- Email: [email protected]
Contrast-enhanced imaging features and differentiation of benign and malignant focal splenic lesions
Clinical Imaging 49 (2018) 58–64
Contents lists available at ScienceDirect
Clinical Imaging journal homepage: www.elsevier.com/locate/clinimag
Contrast-enhanced imaging features and differentiation of benign and malignant focal splenic lesions Fang Caoa, Wei Qianb, Yanqing Maa, Yinbo Wua, Jianguo Zhonga, a b
MARK
⁎
Department of Radiology, Zhejiang Provincial People's Hospital, People's Hospital of Hangzhou medical College, Hangzhou, Zhejiang, China Department of Radiology, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou, Zhejiang, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Spleen Computed tomography Magnetic resonance imaging Mass
To assess the value of imaging features for differentiating malignant from benign focal splenic lesions, 79 pathologically proved cases with contrast-enhanced CT or MRI were retrospectively studied. The morphological characteristics were assessed and the enhancement patterns were classified into five categories. After multivariate logistic analysis, the lesion margin and enhancement patterns were significantly different between benign and malignant lesions. The combination of ill-defined margin and hypovascular enhancement for suggesting malignant lesions had a good specificity (94.9%) and accuracy (89.9%). Morphological and enhancement characteristics on CT/MRI may be valuable in differentiating malignant from benign focal splenic lesions.
1. Introduction The conspicuity of atraumatic splenic lesions has been increased with the development in imaging technology, however, it is still difficult to differentiate malignant from benign focal splenic lesions. The incidence of focal lesions within spleen is low, and most of them are incidental findings on abdominal imaging examinations for unrelated diseases, which lead to the non-specificity of clinical symptoms. Besides, the classical imaging features of some diseases are often absent in splenic lesions. For example, the typical nodular enhancement with centripetal filling of hepatic hemangiomas is uncommon in splenic hemangiomas [1–3]. Therefore, splenic biopsy or splenectomy is often considered for the histopathological confirmation of a splenic mass, especially in those with extrasplenic malignancies. However, invasive procedure of the spleen may have some severe complications such as bleeding, pneumothorax and even immune function impairment. As a result, it is important to distinguish between malignant and benign splenic lesions noninvasively on imaging, and avoid unnecessary splenectomy in benign pathologies. Beyond case reports and description of single disease entity, there have been only a few studies about imaging differentiation of malignant and benign splenic lesions in the literature up to now [4–11], most of which were ultrasound studies [4–8]. Although CT and MRI were major imaging modalities in abdomen, studies on them were rare, and no recognized consensus criteria were reached. In this study, we investigated and compared the contrast-enhanced
⁎
CT and/or MRI findings of focal splenic lesions. Our purpose was to extract the imaging features valuable for differentiating malignant from benign focal splenic lesions. 2. Materials and methods 2.1. Study population The reporting databases for CT and MRI examinations of two large general hospitals between July 2011 and June 2016 were reviewed and searched for reports that mentioned splenic focal lesions or masses. Only patients who had a histopathological examination (biopsy or partial/total splenectomy) of the splenic lesions within 3 months after enhanced CT/MR scan were included in the study. Simple cystic lesions that had a homogeneous CT value of 0-20HU or isointense to cerebrospinal fluid on MRI without septa/thickened walls were excluded, since these lesions were almost all benign revealed in previous literature [10,12]. 2.2. Image examination CT imaging was performed using a SOMATOM Definition AS 40 helical scanner or a SOMATOM Sensation 16 helical scanner (Siemens Healthcare, Erlangen, Germany). The scanning range was from the right diaphragmatic dome to the inferior pole of kidneys. The scanning parameters were as follows: 120 kV, 180–210 mAs, 300–400 mm field
Corresponding author at: Department of Radiology, Zhejiang Provincial People's Hospital, People's Hospital of Hangzhou medical College, Hangzhou, Zhejiang 310014, China. E-mail address: [email protected] (J. Zhong).
https://doi.org/10.1016/j.clinimag.2017.10.020 Received 12 January 2017; Received in revised form 16 August 2017; Accepted 30 October 2017 0899-7071/ © 2017 Elsevier Inc. All rights reserved.
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
of view (FOV), and 5-mm slice thickness reconstruction. The arterialphase and portal-phase images were obtained 25–30 s and 60–70 s after the intravenous bolus (3 ml/s) administration of 1.0 ml/kg of an iodinated nonionic contrast agent, iopromide (Ultravist, 370 mgI/ml; Bayer Schering, Berlin, Germany). Some patients also had a 120–150 s equilibrium-phase scan. MRI was performed on a GE Discovery MR750 3.0 T MR scanner (GE Healthcare, Little Chalfont, UK) or a Siemens Magnetom Avanto 1.5 T MR scanner (Siemens Healthcare, Erlangen, Germany). The scan parameters of 3.0 T MRI were as follows: 1) T2-weighted fast spin echo (FSE) sequence: respiratory-triggered; repetition time (TR), 2 respiratory cycles; echo time (TE), 91 ms; flip angle (FA), 110°; slice thickness, 5.0 mm; FOV 320–400 mm. 2) T1-weighted FSE sequence: TR/TE, 255/3.6 ms; FA, 80°; slice thickness, 5.0 mm; FOV, 320–400 mm. 3) contrast-enhanced liver acquisition with 3D volume acceleration (LAVA) sequence: TR/TE, 4.2/1.9 ms; FA, 12°; slice thickness, 3.0 mm; FOV, 320–400 mm. The scan parameters of 1.5 T MRI: 1) T2-weighted FSE sequence: TR, 1 respiratory cycle; TE, 87 ms; FA, 132°; slice thickness, 6.0 mm; FOV, 320–400 mm. 2) T1-weighted in- and opposed-phase sequence: TR/TE 160/2.2 ms(in-), 160/4.9 ms (opp-); FA, 70°; slice thickness, 6.0 mm; FOV 320–400 mm. 3) contrastenhanced volumetric interpolated breath-hold examination (VIBE) sequence: TR/TE, 4.7/2.3 ms; FA, 10°; slice thickness, 3.0 mm; FOV, 320–400 mm. An intravenous dose of 0.1 mmol/kg of contrast agent (Gadolinium-diethylenetriamine pentaacetic acid, Gd-DTPA; Magnevist; Bayer Schering, Berlin, Germany) was administered, and the arterial-, portal-, equilibrium- (if one existed) phase images were obtained after 30 s, 60 s, 130 s after injection.
