Article: Arvig MD, Lassen AT, Gæde PH, et al. Impact of serial cardiopulmonary point-of-care ultrasound exams in patients with acute dyspnoea: a randomized, controlled trial. Emerg Med J. 2023;40(10):700-707. PMID: 37595984
Clinical Question: In adult patients presenting to the ED with dyspnea, does treatment guided by serial lung and cardiac POCUS plus usual care compared to usual care alone reduce perceived dyspnea on the Verbal Dyspnea Score (VDS)?
What They Did:
- Design: Randomized, controlled, blinded-outcome trial
- Sites: Three emergency departments in Denmark
- Duration: October 9, 2019 to May 26, 2021.
- Patients:
- Compared standard of care to serial US plus stand care in patients with dyspnea.
- Initial assessment included routine physical examinations, medical history, measurement of vital signs, labs, ABG, CXR, and initial PoCUS exam within the first hour.
- PoCUS evaluations included lung ultrasound (LUS) and focused cardiac ultrasound (FoCUS).
- Verbal Dyspnea Score (VDS) was measured on a scale of 0 to 10, with 0 indicating no dyspnea and 10 indicating the worst dyspnea imaginable (2)
- Funding: The authors received funding from various sources, including local hospital department of Emergency Medicine, the hospital’s research fund, and research funds of region Zealand and region of Southern Denmark, as well as the University of Southern Denmark.
- Trial Registry: ClinicaTrials.gov
Population:
Inclusion Criteria:
- Participants who were 18 years of age or older
- Participants with primary complaint of dyspnea in the ED
- Participants who could provide informed consent
- First evaluation of the patient, including the first PoCUS evaluation, could be done within 1 hour of arrival in the ED
Exclusion Criteria:
- Trauma patients
- Patients who were intubated within the first hour after arrival to the ED
- An investigator was not present in the ED
Intervention:
- Serial US + Standard Care
- Lung ultrasound was performed with an 8-zone scanning protocol with the patient in a semi-supine position.
- Cardiac ultrasound assessed RV for dilatation, LV function, presence of pericardial effusion, and calculation of the IVC collapsibility index (IVC-CI)
- Subsequent assessments for patients in the study group performed at 2,4 and 5 hours consisted of clinical evaluations, vital signs, VDS, and PoCUS
Control:
- Standard Care
- Subsequent assessments in the ED for patients in the control group performed at 2,4 and 5 hours consisted of clinical evaluations, vital signs, and VDS.
Outcomes:
Primary Outcome:
- Reduction of dyspnea measured on a verbal dyspnea scale (VDS) from 0 to 10 recorded at enrollment, then at 2, 4, and 5 hours after arrival.
- 1 point difference on VDS was considered minimally clinically important (Pts in the serial ultrasound group were expected to have a 2-point change in VDS compared with a 1-point change in the control group at the final evaluation of patients in the ED)
Secondary Outcomes:
- Length of hospital stay
- The proportion of readmissions within 0-7 and 8-30 days from discharge date
- In-hospital mortality
- 0-7 and 8-30 days mortality from admission date
- Proportion of patients with final ED diagnosis matching the audit diagnosis
- Inferior vena cava collapsibility index (IVC-CI) correlated to vital signs and VDS
- B-line count correlated to vital signs and VDS
- Dynamic changes in IVC-CI between PoCUS evaluations
- Dynamic changes in B-line count between PoCUS evaluations
- Medications and fluids administered in both groups
- Proportions of differential diagnoses during ED stay
- Intra-rater and inter-rater reliability of the PoCUS findings
- Image quality of PoCUS examinations
Results:
Study Characteristics
- 436 patients were assessed for eligibility
- 230 excluded
- 206 patients were randomized
- 102 were allocated to the intervention group
- 104 allocated to the control group
- The mean age of study participants was 76 years
- The most common reason for exclusion was the absence of a primary complaint of dyspnea on arrival to the ED
- A larger proportion of patients in the serial ultrasound group had heart failure compared with those in the control group (28 vs.19)
- The second most common complaint following dyspnea was cough
- One-third of patients had bilateral lower extremity edema in both groups
- One-third of patients had consolidations or pleural effusion on PoCUS examination
- B-lines were present in 85.3% of patients in the serial ultrasound group and 74% of patients in the control group
- Approximately half of patients between both groups had reduced ejection fraction
- The proportion of pathologic ultrasound findings was higher in the serial ultrasound group (i.e. B-lines, consolidation, pleural effusion)
Primary Outcomes:
- Patients in both the control and serial ultrasound groups reported a decrease in the severity of dyspnea over the course of the study
- At 2 hours status post arrival in the ED, patients in the serial ultrasound group reported a VDS difference of -0.08 (-0.51 to 0.35) compared with the control group
- At 4 and 5 hours from inclusion the mean difference in VDS between the patients in the serial ultrasound and the control group was −1.09 (95% CI −1.51 to −0.66) and −1.66 (95% CI −2.09 to −1.23)
