This page separates reported trial results from statistical interpretation. Numerical results are limited to the ClinicalTrials.gov record. Where the registry extract provides an analysis method or p-value without a corresponding effect estimate or confidence interval, that limitation is made explicit rather than filled with an unreported value.
1. Trial at a Glance
SPYRAL HTN-OFF MED was a randomized, parallel-group, single-masked trial evaluating the Symplicity Spyral™ multi-electrode renal denervation system against a sham procedure in participants with hypertension and related vascular and cardiovascular conditions. The ClinicalTrials.gov record reports an enrollment of 366 participants and two treatment arms.
| Feature | SPYRAL HTN-OFF MED |
|---|---|
| Trial name | SPYRAL HTN-OFF MED |
| Brief title | SPYRAL PIVOTAL - SPYRAL HTN-OFF MED Study |
| Condition(s) | Hypertension; Vascular Diseases; Cardiovascular Diseases |
| Phase | NA |
| Design | Randomized, parallel, single-masked |
| Primary purpose | Treatment |
| Enrollment | 366 |
| Interventions | Symplicity Spyral™ multi-electrode renal denervation system; Sham Procedure |
| Lead sponsor | Medtronic Vascular |
| Sponsor type | Industry |
| Status | COMPLETED |
| ClinicalTrials.gov | NCT02439749 |
2. Clinical Question
The statistical question is whether randomized assignment to renal denervation differs from assignment to the sham procedure with respect to the prespecified clinical outcomes, particularly ambulatory systolic blood pressure measured by 24-hour ambulatory blood pressure monitoring.
Population
Participants enrolled in a study of hypertension, with the registry also listing vascular diseases and cardiovascular diseases among the conditions.
Intervention
The Symplicity Spyral™ multi-electrode renal denervation system.
Comparator
A sham procedure.
Primary question
Does randomized assignment to renal denervation produce a different outcome from the sham procedure for the registered primary endpoints?
The ClinicalTrials.gov record classifies the hypothesis type as superiority. That matters because the statistical question is framed around detecting a difference between randomized groups rather than demonstrating that two treatments are sufficiently similar within a prespecified non-inferiority margin.
3. Trial Design
Symplicity Spyral™ system
- Symplicity Spyral™ multi-electrode renal denervation system
- Randomized treatment assignment
- Blood-pressure outcomes assessed during follow-up
Sham procedure
- Sham procedure
- Randomized comparator assignment
- Blood-pressure outcomes assessed during follow-up
the ClinicalTrials.gov record does not provide arm-specific enrollment counts for the randomized population. The safety results do, however, identify 182 participants at risk in the renal-denervation group and 184 in the sham-procedure group for the reported serious-adverse-event measure.
4. Endpoints
| Endpoint | Registered definition / time frame | Type |
|---|---|---|
| Number of Participants With Major Adverse Events (MAE) Defined as a Composite of Events | From baseline to 1 month post-procedure (6 months for new renal artery stenosis) | Binary |
| Baseline Adjusted Change in Systolic Blood Pressure as Measured by 24-hour Ambulatory Blood Pressure Monitoring | From baseline to 3 months post-procedure | Other / unclear |
Major Adverse Events
The registered MAE endpoint is a composite. Its definition includes all-cause mortality, end-stage renal disease (ESRD), significant embolic events resulting in end-organ damage, renal artery perforation requiring intervention, renal artery dissection requiring intervention, vascular complications, hospitalization for hypertensive crisis not related to confirmed non-adherence with medications or the protocol, and new renal artery stenosis greater than 70%.
The registry specifies a 1-month post-procedure assessment window for the composite, with a 6-month window for new renal artery stenosis.
Ambulatory Systolic Blood Pressure
The second primary endpoint is the baseline adjusted change in systolic blood pressure measured by 24-hour ambulatory blood pressure monitoring from baseline to 3 months post-procedure. The ClinicalTrials.gov record states that the unadjusted treatment difference between renal denervation and sham control groups was -3.9 mmHg and that the baseline adjusted treatment difference was also -3.9 mmHg.
