Evaluation of Coronary Artery Calcification Using Gated Stationary Chest Tomosynthesis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluation of Coronary Artery Calcification Using Gated Stationary Chest Tomosynthesis (NCT03169062), a na na study. The primary outcome is Comparison of CT Derived CACS and Tomosynthesis Scores Correlation.

2. Clinical Question

The study concerned individuals with coronary artery disease and evaluated gated stationary chest tomosynthesis as a method for assessing coronary artery calcium. The registry record describes the intended primary question as whether scores from tomosynthesis correspond to coronary artery calcium scores derived from CT.

This question is about measurement agreement or association, rather than a comparison of clinical outcomes between assigned treatments. A relationship between two measurements could help characterize how a new imaging approach corresponds to an established scoring approach, but the registry record contains no observed correlation estimate or agreement results. It therefore does not answer whether the two scoring methods were related in this study.

3. Trial Design

The registry record describes a completed, single-group study of gated stationary chest tomosynthesis. The allocation is recorded as not applicable, and masking is recorded as none. The study enrolled 10 participants. The listed condition is coronary artery disease, and the named intervention is a device: gated stationary chest tomosynthesis. The arm description identifies all patients as receiving that imaging approach.

The study began on 2017-06-26 and reached primary completion on 2019-04-10. Its status is completed. The listed phase is not applicable. These administrative details describe the registration and conduct timeline; they do not establish that the planned endpoint was successfully measured. In this record, the primary analysis could not be conducted because calcium scoring was not performed.

The planned primary endpoint concerned the relationship between CT-derived coronary artery calcium scores and tomosynthesis scores. Its listed timeframe is 1 year. The registry description says that linear regression and Bland-Altman analysis would be used. However, because visualization of coronary artery calcium was limited by the small effective angular span of the current hardware, scoring was not performed. As a result, there are no posted primary-endpoint estimates to interpret.

This is not a randomized comparison, and there is no like-for-like control arm. CT is described as the reference measurement for the intended score comparison, not as a separately randomized study group. The record does not report a protocol rationale for the enrollment target, a sample-size justification, or additional details about participant selection, imaging procedures, or the conduct of follow-up. Such information 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

Analysis population for the primary outcome, as defined in the registry: Visualization of coronary artery calcium was limited given the small effective angular span of the current hardware, thus calcium scoring was not performed. Since calcium scoring could not be performed, the planned analysis could not be conducted.

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 Comparison of CT Derived CACS and Tomosynthesis Scores Correlation, measured over 1 year.

The primary endpoint was intended to assess the relationship between CT-derived coronary artery calcium scores and scores from tomosynthesis. The registry lists a timeframe of 1 year and describes a plan to use linear regression and Bland-Altman analysis. These approaches address related but distinct questions: regression can characterize how two measurements vary together, while Bland-Altman analysis is used to examine the differences between paired measurement methods and assess their agreement.

In general, comparing a new measurement method with a reference method can help determine whether the new method produces values that track with, or agree sufficiently with, the reference. Association alone does not establish agreement: two methods may vary together while yielding systematically different values. Agreement analysis therefore considers the pattern and spread of paired differences, rather than relying only on a correlation or regression relationship.

In this study, the planned endpoint was not measured. The registry says visualization of coronary artery calcium was limited by the small effective angular span of the current hardware, calcium scoring was not performed, and the planned analysis could not be conducted. No endpoint value, regression estimate, correlation estimate, or Bland-Altman result is reported in the registry record. Thus, the endpoint describes the intended measurement question, not an observed result.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The record names linear regression and Bland-Altman analysis as the planned methods for comparing CT-derived coronary artery calcium scores with tomosynthesis scores. Neither method produced a reported result because the required tomosynthesis calcium scoring was not performed. The explanations below describe these methods in general; they are not findings from this study.

What does regression contribute?

In general, linear regression describes how an outcome measurement changes in relation to a predictor measurement. In a paired-method setting, it can summarize whether higher values from one method tend to accompany higher values from another, and it can describe the form of that relationship. The fitted relationship depends on the observed paired measurements and the assumptions used for the model. The registry does not report a fitted slope, intercept, uncertainty interval, or diagnostic assessment for this study.

A regression relationship should not automatically be read as proof that two methods agree. For example, a consistent offset or a different scale between methods could coexist with a strong tendency for their measurements to move together. Agreement is a distinct measurement property. For this reason, a regression analysis and a method-agreement analysis can complement one another rather than answer identical questions.

Why use Bland-Altman analysis?

In general, Bland-Altman analysis uses paired measurements to examine their differences. It can describe the average difference between methods and the spread of individual differences around that average. This helps assess whether discrepancies are small or large in the context of the measurement task and whether the discrepancy changes across the measurement range. The interpretation requires the paired data and an appropriate understanding of what differences would matter; those results and decision criteria are not reported in the registry record.

