Randomized Controlled Trial of Argatroban With Tissue Plasminogen Activator (tPA) for Acute Stroke: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Randomized Controlled Trial of Argatroban With Tissue Plasminogen Activator (tPA) for Acute Stroke (NCT01464788), a randomized phase 2 study. The primary outcome is Number of Participants With 0 or 1 on Modified Rankin Scale.

2. Clinical Question

3. Trial Design

ARTSS-2 was described in its official title as a pilot, phase 2b, randomized, multi-center trial of argatroban in combination with recombinant tissue plasminogen activator for acute stroke. The registry classifies it as phase 2. Its design was randomized, single-masked, and parallel-group, with three arms: low-dose argatroban plus rt-PA, high-dose argatroban plus rt-PA, and rt-PA alone. The record lists enrollment of 90 participants.

The treatment regimens differed by argatroban infusion dose. Both argatroban arms began with a 100 micrograms/kilogram bolus and continued with an IV infusion for 48 hours; the low-dose infusion was 1 microgram/kilogram/minute and the high-dose infusion was 3 micrograms/kilogram/minute. All arms received rt-PA at 0.9mg/kg, up to a maximum dose of 90mg, as a 10% bolus followed by 90% over 1-hour. Further details about treatment administration, adherence, or treatment switching are not reported in the registry record.

The registry lists a start date of 2011-10 and a primary completion date of 2015-06-11. The current status is terminated. The reason for termination and the timing or circumstances of that decision are not reported in the registry record. The sponsor is listed as Andrew D. Barreto, MD. Although the record identifies the trial as single-masked and says that study personnel assessing the primary outcome were blinded to treatment, it does not specify which other trial participants or personnel were masked.

The arm counts in the posted primary outcome are 30 for low-dose argatroban plus rt-PA, 31 for high-dose argatroban plus rt-PA, and 29 for rt-PA alone. One high-dose participant did not receive argatroban and another was lost to follow-up; the registry states that all 31 enrolled participants in that arm were included in the intention-to-treat analysis. Reasons for the other arm counts and fuller details of randomization, allocation concealment, and the original design assumptions 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: Patients who met the inclusion criteria received a head CT scan prior to initiation of rt-PA and the Argatroban infusion. If available, patients also underwent intracranial vessel imaging performed before or immediately after IV-tPA bolus (but before Argatroban bolus). Patients could not be randomized until after the CTA demonstrated an occlusion.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: One patient in the high dose group did not receive Argatroban and another in the high dose group was lost to follow up; however, an intention-to-treat analysis was used and all 31 enrolled patients were included.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With 0 or 1 on Modified Rankin Scale, measured over 90 days.

The primary endpoint counts participants who had a modified Rankin Scale score of 0 or 1 at 90 days. The registry describes this as excellent functional outcome. The modified Rankin Scale is a measure of disability, so the endpoint identifies participants meeting the stated threshold rather than describing every level of disability or the full distribution of scores. A participant is counted once as meeting or not meeting the threshold for this binary outcome.

A binary endpoint can be summarized as a count and a proportion within each randomized arm. The registry reports 9 of 30 participants in the low-dose argatroban plus rt-PA arm, 10 of 31 in the high-dose arm, and 6 of 29 in the rt-PA-alone arm as having a score of 0 or 1. These counts make the outcome readily interpretable as the number meeting the threshold in each group. They do not, by themselves, show the participants’ scale scores outside that threshold or changes over time.

The endpoint is relevant to the stated trial aim because the brief summary describes estimating treatment benefit in terms of improvement in disability among stroke patients receiving rt-PA. Assessors were described as blinded to treatment, which is pertinent because outcome assessment can involve judgment. The registry does not report further details about assessment procedures, the handling of missing outcome values beyond the stated intention-to-treat inclusion, or whether any additional disability measure was designated as a primary endpoint.

