Contrast-enhanced MR Imaging of Postoperative Scars and VX2 ...

Radiology

Joon Woo Lee, MD Woo Kyung Moon, MD Hanns-Joachim Weinmann, PhD Soo Jeoung Kim, PhD Jong Hyo Kim, PhD Seong Ho Park, MD Tae Jung Kim, MD Chang Jin Yoon, MD Young Hoon Kim, MD Eun Yoon Cho, MD Sung Whan Ha, MD Wee-Saing Kang, MD Kee Hyun Chang, MD Index terms: Animals Contrast media, comparative studies Experimental study Gadolinium Magnetic resonance (MR), contrast enhancement Magnetic resonance (MR), contrast media Neoplasms, MR Published online 10.1148/radiol.2291020218 Radiology 2003; 229:132–139 1 From the Department of Radiology and Clinical Research Institute, Seoul National University Hospital and the Institute of Radiation Medicine, Seoul National University Medical Research Center, 28 Yongon-dong, Chongno-gu, Seoul 110-744, Korea (J.W.L., W.K.M., S.J.K., J.H.K., S.H.P., T.J.K., C.J.Y., Y.H.K., K.H.C.); Department of Contrast Media Research, Schering, Berlin, Germany (H.J.W.); Department of Pathology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Korea (E.Y.C.); and Department of Therapeutic Radiology, Seoul National University Hospital, Korea (S.W.H., W.S.K.). From the 1999 RSNA scientific assembly. Received March 13, 2002; revision requested May 22; final revision received January 12, 2003; accepted February 24. Supported by a grant from Schering, Berlin, Germany. Address correspondence to W.K.M. (e-mail: [email protected]).

Author contributions: Guarantor of integrity of entire study, W.K.M.; study concepts and design, all authors; literature research, J.W.L., W.K.M.; experimental studies, J.W.L., S.J.K., T.J.K., C.J.Y., Y.H.K., E.Y.C.; data acquisition, J.W.L., W.K.M., S.J.K., J.H.K.; data analysis/interpretation, all authors; statistical analysis, J.W.L., J.H.K., S.H.P.; manuscript preparation and definition of intellectual content, all authors; manuscript editing, J.W.L., W.K.M., S.H.P.; manuscript revision/review and final version approval, all authors ©

RSNA, 2003

132

Contrast-enhanced MR Imaging of Postoperative Scars and VX2 Carcinoma in Rabbits: Comparison of Macromolecular Contrast Agent and Gadopentetate Dimeglumine1 PURPOSE: To compare the magnetic resonance (MR) imaging enhancement patterns of a blood pool contrast agent, SH L 643A, with those of gadopentetate dimeglumine in postoperative scars and VX2 carcinomas in rabbits and to compare these enhancement patterns with microvessel density in pathologic specimens. MATERIALS AND METHODS: Eighteen rabbits with experimentally induced postoperative scars (n ⫽ 12) or VX2 carcinoma (n ⫽ 6) in the thighs underwent sequential MR imaging first with gadopentetate dimeglumine and then, 24 hours later, with SH L 643A. The enhancement ratios (ie, the ratios of postcontrast to precontrast signal intensity) and the microvessel densities of postoperative scars and VX2 carcinomas were assessed. Differences were tested for by using the MannWhitney U and Wilcoxon signed rank tests. RESULTS: In postoperative scars, enhancement ratios were consistently lower with injection of SH L 643A than with injection of gadopentetate dimeglumine for up to 30 minutes (P ⬍ .05). In postoperative scars, mean peak enhancement ratios were 1.29 ⫾ 0.15 (SD) with injection of SH L 643A and 1.61 ⫾ 0.31 with injection of gadopentetate dimeglumine (P ⬍ .01). In VX2 carcinomas, the enhancement ratios were not significantly different with injection of SH L 643A than with injection of gadopentetate dimeglumine at all time points. The mean difference between the enhancement ratios of the VX2 carcinomas and postoperative scars was 0.64 ⫾ 0.10 (range, 0.50 – 0.77) with SH L 643A and 0.36 ⫾ 0.16 (range, 0.17– 0.66) with gadopentetate dimeglumine (P ⬍ .01). The mean microvessel density (in terms of vessels per field of view) was 10.7 ⫾ 5.5 for postoperative scars and 30.0 ⫾ 7.7 for VX2 carcinoma (P ⬍ .001). CONCLUSION: The difference between the enhancement ratios of postoperative scars and VX2 carcinomas with SH L 643A was greater than that with gadopentetate dimeglumine. Enhancement ratios at SH L 643A– enhanced MR imaging corresponded well with microvessel density in postoperative scars and VX2 carcinomas. ©

RSNA, 2003

The differentiation between posttherapeutic changes and residual or recurrent tumor is of major importance in the follow-up of patients with most treated cancers. Computed tomography and ultrasonography (US) are widely accepted imaging modalities for evaluation of tumor recurrence, but both techniques are based on lesion morphology and are thus limited in that they do not demonstrate the difference between fibrotic and tumorous tissues. The signal intensity on T2-weighted magnetic resonance (MR) images can also be

