Efficacy of Intrathecal Oxytocin in Patients With Neuropathic Pain: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy of Intrathecal Oxytocin in Patients With Neuropathic Pain (NCT02100956), a randomized phase 2 study. The primary outcome is Spontaneous Pain From Their Neuropathic Pain Disease.

2. Clinical Question

3. Trial Design

The registry classifies this as a phase 2, randomized, triple-masked crossover study of intrathecal oxytocin for neuropathic pain. Its planned intervention comparison was oxytocin 100 micrograms against preservative-free normal saline. The trial status is terminated. The registry lists a start date of 2014-06-19 and a primary completion date of 2022-10-10.

In a crossover design, participants receive the study conditions in sequence rather than remaining in a single treatment group throughout. The record describes an oxytocin-then-placebo sequence and a placebo-then-oxytocin sequence. The second injection took place after at least one week. This means the posted injection-condition summaries should not automatically be read as results from separate groups of unrelated participants. The record does not explain how the analysis handled sequence, period, or possible carryover effects.

The registry reports enrollment of 8. For the primary outcome, the displayed oxytocin and placebo groups each have n=5. The stated analysis population was defined as individuals who were randomized and received intrathecal drug on both study days, including subjects who completed both injections. The record does not explain why the enrollment count and outcome-group counts differ, or whether particular observations were excluded from the displayed summaries.

Triple masking indicates masking across multiple parties, but the registry does not specify which people were masked or how masking was maintained. It lists a mixed models analysis for the primary comparison. The record does not report the model specification, the randomization procedure, or a fuller statistical analysis plan. These omissions limit the detail with which the design and analysis can be reconstructed from the registry entry alone.

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: Two subjects were not randomized, one because of an abnormal electrocardiogram and the other for lack of spontaneous neuropathic pain at rest. Six subjects were randomized and received the first intrathecal injection. Participant flow section includes randomized subjects only.

First Injection

At Least One Week Between Injections

Reasons for not completing this period:

Second Injection

Analysis population for the primary outcome, as defined in the registry: The intention to treat population was defined as individuals who were randomized and received intrathecal drug on both study days of the trial. This includes subjects who completed both injections.

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 Spontaneous Pain From Their Neuropathic Pain Disease, measured over 240 minutes post injection.

The primary endpoint was spontaneous pain from the participants’ neuropathic pain disease, measured 240 minutes post injection. The registry describes a visual analog scale: a 10 cm line with “no pain” at the left end and “pain as worse as can be” at the right. Participants marked the line, and the score was determined by the distance in cm from the no-pain end. The outcome is continuous, and the posted group summaries are means with standard errors.

Pain intensity is the direct focus of the stated study purpose, which was to assess the effect of oxytocin on pain as well as hyperalgesia and allodynia. The primary endpoint captures a participant’s reported spontaneous pain at a specified post-injection time. It does not by itself describe other pain characteristics, the distribution of responses, or changes at other times. The registry lists one outcome measure and no posted secondary outcomes.

A mean gives a summary of the observed scores, while the standard error conveys uncertainty around that estimated mean. Neither summary alone gives the adjusted within-person contrast that may arise from a crossover analysis. For that, the model-based treatment comparison and its uncertainty would be informative; those values are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The posted outcome is a continuous measure summarized by a mean and standard error. A mean represents the average score in the analyzed observations; it does not show the shape of the score distribution or how much participants varied individually. A standard error describes the statistical uncertainty in an estimated mean. It is not the same as the standard deviation, which describes variation among observations. The registry reports standard errors of 0.87 for oxytocin and 0.89 for placebo.

What does a mixed models analysis do?

In general, a mixed model can represent both treatment effects and correlations among measurements from the same participant. This can be useful in repeated-measures or crossover studies, where a participant may contribute observations under more than one condition. A model can also account for design features if those features are included in its specification. The registry identifies the method as “Mixed Models Analysis,” but does not report the model formula, which effects were included, the covariance structure, or how missing observations were handled. It is therefore not possible to determine from the record exactly how the crossover structure was represented.