7. Results Posted in the Registry

• High Dose Argatroban + Rt-PA: 32.3% (10/31) versus 30.0% (9/30) on control; risk difference 2.3 percentage points (95% CI -21.0 to 25.5), risk ratio 1.08 (0.51–2.27); unadjusted two-proportion z-test p = 0.849.
• Rt-PA (Alteplase): 20.7% (6/29) versus 30.0% (9/30) on control; risk difference -9.3 percentage points (95% CI -31.4 to 12.7), risk ratio 0.69 (0.28–1.69); unadjusted two-proportion z-test p = 0.412.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is binary: each participant either met the specified modified Rankin Scale threshold at 90 days or did not. Counts and proportions are natural summaries for this type of endpoint. Comparisons can be expressed on an absolute scale as a risk difference, or on relative scales as a risk ratio or an odds ratio. The reported pairwise estimates are unadjusted recalculations from the posted arm summaries.

What does a risk difference express?

A risk difference subtracts the event proportion in one group from the event proportion in the comparison group, with the order of groups determining the sign. A value above zero means the endpoint proportion is higher in the named comparison arm than in the reference arm; a value below zero means it is lower. Unlike a relative measure, a risk difference is expressed on the absolute probability scale. Its interpretation still depends on the definition of the event and which group is treated as the reference.

For high-dose versus low-dose argatroban plus rt-PA, the reported risk difference is 0.023. For rt-PA alone versus low-dose argatroban plus rt-PA, it is -0.093. These are the reported recalculated estimates, not adjusted treatment effects. The comparisons are based on the observed counts and do not account for covariates or other adjustments that are not described in the record.

How do risk ratios and odds ratios differ?

A risk ratio compares the probability of the endpoint in one group with its probability in another. A value of 1 corresponds to equal probabilities on that scale; values on either side describe relative increases or decreases. An odds ratio compares odds rather than probabilities. When outcomes are not rare, odds and probabilities can differ appreciably, so the odds ratio should not be read as though it were a risk ratio. The reported relative estimates complement the risk differences but do not remove the uncertainty indicated by their intervals.

The risk ratio for high-dose versus low-dose argatroban is 1.075, and its interval is 0.509 to 2.271. The corresponding odds ratio is 1.111, with an interval of 0.376 to 3.288. For rt-PA alone versus low-dose argatroban, the risk ratio is 0.690, with an interval of 0.281 to 1.693; the odds ratio is 0.609, with an interval of 0.185 to 2.002. Those intervals are broad, and each includes 1, the value that represents no relative difference on the respective scale.

What does a confidence interval communicate?

A confidence interval describes uncertainty in an estimate under the statistical method used. It gives a range of effect values compatible with the observed data and method; it is not a list of effects that are equally likely, and it does not state the probability that the true effect lies in the particular interval. Narrow intervals indicate greater precision than wide intervals, all else being equal. Here the intervals around the primary comparisons are wide, reflecting substantial uncertainty in the estimated effects. The confidence level used for these intervals is not reported in the registry record.

What is the role of a p-value?

A p-value quantifies how unusual the observed data, or data more extreme, would be under a specified null model and calculation. It is not the probability that the null hypothesis is true, and it does not measure clinical importance. The reported p-values, 0.849 and 0.412, are unadjusted recalculations from the posted summary counts. The record does not report a prespecified analysis plan, testing hierarchy, or multiplicity adjustment for these comparisons. They should therefore be interpreted as unadjusted statistical summaries rather than as definitive tests of all trial hypotheses.

What does intention-to-treat mean?

An intention-to-treat approach analyzes participants according to the randomized group, regardless of whether they received the assigned treatment as planned. The registry explicitly states that all 31 participants enrolled in the high-dose group were included, including one who did not receive argatroban and one who was lost to follow-up. The record does not explain how the missing outcome for the participant lost to follow-up was handled in the endpoint count. Inclusion in an intention-to-treat population does not, by itself, tell the reader whether missing data were imputed, treated as not meeting the endpoint, or addressed in another way.

General statistical context: randomization is intended to make treatment groups comparable on average, while masking can reduce some forms of bias in assessment or care. The registry records randomized allocation and single masking, and specifically states that personnel assessing the primary outcome were blinded to treatment. It does not provide enough detail to assess allocation concealment or identify all masked parties. No model-based adjustment or covariate analysis is described in the registry record.