- Overall, patients in the serial ultrasound group reported a greater decrease in the severity of dyspnea at all time points compared with the control group
- In patients with a presumptive ED diagnosis of acute heart failure (AHF), the difference in VDS at 4 and 5 hours status post arrival to the ED were -1.52 (95% CI −2.52 to −0.78) and -1.97 (95% CI −2.70 to −1.23), respectively
- Patients in the serial ultrasound group received more diuretics, inhaled beta 2 adrenergic agonists, and supplemental oxygen
- Patients in the serial ultrasound group received a dose between 6 and 8 times greater than those in the control group at time points 2 and 4 hours after arrival
Secondary Outcomes:
- No difference between the study groups was observed for the parameters of length of stay (LOS), readmission within 0-7 days and 8-30 days, mortality in the hospital, and mortality at 0-7 days and 8-30 days
- The most common final ED diagnoses were AHF, pneumonia, and COPD exacerbations, respectively
- The number of final ED diagnoses that matched final audit diagnoses was higher in the serial ultrasound group (64% vs 59%), but investigators found this difference to be statistically not significant
- Overall agreement regarding final audit diagnoses between raters was 96%
- In the serial ultrasound group number of B-lines was nearly identical between the initial PoCUS exam and that done at 2 hours, however this number was decreased at the 5-hour time point
- There was no statistically significant difference found between scans for parameter of IVC-CI
- No correlation was found between IVC-CI or B-lines and VDS or vital signs
- The agreement of intra-rater and inter-rater reliability of ultrasound clips was 96%
- The median image quality of ultrasound clips was 4 on a scale of 1-5
Strengths:
- The study employed a randomized control design, which can mitigate bias.
- Patients were analyzed using an intention-to-treat protocol, ensuring that all participants were included in the analysis, regardless of their study group.
- All patients, regardless of their group, received at least one PoCUS evaluation.
- The outcome assessors responsible were independent healthcare professionals who were blinded to the study group allocations.
- Independent reviewers assessed the quality of ultrasound images and evaluated both inter and intra-rater reliability, ensuring the robustness of the ultrasound findings.
Limitations:
- This trial was conducted in a single country, which may limit the generalizability of the findings to other healthcare settings and populations.
- Patients and treating physicians were not blinded to the treatment allocation, which could introduce bias into the study, as they were aware of whether patients were receiving serial ultrasounds or not.
- The primary outcome, which focused on changes in Verbal Dyspnea Scores (VDS), is subjective and introduces potential variability in the results.
- The study had some potential sources of selection bias. Most patients were recruited from a single Emergency Department (ED) by only two study investigators, which might not represent a broader population. Patients were screened for eligibility on only 72% of the days the trial ran, likely due to investigator availability, raising questions about the representativeness of the included population.
- There were baseline demographic discrepancies between the study groups, such as the male-to-female ratio. Patients in the serial ultrasound group had a higher proportion of pathologic ultrasound findings, possibly due to the increased number of PoCUS exams they received.
- All investigators had certification in Point-of-Care Ultrasound (PoCUS) with at least five years of experience, which may not be representative of all clinicians, potentially affecting the applicability of the results.
- Standard care included at least one PoCUS examination, which may not be standard practice in most EDs, potentially affecting the control group’s outcomes.
- The first evaluation of the patient was to be done within 1 hour of arrival to ED (Including the first POCUS). This may not be feasible in busier EDs in the US, making this portion of the protocol not generalizable to all environments.
- Point-of-care ultrasound (PoCUS) is inherently operator-dependent, which may have led to inconsistencies in the quality and findings of the ultrasound exams across different operators.
- Patients in the serial US group received more medical therapies yet there was no discussion on adverse events.
Discussion:
Baseline Characteristics:
Randomization aims to mitigate both known and unknown variables that could influence the study’s outcome, ensuring that the intervention and control groups are comparable except for the intervention itself. In small trials, differences may persist due to chance despite randomization. In this case, the two groups were not identical and didn’t begin the trial with the same prognosis. While the control group had an equal distribution of males and females, the serial US group had a higher percentage of males compared to females (58.8% vs. 41.2%). Perhaps the difference in gender distribution and potential differences in body composition render it easier for clinicians to obtain ultrasound images and, subsequently, arrive at a diagnosis using US.
