Pocket-sized Ultrasound versus Cardiac Auscultation in Diagnosing Cardiac Valve 1 Pathologies : A Prospective Cohort Study 2 3

Background: Pocket-sized ultrasound devices are used to perform focused ultrasound studies (POCUS). We compared valve malfunction diagnosis rate by cardiac auscultation to POCUS (insonation), both conducted by medical students. 
Methods: A prospective cohort study was conducted among patients with and without clinically relevant valve dysfunction. Recruitment to the study group was based on the presence of at least one valve pathology of at least moderate severity identified on recent echocardiography study that was required for clinical reasons. Three final-year medical students examined the patients. Each patient underwent auscultation and a POCUS using a pocket-sized ultrasound machine. Sensitivity was defined as the percentage of patients correctly identified as having a valve disorder. Specificity was defined as correct identification of the absence of valve pathology. 
Results: The study included 56 patients. In 18 (32%), no valve pathology was found. Nineteen patients (34%) had at least two valvular pathologies. Sixty valve lesions were present in the whole cohort. Students' sensitivity for detecting any valve lesion was 32% and 64% for auscultation and insonation; respectively, specificity was similar.  
 The sensitivity for diagnosing mitral regurgitation, mitral stenosis, and aortic regurgitation rose significantly by using POCUS compared to auscultation alone. When using POCUS, Students identified valve pathologies in 22 cases (39%) from the patients with at least two valve dysfunctions, and none when using auscultation. 
Conclusions: Final-year medical students' competency to detect valve dysfunction by performing cardiac auscultation is poor. Cardiac ultrasound-focused training significantly improved medical students' sensitivity for diagnosing a variety of valve pathologies. 


18
As such, ultrasound is gradually being incorporated into the curriculum of medical schools 19 worldwide (11). Finally, insonation meaning "exposure to or the use of ultrasound" has been 20 proposed to become the fifth pillar of the physical examination after inspection, percussion, 21 palpation and auscultation (12).

23
To date, assessment of the additional value of insonation for diagnosing left-sided valvular 24 dysfunction has been evaluated on patients with single valvular lesions. In the current study, we 25 aim to compare auscultation to insonation in the diagnosis of valve malfunction in a population in 26 whom part of them had multiple valve lesions performed by medical students after a relatively short 27 training in cardiac ultrasound. We hypothesized that insonation will outperform auscultation in the 28 diagnosis of valvular pathologies. The study population. Three students in their final year of medical school received 12 hours of 2 training on the operation of a pocket-size ultrasound device (PUD) in order to diagnose common 3 valve disorders. The three students were part of a pilot study with the purpose of evaluating the 4 convenience of implementing this type of course as part of a one-week clerkship in cardiology. The 5 students were not picked by their performance or by their grades but rather arbitrarily. The training 6 process took place in a series of two-hours sessions over the course of approximately a month, 7 beginning with a one-hour lecture on the physics of ultrasound, cardiac ultrasound anatomy, and 8 the examination technique. Next, there was a three-hours bedside, guided lesson on main cardiac 9 ultrasound views, identifying anatomic points, and a two-hours review of normal and abnormal 10 echocardiographic cases focused on valve pathologies in the echocardiography lab. These were 11 followed by one hour of hands-on exercise using PUD under the guidance of an echocardiography 12 technician and seven additional hours of practice on volunteer healthy subjects. Prior to the 13 initiation of the study, the students listened to sound characteristics of murmurs on a Blaufuss 14 sound builder website under supervision and explanation by the principal investigator.

16
The students were proficient in cardiac auscultation that had been taught in the previous years and 17 used it as part of the physical examination they performed in different teaching scenarios during 18 the last three years of the medical school.

20
The session on auscultation took an hour and focused on the recognition of the individual 21 pathologies and the characteristics that allow the examiner to differentiate pathologies that cause 22 systolic and diastolic murmurs. The auscultatory skills of the students were not assessed prior to 23 the initiation of the study.

