Effect of Nitrous Oxide in Treating Neuropathic Pain: A Study in Chronic Low Back Pain Patients: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Nitrous Oxide in Treating Neuropathic Pain: A Study in Chronic Low Back Pain Patients (NCT01172600), a randomized na study. The primary outcome is Change in VAS Pain Score From Baseline to 3 Month Follow-up.

2. Clinical Question

The outcome was measured on a 10-cm visual analog scale, with 0 representing no pain and 10 representing the worst pain imaginable. The registry describes the primary comparison as change from before the first block to the 3-month follow-up after the last block. The number of blocks could be as high as three, with a typical interval of 1 month.

This was a comparison of Entonox added to an interventional block against oxygen added to that block, rather than a comparison of the block itself with no procedure. The registry notes that procedural choices such as steroids and local anesthetic agents were left to the physician and were not controlled by the study. The central statistical question is therefore about the groups’ recorded pain-score changes under the study’s care conditions, not about the isolated effect of the epidural procedure or any one procedural medication.

3. Trial Design

The registry record describes a randomized, double-masked, parallel-group pilot study. It lists enrollment of 78 and records the study as completed. The study start is recorded as 2011-01 and primary completion as 2013-01. The phase is listed as not applicable.

Participants were assigned to Entonox or oxygen. The intervention descriptions specify inhalation through a mouthpiece during the scheduled interventional block and throughout recovery for 4 hours. The registry says participants were blinded to which inhaled treatment they received. Although the study is classified as double-masked, the registry record does not identify every party who was masked.

The epidural injection was to be performed under fluoroscopy. The brief summary describes midazolam before the procedure and possible intravenous fentanyl if needed. It also states that radiopaque contrast, steroids, and local anesthetic agents were used according to the physician performing the block and were not controlled by the study. Participants were monitored in recovery, side effects were to be recorded, and pain score was recorded before discharge. The study could repeat the procedure up to the maximum of three blocks; the registry describes a typical interval of 1 month.

Follow-up was planned during each block and at 1, 3, 6, and 12 months. The primary endpoint focused on baseline-to-3-month change after the last block. Standard instructions about physical back exercises were given to all participants. The registry does not report the trial’s original sample-size justification, detailed allocation procedures, or a fuller description of who was masked. The information about follow-up questionnaires in the brief summary is incomplete, so their full content and administration 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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: For 10 patients with missing VAS at 3 month follow up: 5 patients who had 2nd or 3rd epidural block, we assigned the last VAS observation (i.e., from VAS before 2nd or 3rd block to 3 month follow-up); and for 5 patients who only had 1st epidural, we assigned the worst VAS (10) for Entonox patients and the best VAS (0) for Oxygen patients.

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 Change in VAS Pain Score From Baseline to 3 Month Follow-up, measured over At baseline (before 1st block) and 3 months follow-up after last block (maximum of 3 blocks with a typical 1 month interval between blocks).

The primary endpoint was change in pain score from baseline, before the first block, to the 3-month follow-up after the last block. Pain was measured with a 10-cm visual analog scale ranging from 0, no pain, to 10, the worst pain imaginable. The registry states that the scale was measured before the first, second, and third blocks and at the 3-month follow-up.

A change score describes movement from a participant’s starting measurement, rather than pain at follow-up alone. Negative mean changes in the posted results indicate lower scores relative to baseline. Comparing mean changes between randomized groups is intended to assess whether the recorded changes differed between the Entonox and oxygen groups.

This endpoint is clinically relevant to the question because pain is the symptom the intervention was intended to address. Statistically, a continuous score retains more gradation than a simple classification such as improved or not improved. Its interpretation still depends on how measurements were collected and how missing follow-up scores were handled. The registry describes a special assignment rule for some missing VAS observations, but the description is cut off and does not provide the complete rule. No threshold for a clinically important change is reported in the registry record, so the numerical group difference cannot be classified against such a threshold using this record alone.

7. Results Posted in the Registry

• Oxygen: mean -1.20 versus -1.60 on control; difference 0.40 (95% CI -0.85 to 1.65), standardized effect d = 0.14; z-test p = 0.529.

8. Secondary Efficacy Endpoints

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

Change in Oswestry Score (% of Disability) From Baseline to 2nd Block

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Oxygen: mean -6.00 versus -7.00 on control; difference 1.00 (95% CI -6.27 to 8.27), standardized effect d = 0.07; z-test p = 0.787.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Oxygen and the control. The interval gives the range of effects that remain compatible with the data.

