Pharmacokinetics of Empagliflozin (BI 10773) in Patients With Impaired Liver Function: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pharmacokinetics of Empagliflozin (BI 10773) in Patients With Impaired Liver Function (NCT01111318), a non-randomized phase 1 study. The primary outcome is Area Under the Curve 0 to Infinity (AUC0-∞).

2. Clinical Question

The study is framed as a pharmacokinetic comparison across liver-function categories, not as a trial of clinical benefit. Its central measurement is plasma exposure after dosing. The registry also lists measures of peak timing, elimination, distribution, residence time, and urinary elimination, along with safety and tolerability as objectives. Those additional measurements can describe different parts of the drug's disposition, but their numerical results are not reported in the registry record available for this analysis.

Because participants were not randomly assigned, observed group differences could reflect differences between the groups as well as liver-function category. The registry record does not report enough information to assess how groups were assembled or whether other characteristics were comparable. Accordingly, the design supports describing the measured pharmacokinetics across the listed categories; it does not by itself establish that liver impairment caused any observed difference.

3. Trial Design

This is a Phase I, non-blinded, parallel-group study with non-randomized allocation. It enrolled 36 participants and is marked completed in the registry record. The listed groups are healthy subjects and subjects with mild, moderate, or severe hepatic impairment. The official title identifies the impairment categories as Child-Pugh A, B, and C, respectively. The intervention was BI 10773, given as a single oral dose of 50 mg.

The stated objective was to assess the effect of mild, moderate, and severe hepatic impairment on pharmacokinetics, safety, and tolerability after the single dose. The primary endpoint is the area under the plasma concentration-time curve from zero extrapolated to infinity. Secondary outcomes include area under the curve to the last quantifiable data point, time to maximum concentration, terminal rate constant, terminal half-life, mean residence time, apparent clearance, apparent terminal-phase volume of distribution, and the amount eliminated in urine over the specified collection interval.

Non-blinding means that masking was not used; the record lists masking as none. Parallel-group means the categories are represented as separate groups rather than as repeated treatment periods in the same participants. Since allocation was non-randomized, the healthy group is a reference group, not a randomized control arm. Differences in measured outcomes therefore require cautious interpretation and cannot automatically be attributed to hepatic status alone.

The registry lists Boehringer Ingelheim as sponsor, a start in 2010-07, and primary completion in 2010-11. Details such as the recruitment process, the criteria used to define each group beyond the Child-Pugh categories, and any procedures intended to improve comparability are not reported in the registry record. Nor is a protocol rationale for the enrollment target reported. These omissions limit assessment of how closely the enrolled groups represent the intended categories and how the study's size was chosen.

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

Analysis population for the primary outcome, as defined in the registry: Pharmacokinetic (PK) set included all subjects who were documented to have taken the investigational treatment, who provided at least one observation for at least one primary PK endpoint, without important protocol violations relevant to the evaluation of PK, provided no vomiting occurred at or before two times median tmax.

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 Area Under the Curve 0 to Infinity (AUC0-∞), measured over Pre-dose and 20minutes (min), 40min, 1h, 1.5h, 2h, 2.5h, 3h, 4h, 6h, 8h, 10h, 12h, 16h, 24h, 36h, 48h, 72h and 96h after drug administration.

The primary endpoint is the area under the empagliflozin plasma concentration-time curve from zero extrapolated to infinity, abbreviated AUC0-∞ and measured in nmol*h/L. It represents integrated drug concentration over time, including an estimated portion of the curve beyond the final quantifiable observation. In general, area under a concentration-time curve is used as a summary of systemic exposure: it captures both the concentration levels and how long they persist, rather than focusing on a single sampling time.

The registry lists blood sampling from pre-dose through 96h after administration, with measurements at multiple scheduled times. Those observations form the concentration-time profile from which the exposure measure is derived. The record's endpoint description says the standard deviation is actually the coefficient of variation. It also distinguishes the variability displayed among measured values, described as inter-individual, from the variability used in statistical analyses, described as intra-individual. These distinctions matter because a dispersion measure communicates how much observations vary, and the interpretation depends on what kind of variability it represents.

AUC0-∞ is useful for comparing exposure across liver-function groups, but the record does not report numerical group results for this endpoint. It also does not state the extrapolated fraction of the estimate or other details needed to independently assess how much of the endpoint depended on extrapolation. Thus, the endpoint's purpose and measurement schedule are described, while its empirical result and some estimation details are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a continuous primary outcome summarized by a mean, with a dispersion label that it clarifies is actually a coefficient of variation. The study is non-randomized and parallel-group, and the stated primary endpoint is AUC0-∞. The statistical analysis methods themselves are not reported in the registry record. The explanations below describe general statistical concepts rather than claiming that a particular unreported method was used.

