This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Levo-Dexa vs. Tobra+Dexa for Prevention and Treatment of Inflammation and Prevention of Infection in Cataract Surgery (NCT03739528), a randomized phase 3 study. The primary outcome is Number of Participants Without Signs of Anterior Chamber Inflammation.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03739528 |
| Phase | Phase 3 |
| Condition | Cataract |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Single |
| Enrollment | 808 |
| Arms in the primary outcome | n/a |
| Lead sponsor | NTC srl |
| Results posted | Yes |
2. Clinical Question
Population
Adults undergoing cataract surgery, with cataract listed as the condition. The record states that the primary analysis population included randomized patients who received at least one dose of study treatment. It does not report further eligibility details in the information available for this analysis.
Intervention
Levofloxacin 5 mg/ml plus dexamethasone 1 mg/ml eye drops for 7 days, followed by dexamethasone 1 mg/ml alone for an additional 7 days. Both treatments were described as 1 drop/4 times a day.
Comparator
Tobramycin plus dexamethasone eye drops for 14 days, also described as 1 drop/4 times a day. The registry names this as the standard treatment for the trial comparison.
Primary question
At Day 15, was the proportion of participants without signs of anterior chamber inflammation under the levofloxacin–dexamethasone sequence no worse than under tobramycin–dexamethasone by more than the prespecified non-inferiority margin? The registry states a margin of 10%. The primary comparison needed to answer that question is not reported in a single reduced form in the registry record.
The study also lists outcomes concerning endophthalmitis, ocular signs and symptoms, and use of rescue therapy. Those measures broaden the clinical questions beyond the primary inflammation endpoint, but their presence does not substitute for the primary comparison. The registry record does not report numerical results for these secondary outcomes in the registry record for interpretation.
3. Trial Design
This was a Phase 3, randomized, single-masked, parallel-group study. Its official title describes an international, multicenter, randomized study with a blinded assessor. The record lists enrollment of 808 and status as completed. The treatment comparison was between a levofloxacin–dexamethasone sequence and tobramycin–dexamethasone. The record does not specify which participants were masked or how masking was maintained, beyond the single-masked designation and the blinded-assessor wording in the official title.
The experimental regimen paired levofloxacin with dexamethasone for 7 days and then continued dexamethasone alone for 7 days. The comparator combined tobramycin and dexamethasone for 14 days. Both descriptions specify 1 drop/4 times a day. These are different antibiotic regimens and different schedules for the combination treatment; the trial therefore evaluates the regimens as assigned rather than isolating the effect of a single ingredient.
The registry gives a start date of 2018-09-03 and a primary completion date of 2018-12-19. These dates describe the study timeline recorded in the registry; they do not by themselves establish when every participant was followed or when each outcome was assessed. The listed primary endpoint timeframe is Day 15.
The analysis population for the primary endpoint is described as all randomized patients who received at least one dose of study treatment, called the Full Analysis Set in the record. Enrollment and analysis-population size are not interchangeable: the record does not give the number in that population for the primary endpoint. It also does not report arm-specific counts for the primary endpoint or a participant flow account in the registry record. Further protocol details, including the randomization procedure and the identity of the masked parties, 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: Planned sample size n.800; randomized patients n.808; screened patients n.863
Overall Study
| Stage | Levofloxacin + Dexamethasone Followed by Dexamethasone | Tobramycin + Dexamethasone |
|---|---|---|
| STARTED | 403 | 405 |
| Received at Least One Treatment | 395 | 393 |
| COMPLETED | 388 | 388 |
| NOT COMPLETED | 15 | 17 |
Reasons for not completing this period:
| Reason | Levofloxacin + Dexamethasone Followed by Dexamethasone | Tobramycin + Dexamethasone |
|---|---|---|
| Not treated | 8 | 12 |
| Treatment and study discontinued | 6 | 2 |
| Treatment discontinued + study completed | 1 | 3 |
Analysis population for the primary outcome, as defined in the registry: The analysis population is represented by all randomized patients who received at least one dose of study treatment (Full Analysis Set)
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: The analysis population is represented by all randomized patients who received at least one dose of study treatment (Full Analysis Set)
| Characteristic | Unit | Levofloxacin + Dexamethasone Followed by Dexamethasone | Tobramycin + Dexamethasone | Total |
|---|---|---|---|---|
| Participants analysed | 395 | 393 | 788 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 70 | 71 | 141 |
| Age, Categorical: >=65 years | Participants | 325 | 322 | 647 |
| Age, Continuous | years | 72.12 (SD 8.55) | 71.84 (SD 8.64) | 71.98 (SD 8.59) |
| Sex: Female, Male: Female | Participants | 229 | 239 | 468 |
| Sex: Female, Male: Male | Participants | 166 | 154 | 320 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 1 | 0 | 1 |
| Race (NIH/OMB): Asian | Participants | 1 | 1 | 2 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 1 | 1 |
| Race (NIH/OMB): Black or African American | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): White | Participants | 393 | 391 | 784 |
| Race (NIH/OMB): More than one race | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 |
6. Primary Endpoint
The registry lists the primary outcome as Number of Participants Without Signs of Anterior Chamber Inflammation, measured over Day 15.
