a b s t r a c t

Objectives: This study is to present the diagnostic values of the novel sonographic visualization of the inferoposterior thoracic wall (VIP) and boomerang signs in detecting right pleural effusion by sonologists with little to no experience in ultrasound.

Methods: A Prospective analysis of a convenience sample of patients who were assessed by junior intensive care physicians was performed. The patients all underwent computed tomography (CT) of the chest or abdomen with lung bases as part of their care regardless of indication; the results were interpreted by radiologists and were con- sidered the gold standard. Sonography was performed to assess for the presence of the VIP and boomerang signs. Sonographic and chest radiographic findings were compared against CT results.

Results: 73 patients were enrolled. The sensitivity and specificity for the VIP sign were 0.85 (95% confidence in-

terval [CI], 0.67-0.94) and 0.86 (95% CI, 0.70-0.95). The sensitivity and specificity for the boomerang sign were

0.78 (95% CI, 0.60-0.90) and 0.87 (95% CI, 0.71-0.95). However, the sensitivity and specificity for the traditional approach of detecting an anechoic collection above the diaphragm to indicate pleural effusion were only 0.54 (95% CI, 0.37-0.71) and 0.86 (95% CI, 0.80-0.99).

Conclusions: Despite inexperience in sonography, the novel VIP and boomerang signs show high diagnostic values in detecting right pleural effusion compared to the Traditional methods.

(C) 2017

Introduction

Pleural effusion, which accounts for 17% of patients with dyspnea in emergency departments [1] and 62% of dyspneic patients in ICU [2], entails a list of differential diagnoses such as pneumonia, heart failure, malignancy, etc. Traditionally, chest radiography is the first- line imaging modality for detecting pleural effusion; nevertheless, it is insensitive as it requires the accumulation of more than 200 ml of pleural fluid for the blunting of costophrenic angle to appear [3] and only demonstrates a sensitivity of 0.45 in a meta-analysis [4]. On the other hand, though the gold standard, it is not feasible to order a computed tomography (CT) scan for merely monitoring every patient due to radiation, cost and Patient transport issues es- pecially for ICU patients.

In recent years, Point-of-care ultrasound has become the first-line Diagnostic modality for diagnosing pleural effusion in emer- gency settings. While it is inexpensive, noninvasive and easily repeat- able, ultrasonography can detect 100 ml of pleural fluid with up to

* Corresponding author.

E-mail address: [email protected] (J.S.K. Lau).

100% sensitivity [5]. Most studies rely on a lateral approach by plac- ing the transducer on the posterior axillary line of the patient to visu- alize an anechoic fluid collection superior to the diaphragm [6-10]. However, there are clinical situations such as an immobilized arm, the presence of a wound, chest drain or LUCAS making the lateral ap- proach impossible. In view of this, we hereby advocate an anterior approach for the detection of right pleural effusion and present two novel sonographic markers of right pleural effusion: Visualization of inferoposterior thoracic wall (VIP) sign (Fig. 1c) and boomerang sign (Fig. 1f).

The aim of this study is to prospectively evaluate the diagnostic ac- curacy of the novel VIP and boomerang signs as an indicator of right pleural effusion when performed by physicians with little to no experi- ence in ultrasound.

Materials and methods

From 15th January 2017 to 30th June 2017, patients who were ad- mitted in the ICU at a tertiary care hospital or assessed by junior ICU physicians through in-patient consultations and trauma calls were eligi- ble. Patients 18 years or older who undergo CT thorax and/or abdomen

https://doi.org/10.1016/j.ajem.2017.11.023

0735-6757/(C) 2017

Fig. 1. a. Probe position (long axis) for the visualization of the inferoposterior thoracic wall (VIP) sign. b. Posterior thoracic wall (sagittal plane) does not extend cranial to the diaphragm, indicating the absence of pleural effusion. Note the anechoic space cranial to the diaphragm despite the lack of pleural effusion. c. VIP sign. Visualization of the inferoposterior thoracic wall cranial to the diaphragm is seen (white arrow), indicating the presence of pleural effusion. d. Probe position (short axis) for the boomerang sign. e. Posterior thoracic wall (axial plane) does not extend from the thoracic vertebra (white arrow), indicating the absence of pleural effusion. f. Boomerang sign. Extension of the posterior thoracic wall from the thoracic vertebra is seen (white arrow), indicating the presence of pleural effusion. g. Probe position for the traditional approach of detecting pleural effusion on the right side of the body. h. Note the anechoic space cranial to the diaphragm despite the lack of pleural effusion. i. An anechoic collection cranial to the diaphragm is seen, indicating the presence of pleural effusion. Sonographic images were obtained on a GE LOGIQ e (GE Healthcare, Wauwatosa, Wisconsin) with a C1-5 convex transducer.