What does a mean summarize?

In general, a mean is a measure of the center of a set of numeric observations. For a pharmacokinetic endpoint, a group mean can summarize the measured exposure in the analyzed participants. A mean alone does not show how observations are distributed, how much they vary, or how precisely the group value is estimated. Those questions require dispersion and uncertainty information, such as a standard deviation, coefficient of variation, or confidence interval. The registry identifies the parameter as a mean, but numerical group means and confidence intervals are not reported in the registry record.

Why distinguish standard deviation from coefficient of variation?

In general, standard deviation describes spread in the same units as the measured outcome. The coefficient of variation describes spread relative to the mean, so it is often useful when variability is considered in relation to the scale of the measurement. They answer related but different questions and should not be substituted for one another without explanation. The endpoint description explicitly notes that the reported standard deviation is actually a coefficient of variation. It also says that measured values show inter-individual variability while statistical analyses show intra-individual variability. Inter-individual variation concerns differences between people; intra-individual variation concerns variation within a person. A statistic based on one source of variation should not be read as if it represented the other.

In general, variability among people can be substantial in pharmacokinetic measurements, and a group summary can conceal that distribution. A narrow-looking estimate of a group mean does not necessarily imply that individual values are tightly clustered, and a broad spread among individual observations does not by itself establish that the group mean is estimated imprecisely. The latter also depends on the amount and structure of available information. The registry's distinction between measured-value variability and analysis variability is therefore important, but the detailed model and its assumptions are not reported in the registry record.

What does a confidence interval convey?

In general, a confidence interval gives a range of values compatible with the data and the statistical procedure used, reflecting uncertainty in an estimate. It is not a guarantee that the true value lies in the interval, and it does not describe the range containing most individual participants' measurements. For a comparison, an interval around a contrast can help show both the estimated direction and the precision of the comparison. No confidence-interval values for the primary endpoint are reported in the registry record, so the precision of the group estimates cannot be assessed from the listed details.

Why does the non-randomized design matter?

Random allocation, when feasible and appropriate, helps balance measured and unmeasured characteristics on average between treatment groups. This study instead compares groups defined by hepatic status, with non-randomized allocation. The comparison can describe differences in the observed groups, but potential group differences beyond liver function may also contribute. Statistical adjustment can sometimes address measured factors, but it cannot automatically remove unmeasured confounding. The registry record does not report an adjustment strategy or enough group-characteristic information to judge comparability. General causal interpretation should therefore be restrained.

Parallel-group designs observe separate groups rather than exposing the same participant to multiple conditions in sequence. That design avoids needing to interpret within-person crossover periods, but it leaves group comparisons dependent on how similar the groups are. The record does not report procedures or matching methods used to establish comparability. It is not appropriate to assume that the groups were matched or balanced.

What do the planning calculations mean?

The sample-size assumptions describe a planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. Such calculations ask, in general, how many participants per arm might be needed under specified assumptions to detect an effect of a chosen size with a chosen power. They are not evidence of the study's original design rationale, and they are not observed power. The registry does not report a sample-size justification from the original protocol. Values from retrospective planning illustrations should be read as hypothetical design scenarios, not as results of the completed study.

A normal approximation is a mathematical simplification used to estimate the distribution of a statistic. Its suitability depends on the data and analysis assumptions; the registry record does not provide diagnostics or other information to evaluate that suitability. Equal allocation is likewise an assumption for the planning illustration and should not be taken as a description of actual enrollment in each group. The record does not report the original allocation target by group.

How should multiple outcomes affect interpretation?

