Non-Contrast 4DCT to Detect Pulmonary Thromboembolic Events: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Non-Contrast 4DCT to Detect Pulmonary Thromboembolic Events (NCT03183063), a non-randomized na study. The primary outcome is Correlation of 4DCT Identified Perfusion With SPECT/CT Identified Perfusion.

2. Clinical Question

The study addresses a diagnostic-imaging question rather than a treatment comparison. Its central idea is that changes in lung blood mass during breathing may reveal regions with reduced perfusion without requiring intravenous contrast. The primary outcome description specifies identifying regions of interest on SPECT perfusion and regions of respiratory-induced blood mass change, then assessing their spatial overlap and reporting a Spearman correlation.

The registry describes three cohorts: participants with CTA-confirmed segmental or lobar pulmonary emboli who were to receive 4DCT and SPECT/CT; a planned group receiving 4DCT with BiPAP and SPECT/CT; and participants with CTA ordered or performed for suspected PE who were to receive 4DCT. The primary-outcome population was narrower than the overall study population: Cohort 2 participants were withdrawn, and Cohort 3 participants were not analyzed for this outcome. The outcome therefore does not directly summarize all enrolled participants or all three cohorts.

The record does not report a numerical correlation estimate, a confidence interval, or a test result for the primary question. It also does not report diagnostic accuracy measures such as sensitivity or specificity. The registered question concerns correspondence between image-derived perfusion regions in the analyzed subgroup; it should not be read as a complete assessment of 4DCT performance in every suspected-PE participant.

3. Trial Design

This was a completed, non-randomized, unmasked study with a sequential model. The registry lists phase as not applicable and enrollment as 139. The study title and description identify non-contrast 4DCT as the imaging approach under evaluation for pulmonary thromboembolic events. No randomized allocation or like-for-like control comparison is described.

The registry lists three intervention groups. Cohort 1, “4DCT and SPECT/CT,” anticipated 15 participants with segmental or lobar PE on CTA. They were to receive SPECT/CT and 4DCT on the same day, with both 4DCT scans obtained during normal breathing. Cohort 2, “4DCT with BiPAP and SPECT/CT,” anticipated 5 participants with segmental or lobar PE on CTA. The second of two 4DCT scans was to be obtained with positive airway breathing via BiPAP. Cohort 3, “4DCT with CTA in suspected PE,” anticipated 124 participants who had CTA ordered or performed for suspected PE; the analysis goal was 62 with positive CTA results and 62 with negative CTA results.

The primary outcome was not analyzed across all these groups. Its population description says that correlation coefficients were calculated only using the 15 subjects with documented PE in Cohort 1. All subjects in Cohort 2 were withdrawn, and Cohort 3 was not analyzed for this outcome. The registry reports study start on 2017-04-12, primary completion on 2019-04-23, and status as completed.

Because the study was non-randomized and its arms had different purposes and eligibility descriptions, they should not be treated as randomized treatment groups or as equivalent comparison arms. The registry record does not report further details of the sequence, a formal control-group plan, or a sample-size justification.

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: Correlation coefficients were only calculated using the 15 subjects with documented PE in Cohort 1 (4DCT and SPECT/CT). All subjects in Cohort 2 (4DCT with BiPAP and SPECT/CT were withdrawn. Subjects in Cohort 3 (4DCT with CTA in suspected PE) were not analyzed for this outcome.

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: Baseline data included for 15 subjects in Cohort 1, following removal of screen failures and withdrawn subjects. All patients withdrawn from study in Cohort 2, 0 subjects analyzed. 115 subjects included in baseline characteristics for Cohort 3 includes 4 later withdrawn, total 130 subjects.

6. Primary Endpoint

The registry lists the primary outcome as Correlation of 4DCT Identified Perfusion With SPECT/CT Identified Perfusion, measured over 1 hour.

The primary outcome is titled “Correlation of 4DCT Identified Perfusion With SPECT/CT Identified Perfusion.” The record describes a custom automated PE detection algorithm that delineates hypo-perfused regions of interest on SPECT perfusion and regions of respiratory-induced blood mass change, or RIBMC, using 4DCT. A Dice similarity coefficient was to assess spatial overlap between the two sets of regions, with Spearman correlation reported. The outcome unit is Spearman correlation, and the record specifies a median with full range as its summary format. The stated timeframe is 1 hour.