For a crossover comparison, the treatment contrast may depend on paired observations, treatment sequence, and period. In general, order and carryover can matter if responses differ by period or if the effect of an earlier condition persists. These are statistical considerations, not findings about this trial. The registry does not report a carryover assessment, period-specific results, sequence-specific estimates, or details of any washout rationale. The stated interval between injections was at least one week; the record does not describe further rationale for that interval.

How should the p-value be read?

The sponsor’s reported result is p < 0.05 for a superiority analysis. In general, a p-value evaluates how compatible the observed data are with a specified null model, assuming the analysis model and its assumptions are appropriate. It is not an effect-size measure, and it does not show how large a difference is clinically meaningful. Since the registry provides neither the exact p-value nor the model-based contrast, a reader cannot use this result to judge the range of treatment effects compatible with the analysis.

A confidence interval is a range of estimates generated by a statistical procedure, expressing uncertainty about an effect estimate under that procedure. In general, it helps readers assess both the direction and precision of an estimated contrast. No confidence interval for the mixed-model comparison is reported. The standard errors attached to the condition means do not replace an interval for the treatment contrast, particularly when measurements may be related within participants.

Why does the crossover structure matter?

General analysis of crossover data must account for the fact that observations from the same participant may be correlated. Treating all condition observations as independent can misrepresent uncertainty when that correlation is ignored. Mixed models can accommodate within-participant correlation, but the method label alone does not establish which correlation structure or design terms were used. The record also does not report whether the sponsor’s model included sequence or period effects.

Over-dispersion is a general issue when observed variability exceeds what a chosen statistical model expects. In that situation, standard errors and tests based on the simpler model may not represent uncertainty well. The registry does not report a dispersion assessment, and the available record does not support a conclusion about whether over-dispersion occurred in this trial. More generally, the fit and assumptions of the mixed model cannot be evaluated without its specification and supporting diagnostics.

What can be recalculated from the posted summaries?

The posted data include condition means and standard errors, but the record indicates that values were missing or not estimable for a two-arm comparison of this continuous endpoint. No unadjusted recalculated treatment contrast or confidence interval is therefore reported. In a crossover trial, an unadjusted comparison based only on the two marginal means would also omit the within-participant relationship that may contribute to the sponsor’s model-based result. The sponsor’s mixed-model p-value is consequently the only reported inferential comparison in the registry record.

Planning illustrations based on an observed effect, when available, are hypothetical design calculations rather than observed power. They depend on assumptions about the effect to detect, allocation, significance level, and model approximation. They cannot establish that a completed trial had a particular probability of detecting an effect. The registry states general assumptions for such illustrations, but the posted record does not report a recalculated effect for this endpoint from which to interpret a trial-specific planning estimate.

Finally, multiplicity and analysis-population choices affect interpretation. The registry lists one outcome measure and one analysis, with no secondary outcomes posted. It does not report whether any multiplicity adjustment was planned or used. The stated analysis population requires receipt of study drug on both study days; the record does not describe other analysis populations or sensitivity analyses. Those omissions leave uncertainty about robustness to alternative analysis choices.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Oxytocin Injection", "Placebo Injection")
n   <- c(5, 5)
m   <- c(1.67, 2.43)
sp  <- c(0.87, 0.89)
sd  <- sp * sqrt(n)      # standard errors converted to SDs
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Oxytocin Injection  5  1.670  1.945  -0.746  4.086
CARM  Placebo Injection  5  2.430  1.990  -0.041  4.901

11. Analyses Reported by the Sponsor

The registry reports the sponsor’s own primary comparison as a mixed models analysis with a superiority result of p < 0.05. It does not report a model-based treatment difference, confidence interval, or exact p-value. The available group means and standard errors are descriptive summaries for the oxytocin and placebo conditions.

The registry also indicates that values were missing or not estimable for a two-arm comparison of this continuous endpoint. Accordingly, no simple unadjusted recalculation of a treatment contrast is reported alongside the sponsor’s analysis. Such a recalculation would not reconstruct the mixed model, and the crossover structure means that a comparison based only on marginal means would not capture the participant-level relationship between conditions.

