Background: The objective of this study is to determine the reliability and validity of ultrasonography (US) in diagnosing incisional hernias in comparison with computed tomography (CT). The CT scans were assessed by two radiologists in order to estimate the inter-observer variation and twice by one radiologist to estimate the intra-observer variation. Patients were evaluated after reconstruction for an abdominal aortic aneurysm or an aortoiliac occlusion.
Methods: Patients with a midline incision after undergoing reconstruction of an abdominal aortic aneurysm or aortoiliac occlusion were examined by CT scanning and US. Two radiologists evaluated the CT scans independently. One radiologist examined the CT scans twice. Discrepancies between the CT observations were resolved in a common evaluation session between the two radiologists.
Results: After a mean follow-up of 3.4 years, 40 patients were imaged after a reconstructed abdominal aortic aneurysm (80% of the patients) or aortoiliac occlusion. The prevalence of incisional hernias was 24/ 40 = 60.0% with CT scanning as the diagnostic modality and 17/40 = 42.5% with US. The measure of agreement between CT scanning and US expressed as a Kappa statistic was 0.66 (95% confidence interval [CI] 0.45-0.88). The sensitivity of US examination when using CT as a comparison was 70.8%, the specificity was 100%, the predictive value of a positive US was 100%, and the predictive value of a negative US was 69.6%. The likelihood ratio of a positive US was infinite and that of a negative US was 0.29. The inter- and intra-observer Kappa statistics were 0.74 (CI 0.54-0.95) and 0.80 (CI 0.62-0.99), respectively.
Conclusions: US imaging has a moderate sensitivity and negative predictive value, and a very good specificity and positive predictive value. Consistency of diagnosis, as determined by calculating the inter- and intra-observer Kappa statistics, was good. The incidence of incisional hernias is high after aortic reconstructions.