the same as a statement about the importance of the difference to an individual participant or a recommendation about treatment. The confidence intervals characterize uncertainty in the group summaries, not the range of times each participant might experience. The posted result should be understood alongside the analysis population, missing-time imputation, and the limitations in reporting about the analysis plan and model.

The record is a registry account of a completed trial, not a complete report of every methodological detail or outcome. The trial’s sample-size justification, multiplicity strategy, imputation method, and fuller analysis plan are not reported in the registry record. Those missing details constrain independent assessment of robustness and generalizability, even though the posted primary values can be described directly.

14. Why This Trial Matters Statistically

This trial illustrates why clinical endpoint definitions matter. Recovery is not represented by a vague impression of being “reversed”; it is operationalized as the time required for a monitored T4/T1 ratio to reach 0.9. The monitoring procedure, threshold, and time origin determine what the estimate means. A difference in this endpoint is interpretable only in relation to the specific procedure used to measure it.

The study also illustrates how the treatment contrast includes more than a drug name. Sugammadex was administered at 1-2 PTC, while neostigmine was administered at reappearance of T2. The randomized comparison therefore evaluates two strategies that differ in both medicine and administration point. Clear interpretation keeps that design feature visible rather than attributing the entire contrast to a single component.

Reporting geometric means, confidence intervals, and a formal group comparison gives readers more than a single point estimate. The arm-level estimates describe the outcome scale, while the interval and p-value express statistical uncertainty and evidence under a model. Those statistics answer different questions and should not be collapsed into one claim. In particular, a small p-value does not quantify clinical importance or individual experience.

The record also demonstrates the value of transparent information about analysis populations and missing data. It identifies the efficacy population and states that missing recovery times were imputed, but does not describe the imputation procedure or amount of missingness. These details affect how readers assess the reliability and reproducibility of a result. Differences between efficacy and safety denominators are another reason to keep outcome-specific populations distinct.

Finally, a completed trial can have a clearly reported primary result while leaving important questions unresolved in the registry record. Details about multiplicity, interim analyses, model diagnostics, and sample-size justification are not reported. Such gaps do not erase the posted result, but they define the boundary between interpreting a registry result and conducting a full independent methodological review.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The main design-specific limitation for interpretation is that the reversal strategies differed in both the medicine and the stated time of administration. Sugammadex was given at 1-2 PTC, and neostigmine was given at reappearance of T2. The primary contrast therefore cannot separate the effect of the medicine from the effect of the differing administration point. The registry record does not report a comparison that would isolate those components.

The primary efficacy population was not defined as every randomized participant: it included randomized participants who received investigational medicinal product and had at least one efficacy measurement. Missing recovery times were imputed, but the method, number of missing times, and sensitivity analyses are not reported in the registry record. The impact of these choices cannot be assessed from the posted details.

The registry reports arm-specific estimates and an ANOVA ratio, but does not give the ratio’s numerator convention or a full model specification. It also does not report model diagnostics, covariate adjustments, a multiplicity plan, or interim analyses. The record therefore does not support a complete reconstruction of the analysis or assessment of all modeling assumptions.

The safety denominators differ from the efficacy denominators, and the record does not explain why. It reports participant counts and selected serious adverse-event terms, but does not provide enough information to assess event duration, severity, relatedness, or comparative uncertainty. Death counts are not reported in the registry record.

Although the trial’s aims included operating-room and post-anesthetic-care-unit length of stay, results for those outcomes are not reported in the registry record. The record also does not report the sample-size justification. These omissions limit evaluation of how the trial was planned and of outcomes beyond the primary physiological measure. They do not alter the posted primary estimates, but they constrain conclusions about their broader scope.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Comparison of Sugammadex With Neostigmine During Laparoscopic Cholecystectomy or Appendectomy (P05699) — NCT00724932.
• Registry record last updated 2017-05-16; status last verified 2017-04.
• Results first posted 2013-04-29.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

NCT00724932 is a completed phase 3 randomized, parallel-group, active-controlled trial of sugammadex and neostigmine in adults undergoing laparoscopic cholecystectomy or appendectomy under propofol anesthesia. The registry lists enrollment of 140 and primary completion on 2009-04-01. Sugammadex was administered at 1-2 PTC; neostigmine, with atropine, was administered at reappearance of T2.

The primary outcome was time to recovery of the T4/T1 ratio to 0.9. The posted geometric mean times were 2.4 minutes for sugammadex and 8.4 minutes for neostigmine. The reported ANOVA comparison gave a geometric mean ratio of 3.4, a 95% confidence interval from 2.8 to 4.1, and p < 0.001. Serious adverse events were reported for 5 of 66 and 6 of 67 participants, respectively. The registry does not report a full analysis plan, the imputation method, or results for operating-room and post-anesthetic-care-unit length of stay.