2.4. Statistical analysis The statistical analysis was performed with SPSS 19.0 software (IBM Corp., Armonk, NY, USA). Inter-observer agreement was calculated by using the kappa coefficient. For qualitative comparisons between malignant and benign lesions, the χ2 test or Fisher exact test was used. The Mann-Whitney test was used for continuous variables. Multivariate logistic regression analysis was performed on statistically significant imaging features in univariate analysis. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of significant variables were then calculated. A P value < 0.05 was considered statistically significant. 3. Results A total of 79 patients were finally identified, including 35 males and 44 females, with a mean age of 50 years (range, 19–80 years). Seventyseven cases had partial/total splenectomy, and the other two patients underwent ultrasound-guided biopsy. There were 59 benign cases (24 males and 35 females; mean age, 48 years; range, 19–80 years) and 20 malignant lesions (11 males and 9 females; mean age, 56 years; range, 40–80 years). The histopathological results of all the cases were listed in Table 1. Among all the 79 patients, 70 cases received CT examinations (25 with triple-phase enhancement scanning and 45 with dual-phase enhancement scanning), and 24 cases received MRI (9 with triple-phase and 15 with dual-phase enhancement scanning). The mean delay between histopathological examination and CT/MR scan was 15 days (range, 2–64 days).
2.3. Image assessment
3.1. Imaging findings of malignant and benign focal splenic lesions
Two radiologists with > 5 years of experience in diagnostic imaging, who were blinded to the clinical information and pathological results, assessed CT and MRI images. CT images of all the patients were reviewed first, and MR images were evaluated a month later. For patients with multiple, similar coexisting focal lesions, image evaluation were focused on the largest lesion. If there were any inconsistencies, a consensus would be achieved through discussion by a group consisting of three observers (the above two observers and a third radiologist with 20 years of experience in abdominal imaging). The enhancement patterns of lesions on dynamic enhanced imaging were classified as follows: 1) no enhancement; 2) cysts with septa/wall enhancement, but without enhancement of nodular or mass-like soft tissue; 3) hypovascular enhancement of solid part (mild enhancement, and hypo-density/signal intensity (SI) compared to the normal spleen throughout all phases of enhancement); 4) delayed hypervascular enhancement of solid part (hypo-density/SI compared to spleen in arterial-phase, with partial or entire iso-/hyper-enhancement in portaland/or equilibrium-phase); 5) arterial hypervascular enhancement (partial/entire lesion showing iso-/hyper-enhancement in arterialphase, with persistent enhancement or washout in portal-/equilibriumphase). Other imaging features of each case were recorded: number of lesions (single, multiple), size of the largest lesion (maximum diameter), margin (smooth and well-defined, rough and ill-defined; for a lesion with cystic portion, the inner surface of the cyst was also included into assessment), the presence of splenomegaly (maximum width, length, or thickness at the splenic hilum > 12 cm), calcification (only assessed on CT images), hemorrhage (only on MR images, except hemosiderin deposition spots) within lesion, perilesional splenic infarction, and perisplenic fluid. Some extrasplenic findings, such as intraperitoneal/retroperitoneal lymphadenopathy and lesions with similar imaging features in other abdominal organs, were also assessed. Lymphadenopathy was defined as a lymph node with the shortest diameter ≥ 1.0 cm and homogeneous or heterogeneous enhancement.
The comparison of imaging findings between malignant and benign focal splenic lesions were conducted, and the results were described in Table 2. The inter-observer agreement of all the imaging features was almost perfect (all the kappa coefficients > 0.8). In univariate analysis, there were significant differences in lesion margin, enhancement patterns, the presence of perilesional splenic infarction and splenomegaly between malignant and benign cases (P < 0.05). Compared with benign lesions, focal malignancies more frequently showed ill-defined margin, coexisting splenic infarction and splenomegaly (Fig.1). The most common enhancement pattern of malignancies was hypovascular Table 1 Included cases with focal splenic lesions. Histopathological diagnosis Benign Hemangioma Cyst Hemolymphangioma Lymphangioma SANT Littoral cell angioma Granulomatous disease Abscess Hamartoma Malignant Metastases Colorectal carcinoma Ovarian carcinoma Hepatocellular carcinoma Renal clear cell carcinoma Colon cancer direct invasion Lymphoma Angiosarcoma Carcinosarcoma IPT-like-FDCT Total
59
No. of cases
16 10 7 6 6 5 4 3 2
5 2 1 1 1 7 1 1 1 79
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
Table 2 Imaging findings of malignant and benign focal splenic lesions. Imaging findings Lesion number Single Multiple Margin Well-defined Ill-defined Lesion size (mm) Enhancement pattern No enhancement Arterial hypervascular Hypervascular delayed Hypovascular Septa/wall enhancement Perilesional splenic infarction With Without Splenomegaly With Without Hemorrhage⁎ With Without Peri-splenic fluid With Without Calcification⁎ With Without Abdominal lymphadenopathy With Without Concomitant lesions in other abdominal organ With Without ⁎
Malignant (n = 20)
Benign (n = 59)
12 8
48 11
3 17 62.75 ± 38.57
39 20 61.85 ± 34.25
0 0 1 17 2
5 11 15 13 15
P 0.071
< 0.001
0.960 < 0.001
0.013 4 16
1 58
12 8
17 42
1 2
5 16
2 18
2 57
1 18
15 36
0.017
1.000
0.264
0.052
0.171 6 14
8 51 0.220
4 16
5 54
Evaluation of hemorrhage on MR images, and calcification on CT images.
enhancement in solid portions (17/20, 85.0%), and a minority presented thickened wall/septa enhancement or delayed hypervascular enhancement. While enhancement patterns of benign masses were various. The patterns of arterial hypervascular enhancement and no enhancement appeared only in benign cases (Figs.2 and 3). The percentage of hypovascular enhancement was lower in benign group than that in malignant ones (13/59 vs 17/20, P = 0.002). No significant differences were found in lesion number, lesion size, presence of hemorrhage, peri-splenic fluid, calcification or extrasplenic findings between malignant and benign group.
Fig. 1. A 40-year-old male, B-cell lymphoma. CT images showed a cystic-solid mass in splenomegaly with irregular inner wall and ill-defined margin. The peripheral solid portion of lesion had a hypovascular enhancing pattern (A: unenhanced, B: arterial-phase, C: portal-phase).
