23 May, 16 | by rlloyd
Diagnosing small bowel obstruction (SBO) is bread-and-butter work for the emergency physician. It accounts for 2% of patients presenting to the ED with abdominal pain, and 20% of all surgical admissions. In the developing world the majority of SBO patients have had previous intra-abdominal surgeries causing adhesions… But I won’t delve into aetiology, let’s talk diagnostics.
The May 2016 EMJ issue’s ‘Image Challenge’ is a classic case. An adult male with significant surgical history (caecal adenocarcinoma with subsequent right hemicolectomy) presents to the ED complaining of abdominal pain and vomiting. His abdomen is distended and diffusely tender. Slam dunk.
How would I manage this patient in the ED? Having made them nil by mouth, started IV fluids, and given adequate analgesia, I’ll request the routine plain x-rays (abdominal and erect chest) that almost all of my acute abdomen patients get. A positive AXR for SBO (centrally distributed dilated loops with valvulae conniventes/air fluid levels) will prompt me to insert a nasogastric tube, and call the surgeons with a view to CT/a trip to theatre. I imagine that’s fairly common practice up and down the UK.
The image provided for the discussed case is an ultrasound image showing dilated, fluid-filled loops of bowel – suggestive of SBO. It turns out AXR is pretty useless at detecting SBO, particularly when you consider how much we rely on it traditionally, with a sensitivity of 50-60%. Ultrasound is a quick, cheap, radiation-free option available to us in the ED. And guess what? It’s more reliable than AXR for detecting SBO. Some evidence was published in the EMJ back in 2013 – let’s take a look.
Jang et al prospectively enrolled 76 adults in the ED who were suspected to have SBO, and going for a CT. All patients had an ultrasound exam performed by an EM resident, along with an AXR interpreted by a radiologist. The reference standard for SBO diagnosis was the CT result.
Each EM resident already had a basic understanding and experience of point-of-care scanning, having all undergone a prior introductory course. They were given only a 10 minute (!!) practical tutorial in SBO ultrasonography, and then 5 practice scans prior to being let loose on the study patients. A 10 minute tutorial is fairly minimal prep I think most would agree.
A positive ultrasound was defined as either:
- Dilated loops (>2.5cm) of fluid-filled dilated bowel proximal to normal/collapsed bowel
- Reduced peristalsis – back-and-forth movement of spot echoes inside fluid-filled bowel
Participants were taught to scan in the paracolic gutters bilaterally, epigastric and suprapubic regions. This is the standard approach to SBO ultrasound – see this video for a great tutorial on how to perform the scan. Interestingly, in this study the phased array probe was used instead of the curvilinear – the usual option for transabdominal scanning.
Ultrasonography comfortably outperformed plain radiography in detecting SBO. A sensitivity of 93.9% and specificity of 81.4% left AXR trailing behind with a sensitivity of 46% and specificity of 67%. Dilated loops on ultrasound proved to be far more sensitive than reduced peristalsis – probably because reduced peristalsis is generally considered to be a late finding in SBO, often seen with strangulation.
Of course there are limitations with this small study. There was a disproportionately high prevalence of SBO in the study population (33 out of 76 patients – 43%), bringing into question its external validity. The doctors performing the US exams volunteered themselves, indicating they were enthusiasts – potentially introducing what the authors describe as ‘ultrasound-interest’ bias. And of course, the participants knew they were being compared to a standard, bringing the Hawthorne Effect into play. Nonetheless, pretty convincing stuff.
And there’s more. Here is some further reading:
Something else to consider, is that additional information can be picked up when performing a bedside scan on a patient in whom there is a concern for SBO – free fluid between bowel loops, no peristalsis, or >3mm bowel wall thickening suggests bowel wall ischaemia. Gallstones in the presence of dilated loops? Think gallstone ileus. A lurking AAA might even be picked up.
I’m not suggesting that plain radiography no longer has a role in the suspected SBO patient. Surely though, adding bedside ultrasound to our list of investigative options is an opportunity to improve patient care. A negative scan would provide added reassurance when ruling the diagnosis out in less concerning patients; and we can expedite initial/definitive management in the high-risk patients who have an equivocal AXR.
Ultrasound will always be operator-dependent, but Jang et al have demonstrated that scanning for SBO is a relatively easy skill to acquire. Encouraging stuff.
What is your approach? Does ultrasound have a role in these patients? We would love to hear your thoughts in the comments.
Some more online resources
- Delabrousse, E., et al., CT of small bowel obstruction in adults. Abdom Imaging, 2003. 28(2): p. 257-66.
- John Eicken, S.E.F., Image challenge: Adult male with abdominal pain and vomiting. Emerg Med J 2016;33:5 337 doi:10.1136/emermed-2015-205181.
- Dr Henry Knipe, D.J.J., Small Bowel Obstruction, in Radiopaedia.org.
- Jang, T.B., D. Schindler, and A.H. Kaji, Bedside ultrasonography for the detection of small bowel obstruction in the emergency department. Emerg Med J, 2011. 28(8): p. 676-8.
- Guttman, J., et al., Point-of-care ultrasonography for the diagnosis of small bowel obstruction in the emergency department. CJEM, 2015. 17(2): p. 206-9.
- Alice Chao, M.a.L.G., MD, FACEP, Tips and Tricks: Clinical Ultrasound for Small Bowel Obstruction – A Better Diagnostic Tool? Emergency Ultrasound Section Newsletter – October 2014.