Handheld Ultrasound Evaluation of the Inferior Vena Cava to Guide Heart Failure Treatment: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Handheld Ultrasound Evaluation of the Inferior Vena Cava to Guide Heart Failure Treatment (NCT01962688), a randomized na study. The primary outcome is Number of Participants Hospitalized for Cardiovascular Reasons.

2. Clinical Question

3. Trial Design

The registry describes a randomized, single-masked, parallel study with an enrollment of 37. It lists the condition as heart failure and the status as terminated. The study description says it comprised two independent, non-interacting trials: one in the outpatient setting and one in the inpatient setting. Both settings compared handheld ultrasound-guided diuretic therapy with conventional symptom-guided care, using sham ultrasound in the comparator group to facilitate blinding.

The primary outcome description is specifically about hospitalization information recorded throughout the study for the outpatient arms, and the posted result compares Handheld Ultrasound with Clinical Assessment Only. The record does not explain how the posted counts relate to any inpatient outcome data, or provide separate results for an inpatient comparison. The study timeframe for the primary endpoint is up to 6 months. The record lists a start date of 2013-08 and primary completion of 2015-12. It does not report why the study was terminated.

The arm descriptions state that handheld ultrasound measurement of IVC diameter was used to guide diuretic therapy, while the comparator received conventional care. The ambulatory description places ultrasound use during normal clinic visits. These descriptions identify the intended contrast, but the registry record does not report treatment algorithms, thresholds for changing therapy, visit schedules, or how consistently the intervention was delivered. It also does not report the allocation sequence or the identity of the person or people masked.

Three secondary outcomes are listed: New York Heart Association class, change in health-related quality of life, and length of stay. The record lists six outcome measures in total, but the posted results described for the primary comparison do not provide those secondary outcome findings. Their results 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

5. Baseline Characteristics

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

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants Hospitalized for Cardiovascular Reasons, measured over up to 6 months.

The primary endpoint counts participants hospitalized for cardiovascular reasons. It is a binary participant-level outcome for the comparison: each participant either had a qualifying hospitalization during the stated observation period or did not. The record gives a timeframe of up to 6 months and says hospitalization information was recorded throughout the study for the outpatient arms. The reported counts are 4.0 of 19.0 in Handheld Ultrasound and 5.0 of 18.0 in Clinical Assessment Only.

Hospitalization is used as a clinically relevant event measure because the study description frames the intervention around preventing heart-failure readmissions and the primary outcome specifically concerns cardiovascular hospitalization. The posted endpoint, however, is broader in wording than heart-failure readmission alone: it counts participants hospitalized for cardiovascular reasons. The record does not report a more detailed definition of qualifying cardiovascular hospitalization, how events were adjudicated, whether recurrent admissions were counted separately, or how deaths were handled in the endpoint analysis.

Counting participants with an event makes the endpoint interpretable as the proportion of each group experiencing at least one qualifying event, rather than the total number of hospital stays. The posted summaries do not include individual follow-up times or event dates. Consequently, the comparison is presented as a binary count rather than as a time-to-event analysis. The registry record does not report additional details about censoring or missing outcome status.

7. Results Posted in the Registry

• Clinical Assessment Only: 27.8% (5/18) versus 21.1% (4/19) on control; risk difference 6.7 percentage points (95% CI -20.9 to 34.4), risk ratio 1.32 (0.42–4.15); unadjusted two-proportion z-test p = 0.634.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a risk difference describe?

A risk difference subtracts the event risk in one group from the risk in the other. It is an absolute-scale comparison: it describes how far apart the observed event proportions are, rather than expressing one as a multiple of the other. For this comparison, the reported group risks are 0.210 for Handheld Ultrasound and 0.278 for Clinical Assessment Only, and the reported risk difference is 0.067 for Clinical Assessment Only compared with Handheld Ultrasound. The direction depends on which group is named first. A positive reported difference means the estimated risk is higher in the first group of that stated comparison.

How do risk ratios and odds ratios differ?

A risk ratio compares the event risks as a ratio, while an odds ratio compares the odds of an event. Both are relative measures, and they answer different questions from an absolute risk difference. The registry reports a risk ratio of 1.319 and an odds ratio of 1.442 for Clinical Assessment Only compared with Handheld Ultrasound. Those point estimates are not substitutes for the counts or the uncertainty intervals. With sparse events, relative measures can be especially sensitive to the particular observed counts, so they should be read alongside the interval estimates.

