Metronidazole for Pulmonary Tuberculosis (South Korea): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Metronidazole for Pulmonary Tuberculosis (South Korea) (NCT00425113), a randomized phase 2 study. The primary outcome is Changes in TB Lesion Sizes Using High Resolution Computed Tomography (HRCT)..

2. Clinical Question

3. Trial Design

This was a randomized, double-blind, parallel-group Phase II study of metronidazole added to antituberculosis chemotherapy versus placebo added to antituberculosis chemotherapy. Its registered title identifies the population as subjects with multi-drug resistant pulmonary tuberculosis. The study status is completed, and the posted enrollment is 35. The record lists the National Institute of Allergy and Infectious Diseases (NIAID) as sponsor.

The treatment contrast was metronidazole or placebo on top of a background tuberculosis treatment regimen during the initial 2 months. The brief summary describes an 8-week course of standard second-line treatment in each group. The record does not provide a complete account of the background regimens actually received, treatment adherence, later treatment, or any changes made during follow-up. These details are not reported in the registry record.

The registered study start is 2006-12 and primary completion is 2012-10. The brief summary describes sputum collection, blood testing, PET and CT imaging. The primary endpoint was based on high-resolution CT assessments. The record says that three radiologists independently scored lung zones while blinded to treatment arm, with a fourth radiologist adjudicating scores that were widely discrepant among the initial readers.

Randomization and masking describe design features; they do not by themselves establish that groups remained comparable in follow-up, that masking was maintained, or that all randomized participants were included in each analysis. The registry record does not report the allocation method, concealment procedures, the number assigned to each arm for the primary endpoint, or the analysis population. Those details are not reported in the registry record.

Although the study is marked completed, the registry record lists no analyses for the primary outcome and does not provide its results. Completion status therefore should not be read as evidence that the registered endpoint was reported or that a treatment effect was established.

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:

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: 2 subjects excluded: 1 for uncontrolled diabetes mellitus and 1 for diagnosis of extrapulmonary TB. 2 additional subjects withdrew: 1 self- withdrew and the other was withdrawn for missing \>14 days of study drug.

6. Primary Endpoint

The registry lists the primary outcome as Changes in TB Lesion Sizes Using High Resolution Computed Tomography (HRCT)., measured over 6 months..

The primary endpoint is change in TB lesion sizes assessed using high-resolution computed tomography (HRCT), reported as a reader score. The registry describes lesions including nodules, consolidations, collapse, cavities, fibrosis, bronchial thickening, tree-in-bud opacities and ground glass opacities. Each CT was divided into six lung zones, and readers scored the extent of each lesion type in each zone.

For each lesion in each zone, the scoring scale ran from 0 for no involvement to 4 for 76-100% involvement. A composite score for a lesion was calculated from the scores across the six zones, and the composite sums of all 10 lesion scores were reported. Change in the composite score was measured at 2 and 6 months compared with baseline. The primary endpoint description identifies a 6-month timeframe, while the measurement description also specifies change at 2 months.

Imaging provides a structured way to quantify radiographic abnormalities over time. Composite scoring can summarize multiple lesion types and lung regions, while independent blinded readings and adjudication are intended to address reader variability. These are general features of imaging endpoints; they do not establish that a score is a direct measure of symptoms, bacterial clearance, or long-term clinical outcome.

The endpoint is continuous, and the registry identifies the summary parameter as a mean with standard error. However, no mean changes, standard errors, group contrast, confidence interval or test result for this endpoint are reported in the registry record. Its observed pattern and precision therefore cannot be assessed from the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registered design is randomized, double-blind and parallel. In general, random assignment is used to reduce systematic differences between groups at the start of a trial. A parallel design assigns participants to a treatment group that is compared with another group, rather than having each participant receive both treatments in sequence. Double-blind masking is intended to limit the influence of treatment knowledge on care, reporting or assessment. The registry record identifies these design features but does not report enough operational detail to assess how assignment and masking were implemented or maintained.

