STREAM-Strategic Reperfusion (With Tenecteplase and Antithrombotic Treatment) Early After Myocardial Infarction: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for STREAM-Strategic Reperfusion (With Tenecteplase and Antithrombotic Treatment) Early After Myocardial Infarction (NCT00623623), a randomized phase 3 study. The primary outcome is Number of Patients With All-cause Mortality, Cardiogenic Shock, Congestive Heart Failure and Recurrent Myocardial Infarction Within 30 Days for FAS..

2. Clinical Question

This was a strategy comparison, not simply a comparison of two individual drugs or procedures. The tenecteplase strategy included additional antiplatelet and antithrombin therapy and a planned route to catheterisation or rescue intervention. The comparator was primary PCI. Accordingly, the clinical question concerns the outcomes associated with assignment to these treatment pathways in the registry-defined population.

The primary endpoint combines serious but different clinical events. A composite can capture several ways in which a treatment strategy might affect patients, while a single summary outcome may not show which component contributes most to any observed difference. The registry record reports the composite outcome and separately lists several secondary outcomes. It does not report component-specific primary-endpoint counts in the information available for this analysis.

3. Trial Design

The registry record identifies STREAM as a phase 3, randomized, parallel-group trial with no masking. Enrollment is recorded as 1899, and the status is completed. The primary endpoint is described for the full analysis set, which the record also calls the intention-to-treat population. It includes consented patients randomized through the Interactive Voice Response System or manually allocated to a treatment group by the investigator.

The trial compared two care strategies for patients with acute myocardial infarction within 3 hours of symptom onset. One strategy began with tenecteplase, clopidogrel, and enoxaparin, with routine catheterisation followed by intervention when appropriate, or rescue coronary intervention if required. The other strategy was standard primary PCI according to local standards. The study description frames the setting as pre-hospital care or a community hospital emergency room without a PCI facility.

The primary outcome was the number of patients with all-cause mortality, cardiogenic shock, congestive heart failure, or recurrent myocardial infarction within 30 days. The record also lists secondary outcomes including all-cause and cardiac mortality, shock, heart failure, reinfarction, rehospitalisation, and repeat target-vessel revascularisation. It records 25 outcome measures and 25 analyses, while 24 secondary outcomes are posted.

The trial description calls the study exploratory and says its statistical tests were exploratory, using descriptive p-values for formal hypothesis generation. The registry record does not report the trial's original sample-size justification, detailed allocation procedures, or a broader statistical analysis plan. The specific handling of missing outcomes and any interim analyses are 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.

Before assignment: All subjects were screened for eligibility to participate in trial. Subjects attended specialist sites to ensure that they (the subjects) met all implemented inclusion/exclusion criteria. Subjects were not to be randomised to trial drug if any of the specific entry criteria was violated.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Full analysis set (FAS): All consented patients either randomised by Interactive Voice Response System (IVRS) or treated (i.e., manually allocated to treatment group) by the investigator. This population is also referred to as the intention-to-treat population (ITT).

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: Full analysis set (FAS): All consented patients either randomised by Interactive Voice Response System (IVRS) or treated (i.e., manually allocated to treatment group) by the investigator. This population is also referred to as the intention-to-treat population (ITT).

6. Primary Endpoint

The registry lists the primary outcome as Number of Patients With All-cause Mortality, Cardiogenic Shock, Congestive Heart Failure and Recurrent Myocardial Infarction Within 30 Days for FAS., measured over 30 days.

The primary endpoint counts patients in the full analysis set who experienced all-cause mortality, cardiogenic shock, congestive heart failure, or recurrent myocardial infarction within 30 days. It is a binary composite: for each patient, the relevant distinction is whether an event meeting the composite definition was recorded during the stated timeframe. The registry record describes the outcome as a count of participants, rather than a count of repeated events.

A composite endpoint is used to bring several clinically important outcomes into one primary measure. In general, this can capture a broader range of possible consequences of a treatment strategy and can make the main comparison more informative than focusing on a single event type alone. Its interpretation nevertheless depends on the components: events can differ in severity, frequency, and how they respond to treatment. A composite summary does not by itself establish that each component moved in the same direction.