Change in Oswestry Score (% of Disability) From Baseline to 3rd Block

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Oxygen: mean -7.00 versus -10.00 on control; difference 3.00 (95% CI -8.48 to 14.48), standardized effect d = 0.20; z-test p = 0.609.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Oxygen and the control. The interval gives the range of effects that remain compatible with the data.

Change in Oswestry Score (% of Disability) From Baseline to 3 Months Follow-up

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Oxygen: mean -6.00 versus -11.00 on control; difference 5.00 (95% CI -4.52 to 14.52), standardized effect d = 0.25; z-test p = 0.303.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Oxygen and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary results include three Oswestry disability-score change outcomes: change before the second block, before the third block, and at 3-month follow-up. Across these outcomes, the group means are negative in both arms, while the unadjusted between-group confidence intervals span zero. The pattern is broadly consistent with decreases in the disability scores in both groups, without a clearly resolved difference between groups in these comparisons.

These secondary results complement the primary VAS result by addressing disability rather than pain intensity. They do not overturn or independently settle the primary comparison: the endpoint measures differ, and the secondary comparisons are also imprecise. The registry lists additional secondary outcomes concerning opioid use and biomarkers, but numerical results for those outcomes are not reported in the registry record among the posted secondary summaries.

The number of observations available varies across the Oswestry assessments, with fewer participants represented at later block assessments than in the primary group summaries. These changing analysis sets make direct comparisons across the secondary endpoints less straightforward. The registry does not explain the reasons for missing or unavailable measurements for each endpoint.

Unadjusted secondary findings carry less weight when many outcomes or comparisons are examined. General statistical explanation: with multiple opportunities to observe a nominally notable result, some results can appear notable by chance even if no underlying effect exists. A prespecified hierarchy or adjustment can help control that issue, but the registry record does not report whether multiplicity procedures were planned or used. The secondary results are therefore best treated as descriptive and supportive rather than as independent confirmatory evidence.

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

• Change in Oswestry Score (% of Disability) From Baseline to 2nd Block
• Change in Oswestry Score (% of Disability) From Baseline to 3rd Block
• Change in Oswestry Score (% of Disability) From Baseline to 3 Months Follow-up
• Usage of Opioids
• Usage of Opioid
• Usage of Opioid
• Biomarkers
• Biomarkers

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a score on a visual analog scale, and the registry reports group means and standard deviations. The posted summaries allow a straightforward comparison of average change between the randomized groups. The re-calculated estimate is unadjusted, meaning it uses the group summary statistics without adjusting for other variables in a regression model.

What does a mean change represent?

A mean change is the average of participants’ follow-up scores minus their baseline scores, following the registry’s stated change-score definition. A negative value indicates that the follow-up scores were lower on average than baseline scores. A group mean alone does not reveal how individual participants’ changes were distributed: participants can have different responses even when the average is the same. The standard deviation describes the spread of individual values around the group mean.

For the primary outcome, the Entonox mean change was -1.6 units with a standard deviation of 3.0, and the oxygen mean change was -1.2 units with a standard deviation of 2.6. The reported group size was 39 for each arm. The comparison is based on those summary values and the direction of the contrast is stated as oxygen relative to Entonox. No adjustment for baseline score or other covariates is included in this simple re-calculation.

How should the mean difference be read?

The re-calculated mean difference is 0.4 units. This is the difference between the two group mean changes under the stated comparison direction; it is not an estimate of change within either group alone. The confidence interval spans zero, so the summary-statistic comparison does not rule out either a negative or a positive between-group difference. The unadjusted p-value is 0.529. General statistical interpretation: a p-value describes how unusual data at least this inconsistent with a specified null model would be if that model were true. It is not the probability that the null hypothesis is true, nor a measure of clinical importance.

What does the regression analysis add?

The registry also reports two analyses using linear regression and labels their parameter as mean difference in final values. Their reported estimates are -0.13, with a 95% confidence interval from -1.43 to 1.17 and p = 0.84, and -1.03, with a 95% confidence interval from -2.34 to 0.28 and p = 0.12. Both intervals include zero. The record does not explain why there are two analyses, what covariates or participant sets were used, or how these final-value estimates relate to the change-score estimate. The word “regression” identifies a general model class, but the registry record does not contain enough detail to describe the specific model adjustments.

General statistical explanation: regression can estimate a group contrast while accounting for other measured variables, if those variables are included in the model. That may make the estimate answer a different question from a raw comparison of change-score means. It is not possible to determine from the registry description whether or how that distinction applies to the two reported analyses.

Why does variation matter?