What does the primary measure summarize?

The primary measure is a continuous composite score based on HRCT lesion assessments. The record describes scoring each of six lung zones for several forms of abnormality and combining zone-level scores for each lesion. It then describes composite sums across all 10 lesion scores. A continuous measure can represent a range of values rather than only a category such as improved or not improved. In general, a mean is a measure of the center of such values. The registry identifies the mean as the summary parameter, but it does not report the group means or mean changes.

Composite outcomes combine information from more than one component. This can provide a broad summary, but different components may contribute differently to the total and may change in different directions. The registry does not report component-specific results or enough outcome data to examine how particular lesion types contributed to any composite change. Without those results, the composite’s observed behavior cannot be described.

What does the standard error convey?

The registry identifies standard error as the dispersion measure. In general, a standard error describes the precision of an estimated quantity, such as a group mean, rather than the spread of individual participants’ scores. A smaller standard error generally indicates a more precise estimate under the analysis assumptions. It is not a substitute for reporting the estimate itself, and it does not show the range of individual outcomes. No primary endpoint standard errors are reported in the registry record.

A confidence interval is a range of values compatible with the data and the method used to construct it. In general, it helps readers assess both the direction and precision of an estimated contrast. A p-value addresses how compatible data are with a specified null hypothesis under a statistical model; it does not measure the size or clinical importance of an effect. No primary endpoint confidence interval or p-value is reported in the registry record.

Why might the analysis account for repeated measurements?

The endpoint description includes change from baseline at 2 and 6 months. Repeated measurements on the same participant are generally correlated, because a participant’s scores across visits are not independent observations. An analysis that uses repeated measurements should account for this dependence rather than treating each observation as unrelated. The statistical model and analysis approach used for this endpoint are not reported in the registry record. The record also does not say how missing visits were handled or whether the analysis adjusted for baseline score.

The endpoint has several lesion categories and time points. If multiple comparisons are made, the chance of at least one apparently unusual result can increase when no adjustment or prespecified hierarchy is used. This is a general statistical concern, not a claim about the analysis of this trial. The registry record does not report a multiplicity strategy, a primary contrast across the time points, or how category-level findings were to be interpreted.

How should the imaging readings be understood?

The registry describes three radiologists independently scoring the scans while blinded to treatment arm, with a fourth radiologist adjudicating scores that were widely discrepant. Independent assessment can reduce dependence on a single reader, and blinding is intended to limit bias from knowing treatment assignment. Adjudication can resolve selected disagreements. These procedures do not, by themselves, quantify reader agreement or guarantee that measurement error is absent. The record does not report reader agreement statistics, adjudication frequency, or reliability estimates.

The scoring scale represents estimated lesion involvement in ranges. Such scoring can impose ordered categories on a visual assessment. A composite derived from these scores can be analyzed as a continuous measure, as the registry indicates, but the record does not report model diagnostics or justify the assumptions used for that analysis. In general, statistical interpretation depends on whether the analysis method adequately represents the data and their variability.

What is the role of planning calculations?

Sample-size calculations are planning illustrations, not observed power. In general, a planning calculation depends on assumptions about the effect to detect, variability, significance level, allocation and analysis method. The record’s planning assumptions describe a two-sided significance level of 0.05, equal allocation, a normal approximation and effects expressed as fractions of an observed effect. Such calculations do not establish what the trial actually observed, and they cannot replace an estimated effect with an uncertainty interval. The primary outcome results needed to interpret the observed effect are not reported in the registry record.

For this trial, no numerical sample-size illustration values are stated in the registry record. The record also does not report a sample-size justification from the trial’s own design. The planning assumptions therefore should not be mistaken for a reported design rationale, an observed power calculation, or evidence that the primary endpoint was adequately precise.