3.2. Multivariate analysis of imaging features between malignant and benign lesions
enhancement, two cases of metastases or invasion from colon cancer appearing as complicated cystic lesions with thickened wall enhancement, and the other two cases (lymphoma and metastases from renal cell carcinoma, respectively) appearing as hypovascular-enhanced masses with well-defined margin (Fig.4). The three false-positive cases included two vascular lesions (one cavernous hemangioma and one hemolymphangioma) and one granulomatous disease, all of which manifested as ill-defined hypovascular masses (Fig.5), with cystic components in the case of hemolymphangioma.
Imaging features that suggesting splenic malignancies (ill-defined margin, perilesional infarction, splenomegaly and hypovascular enhancement pattern) were included in multivariate logistic regression analysis. Ill-defined margin [odds ratio (OR) = 10.567; 95% confidence interval (CI): 2.046, 54.568; P = 0.005] and hypovascular enhancement pattern [OR = 26.446; 95%CI: 4.584, 152.561; P < 0.001] were statistically significant after multivariate analysis. When combining these two findings as a predictor for splenic malignant lesions, the sensitivity, specificity, PPV, NPV and accuracy were 15/20 (75.0%), 56/59 (94.9%), 15/18 (83.3%), 56/61 (91.8%), 71/79 (89.9%), respectively. There were five false negative patients: one case of inflammatory pseudotumor-like follicular dendritic cell tumor (IPT-like-FDCT) manifesting as a well-defined mass with delayed hypervascular
3.3. MRI findings of focal splenic lesions There were 24 patients received MRI scan; among them, 15 patients underwent both CT and MRI. Morphological and enhancement characteristics on MRI of each case were the same as that on CT. 60
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
Fig. 2. A 37-year-old male, arteriovenous hemangioma. On T2WI (A), the mass was mainly hyperintense with hypointense foci. The lesion presented an arterial hypervascular enhancement with well-defined margin (B: MR arterial-phase, C: MR portal-phase, D: CT arterial-phase).
features in distinguishing between malignant and benign splenic lesions [4–11], most of which were ultrasound studies [4–8]. Goerg et al. [4] evaluated a variety of echo patterns in splenic lesions, and found that a differential diagnosis of splenic diseases was often impossible without contributory clinical data. Then some contrast-enhanced ultrasound (CEUS) studies [5–7] showed that hypovascular enhancement on CEUS might suggest malignant lesions, while hypervascular lesions were usually benign. Mainentia et al. [10] classified the portal-phase enhancement features on CT into 4 patterns for differentiating splenic diseases, however, they included a subset of splenomegaly cases who had no focal splenic lesions on images (14/47, 29.8%), leading to a higher proportion of malignant cases in their series (25/47, 53.2%) than other studies. Jang et al. [11] observed 53 cases of pathological or clinical diagnosed splenic lesions, and suggested that almost all the splenic malignancies (10/11, 90.9%) presented hypointense in triplephase enhancement. In our series of focal splenic lesions, we revealed that the most common enhancement pattern of malignancies was hypovascular enhancement, and the patterns of arterial hypervascular enhancement and no enhancement appeared only in benign cases. Combining the two findings of enhancement and margin demonstrated a good specificity, NPV and accuracy in predicting splenic malignant lesions. It was important to note, however, there were a few false-negative and falsepositive cases. The complicated cystic lesions with wall/septa enhancement and hypovascular-enhanced lesions with clear margin were the two major imaging patterns of false-negative cases, while the only imaging pattern of false-positive cases was ill-defined masses with hypovascular enhancement. A complicated cystic lesion could mostly be a benign lesion such as complicated cyst, abscess and lymphangioma, and infrequently be an extensively necrotic or cystic metastasis. In these cases, the presence of enhancing mural nodule or irregular inner surface of cyst was important in indicating malignancies, although it may be difficult when the lesion was small and solitary. In our five false-
Five cases were complicated cystic lesions with abnormal or mixed signals on MRI. Hemangioma (n = 9), hemolymphangioma (n = 2), lymphangioma (n = 1) and littoral cell angioma (n = 1) all showed slightly hypointense or isointense on T1WI and hyperintense on T2WI compared with spleen, occasionally containing a few low-signal spots or patches on T2WI. However, sclerosing angiomatoid nodular transformation (SANT) (n = 3) presented diffuse low or mixed low-high signals on T2WI (Fig. 6). Three malignant cases on MRI (1 lymphoma and 2 metastases cases) were all hyperintense on T2WI and hypo- or iso-intense on T1WI, which were the same as most of the benign vascular lesions. 4. Discussion In this study, we retrospectively evaluated cases of histopathologically confirmed focal splenic lesions in 2 large general hospitals in the past 5 years and suggested that among CT/MRI findings, ill-defined margin and hypovascular enhancement were statistically significant for distinguishing malignant lesions from benign ones. The combination of these two imaging findings had a specificity of 94.9% and an accuracy of 89.9% for suggesting malignant focal splenic masses. Several previous studies revealed that the majority of splenic masses were benign, while malignancies accounted for 20–35% (except simple cystic lesions) [11–13], which agreed with our result (20/79, 25.3%). Some literature suggested that certain morphological appearances on images, such as ill-defined lesion margin and presence of solid, enhancing components in the cystic lesions must draw the attention of radiologists to a potential malignancy [14]. In our study, we also indicated ill-defined margin as a significant predictor of focal malignant lesions, furthermore, we demonstrated that the enhancement patterns of solid components could make important sense in differential diagnosis between benign and malignant focal lesions. There were only a few studies investigating the role of imaging 61
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
Fig. 3. A 27-year-old female, true splenic cyst. There was a low-signal nodule on T2WI (A) in the cyst with well-defined margin. T1WI (B) showed hyperintense fluid in the cyst without enhancement (C, MR portal-phase). Intraoperative finding revealed dark brown cystic fluid with a yellow, soft nodule (no connection with the cystic wall).