regardless of indication in 24 h were eligible for inclusion. As lung bases were included in a CT abdomen scan, basal pleural effusion would be seen. On the other hand, patients were excluded if they were unable lung sonography“>to provide an informed consent, had a known pleural effusion, history of lung surgery or pleurodesis. The patients were enrolled by one of the four junior ICU physicians who would refer the case to each other without the patient’s history for sonography on a convenience basis. They were asked to conduct the sonographic examinations prior to performing any physical examinations and reviewing any imaging re- sults such as chest radiography.

The four junior ICU physicians who performed the ultrasound exam- inations included trainees possessing one to three years of clinical expe- rience rotating from internal medicine and emergency medicine with little to no experience in point-of-care ultrasound. Two of the physicians were never exposed to ultrasound. One attended a 2-day introductory course with one year experience of bedside application. The last one had one year of bedside experience with basic echocardiography. With CT results as the gold standard, this study examined the ability to sonographically detect the extension of the posterior thoracic wall from the vertebra and visualize the extension of the inferoposterior tho- racic wall. In addition, the traditional approach of detecting pleural effu- sion based on an anechoic collection superior to the diaphragm was examined. All CT scans were interpreted by radiologists. The study sonologists were briefed in a 30-minute lecture and bedside application on the sonographic diagnosis of pleural effusion prior to data collection. Furthermore, the diagnostic accuracy of erect chest radiography of the enrolled cases was analysed. The principal investigator interpreted

whether any pleural effusion was present by blunting of costophrenic angle. The study was approved by the Research Ethics Committee/Insti- tutional Review Board of the Hospital.

Lung sonography

ultrasound examinations of the lung bases were performed with a GE Venue 50 ultrasound system (GE Healthcare, Wauwatosa, Wiscon- sin) using a 2.5-6.0 MHz curvilinear transducer. The patients were posi- tioned supine with the head of stretcher slightly inclined. By placing the transducer on the right posterior axillary line of the patient with the probe marker pointing to the head (Fig. 1g), the coronal plane of the lung base (Fig. 1h, i) can be obtained. The sonologists first identified the diaphragm as the landmark, then they were asked to describe the echogenicity of the collection as either anechoic or heteroechoic.

Next, the probe was placed subcostally at the mid-clavicular line on the right side of patient with probe marker pointing to the head (Fig. 1a), and a sagittal plane (long-axis) of the lung base (Fig. 1b) was ob- tained. The diaphragm and the posterior thoracic wall inferior to the di- aphragm were identified, then visualization of the inferoposterior thoracic wall superior to the diaphragm was assessed (Fig. 1c). The identification of the inferoposterior thoracic wall can be confirmed by its movement sonographically when it is “balloted” underneath (Video 1). Similarly, to obtain the axial plane (short-axis) (Fig. 1e), the probe was placed subcostally at the mid-clavicular line on the right side of the patient with probe marker pointing to the right (Fig. 1d). In this view, with the acoustic shadow provided by the pleural collection

Table 1 Frequencies of true positive, true negative, false positive, and false negative of the methods in detecting right pleural effusion.

The frequencies of true positive, true negative, false positive and false negative of each method are listed on Table 1. The sensitivity, spec- ificity, prevalence, PPV, NPV, LR+ and LR- of the VIP sign, boomerang

VIP

sign

Boomerang sign

Traditional approach

Chest radiography

sign, traditional anechoic collection, and blunting of costophrenic angle on chest radiography as indicators for right pleural effusion against CT

True positive 28 25 19 15

True negative 32 33 35 31

False positive 5 5 2 5

False negative 5 7 16 21

posterior to the liver and diaphragm, the posterior thoracic wall would be seen extending from the thoracic vertebra like a boomerang (Fig. 1f). This short-axis view can be repeated in the lowest two intercostal spaces to ascertain the diagnosis if necessary. For all three planes, the sonologists noted any specific impediments in obtaining the images.

