Effect of Cosopt Versus Combigan on Retinal Vascular Autoregulation in Primary Open Angle Glaucoma (POAG): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Cosopt Versus Combigan on Retinal Vascular Autoregulation in Primary Open Angle Glaucoma (POAG) (NCT00824824), a randomized na study. The primary outcome is Presence of Retinal Vascular Dysregulation (RVD).

2. Clinical Question

3. Trial Design

The registry identifies this as a completed study sponsored by Massachusetts Eye and Ear Infirmary. The recorded start is 2009-01 and the primary completion is 2012-08. The phase is listed as NA. Enrollment is 21, and the condition is glaucoma. The allocation field says randomized and the masking field says quadruple, but the model field says single-group. The record does not explain how those design labels relate to one another.

The study summary describes assessing the retinal circulation during a posture change from sitting to lying down. The registry says that 21 participants were tested for RVD after 6 weeks of timolol treatment. Seven participants with RVD were randomized to dorzolamide-timolol or brimonidine-timolol; 14 had normal autoregulation. Treatment descriptions specify use in both eyes for 6 weeks. The outcome timeframe is 6 weeks post treatment.

The record lists 7 participants in the dorzolamide-timolol arm and 6 in the brimonidine-timolol arm for the outcome. It also says one participant was removed from analysis in the Brimonidine-Timolol arm, but the available description ends before the reason is stated. The registry does not provide a fuller participant-flow account or explain how the listed enrollment, randomized subset, and analyzed arm counts fit together.

Although the title presents a comparison between the two treatments, the posted outcome data do not include a reported inferential comparison. The registry records no analyses. The protocol’s statistical plan, sample-size justification, intended analysis populations, and interim-analysis approach 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: Prior to the initial testing visit, subjects ceased using their current IOP-lowering medications and were run in for 6 weeks on timolol 0.5% BID OU. After this 6 week period baseline measurements were taken and 7 were identified with retinal vascular dysregulation. These 7 patients were then randomized into one of two treatment arms.

Overall Study

Analysis population for the primary outcome, as defined in the registry: 21 participants were tested for RVD after 6 weeks of timolol treatment. Of the 21 participants who were tested, 7 had RVD and were randomized to Dorzolamide-Timolol and Brimonidine-Timolol; 14 had normal autoregulation. One participant was removed from the analysis in the Brimonidine-Timolol arm due to technical difficulties with equipment.

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: There are 14 subjects in this study whop have PAOG without RVD. There are 7 subjects with POAG with RVD. The 14 subjects without RVD did not continue in the two treament arms of this study.

6. Primary Endpoint

The registry lists the primary outcome as Presence of Retinal Vascular Dysregulation (RVD), measured over 6 weeks post treatment.

The primary endpoint is the presence of retinal vascular dysregulation, recorded as a number of participants, with assessment at 6 weeks post treatment. The registry defines RVD using the change in retinal blood flow between reclining for 30 minutes and a baseline measurement while seated. Participants whose posture-related change fell outside the specified normal range were classified as having RVD.

The threshold came from a previous study of healthy subjects. The registry reports a mean change of +6.5% ± 12% in that group and defines the normal range as ± 2 standard deviations around the mean, expressed as 6.5% ± 24.0%, or -17.5% to +30.5%. A change outside that range was treated as evidence of dysregulation for the study’s classification.

This endpoint turns a continuous blood-flow change into a binary classification: RVD present or not present. Such a classification can make counts easy to interpret, but it does not show how far an individual’s measurement was from the boundary or describe the full distribution of blood-flow changes. The registry’s posted results report participant counts rather than individual measurements or a continuous treatment contrast.

The study’s screening and treatment stages matter to interpretation. The record says participants were assessed after timolol, and those with RVD were randomized to a treatment. The primary endpoint is named as RVD presence after treatment, but the record does not give additional detail about repeated measurements, the exact handling of values at a boundary, or the measurement procedures beyond the posture comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a binary endpoint show?

