Hyperattenuating adrenal lesions in lung cancer: biphasic CT with unenhanced and 1-min enhanced images reliably predicts benign lesions

To investigate usefulness of biphasic computed tomography (CT) in characterizing hyperattenuating adrenal lesions in lung cancer.

This retrospective study included 239 patients with lung cancer who underwent adrenal CT for hyperattenuating (> 10 Hounsfield unit) adrenal lesions. Adrenal CT comprised unenhanced and 1-min and 15-min enhanced images. We dichotomized adrenal lesions depending on benign or metastatic lesions. Reference standard for benignity was histologic confirmation or ≥ 6-month stability on follow-up CT. Two independent readers analyzed absolute (APW) or relative percentage wash-out (RPW) using triphasic CT, and enhancement ratio (ER) or percentage wash-in (PWI) using biphasic CT (i.e., unenhanced and 1-min enhanced CT). Criteria for benignity were as follows: criteria 1, (a) APW ≥ 60% or (b) RPW ≥ 40%, and criteria 2, (a) ER > 3 and (b) PWI > 200%. We analyzed area under the curve (AUC) and accuracy for benignity, and inter-reader agreement.

Proportion of benign adrenal lesion was 71.1% (170/239). For criteria 1 and 2, AUCs were 0.872 (95% confidence interval [CI], 0.822–0.911) and 0.886 (95% CI, 0.838–0.923), respectively, for reader 1 (p = 0.566) and 0.816 (95% CI, 0.761–0.863) and 0.814 (95% CI, 0.759–0.862), respectively, for reader 2 (p = 0.955), and accuracies were 87.9% (210/239) and 86.2% (206/239), respectively, for reader 1 (p = 0.479) and 81.2% (194/239) and 80.3% (192/239), respectively, for reader 2 (p = 0.763). Weighted kappa was 0.725 (95% CI, 0.634–0.816) for criteria 1 and 0.736 (95% CI, 0.649–0.824) for criteria 2.

Biphasic CT can reliably characterize hyperattenuating adrenal lesions in patients with lung cancer.