a b s t r a c t

Objective: Full visualization of the abdominal aorta using the standard midline view is often inadequate for the detection of Abdominal aortic aneurysm. We evaluated whether the addition of a lateral midaxillary right upper quadrant view could improve visualization of the abdominal aorta.

Methods: This was a prospective observational proof-of-concept study of patients older than 50 years undergoing abdominal computed tomographic scan for any indication in the emergency department. Ultrasounds were performed by American College of Emergency Physicians-credentialed study sonographers, and images were reviewed by an ultrasound fellowship-trained reviewer. The standard midline aortic images were obtained as well as additional lateral images from a midaxillary right upper quadrant approach. Visualization of the aorta was determined to be adequate if more than two-thirds of the abdominal aorta was visualized.

Results: We enrolled 60 patients. Six patients were excluded due to missing data. A total of 54 patients were analyzed. The median age was 67.5, and 37% were male. The mean body mass index was 25.9 (SD, 5.8), and mean abdominal circumference was 105.3 cm (SD, 18.1). Visualization of the aorta using the midline approach was adequate in 26 (48.2%) of 54 of the patients and 32 (59.2%) of 54 of the lateral approach. Addition of a lateral view in examinations with an inadequate midline view increased adequate aortic visualization to 41 (75.9%) of 54.

Conclusion: Combining a lateral view to the Standard midline approach improves adequate visualization of the abdominal aorta by approximately 28%. Further study is required to determine if the lateral view is equivalent for detecting abdominal aortic aneurysm.

(C) 2014

Introduction

clinician-performed ultrasound (CPU) of the abdominal aorta to detect Abdominal aortic aneurysm is one of the standard emergency ultrasound applications. Diagnostic accuracy of history and physical examination are notoriously poor, and AAA carries significant mortality [1-3]. For these reasons, emergency physicians (EPs) have increasingly adopted CPU of the aorta for the rapid bedside diagnosis of AAA, and the American College of Emergency Physicians (ACEP) has developed training guidelines and credentialing standards [4].

Previous studies reported high diagnostic accuracy of EPs for the detection of AAA [5-9]. However, many of these initial studies may not be generalizable as they excluded patients in which the aorta was

? Results of the study were presented at 2012 ACEP Scientific Assembly in Denver, CO.

* Corresponding author. Tel.: +1 617 732 5636; fax: +1 617 264 6848.

E-mail address: [email protected] (H.H. Kimberly).

not fully visualized, had only very experienced sonographers performing the examinations, or were limited by methodology such as convenience sampling, which could have resulting in excluding challenging patients.

More recent literature suggests that a significant portion of CPU aorta examinations are limited by inadequate visualization of the entire aorta. A study by Blaivas and Theodoro [10] retrospectively analyzed 207 patients who had received emergency ultrasound to rule out AAA and found that the aorta was not seen in its entirety in 17% of patients, and in 8% of patients, a significant portion of the aorta (N 1/3) was not visualized. Moore et al [11] performed Screening examinations of the aorta on 179 asymptomatic male emergency department (ED) patients reporting complete visualization in only 62.6%, and in 4.5% of patients, the aorta was not visualized at all. More recently, Hoffmann et al [12] enrolled 196 asymptomatic men with AAA risk factors and found that a diverse group of ACEP-credentialed sonographers adequately visualized the aorta in 71.4%, incompletely visualized in 20.4%, and incorrectly measured in 8.2%. They found that

0735-6757/$ - see front matter (C) 2014 http://dx.doi.org/10.1016/j.ajem.2013.10.031

M. Studer et al. / American Journal of Emergency Medicine 32 (2014) 256-259 257

providers with more than 3 years of experience post credentialing demonstrated better performance.

These recent studies demonstrate that in day-to-day practice, CPU of the aorta as practiced using a standard midline approach is significantly limited by inadequate visualization of the vessel. The most common factors resulting in poor ultrasound visualization include bowel gas obscuring the aorta, body habitus, symptoms such as abdominal pain causing inability to tolerate the examination, Experience level of the sonographer, and limited time at the bedside. Unfortunately, the standard countermeasures for poor visualization including graded compression and patient positioning are often not well tolerated by patients. Using the liver as an acoustic window from a midaxillary lateral right upper quadrant (RUQ) approach (such as the RUQ FAST view) visualizes the aorta longitudinally and may provide additional information when a midline view is inadequate (see Fig. 1). To our knowledge, the usefulness of this lateral approach has never been specifically studied and is not typically taught or used by point-of-care sonographers.

