Sunday, December 29, 2013

Small Bowel Obstruction - A Likely Story?

The gist:  Small bowel obstruction (SBO) often presents with murky historical and physical features but quick bedside ultrasound exists as a modality that may expedite a patient's workup, with limited training and good sensitivity.  Uptake of US for SBO in US EDs has not been quick, despite nearly 20 years of literature, but perhaps Free Open Access Medical education (FOAM) such as this stellar tutorial on ultrasound (US) diagnosis will reduced the knowledge translation gap [1].

The case:  A 62 y/o male presented to the Janus General ED with a one day history of abdominal pain, mostly located in his right upper quadrant. He reported no fevers, or chill but did have nausea, poor appetite, and a few episodes of non-bloody, non-bilious emesis containing food/clear.  He reported a "normal" bowel movement last night and had a soft abdomen with normoactive bowel sounds.  As the chief complaint was "RUQ pain," a bedside ultrasound was placed on the patient's abdomen during the initial H&P.  Merely touching the probe the the patient's abdomen yielded the following image
Having watched the Ultrasound Podcast's episode on SBO merely weeks prior, the diagnosis of SBO was made expediently.  The workup at this point changed - the patient was admitted to the surgery service, who ordered a CT to localize the cause of the obstruction.

Recently, I've seen cases of SBO blending into the barrage of gastroenteritides fluxing into the ED, prompting me to review the literature.  I found a Taylor et al review of the likelihood ratios (LR - a concept FOAM helped me understand) of various diagnostic modalities helpful.
  • LR+ >5-10: Significantly increases likelihood of the disease
  • LR+  0.2 -5 (esp if close to 1): Does not modify the likelihood of the disease
  • LR+ below 0.1-0.2: Significantly decreases the likelihood of the disease
History and Physical - Sparse ED literature exists on diagnosis of SBO, particularly with regard to history and physical where the literature is minimal and consists of 2 studies.  The theme of these findings seem to be that if these clues are present, it increases the likelihood, but the absence of these findings doesn't really alter the likelihood of SBO.

Patients often endorse vague abdominal complaints, typically comprising intermittent pain, nausea, vomiting, constipation, and have a history of prior abdominal surgery [2,3]. Unfortunately these are non-specific and variably present.  While insensitive, the following demonstrated the most useful LRs:
  • Constipation  +LR = 8.8 and –LR = 0.59 
    • Sensitivity 37-44%
  • History of abdominal surgery  +LR = 3.86 and –LR = 0.19 
    • Sensitivity 69-85% [4]
The physical exam can play a role in determining the patient's degree of illness (ex: peritonitis, fever, etc) but physical signs are not consistent.  Impressive distention and "tinkling" bowel sounds are board answers and impressive for SBO, but these are not sensitive.
  • Abdominal distention +LR between 5.64-16.8 and –LR of 0.34 to 0.43 depending on the study 
  • Abnormal bowel sounds had a +LR of 6.33 and a –LR of 0.27 [4]
X-ray - the historical initial screening test of choice, yet has poor sensitivity, especially for partial SBO (PSBO) and rarely yields a diagnosis/changes management [5].  While x-ray is cheap with minimal radiation, results may falsely reassure providers and this may re
  • Sensitivities range between 59-93% [5].
  • +LR 1.55 (95% CI = 1.10-2.19) and -LR 0.59 (95% CI = 0.43-0.82) (*after the removal of a study with significant heterogeneity) [4]
Bedside Ultrasound in the ED- this modality is gaining momentum as it can provide a very rapid rule in/rule out, but it is still not recommended in core EM texts or become commonplace in many EDs [1,2,3].  The Eastern Association for the Surgery of Trauma (EAST) guidelines give US a cursory nod, in the event the surgery service questions the ability of US to diagnose SBO [6].
  • Sensitivity: 97% (95% CI= 92–99); Specificity 90% (95% CI=84–95) [4] 
  • +LR 9.5 and -LR 0.04 [4]
  • Good interrater reliability compared with formal US- Kappa of 0.8 [4]
The quick run down: place a curvilinear or phase array probe on the abdomen and search for dilated loops of bowel >2.5 cm.  Decreased peristalsis is a later finding also present in SBO, with lower sensitivity [7].