What does a confidence interval communicate?

A confidence interval expresses uncertainty around an estimate using a range of values compatible with the observed data under the method used. It is not a guarantee that the true effect lies within the interval, and it does not mean that values outside it are impossible. For a risk difference, the no-difference reference is zero. For a ratio measure, the no-association reference is the ratio-neutral value. If an interval spans its relevant reference, the data do not distinguish the estimate from that reference at the interval's confidence level. The registry record does not report the confidence level or the method used to construct these intervals, so those details should not be assumed.

What does the p-value add?

The unadjusted p-value is 0.634. In general, a p-value describes how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The record labels the value unadjusted but does not report the statistical test used, the model, or whether the analysis accounted for any design features. The p-value should therefore be read as a reported summary, not as a complete description of the analysis.

Why does sparse binary data matter?

The endpoint counts participants with an event, and the posted group counts are 4.0 of 19.0 and 5.0 of 18.0. With few events and limited group sizes, each observed event carries substantial influence on the estimated risks. This can yield broad intervals and unstable relative estimates. General statistical methods for binary outcomes may use different approximations or tests, and those choices can matter when data are sparse. The registry record does not identify the method used to calculate the reported estimates and intervals.

Over-dispersion is a general issue when observed variation exceeds what a chosen statistical model expects. It can make model-based uncertainty too small if not accounted for. The posted endpoint is a participant count, and no dispersion measure or model information is reported for this comparison. There is therefore no basis in the registry record to assert that over-dispersion was assessed or that a particular adjustment was used.

Randomization is intended to support a fair comparison by assigning participants to groups without choosing treatment based on their characteristics. It does not guarantee that small groups will be balanced on every prognostic factor. The record says the study was randomized but does not report the randomization procedure, stratification factors, or baseline characteristics in the results described. The comparison is therefore interpreted from the posted randomized-arm counts without assuming additional adjustment or balance.

10. Confidence Intervals and Precision

The principal precision issue is the breadth of the reported uncertainty around the risk difference. The estimate is 0.067, and its interval runs from -0.209 to 0.344. Because the interval includes values on both sides of no difference, it permits interpretations in either direction. Its span also indicates that the observed counts do not estimate the absolute group contrast tightly.

The relative measures are similarly uncertain. The reported risk ratio is 1.319, with an interval from 0.419 to 4.15. The odds ratio is 1.442, with an interval from 0.319 to 6.529. Both intervals include the ratio-neutral reference, so the estimates do not establish a relative difference. The displayed bounds are wide in relation to their point estimates; a reader should not treat the point estimates as precise forecasts.

The record does not report the confidence level or the interval calculation method. Those details affect how an interval should be interpreted, and they cannot be inferred solely from the endpoints. The unadjusted p-value of 0.634 is another summary of uncertainty, but it does not replace the interval or tell readers which effect sizes remain compatible with the data.

The modest group sizes and small event counts are important context for all of these estimates. The data provide limited information about the magnitude and direction of the difference. Precision is about how tightly an effect is estimated, not whether a result is important or whether an intervention should be used. In this comparison, the broad intervals are a reason to avoid strong claims from the point estimates.

11. Planning a Similar Trial: Sample Size

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

The sample-size figures are planning illustrations based on fractions of the observed effect, not numbers of participants observed in the trial. They are expressed as participants per arm for 80% power and participants per arm for 90% power. 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 the stated fraction of the observed effect. The trial’s own original design assumptions are not reported in the registry record.

For planning at 100% of the observed effect, the figures are 640 participants per arm for 80% power and 856 participants per arm for 90% power. For planning at 75% of the observed effect, the figures are 1,111 and 1,487 participants per arm, respectively. For planning at 50% of the observed effect, the figures are 2,438 and 3,263 participants per arm, respectively. These are illustrations under the stated assumptions, not a claim about a required design for a future study.

In general, power is the probability that a planned analysis will detect an effect of a specified size under the assumptions used, if that effect is present. The 80% and 90% columns represent different target probabilities of detection; the higher target requires more participants for each listed effect fraction. Smaller effects also require more participants because they are harder to distinguish from random variation. These figures are not observed power.

The listed assumptions do not describe every possible design choice. The record does not report the trial’s original sample-size justification, anticipated event rate, allowance for missing data, or planned statistical model. The illustration is based on the observed effect, which is uncertain in this small study, so it should be treated as a planning reference rather than a definitive sample-size recommendation.