5. Results
Primary Endpoint: Major Adverse Events
This distinction is important. The presence of a posted outcome measure does not by itself provide the statistical evidence needed to quantify a treatment difference. Without a reported event count or effect estimate for the MAE composite in the ClinicalTrials.gov record, it would not be appropriate to calculate a risk difference, relative risk, odds ratio, or confidence interval independently and present it as a reported trial result.
Primary Endpoint: Baseline-Adjusted Change in 24-Hour Ambulatory Systolic Blood Pressure
Baseline-adjusted treatment difference
Renal denervation versus sham procedure · From baseline to 3 months post-procedure
The ClinicalTrials.gov record reports the same value, -3.9 mmHg, for the unadjusted and baseline-adjusted treatment difference.
| Measure | Reported result |
|---|---|
| Unadjusted treatment difference | -3.9 mmHg |
| Baseline-adjusted treatment difference | -3.9 mmHg |
| Time frame | From baseline to 3 months post-procedure |
| Measurement | 24-hour ambulatory systolic blood pressure |
| Adjustment method | Analysis of covariance (ANCOVA) |
| Confidence interval | Not provided in the ClinicalTrials.gov record |
| P-value | Not provided in the ClinicalTrials.gov record for this primary endpoint |
The reported treatment difference of -3.9 mmHg means that the change in 24-hour ambulatory systolic blood pressure was estimated to be 3.9 mmHg lower in the renal-denervation group than in the sham group at the specified 3-month assessment, according to the treatment-difference convention used in the registry result.
It does not mean that every participant experienced a 3.9 mmHg reduction. It is a group-level comparison of changes from baseline. Nor does the number by itself establish the statistical uncertainty around the estimate because the ClinicalTrials.gov record does not include a confidence interval.
The fact that the adjusted and unadjusted treatment differences are both reported as -3.9 mmHg should not be interpreted as evidence that baseline adjustment is universally unnecessary. ANCOVA and unadjusted comparisons answer related but not identical modeling questions, and baseline adjustment can improve precision when baseline values explain outcome variability.
Finally, the absence of a registry-reported p-value means that this result should not be assigned a statistical-significance label based on the effect estimate alone. The magnitude of an estimate and the strength of statistical evidence are separate quantities.
Secondary Blood Pressure Analyses
| Outcome measure | Time frame | Method | P-value |
|---|---|---|---|
| Change in Diastolic Blood Pressure as Measured by 24-hour ABPM | From baseline to 3 months post procedure | ANCOVA | <0.001 |
| Change in Systolic Blood Pressure as Measured by 24-hour ABPM | From baseline to 36 month post-procedure | ANCOVA | 0.026 |
| Change in Diastolic Blood Pressure as Measured by 24-hour ABPM | From baseline to 36 months post-procedure | ANCOVA | 0.052 |
| Change in Office Systolic Blood Pressure | From baseline to 36 months post procedure | ANCOVA | 0.070 |
| Change in Office Diastolic Blood Pressure | From baseline to 36 months post procedure | ANCOVA | 0.009 |
The ClinicalTrials.gov record provides p-values for these five secondary continuous outcomes but do not provide corresponding treatment-effect estimates or confidence intervals. Accordingly, the p-values can be discussed as measures of compatibility with the respective null hypotheses, but they cannot be converted into a magnitude of blood-pressure difference from the ClinicalTrials.gov record.
6. Statistical Methodology
Analysis of Covariance (ANCOVA)
The registry identifies ANCOVA as the reported statistical method for all five posted secondary statistical analyses and explicitly identifies ANCOVA for the baseline-adjusted primary systolic blood-pressure analysis.
For a continuous outcome such as change in blood pressure, ANCOVA models the outcome while accounting for a baseline covariate. In a randomized trial, the treatment indicator is the primary comparison of interest, while baseline measurements can be incorporated to improve statistical precision and account for variation in the outcome that was already present before treatment assignment.
Here, Y represents the follow-up outcome, T represents randomized treatment assignment, and B represents the baseline measurement. The treatment coefficient describes the adjusted difference between randomized groups under the specified model.
Why ANCOVA fits a blood-pressure endpoint
Blood pressure is a continuous measurement, making a linear-model framework natural for comparing groups. A baseline-adjusted analysis can be particularly informative because participants can begin the study with different blood-pressure values even after randomization.