Agreement is not established solely by showing that two measures are correlated. A correlation can be high when the values from one method are systematically higher or lower than those from another. Conversely, the pattern of paired differences may be the more direct evidence for whether two methods give sufficiently similar values. No correlation coefficient, average difference, spread of differences, or limits of agreement are reported for the planned primary comparison.

What does a confidence interval mean?

In general, a confidence interval expresses statistical uncertainty around an estimated quantity under a specified analysis procedure. A narrower interval indicates greater precision under that procedure, while a wider interval indicates less precision. It does not correct for a measurement process that failed to produce the endpoint, and it cannot be constructed from an absent estimate without underlying data. The registry record contains no confidence interval for the primary comparison.

Likewise, a p-value would address a specific statistical hypothesis under a specified model; it would not by itself quantify the agreement between measurement methods or establish practical acceptability. No primary analysis or p-value is reported. Since no paired scores were analyzed, there is no reported inferential result to distinguish from a descriptive finding.

Why do measurement feasibility and analysis matter?

Statistical methods depend on the measurements needed to answer the research question. Here, the stated limitation arose before the planned comparison: visualization was limited by the small effective angular span of the current hardware, and calcium scoring was not performed. This is different from a completed analysis that found little association or poor agreement. A missing analysis does not demonstrate that the methods disagree; it means the intended comparison is unavailable in the posted results.

The registry does not report whether alternative scoring approaches were considered, whether technical changes were made, or whether other analytic methods were used. Such details are not reported in the registry record. Nor does the record provide distributions of attempted or obtained measurements that would permit assessment of the technical data independently of the primary analysis.

The distinction between planned and observed analysis is important. The registry states that linear regression and Bland-Altman analysis would be used, while the stated technical limitation prevented scoring and the planned analysis. Therefore, those methods should be described as planned, not as completed analyses. The interpretation should not attribute a numerical result to them or infer method performance from the study's enrollment or completion status.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary comparison of CT-derived coronary artery calcium scores and tomosynthesis scores, along with a secondary outcome titled mean correlation coefficient of gating. It also records 2 outcome measures and 0 analyses. No results for either outcome are reported. The registry record does not describe a multiplicity adjustment, a hierarchy among secondary tests, or how the interpretation would account for several statistical comparisons. These details are not reported in the registry record.

In general, multiplicity arises when several statistical hypotheses are tested. If each is assessed without accounting for the others, the chance of at least one apparently notable result can increase. Approaches to address this include defining a primary hypothesis in advance, specifying an order for testing, or adjusting inference across a family of tests. The appropriate approach depends on the aims and analysis plan. No such plan is reported, and no completed analyses are available to evaluate.

The record does not report how missing or unobtainable measurements were handled, whether any participants lacked usable data, or what analysis population was intended for the primary endpoint. It also does not describe whether the secondary outcome was collected. These matters are not reported in the registry record. The stated reason the planned primary analysis could not be conducted is that calcium scoring was not performed because visualization was limited by the current hardware's small effective angular span. This is a measurement-feasibility limitation, not a reported statistical procedure for handling missing values.

Without a reported analysis plan and outcome data, it is not possible to assess the effects of missing measurements, define the analysis population used, or judge how multiplicity might affect the interpretation of any estimates. No estimates are available to which those considerations could be applied.

11. 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.

The registry reports adverse-event counts for the all-patients group. For serious adverse events, 10 participants were at risk and 0 were affected. For other adverse events, 10 were at risk and 0 were affected. The death count is 0. The registry lists no serious adverse-event terms.

These are counts recorded for this enrolled group, not estimates that establish the likelihood of an event in a broader population. In general, observing no events in a study does not demonstrate that the underlying risk is zero; the precision of a safety estimate depends on the amount and quality of follow-up and the number of participants observed. The registry record does not report additional detail about adverse-event collection, severity categories beyond the listed serious and other groupings, or follow-up completeness.

The figures should also be kept separate from the primary imaging question. They do not indicate whether tomosynthesis scores correspond to CT-derived scores, because the planned calcium scoring and comparison were not conducted. The safety counts are the posted event data, while the primary measurement endpoint has no reported analysis.

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

The record documents an attempt to evaluate gated stationary chest tomosynthesis as a way to assess coronary artery calcium in individuals with coronary artery disease. The intended primary question was whether tomosynthesis scores related to CT-derived scores, with linear regression and Bland-Altman analysis planned. The study enrolled 10 participants and is marked completed.

The key limitation is that the intended score comparison was not obtained. The registry states that visualization was limited by the small effective angular span of the current hardware, calcium scoring was not performed, and the planned analysis could not be conducted. There is consequently no reported estimate of association, agreement, or uncertainty for the primary endpoint.