Statistics

Based on CT findings interpreted by radiologists, sonographic find- ings of the visualization of the inferoposterior thoracic wall (VIP) and boomerang signs were categorized as correct or incorrect. The sono- graphic findings were positive for pleural effusion based on only the presence of the signs. For the traditional approach, pleural effusion was present if it was anechoic. The percentage of trace amount of pleu- ral effusion out of those who were reported of having pleural effusion was calculated. In addition, the blunting of costophrenic angle as an in- dicator of basal pleural effusion on chest radiography was categorized against CT findings. Cases were excluded from the analysis if images were not obtainable. All data were entered into a statistical spreadsheet (Excel 2007, Microsoft).

A power analysis estimated that a minimum of 62 patients was needed to achieve 80% power with an alpha of 0.05 and sensitivity and specificity of 70% [11]. The online clinical calculator on http:// faculty.vassar.edu/lowry/clin1.html was used for statistical analysis and calculating the diagnostic values of the sonographic signs and radio- graphic findings against CT results with 95% confidence intervals (CIs). The sensitivity, specificity, positive predictive value (PPV), negative pre- dict value (NPV), positive likelihood ratio (LR+), and negative likeli- hood ratio (LR-) of the signs were determined.

Results

A total of 73 patients (33 males and 40 females) with a mean and median age of 62.5 and 66 (ranged from 18 to 91) were enrolled. On the other hand, the 36 patients with pleural effusion (14 males and 22 females) were slightly older with a mean and median age of 69.9 and 71 respectively. 25 out of 36 (69%) with pleural effusion was reported as having trace/mild/small amount of right pleural effusion by radiolo- gists. There were multiple cases excluded from analysis due to technical difficulties. The three cases excluded in the right subcostal long-axis view were due to generous body habitus, whereas excessive breast tis- sue was the hindrance for the three cases excluded in the right subcostal short-axis view. For the traditional approach, one case was excluded due to the presence of dressing material. On chest radiography, there was one right costophrenic angle excluded due to inadequate view.

findings are listed on Table 2. The sensitivity and specificity for the VIP sign were 0.85 (95% confidence interval [CI], 0.67-0.94) and 0.86 (95% CI, 0.70-0.95), respectively. For the boomerang sign, the sensitivity and specificity were 0.78 (95% CI, 0.60-0.90) and 0.87 (95% CI, 0.71- 0.95) respectively. The sensitivity and specificity for the traditional ap- proach of identifying an anechoic collection above the diaphragm were 0.54 (95% CI, 0.37-0.71) and 0.86 (95% CI, 0.80-0.99).

Discussion

This prospective study demonstrates the high diagnostic values of the novel sonographic VIP and boomerang signs in detecting right pleu- ral effusion. In the long-axis view, visualization of inferoposterior tho- racic wall cranial to the diaphragm should not be seen because of the high acoustic impedance of aerated lungs (Fig. 1b), whereas visualiza- tion of inferoposterior thoracic wall superior to the diaphragm can be demonstrated if a pleural collection is present to provide an acoustic window (Fig. 1c). Similarly, appearing like a boomerang in the short- axis view, the posterior thoracic wall is seen extending from the thoracic vertebra in the presence of a pleural collection (Fig. 1f). Due to stomach air on the left, these two signs are applicable on the right side only. As a result, we only conducted the study on the right hemithorax. However, as in a patient of this study, this additional probe location allows for as- sessment of pleural effusion when the traditional E-FAST view of placing the probe on the axillary line is not available due to wound dressings and chest drain placement. Furthermore, in a patient with suboptimal images due to challenging body habitus, visualizing the basal area with all three anatomical planes can improve accuracy.