12. How the Sample Size Is Derived

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

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.211; treatment proportion p₁ = 0.278; average p̄ = 0.244.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6075 = 1.1907.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6057 = 0.5097.
4. n = (1.1907 + 0.5097)² / (0.0673)² = 639.3, rounded up to 640 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.210526315789474                          # control proportion
p1 <- 0.277777777777778                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(1962688)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 640 (80% power), 856 (90% power)
 75% of observed effect: n per arm = 1,111 (80% power), 1,487 (90% power)
 50% of observed effect: n per arm = 2,438 (80% power), 3,263 (90% power)
power.prop.test: 639.3 
Simulated power at n = 640 per arm: 0.802 (20000 replicates)

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

13. Multiplicity, Missing Data and Analysis Populations

The registry lists three secondary outcomes: the number of participants in each New York Heart Association class, change in health-related quality of life, and length of stay. It also lists six outcome measures in total and one analysis. The posted comparison described for the primary endpoint is an unadjusted analysis. The registry record does not report whether there was a prespecified hierarchy among outcomes, whether a multiplicity adjustment was planned, or how multiple comparisons were handled. Without that information, the one posted p-value should not be assumed to account for all outcomes or analyses.

Missing outcome data can affect a comparison if participants with unknown outcomes differ from those with recorded outcomes. The record does not report the amount or pattern of missing data, reasons for missingness, or how missing outcome status was handled. It also does not report whether all randomized participants were included in the primary analysis or specify an analysis population such as intention-to-treat or per-protocol. Those aspects of the analysis are therefore not reported in the registry record.

The record describes the study as randomized and single-masked, but it does not state who was masked or whether masking was maintained. It does not report the allocation procedure, deviations from assigned treatment, or use of an adjusted model. The endpoint description says hospitalization information was recorded throughout the study for the outpatient arms, but does not give details about event verification, completeness of follow-up, or handling of deaths.

These omissions limit assessment of how the reported estimates were produced and whether the analysis fully followed a planned approach. They do not show that a particular method was or was not used; they identify details that are not reported in the registry record. The reported result can still be described as posted, but its analysis context is incomplete.

14. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: all cause hospitalization (50.0%), cardiac related hospitalization (27.8%), non cardiac hospitalization (22.2%).

The registry reports serious adverse events in 7 of 19 participants in Handheld Ultrasound and 10 of 18 in Clinical Assessment Only. It reports 0 other adverse events among 19 participants in Handheld Ultrasound and 0 among 18 in Clinical Assessment Only. Deaths are listed as 0 in both groups. These are counts of participants affected, as presented in the adverse-event record; they should not be confused with the primary endpoint counts.

The listed serious-event terms include All cause Hospitalization at 50.0%, Cardiac Related Hospitalization at 27.8%, and Non Cardiac Hospitalization at 22.2%. The registry record does not provide enough detail to determine whether these terms overlap, how each percentage was calculated, or how events were classified. Serious adverse-event summaries describe events recorded during the study; they do not by themselves establish that an event was caused by an intervention.

The adverse-event figures should be read in light of the limited group sizes and the absence of detailed event descriptions in the registry record. The record does not report event severity beyond the serious classification, timing, treatment relationship, or the method used to collect and adjudicate events. These counts are descriptive and do not support a precise comparison of safety between groups.

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

The posted primary result is a comparison of cardiovascular hospitalization counts in the Handheld Ultrasound and Clinical Assessment Only groups. The observed counts were 4.0 of 19.0 and 5.0 of 18.0, with reported risks of 0.210 and 0.278. The estimated risk difference is 0.067 for Clinical Assessment Only compared with Handheld Ultrasound, but its interval extends from -0.209 to 0.344. The unadjusted p-value is 0.634.

These statistics indicate that the observed data do not precisely distinguish the groups. The interval includes differences in both directions, and the relative estimates also have broad intervals. The result should not be described as proof that the intervention reduces hospitalization, nor as proof that it has no effect. A non-significant comparison does not demonstrate equivalence, especially when the estimate is imprecise.

The endpoint is the number of participants hospitalized for cardiovascular reasons, with a timeframe of up to 6 months. It is not reported as a count of all admissions, and the registry record does not provide a detailed event definition or information about recurrent hospitalizations. The outcome description specifically identifies the outpatient arms; the record also describes an inpatient trial but does not provide an inpatient primary-outcome comparison in the posted result.

