Impact of Fructose on Metabolism, Energy Homeostasis and Magnetic Resonance Biomarkers in Nonalcoholic Fatty Liver Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Impact of Fructose on Metabolism, Energy Homeostasis and Magnetic Resonance Biomarkers in Nonalcoholic Fatty Liver Disease (NCT01930123), a non-randomized phase 2 study. The primary outcome is Change in Blood Glucose Level From Pre-fructose Administration to Post-fructose Administration in Control Cohort.

2. Clinical Question

The broader study description sets out goals involving liver imaging methods, magnetic resonance biomarkers, and metabolic responses related to fructose. It also describes a cross-sectional and longitudinal dietary-intervention aim. Those goals provide context, but the primary endpoint listed for this analysis is narrower: it concerns pre- to post-challenge blood glucose in healthy controls. The registry record does not establish whether every broader goal was completed.

3. Trial Design

The registry classifies the study as Phase 2, nonrandomized, unmasked, and parallel. It lists an enrollment of 118 and a status of terminated. The recorded start is 2013-10 and primary completion is 2020-01. A reason for termination is not reported in the registry record.

The arm descriptions identify patients with biopsy-proven NAFLD and healthy controls. Both groups are described as receiving a fructose infusion after 12 hours fasting. The description states that 70 subjects with NAFLD and 15 healthy controls were planned or described for the challenge. These arm-description counts are not the same as the counts listed for adverse-event risk sets, and the registry record does not explain the difference. It separately states that the primary outcome data were collected on 19 healthy controls.

Because allocation was nonrandomized, differences between the groups cannot automatically be attributed to NAFLD status or to any other single cause. The registry record itself warns that the arms are not a like-for-like control comparison. The primary endpoint does not make a between-arm comparison: it concerns change within the healthy control cohort. No random allocation procedure, masking procedure, or rationale for the design is reported beyond the classification fields.

The study lists an intravenous fructose challenge and blood draw. A secondary outcome is described as dynamic 31P changes in liver beta-ATP due to fructose injection. The record notes three outcome measures and three analyses, but does not provide details linking each analysis to a particular outcome. More detail about the planned measurement schedule, imaging protocols actually used, and analysis populations is not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Not applicable to the Patients with NAFLD arm. Data collected on 19 healthy controls.

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: Participants who completed the study.

6. Primary Endpoint

The registry lists the primary outcome as Change in Blood Glucose Level From Pre-fructose Administration to Post-fructose Administration in Control Cohort, measured over Baseline to approximately one hour post-fructose administration.

The primary endpoint is the change in blood glucose from pre-fructose administration to post-fructose administration in the control cohort. Its unit is mg/dL, its summary measure is the mean, and variability is to be described with the standard deviation. The stated timeframe is baseline to approximately one hour post-fructose administration. The endpoint description says it is not applicable to the Patients with NAFLD arm and that data were collected on 19 healthy controls.

As a general statistical explanation, a mean change summarizes the average direction and magnitude of within-cohort change. It can be useful for describing the short-term response to a challenge, while the standard deviation describes how spread out individual measurements are around the mean. Those summaries answer different questions: the mean concerns the average response, and the standard deviation concerns variation among observations.

The endpoint’s focus on change distinguishes it from a single post-challenge glucose measurement. However, the registry record does not report the baseline mean, post-challenge mean, mean change, standard deviation, confidence interval, or test result. It also does not state whether the reported change was calculated from paired observations or which statistical test was used. Therefore, the endpoint definition identifies what was intended to be measured, but it does not establish the size, precision, or statistical significance of the observed response.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record gives some information about the planned endpoint and design, but does not report the statistical analysis methods used for the primary result. The explanations below distinguish the statistical meaning of the listed measures from methods that are not documented in the record.

What does a mean change describe?

In general, a mean change is an average of changes measured over time for participants in a defined cohort. For this endpoint, the stated comparison is between glucose before fructose administration and glucose afterward, at approximately one hour. A mean conveys the average change in mg/dL; it does not indicate whether every participant changed in the same direction. Individual responses can vary, which is why the record identifies the standard deviation as a measure of dispersion.

For repeated measurements on the same participants, a paired analysis can account for the fact that each person's measurements are related. That is a general statistical option, not a reported method for this study. The registry record does not specify how the change was calculated, whether observations were paired in the analysis, how incomplete pairs were handled, or which test was used.

What does the standard deviation tell readers?

As a general explanation, the standard deviation describes the spread of participant-level values around their mean. It is not the same as the standard error, which describes uncertainty in an estimated mean, and it is not a confidence interval. A small spread does not by itself demonstrate a meaningful average change, while a larger spread can make an average harder to estimate precisely. The actual standard deviation for the primary outcome is not reported in the registry record.

How should a confidence interval be understood?