Importantly, ANCOVA does not "correct" the trial after the fact. Randomization remains the foundation of the comparison. Baseline adjustment instead uses measured baseline information to estimate the treatment contrast more precisely within the statistical model.
Analysis population
The registry wording for the posted statistical analyses states that the number of participants analyzed reflects the number of subjects within each treatment arm who have data available for outcome-measure analysis and that this number can vary from the number of subjects in the study. This is an important qualification for interpreting continuous-outcome analyses.
The reported enrollment of 366 therefore should not automatically be substituted for the analysis sample size for any individual blood-pressure endpoint. The statistical-analysis record explicitly warns that outcome-specific availability can lead to different analyzed numbers.
Superiority framework
The statistical analyses are identified as having a superiority hypothesis type. In a superiority framework, the question is whether the observed treatment groups differ, rather than whether the treatment is sufficiently close to a comparator to satisfy a non-inferiority margin.
Randomization as the causal foundation
Randomization is one of the most important statistical features of this trial. When treatment assignment is randomized, systematic differences in prognosis that are unrelated to treatment are expected to balance on average across randomized groups. The resulting treatment comparison can therefore be interpreted causally under the assumptions required for the randomized trial design.
Randomization does not guarantee that every measured baseline characteristic will be numerically identical between groups. Nor does it prevent missing follow-up data or measurement error. Its principal role is to establish a defensible framework for comparing outcomes between treatment assignments.
7. Statistical Methods Explained
Why was ANCOVA used?
ANCOVA is well suited to a continuous endpoint such as change in ambulatory blood pressure. It allows the analysis to incorporate baseline blood pressure while estimating the treatment-group difference. In this trial, the registry specifically reports ANCOVA for the baseline-adjusted systolic blood-pressure endpoint and for the posted secondary blood-pressure analyses.
What does a treatment difference of -3.9 mmHg mean?
For the primary ambulatory systolic blood-pressure result, the reported treatment difference is -3.9 mmHg for renal denervation versus sham. Under the direction used in the registry result, the negative value indicates a lower change in the renal-denervation group relative to the sham group. It is a between-group estimate, not a statement about every participant's individual change.
Why does baseline adjustment matter?
Two randomized groups can have different baseline measurements simply through sampling variation. ANCOVA uses baseline information to account for some of the outcome variability associated with those starting measurements. If baseline blood pressure is predictive of follow-up blood pressure, this can make the treatment comparison more statistically efficient.
What does a p-value tell us?
A p-value quantifies how unusual the observed data would be, under the specified null hypothesis and statistical model, if the null hypothesis were true. A small p-value can provide evidence against that null hypothesis. It does not measure the size of the treatment effect, the probability that the treatment hypothesis is true, or the clinical importance of an effect.
What does p < 0.001 mean for the 3-month diastolic blood-pressure analysis?
The registry analysis reports p < 0.001 for change in diastolic blood pressure measured by 24-hour ABPM from baseline to 3 months. This indicates strong statistical evidence against the null hypothesis specified for that ANCOVA comparison under the reported analysis. The ClinicalTrials.gov record does not provide the treatment difference or its confidence interval, so the p-value alone cannot tell us how large the blood-pressure difference was.
Why should the 36-month p-values be read individually?
The ClinicalTrials.gov record reports p-values of 0.026, 0.052, 0.070, and 0.009 for four different 36-month secondary blood-pressure outcomes. Each belongs to a distinct endpoint and measurement type. A p-value should therefore be interpreted in the context of its endpoint, time frame, analysis population, and prespecified role. The registry-reported extract does not establish a multiplicity-adjustment procedure that would justify treating these results as a single unadjusted family of confirmatory tests.
8. Primary Result Interpretation in Statistical Context
The primary ambulatory systolic blood-pressure result supplies an effect estimate of -3.9 mmHg, while the registry-reported secondary analyses supply p-values without effect estimates. These are different types of statistical information. An effect estimate tells the reader about the direction and magnitude of a comparison; a confidence interval describes uncertainty around that estimate; and a p-value addresses compatibility with a specified null hypothesis.
A statistically small p-value cannot compensate for a missing effect estimate when the clinical question concerns magnitude. Conversely, an effect estimate without an uncertainty interval cannot fully communicate how precisely it was estimated.
