Effect of Steroid Injections in a Knee With Osteoarthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Steroid Injections in a Knee With Osteoarthritis (NCT01230424), a randomized phase 4 study. The primary outcome is Change in Mean Cartilage Thickness in the Index Compartment (Compartment With the Most Damage).

2. Clinical Question

3. Trial Design

This was a randomized, double-masked, parallel-group Phase 4 trial of intra-articular steroid injections for osteoarthritis of the knee. The registry lists an enrollment of 140 and reports 70 participants in each primary-outcome group. The sponsor was Tufts Medical Center. The recorded start was 2011-03, primary completion was 2015-01, and trial status is completed.

The intervention was triamcinolone acetonide, with 40 mg injected into the study knee every 12 weeks for a total of 8 injections. The comparator was 0.9% sodium chloride injection as placebo, administered on the same schedule. The brief summary describes assessment of injection tolerability over time and safety through reported adverse effects, knee examinations, and clinical laboratory tests. It also mentions participant questionnaires, X-ray, MRI, and bone density testing.

The main structural outcome was change in mean cartilage thickness in the index compartment, measured on knee MRI from baseline to 2 years. The registry identifies the index compartment as the compartment with the most damage. Other listed outcomes address MRI measures, questionnaires, patient global assessment, and walking time. The exact allocation procedure, masking procedures, eligibility criteria, and details of follow-up are not reported in the registry record.

Randomization is intended to make treatment groups comparable on average at assignment, while masking can reduce the influence of treatment knowledge on participant behavior and outcome assessment. These are general design principles; the registry record identifies the trial as randomized and double-masked but does not describe how those procedures were implemented.

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: All individuals were included in analysis, which was performed on multiply imputed data.

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 Change in Mean Cartilage Thickness in the Index Compartment (Compartment With the Most Damage), measured over Baseline to 2 years.

The primary endpoint is change in mean cartilage thickness in the index compartment of the study knee, measured by MRI from baseline to 2 years. The index compartment is defined in the registry as the compartment with the most damage. Cartilage thickness is a continuous structural measure, reported in millimeters, and the registry gives a mean change for each randomized group with a 95% confidence interval.

The mean summarizes the average observed change across the analyzed participants. A negative mean change indicates that the measured thickness was lower at follow-up than at baseline, on average. This endpoint focuses on structural change in the knee rather than a reported symptom or functional score. MRI provides the measurement method specified in the registry; the scanner is identified as a Philips Achieva X-Series 3.0 Tesla scanner.

Structural measures can be useful in a trial when the research question concerns changes in joint tissue. However, a group mean alone does not show how individual participants changed, and a within-group change does not establish how the groups differed from each other. To assess the randomized treatment contrast, the analysis needs an estimate comparing the groups, with its uncertainty. The registry reports group-level means and confidence intervals but does not provide an estimable two-arm comparison for this continuous endpoint. The clinical interpretation of the endpoint is therefore limited to the posted group estimates and their precision.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous: cartilage thickness is measured in millimeters, and change is summarized by a mean. A mean is the arithmetic average of measurements within an analysis group. In general, it offers a concise description of the group’s average change but does not describe the full distribution of individual changes. The registry reports a mean for each group and a 95% confidence interval; a separate measure of spread such as a standard deviation is not reported in the primary results.

What does a confidence interval express?

General statistical explanation: a confidence interval describes uncertainty around an estimate under a specified analysis and its assumptions. A narrower interval indicates greater precision for that estimate, while a wider interval indicates greater uncertainty. The 95% confidence intervals in this record are attached to the group-specific means. They should not be read as intervals for the difference between groups. The registry does not report a confidence interval for the randomized treatment contrast.

The triamcinolone group’s mean change is -0.21 mm, with a 95% confidence interval from -0.29 to -0.14 mm. The sodium chloride group’s mean change is -0.1 mm, with a 95% confidence interval from -0.16 to -0.04 mm. The estimates and intervals describe change from baseline to 2 years within each group. They do not tell how much more or less change occurred under one assignment than the other, after accounting for uncertainty.

Why is the comparison the key randomized estimate?

General statistical explanation: in a randomized trial, the main comparative question concerns the outcome difference between the assigned groups. A within-group change can reflect treatment, natural course, measurement variation, or other influences over time. Randomization supports a comparison between groups, but the comparison still needs to be estimated and reported. Here, the registry marks the two-arm comparison as having values missing or not estimable for a continuous endpoint. The group means cannot substitute for that missing comparative estimate. No p-value or confidence interval for the between-group contrast is reported in the registry record.