10. Multiplicity, Missing Data and Analysis Populations

The primary endpoint description includes more than one time point and multiple lesion categories. The registry record explicitly indicates that the outcome was not reduced to one comparison because of its several categories or time points. This makes it important to distinguish the registered endpoint definition from a single, clearly identified treatment contrast. The record does not report which comparison was designated as the primary statistical test, whether the time points were ordered in a testing hierarchy, or whether any adjustment for multiple comparisons was planned or used. These details are not reported in the registry record.

Missing data can affect longitudinal outcomes when participants lack a measurement at one or more visits. In general, the impact depends on how much data are missing, why they are missing and the analysis assumptions used. The registry record does not report the number of participants with HRCT data at each assessment, missingness by treatment group, reasons for missing observations, or the method used to handle missing outcome data. It is therefore not possible to assess how missing follow-up may have affected the primary endpoint estimate.

The record also does not identify the primary analysis population or state whether the primary analysis followed randomized assignment, included only participants who completed treatment, or used another definition. It does not report protocol deviations or exclusions from the endpoint analysis. Randomization is a design feature, but interpretation of a randomized comparison also depends on how participants are handled in the analysis.

The sponsor’s statistical plan, any interim analyses and any prespecified multiplicity procedure are not reported in the registry record. Because no primary outcome analyses or results are posted, readers cannot verify the analysis against a prespecified plan or evaluate sensitivity to missing-data and modeling choices.

11. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: peripheral neuropathy (26.7%), seizure (20.0%), hepatitis (13.3%), aspergilloma (6.7%), hemoptysis (6.7%).

The registry reports serious adverse events for 9 of 15 participants at risk in the metronidazole group and 4 of 16 at risk in the placebo group. Other adverse events are reported for 11 of 15 and 8 of 16 at risk, respectively. These are counts of affected participants in the registry’s group-level fields, not estimates adjusted for follow-up time or other differences.

The serious-event terms listed are peripheral neuropathy, seizure, hepatitis, aspergilloma, hemoptysis and gastritis. The registry gives percentages for these terms but does not provide the corresponding event counts or explain in detail how repeated events or participants with multiple terms are represented. The term percentages should therefore be read as registry descriptions, not as a complete comparative analysis.

These figures do not establish that treatment caused an event or that event risks differed between groups. The record does not report a statistical comparison, confidence intervals, event timing, exposure duration by participant or a detailed severity breakdown. Deaths are not reported in the registry record. Interpretation of these safety data is limited to the posted counts and term descriptions; no additional causal or comparative conclusion can be drawn from them alone.

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

The trial addresses a specific comparison: metronidazole added to background tuberculosis treatment versus placebo added to background treatment in a randomized, double-blind Phase II study of multi-drug resistant pulmonary tuberculosis. The primary endpoint is an HRCT-based composite score intended to describe changes in lung lesions. The registry record supplies the endpoint construction and the planned assessment timing, including change at 2 and 6 months compared with baseline.

What the record does not provide is the primary endpoint result. Without group estimates, an estimated between-group contrast and a measure of uncertainty, readers cannot tell whether the scores changed, whether the groups differed, or how precise any difference may have been. The record also does not provide the analysis methods needed to assess how the multiple lesion categories and time points were handled. The study’s completed status is not a substitute for reported outcome data.

The adverse-event entries offer descriptive counts for serious and other adverse events, along with selected serious-event terms. They do not establish causation or a statistically supported difference between groups. Missing details about event timing, analysis and deaths further limit interpretation. The registry information therefore supports a description of what was recorded, not a determination of treatment benefit or harm.

More generally, an endpoint can be carefully defined and still leave the main comparative question unanswered if its results are unavailable. A measured imaging score is not interchangeable with sputum culture conversion, symptoms or other outcomes. The registry lists time to sputum culture conversion to negative on solid medium as a secondary outcome, but does not provide its results either.

The appropriate conclusion from this record is limited: it documents the study design, intervention contrast, imaging endpoint and certain adverse-event counts, while the primary and secondary outcome analyses are not reported in the registry record. It does not support a claim that metronidazole did or did not change the imaging score or any other clinical outcome.