No confidence interval for the primary -3.9 mmHg treatment difference is included in the ClinicalTrials.gov record. A confidence interval would provide an interval of values compatible with the statistical model and sampling framework. It would help distinguish a precisely estimated treatment difference from one with substantial statistical uncertainty.
The secondary p-values range from <0.001 to 0.070, but those values cannot be used to reconstruct the corresponding blood-pressure differences. P-values depend on both the size of the observed difference and the amount of statistical information available. Two analyses with similar effect sizes can have different p-values if their precision differs.
A 3-month ambulatory blood-pressure analysis and a 36-month ambulatory blood-pressure analysis are not interchangeable. They answer questions about different follow-up periods. Longitudinal treatment comparisons can change over time because the amount of available outcome information, treatment exposure, follow-up, and missingness can all differ at later assessments.
9. Secondary Endpoint Results
24-Hour Ambulatory Blood Pressure at 3 Months
Change in diastolic blood pressure
From baseline to 3 months post procedure · ANCOVA
The statistical analysis compares renal denervation with the sham procedure for change in diastolic blood pressure measured by 24-hour ABPM. The ClinicalTrials.gov record identifies the endpoint as continuous and the hypothesis type as superiority.
The reported p-value supports a statistical comparison under the specified analysis, but the ClinicalTrials.gov record does not include the corresponding treatment difference or confidence interval. Therefore, the result should not be translated into a numerical magnitude of diastolic blood-pressure change beyond what is explicitly reported.
36-Month Ambulatory Systolic Blood Pressure
Change in systolic blood pressure
From baseline to 36 month post-procedure · ANCOVA
This analysis extends the ambulatory systolic blood-pressure comparison to 36 months. The p-value of 0.026 is reported under a superiority hypothesis. The registry extract does not supply an effect estimate or confidence interval for this analysis.
36-Month Ambulatory Diastolic Blood Pressure
Change in diastolic blood pressure
From baseline to 36 months post-procedure · ANCOVA
The 36-month ambulatory diastolic blood-pressure analysis reports a p-value of 0.052. This value should be interpreted as the result of the specified statistical test, not as a measure of treatment magnitude. Because the extract does not provide the estimated treatment difference or its confidence interval, the direction and size of the between-group effect cannot be established from this statistical-analysis record alone.
36-Month Office Systolic Blood Pressure
Change in office systolic blood pressure
From baseline to 36 months post procedure · ANCOVA
This endpoint uses office systolic blood pressure rather than 24-hour ambulatory monitoring. That measurement distinction matters: the two endpoints are related but represent different methods of assessing blood pressure. The registry-reported analysis reports p = 0.070 and ANCOVA, but no effect estimate or confidence interval.
36-Month Office Diastolic Blood Pressure
Change in office diastolic blood pressure
From baseline to 36 months post procedure · ANCOVA
The office diastolic blood-pressure analysis reports p = 0.009 under an ANCOVA superiority analysis. As with the other secondary analyses, the ClinicalTrials.gov record does not include the corresponding treatment difference or confidence interval.
| Secondary endpoint | Time frame | Analysis | Reported p-value | Effect estimate reported? |
|---|---|---|---|---|
| Change in diastolic BP, 24-hour ABPM | 3 months | ANCOVA | <0.001 | No |
| Change in systolic BP, 24-hour ABPM | 36 months | ANCOVA | 0.026 | No |
| Change in diastolic BP, 24-hour ABPM | 36 months | ANCOVA | 0.052 | No |
| Change in office systolic BP | 36 months | ANCOVA | 0.070 | No |
| Change in office diastolic BP | 36 months | ANCOVA | 0.009 | No |
10. Safety Results
the ClinicalTrials.gov record reports serious adverse events by arm as follows:
| Safety measure | Renal Denervation | Sham Procedure |
|---|---|---|
| Serious adverse events, affected / at risk | 38 / 182 | 36 / 184 |
Renal denervation
38 participants were affected among 182 participants at risk for the reported serious-adverse-event measure.
Sham procedure
36 participants were affected among 184 participants at risk for the reported serious-adverse-event measure.