The posted primary outcome reports 116 patients in the tenecteplase group and 135 in the primary PCI group. These counts are presented for the full analysis set, with the group denominators recorded as 949 and 948. The registry description does not provide primary component-level counts, so the contribution of mortality, shock, heart failure, and reinfarction to the combined result cannot be separated from these figures.

The 30-day timeframe defines the period being summarized. It does not describe longer-term outcomes, and the registry record does not provide a basis for extending the primary finding beyond that period.

7. Results Posted in the Registry

• Primary PCI (Group B): 14.2% (135/948) versus 12.2% (116/949) on control; risk difference 2.0 percentage points (95% CI -1.0 to 5.1), risk ratio 1.17 (0.92–1.47); unadjusted two-proportion z-test p = 0.195.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Number of Patients With All Cause Mortality

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Primary PCI (Group B): 4.4% (42/948) versus 4.5% (43/949) on control; risk difference -0.1 percentage points (95% CI -2.0 to 1.8), risk ratio 0.98 (0.65–1.48); unadjusted two-proportion z-test p = 0.916.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Primary PCI (Group B) and the control. The interval gives the range of effects that remain compatible with the data.

Number of Patients With Cardiac Mortality

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Primary PCI (Group B): 3.4% (32/948) versus 3.3% (31/949) on control; risk difference 0.1 percentage points (95% CI -1.5 to 1.7), risk ratio 1.03 (0.64–1.68); unadjusted two-proportion z-test p = 0.895.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Primary PCI (Group B) and the control. The interval gives the range of effects that remain compatible with the data.

Number of Patients With Cardiogenic Shock

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Primary PCI (Group B): 5.9% (56/948) versus 4.3% (41/949) on control; risk difference 1.6 percentage points (95% CI -0.4 to 3.6), risk ratio 1.37 (0.92–2.02); unadjusted two-proportion z-test p = 0.117.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Primary PCI (Group B) and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists secondary outcomes spanning mortality, cardiac mortality, cardiogenic shock, heart failure, reinfarction, rehospitalisation for cardiac or non-cardiac reasons, and repeat target-vessel revascularisation. The secondary outcome entries available for detailed comparison include all-cause mortality, cardiac mortality, and cardiogenic shock. Their estimates do not form a uniformly decisive pattern: event counts are close for the mortality outcomes, while the shock counts differ more visibly, but the reported uncertainty remains substantial.

For the available secondary comparisons, the recalculated confidence intervals include the no-difference value for the risk difference. The registry's model-based entries also have confidence intervals spanning the relative-risk no-difference value. These results do not establish a clear secondary-outcome advantage and should be viewed as contextual rather than as confirmation of the primary finding. They are broadly compatible with the primary result's uncertainty, not an independent resolution of it.

The registry reports that the study was exploratory and that p-values were descriptive for hypothesis generation. When many outcomes are considered, unadjusted tests carry less evidential weight than a single prespecified primary test because multiple opportunities for chance findings exist. The registry record does not report a multiplicity correction or testing hierarchy. A nominal p-value for one secondary outcome, if considered alone, would therefore not establish a reliable treatment effect.

Secondary endpoints can help identify questions for later research, but their interpretation depends on outcome definitions, missing-data handling, and the analysis plan. The record does not provide enough detail to assess all these aspects across the full set of posted secondary measures. Nor do the listed results establish that each component of the primary composite followed the same pattern. The primary endpoint remains the main summary comparison specified in the registry.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Number of Patients With All Cause Mortality
• Number of Patients With Cardiac Mortality
• Number of Patients With Cardiogenic Shock
• Number of Patients With Congestive Heart Failure (CHF)
• Number of Patients With Recurrent Myocardial Infarction (Reinfarction)
• Number of Patients With Rehospitalisation for Cardiac Reasons
• Number of Patients With Rehospitalisation for Non-cardiac Reasons
• Number of Patients With Serious Repeat Target Vessel Revascularization

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary at the participant level: a patient either met the composite endpoint within 30 days or did not. The posted group counts and denominators support an unadjusted comparison of the observed proportions. The registry also reports a model-based analysis. These methods answer related but not identical questions, so their effect estimates should be interpreted on their own scales.

