Transthoracic echocardiography (TTE) is practised in emergency departments by emergency physicians at the patient’s bedside as a routine special investigation procedure following a detailed physical examination. The purpose of our study is to evaluate the performance of TTE in emergency departments by emergency physicians by comparing the ndings obtained to those given by an echoDoppler procient cardiologist.
A randomised prospective study of consecutive patients older than 16 years in whom there was an urgent need to practise a TTE. Each patient had to undergo a double echocardiographic examination: an investigation carried out by an emergency physician followed by
an echocardiographic examination achieved by a cardiologist. An inter-rater agreement analysis was performed between the emergency physician and a board-certied cardiologist.
The concordance of the ndings obtained by the emergency physician and by the cardiologist for the visual estimation of the LVEF was Kappa = 0.82 [95% CI 0.63-1] with an agreement = 0.90 [95% CI 0.74-0.99].
e concordance for measurement of the diameter of the IVC was Kappa = 0.95 [95% CI 0.63-1] with an agreement = 0.95 [95 % CI 0.64-0.99] and for assessment of its compliance Kappa=1 with an agreement = 1. e concordance of the ndings obtained for the
diagnosis of pericardial e!usion was Kappa=0.86 [95% CI 0.71-1] with an agreement = 0.92 [95% CI 0.64-0.99] and the concordance for the detection of echocardiographic signs of compressive e!usion was Kappa = 1 with an agreement= 1.
The concordance of the ndings obtained by both operators was excellent. Emergency physicians should then be encouraged to practise TTE at the patient’s bedside. A prior training of 3 months in Doppler echocardiography is nevertheless necessary.
KEY WORDS : Transthoracic echocardiography - emergency physician - cardiologist
Emergency physicians (EPs) are routinely called on to manage critically ill patients who may present with an indeterminate or changing hemodynamic status. Early in the patient’s course, it may be di"cult to rmly identify the underlying etiology. Bedside echocardiography o!ers a noninvasive, brief, rapid method of evaluating cardiac function. It should be practised by EPs physicians as a routine special investigation procedure following a detailed physical examination. In fact it has been shown that bedside echocardiography helps emergency physicians make the correct diagnosis, treatment and enhanced disposition decision, and signicantly improve patient care. [1,2,3] is study aimed to evaluate the performance of tranthoracic echocardiography (TTE) in emergency departments by emergency physicians by comparing the ndings obtained to those given by an echoDoppler procient cardiologist.
This was a randomised prospective study carried out in the emergency department of the military hospital of Tunis (Tunisia) during the period going from 1 January 2016 to 31 December 2016. It included all patients aged > 16 years in whom there was an urgent need to practise a TTE. e patients in the study had to undergo a double echocardiographic examination: an initial echocardiographic investigation carried out by an emergency physician who had previously received a three-month training in Doppler echocardiography, followed by a subsequent investigation performed by an echo-Doppler procient cardiologist.
An inter-rater reliability analysis using the K statistics with 95% CI¹º was performed to determine the degree of agreement between the emergency physician and the cardiologist for each echocardiographic parameter. The evaluation considered the global visual estimation of the left ventricular ejection fraction (LVEF), the presence or absence of pericardial effusion (independently of the site), and the diameter and compliance of the inferior vena cava (IVC).
Two hundred and four patients were involved in the study.
The concordance of the ndings obtained by the emergency physician and the cardiologist for the visual estimation of the LVEF was Kappa = 0.82 [95% CI 0.63-1] with an agreement =
0.90 [95% CI 0.74-0.99]. The concordance for measurement of the diameter of the IVC was Kappa = 0.95 [95% CI 0.63-1] with an agreement = 0.95 [95 % CI 0.64-0.99] and for assessment of its compliance Kappa=1 with an agreement = 1.
The concordance of the ndings obtained for the diagnosis of pericardial e!usion was Kappa=0.86 [95% CI 0.71-1] with an agreement = 0.92 [95% ICI 0.64-0.99] and the concordance
for the detection of echocardiographic signs of compressive effusion was Kappa = 1 with an agreement = 1.