| Element | Registry entry |
|---|---|
| Result type | Count Of Participants |
| Measure of spread | not applicable |
| Unit | Participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint counts participants without signs of anterior chamber inflammation at Day 15. A participant meets the endpoint definition when the sum of the anterior chamber cell and flare scores is 0. The outcome is therefore a binary participant-level status at a specified assessment: meeting or not meeting the stated inflammation-free definition.
This measure is suited to the stated clinical question because it expresses the proportion reaching a defined postoperative inflammation status in each treatment group. A proportion comparison can be framed as a difference between the test regimen and the standard regimen. For non-inferiority, the question is not simply whether the groups differ; it is whether the test regimen’s result is not worse than the comparator by more than the chosen margin. The registry specifies a non-inferiority margin of 10% and states hypotheses using the test-minus-standard contrast.
Any endpoint based on a score threshold depends on how the score is assessed and how missing or unevaluable assessments are handled. The record defines the threshold but the missing-data procedure is not reported in the registry record. Nor does the available record give the numbers meeting the definition in either arm, the estimated difference, or its uncertainty. The result therefore cannot be interpreted as evidence for or against the non-inferiority objective from the endpoint definition alone.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry identifies a non-inferiority objective: compare the proportion without signs of anterior chamber inflammation in the test and standard treatment groups, using a margin of 10%. It states the contrast as test treatment minus standard treatment. Under that framing, the non-inferiority claim would depend on whether the data rule out a disadvantage for the test treatment that reaches or exceeds the specified margin. The record’s stated hypotheses make the direction of the comparison important.
However, the primary outcome is not reduced to one comparison in the registry record: the listed results include several categories or time points. A single arm-level count, treatment contrast, confidence interval, or p-value for the primary non-inferiority test is not reported in the registry record. Without those quantities, it is not possible to establish whether the criterion was met, estimate the size of any treatment difference, or describe how precisely that difference was estimated. The non-inferiority objective and margin describe the intended question, not the outcome of the test.
That distinction matters because a non-inferiority conclusion cannot be inferred merely from a nonsignificant conventional test for difference. A precise estimate compatible with the margin and an imprecise estimate spanning materially different possibilities would support very different interpretations, even if neither produced a conventional superiority result. The registry record does not provide the estimate and interval needed to make that distinction. No conclusion about comparative efficacy follows from the design description alone.
The record lists secondary outcomes that include participants with endophthalmitis, participants without signs of anterior ocular chamber inflammation, conjunctival hyperemia, total ocular symptoms score, ocular pain or discomfort, and use of rescue therapy. It reports 13 outcome measures, but the registry entries do not give numerical comparisons for those measures. Their listing indicates intended or recorded assessments; it does not show whether events occurred, whether groups differed, or how estimates were distributed.
Safety information is more limited but includes serious adverse-event counts, other adverse-event counts, and deaths. The registry reports 4 serious adverse events among 388 at risk in the levofloxacin–dexamethasone sequence group, and 2 among 388 at risk in the tobramycin–dexamethasone group. Other adverse events are recorded as 0 among 395 and 0 among 393, respectively. Deaths are recorded as 1 and 0. These are descriptive registry counts. They do not establish that treatment caused an event, and the record does not provide enough information to assess event timing, clinical context, or comparative uncertainty.
Consequently, the most defensible interpretation is about reporting, not treatment performance: the registry documents a completed randomized trial and its intended non-inferiority framework, but the primary result is not presented as a single interpretable comparison. The observed treatment effect and its precision are not reported in the registry record in the form needed to answer the central statistical question.