What does the risk difference describe?

General statistical explanation: A risk difference subtracts one group's event proportion from the other group's event proportion. It is an absolute measure, expressed on the probability scale. A value of zero represents equal proportions. The recalculated estimate is 0.020, defined as primary PCI minus tenecteplase, and its confidence interval ranges from -0.010 to 0.051. Because that interval crosses zero, it includes both a modest difference favoring either group and no absolute difference. The estimate is unadjusted, meaning it is calculated from the posted group-level summaries rather than from a model adjusting for participant characteristics.

What does a relative risk describe?

General statistical explanation: A relative risk divides one group's event probability by the other's. A value of 1 represents equal probabilities; values on either side indicate a relative contrast. The registry reports a modified Poisson regression relative risk of 0.86, with a 95% confidence interval from 0.68 to 1.08. The registry entry lists the analysis groups but does not make its reference direction explicit in the displayed result description, so the estimate is best read as the model's reported relative measure rather than assigned a direction beyond its documented group ordering. The interval includes 1 and is compatible with no relative difference.

Why use modified Poisson regression?

General statistical explanation: Poisson regression is commonly associated with counts, but a modified form can be used to estimate relative risks for binary outcomes. The modification addresses the fact that binary data do not have the same variance structure as unrestricted count data. The registry names modified Poisson regression for the primary analysis. It does not report further implementation details, such as covariates or variance estimation, so those should not be inferred.

How should the p-value be read?

General statistical explanation: A p-value measures how unusual results at least as incompatible with a specified null hypothesis would be under the assumptions of a statistical test. It is not the probability that the null hypothesis is true, nor does it measure clinical importance. For the primary comparison, the registry reports p = 0.195, and the unadjusted re-calculation also gives p = 0.195. That value does not establish equivalence, prove there is no effect, or tell the probability that either strategy is better. The registry describes its tests as exploratory and its p-values as descriptive, intended for hypothesis generation.

What is the role of the full analysis set?

The registry defines the full analysis set as consented patients randomized through IVRS or manually allocated to a treatment group. It also identifies this population as the intention-to-treat population. In general, an intention-to-treat approach retains participants according to their assigned or allocated group and aims to preserve the comparison created by randomization. The record does not detail how missing endpoint values, withdrawals, or protocol deviations were handled, so the exact implementation cannot be assessed from the registry description.

The primary outcome is a composite, and its component events are not interchangeable. In general, combining outcomes can increase the scope of a primary measure, but the aggregate estimate can be driven by more frequent components. Without component-specific primary results, the overall comparison cannot reveal whether the event types had similar patterns. The registry separately lists secondary outcomes, but these do not provide a substitute for unavailable component counts within the primary composite.

The trial description identifies the analysis as exploratory. It does not report a confirmatory testing hierarchy, a prespecified multiplicity adjustment, or detailed model specifications beyond the named regression method. The absence of such detail in the registry record means those features cannot be assumed. General statistical practice treats multiple unadjusted tests cautiously because the chance of at least one apparently notable result can rise as more hypotheses are examined.

11. Confidence Intervals and Precision

The primary risk-difference estimate is 0.020, with a 95% confidence interval from -0.010 to 0.051. A confidence interval gives a range of effect values compatible with the data and the method's assumptions. It is not a guarantee that the true effect lies within the interval, and it does not show the probability that a particular value is correct. Its width is useful for judging how precisely the effect has been estimated.

For this comparison, the interval crosses zero, the no-difference value for a risk difference. It therefore does not rule out no absolute difference. It also extends to values on either side of zero, so the data leave uncertainty about the direction as well as the magnitude of the contrast. The point estimate alone should not be treated as a settled estimate of the treatment strategy's effect.

The registry's model-based relative risk is 0.86, with a 95% confidence interval from 0.68 to 1.08. For a relative risk, 1 is the no-difference value. This interval includes 1 and permits a range of relative effects around that value. Because the risk difference and relative risk use different scales, their interval endpoints are not directly comparable as numbers; both express uncertainty in their respective measures.