10. Confidence Intervals and Precision

The precision of an estimate is conveyed by the range of values that remain compatible with the data under the method used. For the primary endpoint, the high-dose versus low-dose risk difference is 0.023, with a confidence interval from -0.210 to 0.255. Because the interval extends on both sides of zero, the data are compatible with differences in either direction, as well as values near no difference. The point estimate alone does not capture this uncertainty.

For rt-PA alone versus low-dose argatroban plus rt-PA, the risk difference is -0.093, with a confidence interval from -0.314 to 0.127. This interval also crosses zero and covers a wide range of possible differences. Neither comparison is estimated with enough precision to settle the size or direction of the underlying effect from these posted counts alone.

The relative-effect intervals reinforce the same caution. The high-dose versus low-dose risk ratio interval runs from 0.509 to 2.271, and its odds ratio interval runs from 0.376 to 3.288. For rt-PA alone versus low-dose argatroban, the risk ratio interval runs from 0.281 to 1.693, and the odds ratio interval runs from 0.185 to 2.002. These ranges include substantially different relative effects and include the no-difference value of 1.

The small arm counts—30, 31, and 29—are relevant context for the width of the intervals. A confidence interval does not correct for limitations in trial conduct, missing data, or an analysis plan that is not described. The confidence level used for the intervals and the trial’s original precision targets are not reported in the registry record. The intervals should be read as uncertainty around the recalculated comparisons, not as proof that clinically important differences are absent.

11. 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 sample-size illustrations use the observed effect and describe hypothetical planning under stated assumptions; they do not report the trial’s original sample-size calculation. For these binary outcomes, the figures are participants per arm, not total events. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect equal to the stated fraction of the observed effect. The trial’s own original design assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm for a design with an 80% chance of detecting the specified effect under those assumptions, if that effect were the true one. The 90% power column describes the corresponding planning number for a 90% chance. These are planning targets, not guarantees about the results of a future study. They are also not observed power: observed power is not a useful substitute for interpreting the estimated effect and its confidence interval after the trial.

The rows for planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect show how the calculation changes when the assumed effect is reduced. Smaller effects need more participants to detect under otherwise unchanged assumptions. In the high-dose versus low-dose comparison, the planning figures are much larger than the enrollment in the trial, especially when planning around a fraction of the observed effect. The rt-PA-alone comparison has its own illustrations based on its observed effect. These figures should be treated as planning examples, not as a judgment about the adequacy of the trial’s original design, which is not described in the registry record.

12. 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 = 0.300; treatment proportion p₁ = 0.323; average p̄ = 0.311.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6548 = 1.2834.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6546 = 0.5509.
4. n = (1.2834 + 0.5509)² / (0.0226)² = 6,599.2, rounded up to 6,600 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 <- 0.3                          # control proportion
p1 <- 0.32258064516129                          # 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(1464788)
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 = 6,600 (80% power), 8,834 (90% power)
 75% of observed effect: n per arm = 11,675 (80% power), 15,628 (90% power)
 50% of observed effect: n per arm = 26,134 (80% power), 34,986 (90% power)
power.prop.test: 6599.2 
Simulated power at n = 6,600 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.

13. Multiplicity, Missing Data and Analysis Populations

The registry record lists two outcome measures and zero posted analyses. It does not report a statistical analysis plan, a prespecified hierarchy of outcomes, or a multiplicity procedure. The primary outcome is identified, but the status and planned role of any other outcome measure are not described in detail. The pairwise p-values available for the primary outcome are unadjusted recalculations, so they do not account for testing across multiple comparisons or outcomes.

There are three randomized arms, and the posted pairwise estimates compare high-dose argatroban plus rt-PA with low-dose argatroban plus rt-PA, and rt-PA alone with low-dose argatroban plus rt-PA. The registry record does not report a multiplicity adjustment for these comparisons. When multiple comparisons are examined, the chance of at least one apparently unusual result can differ from the chance associated with a single prespecified test. Without a reported testing strategy, the p-values should be treated as descriptive, unadjusted summaries.