The registry extract does not provide a formal statistical comparison, confidence interval, or p-value for these serious-adverse-event counts. It would therefore be inappropriate to infer a comparative safety effect from the counts alone.
It is also important to distinguish this safety measure from the registered primary MAE endpoint. The MAE endpoint is a defined composite containing several specific event categories and uses a particular follow-up window. The serious-adverse-event measure reported here is reported separately and should not be treated as interchangeable with the MAE composite.
11. Understanding the Analysis Population
One subtle but important feature of the registry's statistical-analysis wording is that the analyzed number of participants can differ from the overall number enrolled. The statistical-analysis records state that the number analyzed reflects participants in each treatment arm who have data available for the particular outcome measure.
A participant can be randomized and included in the trial while lacking an evaluable measurement for a particular follow-up endpoint. Outcome-specific analysis populations therefore need to be identified before interpreting an estimate or p-value.
This issue is especially relevant to longitudinal blood-pressure endpoints. A 3-month analysis and a 36-month analysis can have different numbers of evaluable participants because follow-up availability can change over time.
12. Randomization and the Meaning of the Sham Comparator
The trial's randomized parallel design creates a direct comparison between two assigned procedures: renal denervation and a sham procedure. From a statistical perspective, the sham group provides a contemporaneous randomized comparator rather than a comparison with an historical blood-pressure level or an external control population.
This is particularly relevant for outcomes measured from baseline. Both groups can change over time for reasons unrelated to the active intervention, including the natural course of blood pressure, measurement variation, study participation, and other factors. The randomized comparison focuses on the difference between those changes rather than interpreting a within-group change in isolation.
Within-group change
Asks how an outcome changes from baseline within one treatment group.
Between-group difference
Asks whether the changes differ between randomized treatment assignments.
Why randomization helps
Randomized assignment establishes the primary framework for attributing systematic differences in outcomes to treatment assignment.
Why the sham matters
The comparator supplies a randomized reference group experiencing the same general trial framework without the renal-denervation intervention.
13. Three-Month Versus 36-Month Analysis
The ClinicalTrials.gov record provides one primary blood-pressure result at 3 months and four secondary blood-pressure analyses at 36 months. These analyses should be read as separate statistical questions rather than as a single repeated measurement.
| Time point | Outcome | Role | Reported statistical information |
|---|---|---|---|
| 3 months | 24-hour ambulatory systolic BP | Primary | Baseline-adjusted difference = -3.9 mmHg |
| 3 months | 24-hour ambulatory diastolic BP | Secondary | P < 0.001 |
| 36 months | 24-hour ambulatory systolic BP | Secondary | P = 0.026 |
| 36 months | 24-hour ambulatory diastolic BP | Secondary | P = 0.052 |
| 36 months | Office systolic BP | Secondary | P = 0.070 |
| 36 months | Office diastolic BP | Secondary | P = 0.009 |
The differing p-values should not be interpreted as evidence of a simple "trajectory" unless the longitudinal analysis explicitly models repeated measurements and their covariance structure. The statistical analyses posted on ClinicalTrials.gov identify ANCOVA for each reported comparison but do not provide a repeated-measures model in the extract.
14. Why the Primary Estimate Is More Informative Than a P-value Alone
The reported -3.9 mmHg primary treatment difference directly communicates the direction and magnitude of the between-group comparison. By contrast, a p-value such as 0.026 indicates the result of a statistical test but does not tell the reader the size of the underlying treatment difference.
1. Effect estimate: How large is the observed difference?
2. Confidence interval: How precisely has that difference been estimated?
3. P-value: How compatible are the data with the specified null hypothesis?
The registry-reported SPYRAL HTN-OFF MED extract gives the first component for the primary systolic endpoint and the third component for several secondary endpoints. It does not supply the corresponding confidence intervals or most secondary effect estimates.
This is why statistical reporting should resist the temptation to reduce a clinical trial to a binary "significant/non-significant" label. The scientific question concerns the estimated treatment difference, its uncertainty, its clinical context, and the design that generated the comparison.
15. P-values in This Trial
The statistical-analysis records contain five secondary p-values. They range from <0.001 to 0.070. Each p-value belongs to a specific endpoint and time frame.