Furthermore, the serial US group included more patients with conditions detectable with US. For instance, patients in the serial US group were more likely to have CHF compared to the control group (27.5% vs. 18.3%). CHF, in particular, lends itself well to diagnosis and treatment with PoCUS, as it readily reveals the presence of B lines. In fact, those assigned to receive serial US exhibited a higher percentage of abnormal ultrasound findings, including B lines (85.3% vs. 74%), consolidation (37.3% vs. 27.9%), pleural effusion (45.1% vs. 27.9%), and severely reduced cardiac function (13.7% vs. 6.7%), in comparison to the control group.
Outcomes:
The primary outcome, which focused on reducing dyspnea as measured on a visual analog scale, raises some concerns. While the reduction of dyspnea is technically a patient-oriented outcome, the use of VDS introduces subjectivity. Additionally, the investigators established that the minimally clinically important difference on the VDS was 1 point, but it is unclear how clinically relevant a 1-point difference truly is.
Due to the nature of the intervention, both participants and physicians were aware of enrollment allocation. Patients in the serial US arm might have rated their dyspnea more favorably on the VDS due to what the authors called the “ultrasound assessment placebo effect.” Like the Hawthorne effect, patients may have changed their behavior due to the perception of receiving “special treatment.”
Similarly, physicians potentially exhibited different behaviors. As mentioned earlier, patients in the serial US arm had more conditions suitable for diagnosis with US and likewise had more abnormal findings. Consequently, these patients received more aggressive medical therapy. Notably, the serial US group received diuretic doses 6–8 times greater at 2 and 4 hours from inclusion compared to the control group. Surprisingly, there is no discussion regarding potential adverse reactions to such a substantial increase in diuretic dosage.
Interestingly, though not statistically significant, patients in the serial US group had a longer length of hospital stay compared to the control group (4d vs 3d). Additionally, the serial US group had higher hospital readmission rates on days 0-7 and 8-30 without improving hospital or 30-day mortality. Perhaps these patients were sicker, or possibly the higher diuretic dosages led to downstream consequences such as acute renal injury, hypokalemia, etc., thus prolonging their hospital stay and increasing readmission rates.
The authors emphasize the differences in VDS scores between the two groups. However, there were no statistically significant differences in more clinically relevant secondary outcomes, such as hospital length of stay, readmission rates, and mortality.
Bias:
The enrolled cohort represents a convenience sample, and investigators can introduce selection bias by enrolling participants who are not representative of the target population, ultimately affecting the study’s outcomes. Additionally, recruitment for the study had limitations, as it depended on the availability of one of two investigators, and patients were only recruited on 72% of the days when the trial was active. Furthermore, although the trial was registered as a multicenter study, the vast majority of patients were actually recruited from a single center.
Patients in the serial US group were more likely to have conditions detectable with US. In small trials, despite randomization, there is a risk of persistent imbalances in patient characteristics. Furthermore, co-intervention bias is essentially written into the study protocol. Ideally, in a clinical trial, we want patients in the intervention and comparator groups to be identical except for the studied intervention—serial US. However, in this paper, physicians understandably acted on the US findings, and patients in the serial US arm received many more diuretics.
Open-label trials, like the one conducted here, can introduce bias at various stages of the study. To minimize the impact of bias in open-label trials, it’s essential to use objective outcomes whenever possible. While the secondary outcomes were relatively objective, the primary outcome, which relied on a visual analog scale, remained subjective, potentially introducing bias into the results.
Pragmatic use of US:
The lung US protocol in this study required scanning eight zones of the chest, along with a focused cardiac US. Performing these exams four times for a single patient can be quite time-consuming. Considering the fast-paced environment of a busy ED, it’s challenging to identify where physicians would find the extra time needed to conduct these extensive studies. In real-world clinical practice, outside the context of a clinical trial, it’s likely that physicians would streamline their approach and perform only the most crucial scans on a select group of patients where it’s clear that the additional US images would impact their management.
Author’s Conclusion: “Our study establishes that serial cardiopulmonary PoCUS serves as an effective treatment guide for patients with dyspnea, offering valuable support alongside standard care to alleviate the discomfort linked to dyspnea. Notably, the observed impact is predominantly found in patients with acute heart failure. These findings endorse the use of serial cardiopulmonary PoCUS as a beneficial tool in managing dyspnea, with particular attention to patients with acute heart failure.”