Fig. 4. A 59-year-old female, diffuse large B-cell lymphoma. CT images (A: unenhanced, B: portal-phase, C: equilibrium-phase) presented a well-defined solid splenic mass with hypovascular enhancement.
they may contain hypo-enhancing regions, even in delayed phase [3,17]. In these confusing cases, some other imaging findings of the spleen could provide additional clues for differential diagnosis. As mentioned above, perilesional infarction and splenomegaly tended to present in malignancies, while the presence of calcification may mainly suggest benign lesions. In our study, the only benign case with splenic infarction was a patient with an abscess secondary to previous splenic partial infarction. Some previous literature mentioned that a series of extrasplenic findings, such as retroperitoneal/peritoneal lymphadenopathy and widespread similar lesions in other organs might also help the diagnosis. In our study, we found that malignant cases were more likely to show these extrasplenic findings; however, there was no statistical significance. Benign diseases, such as granulomatous diseases, abscess, hemangiomas and hamartomas, could also have concomitant lesions and/or lymphadenopathy. There were only a few researches focusing on MRI findings of focal
negative cases, IPT-like-FDCT was the only solid case that showed neither hypovascular enhancement nor ill-defined margin. IPT-likeFDCT was extremely rare, and only some case reports mentioned its imaging features [15]. In the literature, IPT-like-FDCT was depicted as a well-defined mass with delayed enhancement, and capsular-like rim on enhanced MRI might aid its diagnosis [15]. However, the rim enhancement on CT in our case was not conspicuous. Hypovascular enhancement was the main enhancement pattern of splenic malignancies, however, some benign lesions could present as hypovascular enhancing masses, mainly granulomatous diseases and vascular lesions. As described in previous literature, most granulomatous disease, such as tuberculosis granuloma, presented low-attenuation lesions with mild or no enhancement [16]. Splenic vascular lesions, especially cavernous hemangiomas, usually developed a range of secondary changes, such as hemorrhage, necrosis, micro-thrombosis and fibrosis, and as a result, 62
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
Fig. 6. A 49-year-old male, SANT. A well-defined solid mass presented as diffuse mixed low and slightly high SI on T2WI (A). The enhanced images showed a few arterial hypervascular-enhanced spots and lines peripherally (B) with centripetal progression on portal-phase (C).
Fig. 5. A 57-year-old female, cavernous hemangioma. On CT images (A: unenhanced, B: arterial-phase, C: portal-phase), the splenic mass with rough and ill-defined margin, presented heterogeneous hypovascular enhancement in portal-phase image.
splenic masses. In our series, signal intensities on T1WI and T2WI have no significant differential diagnosis value between benign and malignant lesions; however, signal features on T2WI may be valuable for differentiation among splenic vascular masses. The predominant and extensive hypointense of SANT on T2WI (Fig. 6) was special and might reflect its pathology of diffuse interstitial fibrotic bands surrounding angiomatoid nodules with hemosiderin deposition [18]. While other vascular masses, such as hemangioma and lymphangioma, occasionally showed relative localized hypointense foci within the hyperintense lesions, which was usually due to focal hemosiderin deposition or calcification. There were several limitations in this study. Firstly, due to the retrospective design of the study, we could not get the complete clinical information and follow-up data of all the patients, which might have suggestive values for diagnosis. Meanwhile, scanning was performed on different CT and MRI scanners, and the multi-phase enhanced CT/MRI
was either dual- or triple-phase scanning, which might have influences on imaging evaluation. Although we founded that no change on enhancement pattern was made by the equilibrium-phase scanning in any case, compared with dual-phase enhancing images. Besides, to our knowledge, this study included a larger sample size than most previous imaging studies on focal splenic masses; however, the size of malignant group was still relatively small due to the low incidence of focal splenic malignant tumors. Finally, the number of cases with MRI examinations was small, and only 15 patients had both CT and MRI scan. Therefore, further prospective studies of larger sample size with combination of clinical characteristics and imaging features should be conducted. In conclusion, our study showed that the morphological and enhancement characteristics on contrast-enhanced CT/MRI might be valuable for distinguishing malignant and benign focal splenic lesions. The combination of ill-defined lesion margin and hypovascular 63
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
enhancement pattern had a good specificity and accuracy in predicting malignant splenic lesions, which could help avoiding unnecessary splenectomy in some cases with benign pathologies.
splenic masses. Semin Ultrasound CT 2006;27:420–5. [10] Mainentia PP, Iodiceb D, Cozzolinoc I, Segreto S, Capece S, Sica G, et al. Tomographic imaging of the spleen: the role of morphological and metabolic features in differentiating benign from malignant diseases. Clin Imaging 2012;36:559–67. [11] Jang KM, Kim SH, Hwang J, Lee SJ, Kang TW, Lee MW, et al. Differentiation of malignant from benign focal splenic lesions: added value of diffusion-weighted MRI. Am J Roentgenol 2014;203(4):803–12. [12] Ricci ZJ, Oh SK, Chernyak V, Flusberg M, Rozenblit AM, Kaul B, et al. Improving diagnosis of atraumatic splenic lesions, part I: nonneoplastic lesions. Clin Imaging 2016;40(4):769–79. [13] Dhyani M, Anupindi SA, Ayyala R, Hahn PF, Gee MS. Defining an imaging algorithm for noncystic splenic lesions identified in young patients. AJR Am J Roentgenol 2013;201(6):W893–9. [14] Karlo CA, Stolzmann P, Do RK, Alkadhi H. Computed tomography of the spleen: how to interpret the hypodense lesion. Insights Imaging 2013;4(1):65–76. [15] Kitamura Y, Takayama Y, Nishie A, Asayama Y, Ushijima Y, Fujita N, et al. Inflammatory pseudotumor-like follicular dendritic cell tumor of the spleen: case report and review of the literature. Magn Reson Med Sci 2015;14(4):347–54. [16] Harisinghani MG, McLoud TC, Shepard JA, Ko JP, Shroff MM, Mueller PR. Tuberculosis from head to toe. Radiographics 2000;20(2):449–70. [17] Ozcan HN, Oguz B, Talim B, Ekinci S, Haliloglu M. Unusual splenic hemangioma of a pediatric patient: hypointense on T2-weighted image. Clin Imaging 2014;38(4):553–5. [18] Thacker C, Korn R, Millstine J, Harvin H, Van Lier Ribbink JA, Gotway MB. Sclerosing angiomatoid nodular transformation of the spleen: CT, MR, PET, and 99 (m)Tc-sulfur colloid SPECT CT findings with gross and histopathological correlation. Abdom Imaging 2010;35(6):683–9.