Radiology

used for distinguishing tumor from fibrosis (1–3). However, results of more recent studies have suggested that high signal intensity on T2-weighted MR images is not specific for recurrent tumor and that low signal intensity can be found in both fibrosis and tumor (4 – 8). MR imaging with a conventional extracellular agent, gadopentetate dimeglumine, has been evaluated by several researchers and has shown promise in the distinction of late (⬎6 months) fibrosis from malignancy (4,8 –10). Late fibrosis showed no substantial enhancement on T1-weighted MR images obtained after gadopentetate dimeglumine administration, whereas recurrent tumors demonstrated early enhancement. However, early fibrosis (⬍6 months) occasionally showed enhancement owing to leaky junctional complexes and intercellular gaps in the endothelium (4,8,11,12). Low-molecularweight (546 Da) gadopentetate dimeglumine not only enhances the vascular space but also diffuses rapidly into the interstitial space (13). Blood pool contrast agents that remain exclusively within the intravascular space are currently being investigated to see if their use can minimize the problems associated with extracellular contrast agents (14). A macromolecular contrast agent, 24gadolinium-tetraazacyclododecane tetraacetic acid dendrimer (SH L 643A, Gadomer-17; Schering, Berlin, Germany), has an apparent molecular weight of 35,000 Da—small enough to guarantee renal excretion and large enough to reduce diffusion through the endothelial cells of intact blood vessels (15,16). Potential clinical applications for such macromolecular contrast agents include MR angiography and determination of tissue perfusion, angiogenesis, and capillary integrity (17–24). Comparative studies of blood pool and extracellular contrast agents in animal models revealed that blood pool contrast agents provide improved lesion conspicuity and tumor characterization (25,26). The hyperpermeability of tumor microvessels to macromolecular contrast agents has been demonstrated in malignant tumors (27, 28). To our knowledge, a comparison study of blood pool and extracellular contrast agents for distinguishing postoperative scars from malignant tumors has not previously been performed. We performed this experimental study to compare the enhancement patterns of the blood pool contrast agent SH L 643A with those of the conventional extracellular agent gadopentetate dimeglumine in postoperative scars and VX2 carcinomas in rabbits and to compare the MR Volume 229

䡠

Number 1

imaging enhancement patterns with microvessel density in pathologic specimens.

MATERIALS AND METHODS Animals and Experimental Model The experiments were performed with 18 New Zealand white rabbits that ranged in weight from 2 to 3 kg. The animals were sedated with an intramuscular injection of 50 mg of ketamine hydrochloride (Ketalar; Yuhan Yanghang, Seoul, Korea) and 20 mg of xylazine hydrochloride (Rompun; Bayer Korea, Seoul, Korea). This experiment was approved by the animal care committee of Seoul National University Medical Research Center. Postoperative scars were made in the right thigh of 12 rabbits. After making a vertical incision in the skin of the lateral side of the thigh with aseptic technique, we (J.W.L., T.J.K., C.J.Y.) exposed the thigh muscles. We incised and dissected the muscles and then folded and packed a 4-inch square of gauze between the muscle bellies. Layer by layer closure from the muscle to the skin was then performed. Two weeks later, the gauze was surgically removed in aseptic conditions. To prevent wound infection, 500 mg of cefazolin sodium (Cefazoline; Chongkeun Dang Pharmacy, Seoul, Korea) and 10 mg of gentamicin sulfate (Gentamicin; Korea United Pharmacy, Seoul, Korea) were injected into the contralateral buttock daily for 4 weeks after initial gauze packing. Each of 12 rabbits was placed into one of three groups according to the following intervals between scar induction and sacrifice: 1 month (n ⫽ 4), 2 months (n ⫽ 4), and 3 months (n ⫽ 4). In each group, MR imaging examinations were performed at 1-month intervals after scar induction. However, an MR imaging examination was not performed 1 month after scar induction in two of four rabbits in the 3-month group because of technical problems with the MR imaging system; therefore, a total of 22 MR imaging examinations were performed in the 12 rabbits. In six rabbits, VX2 carcinoma was inoculated in the right thigh by injecting 0.5 mL of tumor suspension with an 18gauge needle. The experimental VX2 rabbit carcinoma was prepared in the manner previously described by Moon et al (15). An animal was included in this study if a mass larger than 2 cm in diameter was noted at US. MR imaging was performed 3 weeks after inoculation.

MR Imaging All examinations were performed with a 1.5-T MR imaging system (Magnetom Vison Plus; Siemens, Erlangen, Germany) and a knee coil. All images were acquired in the transverse plane. Animals were imaged in the prone position. After routine localization images and transverse T2weighted spin-echo images (repetition time msec/echo time msec, 4,000/96; 4-mm section thickness) were acquired, two sets of sequential T1-weighted images were acquired. The first set of sequential T1-weighted images was acquired after 0.10 mmol per kilogram of body weight of gadopentetate dimeglumine (Magnevist; Schering) was administered to each animal via the ear vein by means of manual fast bolus injection; then, after a 24-hour interval, the second set of sequential T1-weighted images was acquired after administration of 0.05 mmol/kg of SH L 643A in the same fashion. SH L 643A was supplied in an aqueous formulation with a concentration of 500 mmol/L. In all animals, MR imaging was performed first with gadopentetate dimeglumine and then (24 hours later) with SH L 643A. Sequential MR imaging was performed with a fast spin-echo sequence (450/16) before and 1, 2, 3, 4, 5, 10, 15, 20, 25, and 30 minutes after bolus injection of the contrast agent. The echo train length was 10, and the bandwidth was 16 kHz. The section thickness was 4 mm, and the field of view was 15 cm with an acquisition matrix of 256 ⫻ 128.