Clinical Bottom Line:
We acknowledge the potential benefits of cardiopulmonary PoCUS in managing dyspnea, yet this trial presents several limitations that raise questions about the utility of serial ultrasound in dyspneic ED patients. Notably, patients in the serial ultrasound group had conditions more amenable to ultrasound diagnosis and received significantly higher diuretic doses, which likely influenced the primary outcome. Furthermore, the reliance on subjective outcomes in an open-label trial introduces bias, possibly inflating the study’s findings. Lastly, the practicality of dedicating the necessary time for multiple ultrasound examinations on a single patient in a busy ED may pose challenges for emergency physicians. While the study highlights some potential benefits of serial PoCUS, further investigation is warranted to establish its clinical utility and address these limitations.
References:
- Arvig MD, Lassen AT, Gæde PH, et al. Impact of serial cardiopulmonary point-of-care ultrasound exams in patients with acute dyspnoea: a randomised, controlled trial. Emerg Med J. 2023;40(10):700-707. PMID: 37595984.
- Arabiat M, Foderaro AE, Levinson AT. Lung Ultrasound for Diagnosing Patients with Severe Dyspnea and Acute Hypoxic Respiratory Failure. R I Med J (2013). 2019 Dec 2;102(10):34-38. PMID: 31795532.
- Cardinale L, Volpicelli G, Binello F, Garofalo G, Priola SM, Veltri A, Fava C. Clinical application of lung ultrasound in patients with acute dyspnea: differential diagnosis between cardiogenic and pulmonary causes. Radiol Med. 2009 Oct;114(7):1053-64. Epub 2009 Aug 20. PMID: 19697100.
- Staub LJ, Biscaro RRM, Kaszubowski E, et al. Lung ultrasound for the emergency diagnosis of pneumonia, acute heart failure, and exacerbations of chronic obstructive pulmonary disease/asthma in adults: a systematic review and meta-analysis. J Emerg Med. 2018;56(1):53-69. PMID: 30314929.
- Laursen CB, Sloth E, Lassen AT, Christensen Rd, Lambrechtsen J, Madsen PH, Henriksen DP, Davidsen JR, Rasmussen F. Point-of-care ultrasonography in patients admitted with respiratory symptoms: a single-blind, randomised controlled trial. Lancet Respir Med. 2014 Aug;2(8):638-46. Epub 2014 Jul 3. PMID: 24998674.
- Kok B, Wolthuis D, Bosch F, van der Hoeven H, Blans M. POCUS in dyspnea, nontraumatic hypotension, and shock; a systematic review of existing evidence. Eur J Intern Med. 2022 Dec;106:9-38. Epub 2022 Aug 1. PMID: 35927185.
- Wysham NG, Miriovsky BJ, Currow DC, Herndon JE 2nd, Samsa GP, Wilcock A, Abernethy AP. Practical Dyspnea Assessment: Relationship Between the 0-10 Numerical Rating Scale and the Four-Level Categorical Verbal Descriptor Scale of Dyspnea Intensity. J Pain Symptom Manage. 2015 Oct;50(4):480-7. Epub 2015 May 22. PMID: 26004401.
- Heijl C, Mokhtari A, Labaf A, Dryver E, Blomqwist L, Gustav Smith J. Lungultraljud – en uppseglande metod vid dyspné och hjärtsvikt [Lung ultrasound promising method for assessing acute dyspnea and monitoring decompensated heart failure]. Lakartidningen. 2021 June 8;118:20219. Swedish. PMID: 34105735.
For More Information, Check Out:
- Michael Prats. Dedicated POCUS Team vs Primary Team in Undifferentiated Dyspnea. Ultrasound G.E.L. Podcast Blog. Published on February 28, 2022. Accessed on September 28, 2023. Available at https://www.ultrasoundgel.org/121.
- Michael Prats, MD. Integrated Lung Ultrasound for Acute Decompensated Heart Failure. Ultrasound G.E.L. Podcast Blog. Published on March 4, 2019. Accessed on September 28, 2023. Available at https://www.ultrasoundgel.org/64.
Guest Post By:
Touro College of Osteopathic Medicine
Middletown, New York
E-mail: jlacolla@student.touro.edu
Stephanie Midgley, MD, FPD-AEMUS, FACEP, DipABLM
Director of Clinical Ultrasound
Advanced Clinical Ultrasound Fellowship Director
Assistant Emergency Medicine Residency Program Director
Vassar Brothers Hospital, Poughkeepsie, New York
Twitter/X: @nuvancePOCUS
Marco Propersi, DO FAAEM
Vice-Chair, Emergency Medicine
Assistant Emergency Medicine Residency Program Director
Vassar Brothers Hospital, Poughkeepsie, New York
Twitter/X: @marco_propersi
Post Peer Reviewed By: Salim R. Rezaie, MD (Twitter/X: @srrezaie)