References [1] Ferrozzi F, Bova D, Draghi F, Garlaschi G. CT findings in primary vascular tumors of the spleen. Am J Roentgenol 1996;166:1097–101. [2] Mortele KJ. Imaging of tumoral conditions of the spleen. JBR-BTR 2000;83:213–5. [3] Abbott RM, Levy AD, Aguilera NS, Gorospe L, Thompson WM. From the archives of the AFIP: primary vascular neoplasma of the spleen: radiolgic-pathologic correlation. Radiographics 2004;24(4):1137–63. [4] Goerg C, Schwerk WB, Goerg K. Splenic lesions: sonographic patterns, follow up, differential diagnosis. Eur J Radiol 1991;13:59–66. [5] Stang A, Keles H, Hentschke S, von Seydewitz CU, Dahlke J, Malzfeldt E, et al. Differentiation of benign from malignant focal splenic lesions using sulfur hexafluoride–filled microbubble contrast-enhanced pulse-inversion sonography. Am J Roentgenol 2009;193:709–21. [6] Stang A, Keles H, Hentschke S, von Seydewitz CU, Dahlke J, Habermann C, et al. Incidentally detected splenic lesions in ultrasound: does contrast enhanced ultrasonography improve the differentiation of benign hemangioma/hamartoma from malignant lesions? Ultraschall Med 2011;32:582–92. [7] Yu X, Yu J, Liang P, Liu F. Real-time contrast-enhanced ultrasound in diagnosing of focal spleen lesions. Eur J Radiol 2012;81:430–6. [8] von Herbay A, Barreiros AP, Ignee A, Westendorff J, Gregor M, Galle PR, et al. Contrast-enhanced ultrasonography with SonoVue: differentiation between benign and malignant lesions of the spleen. J Ultrasound Med 2009;28:421–34. [9] Metser U, Even-Sapir E. The role of 18F-FDG PET/CT in the evaluation of solid
64
Contents lists available at ScienceDirect
Clinical Imaging journal homepage: www.elsevier.com/locate/clinimag
Contrast-enhanced imaging features and differentiation of benign and malignant focal splenic lesions Fang Caoa, Wei Qianb, Yanqing Maa, Yinbo Wua, Jianguo Zhonga, a b
MARK
⁎
Department of Radiology, Zhejiang Provincial People's Hospital, People's Hospital of Hangzhou medical College, Hangzhou, Zhejiang, China Department of Radiology, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou, Zhejiang, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Spleen Computed tomography Magnetic resonance imaging Mass
To assess the value of imaging features for differentiating malignant from benign focal splenic lesions, 79 pathologically proved cases with contrast-enhanced CT or MRI were retrospectively studied. The morphological characteristics were assessed and the enhancement patterns were classified into five categories. After multivariate logistic analysis, the lesion margin and enhancement patterns were significantly different between benign and malignant lesions. The combination of ill-defined margin and hypovascular enhancement for suggesting malignant lesions had a good specificity (94.9%) and accuracy (89.9%). Morphological and enhancement characteristics on CT/MRI may be valuable in differentiating malignant from benign focal splenic lesions.
1. Introduction The conspicuity of atraumatic splenic lesions has been increased with the development in imaging technology, however, it is still difficult to differentiate malignant from benign focal splenic lesions. The incidence of focal lesions within spleen is low, and most of them are incidental findings on abdominal imaging examinations for unrelated diseases, which lead to the non-specificity of clinical symptoms. Besides, the classical imaging features of some diseases are often absent in splenic lesions. For example, the typical nodular enhancement with centripetal filling of hepatic hemangiomas is uncommon in splenic hemangiomas [1–3]. Therefore, splenic biopsy or splenectomy is often considered for the histopathological confirmation of a splenic mass, especially in those with extrasplenic malignancies. However, invasive procedure of the spleen may have some severe complications such as bleeding, pneumothorax and even immune function impairment. As a result, it is important to distinguish between malignant and benign splenic lesions noninvasively on imaging, and avoid unnecessary splenectomy in benign pathologies. Beyond case reports and description of single disease entity, there have been only a few studies about imaging differentiation of malignant and benign splenic lesions in the literature up to now [4–11], most of which were ultrasound studies [4–8]. Although CT and MRI were major imaging modalities in abdomen, studies on them were rare, and no recognized consensus criteria were reached. In this study, we investigated and compared the contrast-enhanced
⁎
CT and/or MRI findings of focal splenic lesions. Our purpose was to extract the imaging features valuable for differentiating malignant from benign focal splenic lesions. 2. Materials and methods 2.1. Study population The reporting databases for CT and MRI examinations of two large general hospitals between July 2011 and June 2016 were reviewed and searched for reports that mentioned splenic focal lesions or masses. Only patients who had a histopathological examination (biopsy or partial/total splenectomy) of the splenic lesions within 3 months after enhanced CT/MR scan were included in the study. Simple cystic lesions that had a homogeneous CT value of 0-20HU or isointense to cerebrospinal fluid on MRI without septa/thickened walls were excluded, since these lesions were almost all benign revealed in previous literature [10,12]. 2.2. Image examination CT imaging was performed using a SOMATOM Definition AS 40 helical scanner or a SOMATOM Sensation 16 helical scanner (Siemens Healthcare, Erlangen, Germany). The scanning range was from the right diaphragmatic dome to the inferior pole of kidneys. The scanning parameters were as follows: 120 kV, 180–210 mAs, 300–400 mm field
Corresponding author at: Department of Radiology, Zhejiang Provincial People's Hospital, People's Hospital of Hangzhou medical College, Hangzhou, Zhejiang 310014, China. E-mail address: [email protected] (J. Zhong).
https://doi.org/10.1016/j.clinimag.2017.10.020 Received 12 January 2017; Received in revised form 16 August 2017; Accepted 30 October 2017 0899-7071/ © 2017 Elsevier Inc. All rights reserved.