The primary analysis population is described as all individuals included in analysis, performed on multiply imputed data. Multiple imputation is a general method for addressing missing measurements by creating completed versions of a dataset, analyzing each version, and combining results to reflect uncertainty due to missing values. The record does not report the imputation model, the variables used, the number of imputations, or the assumptions underlying the procedure. It also does not report the amount of missing primary-outcome data or sensitivity analyses that test alternative assumptions.

What does the design add?

General statistical explanation: random assignment is designed to reduce systematic differences between groups at the start of a trial. Double masking is intended to limit the influence of knowledge about assignment on care, participant reporting, and outcome assessment. A parallel-group design assigns participants to a group and follows those groups over the study period. The registry identifies these design features but does not give operational details about allocation concealment, who was masked, or how masking was maintained.

For structural outcomes measured repeatedly, the selected analysis can influence how baseline and follow-up measurements are used. The registry identifies change from baseline to 2 years and says missing data were imputed, but it does not report whether the analysis adjusted for baseline cartilage thickness or used any other covariates. Details of the statistical model, assumptions, and handling of MRI measurements are not reported in the registry record. As a result, the posted means and intervals cannot be used to reconstruct a fully specified analysis.

How should planning calculations be interpreted?

General statistical explanation: a sample-size calculation is a planning tool based on assumptions about the effect to detect, variability, allocation, and statistical error rates. It is not a measure of the probability that a completed trial found a true effect. The registry record’s planning assumptions describe a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect expressed as a stated fraction of the observed effect. Any sample-size figures based on those assumptions are planning illustrations, not observed power and not the trial’s original design justification. The trial’s own sample-size justification and statistical plan are not reported in the registry record.

Finally, no multiplicity adjustment is described. General statistical explanation: when a study assesses multiple outcomes or multiple comparisons, the chance of at least one false-positive result can rise if each is assessed without adjustment. The record lists multiple outcomes, but does not report a multiplicity strategy or enough comparative analyses to evaluate how it was handled. Any interpretation of secondary findings would need to account for that uncertainty, as well as the absence of their posted results in the record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 11 outcome measures and 9 posted secondary outcomes. The primary outcome concerns change in mean cartilage thickness in the index compartment. The listed secondary outcomes include other MRI measures, WOMAC function and stiffness scores, patient global assessment, and time to complete a twenty-meter walk. The reason the counts of outcome measures and posted secondary outcomes differ is not reported in the registry record.

Multiplicity arises when a study examines several outcomes or comparisons. General statistical explanation: the more statistical tests considered, the greater the opportunity for a result to appear notable by chance if each test is interpreted independently. A prespecified strategy can address this, for example by controlling the overall error rate or identifying a primary endpoint as the main confirmatory question. Whether such a strategy was used, and whether secondary outcomes were adjusted or treated as exploratory, is not reported in the registry record. The statistical analysis plan is also not reported.

For the primary endpoint, the registry says missing data were imputed and that all individuals were included in analysis of multiply imputed data. The imputation method, missing-data quantities, and assumptions are not reported. Sensitivity analyses assessing the consequences of those assumptions are not reported in the registry record. The group estimates therefore need to be understood in light of incomplete information about missingness and imputation.

The record does not provide an estimable two-arm comparison for the continuous primary endpoint. Thus, multiplicity is not the only limitation: even the main randomized contrast is unavailable as a reported estimate. No inferential results for the posted secondary outcomes are included in the record.

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: asthma excerbation (2.9%), fall (2.9%), cellulitis (2.9%), liver and bone cancer (1.4%), hypertension (1.4%).

The registry lists serious adverse events affecting 15 participants in the triamcinolone group and 27 in the sodium chloride group. For other adverse events, it lists 48 in the triamcinolone group and 60 in the sodium chloride group. Each group has 70 participants listed at risk for the serious and other adverse-event summaries. These are descriptive counts as recorded; the registry figures do not by themselves establish that an event was caused by an injection or by the assigned treatment.

The listed serious-event terms include Asthma excerbation, Fall, and Cellulitis, each reported as 2.9%, and Liver and bone cancer, Hypertension, and Allergic reaction, each reported as 1.4%. The registry does not provide enough detail in these figures to assess event timing, recurrence, severity beyond the serious-event classification, or the basis for attributing an event to treatment. Deaths are not reported in the registry record.

Safety counts should be interpreted with the outcome definitions and follow-up context in mind. The brief summary says that reported adverse effects, knee examinations, and clinical laboratory tests were collected, but the registry figures do not provide event-level narratives or adjusted comparisons. The difference in counts should not be converted into a causal conclusion from these summaries alone.

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

The primary results describe average MRI-measured cartilage-thickness change within each randomized group over the period from baseline to 2 years. The mean change was -0.21 mm for triamcinolone and -0.1 mm for sodium chloride, with the respective 95% confidence intervals reported in the record. These values show that the group averages decreased over that period. The intervals convey uncertainty around each group mean, not around the treatment contrast.