9. Statistical Methods Explained
The study’s primary statistical aim is non-inferiority. General explanation: a non-inferiority design asks whether an experimental regimen is sufficiently close to a comparator on a defined outcome, rather than asking only whether one regimen is superior. The margin sets the largest disadvantage that the design treats as acceptable for that question. Here, the record states a margin of 10% and defines the contrast as the proportion without signs of anterior chamber inflammation in the test group minus the corresponding proportion in the standard group.
How to read the non-inferiority hypotheses
The registry writes the null hypothesis as H0: πT - πS ≤ -Δ and the alternative as H1: πT - πS > -Δ. In this notation, πT and πS refer to the endpoint proportions for test and standard treatment, and Δ is the stated margin. General explanation: the null represents a difference at or beyond the prespecified boundary in the unfavorable direction; the alternative places the difference on the side compatible with non-inferiority. The conclusion depends on the estimate and its uncertainty relative to that boundary. The record does not report the primary estimate or interval needed to apply this rule.
Why the endpoint scale matters
The endpoint is a count of participants satisfying a binary definition at Day 15. The underlying assessment uses the sum of cell and flare scores, with 0 defining absence of signs for this endpoint. A proportion difference is an absolute contrast between groups, not a ratio. General explanation: for a binary outcome, the interpretation of any effect estimate depends on the event definition, the analyzed population, and the time point. A clear definition supports consistent interpretation, but it does not reveal the observed outcome counts. Those counts and the treatment contrast are not reported in a single primary comparison in the registry record.
Confidence intervals describe a range of values compatible with the data and statistical model, while also reflecting sampling uncertainty. General explanation: in a non-inferiority analysis, the location of the interval in relation to the margin is central. An interval that does not resolve the margin question cannot establish non-inferiority, even if a point estimate appears favorable. The registry information does not include a confidence interval for the primary contrast. A p-value alone, even if one were reported, would not replace examination of the estimate, interval, margin, and analysis population.
What randomization and masking contribute
Random assignment is intended to create treatment groups that are comparable on average with respect to measured and unmeasured baseline factors. This supports a causal interpretation when the assignment is preserved and follow-up and analysis are handled appropriately. The record identifies randomization and a parallel-group model, but does not describe the sequence-generation method, allocation concealment, or how departures from assigned treatment were handled. Those details are not reported in the registry record. The single-masked designation and blinded-assessor description may reduce some forms of assessment or performance bias, but the parties masked and the practical masking procedures are not described.
General explanation: over-dispersion is relevant when observed variability exceeds what a statistical model assumes. For count outcomes, that can lead to uncertainty estimates that are too narrow if the model does not account for extra variability. The primary endpoint is described as a participant count meeting a binary criterion, not as a continuous mean with a dispersion measure. The registry record does not specify the model used for the primary comparison, so no claim about over-dispersion or model adequacy can be made for this analysis.
Analysis population and missing observations
The primary analysis population is defined as randomized patients who received at least one dose, identified as the Full Analysis Set. General explanation: a defined analysis population determines which randomized participants contribute to the treatment estimate. In non-inferiority work, population choices and deviations from assigned treatment can matter because exclusions or treatment changes may affect the apparent similarity between groups. The record does not report a per-protocol population, protocol-deviation rules, handling of missing Day 15 assessments, or sensitivity analyses. It is therefore not possible to assess whether conclusions were robust to alternative analysis populations or missing-data assumptions.
Planning calculations are not observed evidence
The registry’s sample-size assumptions describe a planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. General explanation: such calculations illustrate a hypothetical design requirement under specified assumptions; they do not measure the probability that this completed trial found an effect, and they are not observed power. The registry also distinguishes between calculations expressed as participants per arm and time-to-event calculations expressed as total events. No sample-size illustration values are reported in the registry record, and the primary result is not reduced to a single observed effect for interpretation.
The record reports 13 outcome measures and 0 analyses. That entry does not supply the statistical model, confidence interval method, adjustment strategy, or exact testing procedure for the primary non-inferiority comparison. Those details are not reported in the registry record. The written hypotheses and margin establish the conceptual test, but not the analysis implementation or its result.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists 13 outcome measures, including the primary inflammation endpoint and secondary outcomes concerning endophthalmitis, anterior ocular chamber inflammation, conjunctival hyperemia, ocular symptoms, pain or discomfort, and rescue therapy. The record also indicates 0 analyses. The relationship between the outcome measures and the reported results is therefore not clear from the registry information available for this interpretation.