There is no reported recalculated estimate to compare with the sponsor’s result. The sponsor’s p-value should therefore be read as the only inferential test reported in the registry, rather than as a result independently confirmed by a recalculation. Its interpretation remains limited by the absent effect estimate, confidence interval, and model details. The record does not provide enough information to assess whether different reasonable model specifications would produce similar estimates or uncertainty.

12. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and one analysis, with no secondary outcomes posted. It identifies spontaneous pain from the neuropathic pain disease as the primary measure, taken 240 minutes post injection. This provides a clear stated primary endpoint, but it does not establish whether additional outcomes or analyses were considered outside the posted record. Any multiplicity adjustment or strategy for controlling the consequences of multiple testing is not reported in the registry record.

The analysis population is defined as individuals who were randomized and received intrathecal drug on both study days, including subjects who completed both injections. This definition is narrower than a simple description of all randomized participants. The record does not state how many randomized individuals did not meet this definition, why they may have been excluded, or how any missing outcomes were treated. The enrollment of 8 and the displayed n=5 for each condition cannot, by themselves, explain participant flow or the handling of incomplete records.

Missing-data methods are not reported. The registry does not say whether the mixed model used all available measurements, whether any imputation was performed, or whether assumptions about the missingness process were assessed. It also does not report sensitivity analyses to examine how missing observations might affect the comparison.

The record names a mixed models analysis but does not provide a full statistical plan, model specification, or analysis-population alternatives. It is therefore not possible to assess whether the posted analysis followed a prespecified plan or whether conclusions were tested under other plausible assumptions. The registry also does not report interim analyses. That absence is a limitation of the available record, not evidence that such analyses did or did not occur.

13. 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 non-serious adverse event terms

The registry reports serious adverse events for the oxytocin and placebo injection groups. In each group, 6 were at risk and 0 were affected by serious adverse events; deaths were also 0 in each group. The record lists no serious adverse event terms. These figures describe the entries in the posted safety results, not a complete account of every possible harm or symptom.

For other adverse events, the record reports 0 affected among 6 at risk after oxytocin and 1 affected among 6 at risk after placebo. These are the counts as recorded; they should not be converted into a broader safety conclusion. The registry does not give the term for the other adverse event, its severity, timing, relationship to study drug, or whether it was resolved. It also does not explain how safety populations relate to the primary outcome population.

Safety counts should be interpreted in the context of the small recorded group sizes and the registry’s limited detail. A zero count does not demonstrate that an event cannot occur, and a count does not establish that an intervention caused an event. The posted record does not provide enough information to evaluate comparative risks with precision or to characterize uncommon events.

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

The registry’s descriptive primary-outcome summaries show a mean spontaneous-pain score of 1.67 after oxytocin and 2.43 after placebo at 240 minutes post injection. The sponsor reports a mixed-model superiority result of p < 0.05. Together, these entries indicate the direction of the displayed mean comparison and the sponsor’s stated statistical test. They do not provide the model-based magnitude of the comparison or its precision.

The result does not establish how large the adjusted treatment contrast was, because the registry does not report an effect estimate or confidence interval. Nor does the reported p-value indicate the probability that a treatment effect is true. Statistical evidence from a model depends on the model specification and assumptions, details that are not available in the registry entry. The missing inferential estimate also prevents assessment of which effect sizes remain compatible with the analysis.

The finding is specific to the recorded endpoint, population definition, and time point. The primary measure was spontaneous pain at 240 minutes after injection in the stated analysis population of randomized individuals who received drug on both study days. The registry does not report the primary result at other time points, a broader outcome profile, or secondary outcomes. The study’s stated interest in hyperalgesia and allodynia does not mean results for those outcomes are posted.

The crossover design makes within-participant structure relevant, but the registry provides limited information about how that structure was modeled. The record also gives limited information about missing observations and the relation between enrollment and the reported outcome counts. These factors constrain how confidently the numerical summaries can be generalized beyond the recorded analysis.

The trial status is terminated, and the registry reports enrollment of 8. The posted information can describe the trial and its recorded result, but it does not establish whether a similar result would be observed in a larger or different population. It should not be read as a clinical recommendation or as proof that oxytocin does or does not relieve neuropathic pain.