A binary endpoint classifies each participant into one of two outcome states. In this study the states are presence or absence of RVD, based on whether the measured posture-related blood-flow change fell outside the defined normal range. The posted result is expressed as participant counts in each treatment arm. Counts communicate how many recorded participants had the endpoint, but discard information about the size and direction of each person’s underlying blood-flow change.

For a treatment comparison, binary outcomes can be summarized with an absolute difference in event proportions, a relative measure, or a statistical test, often accompanied by a confidence interval. Each approach answers a somewhat different question. An absolute difference describes the contrast in event frequency; a relative measure compares frequencies proportionally; and a test evaluates compatibility with a specified no-difference model. None of these comparative results is reported for this study. The record marks the two-arm binary comparison as not estimable, so a recalculated comparison should not be substituted for a sponsor analysis.

Why does the comparison lack a posted estimate?

Estimating a treatment contrast requires a defined comparison and sufficient outcome information in both groups. This registry record gives the arm counts but does not provide an estimable comparative result. The analysis count is zero, and the record does not include a sponsor-reported test or effect estimate. Therefore, the descriptive outcome counts are the extent of the inferential information available in the registry record.

A confidence interval is a range of values generated by a statistical method to represent uncertainty around an estimate. It is not a guarantee that the true effect lies in the interval, and its interpretation depends on the method and assumptions. Since no comparative estimate or interval is reported, readers cannot use interval width to judge the precision of a treatment effect. The number of participants and the event counts suggest that the available outcome information is limited, but no new precision estimate can be derived from the registry record in this explanation.

How should the planning illustrations be understood?

The stated assumptions for planning illustrations are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The effect targeted is described as a fraction of the observed effect. Such calculations ask what sample size might be needed under chosen assumptions; they do not describe the probability that this completed study detected an effect. They are not observed power and do not replace the study’s own design rationale.

The record does not report the protocol’s original sample-size justification. It also does not report enough detail to reconstruct the assumptions behind the trial’s planned enrollment or to establish that the planning illustration corresponds to the study’s design. Any sample-size planning result should therefore be understood as an illustration based on the specified assumptions rather than evidence about the study’s actual power.

Why can over-dispersion matter?

General statistical context: some count outcomes vary more than a simple probability model expects. This is called over-dispersion. If a method assumes less variability than the data actually have, its standard errors can be too small and its confidence intervals too narrow. If variability is modeled appropriately, uncertainty estimates can better reflect the observed spread. This study’s primary endpoint is binary rather than a count of repeated events per participant, and the registry does not report an over-dispersion analysis. The concept is relevant generally but is not a reported finding about this trial.

What does randomization contribute?

In general, random allocation is intended to make treatment groups comparable with respect to both measured and unmeasured characteristics on average. It does not guarantee identical groups, especially when few participants are randomized. The registry says the seven participants with RVD were randomized to the two treatments, but its design model is labeled single-group and the available fields do not clarify the discrepancy. The record does not report the allocation method, whether allocation was concealed, or a baseline comparison of the randomized groups.

Masking can reduce the influence of treatment knowledge on participant behavior, clinical decisions, measurement, or outcome assessment. The registry lists quadruple masking but does not identify which parties were masked or describe the masking procedures. Thus, the designation is part of the record, while its practical implementation cannot be evaluated from the reported details.

What can be recalculated?

