Diagnostic Accuracy of Abdominal wall Ultrasonography and Local Wound Exploration in Predicting the Need for Laparotomy following Stab Wound
Emergency (Tehran) Epub January 2017 - Pubmed Link
Take Home Points
1. In stable patients, who otherwise do not require the OR, ultrasound may be as specific as local wound exploration in predicting need for laparotomy following anterior stab wounds.
2. Ultrasound is not sensitive enough to rule out the need for laparotomy in these patients.
Background
Here's the deal: if you get stabbed in the abdomen - it matters how deep the knife goes. If it goes into the peritoneum, that = bad. These patients will get an exploratory laparotomy to see if any bowel or organs were damaged. However, about 25% of abdominal stab wound are superficial. These wounds are unlikely to benefit from a laparotomy. We don't want to subject them to the risks of surgery without benefit, so we want to avoid bringing these patients to the operating room (OR) without missing any stab wounds that are deep. Here are the options in determining if the peritoneum was violated - diagnostic peritoneal lavage (rarely used), CT (may not be sensitive for certain injuries), local wound exploration (commonly employed for this purpose), or serial clinical exams. Ultrasound is being offered up here as a potential alternative that is fast, noninvasive, and could potentially help with decision making.
Question
Can POCUS accurately diagnose the need for laparotomy in patients with anterior abdominal stab wounds?
Population
Inclusion
Adults >18
Anterior abdominal stab wound
Exclusion
Hemodynamic instability
Peritoneal signs
Protrusion of abdominal organs
GI bleeding
Pregnancy
Instrument in situ (like knife sticking out of abdomen)
+FAST
Evisceration
Multiple wounds
Need for emergent laparotomy
Design
Cross sectional, from 3/2013 to 3/2015
All patients received ultrasound and local wound exploration.
Both of these methods was compared to decision made by the “in charge surgeon” of whether or not to perform laparotomy. It is unclear if the surgeons were blinded to ultrasound, but it seems they were not blinded to the wound exploration.
Test characteristics were calculated for both ultrasound and local wound exploration.
Anterior abdominal wall defined as superior to inguinal ligaments, medial anterior axillary lines, two fingerbreadths inferior to costal margins
Intervention
Patients with abdominal stab wounds presented to the emergency department
Underwent usual evaluation
Eligible patients underwent abdominal wall ultrasound
Then they underwent local wound exploration by senior surgery residents
Followed until discharge
Who did the scans?
Trained senior emergency medicine residents
Trained by radiologist by practicing on 10 sheep cadaver models
The Scan
Linear probe
Abdominal Wall Ultrasonography (AWU): Scanned the abdominal wall in a 10 x 10 cm area surrounding the wound
A positive study was an impairment of abdominal wall integrity, a hazy fascial defect (see the article for examples because these images are hard to come by)
Results
50 cases
Mean age 28.44
17 (34%) cases received laparotomy
Primary Outcome
Here are the actual numbers without rounding, with the confidence intervals
AWU
Area under ROC 81.96%
Sens 70.58 (44.04-88.62)
Spec 93.33% (76.49-98.83)
+LR 6 (1.63-22.03)
-LR 0.17 (0.07-0.40)
LWE
Area under ROC 90.78%
Sens 88.23% (62.25-97.93)
Spec 93.33% (76.49-98.83)
+LR 7.5 (2.01-27.88)
-LR 0.07 (0.01-0.27)
Limitations
Main problem here is that the outcome was predicting laparotomy. This is a good patient centered outcome BUT there are a lot of factors that may play into the decision that are not controlled for in this study. The surgeons likely trusted the local wound exploration (which is understandable) and since there was no blinding to this, it may have influenced their decision and made the LWE appear more accurate compared to ultrasound. When you are using ultrasound not necessarily to diagnose but to predict a complex decision - it becomes difficult to use the data clinically. This study does not actually tell us how accurate ultrasound is in determine peritoneal penetration, it only tells us that people with a positive ultrasound are more likely to get laparotomy (which could due to any number of confounders).
Most cases were knife inflicted - unclear how accurate in gunshot wounds (where path can be even less predictable) or other penetrating trauma.
Only performed 10 x 10 cm area around wound - may have improved sensitivity if used larger area.
Confidence intervals were fairly large.
Likely would require extra training since this is not standard focused exam.
Take Home Points
1. In stable patients, who otherwise do not require the OR, ultrasound may be as specific as local wound exploration in predicting need for laparotomy following anterior stab wounds.
2. Ultrasound is not sensitive enough to rule out peritoneal penetration in these patients.
Our score