The graphic is only a visual ordering of the reported numerical p-values; it is not a ranking of the clinical importance of the endpoints. A p-value of 0.009 does not automatically describe a more clinically important effect than a p-value of 0.026, because effect size and clinical relevance are not encoded by the p-value alone.
16. Limitations
- Incomplete effect reporting in the registry-reported extract: the primary systolic endpoint includes a treatment difference of -3.9 mmHg, but no confidence interval or p-value is reported. The secondary analyses provide p-values but not their corresponding effect estimates or confidence intervals.
- Primary MAE comparison not quantified: the ClinicalTrials.gov record provides the registered MAE definition and time frame but no formal between-group statistical estimate for that endpoint.
- Outcome-specific analysis populations: the registry states that the number analyzed can vary according to availability of data for each outcome. Overall enrollment should therefore not be substituted for every analysis population.
- Multiplicity: the extract identifies multiple secondary statistical analyses but does not specify a multiplicity-adjustment procedure. The individual p-values should therefore be interpreted with appropriate caution.
- Different measurement methods: 24-hour ambulatory blood pressure and office blood pressure are distinct outcome measurements and should not be treated as identical endpoints.
- Different follow-up windows: the 3-month and 36-month analyses answer different time-specific questions and should not be collapsed into a single estimate.
- ANCOVA assumptions: interpretation of a linear-model estimate depends on the adequacy of the specified model and its assumptions. The registry-reported extract does not provide the diagnostic information needed to assess those assumptions directly.
- Safety versus efficacy populations: the serious-adverse-event figures identify affected and at-risk participants, but the ClinicalTrials.gov record does not provide a formal comparative safety analysis.
- Registry-level interpretation: the analysis presented here is limited to the information contained in the ClinicalTrials.gov record and does not add unreported numerical results from publications or other sources.
17. What Is Not Established by the Supplied Statistical Results
No individual-patient effect
The -3.9 mmHg treatment difference is a group-level estimate. It does not describe the response of any particular participant.
No confidence-interval inference
Because the primary confidence interval is not reported, the precision of the -3.9 mmHg estimate cannot be quantified from this extract.
No effect-size reconstruction
The secondary p-values cannot be converted into blood-pressure differences without additional statistical information.
No multiplicity conclusion
The extract does not establish how the family of secondary endpoints was handled for type I error control.
These distinctions are not merely technical. They determine what can responsibly be inferred from a clinical-trial record. A statistical analysis should preserve the boundary between what the registry reports and what a reader might otherwise be tempted to calculate or assume.
18. Why This Trial Matters Statistically
SPYRAL HTN-OFF MED is a useful teaching case because it demonstrates how a randomized clinical trial can combine a continuous primary endpoint, baseline adjustment, a sham comparator, multiple follow-up periods, and both efficacy and safety outcomes.
| Concept | How it appears in SPYRAL HTN-OFF MED |
|---|---|
| Randomization | Participants were randomly allocated to two parallel treatment arms. |
| Sham control | The comparator was a sham procedure rather than an untreated historical control. |
| Continuous endpoint | Blood-pressure changes were analyzed as continuous outcomes. |
| ANCOVA | Used for the baseline-adjusted primary systolic endpoint and the five posted secondary statistical analyses. |
| Baseline adjustment | The primary systolic endpoint reports both unadjusted and baseline-adjusted treatment differences, each -3.9 mmHg. |
| P-values | Five secondary statistical analyses report p-values ranging from <0.001 to 0.070. |
| Multiple endpoints | The registry contains primary and secondary blood-pressure outcomes across different measurement methods and follow-up windows. |
| Analysis populations | The registry states that analyzed numbers depend on availability of outcome data. |
| Safety comparison | Serious adverse events are reported as affected participants over participants at risk by arm. |
The trial therefore illustrates an important principle in applied biostatistics: The statistical method cannot be separated from the endpoint definition. "Blood pressure" is not one generic outcome. The exact measurement method, baseline adjustment, time frame, analysis population, and statistical model all determine what the reported result means.
19. A Practical Reading Strategy for the Trial
A reader can approach the ClinicalTrials.gov record in a structured sequence.
- Start with randomization. Identify the two assigned groups and recognize that the primary comparison is between randomized treatment assignments.
- Read the endpoint definition. Determine whether the outcome is a binary event, a continuous measurement, or a composite endpoint.
