Am Fam Physician. 2010;81(2):214-216
Background: Acute abdominal pain accounts for up to 10 percent of emergency department visits. The related diagnostic imaging is associated with increases in hospital cost and radiation exposure to patients. Ideally, a rational imaging strategy would allow timely and accurate identification of urgent conditions, and would minimize exposure to unnecessary forms of imaging. Laméris and colleagues conducted a prospective, fully paired diagnostic accuracy study to evaluate various imaging strategies for detecting urgent conditions in patients with acute abdominal pain.
The Study: Eligible patients presented to an emergency department with nontraumatic abdominal pain starting more than two hours but less than five days before evaluation. Patients whose symptoms were not deemed severe enough to warrant imaging were excluded, as were pregnant women and patients in acute shock from gastrointestinal bleeding or ruptured aortic aneurysm.
Patients initially received clinical and pertinent laboratory evaluation, followed by a structured imaging protocol including chest and abdominal radiography, abdominal ultrasonography, and computed tomography (CT). No oral or rectal contrast was used. The rates of diagnostic accuracy of 11 different imaging strategies were compared, including clinical diagnosis with and without plain radiography, ultrasonography, CT, and a combination of various modalities. The American College of Radiology recommendation (i.e., ultrasonography for right upper quadrant pain and CT for pain in other quadrants or diffuse abdominal pain) was also evaluated, as were age and body mass index–driven imaging strategies.
Results: The authors evaluated 1,021 adult patients. Urgent diagnoses were identified in 661 patients (65 percent), with acute appendicitis being the most common final diagnosis. The highest sensitivity in identifying urgent conditions (94 percent) came from a conditional strategy that used CT after negative or inconclusive ultrasonography (see accompanying table). Although 49 percent of patients would still require CT with this strategy, this resulted in substantially less CT-related radiation exposure than with the other strategies reviewed, and maintained high sensitivity.
| Imaging strategy | Sensitivity (%) | Specificity (%) | False negative (%) | False positive (%) |
|---|---|---|---|---|
| Clinical diagnosis | 88 | 41 | 12 | 27 |
| Clinical diagnosis after plain film radiography | 88 | 43 | 12 | 26 |
| Ultrasonography in all patients | 70 | 85 | 30 | 11 |
| CT in all patients | 89 | 77 | 11 | 12 |
| Ultrasonography in all patients; CT if ultrasonography negative or inconclusive | 94 | 68 | 6 | 16 |
| Ultrasonography in all patients; CT if ultrasonography inconclusive | 85 | 76 | 15 | 14 |
| If right upper quadrant tenderness, use ultrasonography; if tenderness diffuse or in other quadrants, use CT (American College of Radiology recommendation) | 89 | 78 | 11 | 12 |
Conclusion: For evaluating acute abdominal pain, a conditional strategy using CT after negative or inconclusive ultrasonography had the highest sensitivity in detecting urgent conditions, with the lowest overall exposure to radiation.
