- Email: [email protected]
Pediatric abdominal masses: diagnostic accuracy of diffusion weighted MRI
Available online at www.sciencedirect.com
Magnetic Resonance Imaging 28 (2010) 629 – 636
Pediatric abdominal masses: diagnostic accuracy of diffusion weighted MRI Murat Kocaoglua,⁎, Nail Bulakbasia , Hatice T. Sanala , Erol Kismetb , Bahadir Caliskanc , Veysel Akguna , Cem Tayfuna a
Department of Radiology, Gulhane Military Medical School, 06018 Etlik, Ankara, Turkey b Department of Pediatric Oncology, 06018 Etlik, Ankara, Turkey c Department of Pediatric Surgery, 06018, Etlik, Ankara, Turkey Received 11 October 2009; revised 14 December 2009; accepted 8 February 2010
Abstract Purpose: To retrospectively identify apparent diffusion coefficient (ADC) values of pediatric abdominal mass lesions, to determine whether measured ADC of the lesions and signal intensity on diffusion-weighted (DW) images allow discrimination between benign and malignant mass lesions. Materials and Methods: Approval for this retrospective study was obtained from the institutional review board. Children with abdominal mass lesions, who were examined by DW magnetic resonance imaging (MRI) were included in this study. DW MR images were obtained in the axial plane by using a non breath-hold single-shot spin-echo sequence on a 1.5-T MR scanner. ADCs were calculated for each lesion. ADC values were compared with Mann–Whitney U test. Receiver operating characteristic curve analysis was performed to determine cut-off values for ADC. The results of visual assessment on b800 images and ADC map images were compared with chi-square test. Results: Thirty-one abdominal mass lesions (16 benign, 15 malignant) in 26 patients (15 girls, 11 boys, ranging from 2 days to 17 years with 6.9 years mean) underwent MRI. Benign lesions had significantly higher ADC values than malignant ones (Pb.001). The mean ADCs of malignant lesions were 0.84±1.7×10−3 mm2/s, while the mean ADCs of the benign ones were 2.28±1.00×10−3 mm2/s. With respect to cutoff values of ADC: 1.11×10−3 mm2/s, sensitivity and negative predictive values were 100%, specificity was 78.6% and positive predictive value was 83.3%. For b800 and ADC map images, there were statistically significant differences on visual assessment. All malignant lesions had variable degrees of high signal intensity whereas eight of the 16 benign ones had low signal intensities on b800 images (Pb.001). On ADC map images, all malignant lesions were hypointense and most of the benign ones (n=11, 68.7%) were hyperintense (Pb.001). Conclusion: DW imaging can be used for reliable discrimination of benign and malignant pediatric abdominal mass lesions based on considerable differences in the ADC values and signal intensity changes. © 2010 Elsevier Inc. All rights reserved. Keywords: Diffusion weighted imaging; MRI; Pediatric imaging; Oncologic imaging; Neoplasms
1. Introduction Diffusion weighted (DW) magnetic resonance imaging (MRI) is a non-invasive method, which is capable of investigating the structure of biologic tissues at a microscopic level and may be used for in vivo tissue characterization [1,2]. This technique enables visualization of the
⁎ Corresponding author. Tel.: +90 532 6843394; fax: +90 312 3044702. E-mail addresses: [email protected], [email protected] (M. Kocaoglu). 0730–725X/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.mri.2010.02.010
arbitrary movement of water molecules (Brownian motion) in microscopic level that causes phase dispersion of the spins and consequential signal loss on diffusion sensitive sequences [2–7]. By using different b values and calculation of apparent diffusion coefficient (ADC) values, quantification of this signal loss can be achieved [5,6]. It has been shown that the areas of restricted diffusion in highly cellular areas demonstrate low ADC values compared with less cellular areas that show higher ADC values. In addition, by observing the comparative signal intensity attenuation, tissue characterization can also be possible depending on variation in molecular diffusion. For instance, cystic or necrotic
630
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
portion of the tumor will show greater signal attenuation on high b-value images because water diffusion is less restricted and correlate with areas of high signal intensity in ADC maps. Contrary to this, the more cellular solid tumor regions will continue to show relatively high signal intensity on high bvalue images and correlate with hypointense signal in ADC maps. One of the pitfalls of subjective assessment of DW images is that the signal depends on both molecular diffusion and the T2 relaxation time. A region with a very long T2 relaxation time may display high signal on high b-value images and may be mistaken for restricted diffusion. This is known as the “T2 shine-through” effect, which is always present in DW images. To eliminate “T2 shine-through” effect in DW imaging, ADC maps should be calculated. DW MRI has been used for the assessment of early and minor intracranial changes before observation of any perceptible signal abnormality on conventional MR sequences [8,9]. With the development of echo-planar imaging, parallel imaging, high gradient amplitudes and coil technologies, it has been possible to apply DW imaging in the abdomen and pelvis [10–15]. However, DW imaging in the abdominal region is problematic, due to susceptibility and motion artifacts. Published articles describing diffusion changes in pediatric abdominal mass lesions are very limited [16–19]. Feasibility of this technique has been shown by Olsen and Sebire in pediatric age group without calculating the ADC values [17]. In addition, the potential use of DW imaging and ADC values (without assessing the value of signal intensity) in discriminating benign and malignant pediatric extracranial mass lesions has been evaluated in another study; however, in this study, ADC values were found to be unable to differentiate benign masses from the malignant ones [16]. The purpose of this study was to identify ADC values of pediatric abdominal mass lesions, to determine whether ADCs of the lesions and signal intensities on DW images allow discriminating benign and malignant mass lesions. 2. Materials and methods 2.1. Patients In July 2006, we incorporated DW MRI with calculation and mapping of ADC values to the standard clinical MRI protocol of abdominal mass lesions. Consecutive children with abdominal mass lesions, which were examined by MRI and DW imaging before the chemotherapy/radiotherapy treatment and any interventional procedures in our radiology department were included in this retrospective study. Patients who had preoperative examinations at other institutions (n=2), patients who were treated before imaging (n=5) and patients who had images distorted by artifacts (n=3) were excluded. Pathologic results or clinical and radiologic follow-up findings were regarded as diagnostic reference. Children younger than 5 years were sedated. Approval for this retrospective study
was obtained from the institutional review board. Thirtyone abdominal mass lesions (16 benign, 15 malignant) in 26 patients were included in this study between July 2006April 2008. The study enrolled 11 male, 15 female, with an age range of 2 days to 17 years (mean 6.9 years), who had adrenal gland neuroblastoma (n=3), hepatoblastoma (n=2), Wilms tumor (n=3) urinary bladder rhabdomyosarcoma (n=2), undifferentiated presacral germ cell tumor (n=1) and its hepatic metastasis (n=1), gastric lymphoma (n=1), hepatic lymphoma (n=1), renal cell carcinoma (n=1), hydatid cyst of liver and spleen (n=2 [total 4 lesions]), retroperitoneal lymphangioma (n=1), peliosis hepatis (n=3), hepatic abscess (n=1), retroperitoneal abscess (n=1), mesenchymal hamartoma of the liver (n=1), splenic epidermoid cyst (n=1), abdominal wall hemangiomas (n=3), and ovarian cyst (n=1). 2.2. MR Technique All MR examinations were obtained in a 1.5T MR scanner (Symphony, Siemens, Erlangen, Germany) with 30 mT/m gradients. Both on the conventional and DW MRI, a standard head and neck coil or flexed coil were used depending on patients' size. We gave earplugs to the patients to minimize potential harmful effect of acoustic noise during MRI. Our routine abdominal MRI protocol included an unenhanced axial and coronal T1-weighted fast low angle shot (FLASH) sequences (TR/TE, 214/4.8; flip angle, 70°; slice thickness, 5–7 mm; number of excitations, 1), axial and coronal T2weighted turbo spin echo (TSE) sequences (TR/TE, 4000/ 160; slice thickness, 5–7 mm; number of excitations, 1; TSE factor, 29), axial fat saturated T2-weighted TSE sequence (TR/TE, 4400/76; slice thickness, 5–7 mm; number of excitations, 1; TSE factor, 29), axial and coronal true fast imaging with steady-state free precession (true FISP) (TR/ TE, 3.7/1.5; slice thickness, 5–7 mm; number of excitations, 1) sequences, in phase-out of phase imaging (TR/TE, 187/ 2.4–4.8 ms; flip angle, 70°; slice thickness, 5–7 mm; number of excitations, 1) and contrast-enhanced axial and coronal T1weighted sequences following the administration of a bolus injection of 0.1 mmol/kg meglumin gadoterat (Dotarem, Guerbet, France). The mean time of conventional MR examination was approximately 20–25 min. DW MRI was performed before contrast-enhanced sequences, and a fat saturated pulse was used to exclude severe chemical shift artifact. DW MR images were obtained in the axial plane by using a non breath-hold single-shot spin-echo sequence using the following image parameters: TR/TE, 2800/78; slice thickness, 5 mm; intersection gap, 0.5 mm; echoplanar readout matrix, 128×96; bandwidth, 1345 Hz/pixel. Two DW sequences were acquired with b values of 0 and 800 sec/mm2, and three DW MR images were gained for each b value. The isotropic DW MR image was generated on a pixel-by-pixel basis. ADC maps were created automatically. The mean time of the DW examination for each patient was approximately 3–5 min.
