• 
• 

Three Part Question

3-part question derived from PICOS formulation as“In [patients presenting to emergency department with suspected ruptured AAA]can [ultrasonography by emergency physicians] accurately [detect an AAA]?”

Clinical Scenario

A 60 year old male, brought in by ambulance crew,with the complaint of sudden onset severe right flank pain. He is also known to have cardiac problems and is on several medications. On examination his systolic blood pressure is 100 mm of Hg and the pulse rate of 65 per minute. The immediate concern is a ruptured AAA. Can ultrasonography by emergency physicians detect accurately the presence or absence of an abdominal aortic aneurysm,AAA and affect the immediate management strategy in this patient?

Search Strategy

A Literature search was conducted on MEDLINE, EMBASE and CINAHL through the NHS Evidence Health Information Resource via Athens portal. The search was based on 3-part questionnaire derived from PICO formulation.
The search was conducted separately on each of these three databases EMBASE, MEDLINE & CINAHL using a combination of keywords and controlled vocabulary (eg MeSH and EMTREE) with Boolean operators (AND, OR). Truncation (*) was used where appropriate. For EMBASE, the search strategy was as follows; similarly the strategies were adapted for MEDLINE and CINAHL.
Date last searched 08.02.2013
1. exp ABDOMINAL AORTA ANEURYSM/ di [di=Diagnosis]
2. EMERGENCY.ti,ab
3. ultraso*.ti,ab
4. 1 AND 2 AND 3 (Limits: Humans and Age 18+)

OTHER DATABASES & BIBILIOGRAPHY:

Further searches were carried out using the keywords “abdominal aortic aneurysm”, ultrasound and emergency on Google scholar, Cochrane Library, Database of Abstracts of Reviews of Effects (DARE), Health Technology Assessment (HTA) and the hand search.Bibliographies were also searched for any further relevant papers. The papers were finally selected based on inclusion and exclusion criteria, which are as follows,

Inclusion criteria for the papers for this CTR were set as: 1. Papers with relevant abstract and content answering 3-part question 2. Ultrasound by emergency physicians in suspected AAA’s presenting to ED 3. Accuracy of EMUS ± mortality or the change in the management based on ED ultrasound findings. Exclusion criteria: 1.Irrelevance by Title or Abstract 2. Irrelevance by content 3. Unavailable.