Standard deviations describe variability among observed scores, while the standard error describes uncertainty in an estimated mean. Greater variability generally makes a group mean less precise when the group size is held fixed. The primary group standard deviations show that individual change scores varied around the group averages. General statistical explanation: when outcomes are more variable, a study often needs more information to estimate a group difference with a given degree of precision.

The simple re-calculation assumes that the posted group summaries are suitable for an unadjusted comparison and uses the reported group sizes. It does not reproduce a participant-level model, apply covariate adjustment, or resolve missing-data questions. The registry’s stated approach to missing VAS values is only partially described. Consequently, the re-calculated estimate is useful as a transparent summary-statistic comparison, but it cannot substitute for a fully specified analysis of the participant-level data.

What assumptions are not described?

The registry record does not report the original sample-size justification, a complete statistical analysis plan, an interim-analysis plan, or details of analysis populations. It also does not report whether multiplicity adjustments were used for the set of outcomes or analyses. These omissions limit assessment of how the posted analyses were selected and how uncertainty should be interpreted across the full set of endpoints. They do not change the arithmetic of the unadjusted re-calculation, but they matter when evaluating the broader evidential weight of the results.

General statistical explanation: a confidence interval reflects uncertainty under the calculation or model used to construct it. Its stated coverage level does not mean there is a corresponding probability that the fixed, true effect lies inside this particular interval. Interpretation relies on assumptions about the data and analysis; a confidence interval does not account automatically for missing data, selective reporting, or multiple comparisons.

11. Confidence Intervals and Precision

The unadjusted primary comparison has a confidence interval that extends from negative to positive values. Because it includes zero, the posted summary statistics do not identify a precise direction for the between-group difference. The interval is also broad enough that effects of differing magnitudes remain compatible with the calculation. The point estimate of 0.4 units should therefore not be read as a precise estimate of the true difference.

General statistical explanation: a confidence interval is a range of estimates compatible with the data and the method used to calculate it, at the stated confidence level. A narrower interval indicates greater precision; a wider interval indicates less precision. An interval that includes zero does not prove there is no difference. It indicates that the data and method do not distinguish the estimate from zero at the corresponding level of uncertainty.

The sponsor-reported regression intervals also include zero: one ranges from -1.43 to 1.17 and the other from -2.34 to 0.28, each at the 95% confidence level. These intervals do not provide a clear separation from zero either. Since the registry record does not explain the distinction between the two regression analyses, their estimates and intervals should not be treated as interchangeable or as a single combined result.

Precision is affected by both the spread of scores and the amount of information available. The primary summaries include 39 participants in each arm, and the registry reports missing VAS follow-up for 10 participants. The incomplete missing-data description adds uncertainty about how the primary endpoint was constructed for some participants. The interval from the simple re-calculation reflects its own summary-statistic assumptions; it does not capture every uncertainty arising from incomplete reporting or missing-data choices.

The confidence intervals are more informative about uncertainty than a simple statement that a result is or is not statistically significant. Here, both the interval and the p-value indicate that the posted results do not give a precise estimate of the between-group contrast. They do not establish equivalence, and the record reports no threshold for judging whether a particular difference would be clinically important.

12. Planning a Similar Trial: Sample Size

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

The planning estimates are participants per arm, not total participants, for the continuous outcomes in this study. The assumptions listed for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The trial’s original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column describes the number of participants per arm the planning calculation indicates for an 80% chance of detecting the specified effect under those assumptions, if that effect and the assumptions were correct. The 90% power column gives the corresponding planning number for a 90% chance. These are design illustrations, not a description of what happened in the completed trial.

The rows that plan for 100%, 75%, or 50% of the observed effect explore different assumed effect sizes. Smaller effects are harder to distinguish from random variation, so the planning calculation requires more participants per arm for the smaller-effect rows. The displayed values can therefore rise substantially as the assumed effect becomes a smaller fraction of the observed one. They are based on the observed estimate, which itself is uncertain, and should not be read as a definitive recommendation for a future trial.

This is not observed power. Observed power is not a useful substitute for interpreting the estimate and its confidence interval after a study has been completed. The planning figures are best understood as conditional illustrations of how a future design’s participant requirements vary with the assumed effect and target power. They do not recover the trial’s original design rationale, which is not reported in the registry record.