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
2.3. Image analysis 2.3.1. Quantitative analysis All MR images were analyzed retrospectively by two radiologists with 6 years experience in pediatric imaging and 9 years experience of MRI, without the knowledge of final diagnosis. They drew a rounded uniform region of interest (ROI) of 10 mm2 in the lesions with consensus for the ADC value recording to eliminate interobserver variability. Care was taken to exclude vessels, surrounding normal parenchyma and motion artifact from the ROIs. All ROIs in each lesion were placed in the solid/contrast-enhancing portion with the exclusion of any necrotic areas as judged from the T2- and contrast-enhanced T1-weighted MR images. For purely cystic lesions ROI was placed central portions of the lesions. ADC values were measured at least for five times, and these measurements were averaged. 2.3.2. Qualitative analysis The b800 diffusion trace images and ADC maps were evaluated with focus on the signal intensity of the lesions. On visual assessment, liver was considered as reference organ
631
and signal intensity of the lesions was subjectively given of the three grades; hyperintense, isointense and hypointense. Signal intensity assessment was made by two radiologists independently without the knowledge of the pathologic diagnosis. Results of the readings were compared and final score was reached by consensus. 2.4. Statistical analysis The ADC values were compared with Mann Whitney U test assuming unequal variances for the survival data between the benign and malignant lesions. Receiver operating characteristic (ROC) curve analysis was performed to determine the optimum cutoff values for ADC that discriminate benign and malignant mass lesions in terms of maximum sensitivity and minimum number of false positive results. The area under the ROC curve for each parameter was calculated. The results of visual assessment on b800 and ADC map images were compared with chi-square test. All statistical analyses were made by using SSPS release 15.0 program (SPSS for Windows; SPSS, Chicago, IL, USA). P values less than .05 were accepted as significant.
Table 1 Final diagnosis, ADC values and signal intensities of lesions on b=800 s/mm2 and ADC map images No/gender/age
Diagnosis
ADC (x10−3 mm2/s)
b=800 s/mm2
ADC map
1/M/17 m 2/F/11 m 3/M/3 y 4/M/6 y 5/F/5 y 6/F/13 m 7/F/18 m 8/F/6 m 9/M/6 m 10/F/7 y 11/F/6 y
16/M/9 y 17/M/9 y
Hepatoblastoma Hepatoblastoma Wilms Wilms Wilms Neuroblastoma a Neuroblastoma a Neuroblastoma a Gastric lymphoma Rhabdomyosarcoma b Germ cell tumors Hepatic metastasis Renal cell carcinoma Rhabdomyosarcoma b Hepatic lymphoma Hepatic hydatid cyst (Type 1) Hepatic hydatid cyst (Type 3) Splenic hydatid cyst (Type 1) Retroperitoneal lymphangioma Peliosis hepatis
18/M/12 y 19/F/19 m 20/F/2 d 21/F/5 y 22/F/12 y 23/F/3 y 24/F/12 y 25/M/9 y 26/M/14 y
Hepatic hydatid cyst (Type 3) Mesenchymal Hamartoma Ovarian cyst Hepatic abscess Retroperitoneal abscess Hemangioma c Hemangioma c Hemangioma c Splenic epidermoid cyst
0.95 1.11 0.72 0.73 0.77 0.77 0.78 0.70 0.43 0.94 0.93 0.82 1.02 1.09 0.97 2.65 2.59 2.14 4.11 1.02 1.06 0.98 3.37 2.91 2.96 0.60 0.64 1.95 2.01 2.38 2.72
Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense/hypointense d Hyperintense/hypointense d Hyperintense Hyperintense Hyperintense Hypointense Hypointense Hypointense Hypointense Hyperintense hyperintense hyperintense Hypointense Hypointense Hypointense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hypointense
Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense/hyperintense d Hypointense/ hyperintense d Hypointense Hypointense Hypointense Hyperintense Hyperintense Hyperintense Hyperintense Hypointense hypointense hypointense Hyperintense Hyperintense Hyperintense Hypointense Hypointense Hyperintense Hyperintense Hyperintense Hyperintense
12/F/14 y 13/M/7 y 14/F/17 y 15/M/16 y
a b c d
Adrenal gland neuroblastoma. Urinary bladder rhabdomyosarcoma. Abdominal wall muscle hemangioma. Peripheral/central regions.
632
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
3. Results In all patients, ADC maps were performed and ADC values were obtained. All lesions except peliosis hepatis, retroperitoneal abscess, and retroperitoneal lymphangioma had pathologic examination of surgical specimens. The ADC values of
Fig. 2. Necrotic metastasis of undifferentiated presacral germ cell tumor to the liver in a 6-year-old female. (A) Axial DW MR image (2800/78) acquired by using b value of 800 s/mm2 at the level of the liver shows hepatic mass lesion with a hypointense central necrotic area surrounded by a hyperintense peripheral rim. (B) ADC map demonstrates hypointense peripheral rim and hyperintense necrotic core consistent with restricted and increased diffusion, respectively. ADC value at the peripheral rim was 0.93×10−3 mm2/s.
Fig. 1. Right sided adrenal gland neuroblastoma in a 13-month-old female. (A) Axial T2-weighted image (4000/160) demonstrates well-defined hyperintense mass compressing the right kidney (arrowheads). (B) Axial DW MR image (2800/78) obtained by using b value of 800 s/mm2 confirm the hyperintense mass. (C) ADC map shows hypointense mass. The ADC value was 0.77×10−3 mm2/s.
benign and malignant mass lesions were calculated and showed in Table 1 (Figs. 1–4). The differences between ADC values of benign and malignant mass lesions were significant. Benign tumors had significantly higher ADC values than malignant tumors (Pb.001). The ADC values of malignant masses ranged between (0.43 and 1.11)×10−3 mm2/ s (mean, 0.84×10−3 mm2/s; standard deviation 1.7; median 0.82×10−3 mm2/s) while the ADCs of the benign masses were ranging between (0.60 and 4.11)×10 −3 mm 2 /s (mean 2.28×10 −3 mm 2 /s; standard deviation, 1.00; median 2.26×10−3 mm2 /s). The ADC value of retroperitoneal lymphangioma was the highest. Among the benign lesions, hepatic abscess had the lowest ADC value. Among the malignant lesions the lowest ADC values measured in gastric lymphoma. Of all malignant lesions, hepatoblastoma had the highest mean ADC value. All benign lesions but peliosis hepatis, hepatic and retroperitoneal abscesses showed values of more than 1.1×10−3 mm2/s. With respect to cut-off values of ADC: 1.11×10−3 mm2/s (area under ROC curve = 0,860, P =
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
633
are shown in Table 1 (Figs. 1–4).Two readers agreed on findings on DW images and ADC map appearances of 26 of 31 (84 %) lesions and disagreements were resolved with consensus in 5 (16%) lesions. On b800 and ADC map images, there were statistically significant differences on visual assessment between benign and malignant lesions. All malignant lesions had variable degrees of high signal whereas 8 of the 16 benign ones had low signal intensities on b800 images (Pb.001) (Fig. 1B and C; Fig. 2). On ADC map images, all of the malignant lesions were hypointense and most of the benign ones (n=11, 68.7%) were hyperintense (Pb.001). Hemangioma (n=3), abscess (n=2) and peliosis hepatis also showed variable degrees of hyperintensity on b800 images. Hemangiomas had high signals on ADC maps similar to the most of the other benign lesions, whereas peliosis hepatis and abscesses had low signal intensities similar to malignant ones. On visual assessment, the central content of the abscess was markedly hyperintense on b800 DW images, whereas the cystic content of the tumors and cystic metastasis were markedly hypointense (Figs. 2,4).