The study status is terminated, and the reason is not reported in the registry record. The posted result does not establish whether termination affected recruitment, follow-up, or analysis. Nor does the result supply findings for the listed secondary outcomes. The statistics should therefore be interpreted as a limited account of the reported endpoint, not as a complete summary of all trial questions or outcomes.

In general, randomized comparisons can help estimate differences between assigned groups, but the reliability and scope of the conclusion depend on the design, conduct, outcome measurement, and analysis. Several of those details are not reported in the registry record. The observed counts are informative about what was recorded in these arms, while the wide uncertainty means they provide limited precision about the underlying group contrast.

16. Why This Trial Matters Statistically

This trial illustrates how a randomized comparison can produce an intuitively simple result—counts of participants with an event—while still leaving substantial uncertainty about the effect size. The primary outcome counts are 4.0 and 5.0, in groups of 19.0 and 18.0. Such small counts make the estimated contrast sensitive to individual events. Reporting the risk difference, relative measures, intervals, and p-value together makes that uncertainty more visible than a point estimate alone.

The comparison also highlights the importance of defining the estimand clearly. The study description discusses preventing heart-failure readmissions, while the registered primary outcome is participants hospitalized for cardiovascular reasons. The posted outcome description refers to the outpatient arms. Careful interpretation therefore depends on distinguishing the broader study aim from the endpoint and population attached to the reported result.

Planning figures based on the observed effect show how sample-size requirements vary with the effect size and target power under stated assumptions. They are useful for illustrating the scale of uncertainty and the effort that detecting smaller effects can require. They are not evidence that the original trial had those assumptions, and they do not establish that the observed effect is stable.

Finally, the record underscores why analysis details matter. The statistical test, confidence level, missing-data handling, analysis population, and multiplicity strategy are not reported. These omissions constrain how fully readers can evaluate the result. Transparent reporting of design and analysis choices is important for judging precision and for comparing the posted findings with the study’s broader questions.

17. Limitations

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

The primary comparison is based on a small number of participants and events: 19.0 in Handheld Ultrasound and 18.0 in Clinical Assessment Only, with 4.0 and 5.0 participants hospitalized, respectively. The resulting risk-difference and ratio intervals are broad. The point estimates should not be treated as precise measures of the difference between groups.

The registry describes two separate, non-interacting trials, one inpatient and one outpatient, while the primary outcome description specifies hospitalization information for the outpatient arms. The posted comparison does not provide separate inpatient results or explain how the displayed arms map to the two trial settings. This limits the scope of conclusions that can be drawn from the reported endpoint.

The record does not report why the study was terminated, or whether termination affected enrollment, outcome collection, follow-up, or analysis. It also does not report the detailed definition or adjudication procedure for cardiovascular hospitalization, the handling of recurrent events, or individual follow-up completeness. The specified timeframe is up to 6 months, but further details about timing and censoring are not reported.

Several analytic details are also absent: the statistical test, confidence level and construction method for intervals, the analysis population, missing-data approach, and multiplicity plan. Baseline characteristics, allocation details, and the identity of those masked are not stated in the result description. These are uncertainties in the record, not evidence that the study did or did not use a particular method. The secondary outcomes are listed, but their results are not reported in the registry record.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Handheld Ultrasound Evaluation of the Inferior Vena Cava to Guide Heart Failure Treatment — NCT01962688.
• Registry record last updated 2018-06-06; status last verified 2018-06.
• Results first posted 2018-06-06.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

The registry describes a randomized, single-masked, parallel study of handheld ultrasound-guided diuretic management compared with conventional clinical assessment for heart failure. It reports enrollment of 37, a start date of 2013-08, primary completion of 2015-12, and terminated status. The study description identifies separate inpatient and outpatient trials.

The primary outcome is the number of participants hospitalized for cardiovascular reasons up to 6 months, with the description referring to the outpatient arms. The posted counts are 4.0 of 19.0 for Handheld Ultrasound and 5.0 of 18.0 for Clinical Assessment Only. The reported risk difference is 0.067, with an interval from -0.209 to 0.344, and the unadjusted p-value is 0.634. The record lists three secondary outcomes, but their results and the reason for termination are not reported in the registry record.