13. Why This Trial Matters Statistically

This trial illustrates why a clear endpoint definition is not the same as an interpretable treatment result. The registry describes the construction of an HRCT composite from lesion assessments across lung zones, the use of blinded readers and the planned comparison of change from baseline. Those details help explain what the study intended to measure. Without the endpoint estimates and their uncertainty, however, the comparative effect remains unknown.

It also highlights the importance of prespecifying how multiple components and time points will be interpreted. The primary endpoint description includes several lesion categories and measurements at 2 and 6 months, while the registry indicates that the outcome was not reduced to one comparison. In general, an analysis plan should clarify which contrast carries the primary inference and how other comparisons will be handled. That plan is not reported in the registry record.

Imaging outcomes involve measurement decisions as well as statistical ones. Reader scoring, aggregation into composites and the handling of disagreement can influence the resulting data. The registry provides useful procedural detail about independent readings and adjudication, but it does not report reader reliability or the statistical model applied to the scores. This makes it difficult to evaluate how measurement variability was incorporated into the endpoint analysis.

Finally, the record demonstrates the distinction between completion, safety reporting and efficacy reporting. The study is marked completed and includes group-level adverse-event counts, yet the primary outcome analyses are not reported. Statistical interpretation depends on estimates, uncertainty and transparent analysis methods, not status alone. When these are absent, conclusions should remain bounded by the information actually recorded.

14. 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 important limitation for interpreting this trial is that the registry record does not report primary outcome results. It gives no group means, changes from baseline, between-group contrast, standard errors, confidence intervals or test results for the HRCT endpoint. The direction, size and precision of the primary effect therefore cannot be assessed.

The endpoint description identifies change at 2 and 6 months but does not specify a single reported comparison. It covers multiple lesion types combined into composite scores. The record does not report the analysis plan for these time points and components, or any adjustment for multiplicity. It also does not provide component-level results that could show whether the composite reflected consistent changes across lesion types.

Although the registry describes blinded, independent image readings and adjudication of widely discrepant scores, it does not report reader agreement, adjudication frequency or measurement reliability. The analysis model and its assumptions are also not reported. These omissions limit assessment of the endpoint’s measurement and statistical properties.

The record does not report analyzed sample sizes for the primary endpoint, follow-up completeness by group, missing-data handling, the analysis population or exclusions. It also does not report the allocation procedure, adherence, complete background treatment details or any interim analyses. These details are not reported in the registry record, so their effects on interpretation cannot be evaluated.

Safety reporting is limited to group-level counts, selected serious-event terms and percentages. Event counts for each term, timing, detailed severity information, statistical comparisons and deaths are not reported in the registry record. The study’s enrollment and completed status do not resolve these reporting limitations.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Metronidazole for Pulmonary Tuberculosis (South Korea) — NCT00425113.
• Registry record last updated 2013-07-10; status last verified 2013-05.
• Results first posted 2013-07-10.
• 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.

18. Record Summary

NCT00425113 is a completed, randomized, double-blind, parallel-group Phase II study sponsored by the National Institute of Allergy and Infectious Diseases (NIAID). Its posted enrollment is 35. The registered comparison was metronidazole added to background tuberculosis treatment versus placebo added to background treatment during the initial 2 months, for multi-drug resistant pulmonary tuberculosis.

The primary outcome concerned changes in TB lesion sizes measured with HRCT, using composite reader scores. The registry describes change from baseline at 2 and 6 months and identifies a 6-month timeframe. The secondary outcome listed is time to sputum culture conversion to negative on solid medium.

The registry record reports serious adverse events affecting 9 of 15 at risk in the metronidazole group and 4 of 16 at risk in the placebo group, and other adverse events affecting 11 of 15 and 8 of 16, respectively. Primary and secondary outcome analyses and results are not reported in the registry record.