The primary outcome description states that one high-dose participant did not receive argatroban and another was lost to follow-up, but that all 31 enrolled participants in the high-dose arm were included in an intention-to-treat analysis. The outcome counts are reported as 9 of 30, 10 of 31, and 6 of 29 across the low-dose, high-dose, and rt-PA-alone arms. The record does not explain the specific handling of the participant lost to follow-up or give missing-outcome counts and methods for the other arms.

The registry therefore identifies the analysis population for the high-dose arm but does not provide enough information to assess the missing-data method. It does not report whether a sensitivity analysis was conducted, whether outcome values were imputed, or how missing outcomes were classified. Randomized assignment, treatment received, and availability of an outcome measurement are distinct pieces of information; the record does not fully describe how they were reconciled in the final analysis. The sponsor’s planned analysis population and analysis methods are not reported in the registry record.

14. 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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: nervous system disorders (36.7%), electrolyte imbalance (17.2%), death (17.2%), respiratory, thoracic and mediastinal disorders (16.7%), skin and subcutaneous tissue disorders (10.0%).

The adverse-event summaries distinguish serious adverse events, other adverse events, and deaths. The serious adverse event counts are 15 of 30 participants in the low-dose argatroban plus rt-PA arm, 15 of 31 in the high-dose arm, and 14 of 29 in the rt-PA-alone arm. The other adverse event counts are 27 of 30, 25 of 31, and 20 of 29, respectively. The recorded death counts are 5, 3, and 5, respectively. These are counts of participants in the posted group summaries and should be read in the context of each arm’s stated denominator.

The registry also lists serious-event terms by percentage, including Nervous system disorders at 36.7%, Electrolyte imbalance at 17.2%, Death at 17.2%, Respiratory, thoracic and mediastinal disorders at 16.7%, Skin and subcutaneous tissue disorders at 10.0%, and Infections and Infestations at 10.0%. These categories describe reported event terms; the registry information does not establish that an event was caused by a particular treatment. The record does not provide further detail about event severity within categories, timing, recurrence, or how attribution was determined. Arm counts are descriptive and do not by themselves establish comparative safety.

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

The posted results describe how many participants in each randomized group met the primary endpoint at 90 days and give unadjusted recalculated comparisons. The endpoint counts were 9 of 30 for low-dose argatroban plus rt-PA, 10 of 31 for high-dose argatroban plus rt-PA, and 6 of 29 for rt-PA alone. The risk-difference estimates are 0.023 for high-dose versus low-dose argatroban and -0.093 for rt-PA alone versus low-dose argatroban. Their confidence intervals are broad and include zero, indicating that the observed data do not precisely establish the direction or size of either comparison.

These results do not establish that the groups are equivalent. A comparison that is statistically uncertain is not the same as evidence that any difference is absent; the intervals allow a range of possible effects. Conversely, the point estimates and observed counts do not by themselves establish a treatment benefit or harm. The unadjusted p-values, 0.849 and 0.412, are not probabilities that the treatments are identical or that a result occurred by chance.

The trial was described as a pilot phase 2b study and enrolled 90 participants across three arms. Such a study can contribute information about the observed outcomes and the uncertainty of comparisons, but these posted data do not resolve the broader question of treatment effects. The record is terminated, and the reason is not reported. The original sample-size justification and detailed analysis plan are also not reported, limiting assessment of the intended inferential framework.

The safety summaries likewise report event counts and terms, but do not establish causation. The primary outcome assessment was stated to be conducted by personnel blinded to treatment. That feature is relevant to assessment, yet the registry does not identify all masked parties or provide complete details on outcome missingness. Interpretation should therefore stay close to what is documented: observed counts, recalculated comparisons, and substantial uncertainty, without extending these data into a definitive claim about treatment effectiveness or safety.

16. Why This Trial Matters Statistically

This trial illustrates why an observed difference and a precise estimate are not the same thing. The arm-level outcome counts differ, but the recalculated risk differences have broad confidence intervals. Reporting the estimate together with its interval makes clear that the data support a range of possible effects rather than a single settled value. For readers assessing a small randomized study, that distinction is essential: the numerical point estimate is not the whole statistical result.