In general, testing many outcomes can increase the chance that at least one apparently notable result arises by chance, especially if each is judged separately without adjustment. A multiplicity strategy may control that risk, prioritize a primary endpoint, or distinguish confirmatory from exploratory outcomes. The registry identifies one primary outcome and lists several secondary pharmacokinetic outcomes, but it does not report whether or how multiplicity was addressed. Nor does it report the tests, model specifications, or decision thresholds for the analyses. The outcome list alone is not enough to establish the statistical plan.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Healthy", "Mild", "Moderate", "Severe")
n   <- c(12, 8, 8, 8)
m   <- c(10800, 13800, 16100, 19000)
sp  <- c(22.6, 38.6, 26.2, 27.1)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Healthy  12  10800.000  22.600  10785.641  10814.359
CARM  Mild  8  13800.000  38.600  13767.730  13832.270
CARM  Moderate  8  16100.000  26.200  16078.096  16121.904
CARM  Severe  8  19000.000  27.100  18977.344  19022.656

11. Multiplicity, Missing Data and Analysis Populations

The registry identifies AUC0-∞ as the primary endpoint and lists secondary outcomes covering AUC0-tz, Tmax, λz, t1/2, MRTpo, CL/F, Vz/F, and Ae0-96. It records 14 outcome measures and 6 analyses. Having a designated primary endpoint can help distinguish the main question from supporting pharmacokinetic descriptions, but the registry record does not report a hierarchy among outcomes, a testing sequence, or a multiplicity adjustment. It is therefore not possible to determine whether the analyses were treated as confirmatory, exploratory, or a mixture of both.

Rules for missing observations are not reported in the registry record. The sampling schedule is described, but the record does not state how incomplete concentration-time profiles were handled, what minimum number of samples was required for each endpoint, or how an endpoint was calculated when data were unavailable. The endpoint description notes extrapolation to infinity, but the record does not report detailed rules for deciding whether that estimate was acceptable. These details matter because missing or unquantifiable measurements can affect which participants contribute to a given pharmacokinetic summary.

The primary endpoint population is described as a pharmacokinetic set including subjects documented to have taken treatment and to have at least one observation for at least one primary pharmacokinetic endpoint, with exclusions for important protocol violations relevant to pharmacokinetic evaluation. The description then becomes incomplete in the registry record. The full criteria and the numbers included in each analysis are not reported in the available record. It is also not clear from the listed details whether all outcome analyses used the same population.

These omissions limit reproducibility and make it difficult to assess whether exclusions, missing samples, or multiple analyses influenced the reported findings. Numerical results for the listed endpoints and a complete statistical analysis plan are 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 non-serious adverse event terms

The registry lists serious adverse events separately from other adverse events for each group. It records 0 serious adverse events among 12 participants at risk in the healthy group, 0 among 8 in the mild group, 0 among 8 in the moderate group, and 0 among 8 in the severe group. For other adverse events, it records 6 affected participants in the healthy group, 0 in the mild group, 3 in the moderate group, and 2 in the severe group. These are the recorded counts; they should not be converted into percentages or combined across groups to create a new total.

The record does not identify serious adverse-event terms in the listed serious-event summary. Death counts are not reported in the registry record. The other-adverse-event counts do not by themselves indicate event severity, whether an event was considered related to study drug, when it occurred, or whether participants experienced more than one event. Those details are not reported in the registry record.

These figures are descriptive safety information from a single-dose study. With no randomized allocation, the group counts are not a like-for-like controlled comparison. Counts of affected participants alone do not establish comparative risk or causation, and the absence of recorded serious events in these groups should not be interpreted as proof that serious events cannot occur.

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

The study asks a focused pharmacokinetic question: how exposure after a single oral dose of BI 10773 varies across healthy subjects and subjects categorized by degree of hepatic impairment. AUC0-∞ is an appropriate type of summary for describing overall plasma exposure because it incorporates concentration across time rather than relying on one measurement. The registry also identifies a broader set of pharmacokinetic outcomes and a safety and tolerability objective.

What the record does not establish is the size or direction of any exposure difference. Numerical results for the primary endpoint are not reported in the registry record, and the listed details do not include confidence intervals or a numerical comparison. It is therefore not possible to say whether exposure was higher, lower, or similar across groups, or how precisely any difference was estimated. The other listed pharmacokinetic outcomes likewise cannot be interpreted without their numerical results.

Even if group results are considered, the non-randomized design means that an observed contrast is descriptive rather than automatically causal. Participants in different hepatic-status groups may differ in other ways. The registry does not report enough information about group characteristics, adjustment, or allocation procedures to determine whether such differences were accounted for. The healthy group provides a reference, but not a randomized counterfactual.

The adverse-event counts describe what was recorded among the listed groups, including no recorded serious adverse events in the group entries and counts of other affected participants. These data do not establish a treatment-related effect or rule out uncommon harms. Details about event terms, causal attribution, and deaths are not reported in the registry record.