Image Analysis The maximal diameters and signal intensities of postoperative scars and VX2 carcinomas at MR imaging were measured manually by one author (J.W.L.). MR imaging signal intensities were measured by using a 2– 6-mm-diameter (mean diameter, 4 mm) circular region of interest placed over the most highly enhancing areas of the lesions. For each time point, the enhancement ratio determined by comparison of the signal intensity on the unenhanced images with that on the contrast material– enhanced images was calculated by using the following equation: ER ⫽ postcontrast SI/precontrast SI, where ER is the enhancement ratio and SI is signal intensity. The enhancement ratio curve was obtained by plotting the enhancement ratios against time in each case. In each case, the peak enhancement ratio (ERpeak) was determined as the value at a time point (peak time, Tmax) beyond which the sum of

Contrast-enhanced MR Imaging of Scars and Carcinoma

䡠

133

Radiology

time point was calculated for each contrast agent. Lesion enhancement ratio curves were classified according to their shape as having a steady pattern, a plateau pattern, or a washout pattern (31,32) by two radiologists (C.J.Y., Y.H.K.) in consensus who were blinded to the contrast agent used. The steady pattern of enhancement corresponded to enhancement curves with continuing enhancement during the dynamic study after the initial upstroke. The plateau pattern of enhancement corresponded to enhancement curves with a sharp bend after the initial upstroke that meant there was no further increase or decrease in enhancement. The washout pattern of enhancement corresponded to enhancement curves that showed a downward curve after the initial upstroke. The parameters representing the characteristics of enhancement such as peak enhancement ratios, peak time, slope, and the shape of the enhancement ratio curves were compared according to contrast agents for postoperative scars and VX2 carcinomas. In addition, the same enhancement parameters were calculated for each scar maturation interval, and differences between the two contrast agents were compared according to maturation interval. The difference in peak enhancement ratios observed in each pair of sequential MR images (ie, the ERpeak for the image obtained with gadopentetate dimeglumine minus the ERpeak for the image obtained with SH L 643A) was also calculated and compared according to scar maturation interval.

Histologic Analysis

Figure 1. Transverse contrast-enhanced fast spin-echo MR images (450/16) of a postoperative scar that was experimentally induced 1 month previously in the right thigh of a rabbit. Images were obtained (a) before and (b) 1, (c) 5, and (d) 30 minutes after injection of SH L 643A and (e) before and (f) 1, (g) 5, and (h) 30 minutes after injection of gadopentetate dimeglumine. (All images were obtained in the same rabbit; a– d were obtained with the rabbit in one position, and e– h were obtained with the rabbit in another position.) The postoperative scars (arrows) are more highly enhanced after injection of gadopentetate dimeglumine than after injection of SH L 643A.

slopes measured for the two intervals between three consecutive time points on each curve was 10% per minute or less. The slope of the curve defined as the percentage increase in enhancement ratio per minute over the baseline value was derived from the following equation 134

䡠

Radiology

䡠

October 2003

(29,30): slope ⫽ [(ERpeak ⫺ ERpre) ⫻ 100]/ (ERpre ⫻ Tmax), where ERpre represents the enhancement ratio of a given region of interest before the injection of a contrast agent. The difference between the mean enhancement ratios of postoperative scars and VX2 carcinomas at each

For analysis of histologic features of the postoperative scars and VX2 carcinomas, animals were sacrificed by injecting a lethal dose (90 mg/kg) of sodium pentobarbital (Pentothal; Choong Wae Pharmacy, Seoul, Korea). The pathologic specimens were sectioned in the same transverse plane used at MR imaging with 7-mm intervals and stained with hematoxylin-eosin for light microscopic evaluation. The microvessel density of postoperative scars and VX2 carcinomas was determined by averaging the counted numbers of capillary vessels in five fields of view at a magnification factor of 200 in the areas corresponding to areas of enhancement on MR images. In postoperative scars, degree of collagen formation, cellularity, extent of muscle involvement, vascularity, and inflammation were analyzed and graded on a scale from 0 to 3 (where 0 indicated “absent”; Lee et al

Radiology

1, “minimal”; 2, “moderate”; and 3, “severe”) by comparing the slides to one another. A pathologist (E.Y.C.) who was blinded to the MR imaging information reviewed all slides.

Statistical Analysis The SPSS software package version 9.0 (SPSS, Chicago, Ill) was used for statistical data analysis. The Mann-Whitney U test was applied to assess the statistical significance of the differences in peak enhancement ratio, enhancement ratios at each time point, and slope values between the two contrast agents. The Wilcoxon signed rank test was used to evaluate the statistical significance of the difference between the mean enhancement ratio of postoperative scars and that of VX2 carcinomas for the two contrast agents. The Fisher exact test was used to assess the differences in curve shapes between the contrast agents. The MannWhitney U test was also used to evaluate the statistical significance of the differences in enhancement ratios at each time point, the differences in peak enhancement ratios among postoperative scar maturation intervals, and the difference in microvessel density between postoperative scars and VX2 carcinomas. A P value of less than .05 was considered to indicate a statistically significant difference.