The three-arm structure also matters. The primary comparison estimates posted in the results use the low-dose argatroban plus rt-PA arm as the reference for both the high-dose combination and rt-PA alone. That makes the role of the comparator explicit, while also leaving other possible pairwise comparisons unreported. The record does not describe a multiplicity plan, so the unadjusted p-values should not be mistaken for a complete testing strategy.

The example also shows that relative and absolute measures answer different questions. Risk differences describe changes on an absolute scale, whereas risk ratios and odds ratios express relative comparisons. Their estimates and intervals should be interpreted on their respective scales, not substituted for one another. Here, uncertainty is apparent across all these measures.

Finally, intention-to-treat terminology does not by itself answer every question about missing outcomes or treatment received. The record states that all high-dose participants were included, while also noting that one participant did not receive argatroban and one was lost to follow-up. The unspecified handling of the missing outcome limits independent assessment of the analysis. Taken together, the trial is statistically instructive because it connects design, endpoint definition, comparator choice, missingness, and precision, while underscoring the boundary between descriptive trial results and firm conclusions.

17. 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 marks the trial as terminated, but does not report why it was terminated or whether the termination affected recruitment, follow-up, or the planned analysis. Enrollment is listed as 90, and the primary outcome counts are available for three arms, but recruitment targets and the original sample-size justification are not reported. The planning illustrations based on the observed effects cannot fill that gap because they are not the trial’s original design calculation.

Eligibility criteria beyond the diagnosis of ischemic stroke and treatment with rt-PA are not reported in the registry record. Baseline characteristics, the balance of prognostic factors across arms, and details of random sequence generation and allocation concealment are also not reported. The record says the study was single-masked and that outcome assessors were blinded, but does not specify who else was masked or describe how masking was maintained.

The primary outcome is a threshold count at 90 days. The registry does not report the full modified Rankin Scale distribution, supporting outcome summaries, or assessment details beyond the use of blinded study personnel. It notes one high-dose participant lost to follow-up but does not state the precise method used for the missing endpoint value. This leaves uncertainty about how missingness may affect the reported arm counts.

The record lists two outcome measures but posts no analyses. It does not provide a full statistical analysis plan, adjusted estimates, a multiplicity strategy, or details of sensitivity analyses. The reported pairwise estimates are unadjusted recalculations, and the confidence level for the intervals is not reported. These limitations matter when interpreting both the p-values and the breadth of uncertainty around the effect estimates.

Adverse-event summaries give group counts and selected serious-event terms, but further details on event timing, attribution, recurrence, and ascertainment are not reported. These gaps constrain interpretation of both efficacy-related and safety-related results. They do not invalidate the observed summaries, but they limit what can be concluded from the registry record alone.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Randomized Controlled Trial of Argatroban With Tissue Plasminogen Activator (tPA) for Acute Stroke — NCT01464788.
• Registry record last updated 2017-05-11; status last verified 2017-03.
• Results first posted 2017-05-11.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

ARTSS-2 was a randomized, single-masked, parallel-group phase 2 trial described as a pilot phase 2b study of argatroban added to rt-PA for acute stroke. The registry reports enrollment of 90 and a terminated status. Participants were assigned to low-dose argatroban plus rt-PA, high-dose argatroban plus rt-PA, or rt-PA alone.

The primary outcome was the number of participants with a modified Rankin Scale score of 0 or 1 at 90 days. Counts were 9 of 30 in the low-dose arm, 10 of 31 in the high-dose arm, and 6 of 29 in the rt-PA-alone arm. The unadjusted recalculated risk differences were 0.023 for high-dose versus low-dose argatroban and -0.093 for rt-PA alone versus low-dose argatroban; both estimates have broad confidence intervals spanning zero.

The registry states that all 31 high-dose participants were included in an intention-to-treat analysis, despite one not receiving argatroban and one being lost to follow-up. Details of missing-outcome handling, the original sample-size justification, and the full statistical analysis plan are not reported in the registry record. The results are best read as observed trial summaries with substantial uncertainty.