Fig. 3. Mesenchymal hamartoma of the liver in a 19-month-old female. (A) Axial T2-weighted image (4000/160) shows a well-defined hyperintense hepatic mass with multiple septations. (B) Axial DW MR image (2800/78) obtained by using b value of 800 sec/mm2 shows significant signal attenuation of the lesion suggesting increased diffusion. (C) ADC map at the same level shows the hyperintense lesion. ADC value was 2.91×10−3 mm2/s.
0.001) sensitivity and negative predictive values were 100%, specificity was 78.6% and positive predictive value was 83.3% in discrimination of benign and malignant lesions. Results of the visual assessment of the signal intensities of mass lesions in diffusion traces and ADC map images
Fig. 4. Retroperitoneal abscess following perforated appendicitis in a 12year-old female. (A) Axial DW MR image (2800/78) acquired by using b value of 800 s/mm2 reveals a high-signal-intensity lesion (arrows) posterior to the right kidney (arrowheads) (B) ADC map demonstrates hypointense lesion. ADC value was 0.64×10−3 mm2/s.
634
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
4. Discussion Reduced ADC values have been reported for most malignant tumors and high ADC values for most benign ones [15,20–22]. DW imaging can be used in several ways in patients with cancer including tumor detection, prediction and monitoring tumor response to treatment [3,19,23,24]. Moreover, DW imaging can improve the diagnostic accuracy of biopsies by sampling more cellular areas instead of areas with necrosis. In our study, the mean values for the ADC of benign lesions were significantly higher than those of malignant ones (2.28×10−3 mm2 vs. 0.84×10−3 mm2/s). It has been shown that the differences in diffusion restriction between benign and malignant mass lesions may be attributable to the differences in cellularity, necrosis, perfusion, and nuclei/ cytoplasm ratios [25–29]. The very densely packed and randomly organized tumor cells inhibit the effective molecular movement of extracellular water molecules. Tissue cellularity and the integrity of the cell membranes inversely affect the extent of diffusion restriction [25–27]. Extracellular tortuosity is another important factor for observed diffusion within a tissue [28]. Finally, the high nucleus to cytoplasm ratio of tumor cell limits intracellular motion of the free water molecules [29,30]. ADC value calculation for extraneurological lesions has been mostly studied in adult population [10–15] and only in very limited pediatric cohorts [16–20]. Abdel Razek et al. [20] found statistically significant differences in ADC values between the benign and malignant pediatric head and neck tumors by using b values of 0, 500 and 1000 s/mm2 per second. Uhl et al. [18] evaluated seven children with neuroblastoma by using b values of 0 and 1000 sec/ mm2/s and found a mean ADC value of 1.1×10−3 mm2/s (range 0.9–1.2×10−3 mm2/s). Humphries et al. [16] evaluated extracranial pediatric masses by using b values of 0, 500 and 1000 s/mm2 and found that cellularity is inversely related to ADC values. They also found relatively low ADC values in malignant masses (mean, 1.00×10−3 mm2/s) and relatively high ADC in benign ones (mean, 1.35×10−3 mm2 /s); however, differences between the benign and malignant lesions were not statistically significant. They used larger ROIs of 1.0 to 82.3 cm2, median 32 cm2 in their studies and presumably included micronecrotic/microcystic areas, which probably caused ineffective ADC measurements in the discrimination of benign and malignant lesions. Among statistically tested probable cut-off values, the ADC value of 1.11×10−3 mm2/s can be used as a cutoff value in the distinction of benign and malignant pediatric abdominal mass lesions with 100% sensitivity and negative predictive values, 78.6% specificity and 83.3% positive predictive value. These values can easily be used in daily practice in combination with MR images. Only three lesions, two abscesses and peliosis hepatis had ADC values less than 1.11×10−3 mm2/s and were below the cutoff values. Low ADC values in peliosis are most likely because of blood
products, which are more viscous than cystic fluid [31]. The restriction of diffusion and low ADC values in abscess are attributable to high viscosity of the abscess content. This finding is also helpful in distinction from cystic/necrotic components of mass lesions, which have low viscosity and, hence, high ADC values [32]. In this study, we also found significant differences between the signal intensities of benign and malignant mass lesions. The signal of the DW image contains contributions from the spin density and T1 and T2 relaxation times [32,33]. Therefore, cysts/cystic components of the neoplasms and abscesses may display a strong T2 effect (T2 shine-through effect); thus at low b value, high signal intensity on DW image may be caused either by reduced diffusion or by T2 shine-through effect, as in the liquidcystic content. At higher b values, the contribution of the T2 shine-through to the signal intensity decreases, whereas tissue cellularity makes a greater contribution [34,35]. Hence, the hyperintensities of the solid components of malignant lesions and abscesses on b800 images can be partly attributed to restricted diffusion. Contrary to this, low signal intensity on DW image is due to high tissue diffusivity. On high b-value images (b800) signal of cystic lesions is low, whereas signal of malignant tumors is high. Regardless of the signal intensity on b800, the lesions with high signal on ADC map images are consistent with benign lesions. On the other hand, combination of high b800 signal intensity and low ADC map signal intensity is highly suggestive for malignancy. Abscess and peliosis hepatis also fall into this last category. Because pus in an abscess cavity is a thick, mucoid fluid that consists of inflammatory cells, bacteria, necrotic tissue and proteinaceous exudates with high viscosity, it is associated with restriction in diffusion and high signal intensity on b800 image [36]. Contrary to this, cystic or necrotic cavities have lower viscosity than pus, and demonstrate greater signal intensity attenuation on high b value images [37]. We think that abovementioned signal features of the lesions are a practical way deciding whether the lesions are benign or malignant. However, we suggest that further studies are needed to show the role of the visual assessment in the characterization of the mass lesions. Our study is limited by the relatively small number of patients and by the heterogeneous example of lesions; however, most of the pediatric abdominal mass lesions have been included. Our limited number of each type of mass prevents making analysis to find possible differences in ADC values of different types of malignant and benign mass lesions from one another. Further studies enrolling larger populations are needed to document the role of this modality in the demonstration of an individual ADC value to specify the lesions. Second, we analyzed only two b values, one low and one high. Previous studies used more data points to let more precise calculation of ADC ratio [13,16]. However, repeated measurements with increased number of b values lengthens the examination times, creating problems of repeatability for sedated children. Third, most of our benign
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
