Learning to apply the pocket ultrasound device on the critically ill: comparing six 'quick-look' signs for quality and prognostic values during initial use by novices

Biais and colleagues [1] have shown that echocardiographers can adequately perform a three-viewcardiac examination in the emergency setting using a pocket ultrasound device (PUD).We have similarly noted that an evidence-based 'quick-look', cardiac limitedultrasound examination has diagnostic and prognostic value [2], can affect medical decision-making [3], and can be successfully taught to internal medicine residents [4]. As few data describe the learning curve of ultrasound imaging with PUDs,we observed the initial quality and prognostic value of six 'quick-look' signsobtained by residents learning to use the PUD.

Internal medicine residents in an ultrasound training program [4] recorded a brief, previously described [2] cardiac limited ultrasound examination designed to detect six'quick-look' signs of left ventricular systolic dysfunction, left atrialenlargement, ultrasound lung comet (ULC) tail artifact representing interstitiallung edema, elevated central venous pressure, pleural effusion, and rightventricular enlargement on a convenience-sample of intensive care unit (ICU)patients with respiratory failure, shock, or severe cardiac disease, using a PUD(Vscan, GE Healthcare, Wauwatosa, WI, USA). An expert echocardiographer reviewed theresident-acquired images and assigned a quality score: 0 (no image), 1 (only motiondetected; off-axis), 2 ('suboptimal', poor delineation of structures), 3 ('adequate'for diagnosis of particular sign), or 4 ('optimal', good delineation of allstructures).

Only technically adequate quality views (score >2) were entered into amultivariate logistic regression combining the six signs, clinical presentation andinpatient mortality (SPSS version 12.0). A P- value <0.05 was consideredstatistically significant. The Scripps Institutional Review Board approved thestudy.

Twenty-one residents recorded 749 views on 107 critically ill patients (mean 5.1patients/resident): mean patient age of 65.2 ± 16.8 years,inpatient mortality of 25.2%, and mean quality score of 2.1 ± 1.4.Presentation, mortality and overall percentage adequate quality views were:respiratory failure (n = 55, 32.7%, 48.0%), shock (n = 16,25.0%, 51.6%) and cardiac disease (n = 36, 13.9%, 51.7%). ULC had themost adequate quality images and is the only sign that had statistically significantprognostic value in the residents’ and cardiologist’s interpretations(Table 1).

Table 1 Mortality odds ratios for 'quick-look' signs determined byresidents’ and cardiologist’s interpretations of adequatequality images

Galen cautioned against extrapolating Biais and colleagues’ data for non-expertusers [5]. As few studies address the learning curve of quick-look ultrasoundimaging tasks, this study suggests that novice users learning to use the PUD readilylearn to image ULC, which was prognostic in this ICU population. In light of asubstantial number of initially difficult psternal long-axis and subcostal views,the PUD’s most simple and immediate use may be in the rapid detection oflife-threatening pulmonary edema.

Authors' response

Cédric Carrié and Matthieu Biais

Recently developed, the new generation of PUDs made real the concept of an ultrasonicstethoscope. But at least three questions remained: first, what is the truediagnostic capacities of these PUDs; second, in which clinical settings should theybe used; and third, what is the level of competence needed for its optimal use?

After several years of experience with PUDs, we have demonstrated its reliability forgoal-directed examinations aiming to answer brief and important clinical questionsencountered by front-line physicians in the emergency setting [1, 6]. However, those examinations were performed by operators sensitized to avisual assessment of semi-quantitative pmeters. Therefore, our results could not beextrapolated for non-expert users.

Here, Mai and colleagues report their experience in implementing a trainingcurriculum dedicated to residents learning to use a PUD. Their observations are inaccordance with the literature. Previously published studies evaluated thefeasibility and the efficiency of limited training programs to reach recommendedcompetencies in basic echocardiography and general ultrasound. Most of these studieswere performed in emergency or critical care settings [7, 8]. However, the duration of theoretical and practical sessions variedconsiderably across studies, explaining the lack of uniformity and generallyaccepted standards in basic ultrasound education among emergency medicineresidents.

Thus, we insist on the need to define the specific learning curve of emergencyresidents for the acquisition of technical and cognitive skills in goal-directedemergency ultrasound. We continue to support the concept of a three-level system fortraining in ultrasound, as a limited field of competence cannot substitute for amore comprehensive imaging examination when indicated [9].