Differentiation of small cell lung cancer and non-small cell lung cancer with diffusion-weighted MR imaging ;at 3.0 Tesla and optimization of b values / 实用医学杂志

ABSTRACT

Objective To investigate the feasibility of diffusion-weighted imaging (DWI) for differentia-tion of small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) at 3.0 T MRI. This study also aims to optimize b values. Methods Thirty-six patients with lung cancer proved by pathology were examined by T2WI and DWI (b = 400, 600 and 800 s/mm2) and apparent diffusion coefficient (ADC) values of each lesion were calculated among different b values. The receiver operating characteristic (ROC) curves were used to calcu-late the difference between the ADC of SCLC and NSCLC and to evaluate the diagnostic capability. Analysis of variance was used to determine difference in ADC of NSCLC histological types at the optimal b values. Results The ADC values of SCLC and NSCLC decreased with increasing b values , and the differences were statistically significant (P = 0.002, P < 0.001). ROC analysis showed that the area under curve (AUC) values were 0.775、0.892、0.804 (b = 400, 600, 800 s/mm2, respectively). The AUC with a b value of 600 s/mm2 was the largest. When an ADC of 1.296 × 10-3 mm2/s was considered as a threshold, the sensitivity, and specificity were 76.9%, and 90.0%, respectively. Conclusion DWI reflects the diffusion motion of water protons in lung carcinoma and it has moderate diagnosis potency in differentiating SCLC and NSCLC. The optimal b value is 600 s/mm 2.