The purpose of this study was to evaluate whether it was possible to improve adequate visualization of the abdominal aorta by supplementing the traditional midline view of abdominal aorta with a lateral RUQ midaxillary view.

Methods

This was a prospective observational proof-of-concept study. The study was approved by the institutional review board and was conducted in an academic emergency medicine department with an ED residency program and an annual census of more than 70000 patients. Patients older than 50 years undergoing an Abdominal computed tomography (CT) for any indication while in the ED were eligible for inclusion. Exclusion criteria included pregnancy, inability to provide verbal consent, or unstable clinical condition as determined by the treating physician. Consecutive patients awaiting abdominal CT were approached by study staff based on their availability. Basic demographic information was obtained at the time of consent including age, sex, height, weight, and abdominal circumference.

Ultrasounds were performed by 5 sonographers including an emergency medicine resident, a physician assistant, an emergency medicine fellowship-trained physician, and 2 foreign trained physi- cians obtaining ultrasound Fellowship training. All of them had performed greater than 25 aorta scans and had met ACEP credential- ing requirements. ultrasound evaluations were performed at the bedside using a 2-5 MHz curvilinear transducer on a Siemens Acuson X300 (Siemens AG, Erlangen, Germany) and Zonare Z.One Ultra- system (Zonare Medical Systems, Mountain View, CA).

All patients were scanned in the supine position. Images of the aorta from ciliac axis to the bifurcation were obtained from the following 2 approaches: (1) traditional midline transverse approach (saving a video clip of scanning down the aorta and still images of the proximal, mid, and distal aorta) and (2) lateral midaxillary longitu- dinal view from the RUQ tracking the aorta distally to the bifurcation. Study sonographers graded the adequacy of aortic visualization in the midline view as entire aorta visualized, limited but adequate visualization (N 2/3 of aorta meaning measurements of proximal, mid, and distal but not full visualization between), inadequate visualization (b 2/3 meaning only 1 or 2 of the 3 measurements could be obtained), or no view. The lateral RUQ view was graded as entire aorta visualized, limited visualization (N 2/3 of the aorta if a longitudinal segment of N 10 cm was measured), inadequate visual- ization (b 2/3 of the aorta visualized), or no view. Time to acquisition of images was recorded. Images were saved as Digital Imaging and Communications in Medicine format video clips to the ultrasound machine‘s hard drive and later burned to DVD. Images were reviewed by a blinded ultrasound-trained emergency medicine physician. Abdominal CTs were reviewed and measured by research assistants to determine the length of the abdominal aorta and for the presence of AAA, defined as diameter greater than 3 cm.

Data analysis

Data were entered into an Excel datasheet then analyzed using STATA 12.0 (StataCorp LP, College Station, TX). Baseline characteris- tics were reported in counts and proportions or mean and SD values when appropriate. Univariate comparisons were made with ?² test. Interobserver agreement between sonographer and reviewer was assessed using Cohen’s ? agreement coefficient. All tests were considered statistically significant when P <= .05.

Results

We enrolled a total of 60 patients. Of these, 6 patients were later excluded because no images were saved after an ultrasound machine hard drive failure, leaving a total of 54 patients for review and analysis. The baseline characteristics patients are reported in the Table. The median age was 67.5 years, and 37% of patients were male. The mean body mass index (BMI) was 25.9 (SD, 5.8), and mean abdominal circumference was 105.3 cm (SD, 18.1). The mean time to perform the midline scan was 4.7 minutes (SD, 2.7); the mean time to perform the RUQ scan was 3.8 minutes (SD, 1.5).

We assessed the visualization of the abdominal aorta using the traditional midline view and the supplemental RUQ view. With the

Fig. 1. Midline aorta view with aorta in transverse (left panel). Lateral RUQ view of abdominal aorta with the aorta deep to the liver and inferior vena cava (right panel).