lesions were predominantly cystic and higher ADC values, which made the distinction from the malignant ones easy. Further studies are recommended to assess the full value of DW imaging in benign solid pediatric masses. Finally, ROIs analyzed on MR images may not precisely represent to the biopsy specimens used in the pathologic examination. In conclusion, DW imaging can be used for reliable discrimination of benign and malignant pediatric abdominal mass lesions based on considerable differences in the ADC values and signal intensity changes. This technique can be added to routine imaging protocols without a considerable increase in the study time, and does not necessitate the administration of intravenous contrast media. References [1] Le Bihan DJ. Differentiation of benign versus pathologic compression fractures with diffusion-weighted MR imaging: a closer step toward the ‘holy grail’ of tissue characterization? Radiology 1998;207:305–7. [2] Szafer A, Zhong J, Gore JC. Theoretical model for water diffusion in tissues. Magn Reson Med 1995;33:697–712. [3] Lang P, Johnston JO, Arenal-Romero F, Gooding CA. Advances in MR imaging of pediatric musculoskeletal neoplasms. Magn Reson Imaging Clin N Am 1998;6:579–604. [4] Szafer A, Zhong J, Anderson AW, Gore JC. Diffusion-weighted imaging in tissues: theoretical models. NMR Biomed 1995;8:289–96. [5] Le Bihan D, Delannoy J, Levin RL. Temperature mapping with MR imaging of molecular diffusion: application to hyperthermia. Radiology 1989;171:853–7. [6] Le Bihan D, Breton E, Lallemand D, Grenier P, Cabanis E, LavalJeantet M. MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology 1986;161:401–7. [7] Tien RD, Felsberg GJ, Friedman H, Brown M, MacFall J. MR imaging of high-grade cerebral gliomas: value of diffusion weighted echoplanar pulse sequences. AJR Am J Roentgenol 1994;162:671–7. [8] Eastwood JD, Lev MH, Wintermark M, Fitzek C, Barboriak DP, Delong DM, et al. Correlation of early dynamic CT perfusion imaging with whole-brain MR diffusion and perfusion imaging in acute hemispheric stroke. AJNR Am J Neuroradiol 2003;24:1869–75. [9] Yoshikawa K, Nakata Y, Yamada K, Nakagawa M. Early pathological changes in the parkinsonian brain demonstrated by diffusion tensor MRI. J Neurol Neurosurg Psychiatry 2004;75:481–4. [10] Parikh T, Drew SJ, Lee VS, Wong S, Hecht EM, Babb JS, et al. Focal liver lesion detection and characterization with diffusion-weighted MR imaging: comparison with standard breath-hold T2-weighted imaging. Radiology 2008;246:812–22. [11] Namimoto T, Yamashita Y, Sumi S, Tang Y, Takahashi M. Focal liver masses: characterization with diffusion-weighted echo-planar MR imaging. Radiology 1997;204:739–44. [12] Kilickesmez O, Yirik G, Bayramoglu S, Cimilli T, Aydin S. Nonbreath-hold high b-value diffusion-weighted MRI with parallel imaging technique: apparent diffusion coefficient determination in normal abdominal organs. Diagn Interv Radiol 2008;14:83–7. [13] Ichikawa T, Haradome H, Hachiya J, Nitatori T, Araki T. Diffusionweighted MR imaging with a single-shot echoplanar sequence: detection and characterization of focal hepatic lesions. AJR Am J Roentgenol 1998;170:397–402. [14] Demir OI, Obuz F, Sagol O, Dicle O. Contribution of diffusionweighted MRI to the differential diagnosis of hepatic masses. Diagn Interv Radiol 2007;13:81–6. [15] Yamada I, Aung W, Himeno Y, Nakagawa T, Shibuya H. Diffusion coefficients in abdominal organs and hepatic lesions: evaluation with
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
635
intravoxel incoherent motion echo-planar MR imaging. Radiology 1999;210:617–23. Humphries PD, Sebire NJ, Siegel MJ, Olsen OE. Tumors in pediatric patients at diffusion-weighted MR imaging: apparent diffusion coefficient and tumor cellularity. Radiology 2007;245:848–54. Olsen OE, Sebire NJ. Apparent diffusion coefficient maps of pediatric mass lesions with free-breathing diffusion-weighted magnetic resonance: feasibility study. Acta Radiol 2006;47:198–204. Uhl M, Altehoefer C, Kontny U, Il'yasov K, Buchert M, Langer M. MRI-diffusion imaging of neuroblastomas: first results and correlation to histology. Eur Radiol 2002;12:2335–8. Alibek S, Cavallaro A, Aplas A, Uder M, Staatz G. Diffusion weighted imaging of pediatric and adolescent malignancies with regard to detection and delineation: initial experience. Acad Radiol 2009;16: 866–71. Abdel Razek AA, Gaballa G, Elhawarey G, Megahed AS, Hafez M, Nada N. Characterization of pediatric head and neck masses with diffusion-weighted MR imaging. Eur Radiol 2009;19:201–8. Kim T, Murakami T, Takahashi S, Hori M, Tsuda K, Nakamura H. Diffusion-weighted single-shot echoplanar MR imaging for liver disease. AJR Am J Roentgenol 1999;173:393–8. Lyng H, Haraldseth O, Rofstad EK. Measurement of cell density and necrotic fraction in human melanoma xenografts by diffusion weighted magnetic resonance imaging. Magn Reson Med 2000;43: 828–36. Thoeny HC, De Keyzer F, Chen F, Ni Y, Landuyt W, Verbeken EK, et al. Diffusion-weighted MR imaging in monitoring the effect of a vascular targeting agent on rhabdomyosarcoma in rats. Radiology 2005;234:756–64. Koh DM, Scurr E, Collins D, Kanber B, Norman A, Leach MO, et al. Predicting response of colorectal hepatic metastasis: value of pretreatment apparent diffusion coefficients. AJR Am J Roentgenol 2007;188:1001–8. Guo Y, Cai YQ, Cai ZL, Gao YG, An NY, Ma L, et al. Differentiation of clinically benign and malignant breast lesions using diffusionweighted imaging. J Magn Reson Imaging 2002;16:172–8. Gauvain KM, McKinstry RC, Mukherjee P, Perry A, Neil JJ, Kaufman BA, et al. Evaluating pediatric brain tumor cellularity with diffusiontensor imaging. AJR Am J Roentgenol 2001;177:449–54. Sugahara T, Korogi Y, Kochi M, Ikushima I, Shigematu Y, Hirai T, et al. Usefulness of diffusion-weighted MRI with echo-planar technique in the evaluation of cellularity in gliomas. J Magn Reson Imaging 1999;9:53–60. Lang P, Wendland MF, Saeed M, Gindele A, Rosenau W, Mathur A, et al. Osteogenic sarcoma: noninvasive in vivo assessment of tumor necrosis with diffusion-weighted MR imaging. Radiology 1998;206: 227–35. Norris DG. The effects of microscopic tissue parameters on the diffusion weighted magnetic resonance imaging experiment. NMR Biomed 2001;14:77–93. Kotsenas AL, Roth TC, Maness WK, Faerber EN. Abnormal diffusionweighted MRI in a medulloblastoma: does it reflect small cell histology? Pediatr Radiol 1999;29:524–6. Okamoto K, Ito J, Ishikawa K, Sakai K, Tokiguchi S. Diffusionweighted echo-planar MR imaging in differential diagnosis of brain tumors and tumor-like conditions. Eur Radiol 2000;10:1342–50. Chan JH, Tsui EY, Luk SH, Fung AS, Yuen MK, Szeto ML, et al. Diffusion-weighted MR imaging of the liver: distinguishing hepatic abscess from cystic or necrotic tumor. Abdom Imaging 2000;26: 161–5. Eastwood JD, Engelter ST, MacFall JF, Delong DM, Provenzale JM. Quantitative assessment of the time course of infarct signal intensity on diffusion-weighted images. AJNR Am J Neuroradiol 2003;24: 680–7. Hakyemez B, Aksoy U, Yildiz H, Ergin N. Intracranial epidermoid cysts: diffusion-weighted, FLAIR and conventional MR findings. Eur J Radiol 2005;54:214–20.
636
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
[35] Colagrande S, Belli G, Politi LS, Mannelli L, Pasquinelli F, Villari N. The influence of diffusion- and relaxation-related factors on signal intensity: an introductive guide to magnetic resonance diffusionweighted imaging studies. J Comput Assist Tomogr 2008;32:463–74. [36] Naganawa S, Kawai H, Fukatsu H, Sakurai Y, Aoki I, Miura S, Mimura T, et al. Diffusion-weighted imaging of the liver: technical
challenges and prospects for the future. Magn Reson Med Sci 2005;4:175–86. [37] Ebisu T, Tanaka C, Umeda M, Kitamura M, Naruse S, Higuchi T, et al. Discrimination of brain abscess from necrotic or cystic tumors by diffusion-weighted echo planar imaging. Magn Reson Imaging 1996;14:1113–6.