Search Outcome

The total number of papers found through electronic databases and hand searching were 182, out of which n=59 were duplicates. From remaining 123 papers n=75 were excluded for the irrelevance by title or abstract. 48 (123-75=48) papers were read, bibliographies were searched for any further relevant papers and finally 9 papers were selected using inclusion and exclusion criteria.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Rubano E et al February 2013 USA	18+ adults attending emergency department with suspected abdominal aortic aneurysm, AAA	Systematic review. The studies selected after the search through the electronic databases based on the criteria derived from PICO formulation of the clinical question: Is ED-performed US sufficiently accurate to rule out an AAA in a suspected patient? Patients—patients (18+ years) suspected of having an AAA. Intervention—bedside ED US performed by EPs to detectAAA. Comparator—reference standard for diagnosingAAA was a CT, magnetic resonance imaging (MRI),aortography, ED US reviewed by radiology, or official US performed by radiology, exploratory laparotomy,or autopsy results.Total number of studies appraised were 7 and the total number of patients selected were 655.The operating characteristics were calculated based on the results.	Accuracy of emergency ultrasound for AAA	Sensitivity of emergency ultrasound for AAA= 99% (95% CI 96%-100%)	1. The studies appraised,are mainly from USA and all of them are in english language.This may affect the external validity or the applicability of the study recommendations and the results. 2. This systematic Review has missed few directly relevant studies because of narrow search criteria.
				Specificity of emergency ultrasound for AAA= 99% (95% CI 97%-99%)
				Likelihood Ratio for Positive= 10.8 to ∞
				Likelihood Ratio for Negative= 0.00 to 0.025
Dent B et al 2007 UK	Adults presenting to ED with back or abdominal pain or any other suspicion for AAA during a 12 month period from January to December 2005 were included	Prospective cohort study	Presence of AAA, defined as abdominal aortic diameter > 3cm	Sensitivity for AAA= 96.3% (95% CI 81.0% to 99.9%)	1.Work-up/varification bias which might have affected the accuracy of the index test( ultrasound by emergency physicians) 2.Review bias as diagnostic and index tests results were known to interpreters, reporting the higher sensitivity of the index test 3.Selection bias as no indeterminate scans are mentioned 4. A small sample size
				Specificity for AAA= 100% (95% CI 91.8% to 100%)
				Negative predictive value= 98.6% (95% CI 88.0% to 99.9%)
				Positive predictive value =100% (95% CI 86.8% to 100%)
			Number of patients with rAAAs taken to theatre	4/26
Knaut AL et al 2005 USA	Adults ≥50 years, presenting to ED with abdominal pain, scheduled for contrast CT, randomly enrolled for EMUS	Prospective diagnostic cohort,Double blinded study.	Difference in aortic diameters (EMUS and CT) at SMA, bifurcation and at midway between SMA and bifurcation (longitudinal)	Positive predictive value for AAA measurement by ED ultrasound as compared to CT measurement at SMA= 0.00 (0.00–0.46)	No record of indeterminate scans convenience sampling small number of AAA positive patients
				Negative predictive values for AAA measurement by ED ultrasound as compared to CT measurement at SMA= 0.99 (0.94–0.99)
				Positive predictive value for AAA measurement by ED ultrasound as compared to CT measurement, Longitudinal= 0.67 (0.22–0.96)
				Negative predictive values for AAA measurement by ED ultrasound as compared to CT measurement. Longitudinal= 0.99 (0.94–0.99)
				Positive predictive value for AAA measurement by ED ultrasound as compared to CT measurement at bifurcation= 0.80 (0.28–0.99)
				Negative predictive values for AAA measurement by ED ultrasound as compared to CT measurement at bifurcation= 1.00 (0.96–1.00)
			Presence of AAA, defined as infrarenal abdominal aortic diameter >3cm OR suprarenal to infrarenal ratio of aortic diameter > 1.5	Number of AAAs diagnosed on EMUS and offered repair/total scans performed= 9/104
				Total number of AAAs diagnosed and went to theatre without frther imaging= 4/9
			Average time needed to scan AAAs by enmergency physicians	5 minutes
Costantino TG et al 2005 USA	Adults >55years with abdominal, back, flank, chest pain or hypotension as well as suspicion of AAA. Known stable AAA’s were excluded.	Prospective cohort,Diagnostic study	Presence of AAA defined as transverse aortic diameter >3cm on emergency ultrasound	Sensitivity = 94%	1.No indeterminate scans logged, possibility of convenience sample and selection bias 2.Non blided study, review bias
				Specificity= 100%
			surgical repair based on ultrasound result	n=5
			Comparison of measurements of all AA diameter >3cm by EUS and RAD	Mean difference between EUS and RAD was 0.44 cm.
Blaivas M et al 2004 USA	Patients >50 years with abdominal ,flank or back pain presented to ED, patients who refused and who had AA scanned previously were excluded from the study	Retrospective observational study.	presence of AAAs defined as external diameter >3cm.	Number of positives for the presence of AAAs amongst the total number of scans performed= 29/207	1.Retrospective data 2.No report of false positives, false negatives or indeterminate scans, hence possibility of convenience sampling, selection bias directly affecting the accuracy values.
				False positives for AAAs= 0/207
				False Negatives for AAAs= 0/207
				Number of positives for AAAs who were unstable and underwent surgical repair without further imaging= 9/29
			Number of inadequate scans	8%
Tayal VS et al 2003 USA	All adults presenting to emergency department suspected to have AAA. Inclusion criteria were abdominal, back, flank pain or syncope. Patients known to have AAA’s were excluded.	Prospective diagnostic cohort study	Presence or absence of AAA, AAA defined as AA external diameter more than 3 cm	Sensitivity = 100%	1.Presence of Differential work up bias/verification bias as the gold standard instituted was dependent on the result of index test 2.No indeterminate scans were mentioned hence possibility of convenience sampling and selection bias directly affecting the statistical accuracy values 3.Non blinded study
				Specificity = 98%
				Negative predicitive value= 100%
				Positive predicitive value= 93%
			Improvement of patient care	May have improved patient care in 63% = 17/27 of AAA patients
Kuhn M et al 2000 AUSTRALIA	Adults > 50 years presenting to emergency department, abdominal pain with or without back pain, presumed renal colic and all patients with presumed AAA. Patients with a known AAA OR unable to obtain consent for study entry were the exclusion criteria	Prospective cohort, double-blinded study	Presence of AAA as defined by abdominal aortic diameter > 3 cm.	Sensitive of emergency ultrasound for AAA= 100% (95% CI 87%-100%)	1.Incorporation bias as index test result has been used as gold standard in some cases 2.Differential verification bias as different gold standards used for the subjects in the sample, affecting directly the accuracy results of emergency ultrasound for AAAs. 3.No inclusion of indeterminate scans in final statistical calculations 4.No specifics provided as to how emergency ultrasound could have improved the patient care. 5. Small sample size and possibility of selection bias
				Specificity of emergency ultrasound for AAA= 100% (95% CI 91%-100%)
				Positive predictive value of emergency ultrasound for AAAs= 100% (95% CI 87%-100%)
				Negative predictive value of emergency ultrasound for AAAs= 100% (95% CI 87%-100%)
			Clinical impact of emergency ultrasound on patient management	ED scans would have improved the care of 46 patients and would not have caused any adverse effect.No further specifics provided in the original article
Walker A et al 2004 UK	patients >55 years old presenting to ED with suspected of AAA, with abdominal, back or hip pain, n=5.	Prospective case series n=5.	presence of AAAs defined as external aortic diameter of 3 or more than 3 cm	Total number of positive AAA cases on ED ultrasound= 5	Small group of patients Convenience sampling as patients selected had frank features suggestive of rupturing, rAAAs Study involved patients with bigger aneurysms which are relatively easy to pick up. The study only looked at positive scans.
			Improvement in patient care	Number of positive AAAs on ED ultrasound, with symtoms and signs of leaking, underwent surgical repair without further imaging= 5
				Number of patients survived the repair= 4/5
Bentz S et al 2006 UK	All adults presenting to emergency department with the features suggestive of leaking AAAs	Best evidence topic report, (Best BET) reviewed four diagnostic cohorts.	Accuracy	In patients suspected of having AAA, ED bedside ultrasound is sensitive and specific for AAA	Reviewed diagnostic cohorts with small sample size 3/4 sudies considered are nonblinded No standardisation for the heterogeniety in training and experience of scanners, different machines used No metaanalysis performed
			Patient care	may improve patient care