In general, correlation describes how two measured quantities vary together. Spearman correlation is a rank-based measure of association; it does not by itself establish that two methods agree in their absolute measurements. Spatial overlap is a related but distinct aspect of image correspondence, which is why the description also identifies the Dice similarity coefficient. Neither measure alone establishes the accuracy of a diagnostic method against every possible clinical reference.

The registry record does not report the observed median, full range, or a confidence interval for the primary outcome. It also does not report the Dice coefficient value. Thus, the endpoint definition is available, but the magnitude and precision of the observed correspondence cannot be assessed from the registered result information.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary analysis is described using Spearman correlation, with a median and full range as the summary. The method choice and what it can show are distinct from the missing numerical result: the registry record identifies the planned measure, but does not report its estimate.

What does Spearman correlation describe?

General explanation: Spearman correlation assesses whether observations tend to move together in rank order. It can be useful when the relationship of interest is monotonic but may not be well described by a straight-line relationship, or when a rank-based measure is preferred. The sign indicates the direction of association, and the magnitude indicates how consistently the rankings correspond. These interpretations require the actual coefficient; no coefficient is reported in the registry record.

Correlation is not a direct test of equivalence between imaging methods. For example, a strong association does not ensure that two methods assign the same values or identify precisely the same regions. A correlation also does not, on its own, establish diagnostic accuracy relative to a clinical standard. The primary-outcome description therefore includes a separate spatial-overlap measure, the Dice similarity coefficient, alongside the correlation.

Why assess spatial overlap?

General explanation: a spatial-overlap statistic compares the locations of regions identified by two methods. For image analysis, this can address whether the marked regions occupy similar areas. The registry states that a custom automated algorithm would delineate hypo-perfused regions on SPECT perfusion and RIBMC regions using 4DCT, and that overlap would be assessed with the Dice similarity coefficient. The numerical Dice result is not reported in the registry record. The record’s description also says Spearman correlation would be reported; it does not provide enough result detail to establish how the correlation and overlap values relate to one another in the final analysis.

How should the median and full range be read?

General explanation: the median is the middle value in an ordered set of observations, while the full range gives the minimum and maximum observed values. A range conveys the spread of the observed values but is sensitive to extreme observations and does not, by itself, quantify uncertainty about the median. A confidence interval would address precision in a different way. The record lists median and full range as the primary outcome’s summary format, but neither the values nor a confidence interval are reported.

What is the relevance of the analysis population?

The outcome population is defined as 15 subjects with documented PE in Cohort 1. Cohort 2 participants were withdrawn, and Cohort 3 was not analyzed for this outcome. The endpoint therefore pertains to the stated subgroup, rather than to every participant enrolled in the study. General explanation: restricting an analysis to a subgroup can make the result more directly relevant to that subgroup, but it limits how readily findings can be generalized to populations not included in the analysis. The registry does not provide a broader primary-outcome analysis.

What does the study structure imply?

The registry classifies allocation as non-randomized, masking as none, and the model as sequential. General explanation: random allocation can help balance measured and unmeasured characteristics between comparison groups, while masking can reduce some forms of assessment or treatment-related bias. Neither feature is present according to the registry classifications. That does not by itself invalidate an imaging comparison, but it means the design should not be interpreted as a randomized test of competing strategies. The cohorts also had different planned imaging procedures, and the registry explicitly notes that the arms are not a like-for-like control comparison.

The record does not describe a formal hypothesis test for the primary outcome, the confidence-interval method, adjustments for covariates, handling of repeated observations, or a prespecified sample-size justification. Those details are not reported in the registry record. General statistical explanation: a confidence interval communicates the range of effect estimates compatible with the data under the chosen model and assumptions; it is useful for judging precision, not only whether an estimate crosses a significance threshold. No interval is available for this primary outcome.

The registry includes a sample-size planning assumption stating a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the detectable effect expressed as a fraction of the observed effect. It also distinguishes total events for time-to-event illustrations from participants per arm for other illustrations. These are planning assumptions, not a reported design justification or an observed power calculation. The numerical planning outputs and the observed effect used for such illustrations are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and two secondary outcomes concerning re-imaging variance in radiographic tidal volume and parenchymal lung mass for RIBMC images. It does not report a multiplicity adjustment, a hierarchy for interpreting outcomes, or a designated family of hypotheses. Whether or how multiplicity was addressed is not reported in the registry record.

