Skip to main content
The Ultrasound Journal
The Ultrasound Journal Cover Image
Download PDF

Impact of an ultrasound-driven diagnostic protocol at early intensive-care stay: a randomized-controlled trial

The Ultrasound Journal volume 11, Article number: 24 (2019) Cite this article

Abstract

Background

Point-of-care ultrasound (POCUS) is a tool in increasing use, but there is still a lack of basics for its routine use and evidence of its impact in intensive care.

Objective

To measure the impact of POCUS on resource utilization, diagnostic accuracy, and clinical management in medical-surgical intensive care units (ICUs).

Methods

Prospective, controlled study, in two polyvalent ICUs. The patients were randomly assigned to POCUS or control group.

Interventions

POCUS patients received systematic ultrasound examination of optic nerve, lung/pleura, heart, abdomen, and venous system, performed at the bedside by trained intensivists. Control patients were treated by critical care specialists who do not perform ultrasound in their clinical practice.

Results

We included 80 patients, 40 per group. There were no significant differences in age, sex, APACHE II score, or admission diagnosis. POCUS group used fewer resources per patient in the first 5 days of hospitalization: chest radiography (2.6 ± 2.0 vs 4.1 ± 3.5, P = 0.01), additional ultrasound evaluations performed by a radiology specialist (0.6 ± 0.7 vs 1.1 ± 0.7, P = 0.002), and computed tomography studies (0.5 ± 0.6 vs 0.9 ± 0.7, P = 0.007). Time to perform any requested ultrasound evaluation after ICU admission was 2.1 ± 1.6 h versus 7.7 ± 6.7 h (P = 0.001). Systematic ultrasound evaluation led to better characterization of ICU admission diagnosis in 14 (35%) patients and change in clinical management in 24 (60%). POCUS group had lower fluid balance at 48 and 96 h after admission (P = 0.01) and spent less time mechanically ventilated (5.1 ± 5.7 days vs 8.8 ± 9.4, P = 0.03).

Conclusions

Systematic application of POCUS may decrease utilization of conventional diagnostic imaging resources and time of mechanical ventilation, and facilitate meticulous intravenous fluid administration in critically ill patients during the first week of stay in the ICU.

Trial registration ClinicalTrials.gov Identifier: NCT03608202.

Introduction

Ultrasonography is currently considered almost essential in the management of patients in shock, acute respiratory failure, or multiorgan dysfunction [1,2,3,4]. There is still a broad field of research aimed at generating a transformational change based on scientific evidence in critical care ultrasound [5, 6]. Given the inherent limitations for an accurate physical examination of the critically ill, imaging modalities such as chest, abdominal radiography, and computed tomography (CT) are considered more reliable, useful, and impactful in the critical care setting. However, their portability, versatility, and costs are less than ideal [1], with unnecessary delays in obtaining a diagnosis in the critically ill, and increased risk for hypoxemia or refractory arterial hypotension during transportation [3]. In contrast, ultrasound, a widely available, simpler technique that can be performed with portable equipment at the bedside, has gained progressive acceptance [4].

Recent investigations have demonstrated that multiorganic ultrasound decreases resource utilization and guides clinical management of critical ill patients while enhancing diagnostic accuracy [7,8,9,10,11]. Nevertheless, these previous publications lack a randomized and controlled trial methodology. We hypothesize that routine use of methodic, multisystemic ultrasound protocol in critically ill patients at admission to the intensive-care unit (ICU) reduces utilization of diagnostic resources and consultation and time to definitive diagnosis and tracheal extubation. Thus, the aims of our investigation were to analyze diagnostic and therapeutic implications of a point-of-care ultrasound (POCUS)-driven protocol during the first 5 days of ICU admission.

Materials and methods

Study design

Our prospective, randomized, controlled study was conducted at two major referral hospitals, Asociación Española Primera de Socorros Mutuos and Hospital Pasteur, Administración de Servicios de Salud del Estado. Patients admitted to a 12-bed medical-surgical ICU and 8-bed stepdown unit in those institutions were included in the study. The study protocol was approved by the Institutional Ethics Committee, and informed consent was obtained prior to each patient’s enrollment.

Patients

We screened patients between February 1, 2017 and May 31, 2017, admitted to the ICUs. Inclusion criteria to the study were: (1) 18 years of age or older; (2) mechanical ventilation (MV) required at ICU admission. Patients with more than 24 h of hospitalization and those considered with low likelihood of survival by the treating intensivist at ICU admission were excluded.