In general, a confidence interval expresses uncertainty around an estimated quantity under the assumptions of the analysis. Its width reflects precision: narrower intervals indicate more precise estimates than wider intervals. An interval can also help readers judge which effect sizes remain compatible with the data and analysis. No confidence interval for the primary endpoint is reported in the registry record, so its precision cannot be evaluated from the recorded result.

What can a p-value establish?

In general, a p-value measures how unusual data at least as extreme as those observed would be under a specified null model and its assumptions. It does not give the probability that the null hypothesis is true, measure the size of an effect, or establish clinical importance. The registry record does not report a p-value for the primary endpoint or identify the test used. Statistical significance therefore cannot be assessed from the registry record.

What does the design imply for comparisons?

Nonrandomized allocation means that group membership was not assigned by random allocation, according to the registry classification. In general, this can leave group comparisons open to confounding: characteristics that differ between groups may also be related to the measured outcome. The registry record states that the arms are not a like-for-like control comparison. It does not report a covariate-adjustment plan or enough analysis detail to assess whether potential confounding was addressed.

The primary endpoint is described within the healthy control cohort, so it should not be interpreted as a direct comparison of glucose changes between healthy controls and patients with NAFLD. The endpoint description says it is not applicable to the Patients with NAFLD arm. The record does not report a separate primary estimate for that group.

How should planning calculations be read?

The stated planning illustration uses a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. Its target effect is defined as a stated fraction of the observed effect. These assumptions describe a planning exercise; they are not reported as the study's original design justification. The sample-size justification is not reported in the registry record.

Planning calculations based on an observed effect are hypothetical estimates of what a future study might require under stated assumptions. They are not observed power and do not demonstrate that the present study was adequately powered. If a planning result is expressed as participants per arm for 80% power, that is a prospective planning quantity, not the number analyzed in the current study. The record’s note that non-time-to-event calculations use participants per arm also distinguishes these illustrations from observed outcome counts.

Why can sparse event data be difficult to interpret?

As a general statistical point, event counts that are small or absent can yield imprecise estimates, particularly when comparing groups. For count outcomes with variability greater than a simple model assumes, over-dispersion can make conventional uncertainty estimates too narrow; models that account for that extra variability may be appropriate in some settings. This is general guidance, not a description of a method used in this study. The registry record does not report an over-dispersion assessment or a model for adverse events.

Overall, the registry describes the endpoint and several design features, but does not report enough analysis detail to reconstruct the primary statistical result. The outcome estimate, uncertainty, test, missing-data handling, and adjustment strategy are not reported in the registry record.

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 Controls")
n   <- c(19)
m   <- c(6.26)
sp  <- c(9.98)
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 Controls  19  6.260  9.980  1.450  11.070

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists three outcome measures and three analyses, and names dynamic 31P changes in liver beta-ATP due to fructose injection as a secondary outcome. It does not identify all outcome measures or explain how the analyses map to the outcomes. A plan for controlling multiplicity, such as adjusting for the number of statistical tests or defining a hierarchy of outcomes, is not reported in the registry record.

As a general statistical concern, testing multiple outcomes or conducting multiple analyses can increase the chance of at least one apparently unusual result under a collection of null hypotheses. Without a prespecified strategy, readers cannot tell whether a reported result was treated as confirmatory or exploratory. Here, the registry record does not report such a strategy, and it also does not provide the primary outcome’s test result.

Missing-data methods and the handling of incomplete measurements are not reported in the registry record. The endpoint description states that data were collected on 19 healthy controls, while other registry entries describe different group counts. The record does not clarify how many participants had complete paired glucose measurements, whether any values were excluded, or how the analysis population was defined.

The record does not describe an intention-to-treat population or another formal analysis population for the primary endpoint. It does not report imputation, sensitivity analyses, interim analyses, or rules for handling protocol deviations. Their use should not be assumed. Without these details and the outcome estimate, it is not possible to evaluate the robustness of the primary analysis or the potential effect of missing observations.

12. 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 non-serious adverse event terms

The adverse-event record lists serious adverse events affecting 0 participants in the Patients With NAFLD group among 81 at risk, and 0 participants in the Healthy Controls group among 37 at risk. It lists other adverse events affecting 1 participant among 81 at risk in the Patients With NAFLD group and 0 participants among 37 at risk in the Healthy Controls group. Deaths are recorded as 0 in both groups.

These are registry counts, not a comparison of event rates adjusted for follow-up or participant characteristics. The record does not provide event descriptions for the other adverse event, and no top serious-event terms are listed. It also does not report the observation period or explain how the adverse-event risk sets relate to the arm-description counts. Those counts differ from the group sizes described for the fructose challenge, and the reason is not reported in the registry record.

Zero recorded serious events or deaths in these groups does not establish that risk was absent. The figures describe what was recorded in the registry, not a guarantee about events outside the recorded follow-up or an estimate of comparative safety. Because allocation was nonrandomized and the registry does not give further safety-analysis methods, causal conclusions about the challenge and adverse events are not supported by these counts alone.