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
of view (FOV), and 5-mm slice thickness reconstruction. The arterialphase and portal-phase images were obtained 25–30 s and 60–70 s after the intravenous bolus (3 ml/s) administration of 1.0 ml/kg of an iodinated nonionic contrast agent, iopromide (Ultravist, 370 mgI/ml; Bayer Schering, Berlin, Germany). Some patients also had a 120–150 s equilibrium-phase scan. MRI was performed on a GE Discovery MR750 3.0 T MR scanner (GE Healthcare, Little Chalfont, UK) or a Siemens Magnetom Avanto 1.5 T MR scanner (Siemens Healthcare, Erlangen, Germany). The scan parameters of 3.0 T MRI were as follows: 1) T2-weighted fast spin echo (FSE) sequence: respiratory-triggered; repetition time (TR), 2 respiratory cycles; echo time (TE), 91 ms; flip angle (FA), 110°; slice thickness, 5.0 mm; FOV 320–400 mm. 2) T1-weighted FSE sequence: TR/TE, 255/3.6 ms; FA, 80°; slice thickness, 5.0 mm; FOV, 320–400 mm. 3) contrast-enhanced liver acquisition with 3D volume acceleration (LAVA) sequence: TR/TE, 4.2/1.9 ms; FA, 12°; slice thickness, 3.0 mm; FOV, 320–400 mm. The scan parameters of 1.5 T MRI: 1) T2-weighted FSE sequence: TR, 1 respiratory cycle; TE, 87 ms; FA, 132°; slice thickness, 6.0 mm; FOV, 320–400 mm. 2) T1-weighted in- and opposed-phase sequence: TR/TE 160/2.2 ms(in-), 160/4.9 ms (opp-); FA, 70°; slice thickness, 6.0 mm; FOV 320–400 mm. 3) contrastenhanced volumetric interpolated breath-hold examination (VIBE) sequence: TR/TE, 4.7/2.3 ms; FA, 10°; slice thickness, 3.0 mm; FOV, 320–400 mm. An intravenous dose of 0.1 mmol/kg of contrast agent (Gadolinium-diethylenetriamine pentaacetic acid, Gd-DTPA; Magnevist; Bayer Schering, Berlin, Germany) was administered, and the arterial-, portal-, equilibrium- (if one existed) phase images were obtained after 30 s, 60 s, 130 s after injection.
2.4. Statistical analysis The statistical analysis was performed with SPSS 19.0 software (IBM Corp., Armonk, NY, USA). Inter-observer agreement was calculated by using the kappa coefficient. For qualitative comparisons between malignant and benign lesions, the χ2 test or Fisher exact test was used. The Mann-Whitney test was used for continuous variables. Multivariate logistic regression analysis was performed on statistically significant imaging features in univariate analysis. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy of significant variables were then calculated. A P value < 0.05 was considered statistically significant. 3. Results A total of 79 patients were finally identified, including 35 males and 44 females, with a mean age of 50 years (range, 19–80 years). Seventyseven cases had partial/total splenectomy, and the other two patients underwent ultrasound-guided biopsy. There were 59 benign cases (24 males and 35 females; mean age, 48 years; range, 19–80 years) and 20 malignant lesions (11 males and 9 females; mean age, 56 years; range, 40–80 years). The histopathological results of all the cases were listed in Table 1. Among all the 79 patients, 70 cases received CT examinations (25 with triple-phase enhancement scanning and 45 with dual-phase enhancement scanning), and 24 cases received MRI (9 with triple-phase and 15 with dual-phase enhancement scanning). The mean delay between histopathological examination and CT/MR scan was 15 days (range, 2–64 days).
2.3. Image assessment
3.1. Imaging findings of malignant and benign focal splenic lesions
Two radiologists with > 5 years of experience in diagnostic imaging, who were blinded to the clinical information and pathological results, assessed CT and MRI images. CT images of all the patients were reviewed first, and MR images were evaluated a month later. For patients with multiple, similar coexisting focal lesions, image evaluation were focused on the largest lesion. If there were any inconsistencies, a consensus would be achieved through discussion by a group consisting of three observers (the above two observers and a third radiologist with 20 years of experience in abdominal imaging). The enhancement patterns of lesions on dynamic enhanced imaging were classified as follows: 1) no enhancement; 2) cysts with septa/wall enhancement, but without enhancement of nodular or mass-like soft tissue; 3) hypovascular enhancement of solid part (mild enhancement, and hypo-density/signal intensity (SI) compared to the normal spleen throughout all phases of enhancement); 4) delayed hypervascular enhancement of solid part (hypo-density/SI compared to spleen in arterial-phase, with partial or entire iso-/hyper-enhancement in portaland/or equilibrium-phase); 5) arterial hypervascular enhancement (partial/entire lesion showing iso-/hyper-enhancement in arterialphase, with persistent enhancement or washout in portal-/equilibriumphase). Other imaging features of each case were recorded: number of lesions (single, multiple), size of the largest lesion (maximum diameter), margin (smooth and well-defined, rough and ill-defined; for a lesion with cystic portion, the inner surface of the cyst was also included into assessment), the presence of splenomegaly (maximum width, length, or thickness at the splenic hilum > 12 cm), calcification (only assessed on CT images), hemorrhage (only on MR images, except hemosiderin deposition spots) within lesion, perilesional splenic infarction, and perisplenic fluid. Some extrasplenic findings, such as intraperitoneal/retroperitoneal lymphadenopathy and lesions with similar imaging features in other abdominal organs, were also assessed. Lymphadenopathy was defined as a lymph node with the shortest diameter ≥ 1.0 cm and homogeneous or heterogeneous enhancement.
The comparison of imaging findings between malignant and benign focal splenic lesions were conducted, and the results were described in Table 2. The inter-observer agreement of all the imaging features was almost perfect (all the kappa coefficients > 0.8). In univariate analysis, there were significant differences in lesion margin, enhancement patterns, the presence of perilesional splenic infarction and splenomegaly between malignant and benign cases (P < 0.05). Compared with benign lesions, focal malignancies more frequently showed ill-defined margin, coexisting splenic infarction and splenomegaly (Fig.1). The most common enhancement pattern of malignancies was hypovascular Table 1 Included cases with focal splenic lesions. Histopathological diagnosis Benign Hemangioma Cyst Hemolymphangioma Lymphangioma SANT Littoral cell angioma Granulomatous disease Abscess Hamartoma Malignant Metastases Colorectal carcinoma Ovarian carcinoma Hepatocellular carcinoma Renal clear cell carcinoma Colon cancer direct invasion Lymphoma Angiosarcoma Carcinosarcoma IPT-like-FDCT Total
59
No. of cases
16 10 7 6 6 5 4 3 2
5 2 1 1 1 7 1 1 1 79
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
Table 2 Imaging findings of malignant and benign focal splenic lesions. Imaging findings Lesion number Single Multiple Margin Well-defined Ill-defined Lesion size (mm) Enhancement pattern No enhancement Arterial hypervascular Hypervascular delayed Hypovascular Septa/wall enhancement Perilesional splenic infarction With Without Splenomegaly With Without Hemorrhage⁎ With Without Peri-splenic fluid With Without Calcification⁎ With Without Abdominal lymphadenopathy With Without Concomitant lesions in other abdominal organ With Without ⁎
Malignant (n = 20)
Benign (n = 59)
12 8
48 11
3 17 62.75 ± 38.57
39 20 61.85 ± 34.25
0 0 1 17 2
5 11 15 13 15
P 0.071
< 0.001
0.960 < 0.001
0.013 4 16
1 58
12 8
17 42
1 2
5 16
2 18
2 57
1 18
15 36
0.017
1.000
0.264
0.052
0.171 6 14
8 51 0.220
4 16
5 54
Evaluation of hemorrhage on MR images, and calcification on CT images.