The high diagnostic values of the VIP and boomerang signs in this study demonstrate again the superiority over chest radiography. The di- agnostic values of chest radiography in this study are comparable to other studies [4,9]. Though operator dependent, the four novices with little to no training in ultrasound were able to visualize the signs accu- rately implying they are simple to learn and apply. Furthermore, the signs are highly accurate despite the high proportions of trace amount of Pleural effusions on CT (69%) and non-anechoic description of pleural collections (Table 1, Fig. 2). Being able to identify accurately even small amount of pleural effusion can help in determining diagnoses, clinical progress and treatment decision in settings such as the emergency de- partment. For example, pleural effusion which is a late stage manifesta- tion of congestive heart failure can be quickly and accurately demonstrated sonographically. This is clinically significant as a shorter door-to-furosemide injection time in patients with acute heart failure is associated with lower in-hospital mortality [12].

When compared to the VIP sign, the lower sensitivity for the boo- merang sign can be attributed to the difficulty of the short axis view. One may confuse the diaphragm as the sign if the sonologist does not follow it from the vertebral body, hence one may rely on the existence of a collection and reversely look for the sign. However, the surprising finding of this study is the low sensitivity of the traditional approach of identifying pleural effusion by an anechoic collection cranial to the di- aphragm. Unfortunately, artefacts such as hepatization of the lung,

Table 2

Diagnostic values of the sonographic signs and chest radiography for the right hemithorax, values with 95% CIs in parenthesis.

	Sensitivity	Specificity	PPV	NPV	+LR	-LR	Prevalence
VIP sign	0.85 (0.67-0.94)	0.86 (0.70-0.95)	0.85 (0.67-0.94)	0.86 (0.70-0.95)	6.28 (2.74-14.37)	0.18 (0.08-0.40)	0.47 (0.35-0.59)
Boomerang sign	0.78 (0.60-0.90)	0.87 (0.71-0.95)	0.83 (0.65-0.94)	0.83 (0.67-0.92)	5.94 (2.57-13.71)	0.25 (0.13-0.49)	0.46 (0.34-0.58)
Traditional approach	0.54 (0.37-0.71)	0.95 (0.80-0.99)	0.90 (0.68-0.98)	0.69 (0.54-0.80)	10.04 (2.52-40.00)	0.48 (0.34-0.70)	0.49 (0.37-0.61)
Chest radiography	0.42 (0.26-0.59)	0.86 (0.69-0.95)	0.75 (0.51-0.91)	0.60 (0.45-0.73)	3.00 (1.22-7.38)	0.68 (0.51-0.90)	0.50 (0.38-0.62)

Fig. 2. An example of a patient with trace pleural effusion. a. VIP sign (white arrow). A corresponding video clip demonstrates its movement by ballottement. b. Boomerang sign. c. Note the lack of an anechoic fluid collection despite the presence of pleural effusion in the traditional approach. d. Corresponding posterior-anterior erect chest radiography demonstrates the sharp right costophrenic angle. e. Corresponding CT shows bilateral trace pleural effusion. Sonographic images were obtained on a GE LOGIQ e (GE Healthcare, Wauwatosa, Wisconsin) with a C1-5 convex transducer.

mirror image artefact, and rib shadows can all easily confuse the inexpe- rienced sonologists. Moreover, the high accuracy in literature is easily confounded by the operators’ experience. For example, in another sono- graphic study of pleural effusion, the sonologists were emergency phy- sicians with extensive ultrasonography experience or emergency ultrasound fellowship [13]. In our setting, based on ultrasound findings, we often perform Pigtail catheter insertion to drain pleural effusion in indicated patients. It would be catastrophic to wait for the finding of a chest radiograph in a deteriorating patient, or making an incorrect diag- nosis and inserted the catheter into an otherwise healthy pleura. This further strengthens the importance of the VIP and boomerang signs to those who are not as experienced, who are often the first-call physicians.

Limitations

There are several limitations in this study. The sample size is not large which results in wide confidence intervals. Secondly, the general- izability may be questioned as this study was conducted in an ICU of a single centre. In addition, pulmonary hepatization due to pneumonic consolidation may allow visualization of the thoracic wall and lead to false-positive interpretation. Lastly, the successful detection of pleural effusion relies on its collection at the basal dependent area. We failed

to detect a hemothorax which was collected at the right middle zone consequently. Therefore, increasing the incline angle of the upper body can increase the sensitivity of the signs.