Magnetic Resonance Imaging 28 (2010) 629 – 636
Pediatric abdominal masses: diagnostic accuracy of diffusion weighted MRI Murat Kocaoglua,⁎, Nail Bulakbasia , Hatice T. Sanala , Erol Kismetb , Bahadir Caliskanc , Veysel Akguna , Cem Tayfuna a
Department of Radiology, Gulhane Military Medical School, 06018 Etlik, Ankara, Turkey b Department of Pediatric Oncology, 06018 Etlik, Ankara, Turkey c Department of Pediatric Surgery, 06018, Etlik, Ankara, Turkey Received 11 October 2009; revised 14 December 2009; accepted 8 February 2010
Abstract Purpose: To retrospectively identify apparent diffusion coefficient (ADC) values of pediatric abdominal mass lesions, to determine whether measured ADC of the lesions and signal intensity on diffusion-weighted (DW) images allow discrimination between benign and malignant mass lesions. Materials and Methods: Approval for this retrospective study was obtained from the institutional review board. Children with abdominal mass lesions, who were examined by DW magnetic resonance imaging (MRI) were included in this study. DW MR images were obtained in the axial plane by using a non breath-hold single-shot spin-echo sequence on a 1.5-T MR scanner. ADCs were calculated for each lesion. ADC values were compared with Mann–Whitney U test. Receiver operating characteristic curve analysis was performed to determine cut-off values for ADC. The results of visual assessment on b800 images and ADC map images were compared with chi-square test. Results: Thirty-one abdominal mass lesions (16 benign, 15 malignant) in 26 patients (15 girls, 11 boys, ranging from 2 days to 17 years with 6.9 years mean) underwent MRI. Benign lesions had significantly higher ADC values than malignant ones (Pb.001). The mean ADCs of malignant lesions were 0.84±1.7×10−3 mm2/s, while the mean ADCs of the benign ones were 2.28±1.00×10−3 mm2/s. With respect to cutoff values of ADC: 1.11×10−3 mm2/s, sensitivity and negative predictive values were 100%, specificity was 78.6% and positive predictive value was 83.3%. For b800 and ADC map images, there were statistically significant differences on visual assessment. All malignant lesions had variable degrees of high signal intensity whereas eight of the 16 benign ones had low signal intensities on b800 images (Pb.001). On ADC map images, all malignant lesions were hypointense and most of the benign ones (n=11, 68.7%) were hyperintense (Pb.001). Conclusion: DW imaging can be used for reliable discrimination of benign and malignant pediatric abdominal mass lesions based on considerable differences in the ADC values and signal intensity changes. © 2010 Elsevier Inc. All rights reserved. Keywords: Diffusion weighted imaging; MRI; Pediatric imaging; Oncologic imaging; Neoplasms
1. Introduction Diffusion weighted (DW) magnetic resonance imaging (MRI) is a non-invasive method, which is capable of investigating the structure of biologic tissues at a microscopic level and may be used for in vivo tissue characterization [1,2]. This technique enables visualization of the
⁎ Corresponding author. Tel.: +90 532 6843394; fax: +90 312 3044702. E-mail addresses: [email protected], [email protected] (M. Kocaoglu). 0730–725X/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.mri.2010.02.010
arbitrary movement of water molecules (Brownian motion) in microscopic level that causes phase dispersion of the spins and consequential signal loss on diffusion sensitive sequences [2–7]. By using different b values and calculation of apparent diffusion coefficient (ADC) values, quantification of this signal loss can be achieved [5,6]. It has been shown that the areas of restricted diffusion in highly cellular areas demonstrate low ADC values compared with less cellular areas that show higher ADC values. In addition, by observing the comparative signal intensity attenuation, tissue characterization can also be possible depending on variation in molecular diffusion. For instance, cystic or necrotic
630
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
portion of the tumor will show greater signal attenuation on high b-value images because water diffusion is less restricted and correlate with areas of high signal intensity in ADC maps. Contrary to this, the more cellular solid tumor regions will continue to show relatively high signal intensity on high bvalue images and correlate with hypointense signal in ADC maps. One of the pitfalls of subjective assessment of DW images is that the signal depends on both molecular diffusion and the T2 relaxation time. A region with a very long T2 relaxation time may display high signal on high b-value images and may be mistaken for restricted diffusion. This is known as the “T2 shine-through” effect, which is always present in DW images. To eliminate “T2 shine-through” effect in DW imaging, ADC maps should be calculated. DW MRI has been used for the assessment of early and minor intracranial changes before observation of any perceptible signal abnormality on conventional MR sequences [8,9]. With the development of echo-planar imaging, parallel imaging, high gradient amplitudes and coil technologies, it has been possible to apply DW imaging in the abdomen and pelvis [10–15]. However, DW imaging in the abdominal region is problematic, due to susceptibility and motion artifacts. Published articles describing diffusion changes in pediatric abdominal mass lesions are very limited [16–19]. Feasibility of this technique has been shown by Olsen and Sebire in pediatric age group without calculating the ADC values [17]. In addition, the potential use of DW imaging and ADC values (without assessing the value of signal intensity) in discriminating benign and malignant pediatric extracranial mass lesions has been evaluated in another study; however, in this study, ADC values were found to be unable to differentiate benign masses from the malignant ones [16]. The purpose of this study was to identify ADC values of pediatric abdominal mass lesions, to determine whether ADCs of the lesions and signal intensities on DW images allow discriminating benign and malignant mass lesions. 2. Materials and methods 2.1. Patients In July 2006, we incorporated DW MRI with calculation and mapping of ADC values to the standard clinical MRI protocol of abdominal mass lesions. Consecutive children with abdominal mass lesions, which were examined by MRI and DW imaging before the chemotherapy/radiotherapy treatment and any interventional procedures in our radiology department were included in this retrospective study. Patients who had preoperative examinations at other institutions (n=2), patients who were treated before imaging (n=5) and patients who had images distorted by artifacts (n=3) were excluded. Pathologic results or clinical and radiologic follow-up findings were regarded as diagnostic reference. Children younger than 5 years were sedated. Approval for this retrospective study
was obtained from the institutional review board. Thirtyone abdominal mass lesions (16 benign, 15 malignant) in 26 patients were included in this study between July 2006April 2008. The study enrolled 11 male, 15 female, with an age range of 2 days to 17 years (mean 6.9 years), who had adrenal gland neuroblastoma (n=3), hepatoblastoma (n=2), Wilms tumor (n=3) urinary bladder rhabdomyosarcoma (n=2), undifferentiated presacral germ cell tumor (n=1) and its hepatic metastasis (n=1), gastric lymphoma (n=1), hepatic lymphoma (n=1), renal cell carcinoma (n=1), hydatid cyst of liver and spleen (n=2 [total 4 lesions]), retroperitoneal lymphangioma (n=1), peliosis hepatis (n=3), hepatic abscess (n=1), retroperitoneal abscess (n=1), mesenchymal hamartoma of the liver (n=1), splenic epidermoid cyst (n=1), abdominal wall hemangiomas (n=3), and ovarian cyst (n=1). 2.2. MR Technique All MR examinations were obtained in a 1.5T MR scanner (Symphony, Siemens, Erlangen, Germany) with 30 mT/m gradients. Both on the conventional and DW MRI, a standard head and neck coil or flexed coil were used depending on patients' size. We gave earplugs to the patients to minimize potential harmful effect of acoustic noise during MRI. Our routine abdominal MRI protocol included an unenhanced axial and coronal T1-weighted fast low angle shot (FLASH) sequences (TR/TE, 214/4.8; flip angle, 70°; slice thickness, 5–7 mm; number of excitations, 1), axial and coronal T2weighted turbo spin echo (TSE) sequences (TR/TE, 4000/ 160; slice thickness, 5–7 mm; number of excitations, 1; TSE factor, 29), axial fat saturated T2-weighted TSE sequence (TR/TE, 4400/76; slice thickness, 5–7 mm; number of excitations, 1; TSE factor, 29), axial and coronal true fast imaging with steady-state free precession (true FISP) (TR/ TE, 3.7/1.5; slice thickness, 5–7 mm; number of excitations, 1) sequences, in phase-out of phase imaging (TR/TE, 187/ 2.4–4.8 ms; flip angle, 70°; slice thickness, 5–7 mm; number of excitations, 1) and contrast-enhanced axial and coronal T1weighted sequences following the administration of a bolus injection of 0.1 mmol/kg meglumin gadoterat (Dotarem, Guerbet, France). The mean time of conventional MR examination was approximately 20–25 min. DW MRI was performed before contrast-enhanced sequences, and a fat saturated pulse was used to exclude severe chemical shift artifact. DW MR images were obtained in the axial plane by using a non breath-hold single-shot spin-echo sequence using the following image parameters: TR/TE, 2800/78; slice thickness, 5 mm; intersection gap, 0.5 mm; echoplanar readout matrix, 128×96; bandwidth, 1345 Hz/pixel. Two DW sequences were acquired with b values of 0 and 800 sec/mm2, and three DW MR images were gained for each b value. The isotropic DW MR image was generated on a pixel-by-pixel basis. ADC maps were created automatically. The mean time of the DW examination for each patient was approximately 3–5 min.
