Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

Using ultrasound to detect peritoneal fluid in a pregnant patient with abdominal trauma

Three Part Question

In [a pregnant patient who has suffered a blunt abdominal trauma] is [abdominal ultrasound effective] at [detecting peritoneal fluid]?

Clinical Scenario

A pregnant patient in her 3rd trimester is brought in by ambulance following an accident in a car where she was the driver. You wonder how effective ultrasound would be in detecting the presence of peritoneal fluid in such a late stage of pregnancy.

Search Strategy

Medline 1950 to November Week 1 2008 using Ovid Interface
EMBASE 1980-2008 Week 1 using Ovid Interface
The Cochrane Library

[{exp Pregnancy/ or Pregnancy.mp.} OR { exp Pregnancy Complications/ OR Pregnancy Complications.mp.}AND {exp Abdominal Injuries/ OR Abdominal Injuries.mp. OR Abdominal trauma.mp. }AND { exp Ultrasonography/ OR ultrasound.mp.} AND { exp Ascitic Fluid/ OR peritoneal fluid.mp.}]

Search Outcome

3 papers were found of which 3 were relevent

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Ormsby et al.
2005
USA
328 pregnant women who presented with blunt abdominal trauma and underwent FAST in an Emergency Department in California from January 1995-June 2002. Mean age was 27.1 9.3 years Pregnant women in each of the 3 trimesters were included in this study. Retrospective cohort study.Intra-abdominal injury (IAI)23 patients from 328 who were scanned.The study was dependent on dictated reports. Many patients with negative US scans were followed clinically without further imaging studies which could have meant minor IAI were missed. Not all consecutive blunt abdominal trauma patients underwent FAST. As the USS were performed in the ED patients may have been catheterized before the scan, which may have decreased the visualization of pelvic FF. These are the same patients reported in Richards et al, but analysed in a different way. This study is from the same centre as the study by Goodwin et al and there is overlap of the study period. It also mentions 3 patients with coincidental ruptured ectopic pregnancies, presumably the same 3 seen in the Goodwin study.
Free fluid (FF)29 patients from 328, 14 true positive, 15 false positive. 299 showed no FF (290 true negative, 9 false negative) Sensitivity of FF for IAI = 61%, specificity = 95%, PPV = 48%, NPV = 97%, accuracy = 93%.
Goodwin et al.
2001
USA
127 pregnant women who presented with blunt abdominal trauma underwent FAST in an Emergency Department in California from January 1992-November 1999. Mean age was 24.4 6.4 years Pregnant women in each of the 3 trimesters were included in this study. Retrospective cohort studyIntra-abdominal injury (IAI)7 out of 127 patients. Sensitivity of US for detecting IAI = 71%, specificity = 97%.Only 127 of 280 pregnant blunt trauma patients (61%) had an abdominal ultrasound examination. The majority of patients did not have a Gold Standard investigation (CT or laparotomy). Three patients in the 1st trimester had positive ultrasound scans for free fluid but turned out to have coincidental ruptured ectopic pregnancies. The authors did not include them in their calculations. If they had then the specificity of ultrsound to detect intra-abdominal injury would have been 95%.
Free fluid (FF)US detected 5/6 patients who had FF and IAI. 3 patients had FF but no IAI. Sensitivity of US for detecting FF = 83% (95% CI 36-100), specificity = 98% (95% CI 93-100). PPV for US for detection of FF = 100% (95% CI 76-100), PPV for US for detection of FF from IAI = 63% (95% CI 24-91), NPV for US for detection of patients with FF from IAI = 99%. Error rate (% times US was incorrect) = 2% (1 false negative, 2 false positives)
3rd trimester59 patients, 5 had IAI, 1 US failed to detect FF.
Richards et al.
2004
USA
Review of 2319 US examinations performed in the Emergency Department of a level 1 trauma centre from January 1995-June 2002 for females aged 10-50 years presenting with blunt abdominal trauma.Retrospective coohort studyPerformance of US in detecting FF and/or IAI for all pregnant patients (n=328)Sensitivity = 61% (95% CI 39-80), specificity = 94.4% (95% CI 91-97), PPV = 45% (95% CI 27-64), NPV = 97% (95% CI 94-99), accuracy = 92.1% (95% CI 89-95).A retrospective review. Few patients had a Gold Standard investigation after a negative US scan. Not all blunt abdominal trauma patients had an US scan. Several radiologists interpreted the scans.
Performance of US in detecting FF and/or IAI for 1st trimester patients (n=81)Sensitivity = 90% (95% CI 56-100), specificity = 89% (95% CI 79-95), PPV = 53% (95% CI 28-77), NPV = 98% (95% CI 92-100), accuracy = 89% (95% CI 80-95).
Performance of US in detecting FF and/or IAI for 2nd trimester patients (n=134)Sensitivity = 25% (95% CI 10-81), specificity = 97.7% (95% CI 93-100), PPV = 25% (95% CI 10-81), NPV = 97.7% (95% CI 93-100), accuracy = 95.5% (95% CI 91-98).
Performance of US in detecting FF and/or IAI for 3rd trimester patients (n=113)Sensitivity = 44% (95% CI 14-79), specificity = 94.2% (95% CI 88-98), PPV = 40% (95% CI 12-74), NPV = 95.1% (95% CI 89-98), accuracy = 90.3% (95% CI 83-95).
Performance of US in detecting FF and/or IAI for all non-pregnant patients (n=1991)Sensitivity = 71.2% (95% CI 64-77), specificity = 97.4% (95% CI 97-98), PPV = 76.2% (95% CI 69-82), NPV = 96.7% (95% CI 96-97), accuracy = 94.6% (95% CI 94-96).
Performance of US in detecting FF and/or IAI for all patients pregnant and non-pregnant (n=2319)Sensitivity = 70.1% (95% CI 64-76), specificity = 96.9% (95% CI 96-98), PPV = 72% (95% CI 65-77), NPV = 96.7% (95% CI 96-97), accuracy = 94.3% (95% CI 93-95).

Comment(s)

Focused assessment sonographic for trauma or FAST as it is commonly known is a rapid ultrasound examination that can be done in the emergency department to screen for free fluid in the pelvis and abdomen. Any free fluid in these areas may, in the context of trauma, be due to an intra-abdominal injury. Four areas are scanned: the subxiphoid, the right upper quadrant, the left upper quadrant and the suprapubic area. The speed with which this can be performed is of great value as alternative imaging methods such as CT scans take longer to perform. In a pregnant patient ultrasound has the distinct advantage that the foetus is not exposed to harmful radiation, which is particularly damaging in the first trimester. All three studies come from the same centre. Two studies analysed the same data and there appears to be overlap with the third.

Clinical Bottom Line

FAST can be used in pregnant women in all trimesters presenting with blunt abdominal trauma to detect peritoneal fluid. However as a test it performss less well in pregnant than non-pregnant patients.

References

  1. Ormsby EL. Geng J. McGahan JP. Richards JR. Pelvic free fluid: clinical importance for reproductive age women with blunt abdominal trauma. Journal Article] Ultrasound in Obstetrics & Gynaecology Sep 2005; 271-8
  2. Goodwin H, Holmes JF, Wisner DH. Abdominal ultrasound examination in pregnant blunt trauma patients. Journal of Trauma-Injury Infection & Critical Care. 2001, 50 (4); 689-93
  3. Richards JR, Ormsby EL, Romo MV, Gillen MA, McGahan JP. Blunt abdominal injury in the pregnant patient: detection with US. Radiology 2004, 233: 463-470.