Advantages: Quick, cheap, easy, able to identify other etiologies, may expedite work-up.  In the Jang et al study, the residents had, in addition to their prior training, a 10 minute course and 5 SBO scans prior to enrolling patients - a minimal time investment.
Disadvantages: Images may not be available to inpatient/admitting team, does not demonstrate the cause of the obstruction.  Furthermore, no literature demonstrates that ED bedside US demarcates transition points well, but future studies may elucidate utility in this area.  Ileus and SBO may appear similarly on US, so providers should entertain this in the differential.

Note bene: The bedside ED literature in this realm is small in numbers (2 prospective studies), uses CT or surgical/1 month follow up as the gold standard, and some of the confidence intervals in studies are fairly wide and have sensitivities down to 74.5% when solely looking for dilated bowel.

CT - EAST gives a level I recommendation to CT scans in all patients with SBO to determine etiology and grading [6].  Often, these patient's get CT scans to evaluate for other pathology; however, CT scans can take time, contributing to issues with flow.  CT may be needed for the admitting service, but there's often no need for this to occur prior to getting the admission process started.  With regard to diagnosis, in the two relevant studies that used thin slices, CT demonstrates excellent sensitivity.
  • Sensitivities: 96% (95% CI = 80% to 100%) and 93% (95% CI = 87% to 97%)
  • Specificity of 100% (95% CI = 69% to 100%) and 93% (95% CI = 88% to 96%) [4]
There's also no need for the time consuming oral contrast.
Labs - may help in risk stratifying patient based on degree of illness but add little to the diagnosis.
  • Lactate - if elevated, may indicate ischemia or sepsis
  • WBC >20,000 cited as concerning for ischemia, abscess, or peritonitis [2,3]
  • Elevated BUN or creatinine may indicate volume depletion
How the literature/FOAM has shaped my practice:
  • Do the history and physical, but don't be entirely reassured. Absence of distention or a vomiting patient with diarrhea in the midst of several gastroenteritis patients may cause me to engage in premature closure without recognizing the limitations of these historical and physical factors.
  • Bring the ultrasound into the room of abdominal pain patients while doing the H&P and have a low threshold for taking a quick look when SBO is a consideration.
1.  Carpenter CR, Pines JM. The end of X-rays for suspected small bowel obstruction? Using evidence-based diagnostics to inform best practices in emergency medicine. Acad. Emerg. Med. 2013;20(6):618–20.
2.  Vicario SJ, Price TG.  "Bowel Obstruction and Volvulus." Tintinalli's Emergency Medicine: A Comprehensive Study Guide.  7th ed. pp 581-583.
3.  Torrey SP, Henneman PL.  "Disorders of the Small Intestine."  Rosen's Emergency Medicine. pp 1184-86.
4.  Taylor MR, Lalani N. Adult small bowel obstruction. Acad. Emerg. Med. 2013;20(6):528–44.
5. Böhner H, Yang Q, Franke C, Verreet PR, Ohmann C. Simple data from history and physical examination help to exclude bowel obstruction and to avoid radiographic studies in patients with acute abdominal pain. Eur. J. Surg. 1998;164(10):777–84. 
6.  Maung AA, Johnson DC, Piper GL et al. Evaluation and Management of Small-Bowel Obstruction.  J Trauma. 73(5):S362-S369, November 2012
7. Jang TB, Schindler D, Kaji AH.  Bedside ultrasonography for the detection of small bowel obstruction in the emergency departmentEmerg Med J. 2011 Aug;28(8):676-8.

1 comment:

  1. Great post - loving the trend to US in your recent stuff! Clearly you are moving to the dark side. :-)