15. Why This Trial Matters Statistically

This trial is statistically notable because it combines a randomized crossover design with a continuous pain outcome and a mixed-model analysis. Crossover data can use measurements from the same participant under different conditions, but valid inference depends on representing the dependence between those measurements and the sequence in which conditions were received. The registry names a mixed models analysis, yet does not provide enough detail to evaluate how those design features entered the model.

The registry presents descriptive means and standard errors alongside a sponsor-reported superiority test. That contrast illustrates why a p-value should not stand alone: without a treatment-effect estimate and confidence interval, readers cannot judge the magnitude or precision of the modeled comparison. The record’s indication that a two-arm recalculation was not estimable also highlights the difference between a simple calculation from summary data and an analysis that uses the trial’s design.

The counts underscore the need to distinguish enrollment, outcome analysis, and safety reporting. Enrollment is 8; each displayed primary-outcome condition has n=5; the safety groups list 6 at risk. The registry does not explain these differences, so they should not be combined or treated as interchangeable descriptions of the same analysis population.

Finally, the record gives a defined primary time point and outcome but reports no secondary outcomes. Its statistical result can be read in context, while the missing model details, effect estimate, confidence interval, and missing-data account remain important limits. This is a useful example of why transparent reporting of estimates and analysis assumptions matters alongside a headline significance result.

16. 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 contains limited detail about trial conduct and statistical analysis. It identifies randomized allocation, triple masking, a crossover model, and a mixed models analysis, but does not describe the randomization method, the parties masked, the model specification, or the covariance structure. It does not report how sequence and period were handled, or whether carryover was assessed. These omissions matter because the treatment contrast and its uncertainty can depend on how the crossover design is represented.

The record reports enrollment of 8, while the primary outcome summaries list n=5 for each injection condition and the safety entries list 6 at risk in each group. It does not explain participant flow or reconcile these counts. The analysis population is defined as randomized individuals who received study drug on both study days, including subjects who completed both injections, but the record does not provide a fuller account of exclusions, missing measurements, or other analysis populations.

The primary outcome is reported at 240 minutes post injection, with condition means and standard errors. The registry does not report the model-based treatment difference, confidence interval, or exact p-value. Because the stated comparison could not be recalculated from the posted summary statistics, there is no independent unadjusted estimate available for comparison. Model diagnostics and sensitivity analyses are also not reported.

The registry marks the study as terminated but does not explain the reason for termination. It does not report a sample-size justification, interim analyses, or a complete statistical analysis plan. The record lists no secondary outcomes, although the study purpose also mentions hyperalgesia and allodynia. These limits restrict assessment of the planned scope, robustness of the analysis, and possible reasons the posted results are incomplete.

17. Related Tutorials

18. Related Statistical Calculators

19. Sources

• ClinicalTrials.gov: Efficacy of Intrathecal Oxytocin in Patients With Neuropathic Pain — NCT02100956.
• Registry record last updated 2023-10-03; status last verified 2022-11.
• Results first posted 2023-10-03.
• 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.

20. Record Summary

NCT02100956 is a phase 2 randomized, triple-masked crossover study of intrathecal oxytocin 100 micrograms versus preservative-free normal saline in patients with chronic neuropathic pain. The registry lists Wake Forest University Health Sciences as sponsor, a start date of 2014-06-19, a primary completion date of 2022-10-10, and a terminated status. Enrollment is reported as 8.

The primary endpoint was spontaneous pain on a visual analog scale at 240 minutes post injection. The displayed means were 1.67 after oxytocin and 2.43 after placebo, with standard errors of 0.87 and 0.89. The sponsor’s mixed models superiority analysis is reported as p < 0.05. No model-based effect estimate or confidence interval is reported, and a two-arm recalculation was not estimable from the posted summary statistics.

The record defines the analysis population as randomized individuals who received intrathecal drug on both study days. It lists no secondary outcomes and gives limited detail about missing data, model specification, and participant flow. Safety entries report 0 serious adverse events and 0 deaths in both groups; other adverse events were recorded for 0 at risk after oxytocin and 1 at risk after placebo.