Diffusion-weighted magnetic resonance imaging in preoperative assessment of non–small cell lung cancer

    loading  Checking for direct PDF access through Ovid

Abstract

Background:

Diffusion-weighted magnetic resonance imaging (DWI) frequently shows heterogeneity of signal intensity (SI) in non–small cell lung cancer (NSCLC). The purpose of our study was to examine the association of SI and DWI patterns with histology, tumor invasiveness, lymph node metastasis, and 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) uptake in NSCLC.

Methods:

One hundred forty-five patients with NSCLC underwent preoperative DWI and FDG-PET. DWI patterns were visually classified as homogenous (HOM) (n = 81) or heterogeneous (HET) (n = 64). The former was further classified as faint (faint-HOM) (n = 27) or dark (dark-HOM) (n = 54) according to a cutoff value of SI. Associations of SI and DWI patterns with tumor histology, lymphatic or vascular invasion, pleural invasion, lymph node metastasis, and FDG uptake were evaluated.

Results:

All faint-HOM tumors were well-differentiated adenocarcinoma, whereas dark-HOM and HET tumors were less-differentiated adenocarcinoma or nonadenocarcinoma. Although the dark areas in HET tumors showed a dense aggregation of tumor cells, their faint areas showed abundant fibrovascular stroma or necrosis, or a well-differentiated part of adenocarcinoma. Tumor size and the frequencies of lymphatic or vascular invasion, pleural invasion, and nodal metastasis were highest in HET tumors, followed by dark-HOM and faint-HOM (P = .1-P < .001) tumors. Sixty-five tumors having at least 1 of the invasions or metastasis showed significantly higher SI than the 81 tumors without (P < .001). HET tumors had the highest FDG uptake, followed by dark-HOM and faint-HOM tumors; differences between the groups were significant (P < .01 to P < .001).

Conclusions:

The SI and heterogeneity of DWI reflect the histologic heterogeneity, tumor aggressiveness, and FDG-PET uptake in NSCLC.

Related Topics

    loading  Loading Related Articles