The two primary p-values are both reported as 0.195. A p-value does not replace a confidence interval: it summarizes evidence against a particular null under the test assumptions, whereas the interval displays a range of effect sizes compatible with the analysis. Together, the estimate, interval, and p-value show that the primary comparison is imprecise enough that materially different interpretations remain possible. The registry's exploratory framing further cautions against treating the findings as a definitive test of a prespecified confirmatory claim.

The registry record does not report a separate precision target or clinically important difference against which to judge the interval. Statistical uncertainty can therefore be described, but the clinical importance of every value within the interval cannot be determined from the record alone.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning figures are illustrations based on the observed effect, not the original trial design. For this binary outcome, the figures are participants per arm. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column describes the planned number of participants per arm under the stated assumptions for a design that would have an 80% chance of detecting the specified effect if that effect were the true one. The 90% power column describes the corresponding planning target for a 90% chance. These are design probabilities under assumptions, not guarantees about a future trial or its results. This is not observed power.

Rows that plan for 100% of the observed effect use the full effect estimate from the posted results. Rows that plan for 75% or 50% of the observed effect consider smaller effects. Smaller effects are harder to distinguish from random variation, so they require more participants per arm to attain the same planned power. For example, the primary endpoint's planning illustration lists 4,429 participants per arm for 80% power at 100% of the observed effect, and 17,141 participants per arm at 50% of the observed effect. The associated 90% targets are larger than the corresponding 80% targets.

These figures should be read as hypothetical planning scenarios, not as a reconstruction of the trial's intended sample size. They rely on an effect estimated from the trial itself, which is uncertain, and they do not account for any assumptions not listed in the registry record. The registry does not report that these planning calculations governed the original trial.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.122; treatment proportion p₁ = 0.142; average p̄ = 0.132.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4792 = 0.9392.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4790 = 0.4031.
4. n = (0.9392 + 0.4031)² / (0.0202)² = 4,428.4, rounded up to 4,429 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.122233930453109                          # control proportion
p1 <- 0.142405063291139                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(623623)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 4,429 (80% power), 5,928 (90% power)
 75% of observed effect: n per arm = 7,746 (80% power), 10,370 (90% power)
 50% of observed effect: n per arm = 17,141 (80% power), 22,947 (90% power)
power.prop.test: 4428.4 
Simulated power at n = 4,429 per arm: 0.800 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The registry's own primary analysis is a modified Poisson regression, reporting a relative risk of 0.86, a 95% confidence interval from 0.68 to 1.08, and p = 0.195. The unadjusted re-calculation uses the posted counts to estimate a risk difference of 0.020 for primary PCI relative to tenecteplase, with a 95% confidence interval from -0.010 to 0.051 and p = 0.195.

These figures should not be expected to match numerically because a relative risk and a risk difference express effects on different scales, and the registry result is model-based while the recalculation is unadjusted. Their central inferential message is similar: the primary confidence intervals include the respective no-difference values, and the posted p-values do not distinguish the strategies clearly. The registry record marks no discrepancy between the posted analysis and the re-calculation.

The registry's analysis is the trial's reported result and should be the main reference for what the investigators analyzed. The re-calculation is a transparent comparison based on summary counts; it does not supersede the sponsor's model or reproduce any unreported adjustment. Neither analysis establishes equivalence or proves that the treatment strategies have identical effects.

The registry describes the study as exploratory and the tests as descriptive for hypothesis generation. That context matters: the reported p-value should not be reframed as a confirmatory threshold-based conclusion, and the estimates should be read alongside their confidence intervals and the trial's stated analytic limitations.

15. Multiplicity, Missing Data and Analysis Populations

The registry record says the study was exploratory and that all statistical tests were exploratory, with descriptive p-values intended for formal statistical hypothesis generation. This statement gives important context for reading the reported p-values: they are not presented as a complete confirmatory testing framework. The record does not report a multiplicity adjustment, a hierarchical testing procedure, or a formal plan controlling the overall chance of false-positive findings across the outcome set.