Clinical data

We collected pertinent clinical data, including Acute Physiology and Chronic Health Evaluation, second version (APACHE II) score at admission, reason for admission to the ICU, and other relevant physiologic data.

The Distensibility index of inferior vena cava (%) was utilized as the primary measure for fluid responsiveness based on a previous investigation [12] (Table 1). In addition, we routinely used invasive MAP monitoring, capillary filling, hourly diuresis, central venous saturation, arteriovenous CO2 difference, and plasma lactate measurement alongside measure of cardiac output and LVEF as supportive parameters of the clinical-decision making of administering intravenous fluids to patients suffering from shock.

Table 1 POCUS protocol
Full size table

Intervention

Two groups were compared; the study group received routine evaluation with an ultrasound-driven protocol (POCUS group), while a control group received conventional management according to the pre-established protocols in the ICU (in sepsis, pneumonia, postoperative care, heart failure and myocardial ischemia, trauma, COPD, etc.).

Patients were randomly assigned by permuted blocks to either the POCUS group or the control group.

The POCUS protocol is shown in Table 1; the following variables were evaluated: confirmation of initial diagnosis; change in initial diagnosis; new unsuspected finding; lack of change in initial diagnosis, inability to rule out a condition, and no changes in treatment; association with wrong diagnosis; and association with subsequent pharmacologic, medical, procedural, or surgical measures. In addition, these resource- and outcome-related variables were analyzed for comparison between the two groups: number of additional radiologic, ultrasound, and CT studies; number and type of procedures or interventions ordered; time to for ultrasound evaluation; duration of MV; length of stay in the ICU; and mortality rate.

POCUS studies were conducted with Logiq-e (General Electric Healthcare Japan Corporation) system with digital image storage capacity. Clinical data were saved in the SPSS Statistics 23 program (International Business Machines Corp.).

Statistical analysis

With a sample size of 40 patients per group for an annual population of 400 patients admitted to each ICU, we estimated a margin of error of 14% (95% CI). The qualitative variables were records as percentages and the quantitative as mean ± standard deviation. For the comparison of mean between groups, either the Student’s t test for independent groups or Mann–Whitney test was applied according to the case; and binaries in contingency tables by χ2 or Fisher’s exact test. A P value < 0.05 was considered significant in all tests.

Results

Patient demographics

We included 80 patients, 40 in each group. In the POCUS group, 24 were men (61%) and 16 were women (39%); similarly, in the control group, there were 25 men (62%) and 15 women (38%). There were no statistical differences between the groups regarding age, sex, APACHE II score at 24 h after admission, or reason for admission to ICU (Table 2).

Table 2 Demographic data and diagnosis at admission in each group
Full size table

Clinical information or decision-making following POCUS

The POCUS protocol was performed in 15–30 min in all cases.

In the POCUS group, 48 clinical decisions were made in 36 of 40 patients (90%). Furthermore, 35% of these observations were related to the diagnostic category, and either due to new or changed diagnosis. Modifications in pharmacologic therapy occurred in 24 patients (60%), and invasive maneuvers or procedures were necessary in 9 (23%) (Table 3).

Table 3 Description of changes in clinical information or decisions led by ultrasound
Full size table

Utilization of additional diagnostic testing

Overall, there was lower utilization of resources per patient during the first 5 days of ICU stay in the POCUS group compared with the control group (Table 4): lower number of chest radiography requested (2.6 ± 2.0 vs 4.1 ± 3.5, respectively, P = 0.01); lower utilization of additional ultrasound ICU imaging per radiology specialist (0.6 ± 0.7 vs 1.1 ± 0.7, respectively, P = 0.002); and fewer CT scans performed (0.5 ± 0.6 vs 0.9 ± 0.7, respectively, P = 0.007).

Table 4 Comparison of resource utilization, length of stay and mortality in ultrasound-driven evaluation versus conventional management in ICU
Full size table

Analysis of delay times

We measured time from admission to first ultrasound (delay time). The delay in performing diagnostic ultrasound was significantly reduced in the POCUS group versus control (2.1 ± 1.6 h vs 7.7 ± 6.7, P = 0.001). In the control group, all diagnostic ultrasound evaluations were performed by radiology specialists available via telephone.

Clinical outcomes

Fluid management

The POCUS-driven decision of fluid management in each patient was left at treating intensivist discretion. Similarly, clinical decisions on fluid, inotropic, or vasopressor therapy, and adjustments to MV were made based on POCUS assessment. Fluid balance (FB) was much lower in the POCUS group than in the control at 48 and 96 h after ICU admission (− 1600 ± 2750 mL vs − 400 ± 2130 mL, P = 0.03, and − 3200 ± 3510 mL vs − 890 ± 3900 mL, P = 0.006, respectively). There was a significant positive linear correlation between left-ventricular ejection fraction (LVEF) and FB at 48 (Pearson correlation r = 0.57, P = 0.002) (Fig. 1) and 96 h (r = 0.58, P = 0.03).

