Accuracy of emergency department ultrasound scanning in detecting abdominal aortic aneurysm

Accuracy of emergency department ultrasound scanning in detecting abdominal aortic aneurysm

Report by S Bentz, Resident

Checked by J Jones, Faculty—EM Residency program

Grand Rapids Medical Education and Research, Michigan, USA

Abstract

A short‐cut review was carried out to establish whether emergency department ultrasound scanning had clinical utility for the diagnosis of abdominal aortic aneurysm (AAA). A total of 73 papers were found using the reported searches, of which 4 presented the best evidence to answer the clinical question. The author, date and country of publication, patient group studied, study type, relevant outcomes, results and study weaknesses of these best papers are tabulated. It is concluded that in patients suspected of having AAA, emergency department bedside ultrasound scanning for AAA is sensitive and specific and may improve patient care.

Clinical scenario

A 55‐year‐old man presents to the emergency department with a sudden onset of abdominal and flank pain associated with hypotension. Our concern is that he may have an AAA, but at this time the patient is too unstable to leave the emergency department for formal imaging. The vascular surgeons suggest a bedside ultrasound study. We perform one scan and see no AAA. Can you be confident that an ultrasound scan performed by an emergency physician can accurately rule in or out an AAA?

Three‐part question

In [patients suspected of having abdominal aortic aneurysms] is [ED bedside ultrasound performed by EM physicians as good as that performed by radiologists] at [accurately detecting AAA?]

Search strategy

Medline 1966–week 3 July 2006 using the Ovid interface, and Embase 1980–weel 29 2006 using Multifile searching, and Cochrane Library Issue 3 2006.

Medline/Embase: [(exp aortic aneurysm, abdominal/ultrasonography) and (exp emergency.mp. or emergencies/)]. LIMIT to human and English.

Cochrane Library: Aortic Aneurysm, Abdominal/US [MeSH] and Emergencies [MeSH] 0 records.

Outcome

A total of 73 papers were found, of which 69 were irrelevant or of insufficient quality. Table 1 summarises the contents of remaining four papers.

Table 1.

Author, date, country	Patient group	Study type	Outcomes	Key results	Study weaknesses
Tayal et al, 2003, USA	All patients suspected of having AAA	Diagnostic cohort	Positive or negative for AAA defined as abdominal aorta diameter >3 cm	29/125 scans positive for AAA. 27/29 positive on confirmatory test. Sensitivity = 100%, Specificity = 98%, NPV = 100%, PPV = 93%	Small number of patients. Lack of clear inclusion and exclusion criteria may have led to selection bias
	Emergency bedside ultrasound performed by EM physicians followed by a confirmatory test, ie, radiology ultrasound, abdominal CT, abdominal MRI and laparotomy.
Knaut et al, February 2005, USA	Convenience sample of patients ⩾50 years with abdominal pain and scheduled for abdominal and pelvic CT scan with IV contrast as part of their evaluation.	Diagnostic cohort	AAA diagnosed by ultrasound and confirmed by CT	5/104 diagnosed by US were confirmed by CT. No false negatives. PPV of 0.00 at SMA, 0.67 in longitudinal view and 0.80 at bifurcation. NPV 0.099, 0.99 and 1.00, respectively	Convenience sample could lead to selection bias
	Ultrasound to measure the diameter of the abdominal aorta by either a resident or attending emergency medicine physician.		Difference in measured diameter at SMA, bifurcation and longitudinal view	<1.41 cm at SMA, <1.05 at bifurcation and <0.94 in longitudinal view 95% of the time	Did not include patients who went to OR for repair were not scanned
Constantino et al, November 2005, USA	Patients ⩾55 years with at least one of the following: abdominal, back, flank or chest pain or hypotension as well as clinical suggestion of AAA	Diagnostic cohort	AAA measuring >3.0 cm on ultrasound	36/238 had abnormal aortas, 34 AAA, 1 aortic dissection, 1 intraluminal clot. Sensitivity 100% for aortic abnormality, 94% for AAA. Specificity 100% for both end points	Possible selection bias. Residents who performed EUS may have had more extensive training than most
	EUS performed by 3rd year EM residents
Kuhn et al, September 2000, Australia	Patients in whom AAAs were expected, including those patients over ⩾50 years presenting with abdominal/back pain of unclear origin or presumed renal colic	Diagnostic cohort	Positive for AAA defined as aortic diameter >3 cm	68 patients scanned, aorta visualised in 66. 26 true positives and 40 true negatives Sensitivity = 100%, Specificity = 100%, PPV = 100%, NPV = 100%	Small study, only 68 participants
	ED bedside ultrasound preformed by EM physician who had attended a 3‐day ultrasound training course.		Improvement of patient care	Improved care in 46 patients, 0 adversely affected	Convenience sampling may have led to bias in selecting patients who are easier to scan or who have high clinical suspicion of AAA

AAA, abdominal aortic aneurysm; CT, computed tomography; ED, emergency department; EM, emergency medicine; EUS, endoscopic ultrasound scanning; IV, intravenous; NPV, negative predictive value; OR, operating room; PPV, positive predictive value; SMA, superior mesenteric artery; US, ultrasound scan.

Comments

Although all of these studies probably had some degree of selection bias, these are the patients who are most likely to benefit from early detection of AAA. It would have been helpful if they had included information on patients who may have been excluded because of body habitus or potential difficulty in scanning. The amount of training in ultrasound scanning does not seem to affect the accuracy.

Clinical bottom line

In patients suspected of having AAA, emergency department bedside ultrasound scanning for AAA is sensitive and specific and may improve patient care.