Limits...
Small malignant hepatic tumor detection in gadolinium- and ferucarbotran-enhanced magnetic resonance imaging: does combining ferucarbotran-enhanced T2*-weighted gradient echo and T2-weighted turbo spin echo images have additive efficacy?

Kim YK, Lee YH, Kwak HS, Kim CS, Han YM - Korean J Radiol (2008 Nov-Dec)

Bottom Line: The diagnostic accuracy and sensitivity for each image set and the combined set were evaluated using the alternative-free response receiver operating characteristic method.The mean area under the curve value of the combined set (0.966) tended to be better than that for each individual image set (T2W-TSE [0.910], T2*W-GRE [0.892]).Sensitivities in the combined set were higher than those in each individual image set for detecting HCC (mean, 93.0% versus 81.6% and 86.7%, respectively, p < 0.01).

View Article: PubMed Central - PubMed

Affiliation: Department of Diagnostic Radiology, Chonbuk National University Hospital and Medical School, Chonbuk, Korea. jmyr@dreamwiz.com

ABSTRACT

Objective: To determine if a combination of ferucarbotran-enhanced T2*weighted-gradient echo (T2*W-GRE) and T2-weighted turbo spin echo (T2W-TSE) images in gadolinium- and ferucarbotran-enhanced MRI has additive efficacy compared to each image alone for detecting small (< or = 2.0 cm) hepatocellular carcinoma (HCC) lesions in a group of cirrhotic patients and metastases in a group of non-cirrhotic patients.

Materials and methods: Two readers retrospectively analyzed gadolinium- and ferucarbotran-enhanced T2*W-GRE, T2W-TSE, and combined T2*W-GRE/T2W-TSE images of 119 patients with 157 HCCs and 32 patients with 98 metastases. The diagnostic accuracy and sensitivity for each image set and the combined set were evaluated using the alternative-free response receiver operating characteristic method.

Results: The mean area under the curve value of the combined set (0.966) tended to be better than that for each individual image set (T2W-TSE [0.910], T2*W-GRE [0.892]). Sensitivities in the combined set were higher than those in each individual image set for detecting HCC (mean, 93.0% versus 81.6% and 86.7%, respectively, p < 0.01). Sensitivities in the combined set and the T2W-TSE set were the same for detecting metastases, and both were higher than the sensitivity seen in the T2*W-GRE set (mean, 97.5% versus 85.2 %, p < 0.01).

Conclusion: Combining ferucarbotran-enhanced T2*W-GRE and T2W-TSE has additive efficacy for detecting HCC in cirrhotic patients, but T2W-TSE is preferred for detecting metastases in non-cirrhotic patients.

Show MeSH

Related in: MedlinePlus

50-year-old man with two hepatocellular carcinomas in right hepatic lobe.A. Axial arterial phase three-dimensional dynamic MRI after administration of gadopentetate dimeglumine (4.3/2.0) shows subtle small nodular enhancement (arrow) in subcapsular area of right hepatic lobe. Inferior margin of main mass located at slightly higher level is shown (small arrows).B. Axial ferucarbotran-enhanced breath-hold T2*-weighted gradient echo imaging (180/12) shows only main mass (arrows) in posterior portion of right hepatic lobe. Subcapsular lesion was assigned score of 1 by both observers using T2*-weighted gradient echo image set.C. Axial ferucarbotran-enhanced respiratory-triggered T2-weighted turbo spin echo imaging (4200/76) shows definitive round hyperintense subcapsular mass (arrow) and main mass (small arrows), which were assigned confidence level of 4 by both observers.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2627240&req=5

Figure 2: 50-year-old man with two hepatocellular carcinomas in right hepatic lobe.A. Axial arterial phase three-dimensional dynamic MRI after administration of gadopentetate dimeglumine (4.3/2.0) shows subtle small nodular enhancement (arrow) in subcapsular area of right hepatic lobe. Inferior margin of main mass located at slightly higher level is shown (small arrows).B. Axial ferucarbotran-enhanced breath-hold T2*-weighted gradient echo imaging (180/12) shows only main mass (arrows) in posterior portion of right hepatic lobe. Subcapsular lesion was assigned score of 1 by both observers using T2*-weighted gradient echo image set.C. Axial ferucarbotran-enhanced respiratory-triggered T2-weighted turbo spin echo imaging (4200/76) shows definitive round hyperintense subcapsular mass (arrow) and main mass (small arrows), which were assigned confidence level of 4 by both observers.