Recalculated analyses are simple, unadjusted calculations from posted summary statistics. They can be useful checks when a registry gives sufficient information, but they are not equivalent to a prespecified analysis and may not account for the trial’s design or missing-data procedures. For this endpoint, the comparison is marked not estimable, and no recalculated treatment effect is available to interpret. If a discrepancy between a recalculation and the sponsor’s own analysis were reported, the sponsor’s own analysis would take precedence; the registry record does not report a discrepancy in this case.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Dorzolamide-Timolol", "Brimonidine-Timolol")
n      <- c(7, 6)
events <- c(7, 4)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Dorzolamide-Timolol  7  7  100.0  59.0  100.0  64.6  100.0  63.1  99.7
ARM  Brimonidine-Timolol  6  4  66.7  22.3  95.7  30.0  90.3  29.0  90.1

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome measure and no secondary outcomes. It records no analyses. A multiplicity procedure for handling several outcomes, time points, or statistical tests is not reported in the registry record. With no reported analysis results, it is also not possible to determine whether additional exploratory tests were conducted or how any such tests were handled.

The primary outcome listing gives 7 participants in the dorzolamide-timolol arm and 6 in the brimonidine-timolol arm, with outcome counts of 7 and 4, respectively. The narrative says that one participant was removed from analysis in the Brimonidine-Timolol arm but does not state why; the description ends before completing that account. The registry does not report whether this participant lacked an outcome measurement, was excluded for another reason, or was handled under a particular missing-data rule. No imputation or sensitivity analysis is described.

The analysis population is not fully specified. The record does not say whether the outcome counts are based on all randomized participants, participants who completed treatment, or another population. Nor does it provide a participant flow that reconciles the 21 tested after timolol, the 7 with RVD who were randomized, and the listed arm denominators. These omissions make it difficult to assess how missing outcomes or exclusions might affect the comparison.

The protocol’s statistical plan, interim-analysis approach, and any prespecified multiplicity strategy are not reported in the registry record. No inference should be made that multiplicity was either adjusted or ignored; the record simply does not describe the relevant procedures.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 3.8% reporting threshold.

The registry reports adverse-event groups by treatment sequence: Dorzolamide-Timolol Then Brimonidine-Timolol and Brimonidine-Timolol Then Dorzolamide-Timolol. In each group, 4 participants were at risk for serious adverse events and 0 were reported as affected. For other adverse events, 4 participants were at risk in each group and 0 were reported as affected. The registry lists no serious adverse-event terms.

These figures should be read as the reported counts for the sequence groups, not as a direct adverse-event comparison between the individual treatment arms in the primary outcome. The sequence labels imply a different grouping from the outcome arms, and the record does not describe the safety analysis in enough detail to reconcile them. The number of deaths is not reported in the registry record.

A zero reported count means no events were recorded in the relevant category among the listed participants at risk. It does not establish that an event is impossible or that the treatments have equivalent safety. The small number of participants and the limited safety detail constrain what can be concluded about uncommon events, duration of risk, or differences between treatments. These figures are registry reporting, not medical advice.

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

The posted results describe RVD counts after treatment among participants selected for having RVD following timolol treatment. The registry reports 7 participants with RVD among 7 listed in the dorzolamide-timolol arm and 4 among 6 listed in the brimonidine-timolol arm. These counts are descriptive observations within the recorded groups. They do not, by themselves, establish a treatment effect or its precision.

The results do not show how much retinal blood flow changed for each participant, whether changes were close to or far from the classification thresholds, or how measurements varied across participants. The endpoint reduces a posture-related measurement to whether it falls inside or outside a defined range. A participant counted as having RVD is classified by that rule; the classification does not on its own explain the biological mechanism or the clinical importance of the measured change.

The population is restricted to participants with RVD identified after timolol treatment. The 14 participants reported to have normal autoregulation are not described as randomized to the two treatment arms. Accordingly, the posted counts do not directly answer how the treatments compare among all people with POAG, nor do they establish what would happen in a different population or under different treatment conditions.

No comparative statistical estimate or confidence interval is reported, and the registry identifies the binary comparison as not estimable. The record also lacks details about the reason for the participant removed from analysis, the full participant flow, and the statistical plan. These gaps prevent a reliable assessment of how uncertainty, missing data, and design choices affect the apparent difference in counts.