This means the posted results cannot establish that the two scoring methods agree, that they differ, or that one method can substitute for the other. It would also be incorrect to treat the absence of an analysis as evidence that a statistical test found no relationship. The data needed to answer that question were not analyzed according to the registry description.

The record does document the technical obstacle encountered for the intended assessment. That information is relevant to understanding why the planned measurement comparison is absent, but it does not quantify performance under other hardware configurations or imaging conditions. Such results are not reported in the registry record.

The adverse-event counts show 0 serious events, 0 other adverse events, and 0 deaths among the listed at-risk group of 10. These counts describe the registry's safety entries and should not be generalized beyond the study. The record also lists a secondary gating outcome without a reported result. Taken together, the record supports a limited conclusion about the status of the planned measurements and analyses, not a conclusion about clinical benefit or comparative diagnostic performance.

13. Why This Trial Matters Statistically

This study illustrates an important distinction in clinical measurement research: a statistical analysis can answer a question only if the measurements needed for that analysis can be obtained. The registry's proposed comparison required both CT-derived calcium scores and tomosynthesis scores. Because calcium scoring was not performed under the stated visualization conditions, the planned comparison was unavailable.

That distinction prevents a common interpretive error. An absent estimate is not a null finding. There is no posted result indicating that the methods had no association, and there is no reported evidence that they agreed. The record instead explains that the analysis could not be conducted. Reporting this distinction clearly is essential when reading completed studies whose registered endpoint was not ultimately analyzed.

The planned use of both regression and Bland-Altman analysis also highlights that association and agreement are not interchangeable statistical concepts. Regression can describe how measurements vary together, whereas paired-difference methods address how closely their values match. A measurement study may need both perspectives, depending on its purpose. Neither perspective is available without analyzable paired scores.

The record also demonstrates why technical feasibility belongs in the interpretation of a statistical study. A device's ability to produce a usable measurement is upstream of model choice and inferential precision. Sophisticated analysis cannot compensate for an endpoint that could not be generated. Here, the stated limitation concerns visualization with the current hardware; the registry does not report an alternative analysis that resolves it.

Finally, enrollment, completion status, and safety counts should not be mistaken for evidence about the primary measurement question. Those details describe the study record but do not create an estimate of score correspondence. Keeping these categories separate makes the statistical interpretation more accurate and transparent.

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 primary limitation is that the intended calcium scores were not available for analysis. The registry attributes this to limited visualization of coronary artery calcium given the small effective angular span of the current hardware. Calcium scoring was therefore not performed, and the planned comparison with CT-derived scores could not be conducted. This prevents interpretation of the primary endpoint as a measured relationship or agreement result.

The enrollment was 10, and the study was single-group rather than a randomized comparison. There is no control arm that would support a like-for-like comparison between assigned groups. The intended CT comparison was a measurement reference, not a randomized treatment allocation. The record does not report a sample-size justification or a rationale for the enrollment target.

The record does not report participant-level score data, an estimate for the primary endpoint, a confidence interval, or the results of the named analyses. It also does not provide results for the listed secondary outcome. Details about data completeness, analysis populations, missing-data handling, and the statistical plan are not reported in the registry record.

Further technical details that could clarify the scope of the limitation, such as whether different acquisition or scoring procedures were evaluated, are not reported in the registry record. The record's stated finding should therefore be read narrowly: scoring was not performed under the described conditions. It does not establish how the method would perform under other conditions.

The adverse-event entries report no affected participants among the listed group of 10, but the record does not report additional information about event surveillance or follow-up completeness. The safety counts should be interpreted as registry counts, not as a precise estimate of risk beyond this study.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Evaluation of Coronary Artery Calcification Using Gated Stationary Chest Tomosynthesis — NCT03169062.
• Registry record last updated 2020-02-11; status last verified 2020-01.
• Results first posted 2020-02-11.
• 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

This completed, single-group device study evaluated gated stationary chest tomosynthesis in individuals with coronary artery disease. The registry records 10 enrolled participants, no masking, and allocation as not applicable. The study began on 2017-06-26, with primary completion on 2019-04-10.

The intended primary endpoint was the relationship between CT-derived coronary artery calcium scores and tomosynthesis scores, with a listed timeframe of 1 year. Linear regression and Bland-Altman analysis were planned. The registry states that limited visualization related to the small effective angular span of the current hardware prevented calcium scoring, so the planned analysis could not be conducted. No primary estimate is reported.

The record lists mean correlation coefficient of gating as a secondary outcome but reports no result. It records 0 serious adverse events, 0 other adverse events, and 0 deaths among 10 at risk. The available record therefore describes the study objective, technical limitation, and posted safety counts, but does not provide an analyzed comparison of the imaging scores.