25
The recruitment of subjects was conducted through the Cardiology Section at Soroka Medical

26
Center. Recruitment was based on the presence of at least one valve pathology of at least 27 moderate severity identified on recent echocardiography study that was required for clinical 28 reasons. A control group of subjects without valve disease was recruited as well and was matched 29 by gender and age. Echocardiography is the most efficient tool to diagnose valve disease; 30 accordingly, we use it as the gold-standard method to compare students' ability to diagnose valve 31 disease and rather than the physical examination of expert clinicians which, when based on 32 auscultation, can misdiagnose almost half of the clinically significant valve diseases (2,11,12).

33
The nature of the study and the examinations was explained to all the research subjects, and they 34 signed an informed consent form. The study was approved by the local ethics committee. The Device. The miniaturized device used was the General Electric Vscan ultrasound device, 1 measuring 28 × 73 × 135 mm. The combined weight of the device and transducer is 390 grams.

2
The monitor of the device is 3.5 inches wide, with a resolution of 320×240 pixels, and provides two-3 dimensional and conventional color Doppler, but lacks spectral Doppler. The device is able to save 4 still images and videos in a flash-card memory.

4
In the performance of the model, adjustments were made for tests conducted on the same patient, 5 as well as by the same operator. Variables with two-sided p value < 0.1 in the univariate analysis 6 or as clinically relevant were introduced into the multivariate analysis including age, body mass 7 index, gender, type of valve pathology and severity. A two-sided p-value <0.05 was considered 8 significant.

10
Sample size considerations were as follows: according to study hypothesis, echocardiography has 11 better sensitivity and specificity of finding valve pathology, in comparison to basic physical exam 12 using stethoscope. Basic physical exam sensitivity and specificity is approximately 50%. We 16 17

1
The study included a total of 56 subjects who were examined by the three medical students. The 2 characteristics of the subjects are presented in Table 1. Of the total number of subjects, 18 had no 3 valve pathology and 38 had at least one ³ moderate valve pathology, 19 of them having more than

36
However, better sensitivity (70%) was demonstrated by POCUS, with only a slight drop in specificity 1 (87%) The accuracy was 82% and 85% for insonation and auscultation, respectively. It should be 2 noted that with the use of POCUS, a wide range of level of sensitivity among the three students 3 was apparent, seen as well with auscultation (Table 3)  14 15 16 1 Our study demonstrates that when students based their diagnosis of valve dysfunction on cardiac 2 auscultation, their performance was poor (mean sensitivity 32%, mean specificity 86%), particularly 3 for identifying valve pathologies that cause a diastolic murmur (mean sensitivity 7% and mean 4 specificity 95%). Students noticeably improved their diagnostic ability with the use of POCUS 5 (mean sensitivity 64%, mean specificity 83%). However, the accuracy rate remains unchanged 6 between auscultation-based and insonation-based diagnosis of the left-side valve lesions, except 7 for MR in which insonation has better sensitivity, specificity, and accuracy than auscultation. It is 8 obvious that auscultation's specificity can be outstanding if the sensitivity of the method is so low.

9
These data on the diagnostic rate of cardiac auscultation is similar to the results of historical studies 10 that exist in the field, and have not improved for the last two decades, despite the fact that the 11 innovative methods based on high quality audio and self-study techniques are widely available (1-

17
A serious concern which arises from our study as well as from Stokke et al study is that even when 18 testing only moderate or severe valve dysfunction, students' diagnoses were poor when relying on 19 cardiac auscultation (mean sensitivity 35%) and improved considerably using POCUS (mean  13 physical examination, including cardiac auscultation. We believe that POCUS can be used as an 1 instrument to improve auscultatory skills by providing immediate confirmation or rejection of the 2 auscultatory findings. This feedback is essential for the learning process.

4
The main barriers in incorporating POCUS into the medical school curriculum are time that is added 5 into the busy curriculum for a new course, the necessity of sufficient instructors to teach a growing 6 number of students in small groups, and financial issues related to the cost of the ultrasound 7 devices and cost of the instructions' teaching time (14). Our experience has demonstrated that 8 some of these limitations can be overcome by incorporating students as instructors of their

22
Another limitation relates to the imaging quality of POCUS examination that was not graded.

23
However, none of the recruited subjects was discarded from the analysis due to poor POCUS 24 imaging. Finally, the three students in the study were recruited based on their willingness to 25 participate in a research project; we did not assess before their participation their diagnostic skills.

26
They received the same instruction, and we cannot explain the differences in students' results,