Fig. 1
figure1

Linear correlation between fluid balance at 48 h and left-ventricular ejection fraction. Pearson correlation r = 0.57, P = 0.002. FB fluid balance, LVEF left-ventricular ejection fraction

Full size image

Duration of MV

The POCUS group showed a lower duration of MV than the control (5.1 ± 5.7 days vs 8.8 ± 9.4 days, P = 0.03) (Table 4). There was no significant correlation between the time spent on MV and FB, or between the time spent on MV and LVEF.

Length of stay in ICU and mortality

Mortality rates did not differ between POCUS and control groups. Of note, there was a tendency for decreased ICU length of stay with POCUS (Table 4).

Discussion

We demonstrated the direct impact of a routine POCUS-driven protocol in patients during the early phase after admission to medical–surgical ICUs. A systematic application of POCUS resulted in decreased utilization of conventional diagnostic imaging resources and time of MV, and facilitated a judicious intravenous fluid administration in critically ill patients during the first week of ICU hospitalization. Application of routine POCUS appears to be safe if utilized by appropriately trained intensivist as no complications were directly associated with POCUS use in the study group.

Previous investigations [2, 5] applying particular ICU-sound protocol led to more accurate admission diagnosis in 25.6% to 24.9%. They had an observational design, but are comparable to the 22% accuracy rate of new or corrected diagnoses in our study. Moreover, the majority of our patient population suffered from sepsis. Manno et al. [2] proposed that septic patients may be the most favored subgroup of patients receiving a POCUS protocol because of the myriad of information and therapies that can be tailored to each individual patient.

The main effect of routine utilization of our POCUS protocol was the clinical decision-making, more specifically in pharmacologic management adjustments in shock patients (60%) and ordering of ultrasound-guided procedures (23%). In our study, we found that 23% of our population required ultrasound-guided invasive procedures, which is similar to the 21.6% reported by the Manno group [2]. Vignon et al. [13] described a change in therapeutic management of patients in 51% of the cases, and Bernier-Jean et al. [5] in 44%.

Management of intravascular volume status can be a difficult decision at the bedside. Meticulous fluid management in treating shock patients with heart disease is desirable. In our case, the most common clinical decision was related to fluid management. Similar findings have been described in an observational study [13], but we emphasize the randomized-controlled design of our study. Nonetheless, our direct comparison of timely ultrasound-driven versus conventional management allows us to represent the current practice in the majority of medical centers due to lack of routine utilization of ultrasound by intensivists. Furthermore, positive fluid responsiveness test does not indicated necessity of further fluid administration, our clinical-decision making was determined by the echocardiography measurements (LVEF, IVC distensibility index) in addition to systemic perfusion surrogates (central venous saturation, arteriovenous CO2 difference, and plasma lactate measurements). Finally, in cases of heart failure, we guided restriction of fluids by signs of pulmonary and systemic congestion. However, we acknowledge tolerance to fluids must be assessed with diastolic function parameters and may be in addition to lung ultrasonography [14,15,16].

Our POCUS protocol led to a significantly lower utilization of chest radiography, ultrasound performed by non-intensivist specialists, and CT scans (up to 56% reduction in CT requests). We found a reduction of radiology evaluation utilization from 4.1 in the control group to 2.6 in the POCUS group (63% reduction). This finding differs from other groups reporting 22% to 26% reduction in radiography utilization [17, 18]. Possible explanations include intensivist discretion versus specific protocol, considerable critical care ultrasound expertise by attending intensivist at time of the study, and more availability of fully capable ultrasound devices. Of note, the American College of Radiology expert panel recommended daily radiology for each intubated patient [17]. The direct clinical implications of our findings were associated with lower radiation exposure, less intra-hospital transportation of unstable patients, and inherent economic savings with more sophisticated evaluations, including in improved imaging suites workflow and reduced necessity for personnel to transport patients [19,20,21].

We found a significant correlation between cardiac function (left systolic ventricle function measured by LVEF index) and a negative FB in the first 5 days of ICU management. This might indicate that judicious fluid management and decisions in administration of diuretic agents are facilitated with real-time utilization of POCUS since ICU admission. We did not find a significant correlation between time spent on MV and FB or between duration of MV and LVEF. This can possibly be explained by the heterogeneity of our population. However, there was correlation between lower LVEF and negative FB, since patients with low LVEF are likely to receive less fluids and more diuretics.