RESULTS The maximal diameter of the postoperative scars (12 rabbits) and the VX2 carcinomas (six rabbits) ranged from 2.3 to 3.5 cm (mean, 2.8 cm ⫾ 0.4 [SD]) and from 2.0 to 3.1 cm (mean, 2.4 cm ⫾ 0.4), respectively. After injection of SH L 643A and gadopentetate dimeglumine, strong rim enhancement was seen in all VX2 carcinomas, whereas mild to moderate enhancement was seen in postoperative scars (Figs 1, 2). The peak enhancement ratios of postoperative scars were significantly lower with injection of SH L 643A (mean ⫾ SD, 1.29 ⫾ 0.15) than with injection of gadopentetate dimeglumine (1.61 ⫾ 0.31) (P ⬍ .01) (Table 1), whereas the peak enhancement ratios of VX2 carcinoma were not significantly lower with injection of SH L 643A (2.02 ⫾ 0.76) than with injection of gadopentetate dimeglumine (2.09 ⫾ 0.32) (P ⫽ .240). With SH L 643A, the enhancement ratios of postoperative scars and VX2 carcinomas peaked at 14.30 minutes ⫾ 8.36 (range, 3.00 –30.00 minutes) and 8.80 minutes ⫾ 6.90 (range, Volume 229

䡠

Number 1

Figure 2. Transverse contrast-enhanced fast spin-echo MR images (450/16) of a VX2 carcinoma that was experimentally induced 3 weeks previously in the right thigh of a rabbit. Images were obtained (a) before and (b) 1, (c) 5, and (d) 30 minutes after injection of SH L 643A and (e) before and (f) 1, (g) 5, and (h) 30 minutes after injection of gadopentetate dimeglumine. (All images were obtained in the same rabbit; a– d were obtained with the rabbit in one position, and e– h were obtained with the rabbit in another position.) The peripheral portion of the VX2 carcinoma (arrows) is well enhanced with both contrast agents. The degree of enhancement in VX2 carcinoma is similar between the two contrast agents. Compared with images obtained after injection of SH L 643A, images of VX2 carcinomas obtained after injection of gadopentetate dimeglumine show early enhancement and washout in the later images.

4.00 –20.00 minutes), respectively, while with gadopentetate dimeglumine, they peaked at 9.77 minutes ⫾ 6.19 (range, 1.00 –30.00 minutes) and 3.70 minutes ⫾

3.40 (range, 1.00 –10.00 minutes). The mean slope values for postoperative scars and VX2 carcinomas, respectively, were 2.99% per minute ⫾ 2.72 (range, 0.79%–


䡠

135

TABLE 1 Enhancement Parameters in Postoperative Scars and VX2 Carcinomas

Radiology

Peak Enhancement Ratio* Group Postoperative scars (n ⫽ 22) VX2 carcinomas (n ⫽ 6)

Peak Time (min)*

Slope (%/min)*

SH L 643A

Gd-DTPA

P Value

SH L 643A

Gd-DTPA

P Value

SH L 643A

Gd-DTPA

P Value

1.29 ⫾ 0.15

1.61 ⫾ 0.31

⬍.01

14.30 ⫾ 8.36

9.77 ⫾ 6.19

⬍.01

2.99 ⫾ 2.72

8.99 ⫾ 7.32

⬍.01

2.02 ⫾ 0.76

2.09 ⫾ 0.32

8.80 ⫾ 6.90

3.70 ⫾ 3.40

19.00 ⫾ 17.70

57.60 ⫾ 48.60

.240

.065

.093

* Data are means ⫾ SDs. Gd-DTPA ⫽ gadopentetate dimeglumine.

Figure 3. Graph shows mean enhancement ratio after injection of 0.10 mmol/kg of gadopentetate dimeglumine and after injection of 0.05 mmol/kg of SH L 643A in postoperative scars (n ⫽ 22) and VX2 carcinomas (n ⫽ 6). Although they show a plateau pattern of enhancement after SH L 643A administration, postoperative scars show a remarkable enhancement and washout pattern after gadopentetate dimeglumine administration. In VX2 carcinomas, a washout pattern is seen after both contrast agent injections. In postoperative scars, the enhancement ratios are significantly lower at all time points with SH L 643A than with gadopentetate dimeglumine (P ⬍ .05). F ⫽ enhancement ratio in VX2 carcinomas after SH L 643A administration, E ⫽ enhancement ratio in VX2 carcinomas after gadopentetate dimeglumine administration, Œ ⫽ enhancement ratio in postoperative scars after SH L 643A administration, ‚ ⫽ enhancement ratio in postoperative scars after gadopentetate dimeglumine administration. Error bars represent SDs.