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
2.3. Image analysis 2.3.1. Quantitative analysis All MR images were analyzed retrospectively by two radiologists with 6 years experience in pediatric imaging and 9 years experience of MRI, without the knowledge of final diagnosis. They drew a rounded uniform region of interest (ROI) of 10 mm2 in the lesions with consensus for the ADC value recording to eliminate interobserver variability. Care was taken to exclude vessels, surrounding normal parenchyma and motion artifact from the ROIs. All ROIs in each lesion were placed in the solid/contrast-enhancing portion with the exclusion of any necrotic areas as judged from the T2- and contrast-enhanced T1-weighted MR images. For purely cystic lesions ROI was placed central portions of the lesions. ADC values were measured at least for five times, and these measurements were averaged. 2.3.2. Qualitative analysis The b800 diffusion trace images and ADC maps were evaluated with focus on the signal intensity of the lesions. On visual assessment, liver was considered as reference organ
631
and signal intensity of the lesions was subjectively given of the three grades; hyperintense, isointense and hypointense. Signal intensity assessment was made by two radiologists independently without the knowledge of the pathologic diagnosis. Results of the readings were compared and final score was reached by consensus. 2.4. Statistical analysis The ADC values were compared with Mann Whitney U test assuming unequal variances for the survival data between the benign and malignant lesions. Receiver operating characteristic (ROC) curve analysis was performed to determine the optimum cutoff values for ADC that discriminate benign and malignant mass lesions in terms of maximum sensitivity and minimum number of false positive results. The area under the ROC curve for each parameter was calculated. The results of visual assessment on b800 and ADC map images were compared with chi-square test. All statistical analyses were made by using SSPS release 15.0 program (SPSS for Windows; SPSS, Chicago, IL, USA). P values less than .05 were accepted as significant.
Table 1 Final diagnosis, ADC values and signal intensities of lesions on b=800 s/mm2 and ADC map images No/gender/age
Diagnosis
ADC (x10−3 mm2/s)
b=800 s/mm2
ADC map
1/M/17 m 2/F/11 m 3/M/3 y 4/M/6 y 5/F/5 y 6/F/13 m 7/F/18 m 8/F/6 m 9/M/6 m 10/F/7 y 11/F/6 y
16/M/9 y 17/M/9 y
Hepatoblastoma Hepatoblastoma Wilms Wilms Wilms Neuroblastoma a Neuroblastoma a Neuroblastoma a Gastric lymphoma Rhabdomyosarcoma b Germ cell tumors Hepatic metastasis Renal cell carcinoma Rhabdomyosarcoma b Hepatic lymphoma Hepatic hydatid cyst (Type 1) Hepatic hydatid cyst (Type 3) Splenic hydatid cyst (Type 1) Retroperitoneal lymphangioma Peliosis hepatis
18/M/12 y 19/F/19 m 20/F/2 d 21/F/5 y 22/F/12 y 23/F/3 y 24/F/12 y 25/M/9 y 26/M/14 y
Hepatic hydatid cyst (Type 3) Mesenchymal Hamartoma Ovarian cyst Hepatic abscess Retroperitoneal abscess Hemangioma c Hemangioma c Hemangioma c Splenic epidermoid cyst
0.95 1.11 0.72 0.73 0.77 0.77 0.78 0.70 0.43 0.94 0.93 0.82 1.02 1.09 0.97 2.65 2.59 2.14 4.11 1.02 1.06 0.98 3.37 2.91 2.96 0.60 0.64 1.95 2.01 2.38 2.72
Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense/hypointense d Hyperintense/hypointense d Hyperintense Hyperintense Hyperintense Hypointense Hypointense Hypointense Hypointense Hyperintense hyperintense hyperintense Hypointense Hypointense Hypointense Hyperintense Hyperintense Hyperintense Hyperintense Hyperintense Hypointense
Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense Hypointense/hyperintense d Hypointense/ hyperintense d Hypointense Hypointense Hypointense Hyperintense Hyperintense Hyperintense Hyperintense Hypointense hypointense hypointense Hyperintense Hyperintense Hyperintense Hypointense Hypointense Hyperintense Hyperintense Hyperintense Hyperintense
12/F/14 y 13/M/7 y 14/F/17 y 15/M/16 y
a b c d
Adrenal gland neuroblastoma. Urinary bladder rhabdomyosarcoma. Abdominal wall muscle hemangioma. Peripheral/central regions.
632
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
3. Results In all patients, ADC maps were performed and ADC values were obtained. All lesions except peliosis hepatis, retroperitoneal abscess, and retroperitoneal lymphangioma had pathologic examination of surgical specimens. The ADC values of
Fig. 2. Necrotic metastasis of undifferentiated presacral germ cell tumor to the liver in a 6-year-old female. (A) Axial DW MR image (2800/78) acquired by using b value of 800 s/mm2 at the level of the liver shows hepatic mass lesion with a hypointense central necrotic area surrounded by a hyperintense peripheral rim. (B) ADC map demonstrates hypointense peripheral rim and hyperintense necrotic core consistent with restricted and increased diffusion, respectively. ADC value at the peripheral rim was 0.93×10−3 mm2/s.
Fig. 1. Right sided adrenal gland neuroblastoma in a 13-month-old female. (A) Axial T2-weighted image (4000/160) demonstrates well-defined hyperintense mass compressing the right kidney (arrowheads). (B) Axial DW MR image (2800/78) obtained by using b value of 800 s/mm2 confirm the hyperintense mass. (C) ADC map shows hypointense mass. The ADC value was 0.77×10−3 mm2/s.
benign and malignant mass lesions were calculated and showed in Table 1 (Figs. 1–4). The differences between ADC values of benign and malignant mass lesions were significant. Benign tumors had significantly higher ADC values than malignant tumors (Pb.001). The ADC values of malignant masses ranged between (0.43 and 1.11)×10−3 mm2/ s (mean, 0.84×10−3 mm2/s; standard deviation 1.7; median 0.82×10−3 mm2/s) while the ADCs of the benign masses were ranging between (0.60 and 4.11)×10 −3 mm 2 /s (mean 2.28×10 −3 mm 2 /s; standard deviation, 1.00; median 2.26×10−3 mm2 /s). The ADC value of retroperitoneal lymphangioma was the highest. Among the benign lesions, hepatic abscess had the lowest ADC value. Among the malignant lesions the lowest ADC values measured in gastric lymphoma. Of all malignant lesions, hepatoblastoma had the highest mean ADC value. All benign lesions but peliosis hepatis, hepatic and retroperitoneal abscesses showed values of more than 1.1×10−3 mm2/s. With respect to cut-off values of ADC: 1.11×10−3 mm2/s (area under ROC curve = 0,860, P =
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
633
are shown in Table 1 (Figs. 1–4).Two readers agreed on findings on DW images and ADC map appearances of 26 of 31 (84 %) lesions and disagreements were resolved with consensus in 5 (16%) lesions. On b800 and ADC map images, there were statistically significant differences on visual assessment between benign and malignant lesions. All malignant lesions had variable degrees of high signal whereas 8 of the 16 benign ones had low signal intensities on b800 images (Pb.001) (Fig. 1B and C; Fig. 2). On ADC map images, all of the malignant lesions were hypointense and most of the benign ones (n=11, 68.7%) were hyperintense (Pb.001). Hemangioma (n=3), abscess (n=2) and peliosis hepatis also showed variable degrees of hyperintensity on b800 images. Hemangiomas had high signals on ADC maps similar to the most of the other benign lesions, whereas peliosis hepatis and abscesses had low signal intensities similar to malignant ones. On visual assessment, the central content of the abscess was markedly hyperintense on b800 DW images, whereas the cystic content of the tumors and cystic metastasis were markedly hypointense (Figs. 2,4).