LV systolic function
Visual estimation of global LV systolic function by the EPs in this study showed substantial agreement with that made by the cardiologist (90%, K=0.82, n=44). is degree of agreement is comparable with those found in previous studies of emergency physician-performed echocardiography. [1,4,5] is study suggests that, even with limited training, EPs were able to use visual estimation n for global LV function with reliable accuracy. Similarly, Moore et al [4] concluded that emergency physicians with focused training in echocardiography could accurately determine left ventricular function in hypotensive adult patients.
Another study conducted by Eherman et al founds that agreement between EP-sonographers was good with 95% agreement on normal vs abnormal diastolic function, with K= 0.66 (95% CI,0.39-0.92) and weighted K = 0.77 (95% CI, 0.56-0.96) for grade of diastolic dysfunction(DD). There was also 95% agreement on presence of clinically signicant DD. This demonstrates the high reliability of the EP-performed diastolic examination, which is important because it shows that EPs were able to consistently follow a multistep protocol and reach similar conclusions. [6] Similarly to the study conducted by Bustam et al [1] our study directly compared echocardiography examinations performed by the EPs and the cardiologist immediately after one another. This provides a more direct and real-time comparison.
Furthemore, the visual technique to estimate LV function by EPs is not only accurate but also quicker to perform. In addition it saves time and costs.
Pericardial effusion
Pericardial effusion was identied by the presence of an anechoic stripe within the pericardium surrounding the heart, and was categorically assessed as either absent or present.
It is also important for emergency physicians to be able to accurately diagnose the absence of pericardial e!usion, as it can be mistaken for pericardial fat, other pericardial abnormalities or the more common pleural e!usion. [1,7,8] False positive ndings of pericardial e!usion might lead to inadvertent pericardiocentesis. [9] The detection of pericardial effusion when comparing when comparing the ndings of the EPs with those of the cardiologist was Kappa Kappa=0.86 [95% CI 0.71-1] with an agreement = 0.92 [95% CI 0.64-0.99] and the concordance for the detection of echocardiographic signs of compressive e!usion was Kappa = 1 with an agreement =1. These results are compared to other studies which reported sensitivities of emergency physiciansperformed focused echocardiography in detecting pericardial effusion of between 88% and 100%. [10,11] IVC assessment for "uid volume status IVC assessment in this study involved measuring the changes in the diameter of IVC during inspiration and expiration.
The degree of collapsibility of the inferior vena cava provides information regarding the central venous lling volume and the right atrial pressure represented by the CVP. is is done by viewing the vena cava below the diaphragm in the sagittal plane and observing the change in the IVC diameter during the respiratory cycle. During inspiration, negative intrathoracic pressure causes negative intraluminal pressure and increases venous return to the heart. The compliance of the extrathoracic IVC causes the diameter to decrease with normal inspiration.
In patients with low intravascular volume, the inspiration to expiration diameter ratios change more than in those patients who have normal or high intravascular volume, and therefore
a quick assessment of intravascular volume can be made. [1,12,13,14] IVC evaluation can be particularly helpful in those patients with a signicant respiratory collapse during inspiration, permitting prompt identication of the hypovolemic patient. [15] In our study, the concordance for measurement of the diameter of the IVC was Kappa = 0.95 [95% CI 0.63-1] with an agreement =0.95 [95 % CI 0.64-0.99] and for assessment of its compliance Kappa=1 with an agreement =1. Whereas, in the studies conducted by Bustom and Randazzo the IVC diameter and collapsibility assessment by the trainees showed only a moderate agreement with that made by the cardiologist (64.2%, K=0.45, n=95) and (68.1% K=0.41, n=94) respectively. [1,5] Limitations One of the limitations of our study was the small number of patients. Besides, the emergency physician was not blinded to the study. As he knew he was being evaluated, he may have been more motivated to diagnose and enhance his performance on the criteria being studied. Another limitation of our study is that we enrolled a sample of patients at a single institution, which likely introduced selection bias.
Emergency physicians are able to perform and interpret focused echocardiography with reliable accuracy. Emergency physicians should then be encouraged to practise TTE at the patient’s bedside. Nevertheles a practical training seems necessary.