Use of the Focused Assessment with Sonography for Trauma (FAST) exam in patients with pelvic fractures has been reported as unreliable. We hypothesized that FAST is a reliable method for detecting clinically significant intra-abdominal hemorrhage in patients with pelvic fractures.
All patients with pelvic fractures over a 10-year period were reviewed at a Level I trauma center. The predictive ability of FAST was assessed by calculating the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) against the gold standard of either CT or laparotomy findings. The FAST exam was considered “false negative” if findings at laparotomy indicated traumatic intra-abdominal hemorrhage. Likewise, the FAST exam was considered “false positive” if either CT or findings at laparotomy indicated no intra-abdominal hemorrhage. Hemodynamic Instability Scores (HIS) were calculated for all patients.
1456 patients with pelvic fractures and an initial FAST were reviewed; 1219 (83.7%) underwent FAST and either CT or operative exploration. Median age was 43 years (interquartile range 26-56) and mean Injury Severity Score (ISS) was 18.5 ± 12.3. The sensitivity and specificity for FAST in this group of patients with pelvic fracture was 85.4% and 98.1%, respectively. The PPV and NPV were 78.4% and 98.8%, respectively. Of 21 patients with a false positive FAST 15 (71.4%) were confirmed with a negative CT scan, and 6 (28.6%) underwent laparotomy without findings of intra-abdominal hemorrhage. Of 13 patients with a false negative FAST, all were identified with positive findings at the time of laparotomy. The specificity of the FAST exam remained high regardless of HIS grade.
The false positive rate of FAST exam for intra-abdominal hemorrhage is 1.1%. These data suggest that a positive FAST in this clinical scenario should be considered to represent intra-abdominal fluid. This series contradicts prior reports that FAST is unreliable in patients with pelvic fracture.
Level III Diagnostic Tests and Criteria.

References

PubMed