Fig. 3. Mesenchymal hamartoma of the liver in a 19-month-old female. (A) Axial T2-weighted image (4000/160) shows a well-defined hyperintense hepatic mass with multiple septations. (B) Axial DW MR image (2800/78) obtained by using b value of 800 sec/mm2 shows significant signal attenuation of the lesion suggesting increased diffusion. (C) ADC map at the same level shows the hyperintense lesion. ADC value was 2.91×10−3 mm2/s.
0.001) sensitivity and negative predictive values were 100%, specificity was 78.6% and positive predictive value was 83.3% in discrimination of benign and malignant lesions. Results of the visual assessment of the signal intensities of mass lesions in diffusion traces and ADC map images
Fig. 4. Retroperitoneal abscess following perforated appendicitis in a 12year-old female. (A) Axial DW MR image (2800/78) acquired by using b value of 800 s/mm2 reveals a high-signal-intensity lesion (arrows) posterior to the right kidney (arrowheads) (B) ADC map demonstrates hypointense lesion. ADC value was 0.64×10−3 mm2/s.
634
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
4. Discussion Reduced ADC values have been reported for most malignant tumors and high ADC values for most benign ones [15,20–22]. DW imaging can be used in several ways in patients with cancer including tumor detection, prediction and monitoring tumor response to treatment [3,19,23,24]. Moreover, DW imaging can improve the diagnostic accuracy of biopsies by sampling more cellular areas instead of areas with necrosis. In our study, the mean values for the ADC of benign lesions were significantly higher than those of malignant ones (2.28×10−3 mm2 vs. 0.84×10−3 mm2/s). It has been shown that the differences in diffusion restriction between benign and malignant mass lesions may be attributable to the differences in cellularity, necrosis, perfusion, and nuclei/ cytoplasm ratios [25–29]. The very densely packed and randomly organized tumor cells inhibit the effective molecular movement of extracellular water molecules. Tissue cellularity and the integrity of the cell membranes inversely affect the extent of diffusion restriction [25–27]. Extracellular tortuosity is another important factor for observed diffusion within a tissue [28]. Finally, the high nucleus to cytoplasm ratio of tumor cell limits intracellular motion of the free water molecules [29,30]. ADC value calculation for extraneurological lesions has been mostly studied in adult population [10–15] and only in very limited pediatric cohorts [16–20]. Abdel Razek et al. [20] found statistically significant differences in ADC values between the benign and malignant pediatric head and neck tumors by using b values of 0, 500 and 1000 s/mm2 per second. Uhl et al. [18] evaluated seven children with neuroblastoma by using b values of 0 and 1000 sec/ mm2/s and found a mean ADC value of 1.1×10−3 mm2/s (range 0.9–1.2×10−3 mm2/s). Humphries et al. [16] evaluated extracranial pediatric masses by using b values of 0, 500 and 1000 s/mm2 and found that cellularity is inversely related to ADC values. They also found relatively low ADC values in malignant masses (mean, 1.00×10−3 mm2/s) and relatively high ADC in benign ones (mean, 1.35×10−3 mm2 /s); however, differences between the benign and malignant lesions were not statistically significant. They used larger ROIs of 1.0 to 82.3 cm2, median 32 cm2 in their studies and presumably included micronecrotic/microcystic areas, which probably caused ineffective ADC measurements in the discrimination of benign and malignant lesions. Among statistically tested probable cut-off values, the ADC value of 1.11×10−3 mm2/s can be used as a cutoff value in the distinction of benign and malignant pediatric abdominal mass lesions with 100% sensitivity and negative predictive values, 78.6% specificity and 83.3% positive predictive value. These values can easily be used in daily practice in combination with MR images. Only three lesions, two abscesses and peliosis hepatis had ADC values less than 1.11×10−3 mm2/s and were below the cutoff values. Low ADC values in peliosis are most likely because of blood
products, which are more viscous than cystic fluid [31]. The restriction of diffusion and low ADC values in abscess are attributable to high viscosity of the abscess content. This finding is also helpful in distinction from cystic/necrotic components of mass lesions, which have low viscosity and, hence, high ADC values [32]. In this study, we also found significant differences between the signal intensities of benign and malignant mass lesions. The signal of the DW image contains contributions from the spin density and T1 and T2 relaxation times [32,33]. Therefore, cysts/cystic components of the neoplasms and abscesses may display a strong T2 effect (T2 shine-through effect); thus at low b value, high signal intensity on DW image may be caused either by reduced diffusion or by T2 shine-through effect, as in the liquidcystic content. At higher b values, the contribution of the T2 shine-through to the signal intensity decreases, whereas tissue cellularity makes a greater contribution [34,35]. Hence, the hyperintensities of the solid components of malignant lesions and abscesses on b800 images can be partly attributed to restricted diffusion. Contrary to this, low signal intensity on DW image is due to high tissue diffusivity. On high b-value images (b800) signal of cystic lesions is low, whereas signal of malignant tumors is high. Regardless of the signal intensity on b800, the lesions with high signal on ADC map images are consistent with benign lesions. On the other hand, combination of high b800 signal intensity and low ADC map signal intensity is highly suggestive for malignancy. Abscess and peliosis hepatis also fall into this last category. Because pus in an abscess cavity is a thick, mucoid fluid that consists of inflammatory cells, bacteria, necrotic tissue and proteinaceous exudates with high viscosity, it is associated with restriction in diffusion and high signal intensity on b800 image [36]. Contrary to this, cystic or necrotic cavities have lower viscosity than pus, and demonstrate greater signal intensity attenuation on high b value images [37]. We think that abovementioned signal features of the lesions are a practical way deciding whether the lesions are benign or malignant. However, we suggest that further studies are needed to show the role of the visual assessment in the characterization of the mass lesions. Our study is limited by the relatively small number of patients and by the heterogeneous example of lesions; however, most of the pediatric abdominal mass lesions have been included. Our limited number of each type of mass prevents making analysis to find possible differences in ADC values of different types of malignant and benign mass lesions from one another. Further studies enrolling larger populations are needed to document the role of this modality in the demonstration of an individual ADC value to specify the lesions. Second, we analyzed only two b values, one low and one high. Previous studies used more data points to let more precise calculation of ADC ratio [13,16]. However, repeated measurements with increased number of b values lengthens the examination times, creating problems of repeatability for sedated children. Third, most of our benign
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