13. How the Sample Size Is Derived

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

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 0.400; pooled standard deviation σ = 2.807.
2. n = 2 × (1.96 + 0.842)² × 2.807² / 0.400² = 773.1, rounded up to 774 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
delta <- 0.4                      # observed difference in means
sigma <- 2.80713376952364                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1172600)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 774 (80% power), 1,035 (90% power)
 75% of observed effect: n per arm = 1,375 (80% power), 1,840 (90% power)
 50% of observed effect: n per arm = 3,093 (80% power), 4,140 (90% power)
power.t.test: 774.1 
Simulated power at n = 774 per arm: 0.798 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The registry reports two sponsor analyses using linear regression, both labeled as mean differences in final values. The first estimate is -0.13, with a 95% confidence interval from -1.43 to 1.17 and p = 0.84. The second is -1.03, with a 95% confidence interval from -2.34 to 0.28 and p = 0.12. Both intervals include zero.

The simple re-calculation instead compares the posted mean changes and gives a mean difference of 0.4 units for oxygen relative to Entonox, with an unadjusted p-value of 0.529. This re-calculation is not a reproduction of the sponsor’s regression model. One is derived from group-level change summaries; the registry labels the sponsor analyses as regression comparisons of final values. The record does not explain the distinct purpose or specification of the two regression analyses, or how their contrast direction and participant sets relate to the summary-statistic comparison.

The registry record does not flag a discrepancy between the posted summaries and the reported analyses. The sponsor results should be reported as they appear in the record, while the re-calculation should remain clearly identified as an unadjusted comparison. Neither set of results provides a confidence interval excluding zero. Without more detail on model specification, covariates, missing-data handling, and analysis populations, the estimates cannot be reconciled into one preferred numerical contrast solely from the registry description.

15. Multiplicity, Missing Data and Analysis Populations

The registry record identifies a primary VAS change outcome and multiple secondary outcomes, including Oswestry disability scores, opioid use, and biomarkers. It also reports more than one regression analysis for the primary outcome. A complete statistical analysis plan, including the intended hierarchy among outcomes and analyses, is not reported in the registry record.

Multiplicity arises when several outcomes, time points, or comparisons are evaluated. General statistical explanation: making multiple statistical tests increases the chance that at least one result will appear unusual by chance, even when each test is interpreted at the same nominal threshold. Adjustment methods or a prespecified testing order can address this issue, but whether such procedures were used in this study is not reported in the registry record. For that reason, secondary findings and multiple reported analyses warrant cautious interpretation rather than being treated as a set of independent confirmations.

Missing data are especially relevant to the primary outcome. The registry states that 10 participants lacked a VAS measurement at the 3-month follow-up. It describes assigning the last VAS observation for some participants who had a second or third epidural block, and assigning the worst VAS value of 10 for some Entonox participants who had only the first epidural. The final words of the registry’s description are incomplete, so the full assignment rule, including its complete application across groups, is not reported in the registry record.

The record does not report a complete account of the primary analysis population, whether all randomized participants were included in each analysis, or how missing values were handled for the other endpoints. It also does not provide enough information to assess whether participants with missing outcomes differed from those with observed outcomes. These details matter because estimates can depend on which participants and measurements are included and on the assumptions used to handle missing observations.

Accordingly, the posted summaries support a limited interpretation of the observed estimates, but they do not allow a full audit of the analysis choices. The registry does not report an interim-analysis plan, the original sample-size justification, or a complete description of analysis populations. Those omissions should be distinguished from evidence that a particular procedure was or was not used.

16. 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 0% reporting threshold.

The registry reports no serious adverse events among the 39 participants at risk in the Entonox group and none among the 39 at risk in the oxygen group. It also reports no other adverse events in either group among those at risk. These are the adverse-event counts recorded in the registry.

These figures describe what was recorded; they do not establish that no adverse effects occurred outside the reporting framework or observation period. The study summary says possible side effects were to be monitored and recorded, but the registry does not give detailed event terms, severity categories beyond the serious/other groupings, or a fuller account of event collection and follow-up.

Deaths are listed as null for both groups, rather than as a reported zero. Therefore, deaths are not reported in the registry record. The absence of listed serious-event terms alongside zero reported counts should be read as a registry summary, not as proof that an intervention has no safety risks.

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

The posted results describe average changes in measured pain and disability scores among participants in the study. In the primary comparison, mean VAS scores decreased from baseline in both groups, and the unadjusted estimate of the difference between the group changes was 0.4 units. Its confidence interval spans zero, and the unadjusted p-value is 0.529. The registry’s two regression analyses also have confidence intervals that include zero.

These results do not provide a clear statistical separation between Entonox and oxygen on the primary outcome. They also do not demonstrate that the treatments are equivalent. Showing equivalence requires a design and analysis with a defined equivalence margin and an appropriate interpretation of the interval relative to that margin; no such margin is reported in the registry record.