Multiplicity matters when researchers examine several outcomes or several analyses. General statistical explanation: If each test is interpreted separately without adjustment, the chance of observing at least one apparently unusual result can increase as the number of tests grows. This does not mean every finding is false; it means secondary findings should be interpreted with caution and in light of the prespecified analysis plan. The registry's exploratory description supports treating the secondary results as hypothesis-generating.

The primary endpoint population is defined as the full analysis set and also called the intention-to-treat population. It includes consented patients randomized through IVRS or manually allocated by the investigator. The registry record does not report the number of participants with missing primary outcomes, the imputation method if any, or whether missing observations were handled through exclusions or another approach. It also does not report details of outcome follow-up completeness.

Randomization and analysis population definitions help frame an estimate, but they do not on their own resolve missing-data concerns. General statistical explanation: When outcomes are missing, the effect estimate can depend on how missingness relates to treatment assignment and the outcome. Without the missing-data methods and completeness details, the potential impact cannot be judged from the registry record.

The trial record names modified Poisson regression for the sponsor's primary analysis but does not provide covariate specifications, model diagnostics, or details of variance estimation. It also does not report interim analyses. These items are not reported in the registry record and should not be inferred from the posted estimate or from the exploratory description.

16. 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 5% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: ventricular fibrillation (4.9%), cardiac arrest (1.7%), atrioventricular block (1.5%), pulmonary oedema (1.5%), ventricular tachycardia (1.4%).

The safety listing reports serious adverse events affecting 146 of 949 in the tenecteplase group and 164 of 948 in the primary PCI group. It separately reports deaths of 43 and 42, respectively. The registry also records zero other adverse events in each group. These figures are descriptive safety information; they do not by themselves establish that one strategy causes more or fewer adverse events.

The registry lists serious adverse event terms with percentages, including ventricular fibrillation at 4.9%, cardiac arrest at 1.7%, atrioventricular block at 1.5%, pulmonary oedema at 1.5%, ventricular tachycardia at 1.4%, and atrial fibrillation at 1.2%. The record does not give group-specific percentages for those terms in the information, so they should not be assigned to one treatment group or compared between groups.

Safety summaries require attention to definitions, follow-up, and whether the figures count participants or events. The primary endpoint counts participants with a composite clinical outcome; the safety listing is a separate registry summary. The record does not report further detail about event adjudication or the relationship between individual serious adverse events and primary endpoint events. The listed deaths should therefore be read as safety data, not as a replacement for the primary analysis.

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

The posted data show the number of patients recorded with a four-part composite outcome within 30 days in each randomized strategy group. The unadjusted risk difference and the registry's model-based relative risk both have confidence intervals that include their respective no-difference values. The reported p-value is 0.195. Together, these results indicate uncertainty about the size and direction of the difference between the strategies in this trial.

That uncertainty is not the same as evidence that the strategies are equivalent. A study that does not clearly distinguish groups may still be compatible with effects in either direction, as reflected by the confidence intervals. Equivalence or non-inferiority requires a prespecified margin and an analysis designed to assess that specific claim. Such a margin and analysis are not reported in the registry record.

The primary result concerns a composite of all-cause mortality, cardiogenic shock, congestive heart failure, and recurrent myocardial infarction, measured within 30 days in the full analysis set. It does not show how each component contributed, and it does not describe outcomes beyond the stated timeframe. The secondary outcomes offer additional descriptive information but are exploratory and do not establish a definitive effect for any one outcome.

The result also applies to the treatment strategies as described, including accompanying therapies and the planned approach to catheterisation or rescue intervention. It should not be interpreted as an isolated comparison of tenecteplase with PCI, or as a finding about other care settings or treatment pathways that the registry record does not describe.

These statistics do not determine the best treatment for an individual, and they are not medical advice. The registry's own characterization of the study as exploratory is relevant to the strength of conclusions: the findings can inform questions for further study, but they do not establish that either strategy works or does not work.