enhancement in solid portions (17/20, 85.0%), and a minority presented thickened wall/septa enhancement or delayed hypervascular enhancement. While enhancement patterns of benign masses were various. The patterns of arterial hypervascular enhancement and no enhancement appeared only in benign cases (Figs.2 and 3). The percentage of hypovascular enhancement was lower in benign group than that in malignant ones (13/59 vs 17/20, P = 0.002). No significant differences were found in lesion number, lesion size, presence of hemorrhage, peri-splenic fluid, calcification or extrasplenic findings between malignant and benign group.
Fig. 1. A 40-year-old male, B-cell lymphoma. CT images showed a cystic-solid mass in splenomegaly with irregular inner wall and ill-defined margin. The peripheral solid portion of lesion had a hypovascular enhancing pattern (A: unenhanced, B: arterial-phase, C: portal-phase).
3.2. Multivariate analysis of imaging features between malignant and benign lesions
enhancement, two cases of metastases or invasion from colon cancer appearing as complicated cystic lesions with thickened wall enhancement, and the other two cases (lymphoma and metastases from renal cell carcinoma, respectively) appearing as hypovascular-enhanced masses with well-defined margin (Fig.4). The three false-positive cases included two vascular lesions (one cavernous hemangioma and one hemolymphangioma) and one granulomatous disease, all of which manifested as ill-defined hypovascular masses (Fig.5), with cystic components in the case of hemolymphangioma.
Imaging features that suggesting splenic malignancies (ill-defined margin, perilesional infarction, splenomegaly and hypovascular enhancement pattern) were included in multivariate logistic regression analysis. Ill-defined margin [odds ratio (OR) = 10.567; 95% confidence interval (CI): 2.046, 54.568; P = 0.005] and hypovascular enhancement pattern [OR = 26.446; 95%CI: 4.584, 152.561; P < 0.001] were statistically significant after multivariate analysis. When combining these two findings as a predictor for splenic malignant lesions, the sensitivity, specificity, PPV, NPV and accuracy were 15/20 (75.0%), 56/59 (94.9%), 15/18 (83.3%), 56/61 (91.8%), 71/79 (89.9%), respectively. There were five false negative patients: one case of inflammatory pseudotumor-like follicular dendritic cell tumor (IPT-like-FDCT) manifesting as a well-defined mass with delayed hypervascular
3.3. MRI findings of focal splenic lesions There were 24 patients received MRI scan; among them, 15 patients underwent both CT and MRI. Morphological and enhancement characteristics on MRI of each case were the same as that on CT. 60
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
Fig. 2. A 37-year-old male, arteriovenous hemangioma. On T2WI (A), the mass was mainly hyperintense with hypointense foci. The lesion presented an arterial hypervascular enhancement with well-defined margin (B: MR arterial-phase, C: MR portal-phase, D: CT arterial-phase).
features in distinguishing between malignant and benign splenic lesions [4–11], most of which were ultrasound studies [4–8]. Goerg et al. [4] evaluated a variety of echo patterns in splenic lesions, and found that a differential diagnosis of splenic diseases was often impossible without contributory clinical data. Then some contrast-enhanced ultrasound (CEUS) studies [5–7] showed that hypovascular enhancement on CEUS might suggest malignant lesions, while hypervascular lesions were usually benign. Mainentia et al. [10] classified the portal-phase enhancement features on CT into 4 patterns for differentiating splenic diseases, however, they included a subset of splenomegaly cases who had no focal splenic lesions on images (14/47, 29.8%), leading to a higher proportion of malignant cases in their series (25/47, 53.2%) than other studies. Jang et al. [11] observed 53 cases of pathological or clinical diagnosed splenic lesions, and suggested that almost all the splenic malignancies (10/11, 90.9%) presented hypointense in triplephase enhancement. In our series of focal splenic lesions, we revealed that the most common enhancement pattern of malignancies was hypovascular enhancement, and the patterns of arterial hypervascular enhancement and no enhancement appeared only in benign cases. Combining the two findings of enhancement and margin demonstrated a good specificity, NPV and accuracy in predicting splenic malignant lesions. It was important to note, however, there were a few false-negative and falsepositive cases. The complicated cystic lesions with wall/septa enhancement and hypovascular-enhanced lesions with clear margin were the two major imaging patterns of false-negative cases, while the only imaging pattern of false-positive cases was ill-defined masses with hypovascular enhancement. A complicated cystic lesion could mostly be a benign lesion such as complicated cyst, abscess and lymphangioma, and infrequently be an extensively necrotic or cystic metastasis. In these cases, the presence of enhancing mural nodule or irregular inner surface of cyst was important in indicating malignancies, although it may be difficult when the lesion was small and solitary. In our five false-
Five cases were complicated cystic lesions with abnormal or mixed signals on MRI. Hemangioma (n = 9), hemolymphangioma (n = 2), lymphangioma (n = 1) and littoral cell angioma (n = 1) all showed slightly hypointense or isointense on T1WI and hyperintense on T2WI compared with spleen, occasionally containing a few low-signal spots or patches on T2WI. However, sclerosing angiomatoid nodular transformation (SANT) (n = 3) presented diffuse low or mixed low-high signals on T2WI (Fig. 6). Three malignant cases on MRI (1 lymphoma and 2 metastases cases) were all hyperintense on T2WI and hypo- or iso-intense on T1WI, which were the same as most of the benign vascular lesions. 4. Discussion In this study, we retrospectively evaluated cases of histopathologically confirmed focal splenic lesions in 2 large general hospitals in the past 5 years and suggested that among CT/MRI findings, ill-defined margin and hypovascular enhancement were statistically significant for distinguishing malignant lesions from benign ones. The combination of these two imaging findings had a specificity of 94.9% and an accuracy of 89.9% for suggesting malignant focal splenic masses. Several previous studies revealed that the majority of splenic masses were benign, while malignancies accounted for 20–35% (except simple cystic lesions) [11–13], which agreed with our result (20/79, 25.3%). Some literature suggested that certain morphological appearances on images, such as ill-defined lesion margin and presence of solid, enhancing components in the cystic lesions must draw the attention of radiologists to a potential malignancy [14]. In our study, we also indicated ill-defined margin as a significant predictor of focal malignant lesions, furthermore, we demonstrated that the enhancement patterns of solid components could make important sense in differential diagnosis between benign and malignant focal lesions. There were only a few studies investigating the role of imaging 61
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
Fig. 3. A 27-year-old female, true splenic cyst. There was a low-signal nodule on T2WI (A) in the cyst with well-defined margin. T1WI (B) showed hyperintense fluid in the cyst without enhancement (C, MR portal-phase). Intraoperative finding revealed dark brown cystic fluid with a yellow, soft nodule (no connection with the cystic wall).