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

The registry record establishes the intended primary measurement: mean blood-glucose change from before to after fructose administration in healthy controls, with standard deviation as the stated measure of variability. It also describes the approximate measurement window and specifies the unit. These details help define the question, but the record does not report the result needed to answer it.

Without the mean change and its uncertainty, the magnitude, direction, and precision of the observed glucose response cannot be judged. The record does not report a statistical test or p-value, so statistical significance cannot be assessed either. The absence of a reported result is not evidence that there was no change; it means the result is not reported in the registry record.

The primary endpoint does not directly compare patients with NAFLD against healthy controls. The study is classified as nonrandomized, and the registry cautions that its arms are not a like-for-like control comparison. Even if a group contrast were available, differences could reflect group characteristics or other factors, and no adjustment strategy is reported in the record.

The broader study aims include metabolic and imaging questions, and a secondary outcome concerns dynamic liver beta-ATP changes. The existence of those aims and listed outcomes does not establish that the measurements were completed or that they produced a particular finding. Results for these outcomes are not reported in the registry record.

The adverse-event entries describe recorded counts, including no serious adverse events or deaths in either listed group. They do not establish absence of risk or prove a causal safety relationship. Overall, the record defines aspects of the study question and design, but it does not provide the primary outcome results needed to characterize the glucose response or draw conclusions about group differences.

14. Why This Trial Matters Statistically

This study’s statistical questions span a short-term metabolic response and broader aims involving liver imaging and magnetic resonance biomarkers in NAFLD. The primary endpoint is specific: blood-glucose change after fructose administration in healthy controls. Keeping that endpoint distinct from the wider study aims helps prevent the stated research goals from being mistaken for reported findings.

The design also illustrates why allocation matters when interpreting comparisons. The registry classifies the study as nonrandomized and states that its arms are not a like-for-like control comparison. This matters because statistical differences between groups do not necessarily isolate the effect of NAFLD status or of a challenge; group characteristics may offer alternative explanations. A clearly defined within-cohort endpoint and a between-group contrast answer different questions.

Reporting the mean together with dispersion can help distinguish an average response from variation among participants. But those quantities need to be available, along with an uncertainty measure, for readers to assess the estimate. In this record, the endpoint is described but its results are not reported. That gap demonstrates why outcome definitions alone cannot establish the size or precision of a finding.

The record also contains group counts in different contexts: arm descriptions, adverse-event risk sets, and the primary outcome population. Their relationship is not explained. Clear accounting of analysis populations and missing observations is important for understanding which participants contribute to each result and for judging how representative an estimate may be.

Finally, the listed planning assumptions should be separated from the study’s observed evidence. A planning calculation based on an observed effect is a hypothetical design aid, not observed power or proof that an analysis was adequately powered. This distinction is important when an outcome result is unavailable and no sample-size justification is reported.

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 primary limitation for interpreting the main question is that the registry record defines the blood-glucose endpoint but does not report its numerical result. No mean change, standard deviation, confidence interval, or p-value is included. The direction, magnitude, and precision of the response therefore cannot be assessed.

The primary endpoint applies to healthy controls and is not applicable to the Patients with NAFLD arm. Although the record describes controls as a comparison group, the primary endpoint is not a direct comparison between the groups. Nonrandomized allocation and the registry’s warning that the arms are not a like-for-like control comparison further limit causal interpretation of any group contrast.

The record gives differing participant counts in separate contexts: arm descriptions, adverse-event risk sets, and the stated primary outcome population. It does not explain how these counts relate to enrollment, eligibility, completion, or analysis. The record also does not report missing-data procedures, the number with complete pre- and post-challenge measurements, or exclusions from the primary analysis.

Other analysis details are absent, including the statistical test, any covariate adjustment, the multiplicity strategy, and a sample-size justification. The registry lists three outcome measures and three analyses but does not provide a complete account of their relationship. The reason for the terminated status is not reported in the registry record. These omissions limit the ability to evaluate the analysis plan, robustness, and completeness of the evidence.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Impact of Fructose on Metabolism, Energy Homeostasis and Magnetic Resonance Biomarkers in Nonalcoholic Fatty Liver Disease — NCT01930123.
• Registry record last updated 2022-08-31; status last verified 2022-08.
• Results first posted 2022-08-31.
• 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

The registry describes a Phase 2, nonrandomized, unmasked, parallel study of fructose challenge in patients with NAFLD and healthy controls. It lists enrollment of 118 and a status of terminated. The primary endpoint is mean blood-glucose change from before to approximately one hour after fructose administration in healthy controls, measured in mg/dL with standard deviation as the stated dispersion measure.

The record says the primary outcome data were collected on 19 healthy controls and that the endpoint does not apply to the Patients with NAFLD arm. It lists a secondary outcome involving dynamic 31P changes in liver beta-ATP. The primary result, its uncertainty, and the statistical test are not reported in the registry record. The record also does not explain differences between the arm-description counts, adverse-event risk sets, and primary outcome population.