9.57% per minute) and 19.00% per minute ⫾ 17.70 (range, 3.70%–50.90% per minute) after SH L 643A injection and 8.99% per minute ⫾ 7.32 (range, 0.97%–30.00% per minute) and 57.60% per minute ⫾ 48.60 (range, 11.00%– 110.90% per minute) after gadopentetate dimeglumine injection. The difference in slope between the two contrast agents was statistically significant in postoperative scars (P ⬍ .01). Peak enhancement ratios and peak time and slope values for postoperative scars and VX2 carcinomas are summarized in Table 1. The enhancement ratios of postopera136

䡠

Radiology

䡠

October 2003

tive scars were notably different compared with those of VX2 carcinomas after both contrast agent injections (Fig 3). In postoperative scars, the enhancement ratios with injection of SH L 643A were significantly lower than those with injection of gadopentetate dimeglumine up to the 30-minute time point (P ⬍ .05). In VX2 carcinomas, the enhancement ratios with injection of gadopentetate dimeglumine were higher within 4 minutes and were lower at 5–30 minutes than those with injection of gadopentetate dimeglumine, but this difference was not significant at all time points (P ⬎ .1) (Fig 3). The

difference between the mean enhancement ratios of postoperative scars and those of VX2 carcinomas was greater with SH L 643A than with gadopentetate dimeglumine. With SH L 643A, the differences ranged from 0.50 to 0.77 (mean, 0.64 ⫾ 0.10); with gadopentetate dimeglumine, they ranged from 0.17 to 0.66 (mean, 0.36 ⫾ 0.16) (P ⬍ .01). In postoperative scars only, the shapes of the curves differed significantly between the two contrast agents (P ⬍ .01). With SH L 643A, a plateau pattern of enhancement ratio curves was observed in 20 (91%) of the 22 postoperative scars. A steady pattern of the enhancement curve was observed in one (5%) of the 22 scars, and a washout pattern of the enhancement curve was observed in only one (5%) of the 22 scars. With gadopentetate dimeglumine, a washout pattern of the enhancement curve was observed in 11 (50%) of the 22 scars, and a plateau pattern of the enhancement curve was observed in 11 (50%) of the 22 scars. Among VX2 carcinomas, the shape of the enhancement ratio curve was the washout pattern in five and the plateau pattern in one with SH L 643A, whereas all carcinomas showed a washout pattern after gadopentetate dimeglumine injection. The shapes of the curves in VX2 carcinomas did not differ significantly between the two contrast agents (P ⬎ .05). When the enhancement ratios were compared among the postoperative scars at the 1-month (n ⫽ 10), 2-month (n ⫽ 8), and 3-month (n ⫽ 4) intervals, they were consistently higher in the 1-month postoperative scars than in the 2-month or 3-month postoperative scars after gadopentetate dimeglumine administration. The difference was statistically significant at all time points (P ⬍ .05) except for 4, 25, and 30 minutes after gadopentetate dimeglumine administration. However, with SH L 643A, the enhancement ratios were not significantly higher in 1-month postoperative scars than in Lee et al

Radiology

2-month or 3-month postoperative scars (P ⬎ .1) (Fig 4). The difference in peak enhancement ratios in each pair of sequential MR images was higher in 1-month postoperative scars (mean ⫾ SD, 0.37 ⫾ 0.21) than in 2-month (0.35 ⫾ 0.46, P ⫽ .11) or 3-month (0.18 ⫾ 0.08, P ⬍ .05) postoperative scars. The shapes of the curves after the injection of each contrast agent differed between 1-month and 2- or 3-month postoperative scars. Seven (70%) of 10 1-month postoperative scars showed a washout pattern with gadopentetate dimeglumine, whereas only one (10%) showed a washout pattern with SH L 643A. Among 2-month and 3-month postoperative scars considered together, four scars (33%) showed a washout pattern with gadopentetate dimeglumine, whereas no scar showed a washout pattern with SH L 643A. Other enhancement characteristics of postoperative scars according to maturation interval are summarized in Table 2. In all cases, pathologic findings in the experimentally induced scars were consistent with fibrosis at light microscopy. All VX2 carcinomas had necrotic areas in the center and viable tumor cells in the periphery. In postoperative scars, there were few vessels scattered in the collagen deposits, whereas the periphery of the VX2 carcinomas and the border between skeletal muscle and neoplastic tissue were rich with new vessel formation (Figs 5, 6). The mean microvessel density (in terms of vessels per field of view) was 10.7 ⫾ 5.5 (magnification, ⫻200) for postoperative scars and 30.0 ⫾ 7.7 (magnification, ⫻200) for VX2 carcinomas. The difference was statistically significant (P ⬍ .001). Inflammatory cells were more frequently seen in 1-month and 2-month postoperative scars than in 3-month postoperative scars. There were no differences in the grade of collagen deposition, cellularity, muscle involvement, or vessel count among 1-month, 2-month, and 3-month postoperative scars.