The results do not establish that Entonox caused the changes seen within either group. Both groups received an interventional block, and the registry notes that several procedural choices were determined by the physician rather than controlled by the study. The study therefore compares the inhaled treatment conditions in the context of those procedures and associated care.

The estimates are uncertain, and interpretation is further limited by incomplete information about the missing primary VAS observations, the two regression analyses, analysis populations, and multiplicity procedures. No clinically important difference threshold is reported, so statistical estimates alone cannot determine whether a particular contrast would be meaningful to patients. The registry’s adverse-event counts are also limited summaries and should not be interpreted as proof of absence of risk.

The most defensible conclusion is that this completed pilot study reports no precise, clearly resolved difference between the groups in the posted primary comparison. It does not settle whether a treatment effect exists, whether any effect is clinically important, or what would be found under a different study design or analysis. Those questions cannot be answered from the registry record alone.

18. Why This Trial Matters Statistically

This trial illustrates why a point estimate should be interpreted alongside its uncertainty. The group means indicate decreases in VAS pain scores in both arms, but the between-group estimate is accompanied by an interval that crosses zero. The numerical estimate alone does not establish a difference, and a non-significant p-value does not establish that the groups are identical.

It also illustrates how different analysis choices can answer different questions. The simple re-calculation uses mean changes, while the registry reports linear-regression analyses of final values. The record does not explain the distinction between the two reported regression analyses or fully describe their models. Clear reporting of the estimand, analysis population, model specification, and missing-data rules is important for comparing estimates and understanding what each one represents.

The endpoint and study procedures also require careful framing. Pain was measured repeatedly, the number of blocks could vary, and physician-selected procedural components were not controlled by the study. The reported primary outcome concerns change from baseline to follow-up after the last block, rather than an isolated effect of inhaled gas independent of the rest of care.

Finally, the pilot context and the wide uncertainty make this a useful example of the distinction between exploratory estimates and definitive evidence. The posted results can inform understanding of the measured outcomes and the limitations of the analysis, but they do not establish equivalence or resolve the effectiveness question. Transparent reporting of planning assumptions, outcome handling, and multiplicity would help readers assess how much weight to place on each result.

19. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry provides incomplete detail about the primary missing-data procedure. It reports missing VAS follow-up for 10 participants and describes different assignments for some participants depending on their block history and group, but the description is cut off. This prevents a complete assessment of how the final primary values were constructed.

The two sponsor-reported regression analyses are not distinguished by population, covariates, model specification, or contrast definition in the registry record. Their estimates therefore cannot be fully interpreted or reconciled with the unadjusted change-score calculation from the registry information alone. Analysis populations and the handling of missing data for the secondary outcomes are also not reported.

The registry lists multiple secondary outcomes, but numerical results are posted for only some of those outcomes in the information available. It does not report whether multiplicity was addressed, nor does it provide a complete statistical analysis plan, an interim-analysis plan, or the original sample-size justification. These gaps limit evaluation of the broader analysis strategy and the intended status of the reported comparisons.

Several aspects of the intervention context were not standardized by the study: the registry says steroids and local anesthetic agents were used according to the physician performing the block. The record also does not provide a detailed account of masking beyond the study’s double-masked classification and the statement that participants were blinded to the inhaled treatment. The extent to which other study personnel were masked is not reported in the registry record.

These reporting limitations do not invalidate the posted summaries, but they constrain how specifically they can be interpreted. The available results describe group-level outcomes under the recorded study conditions; they do not provide a complete basis for reconstructing every analysis or separating the inhaled treatment from other aspects of care.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Effect of Nitrous Oxide in Treating Neuropathic Pain: A Study in Chronic Low Back Pain Patients — NCT01172600.
• Registry record last updated 2016-10-17; status last verified 2016-08.
• Results first posted 2016-10-17.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

23. Record Summary

The registry record describes a completed, randomized, double-masked, parallel-group pilot study sponsored by The Cleveland Clinic. It lists enrollment of 78 and compares inhaled Entonox with oxygen during an interventional block for chronic low back pain with radiating pain.

The primary outcome was change in VAS pain score from baseline to the 3-month follow-up after the last block. Both groups had negative mean changes, and the unadjusted between-group estimate had a confidence interval spanning zero. The registry also reports two linear-regression analyses, each with a confidence interval spanning zero, without explaining how the analyses differ.

The record describes missing primary VAS observations but gives an incomplete account of their handling. It reports no serious or other adverse events in either group; deaths are not reported. The original sample-size rationale, complete analysis plan, analysis populations, and multiplicity procedures are not reported in the registry record.