The results therefore support a limited description of the recorded endpoint counts, not a conclusion that either treatment is effective, ineffective, better, or worse. The registry record also does not establish clinical consequences from the observed classifications. Any broader interpretation would require information and analyses not reported in the record.

14. Why This Trial Matters Statistically

This study illustrates why the design and analysis record matter alongside an outcome count. A randomized treatment comparison is most informative when the population, allocation, outcome definition, analysis population, and statistical method are clearly described. Here, the registry reports randomization and two treatment arms but labels the model single-group. It describes a selected subgroup entering randomization, while the enrollment field gives the total number tested. Understanding that pathway is essential to knowing which participants the arm counts represent.

The endpoint also shows how a continuous physiological measurement can be converted into a binary outcome using a threshold derived from a healthy group. That choice can simplify reporting, but it loses information about the size of changes and their distance from the cutoff. Readers should distinguish the endpoint classification from the underlying retinal blood-flow response.

Small arm counts make transparent uncertainty reporting especially important. Counts alone cannot show the precision of a treatment contrast. An effect estimate and confidence interval, together with a clear account of exclusions and missing outcomes, would help readers assess the range of effects compatible with the data. The registry does not report those comparative results.

Finally, planning calculations and observed trial results answer different questions. A sample-size illustration based on assumptions is not evidence of observed power and is not a substitute for the trial’s original design justification. The record’s missing statistical-plan details and lack of an estimable comparison demonstrate why registry results should be interpreted in light of both what is reported and what remains unreported.

15. 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 leaves several trial-specific limitations unresolved. Although it reports randomized allocation and quadruple masking, the model is labeled single-group. The record does not explain this combination or describe allocation concealment, the randomization procedure, or which parties were masked. The treatment comparison is therefore not documented in enough detail to reconstruct all design features.

The randomized population was a subset of those tested: 7 participants had RVD after timolol treatment, while 14 had normal autoregulation. The arm outcome listing includes 7 participants for dorzolamide-timolol and 6 for brimonidine-timolol. The record does not provide a participant-flow account that explains these counts or the relationship between the randomized subset and outcome denominators.

The narrative states that one participant was removed from analysis in the Brimonidine-Timolol arm, but the reason is not reported in the registry record. Missing-data rules, analysis-population definitions, and sensitivity analyses are also not reported. These omissions limit evaluation of whether the reported outcome counts could be affected by exclusions or incomplete follow-up.

The registry gives no comparative effect estimate, confidence interval, or test, and indicates that the binary comparison is not estimable. The protocol’s statistical plan, sample-size justification, interim analyses, and multiplicity procedures are not reported in the registry record. The safety results are grouped by treatment sequence, without an explanation of how those groups relate to the outcome arms, and deaths are not reported in the registry record.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Effect of Cosopt Versus Combigan on Retinal Vascular Autoregulation in Primary Open Angle Glaucoma (POAG) — NCT00824824.
• Registry record last updated 2017-04-04; status last verified 2017-03.
• Results first posted 2016-10-28.
• 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.

19. Record Summary

NCT00824824 is a completed study sponsored by Massachusetts Eye and Ear Infirmary, with a recorded start of 2009-01 and primary completion of 2012-08. The registry lists glaucoma as the condition, enrollment of 21, phase NA, randomized allocation, quadruple masking, and a single-group model. The record describes testing participants for RVD after 6 weeks of timolol treatment; 7 with RVD were randomized to dorzolamide-timolol or brimonidine-timolol, while 14 had normal autoregulation.

The primary outcome was presence of RVD at 6 weeks post treatment. The registry reports counts of 7 among 7 in the dorzolamide-timolol arm and 4 among 6 in the brimonidine-timolol arm. It does not report an estimable statistical comparison, a confidence interval, or an analysis result. One participant was removed from analysis in the Brimonidine-Timolol arm, but the reason is not reported in the registry record. Statistical-plan details, the original sample-size justification, and a complete account of missing data and participant flow are also not reported.