General explanation: when several outcomes or analyses are examined, the chance of at least one apparently notable result can increase if each is assessed separately. Prespecified outcome ordering or an adjustment procedure can help define how evidence should be interpreted. The registry’s outcome descriptions alone do not establish whether such procedures were used.

The primary outcome population is specified: only the 15 subjects with documented PE in Cohort 1 contributed to the correlation coefficients. Cohort 2 participants were withdrawn, and Cohort 3 participants were not analyzed for this outcome. The registry does not report further details of the analysis populations for all secondary outcomes or the reasons for withdrawal from Cohort 2.

Missing-data counts, methods for handling missing observations, and sensitivity analyses are not reported in the registry record. The record also does not report an intention-to-treat or per-protocol designation for the primary analysis. Consequently, it is not possible to determine from the registry information whether missingness could have affected the primary estimate or how robust the analysis was to alternative population definitions.

The registry lists zero analyses in its analysis count. This does not provide a numerical primary-outcome result or explain the analytic procedures. The endpoint’s population statement supplies some information about who was included, but a complete account of multiplicity, missing data, and analysis populations is not reported in the registry record.

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 serious adverse events, other adverse events, and deaths by intervention group. For “4DCT and SPECT/CT,” 20 participants were listed at risk for serious and other adverse events, with 0 affected in each category and 0 deaths. For “4DCT With BiPAP and SPECT/CT,” 4 were listed at risk in each adverse-event category, with 0 affected and 0 deaths. For “4DCT With CTA in Suspected PE,” 115 were listed at risk in each category, with 0 affected and 0 deaths.

These are the counts recorded for the specified groups and categories. They should not be read as proof that no adverse events could occur, or as evidence that one imaging approach is safer than another. General explanation: a zero count in a study record means no events were recorded in the stated group and reporting category; it does not establish the absence of risk beyond those observations. The groups were not randomized, and the registry notes that they are not like-for-like control groups.

The registry lists no top serious adverse-event terms. It does not provide additional detail about event definitions, severity grading, assessment procedures, or follow-up duration in the adverse-event information. The at-risk counts also differ across groups, so raw counts should be interpreted in their stated group context rather than compared as if the groups had equal size or comparable enrollment.

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

The registry record defines a focused imaging question: whether regions of perfusion identified from non-contrast 4DCT correspond to regions identified with SPECT/CT among subjects with documented PE in Cohort 1. It also identifies the intended measurement approach, including a Dice similarity coefficient for spatial overlap and Spearman correlation. These details clarify what the primary outcome was meant to quantify.

However, the numerical primary-outcome result is not reported in the registry record. No observed correlation estimate, full range, confidence interval, or Dice coefficient is available in the stated results. It is therefore not possible to characterize the size or precision of the observed correspondence from this record. The description of an endpoint is not itself evidence of the size or direction of an observed result.

The primary analysis involved 15 subjects with documented PE in Cohort 1. It does not include Cohort 2, whose participants were withdrawn, or Cohort 3, which was not analyzed for the primary outcome. Thus, the primary endpoint as described does not directly answer how the approach performed across the full suspected-PE cohort or among participants with negative CTA results.

General statistical explanation: correlation and spatial overlap address aspects of association and image-region correspondence, but neither alone is equivalent to diagnostic accuracy or clinical utility. The registry record does not report sensitivity, specificity, predictive values, or a comparison of clinical outcomes. It also does not describe a randomized comparison of imaging strategies. No treatment-effect conclusion follows from these data.

The adverse-event entries show zero recorded serious adverse events, other adverse events, and deaths in the listed groups, alongside their respective at-risk counts. Those entries describe the recorded safety counts and do not establish that risk is absent or support a direct safety comparison between the non-randomized groups.

In short, the record provides information about the study structure, intended imaging measurements, analysis subgroup, and recorded adverse-event counts. It does not report the principal numerical result needed to assess the strength or precision of the primary imaging correspondence. The record therefore supports a description of the research question and its limits, not a conclusion about diagnostic performance.