258 M. Studer et al. / American Journal of Emergency Medicine 32 (2014) 256-259

Table

various studies is limited without a standard definition of adequate

Baseline characteristics				view. Hoffman et al [12] classified a study as “correct” if both upper and
Adequate view		Inadequate view	Total	lower segments were visualized and “incorrect” if one segment was
(n = 41)		(n = 13)	(N = 54)	not well visualized or indeterminate. Moore et al [11] labeled scans as
Age, median, y (IQR)	68 (14)	66 (11)	67.5 (15) correct when visualized from celiac axis to bifurcation and limited “if
Sex M/F ratio	0.52	0.86	0.59 only portions of the aorta could be examined.” Blaivas and Theodoro

white race (%)	70.7	53.9	66.7
BMI, mean (SD)	25 (7.1)	29 (5)	25.9 (5.8)
Abdominal circumference,	102.7 (18.4)	113.5 (14.9)	105.3 (18.1)

mean, cm (SD)

midline approach, visualization was adequate in 26 patients (48.2%) and inadequate in 28 (51.8%). In the RUQ view, the aorta could adequately be visualized in 32 patients (59.2%), leaving 22 (40.8%) with an inadequate RUQ view. Of the 28 patients who had an inadequate midline view, 15 of those (54.8%) had an adequate RUQ view. Combining the RUQ view with the midline view resulted adequate visualization in 75.9% (total of 41 of 54) patients as opposed to only 48.2% (total of 26 of 54) patients with a midline approach only (Fig. 2). ? between the sonographer and the reviewer was 0.74 (P b

.01) for midline transverse approach and 0.61 (P b .01) for RUQ view.

In evaluating the effects of body habitus on aortic visualization, we found mean BMI was 24.1 and 27.6 (P = .03) for patients with adequate and inadequate midline view, respectively, and 24.9 and

27.4 for RUQ view (P = .11). Computed tomographic images were reviewed on all 54 patients. The mean length of the aorta measured on CT was 15.6 cm (SD, 3.1). Two AAA were found on review of the CT images, and in both patients, the aneurysm was also seen on the midline and RUQ views.

Discussion

This is the first study to our knowledge evaluating the feasibility and utility of a lateral RUQ view of the abdominal aorta in emergency ultrasound. In this small proof-of-concept study, we found that RUQ view was achieved in all enrolled patients, and the addition of this lateral view improves adequate visualization of the aorta by approximately 28%. A midline plus RUQ implemented approach may especially be helpful in patients with difficult midline aorta views, for instance, patients with increased BMI.

This study highlights the challenge of adequate visualization of the aorta on bedside ultrasound with only 48% of our study patients found to have adequate midline view. Our rate of adequate visualization is comparatively lower than in other recent studies; 62.6% in Moore et al [11], 71.4% in Hoffman et al [12], and 74.8% in Blaivas and Theodoro

[10] and likely due to a variety of reasons. The major difference between our study and the prior studies is the patient population. Our aim was a proof-of-concept study to evaluate visualization of the aorta using a novel view. We enrolled patients who were undergoing abdominal CT for any reason, and thus, the patient population was not specifically targeted toward patients with suspected AAA. Many of these patients had other intra-Abdominal pathology, such as Bowel obstruction, resulting in increased bowel gas making them more challenging to scan. Potentially, these patients were undergoing CT because of difficult physical examination or inconclusive clinician- performed ultrasound, although we did not specifically collect data on non-study-related bedside ultrasound use. In addition, the study sonographers, although ACEP credentialed, had varied experience levels, which may have resulted in more challenging scans. Hoffman et al [12] demonstrated that the rate of indeterminate and incorrect aorta scans varied significantly by Provider experience level. The difference in ? values of the sonographers and reviewer between the RUQ and midline views in our study likely illustrates the effect of varied sonographer experience level with the novel RUQ view demonstrating a lower ? value. Finally, direct comparison between

[10] defined inadequate scans as missing a portion of the aorta at least one-third its length which is the definition we used in this study.

