Objective: The objective of our study was to investigate the accuracy of sonography in patients with symptoms suggestive of a hernia and normal or equivocal clinical examination findings.

Subjects and methods: Fifty-nine consecutive patients (47 men, 12 women; median age, 51 years; range, 19-82 years) were enrolled in a prospective study of sonography and herniography for investigation of inguinofemoral pain. All patients were referred with a history suggestive of hernia but with equivocal clinical features by three experienced surgeons. All patients underwent sonography and herniography examinations performed by experienced radiologists blinded to clinical details. The imaging variables recorded for each side were normal (including posterior inguinal wall bulging), hernia (indirect, direct, femoral, and abdominal wall), or nondiagnostic. The percentage of exact agreement, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for sonography and herniography and were compared with surgery when performed; then all sides for sonography were compared with herniography.

Results: Surgery was performed in 18 patients (31%) on 21 sides and found hernia (n = 20) and patulous posterior inguinal wall (with no hernia) (n = 1). Compared with surgery, the results of sonography versus herniography, respectively, were exact agreement (91% vs 71%), sensitivity (95% vs 70%), specificity (100% vs 100%), PPV (100% vs 100%), and NPV (50% vs 14%). The sensitivity of sonography was significantly higher than that of herniography (McNemar test, p = 0.025). Both techniques had one false-negative in the same patient. Herniography had five additional false-negatives identified as hernias at sonography and surgery. Compared with herniography as the reference, the sonography findings were in exact agreement in 91% (107/118) of the cases; and sensitivity was 90% (19/21); specificity, 91% (88/97); PPV, 68% (19/28); and NPV, 98% (88/90).

Conclusion: Sonography is an accurate technique for the detection of inguinofemoral hernias in patients with clinically equivocal findings.