13. Why This Trial Matters Statistically

This study is statistically notable for using an imaging-region comparison rather than a conventional treatment-versus-control outcome. Its primary question concerns whether regions identified through respiratory-induced blood mass change on non-contrast 4DCT correspond to hypo-perfused regions identified through SPECT perfusion. The registry describes spatial overlap and Spearman correlation as complementary ways to characterize that correspondence.

The distinction between association and agreement is important in imaging research. General explanation: a correlation can describe whether two measurements vary together, while a spatial-overlap measure asks whether identified regions occupy similar locations. Neither measure alone answers every diagnostic question. Clear interpretation depends on reporting the estimates, their variability, and the population from which they were obtained.

The registry makes the primary analysis population explicit: 15 subjects with documented PE in Cohort 1. It also notes that Cohort 2 participants were withdrawn and that Cohort 3 participants were not analyzed for the primary outcome. That specificity helps readers avoid treating the primary outcome as a result from all enrolled participants or as evidence across both positive and negative CTA findings.

The study’s non-randomized, unmasked structure is also relevant. Its cohorts had different planned imaging procedures and purposes, and the registry states that they are not like-for-like controls. This limits interpretation of any between-group comparison and reinforces that the principal question is an imaging correspondence question within a defined cohort.

Finally, the record illustrates the difference between a well-described outcome and a reported outcome result. It specifies a correlation measure, a spatial-overlap method, a summary format, and a timeframe, but does not report the numerical primary result or an interval. Without those values, the magnitude and precision of the observed correspondence cannot be assessed. That reporting gap is central to statistical interpretation of this study.

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 direct limitation in the registry information is that the numerical primary-outcome result is not reported. The record specifies Spearman correlation, median and full range, and a one-hour timeframe, but gives no coefficient, range, confidence interval, or Dice similarity coefficient. The primary endpoint’s magnitude and precision therefore cannot be evaluated from the registry record.

The primary analysis is limited to the 15 subjects with documented PE in Cohort 1. Cohort 2 participants were withdrawn, and Cohort 3 participants were not analyzed for this endpoint. As a result, the primary outcome does not represent all cohorts or the entire enrolled population. The registry does not report further detail on the withdrawal or the implications of the restricted analysis population.

The design was non-randomized and unmasked. The intervention groups had different imaging plans, and the record explicitly cautions that they are not a like-for-like control comparison. The available description therefore does not support treating the groups as randomized comparators or interpreting group differences as causal effects.

The record does not report diagnostic accuracy measures such as sensitivity or specificity, nor does it provide details of the primary-outcome confidence-interval method, hypothesis test, missing-data handling, or sensitivity analyses. Multiplicity procedures and a sample-size justification are also not reported in the registry record. The stated planning assumptions for sample-size illustrations do not fill those gaps or establish observed power.

The safety information is presented as counts and at-risk group sizes, with zero recorded events in the listed categories. Additional event definitions and follow-up detail are not reported in the registry record. These limitations constrain interpretation of both efficacy-related imaging results and safety; the primary numerical imaging result itself is unavailable.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Non-Contrast 4DCT to Detect Pulmonary Thromboembolic Events — NCT03183063.
• Registry record last updated 2020-06-05; status last verified 2020-05.
• Results first posted 2020-06-05.
• 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

The registry record describes a completed, non-randomized, unmasked study of non-contrast 4DCT for pulmonary thromboembolic events, with enrollment of 139 and a sequential model. It lists three cohorts: 4DCT with SPECT/CT, 4DCT with BiPAP and SPECT/CT, and 4DCT with CTA in suspected PE.

The primary outcome concerns correlation between perfusion regions identified by 4DCT and SPECT/CT, with spatial overlap assessed using the Dice similarity coefficient. Correlation coefficients were calculated only for the 15 subjects with documented PE in Cohort 1. Cohort 2 participants were withdrawn, and Cohort 3 was not analyzed for this outcome.

The registry record does not report the numerical primary-outcome estimate or its precision. It lists zero serious adverse events, other adverse events, and deaths in each intervention group. The record does not report a sample-size justification, a multiplicity procedure, or further details of missing-data handling.