- Check the time frame. A 3-month outcome is not interchangeable with a 36-month outcome.
- Identify the statistical model. Here, ANCOVA is the reported method for the posted statistical analyses.
- Find the effect estimate. For the primary ambulatory systolic endpoint, the registry-reported estimate is -3.9 mmHg.
- Look for uncertainty. The registry-reported extract does not provide a confidence interval for that estimate, so precision cannot be fully assessed from the extract.
- Then read the p-value. For secondary analyses, the p-value supplies evidence relative to the null hypothesis but does not provide the effect magnitude.
- Check analysis-population information. The registry explicitly states that participants analyzed can vary according to data availability.
- Separate efficacy and safety. Serious-adverse-event counts should not be substituted for the registered MAE composite.
This workflow helps prevent one of the most common mistakes in clinical-trial interpretation: treating a single p-value as if it contains the entire statistical result.
20. Clinical Interpretation vs Statistical Interpretation
Statistical interpretation
The randomized trial reports a baseline-adjusted 3-month ambulatory systolic blood-pressure treatment difference of -3.9 mmHg. Five secondary blood-pressure analyses are reported using ANCOVA with p-values from <0.001 to 0.070.
Clinical interpretation
The ClinicalTrials.gov record describes differences in blood-pressure outcomes and serious adverse events between randomized groups. Determining the clinical importance of those differences requires considering the magnitude, precision, endpoint definition, follow-up period, and safety information together.
The two forms of interpretation should remain distinct. Statistical evidence addresses uncertainty under a specified model; clinical interpretation asks what the magnitude and nature of an observed difference mean in the context of the condition and intervention. The registry extract does not provide enough information to make every aspect of that clinical interpretation quantitative.
21. Longitudinal Trial History
Trial begins
the ClinicalTrials.gov record lists June 2015 as the study start.
Ambulatory blood-pressure endpoint
The registered baseline-adjusted systolic blood-pressure endpoint is assessed from baseline to 3 months post-procedure. The reported treatment difference is -3.9 mmHg.
Longer-term blood-pressure outcomes
Four secondary blood-pressure analyses are reported at 36 months: ambulatory systolic, ambulatory diastolic, office systolic, and office diastolic blood pressure.
Primary completion
the ClinicalTrials.gov record lists April 30, 2020 as the primary completion date.
22. Statistical Concepts in This Trial
Learn more about the methods used in this trial:
23. Related Statistical Calculators
24. Sources
- ClinicalTrials.gov: SPYRAL PIVOTAL - SPYRAL HTN-OFF MED Study, NCT02439749.
- PubMed: PMID 39526682.
- PubMed: PMID 35852582.
- PubMed: PMID 34474735.
- PubMed: PMID 33957242.
- PubMed: PMID 32234534.
The numerical analysis on this page is restricted to the ClinicalTrials.gov record. The PubMed records are provided as source links; no numerical result has been added from those publications.
Continue through the Clinical Biostats statistical library
Connect this trial's randomized design, ANCOVA analysis, continuous endpoints, and p-value interpretation to deeper statistical tutorials and practical calculators.
25. Record Summary
SPYRAL HTN-OFF MED provides a useful example of randomized-trial analysis centered on continuous blood-pressure outcomes. The study enrolled 366 participants in two parallel randomized arms comparing the Symplicity Spyral™ multi-electrode renal denervation system with a sham procedure. The registered primary endpoints included a composite major-adverse-event measure and baseline-adjusted change in 24-hour ambulatory systolic blood pressure from baseline to 3 months.
The primary ambulatory systolic endpoint reports a treatment difference of -3.9 mmHg, with both the unadjusted and baseline-adjusted treatment differences reported as -3.9 mmHg. The ClinicalTrials.gov record does not provide a confidence interval or p-value for that primary estimate. Five secondary blood-pressure analyses use ANCOVA and report p-values of <0.001, 0.026, 0.052, 0.070, and 0.009 across 3-month and 36-month ambulatory and office measurements.
Statistically, the most important lesson is that these pieces of evidence answer different questions. The treatment difference describes magnitude; a confidence interval would describe precision; and a p-value addresses compatibility with a null hypothesis.