Conclusion

In conclusion, the high diagnostic accuracy of the novel VIP and boo- merang signs add a new and powerful adjunct in detecting even trace amount of pleural effusion where an anechoic collection may not be vi- sualized clearly. Besides providing an additional site of probe place- ment, they allow delineating the right basal pleural space in three- dimension to enhance diagnostic accuracy and safety of interventions. This study also illustrates the ease of acquiring and applying the tech- niques of various sonographic signs presented, even for the inexperi- enced. Not only do detecting these signs at the bedside can aid in diagnosis and alter the timing of management to improve clinical out- come, but it can also monitor clinical progress. However, this does not mean to replace a chest radiograph completely, rather it affects the timing and frequency of performing one. Future studies can look into whether the length of extension of these signs can be correlated to the size of pleural collection and be used as a treatment monitoring tool.

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2017.11.023.

Acknowledgements

We thank Drs. Catherine Tam, Judianna Yu, and Jonathan Lai for spending their invaluable time for participating as the sonologists, Dr. Boris Yow for helping out in displaying the probe positions of the vari- ous sonographic signs, and Dr. Rita Chan for preparing the consent form in the official languages. Last but not least, we thank Dr. Hoi-ping Shum for the guidance of this research.

Conflict of interest

The authors declare that they have no conflict of interest.

References

1. Knudsen CW, Omland T, Clopton P, Westheim A, Abraham WT, Storrow AB, et al. Di- agnostic value of B-type natriuretic peptide and chest radiographic findings in pa- tients with acute dyspnea. Am J Med 2004;116(6):363-8.
2. Mattison LE, Coppage L, Alderman DF, Herlong JO, Sahn SA. Pleural effusions in the medical ICU: prevalence, causes and clinical implications. Chest 1997;111(4): 1018-23.
3. Soni NJ, Franco R, Velez MI, Schnobrich D, Dancel R, Restrepo MI, et al. Ultrasound in the diagnosis & management of pleural effusions. J Hosp Med 2015;10(12):811-6.
4. Yousefifard M, Baikpour M, Ghelichkhani P, Asady H, Nia KS, Jafari AM, et al. Screen- ing performance characteristic of ultrasonography and radiography in detection of pleural effusion; a meta-analysis. Emerg (Tehran) 2016;4(1):1-10.
5. Kalokairinou-Motogna M, Maratou K, Paianid I, Soldatos T, Antipa E, Tsikkini A, et al. Application of Color Doppler ultrasound in the study of small pleural effusion. Med Ultrason 2010;12(1):12-6.
6. Xirouchaki N, Kondili E, Prinianakis G, Malliotakis P, Georgopoulos D. Impact of lung ultrasound on clinical decision making in critically ill patients. Intensive Care Med 2014;40:57-65.
7. Kataoka H, Takada S. The role of thoracic ultrasonography for evaluation of patients with decompensated chronic heart failure. J Am Coll Cardiol 2000;35(6):1638-46.
8. Ma OJ, Mateer JR. Trauma ultrasound examination versus chest radiography in the detection of hemothorax. Ann Emerg Med 1997;29(3):312-5.
9. Rocco M, Carbone I, Morelli A, Bertoletti L, Rossi S, Vitale M, et al. Diagnostic accuracy of bedside ultrasonography in the ICU: feasibility of detecting pulmonary effusion and lung contusion in patients on respiratory support after severe blunt thoracic trauma. Acta Anaesthesiol Scand 2008;52(6):776-84.
10. Lichtenstein D, Goldstein I, Mourgeon E, Cluzel P, Grenier P, Rouby JJ. Comparative diagnostic performances of auscultation, chest radiography, and lung ultrasonogra- phy in acute respiratory distress syndrome. Anesthesiology 2004;100(1):9-15.
11. Bujang MA, Adnan TH. Requirements for minimum sample size for sensitivity and specificity analysis. J Clin Diagn Res 2016;10(10):YE01-6.
12. Matsue Y, Damman K, Voors AA, Kagiyama N, Yamaguchi T, Kuroda S, et al. Time-to-

    furosemide treatment and mortality in patients hospitalized with acute heart failure. J Am Coll Cardiol 2017;69:3042-51.

    Dickman E, Terentiev V, Likourezos A, Derman A, Haines L. Extension of the thoracic spine sign: a new sonographic marker of pleural effusion. J Ultrasound Med 2015; 34(9):1555-61.