DISCUSSION The results of our experimental study indicate that the blood pool contrast agent SH L 643A, which is confined to the vascular space, can reveal the blood volume of lesions more accurately than can gadopentetate dimeglumine. In postoperative scars and in VX2 carcinomas, when the macromolecular agent was used, peak enhancement ratios (which are related to blood volume) correlated better with miVolume 229

䡠

Number 1

Figure 4. Graph shows mean enhancement ratios of postoperative scars after injection of 0.10 mmol/kg of gadopentetate dimeglumine and after injection of 0.05 mmol/kg of SH L 643A for each scar maturation interval. Enhancement ratios of 1-month postoperative scars are significantly higher at all time points with gadopentetate dimeglumine than with SH L 643A (P ⬍ .05). The enhancement ratios were consistently higher in 1-month (n ⫽ 10) postoperative scars than in 2-month (n ⫽ 8) or 3-month (n ⫽ 4) postoperative scars after gadopentetate dimeglumine administration but were not significantly higher in 1-month postoperative scars than in 2- or 3-month postoperative scars after SH L 643A administration. ⽧ ⫽ enhancement ratio in 1-month postoperative scars with SH L 643A, F ⫽ enhancement ratio in 2-month postoperative scars with SH L 643A, ⫻ with dotted line ⫽ enhancement ratio in 3-month postoperative scars with SH L 643A, 〫 ⫽ enhancement ratio in 1-month postoperative scars with gadopentetate dimeglumine, E ⫽ enhancement ratio in 2-month postoperative scars with gadopentetate dimeglumine, ⫻ with solid line ⫽ enhancement ratio in 3-month postoperative scars with gadopentetate dimeglumine. Error bars represent SDs.

crovessel density, and peak time and slope of enhancement (which are related to capillary permeability) were slower and smaller, respectively. Peak enhancement ratios were probably better determined with SH L 643A than with gadopentetate dimeglumine because with gadopentetate dimeglumine, peak enhancement ratio was the blood volume polluted by the capillary permeability (16,18,21,29,33). In general, there are three requirements for contrast enhancement of any tissue with the conventional extracellular agent gadopentetate dimeglumine: (a) a vascular supply, (b) a route out of the vasculature for the contrast material, and (c) the availability of an interstitial space for sequestration of the contrast material. Postoperative scars have all of these attributes (4,11,12). In our study, the scars were significantly less enhanced with SH L 643A, a high-molecular-weight MR imaging contrast agent, than with gadopentetate dimeglumine. The difference

was most pronounced in earlier-stage scars. The difference in mean enhancement ratios between postoperative scars and VX2 carcinomas was also greater with SH L 643A, which supports the idea that SH L 643A reflects blood volume more accurately than does gadopentetate dimeglumine. In this study, we calculated the parameters representing the characteristics of enhancement such as peak enhancement ratio, peak time, and slope from the enhancement ratio curves. However, the use of a more complicated pharmacokinetic model would have enabled the calculation of more precise quantitative parameters like blood volume and capillary permeability (19,20,23,33). With use of dynamic or sequential MR image sets, two different criteria are clinically used to describe lesion enhancement kinetics (31). The first criterion is the signal intensity behavior of the lesion in the early phase after the administration of contrast material. It is evaluated by means of the steepness of the postcon-


䡠

137

TABLE 2 Enhancement Parameters of Postoperative Scars According to Maturation Interval

Radiology

Peak Enhancement Ratio†

Peak Time (min)†

Slope (%/min)†

Interval (mo)*

SH L 643A

Gd-DTPA

P Value

SH L 643A

Gd-DTPA

P Value

SH L 643A

Gd-DTPA

P Value

1 (n ⫽ 10) 2 (n ⫽ 8) 3 (n ⫽ 4)

1.34 ⫾ 0.17 1.28 ⫾ 0.12 1.17 ⫾ 0.05

1.71 ⫾ 0.23 1.61 ⫾ 0.40 1.35 ⫾ 0.10

⬍.01 ⬍.05 ⬍.05

16.10 ⫾ 8.88 12.25 ⫾ 8.21 13.75 ⫾ 8.54

8.30 ⫾ 3.62 8.38 ⫾ 5.53 16.25 ⫾ 9.46

⬍.01 ⬍.05 ⬍.05

3.02 ⫾ 2.68 3.68 ⫾ 3.33 1.55 ⫾ 0.69

10.39 ⫾ 6.13 10.40 ⫾ 9.15 2.67 ⫾ 1.39

⬍.01 ⬍.05 ⬎.05

* Interval between scar induction and MR imaging with both contrast agents. Numbers are numbers of scars. † Data are means ⫾ SDs. Gd-DTPA ⫽ gadopentetate dimeglumine.

Figure 5. Photomicrograph of a 2-month postoperative scar shows prominent fibroblast proliferation and collagen deposition. Few blood vessels are seen among the collagen deposits. (Hematoxylin-eosin stain; original magnification, ⫻200.)

trast signal intensity curve. The second criterion is the signal intensity behavior of the lesion in the intermediate and late postcontrast periods. It is traced to derive diagnostic information on the basis of visual or quantitative evaluation of the shape of the time-versus–signal intensity curve. A strong early enhancement or washout pattern is usually regarded as diagnostic for malignant lesions, whereas a mild or steady enhancement pattern is usually regarded as diagnostic for benign lesions (although there is some overlap) (31,32). In our study, the enhancement curves of VX2 carcinomas were mostly those of the washout pattern with both contrast agents. These results correlate well with those of previous studies (15,31). The enhancement curves of postoperative scars, on the other hand, were predominantly (91%) those of the plateau pattern with SH L 643A but were those of the plateau pattern only half (50%) of the time with gadopentetate dimeglumine. These results suggest that SH L 643A might be more helpful in differentiating fibrosis from recurrent cancer at analysis of lesion enhancement kinetics at contrast-enhanced MR imaging. Previously, improved contrast between postoperative scars and vertebral disks was 138