Mentions: For detection of HCCs, the Az value of the combined set was significantly higher than that of each set alone, for both observers (p = 0.016 [observer 1], 0.013 [observer 2] for the T2*W-GRE set; p = 0.042 for the T2W-TSE set) (Table 2). Although the Az values of the T2*W-GRE set were slightly better than those of the T2W-TSE set for both observers (Fig. 1), no significant difference was noted for either observer (p = 0.635 for observer 1; p = 0.801 for observer 2). Among the 157 HCCs, the T2W-TSE set allowed for the depiction of 127 lesions (sensitivity, 80.9%; 95% CI: 73.9%, 86.7%) by observer 1 and 129 lesions (sensitivity, 82.2%; 95% CI: 75.35%, 87.8%) by observer 2; the T2*W-GRE set allowed the depiction of 135 lesions (sensitivity, 86.0%; 95% CI: 79.6%, 91.0%) by observer 1 and 137 lesions (sensitivity, 87.3%; 95% CI: 81.0%, 92.0%) by observer 2. There was a significant difference in sensitivities between the two image sets for both observers (p = 0.01). The combined approach allowed for the depiction of 145 lesions (sensitivity, 92.4%; 95% CI: 87.05%, 96.0%) by observer 1 and 147 lesions (sensitivity, 93.6%; 95% CI: 88.6%, 96.9%) by observer 2 (mean, 93.0%), which was significantly better than the values obtained using each set alone (mean, 81.6% for T2W-TSE set, 86.7% for T2*W-GRE set) (p = 0.0001 for T2W-TSE set, p = 0.002 for T2*W-GRE set). There were nine and eight HCCs (size range: 0.6-1.2 cm, mean size: 0.9 cm) for each observer, respectively, which were not detected on the T2*W-GRE set, but which were detected on the T2W-TSE set (Fig. 2). There were 17 HCCs (size range: 0.6-1.9 cm, mean size: 1.2 cm) for both observers that were not detected on the T2W-TSE set, but which were detected on the T2*W-GRE set. Of the 51 HCCs with surgical confirmation, four well-differentiated HCCs were not detected by either observer on the T2W-TSE set, but were detected on the T2*W-GRE set by both observers. For the remaining lesions, the two image sets showed similar results.


Small malignant hepatic tumor detection in gadolinium- and ferucarbotran-enhanced magnetic resonance imaging: does combining ferucarbotran-enhanced T2*-weighted gradient echo and T2-weighted turbo spin echo images have additive efficacy?

Kim YK, Lee YH, Kwak HS, Kim CS, Han YM - Korean J Radiol (2008 Nov-Dec)

50-year-old man with two hepatocellular carcinomas in right hepatic lobe.A. Axial arterial phase three-dimensional dynamic MRI after administration of gadopentetate dimeglumine (4.3/2.0) shows subtle small nodular enhancement (arrow) in subcapsular area of right hepatic lobe. Inferior margin of main mass located at slightly higher level is shown (small arrows).B. Axial ferucarbotran-enhanced breath-hold T2*-weighted gradient echo imaging (180/12) shows only main mass (arrows) in posterior portion of right hepatic lobe. Subcapsular lesion was assigned score of 1 by both observers using T2*-weighted gradient echo image set.C. Axial ferucarbotran-enhanced respiratory-triggered T2-weighted turbo spin echo imaging (4200/76) shows definitive round hyperintense subcapsular mass (arrow) and main mass (small arrows), which were assigned confidence level of 4 by both observers.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2627240&req=5

