OBJECTIVEThe purpose of this study is to investigate whether low-dose spiral chest CT scan can replace standard-dose CT scan in detecting pulmonary metastases for patients with gestational trophoblastic tumor (GTT).

METHODSTotally, 34 GTT patients underwent 56 chest CT scans for the assessment of pulmonary metastasis. All patients received CT examination both at standard-dose (120 KV, 150 mAs, pitch 1, and a standard reconstruction algorithm) and low-dose CT (120 KV, 40 mAs, pitch 2, and a bone reconstruction algorithm) simultaneously each time. The images were interpreted by two radiologists independently. A metastasis by CT image was defined as a nodule within lung parenchyma that could not be attributed to a pulmonary vessel. The number of lesions detected at each dose protocol was recorded. The size of each lesion was measured and categorized as < 5 mm, 5 - 10 mm or > or = 10 mm. The differences in detection of the lesions between the standard- and low-dose CT protocols were compared using Wilconxon signed rank test.

RESULTS1417 lesions were detected at the standard-dose, whereas 1214 lesions were found by low-dose CT. Lesions < 5 mm detected by low-dose CT were fewer than that detected by standard-dose CT (Z = -3.368, P = 0.000), though there was no statistically significant difference between the standard- and low-dose CT in detecting lesion > or = 5 mm (Z = -0.055, P = 0.957). Moreover, the risk score of the patients was not affected either. The sensitivity of low-dose CT was 69.16% for all size of lesions, 58.50% for < 5 mm, 87.07% for 5 - 10 mm, and 97.01% for > or = 10 mm. The positive predictive value for different sizes of lesion was 80.71% (all sizes), 73.82% (5 mm), 88.86% (5 - 10 mm), and 98.48% (> or = 10 mm), respectively.

CONCLUSIONLow-dose chest CT can replace the standard-dose chest CT as a screening and follow-up examination to assess the change in pulmonary metastasis for patients with gestational trophoblastic tumor.