Accordingly, the record supports understanding the study's objective, structure, endpoints, and some descriptive safety counts. It does not support a quantitative conclusion about pharmacokinetic differences or a recommendation about use. A planning illustration based on an observed effect, if considered alongside the posted results, would address hypothetical future-study size under specified assumptions; it would not supply missing evidence about this study's precision or represent observed power.

14. Why This Trial Matters Statistically

This trial is statistically informative as an example of a specialized pharmacokinetic comparison in which the groups are defined by liver-function category rather than randomized treatment assignment. That distinction affects what the data can establish. Pharmacokinetic studies often seek to characterize exposure across clinically meaningful groups, but a comparison of separate groups must be interpreted with attention to the possibility that group composition also differs. The design determines the strength and scope of conclusions, not just the choice of statistical test.

The primary endpoint illustrates why outcome definition and uncertainty reporting matter. AUC0-∞ compresses a sequence of plasma concentration measurements into an exposure summary, while depending on both observed data and extrapolation beyond the last quantifiable point. A clear definition helps readers understand the target quantity, but numerical estimates and uncertainty are needed to evaluate the result. Those primary endpoint results are not reported in the registry record.

The record's clarification about dispersion is another important statistical point. A coefficient of variation and a standard deviation describe variability in different ways, and inter-individual and intra-individual variation refer to different sources of spread. Keeping these concepts distinct prevents readers from attributing a statistical summary to the wrong kind of variation. The registry explicitly notes these distinctions, but does not report the full analysis method needed to reproduce them.

Finally, a primary outcome alongside numerous additional measures raises questions about analysis planning and multiplicity. A prespecified hierarchy and transparent missing-data rules help readers distinguish the principal question from supporting analyses. The registry lists outcomes and analyses but does not report those methods or numerical outcome results. The study therefore highlights both the value of a well-defined pharmacokinetic objective and the limits imposed when results and analysis details are unavailable.

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 most consequential limitation for interpreting results is that numerical primary endpoint estimates are not reported in the registry record. Without group-level AUC0-∞ values and uncertainty estimates, it is not possible to assess the magnitude or precision of exposure differences. Numerical results for the secondary pharmacokinetic measures are also not included in the listed details, so their role as supporting evidence cannot be evaluated.

The non-randomized design limits causal interpretation. The registry does not report group characteristics, recruitment details, matching, or statistical adjustment, so it is not possible to judge whether the comparison groups were similar in factors other than hepatic status. Lack of masking is recorded, but the record does not explain how this affected measurement or study conduct.

The endpoint description contains an explicit terminology clarification: the stated standard deviation is actually the coefficient of variation, and measured values represent inter-individual variability while statistical analyses represent intra-individual variability. This is useful context, but the underlying model and calculation details are not reported. The endpoint population description is also incomplete in the registry record, making its full inclusion and exclusion criteria unavailable.

Missing-data procedures, the complete analysis plan, and any strategy for handling multiplicity are not reported. The record gives a sampling schedule but does not provide enough detail to assess incomplete profiles, endpoint-specific exclusions, or the influence of extrapolation on AUC0-∞. The planning assumptions do not replace an original sample-size justification, which is not reported.

Safety interpretation is limited by the absence of event descriptions, attribution details, and death counts in the registry record. The reported counts alone cannot establish comparative risk, especially in a non-randomized study. These limitations mean the record supports description of the design and objectives more strongly than evaluation of the study's pharmacokinetic or safety findings.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Pharmacokinetics of Empagliflozin (BI 10773) in Patients With Impaired Liver Function — NCT01111318.
• Registry record last updated 2014-06-13; status last verified 2014-05.
• Results first posted 2014-06-13.
• 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

NCT01111318 is a completed Phase I, non-blinded, parallel-group, non-randomized study sponsored by Boehringer Ingelheim. It enrolled 36 participants and evaluated a single oral 50 mg dose of BI 10773, identified as empagliflozin, in healthy subjects and subjects with mild, moderate, or severe hepatic impairment categorized by Child-Pugh classification.

The primary endpoint was AUC0-∞, a measure of plasma exposure. Secondary pharmacokinetic outcomes included measures of concentration-time profile, elimination, distribution, and urinary excretion. The registry also describes safety and tolerability as study objectives.

Numerical results for the primary and secondary pharmacokinetic endpoints are not reported in the registry record. The record includes adverse-event counts by group, but does not report event attribution details or death counts. Non-randomized allocation, incomplete analysis information, and unavailable endpoint estimates constrain interpretation of comparative findings.