Figure 2: 50-year-old man with two hepatocellular carcinomas in right hepatic lobe.A. Axial arterial phase three-dimensional dynamic MRI after administration of gadopentetate dimeglumine (4.3/2.0) shows subtle small nodular enhancement (arrow) in subcapsular area of right hepatic lobe. Inferior margin of main mass located at slightly higher level is shown (small arrows).B. Axial ferucarbotran-enhanced breath-hold T2*-weighted gradient echo imaging (180/12) shows only main mass (arrows) in posterior portion of right hepatic lobe. Subcapsular lesion was assigned score of 1 by both observers using T2*-weighted gradient echo image set.C. Axial ferucarbotran-enhanced respiratory-triggered T2-weighted turbo spin echo imaging (4200/76) shows definitive round hyperintense subcapsular mass (arrow) and main mass (small arrows), which were assigned confidence level of 4 by both observers.
Mentions: For detection of HCCs, the Az value of the combined set was significantly higher than that of each set alone, for both observers (p = 0.016 [observer 1], 0.013 [observer 2] for the T2*W-GRE set; p = 0.042 for the T2W-TSE set) (Table 2). Although the Az values of the T2*W-GRE set were slightly better than those of the T2W-TSE set for both observers (Fig. 1), no significant difference was noted for either observer (p = 0.635 for observer 1; p = 0.801 for observer 2). Among the 157 HCCs, the T2W-TSE set allowed for the depiction of 127 lesions (sensitivity, 80.9%; 95% CI: 73.9%, 86.7%) by observer 1 and 129 lesions (sensitivity, 82.2%; 95% CI: 75.35%, 87.8%) by observer 2; the T2*W-GRE set allowed the depiction of 135 lesions (sensitivity, 86.0%; 95% CI: 79.6%, 91.0%) by observer 1 and 137 lesions (sensitivity, 87.3%; 95% CI: 81.0%, 92.0%) by observer 2. There was a significant difference in sensitivities between the two image sets for both observers (p = 0.01). The combined approach allowed for the depiction of 145 lesions (sensitivity, 92.4%; 95% CI: 87.05%, 96.0%) by observer 1 and 147 lesions (sensitivity, 93.6%; 95% CI: 88.6%, 96.9%) by observer 2 (mean, 93.0%), which was significantly better than the values obtained using each set alone (mean, 81.6% for T2W-TSE set, 86.7% for T2*W-GRE set) (p = 0.0001 for T2W-TSE set, p = 0.002 for T2*W-GRE set). There were nine and eight HCCs (size range: 0.6-1.2 cm, mean size: 0.9 cm) for each observer, respectively, which were not detected on the T2*W-GRE set, but which were detected on the T2W-TSE set (Fig. 2). There were 17 HCCs (size range: 0.6-1.9 cm, mean size: 1.2 cm) for both observers that were not detected on the T2W-TSE set, but which were detected on the T2*W-GRE set. Of the 51 HCCs with surgical confirmation, four well-differentiated HCCs were not detected by either observer on the T2W-TSE set, but were detected on the T2*W-GRE set by both observers. For the remaining lesions, the two image sets showed similar results.

Bottom Line: The diagnostic accuracy and sensitivity for each image set and the combined set were evaluated using the alternative-free response receiver operating characteristic method.The mean area under the curve value of the combined set (0.966) tended to be better than that for each individual image set (T2W-TSE [0.910], T2*W-GRE [0.892]).Sensitivities in the combined set were higher than those in each individual image set for detecting HCC (mean, 93.0% versus 81.6% and 86.7%, respectively, p < 0.01).

View Article: PubMed Central - PubMed

Affiliation: Department of Diagnostic Radiology, Chonbuk National University Hospital and Medical School, Chonbuk, Korea. jmyr@dreamwiz.com

ABSTRACT

Objective: To determine if a combination of ferucarbotran-enhanced T2*weighted-gradient echo (T2*W-GRE) and T2-weighted turbo spin echo (T2W-TSE) images in gadolinium- and ferucarbotran-enhanced MRI has additive efficacy compared to each image alone for detecting small (< or = 2.0 cm) hepatocellular carcinoma (HCC) lesions in a group of cirrhotic patients and metastases in a group of non-cirrhotic patients.

Materials and methods: Two readers retrospectively analyzed gadolinium- and ferucarbotran-enhanced T2*W-GRE, T2W-TSE, and combined T2*W-GRE/T2W-TSE images of 119 patients with 157 HCCs and 32 patients with 98 metastases. The diagnostic accuracy and sensitivity for each image set and the combined set were evaluated using the alternative-free response receiver operating characteristic method.

Results: The mean area under the curve value of the combined set (0.966) tended to be better than that for each individual image set (T2W-TSE [0.910], T2*W-GRE [0.892]). Sensitivities in the combined set were higher than those in each individual image set for detecting HCC (mean, 93.0% versus 81.6% and 86.7%, respectively, p < 0.01). Sensitivities in the combined set and the T2W-TSE set were the same for detecting metastases, and both were higher than the sensitivity seen in the T2*W-GRE set (mean, 97.5% versus 85.2 %, p < 0.01).

Conclusion: Combining ferucarbotran-enhanced T2*W-GRE and T2W-TSE has additive efficacy for detecting HCC in cirrhotic patients, but T2W-TSE is preferred for detecting metastases in non-cirrhotic patients.

Show MeSH
Related in: MedlinePlus