䡠

Radiology

䡠

October 2003

Figure 6. Photomicrograph of a VX2 carcinoma shows abundant blood vessels (arrows) among sheets of tumor cells. (Hematoxylineosin stain; original magnification, ⫻200.)

reported with use of blood pool contrast agents. Nguyen-minh et al (18) measured the signal intensity changes in recurrent herniated disks and postoperative scars in a dog model after injection of SH L 643A and gadopentetate dimeglumine. Compared with the small-molecular-weight contrast agent gadopentetate dimeglumine, the high-molecular-weight contrast agent SH L 643A yielded better contrast between scar tissues and disk fragments. Nguyen-minh et al (18) concluded that detection of recurrent herniated disks might be improved by using higher-molecular-weight contrast media. However, their study did not involve dynamic MR imaging, and the enhancement pattern and other parameters such as peak enhancement ratios and slope could not be evaluated. Our study had several limitations. This rabbit model may not reproduce the clinical situation perfectly because the postoperative scars and VX2 carcinomas were induced in different animals. Because, to our knowledge, no standard method for inducing scars in the soft tissue of rabbits has been reported in the literature, we tested several approaches such as ethanol injection, thermal ablation with a Bovie device, and muscle excision for inducing postoperative scars and found that muscle excision followed by temporary gauze packing was the most reliable method for inducing fibrosis in rabbit thighs. A small

number of animals were used for the VX2 carcinoma study because sequential MR imaging findings of VX2 carcinoma after injection of SH L 643A and gadopentetate dimeglumine have previously been reported in detail (15). In this study, the dose of injected SH L 643A was 0.05 mmol/kg—that is, half the dose of gadopentetate dimeglumine (0.10 mmol/kg). However, the T1 relaxivity of SH L 643A is 11.9 L/mmol/sec—more than twice that of gadopentetate dimeglumine (4.9 L/mmol/sec) (16). Thus, we concluded there was no bias in this study in terms of the injected dose. In conclusion, enhancement ratios at MR imaging with a blood pool contrast agent corresponded well with microvessel density in postoperative scars and VX2 carcinomas. Use of the blood pool MR imaging contrast agent SH L 643A may help in differentiating early postoperative scars from recurrent tumors. Practical application: The differentiation between posttherapeutic changes and residual or recurrent tumor is of great importance in the follow-up of patients with most treated cancers. Results of our animal experiments showed that there were notable differences in enhancement patterns in postoperative scars, particularly in early stages, when SH L 643A was used versus when gadopentetate dimeglumine was used. The difference between the mean enhancement ratios of the VX2 carcinomas and postoperative scars with SH L 643A was significantly greater than the difference with gadopentetate dimeglumine. In our study, direct comparison of enhancement patterns between postoperative scars and tumors that were closely located in the same animal could not be performed, so a successful translation of our experimental observations to clinical practice requires further studies of many postoperative conditions, including recurrent tumors in a background of postoperative fibrosis. Lee et al

Radiology

References 1. Glazer HS, Lee JK, Levitt RG, et al. Radiation fibrosis: differentiation from recurrent tumor by MR imaging. Radiology 1985; 156:721–726. 2. Ebner F, Kressel HY, Mintz MC, et al. Tumor recurrence versus fibrosis in the female pelvis: differentiation with MR imaging at 1.5 T. Radiology 1988; 166:333– 340. 3. Krestin GP, Steinbrich W, Friedmann G. Recurrent rectal cancer: diagnosis with MR imaging versus CT. Radiology 1988; 168:307–311. 4. Markus J, Morrissey B, deGara C, Tarulli G. MRI of recurrent rectosigmoid carcinoma. Abdom Imaging 1997; 22:338 –342. 5. Muller-Schimpfle M, Brix G, Layer G, et al. Recurrent rectal cancer: diagnosis with dynamic MR imaging. Radiology 1993; 189:881– 889. 6. de Lange EE, Fechner RE, Wanebo HJ. Suspected recurrent rectosigmoid carcinoma after abdominoperineal resection: MR imaging and histopathologic findings. Radiology 1989; 170:323–328. 7. Ito K, Kato T, Tadokoro M, et al. Recurrent rectal cancer and scar: differentiation with PET and MR imaging. Radiology 1992; 182:549 –552. 8. Rafto SE, Amendola MA, Gefter WB. MR imaging of recurrent colorectal carcinoma versus fibrosis. J Comput Assist Tomogr 1988; 12:521–523. 9. Kinkel K, Ariche M, Tardivon AA, et al. Differentiation between recurrent tumor and benign conditions after treatment of gynecologic pelvic carcinoma: value of dynamic contrast-enhanced subtraction MR imaging. Radiology 1997; 204:55– 63. 10. Dao TH, Rahmouni A, Campana F, Laurent M, Asselain B, Fourquet A. Tumor recurrence versus fibrosis in the irradiated breast: differentiation with dynamic gadolinium-enhanced MR imaging. Radiology 1993; 187:751–755. 11. Ross JS, Delamarter R, Hueftle MG, et al. Gadolinium-DTPA-enhanced MR imaging of the postoperative lumbar spine: time course and mechanism of enhancement. AJR Am J Roentgenol 1989; 152:825– 834. 12. Bundschuh CV, Stein L, Slusser JH, Schinco FP, Ladaga LE, Dillon JD. Distinguishing between scar and recurrent herniated disk in postoperative patients: value of contrast-enhanced CT and MR

Volume 229

䡠

Number 1

13.