Multiplicity arises when several outcomes, time points, subgroups, or statistical tests are examined. General explanation: repeated testing can increase the chance of at least one apparently unusual result, so a statistical plan may specify which outcome is primary and how secondary analyses are interpreted or adjusted. The record identifies a primary endpoint, but the primary results are not reduced to one comparison because several categories or time points are present. A multiplicity strategy, hierarchy for secondary outcomes, or adjustment procedure is not reported in the registry record. It would be inappropriate to assume that testing was adjusted or unadjusted.
The primary analysis population includes randomized participants who received at least one dose. The number analyzed for this endpoint, missing Day 15 assessments, exclusions, and methods for handling missing data are not reported in the registry record. These omissions limit assessment of how closely the analyzed population followed random assignment and whether missing observations could affect the endpoint estimate.
It is also not reported whether a per-protocol analysis or sensitivity analyses were conducted. General explanation: comparing results across prespecified analysis populations can be informative in a non-inferiority study, because different deviations and exclusions can influence apparent similarity. Without those results and the associated rules, robustness cannot be assessed. The available information identifies the intended population and endpoint, but does not establish the completeness or stability of the primary analysis.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Levofloxacin + Dexamethasone Followed by Dexamethasone | 4 | 388 | 1.0% |
| Tobramycin + Dexamethasone | 2 | 388 | 0.5% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Levofloxacin + Dexamethasone Followed by Dexamethasone | 1 of 395 |
| Tobramycin + Dexamethasone | 0 of 393 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Levofloxacin + Dexamethasone Followed by Dexamethasone | 0 of 395 |
| Tobramycin + Dexamethasone | 0 of 393 |
Most frequent serious adverse event terms
| Event term | Organ system | Levofloxacin + Dexamethasone Followed by Dexamethasone | Tobramycin + Dexamethasone |
|---|---|---|---|
| Fall | Injury, poisoning and procedural complications | 2 of 388 | 1 of 388 |
| Myocardial Infarction | Cardiac disorders | 1 of 388 | 1 of 388 |
| Retinal Detachment | Eye disorders | 1 of 388 | 0 of 388 |
Serious events with the highest arm-level rate: fall (0.5%), myocardial infarction (0.3%), retinal detachment (0.3%).
The registry reports serious adverse-event counts separately for the two treatment groups. In the levofloxacin plus dexamethasone followed by dexamethasone group, 4 participants with serious adverse events are reported among 388 at risk. In the tobramycin plus dexamethasone group, 2 are reported among 388 at risk. The record also gives other adverse-event counts of 0 among 395 and 0 among 393, respectively. Death counts are reported as 1 and 0. These figures are descriptive and should be read with their stated denominators; the at-risk counts differ by outcome.
The registry lists Fall at 0.5%, Myocardial Infarction at 0.3%, and Retinal Detachment at 0.3% among the top serious terms. The information available does not assign those term percentages to a particular treatment group or provide further event-level context. The percentages should not be converted into group-specific counts or interpreted as causal effects.
General explanation: adverse-event counts alone do not establish whether an event was caused by a treatment. Interpretation also depends on timing, seriousness definitions, participant follow-up, and clinical context. Those details are not reported in the registry record. The reported death counts likewise do not indicate cause or attribution. No comparative safety conclusion can be drawn from these counts alone.
12. What the Result Does, and Does Not, Mean
The record establishes that a randomized, parallel-group Phase 3 study compared two postoperative eye-drop regimens in adults undergoing cataract surgery. It also defines a Day 15 primary endpoint and a non-inferiority margin of 10%. These elements describe the question the study was designed to address.
They do not establish the answer. The primary endpoint results are not reduced to one comparison in the registry record because several categories or time points are present. No single treatment difference, confidence interval, or p-value for the primary non-inferiority test is reported in the registry record for interpretation. As a result, the record does not support a conclusion that the study did or did not meet its non-inferiority objective.
Nor does the registry information permit a judgment about the size or precision of any difference in inflammation status. A trial can be completed without its results being presented in a form that answers every statistical question. The stated endpoint and margin cannot stand in for the observed data, and the enrollment total cannot be used to infer the outcome.