Comment(s)

Rupture of an abdominal aortic aneurysm (rAAA) is a lethal vascular emergency with a mortality of 80%-90%. Two thirds of patients die before reaching the hospital and those who “arrive alive” have 50-80% mortality. Several studies have attempted to demonstrate the beneficial effects of early detection of AAA with emergency ultrasound (EMUS). AAA is silent in terms of presence and expansion and rupture is the most common way it presents to ED without specific prodrome.The accuracy of the physical examination to detect AAA is 68% at its best. This is even less sensitive with small aneurysms and/or with abdominal girth of more than 100 cm. The classical triad of hypotension, abdominal pain and pulsatile mass is seen in only 30%-50% of patients. EMUS has the potential to decrease the time-to-diagnosis, time-to-CT, and time-to-operative repair. The studies by Rubano et al, Dent et al, Knaut et al, Constantino et al, Kuhn et al and Tayal VS et al have demonstrated high accuracy of EMUS in AAA’s in unstable and stable patients. They have shown benefits of EMUS in ruptured AAA’s with no adverse outcome. Rubano,Knaut, Tayal and Kuhn showed that EMUS can be used as a rule out technique. Some of the studies have reported false positives and false neagative scans for AAAs with no adverse effects and with minimal measurement difference between ultrasound by emergency phtsicians and the CT abdomen.According to Blaivas et al,ED sonographers couldn’t visualise aortas in 8% of patient but this could be very well accorded to the presence of sonographic impediments which radiologists usually don’t encounter in fasted patients.Studies have successfully demonstrated that unstable patients with demonstrable AAA by ED ultrasound underwent surgical repair without further imaging.In this group of unstable patients no false positives or false negatives have been reported.None of the studies have reported any adverse outcomes secondary to ED ultrasound in suspected AAA patients.

Clinical Bottom Line

1. Ultrasound should be used to detect the presence of an AAA in suspected and high risk group by Emergency physicians trained in Emergency Ultrasound. 2. Emergency Ultrasound may be used to rule out the presence of AAAs only if the entire aorta from origin to bifurcation has been visualised and is of normal calibre. 3. High index of suspicion and the low threshold for ultrasound scan in suspected and risk group for AAA is of essence as it shortens the time to diagnosis, time to CT and the time to definitive management

References

1. Rubano E, Mehta N, Caputo W, Paladino L, Sinert R Systematic Review: Emergency Department Bedside Ultrasonography for Diagnosing Suspected Abdominal Aortic Aneurysm Academic Emergency Medicne 2013; 20:128-138
2. Dent B, Kendall R J, Boyle A A, Atkinson P R T. Emergency ultrasound of the abdominal aorta by UK emergency physicians: a prospective cohort study Emergency Medical Journal 2007; 24: 547-49
3. Knaut AL, Kendall JL, Patten R, Ray C Ultrasound measurements of aortic diameter by emergency physicians approximates results obtained by computed tomography J Emerg Med 2005 2005; 28: 119–26
4. Costantino TG, Bruno EC, Handly N Accuracy of emergency medicine ultrasound in the evaluation of abdominal aortic aneurysm J Emerg Med 2005; 29: 455–60.
5. Blaivas M, Theodore D. Frequency of incomplete abdominal visualisation by emergency department bedside ultrasound. Academic Emergency Medicine 2004; 11, 1: 103-105.
6. 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: 867–71.
7. Kuhn M, Bonnin RL, Davey MJ, Emergency department ultrasound scanning for abdominal aortic aneurysm: accessible, accurate, and advantageous. Ann Emerg Med 2000; 36: 219–23.
8. Walker A, Brenchley J, Sloan JP, Lolanda M, Venables H. Ultrasound by emergency physicians to detect abdominal aortic aneurysms: a UK case series. Emerg Med J 2004; 21: 257–9.
9. Bentz S, Jones J Towards evidence based medicine: best BETS from The Manchester Royal Infirmary. Accuracy of emergency department ultrasound scanning in detecting abdominal aortic aneurysm Emergency Medical Journal 2006; 23: 803-4