The results do not quantify whether the average change differed between the assigned groups. The registry does not report an estimable two-arm comparison for the continuous primary endpoint, so the magnitude and precision of the comparative effect cannot be read from the posted results. Nor do the group-specific estimates establish that an observed change was caused by the injection assignment. No claim about treatment benefit or lack of benefit follows from these summaries alone.

The endpoint addresses a structural MRI measure in the compartment with the most damage. It is not interchangeable with pain, stiffness, physical function, or a participant’s overall assessment. Although the registry lists secondary outcomes covering some of those domains, their results are not reported in the primary results. The record also does not report the original statistical plan or the reasoning behind the trial’s sample-size design.

Missing data were imputed, but the details and assumptions of that procedure are not reported. This makes it difficult to assess how robust the group estimates are to alternative approaches for incomplete MRI measurements. The registry’s safety counts are likewise descriptive and do not prove that an event was caused by treatment. Overall, the record supports a careful description of the observed group-level estimates and their uncertainty, but not a definitive comparative or causal conclusion.

13. Why This Trial Matters Statistically

This trial illustrates an important distinction in randomized-trial reporting: a mean change in each group is not the same quantity as the treatment contrast. The primary results give group-level MRI estimates, but do not give an estimable comparison between the randomized assignments. Readers therefore cannot use the reported means alone to determine the size or precision of the difference between groups. This distinction is central when interpreting any trial whose endpoint is measured before and after treatment.

The example also shows why the analysis population and missing-data approach matter. The record says that the primary analysis used multiply imputed data and included all individuals in the analysis. Multiple imputation can incorporate uncertainty due to missing measurements, but its credibility depends on modeling choices and assumptions. Those details are not reported, so readers cannot fully assess their influence on the estimates or compare the primary result with alternative missing-data analyses.

A structural endpoint measured by MRI is also distinct from outcomes that describe symptoms or daily function. The registry lists several outcome domains, making it important to know which one was primary and how multiple outcomes were handled. The record identifies the primary outcome but does not describe a multiplicity strategy or provide the secondary results in the material. That limits assessment of the broader outcome pattern.

Finally, this record is a reminder that completed status and detailed group estimates do not replace a complete statistical report. A well-interpreted result depends on the comparative estimate, its uncertainty, transparent analysis methods, and clear handling of missing data and multiple outcomes. The available record permits description of the primary group means, but leaves these central inferential details unresolved.

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 main limitation for statistical interpretation is that the registry does not report an estimable two-arm comparison for the continuous primary endpoint. The group-specific mean changes and their confidence intervals cannot quantify the randomized treatment contrast. No comparative p-value or confidence interval is reported. This prevents a direct assessment of the magnitude and precision of the primary between-group result.

Although the primary analysis is described as using multiply imputed data, the imputation model, assumptions, number of imputations, and sensitivity analyses are not reported in the registry record. The amount and pattern of missing MRI data are also not reported. These omissions limit evaluation of whether the results could be sensitive to the approach used for missing measurements.

Details of the primary statistical model, adjustment for baseline values or other covariates, and the statistical analysis plan are not reported. The record also does not explain its sample-size justification. The planning assumptions associated with illustrative sample-size calculations do not establish the design assumptions used for the original trial.

The registry lists multiple outcomes but does not report a strategy for handling multiplicity. It does not provide the results of the listed secondary outcomes in the primary results. Further, the clinical meaning of a change in this MRI measure cannot be determined from the group estimates alone, and the structural endpoint is not a direct measure of symptoms or function.

Safety summaries are presented as counts and selected serious-event terms, without enough detail to assess timing, recurrence, or causation. Eligibility criteria, allocation procedures, masking implementation, and more detailed follow-up information are not reported in the registry record. These limitations call for a narrow interpretation of the posted findings.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Effect of Steroid Injections in a Knee With Osteoarthritis — NCT01230424.
• Registry record last updated 2017-07-31; status last verified 2017-06.
• Results first posted 2017-07-31.
• 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.

18. Record Summary

NCT01230424 is a completed randomized, double-masked, parallel-group Phase 4 trial sponsored by Tufts Medical Center. It enrolled 140 participants with knee osteoarthritis. The recorded start was 2011-03 and primary completion was 2015-01.

Participants received triamcinolone acetonide 40 mg or 0.9% sodium chloride placebo into the study knee every 12 weeks for a total of 8 injections. The primary endpoint was change in mean cartilage thickness in the index compartment on MRI from baseline to 2 years.

The registry reports group-specific mean changes and 95% confidence intervals, with missing data imputed, but does not report an estimable two-arm comparison for the primary continuous endpoint. Details of the statistical plan, imputation procedure, multiplicity handling, and trial sample-size justification are not reported in the registry record.