The secondary endpoint list indicates additional topics the study assessed, but numerical findings for those outcomes are not reported in the available information. The safety counts offer limited descriptive information, but do not establish causation or a treatment-related difference. The absence of a reported comparison should not be interpreted as evidence that groups were the same.
General explanation: statistical evidence is most informative when the estimate, its uncertainty, the prespecified decision boundary, and the analysis population are considered together. For this trial, the record supplies the margin and analysis-population definition but not the primary estimate and uncertainty. The appropriate conclusion is therefore limited: the registry documents a non-inferiority study and some safety counts, while the central comparative result remains not reported in the registry record in an interpretable single-comparison form.
13. Why This Trial Matters Statistically
This trial illustrates why statistical interpretation begins with the target question rather than with a design label or enrollment total. The registry names a non-inferiority objective and specifies the direction of the test, the endpoint definition, and a margin of 10%. These details are essential because they determine what evidence would be relevant to the stated claim. Without the observed contrast and uncertainty, however, the design framework alone cannot answer the question.
The endpoint is a participant-level binary status at Day 15, defined using the sum of cell and flare scores. That makes the treatment contrast conceptually direct: compare the proportions meeting the inflammation-free definition. Yet the registry record flags several categories or time points and does not reduce the primary outcome to one comparison. This highlights a practical reporting issue: results must be organized in a way that lets readers align the estimate with the endpoint, time point, analysis population, and non-inferiority margin.
The trial also shows why allocation, masking, and analysis-population details should be interpreted separately. Randomization and a blinded assessor are recorded, while more detailed operational and analytic procedures are not reported. A reader can recognize the design strengths named in the record without assuming unreported safeguards or procedures.
Finally, the limited safety counts demonstrate the difference between event description and comparative inference. Counts and denominators are useful context, but they do not establish causation or quantify uncertainty. Overall, the statistical importance of the record lies in distinguishing the intended design from the evidence needed to assess its primary objective, and in identifying where reporting detail is insufficient for that assessment.
14. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The most consequential limitation for interpreting this trial is that the primary endpoint is not presented as one treatment comparison. The registry notes several categories or time points, but the registry entries available do not give arm-specific primary endpoint counts or a single effect estimate. A confidence interval and p-value for the non-inferiority test are also not reported in the registry record. This prevents evaluation against the stated 10% margin.
The record describes the primary analysis population as randomized participants who received at least one dose, but does not give the analyzed number in each arm. Missing assessments, exclusions, protocol deviations, treatment discontinuations, and handling of incomplete Day 15 data are not reported in the registry record. This makes it difficult to assess participant flow or the possible influence of missing outcome data.
Although the study is described as randomized, single-masked, parallel-group, and with a blinded assessor, details about the randomization procedure, allocation concealment, masking implementation, and who remained masked are not reported. The statistical model and variance or interval-estimation procedure are also not reported. The registry lists 13 outcome measures and 0 analyses, but does not give a multiplicity plan or a hierarchy for interpreting secondary outcomes.
The safety information consists of counts and selected serious-event terms, without the event-level detail, timing, attribution, or comparative uncertainty needed for fuller interpretation. The record does not specify whether deaths were considered related to treatment. These limitations restrict the assessment of both the primary efficacy question and the safety context; they do not demonstrate that the study procedures or results were inadequate.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Levo-Dexa vs. Tobra+Dexa for Prevention and Treatment of Inflammation and Prevention of Infection in Cataract Surgery — NCT03739528.
- Registry record last updated 2020-08-18; status last verified 2020-08.
- Results first posted 2020-08-18.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The registry record describes NCT03739528, a completed Phase 3 randomized, single-masked, parallel-group study in adults undergoing cataract surgery. Enrollment is listed as 808. The study compared levofloxacin plus dexamethasone for 7 days followed by dexamethasone alone for 7 days with tobramycin plus dexamethasone for 14 days. Both regimens were described as 1 drop/4 times a day.
The primary endpoint was the number of participants without signs of anterior chamber inflammation at Day 15, defined by a sum of cell and flare scores equal to 0. The stated objective was non-inferiority with a margin of 10%. The primary outcome is not reduced to one comparison in the registry record, and its treatment estimate and uncertainty are not reported in the information available for interpretation. The record lists secondary outcomes and selected adverse-event counts, but these do not resolve the primary question.