We acknowledge several limitations in the current investigation. First, our small patient population avoids finding significant differences in meaningful clinical outcomes, such as mortality. Second, we analyzed only the first 5 days of ICU stay. However, as patients stay longer in the ICU, additional confounding factors not necessarily related with admission diagnosis can affect the outcomes (late-onset, ventilator-associated pneumonia). Mortality was similar between both groups, but it should be noted that future studies with larger patient populations having a shorter time of MV as found in the POCUS group could potentially determine a decrease in mortality related to MV complications. Third, we did not find a significant correlation between normal or high ejection fraction and pulmonary B profile in our study that can imply possible tolerance to fluid administration. However, patients with CHF and preserved EF can be only characterized with advanced echocardiography parameters of diastolic function. We recognize that the lack of diastolic function assessment is another limitation of this study. However, we aimed to evaluate a practical approach for assessment of ventricular function and fluid responsiveness, so we propose further investigation of the correlation between normal or high ejection fraction and pulmonary B profile in addition to diastolic function assessment in septic shock patients.

Finally, the absence of a reposition protocol guided by pre-established ultrasound should be endorsed in multicentric studies.

Conclusions

A systematically applied POCUS protocol upon admission to the ICU can guide diagnostic and therapeutic decisions in critically ill patients. This potentially leads to greater resourcefulness and shorter duration of MV. Definitive associations, such as ultrasound-driven fluid resuscitation and improved outcomes, remain to be proven in larger multicenter studies. Diastolic function assessment can improve overall evaluation of fluid tolerance to the fluid management of septic shock patients.

Availability of data and materials

The data sets used and/or analysed during the current study are available from the corresponding author on reasonable request.

References

  1. 1.

    Frankel HL, Kirkpatrick AW, Elbarbary M et al (2015) Guidelines for the appropriate use of bedside general and cardiac ultrasonography in the evaluation of critically ill patients—part I: general ultrasonography. Crit Care Med 43:2479–2502

    Article  Google Scholar 

  2. 2.

    Manno E, Navarra M, Faccio L et al (2012) Deep impact of ultrasound in the intensive care unit: the “ICU-sound” protocol. Anesthesiology 117:801–809

    Article  Google Scholar 

  3. 3.

    Moore CL, Copel JA (2011) Point-of-care ultrasonography. N Engl J Med 364:749–757

    CAS  Article  Google Scholar 

  4. 4.

    Narasimhan M, Koenig SJ, Mayo PH (2016) A whole-body approach to point of care ultrasound. Chest 150:772–776

    Article  Google Scholar 

  5. 5.

    Bernier-Jean A, Albert M, Shiloh AL, Eisen LA, Williamson D, Beaulieu Y (2017) The diagnostic and therapeutic impact of point-of-care ultrasonography in the intensive care unit. J Intensive Care Med. 32:197–203

    Article  Google Scholar 

  6. 6.

    Bossone E, DiGiovine B, Watts S et al (2002) Range and prevalence of cardiac abnormalities in patients hospitalized in a medical ICU. Chest 122:1370–1376

    Article  Google Scholar 

  7. 7.

    Kristensen MS, Teoh WH, Graumann O, Laursen CB (2014) Ultrasonography for clinical decision-making and intervention in airway management: from the mouth to the lungs and pleurae. Insights Imaging. 5:253–279

    Article  Google Scholar 

  8. 8.

    Laursen CB, Sloth E, Lassen AT et al (2014) Point-of-care ultrasonography in patients admitted with respiratory symptoms: a single-blind, randomised controlled trial. Lancet Respir Med. 2:638–646

    Article  Google Scholar 

  9. 9.

    Oks M, Cleven KL, Cardenas-Garcia J et al (2014) The effect of point-of-care ultrasonography on imaging studies in the medical ICU: a comparative study. Chest 146:1574–1577

    Article  Google Scholar 

  10. 10.

    Pontet J, Yic C, Diaz-Gomez J, et al. Ultrasonografía En Medicina de Urgencias, Terapia Intensiva Y Anestesia. 1st ed: Montevideo; 2016

  11. 11.

    Volpicelli G, Lamorte A, Tullio M et al (2013) Point-of-care multiorgan ultrasonography for the evaluation of undifferentiated hypotension in the emergency department. Intensive Care Med 39:1290–1298

    CAS  Article  Google Scholar 

  12. 12.

