[The diagnostic value of dynamic CT in solitary pulmonary nodules: a prospective study]

To evaluate imaging features of benign and malignant solitary pulmonary nodules (SPN) using dynamic computed tomography (dCT) to improve the accuracy of radiological diagnosis.

Fifty-one patients with SPN were studied by dCT. In this procedure, a bolus of 100 ml contrast medium was administrated intravenously at a rate of 4 ml/sec. The same-located dynamic scans were carried out from 15 sec to 120 sec following the injection. Time-attenuation curves (TAC) were created according to circular or oval ROI drawn over nodules. Histopathological diagnosis was considered as the golden standard in all patients. Double-blind examination and evaluation were carried out and the data were analyzed statistically with Mann-Whitney U test.

Thirty eight cases were diagnosed to be malignant SPN (mSPN) and 13 cases to be benign SPN (bSPN). The benign SPN were further divided into two groups, bSPN(1) consisting 6 cases with chronic pneumonitis, nodular tuberculosis or sclerosing hemangioma and bSPN(2) consisting 7 cases with tuberculoma, pulmonary cyst, pulmonary sequestration or mycosis ball. There were statistically significant differences between mSPN and bSPN(2) in peak heights of enhancement (87.6 HU vs. 57.8 HU, P < 0.01), enhancement values (peak heights- unenhanced CT values, 59.6 HU vs. 11.1 HU, P < 0.01). However, no statistically significant differences of those two values existed between mSPN and bSPN(1). TAC of mSPN reached to peak height more rapidly and remained a plateau. TAC of bSPN(1)s showed similar changes to that of mSPN despite a delayed reach to a peak height or even a descending branch. TAC of the bSPN(2)s was lower and flatter without peak height. If a threshold of 20 HU was taken for dCT, the lesions with < or = 20 HU were diagnosed as benign, and the lesions with > 20 HU were diagnosed as malignant, with a sensitivity of 100%, a specificity of 54.0% and an accuracy of 88.4%.

(1) Absence of the marked enhancement (< or = 20 HU) in dynamic CT is strongly predictive of benignancy. (2) The peak height and enhancement value of dCT are helpful to differentiate malignant SPNs from benign ones. (3) The TAC configuration is helpful in differentiating malignant SPNs from benign ones. Descending branches could be found in some benign lesions, but not in the malignant ones. The TAC of tuberculoma and mycosis ball is usually relatively low and flat without any peak height.