Introduction

Lung cancer is one of the leading causes of cancer-related death from malignant tumors worldwide, with a high incidence 1. Tumor angiogenesis is crucial to its growth, invasion, and metastasis. DCE-CT is widely used in clinical lung cancer screening because of its high spatial resolution, high sensitivity, and rapid imaging. However, the cumulative ionizing radiation associated with repeated CT examinations is a concern. Recent studies have proved that GRASP can be used to optimize the free-breathing DCE-MRI of several organs sensitive to motion-induced artifacts in traditional Cartesian imaging, such as liver, esophagus, urinary system, and so forth [7-8], yielding satisfactory results. Therefore, this study aimed to explore the image quality of the free-breathing GRASP technique for high-resolution imaging of the lungs in a group of patients with lung cancer compared with the image quality of traditional Cartesian breath-holding volume interpolation scan (BH-VIBE).

Materials and Methods

This study was approved by the Anqing Municipal Hospital Medical Ethics Committee (Approval Number: Medical Ethics Review [2022] No. 96). A total of 30 patients diagnosed with lung cancer through CT scans (20 male, 10 female; median age, 67 years; range, 48-91 years) using a 3T magnetic resonance imaging system (MAGNETOM Vida, Siemens Healthcare, Germany) with an 18-channel body phased-array coil from October 2022 to December 2022 were enrolled. Coronal T2-weighted imaging (HASTE sequence), axial diffusion-weighted imaging (EPI sequence), axial T1-weighted imaging (BH-VIBE sequence), axial GRASP (radial VIBE sequence), and axial T1-weighted contrast-enhanced imaging (BH-VIBE sequence) were performed. The total scan time for each patient was approximately 15 minutes. The specific imaging parameters of GRASP were as follows: FOV = 350 × 350 mm2; matrix size = 256 x 256; number of slices = 64; slice thickness = 3 mm; flip angle = 12°; in-plane spatial resolution = 1.2 x 1.2 mm2; TR/TE = 3.50 milliseconds/1.35 milliseconds; and total scan time = 366 seconds. Gadolinium-based contrast agent (GE Healthcare, Shanghai) was administered at a rate of 3.0 mL/s (0.2 mmol/kg body weight) 60 seconds after the start of data acquisition, followed by a 20-mL saline flush at the same rate. All participants were thoroughly informed regarding the examination process and provided informed consent before undergoing MRI scans. The inclusion criteria for patients were as follows: (1) solid lesion diameter greater than 10 mm; (2) absence of metallic implants; (3) patient cooperation with MR examination; and (4) availability of complete clinical and pathologic data. All selected GRASP images were visually assessed for quality compared with the corresponding BH-VIBE images and CT images. Two radiologists (each with 13 and 10 years of experience in chest MRI) blinded to the sequence information independently performed the visual image quality assessment. For each set of images, the readers assigned scores for overall image quality, lesion delineation, overall artifact level, and diagnostic confidence using a 4-point Likert scale: 4 = excellent; 3 = good; 2 = acceptable; and 1 = poor (nondiagnostic). Nonparametric paired-sample two-tailed Wilcoxon signed-rank tests were employed to compare readers’ scores between pre-contrast and post-contrast GRASP and BH-VIBE images. A P value less than .05 indicated a statistically significant difference. Cohen's kappa coefficient was used to assess the agreement between various image quality scores. Statistical analysis was performed using SPSS[Editor1] (version 25.0, Chicago). [Editor1]Please provide manufacturer details.

Results

Finally, the study included 30 patients with lung cancer (20 male and 10 female; age 48-91 years; median age 67.4 years. The tumor sizes ranged from 1.5 to 7.1 cm. Histologically, the study was performed on 21 adenocarcinomas, 4 squamous cell carcinomas, 4 small cell lung carcinomas, and 1 non-small-cell lung carcinoma. The visual image quality scores for conventional Cartesian BH-VIBE and GRASP imaging indicated that all GRASP images had clear lesion delineation. Remarkable imaging performance was achieved before and after contrast injection, with all evaluation categories having P values less than .05. BH-VIBE image quality scores were consistently lower than those of GRASP, with considerable differences in the overall image quality and artifact level (P <.01).

Discussion and conclusion

The study demonstrated the advantages of the GRASP sequence in effectively eliminating motion artifacts related to respiration and autonomous cardiac pulsation. The image quality of GRASP was remarkably superior to that of BH-VIBE (P <.01). Therefore, GRASP is a highly practical clinical lung MR imaging technique, which allows patients to breathe freely during MR scans, and hence is more patient-friendly and minimally impacts the image quality and diagnostic confidence.

Acknowledgements

None