18. Why This Trial Matters Statistically

STREAM is statistically informative because it compares two treatment pathways rather than a single intervention in isolation. One pathway begins with early fibrinolytic and antithrombotic treatment and proceeds to routine or rescue coronary intervention; the other uses standard primary PCI. That structure makes the estimand—the effect associated with assignment to a strategy—important when interpreting the result.

The primary endpoint also illustrates a common trade-off in clinical trial design. A composite can summarize several important outcomes in one measure, but its aggregate estimate does not reveal whether components are similarly affected. The registry reports the composite count but does not provide component counts for the primary endpoint, limiting interpretation of what drives the comparison.

The difference between the unadjusted risk difference and the model-based relative risk is a useful reminder that effect measures answer related questions on different scales. Both estimates need to be read with their confidence intervals. A point estimate alone can sound more conclusive than the underlying data justify, while an interval shows the range of effects that remain compatible with the analysis.

The record's exploratory framing is equally important. It identifies descriptive p-values for hypothesis generation and lists a range of secondary outcomes. Without a reported multiplicity procedure, secondary findings are especially suited to generating questions rather than settling them. The trial therefore offers a case study in combining estimates, precision, endpoint construction, population definition, and analytic intent before drawing conclusions.

Finally, the planning illustrations demonstrate why an observed effect should not be treated as a dependable design target without uncertainty. Sample-size needs rise when planning for smaller effects, and the registry's original design assumptions are not reported. That distinction separates retrospective planning illustrations from the trial's actual design rationale.

19. 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 registry record describes the trial as exploratory and its tests as descriptive for hypothesis generation. This limits the strength of confirmatory claims that can be made from the reported p-values. The record does not report a multiplicity adjustment or a testing hierarchy, despite listing multiple outcome measures and analyses.

The primary outcome is a composite of four clinical events, but component-specific primary endpoint results are not reported in the information available for this analysis. As a result, the composite estimate cannot show which events contributed to the observed group counts or whether component patterns were consistent. The record also does not provide detailed outcome adjudication information.

The registry defines the full analysis set but does not report missing-outcome counts, follow-up completeness, or a missing-data method. It names modified Poisson regression for the sponsor analysis but does not report full model specifications, including covariates or variance-estimation details. Any adjustment beyond what is explicitly described cannot be assessed.

The treatment comparison is between strategies that include multiple therapies and procedures. The results therefore do not isolate the effect of any one component. The study description focuses on patients presenting within 3 hours of symptom onset in pre-hospital or community hospital settings without a PCI facility; generalization beyond that described context is not established by the registry record.

The primary timeframe is 30 days, so these results do not provide a longer-term comparison. The safety summary gives aggregate serious adverse event figures and listed terms, but the registry record does not give group-specific rates for the listed terms or further detail about event attribution. These limitations constrain interpretation without negating the descriptive findings.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: STREAM-Strategic Reperfusion (With Tenecteplase and Antithrombotic Treatment) Early After Myocardial Infarction — NCT00623623.
• Registry record last updated 2019-08-01; status last verified 2019-06.
• Results first posted 2019-08-01.
• 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.

23. Record Summary

The registry record describes NCT00623623, a completed phase 3 randomized, parallel-group trial with no masking and recorded enrollment of 1899. It compared early tenecteplase with clopidogrel and enoxaparin followed by routine or rescue coronary intervention against standard primary PCI in patients with acute myocardial infarction within 3 hours of symptom onset.

The primary endpoint was the number of patients in the full analysis set with all-cause mortality, cardiogenic shock, congestive heart failure, or recurrent myocardial infarction within 30 days. The posted counts were 116 of 949 for tenecteplase and 135 of 948 for primary PCI. The registry's modified Poisson analysis reported a relative risk of 0.86, a 95% confidence interval from 0.68 to 1.08, and p = 0.195. The unadjusted risk-difference re-calculation also had p = 0.195 and an interval spanning zero.

The registry characterizes the study and statistical tests as exploratory, with descriptive p-values for hypothesis generation. The primary result is uncertain and does not establish equivalence or a definitive difference between strategies. Missing-data methods, multiplicity procedures, and the original sample-size justification are not reported in the registry record.