lesions were predominantly cystic and higher ADC values, which made the distinction from the malignant ones easy. Further studies are recommended to assess the full value of DW imaging in benign solid pediatric masses. Finally, ROIs analyzed on MR images may not precisely represent to the biopsy specimens used in the pathologic examination. In conclusion, DW imaging can be used for reliable discrimination of benign and malignant pediatric abdominal mass lesions based on considerable differences in the ADC values and signal intensity changes. This technique can be added to routine imaging protocols without a considerable increase in the study time, and does not necessitate the administration of intravenous contrast media. References [1] Le Bihan DJ. Differentiation of benign versus pathologic compression fractures with diffusion-weighted MR imaging: a closer step toward the ‘holy grail’ of tissue characterization? Radiology 1998;207:305–7. [2] Szafer A, Zhong J, Gore JC. Theoretical model for water diffusion in tissues. Magn Reson Med 1995;33:697–712. [3] Lang P, Johnston JO, Arenal-Romero F, Gooding CA. Advances in MR imaging of pediatric musculoskeletal neoplasms. Magn Reson Imaging Clin N Am 1998;6:579–604. [4] Szafer A, Zhong J, Anderson AW, Gore JC. Diffusion-weighted imaging in tissues: theoretical models. NMR Biomed 1995;8:289–96. [5] Le Bihan D, Delannoy J, Levin RL. Temperature mapping with MR imaging of molecular diffusion: application to hyperthermia. Radiology 1989;171:853–7. [6] Le Bihan D, Breton E, Lallemand D, Grenier P, Cabanis E, LavalJeantet M. MR imaging of intravoxel incoherent motions: application to diffusion and perfusion in neurologic disorders. Radiology 1986;161:401–7. [7] Tien RD, Felsberg GJ, Friedman H, Brown M, MacFall J. MR imaging of high-grade cerebral gliomas: value of diffusion weighted echoplanar pulse sequences. AJR Am J Roentgenol 1994;162:671–7. [8] Eastwood JD, Lev MH, Wintermark M, Fitzek C, Barboriak DP, Delong DM, et al. Correlation of early dynamic CT perfusion imaging with whole-brain MR diffusion and perfusion imaging in acute hemispheric stroke. AJNR Am J Neuroradiol 2003;24:1869–75. [9] Yoshikawa K, Nakata Y, Yamada K, Nakagawa M. Early pathological changes in the parkinsonian brain demonstrated by diffusion tensor MRI. J Neurol Neurosurg Psychiatry 2004;75:481–4. [10] Parikh T, Drew SJ, Lee VS, Wong S, Hecht EM, Babb JS, et al. Focal liver lesion detection and characterization with diffusion-weighted MR imaging: comparison with standard breath-hold T2-weighted imaging. Radiology 2008;246:812–22. [11] Namimoto T, Yamashita Y, Sumi S, Tang Y, Takahashi M. Focal liver masses: characterization with diffusion-weighted echo-planar MR imaging. Radiology 1997;204:739–44. [12] Kilickesmez O, Yirik G, Bayramoglu S, Cimilli T, Aydin S. Nonbreath-hold high b-value diffusion-weighted MRI with parallel imaging technique: apparent diffusion coefficient determination in normal abdominal organs. Diagn Interv Radiol 2008;14:83–7. [13] Ichikawa T, Haradome H, Hachiya J, Nitatori T, Araki T. Diffusionweighted MR imaging with a single-shot echoplanar sequence: detection and characterization of focal hepatic lesions. AJR Am J Roentgenol 1998;170:397–402. [14] Demir OI, Obuz F, Sagol O, Dicle O. Contribution of diffusionweighted MRI to the differential diagnosis of hepatic masses. Diagn Interv Radiol 2007;13:81–6. [15] Yamada I, Aung W, Himeno Y, Nakagawa T, Shibuya H. Diffusion coefficients in abdominal organs and hepatic lesions: evaluation with
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
635
intravoxel incoherent motion echo-planar MR imaging. Radiology 1999;210:617–23. Humphries PD, Sebire NJ, Siegel MJ, Olsen OE. Tumors in pediatric patients at diffusion-weighted MR imaging: apparent diffusion coefficient and tumor cellularity. Radiology 2007;245:848–54. Olsen OE, Sebire NJ. Apparent diffusion coefficient maps of pediatric mass lesions with free-breathing diffusion-weighted magnetic resonance: feasibility study. Acta Radiol 2006;47:198–204. Uhl M, Altehoefer C, Kontny U, Il'yasov K, Buchert M, Langer M. MRI-diffusion imaging of neuroblastomas: first results and correlation to histology. Eur Radiol 2002;12:2335–8. Alibek S, Cavallaro A, Aplas A, Uder M, Staatz G. Diffusion weighted imaging of pediatric and adolescent malignancies with regard to detection and delineation: initial experience. Acad Radiol 2009;16: 866–71. Abdel Razek AA, Gaballa G, Elhawarey G, Megahed AS, Hafez M, Nada N. Characterization of pediatric head and neck masses with diffusion-weighted MR imaging. Eur Radiol 2009;19:201–8. Kim T, Murakami T, Takahashi S, Hori M, Tsuda K, Nakamura H. Diffusion-weighted single-shot echoplanar MR imaging for liver disease. AJR Am J Roentgenol 1999;173:393–8. Lyng H, Haraldseth O, Rofstad EK. Measurement of cell density and necrotic fraction in human melanoma xenografts by diffusion weighted magnetic resonance imaging. Magn Reson Med 2000;43: 828–36. Thoeny HC, De Keyzer F, Chen F, Ni Y, Landuyt W, Verbeken EK, et al. Diffusion-weighted MR imaging in monitoring the effect of a vascular targeting agent on rhabdomyosarcoma in rats. Radiology 2005;234:756–64. Koh DM, Scurr E, Collins D, Kanber B, Norman A, Leach MO, et al. Predicting response of colorectal hepatic metastasis: value of pretreatment apparent diffusion coefficients. AJR Am J Roentgenol 2007;188:1001–8. Guo Y, Cai YQ, Cai ZL, Gao YG, An NY, Ma L, et al. Differentiation of clinically benign and malignant breast lesions using diffusionweighted imaging. J Magn Reson Imaging 2002;16:172–8. Gauvain KM, McKinstry RC, Mukherjee P, Perry A, Neil JJ, Kaufman BA, et al. Evaluating pediatric brain tumor cellularity with diffusiontensor imaging. AJR Am J Roentgenol 2001;177:449–54. Sugahara T, Korogi Y, Kochi M, Ikushima I, Shigematu Y, Hirai T, et al. Usefulness of diffusion-weighted MRI with echo-planar technique in the evaluation of cellularity in gliomas. J Magn Reson Imaging 1999;9:53–60. Lang P, Wendland MF, Saeed M, Gindele A, Rosenau W, Mathur A, et al. Osteogenic sarcoma: noninvasive in vivo assessment of tumor necrosis with diffusion-weighted MR imaging. Radiology 1998;206: 227–35. Norris DG. The effects of microscopic tissue parameters on the diffusion weighted magnetic resonance imaging experiment. NMR Biomed 2001;14:77–93. Kotsenas AL, Roth TC, Maness WK, Faerber EN. Abnormal diffusionweighted MRI in a medulloblastoma: does it reflect small cell histology? Pediatr Radiol 1999;29:524–6. Okamoto K, Ito J, Ishikawa K, Sakai K, Tokiguchi S. Diffusionweighted echo-planar MR imaging in differential diagnosis of brain tumors and tumor-like conditions. Eur Radiol 2000;10:1342–50. Chan JH, Tsui EY, Luk SH, Fung AS, Yuen MK, Szeto ML, et al. Diffusion-weighted MR imaging of the liver: distinguishing hepatic abscess from cystic or necrotic tumor. Abdom Imaging 2000;26: 161–5. Eastwood JD, Engelter ST, MacFall JF, Delong DM, Provenzale JM. Quantitative assessment of the time course of infarct signal intensity on diffusion-weighted images. AJNR Am J Neuroradiol 2003;24: 680–7. Hakyemez B, Aksoy U, Yildiz H, Ergin N. Intracranial epidermoid cysts: diffusion-weighted, FLAIR and conventional MR findings. Eur J Radiol 2005;54:214–20.
636
M. Kocaoglu et al. / Magnetic Resonance Imaging 28 (2010) 629–636
[35] Colagrande S, Belli G, Politi LS, Mannelli L, Pasquinelli F, Villari N. The influence of diffusion- and relaxation-related factors on signal intensity: an introductive guide to magnetic resonance diffusionweighted imaging studies. J Comput Assist Tomogr 2008;32:463–74. [36] Naganawa S, Kawai H, Fukatsu H, Sakurai Y, Aoki I, Miura S, Mimura T, et al. Diffusion-weighted imaging of the liver: technical
challenges and prospects for the future. Magn Reson Med Sci 2005;4:175–86. [37] Ebisu T, Tanaka C, Umeda M, Kitamura M, Naruse S, Higuchi T, et al. Discrimination of brain abscess from necrotic or cystic tumors by diffusion-weighted echo planar imaging. Magn Reson Imaging 1996;14:1113–6.