14. 15.

16.

17.

18.

19.

20.

21.

22.

23.

imaging. AJNR Am J Neuroradiol 1990; 11:949 –958. Weinmann HJ, Brasch RC, Press WR, Wesbey GE. Characteristics of gadolinium-DTPA complex: a potential NMR contrast agent. AJR Am J Roentgenol 1984; 142:619 – 624. Brasch RC. Rationale and applications for macromolecular Gd-based contrast agents. Magn Reson Med 1991; 22:282–287. Moon WK, Chang KH, Weinmann HJ, et al. Dynamic contrast-enhanced MR imaging of bacterial abscess and VX2 carcinoma in rabbits: comparison of gadopentetate dimeglumine and a macromolecular contrast agent. AJR Am J Roentgenol 2000; 174:1385–1390. Misselwitz B, Schmitt-Willich H, Ebert W, Frenzel T, Weinmann HJ. Pharmacokinetics of Gadomer-17, a new dendritic magnetic resonance contrast agent. MAGMA 2001; 12:128 –134. Grist TM, Korosec FR, Peters DC, et al. Steady-state and dynamic MR angiography with MS-325: initial experience in humans. Radiology 1998; 207:539 –544. Nguyen-minh C, Haughton VM, An HS, You JW, Wook S, Ho KC. Contrast media of high and low molecular weights in the detection of recurrent herniated disks. AJNR Am J Neuroradiol 1998; 19:889 – 893. Demsar F, Roberts TP, Schwickert HC, et al. A MRI spatial mapping technique for microvascular permeability and tissue blood volume based on macromolecular contrast agent distribution. Magn Reson Med 1997; 37:236 –242. Brasch R, Pham C, Shames D, et al. Assessing tumor angiogenesis using macromolecular MR imaging contrast media. J Magn Reson Imaging 1997; 7:68 –74. Li D, Zheng J, Weinmann HJ. Contrastenhanced MR imaging of coronary arteries: comparison of intra- and extravascular contrast agents in swine. Radiology 2001; 218:670 – 678. Jeong AK, Choi SI, Kim DH, et al. Evaluation by contrast-enhanced MR imaging of the lateral border zone in reperfused myocardial infarction in a cat model. Korean J Radiol 2001; 2:21–27. Henderson E, Sykes J, Drost D, Weinmann HJ, Rutt BK, Lee TY. Simultaneous MRI measurement of blood flow, blood volume, and capillary permeability in mammary tumors using two different

24.

25.

26.

27. 28.

29.

30.

31.

32.

33.

contrast agents. J Magn Reson Imaging 2000; 12:991–1003. Clarke SE, Weinmann HJ, Dai E, Lucas AR, Rutt BK. Comparison of two blood pool contrast agents for 0.5-T MR angiography: experimental study in rabbits. Radiology 2000; 214:787–794. Tacke J, Adam G, Classen H, Muhler A, Prescher A, Gunther RW. Dynamic MRI of a hypovascularized liver tumor model: comparison of a new blood pool contrast agent (24-gadolinium-DTPA-cascade-polymer) with gadopentetate dimeglumine. J Magn Reson Imaging 1997; 7:678 –682. Adam G, Muhler A, Spuntrup E, et al. Differentiation of spontaneous canine breast tumors using dynamic magnetic resonance imaging with 24-gadoliniumDTPA-cascade-polymer, a new bloodpool agent: preliminary experience. Invest Radiol 1996; 31:267–274. Jain RK. Transport of molecules across tumor vasculature. Cancer Metastasis Rev 1987; 6:559 –593. Su MY, Wang Z, Nalcioglu O. Investigation of longitudinal vascular changes in control and chemotherapy-treated tumors to serve as therapeutic efficacy predictors. J Magn Reson Imaging 1999; 9:128 –137. Erlemann R, Reiser MF, Peters PE, et al. Musculoskeletal neoplasms: static and dynamic Gd-DTPA– enhanced MR imaging. Radiology 1989; 171:767–773. Fletcher BD, Hanna SL, Fairclough DL, Gronemeyer SA. Pediatric musculoskeletal tumors: use of dynamic, contrast-enhanced MR imaging to monitor response to chemotherapy. Radiology 1992; 184: 243–248. Kuhl CK, Mielcareck P, Klaschik S, et al. Dynamic breast MR imaging: are signal intensity time course data useful for differential diagnosis of enhancing lesions? Radiology 1999; 211:101–110. Orel SG. Differentiating benign from malignant enhancing lesions identified at MR imaging of the breast: are time–signal intensity curves an accurate predictor? (editorial). Radiology 1999; 211:5–7. Verhoye M, van der Sanden BP, Rijken PF, et al. Assessment of the neovascular permeability in glioma xenografts by dynamic T1 MRI with Gadomer-17. Magn Reson Med 2002; 47:305–313.


䡠

139