    Feissel M, Michard F, Faller JP, Teboul JL (2004) The respiratory variation in inferior cava diameter as a guide to fluid therapy. Intensive Care Med 30(9):1834–1837

    Article  Google Scholar 

  13. 13.

    Vignon P, Frank MB, Lesage J et al (2004) Hand-held echocardiography with Doppler capability for the assessment of critically-ill patients: is it reliable? Intensive Care Med 30:718–723

    Article  Google Scholar 

  14. 14.

    Geri G, Vignon P, Aubry A, Fedou AL, Charron C, Silva S, Repessé X, Vieillard-Baron A (2019) Cardiovascular clusters in septic shock combining clinical and echocardiographic parameters: a post hoc analysis. Intensive Care Med 45(5):657–667

    Article  Google Scholar 

  15. 15.

    Nair S, Sauthoff H (2019) Assessing extravascular lung water with ultrasound: a tool to individualize fluid management? J Intensive Care Med. https://doi.org/10.1177/0885066619855000 (Epub ahead of print)

    Article  PubMed  Google Scholar 

  16. 16.

    Orme RM, Oram MP, McKinstry CE (2009) Impact of echocardiography on patient management in the intensive care unit: an audit of district general hospital practice. Br J Anaesth 102:340–344

    Article  Google Scholar 

  17. 17.

    Expert Panel on Thoracic Imaging, McComb BL, Chung JH et al (2016) ACR appropriateness criteria routine chest radiography. J Thorac Imaging. 31:W13–15

    Article  Google Scholar 

  18. 18.

    Krinsley JS (2003) Test-ordering strategy in the intensive care unit. J Intensive Care Med. 18:330–339

    Article  Google Scholar 

  19. 19.

    Expert Round Table on Echocardiography in ICU (2014) International consensus statement on training standards for advanced critical care echocardiography. Intensive Care Med 40:654–666

    Article  Google Scholar 

  20. 20.

    Kue R, Brown P, Ness C, Scheulen J (2011) Adverse clinical events during intrahospital transport by a specialized team: a preliminary report. Am J Crit Care. 20:153–161 (quiz 162)

    Article  Google Scholar 

  21. 21.

    Smith I, Fleming S, Cernaianu A (1990) Mishaps during transport from the intensive care unit. Crit Care Med 18:278–281

    CAS  Article  Google Scholar 

Download references

Acknowledgements

Stella Calvo, MD for his technical help in the elaboration of the manuscript.

Preliminary results and abstract presented at 13th WINFOCUS World Congress, Argentina, September 2017, and received the Golden Probe Award as Best Poster Presentation.

Funding

There was no source of funding.

Author information

Affiliations

  1. Intensive Care Unit, Hospital Pasteur, Administración de Servicios de Salud del Estado, Montevideo, Uruguay

    Julio Pontet, Pablo Rodriguez, Igor Sviridenko, Sylvia Noveri & Ana Soca

  2. Intensive Care Unit, Asociación Española Primera de Socorros Mutuos, Montevideo, Uruguay

    Julio Pontet, Christian Yic, Diego Méndez & Mario Cancela

  3. Department of Critical Care Medicine, Mayo Clinic, Jacksonville, USA

    José L. Díaz-Gómez

  4. Larravide 2458, Montevideo, Uruguay

    Julio Pontet

Authors
  1. Julio Pontet
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Christian Yic
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. José L. Díaz-Gómez
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Pablo Rodriguez
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Igor Sviridenko
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Diego Méndez
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Sylvia Noveri
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Ana Soca
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Mario Cancela
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

JP and CY have made substantial contributions to the conception and design of the work, the acquisition, analysis, and interpretation of data, and have drafted the work. JLD have made contributions to the analysis and interpretation of data and substantively revised it. PR, IS, and DM have made contributions to the acquisition and analysis of data. SN, AS, and MC have made contributions to the analysis and interpretation of data and substantively revised it. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Julio Pontet.

Ethics declarations

Ethics approval and consent to participate

The study protocol was approved by the Institutional Ethics Committee (Hospital Pasteur), and informed consent was obtained prior to each patient’s enrollment.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Pontet, J., Yic, C., Díaz-Gómez, J.L. et al. Impact of an ultrasound-driven diagnostic protocol at early intensive-care stay: a randomized-controlled trial. Ultrasound J 11, 24 (2019). https://doi.org/10.1186/s13089-019-0139-2

Download citation

Keywords

  • Clinical impact
  • Critical care
  • Diagnostic techniques and procedures
  • Point-of-care ultrasound
  • Point-of-care systems
  • Ultrasound protocol