Although there is some variability in the current literature regarding the number of adequate scans visualized using the midline approach, it is clear that the midline approach can be challenging. Typical countermeasures, including moving the patient and probe pressure to move bowel gas (which is difficult in patients with abdominal pain), depend on the patient’s clinical status. No other studies to our knowledge have specifically evaluated the addition of the lateral RUQ approach. The benefits to this approach include using the liver as an acoustic window, thus potentially avoiding bowel gas and central obesity.

Based on our small proof-of-concept study, adding the lateral RUQ view should be considered when the midline view is inadequate such as patients with significant bowel gas or increased BMI. Further studies should specifically evaluate the sensitivity and specificity of the lateral view in the diagnosis of AAA. Although our study was not specifically designed to assess AAA and sample size was limited, we did encounter 2 patients with AAA, and both were identified on midline and lateral views.

The results of the study should be considered in the context of the following limitations. The study was a proof-of-concept study with a small sample size, and larger studies should be performed to verify these results. Patients were enrolled based on the convenience of the study sonographers, which could introduce a sample bias. Study sonographers did approach consecutive patients during their scan- ning hours to attempt to minimize potential selection bias. Because we enrolled patients undergoing abdominal CT for any reason and not patients suspected of AAA, we obtained a patient population that is likely different from patients presenting with suspected AAA and described in literature. For example, the study population had a greater proportion of women and an average BMI of 25.9, which may not be generalizable to other settings. Because of the small sample size, we were not able to assess results by individual sonographers who varied in training and prior scans performed. Finally, we designed the study to evaluate the ability to visualize the aorta, not to detect AAA, and thus, the results should be interpreted in this context, and further studies should specifically evaluate the sensitivity and specificity of the lateral view for detecting AAA.

Fig. 2. Number of adequate aorta views using traditional midline approach and midline

+ RUQ approach.

M. Studer et al. / American Journal of Emergency Medicine 32 (2014) 256-259 259

Conclusion

This study found that the addition of a lateral RUQ view of the abdominal aorta improved adequate visualization of the vessel. Providers performing ultrasound of the aorta could consider this additional view when the midline view is inadequate. However, further studies should evaluate this technique in patients suspected of AAA to assess the accuracy of this implemented approach in the diagnosis of AAA.

References

1. Lederle FA, Simel DL. The rational clinical examination. Does this patient have abdominal aortic aneurysm? JAMA 1999;281(1):77-82.
2. Fink HA, et al. The accuracy of physical examination to detect abdominal aortic aneurysm. Arch Intern Med 2000;160(6):833-6.
3. Sakalihasan N, Limet R, Defawe OD. Abdominal aortic aneurysm. Lancet 2005;365(9470):1577-89.
4. American College of Emergency Physicians. ACEP emergency ultrasound guide- lines-2001. Ann Emerg Med 2001;38(4):470-81.
5. Costantino TG, et al. Accuracy of emergency medicine ultrasound in the evaluation of abdominal aortic aneurysm. J Emerg Med 2005;29(4):455-60.
6. Dent B, et al. Emergency ultrasound of the abdominal aorta by UK emergency physicians: a prospective cohort study. Emerg Med J 2007;24(8):547-9.
7. Knaut AL, et al. Ultrasonographic measurement of aortic diameter by emergency physicians approximates results obtained by computed tomography. J Emerg Med 2005;28(2):119-26.
8. Kuhn M, et al. Emergency department ultrasound scanning for abdominal aortic aneurysm: accessible, accurate, and advantageous. Ann Emerg Med 2000;36(3): 219-23.
9. Tayal VS, Graf CD, Gibbs MA. Prospective study of accuracy and outcome of emergency ultrasound for abdominal aortic aneurysm over two years. Acad Emerg Med 2003;10(8):867-71.
10. Blaivas M, Theodoro D. Frequency of incomplete abdominal aorta visualization by emergency department bedside ultrasound. Acad Emerg Med 2004;11(1):103-5.
11. Moore CL, et al. Screening for abdominal aortic aneurysm in asymptomatic at-risk patients using emergency ultrasound. Am J Emerg Med 2008;26(8):883-7.
12. Hoffmann B, et al. Successful sonographic visualisation of the abdominal aorta differs significantly among a diverse group of credentialed emergency department providers. Emerg Med J 2011;28(6):472-6.