Fig. 4. A 59-year-old female, diffuse large B-cell lymphoma. CT images (A: unenhanced, B: portal-phase, C: equilibrium-phase) presented a well-defined solid splenic mass with hypovascular enhancement.
they may contain hypo-enhancing regions, even in delayed phase [3,17]. In these confusing cases, some other imaging findings of the spleen could provide additional clues for differential diagnosis. As mentioned above, perilesional infarction and splenomegaly tended to present in malignancies, while the presence of calcification may mainly suggest benign lesions. In our study, the only benign case with splenic infarction was a patient with an abscess secondary to previous splenic partial infarction. Some previous literature mentioned that a series of extrasplenic findings, such as retroperitoneal/peritoneal lymphadenopathy and widespread similar lesions in other organs might also help the diagnosis. In our study, we found that malignant cases were more likely to show these extrasplenic findings; however, there was no statistical significance. Benign diseases, such as granulomatous diseases, abscess, hemangiomas and hamartomas, could also have concomitant lesions and/or lymphadenopathy. There were only a few researches focusing on MRI findings of focal
negative cases, IPT-like-FDCT was the only solid case that showed neither hypovascular enhancement nor ill-defined margin. IPT-likeFDCT was extremely rare, and only some case reports mentioned its imaging features [15]. In the literature, IPT-like-FDCT was depicted as a well-defined mass with delayed enhancement, and capsular-like rim on enhanced MRI might aid its diagnosis [15]. However, the rim enhancement on CT in our case was not conspicuous. Hypovascular enhancement was the main enhancement pattern of splenic malignancies, however, some benign lesions could present as hypovascular enhancing masses, mainly granulomatous diseases and vascular lesions. As described in previous literature, most granulomatous disease, such as tuberculosis granuloma, presented low-attenuation lesions with mild or no enhancement [16]. Splenic vascular lesions, especially cavernous hemangiomas, usually developed a range of secondary changes, such as hemorrhage, necrosis, micro-thrombosis and fibrosis, and as a result, 62
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
Fig. 6. A 49-year-old male, SANT. A well-defined solid mass presented as diffuse mixed low and slightly high SI on T2WI (A). The enhanced images showed a few arterial hypervascular-enhanced spots and lines peripherally (B) with centripetal progression on portal-phase (C).
Fig. 5. A 57-year-old female, cavernous hemangioma. On CT images (A: unenhanced, B: arterial-phase, C: portal-phase), the splenic mass with rough and ill-defined margin, presented heterogeneous hypovascular enhancement in portal-phase image.
splenic masses. In our series, signal intensities on T1WI and T2WI have no significant differential diagnosis value between benign and malignant lesions; however, signal features on T2WI may be valuable for differentiation among splenic vascular masses. The predominant and extensive hypointense of SANT on T2WI (Fig. 6) was special and might reflect its pathology of diffuse interstitial fibrotic bands surrounding angiomatoid nodules with hemosiderin deposition [18]. While other vascular masses, such as hemangioma and lymphangioma, occasionally showed relative localized hypointense foci within the hyperintense lesions, which was usually due to focal hemosiderin deposition or calcification. There were several limitations in this study. Firstly, due to the retrospective design of the study, we could not get the complete clinical information and follow-up data of all the patients, which might have suggestive values for diagnosis. Meanwhile, scanning was performed on different CT and MRI scanners, and the multi-phase enhanced CT/MRI
was either dual- or triple-phase scanning, which might have influences on imaging evaluation. Although we founded that no change on enhancement pattern was made by the equilibrium-phase scanning in any case, compared with dual-phase enhancing images. Besides, to our knowledge, this study included a larger sample size than most previous imaging studies on focal splenic masses; however, the size of malignant group was still relatively small due to the low incidence of focal splenic malignant tumors. Finally, the number of cases with MRI examinations was small, and only 15 patients had both CT and MRI scan. Therefore, further prospective studies of larger sample size with combination of clinical characteristics and imaging features should be conducted. In conclusion, our study showed that the morphological and enhancement characteristics on contrast-enhanced CT/MRI might be valuable for distinguishing malignant and benign focal splenic lesions. The combination of ill-defined lesion margin and hypovascular 63
Clinical Imaging 49 (2018) 58–64
F. Cao et al.
enhancement pattern had a good specificity and accuracy in predicting malignant splenic lesions, which could help avoiding unnecessary splenectomy in some cases with benign pathologies.
splenic masses. Semin Ultrasound CT 2006;27:420–5. [10] Mainentia PP, Iodiceb D, Cozzolinoc I, Segreto S, Capece S, Sica G, et al. Tomographic imaging of the spleen: the role of morphological and metabolic features in differentiating benign from malignant diseases. Clin Imaging 2012;36:559–67. [11] Jang KM, Kim SH, Hwang J, Lee SJ, Kang TW, Lee MW, et al. Differentiation of malignant from benign focal splenic lesions: added value of diffusion-weighted MRI. Am J Roentgenol 2014;203(4):803–12. [12] Ricci ZJ, Oh SK, Chernyak V, Flusberg M, Rozenblit AM, Kaul B, et al. Improving diagnosis of atraumatic splenic lesions, part I: nonneoplastic lesions. Clin Imaging 2016;40(4):769–79. [13] Dhyani M, Anupindi SA, Ayyala R, Hahn PF, Gee MS. Defining an imaging algorithm for noncystic splenic lesions identified in young patients. AJR Am J Roentgenol 2013;201(6):W893–9. [14] Karlo CA, Stolzmann P, Do RK, Alkadhi H. Computed tomography of the spleen: how to interpret the hypodense lesion. Insights Imaging 2013;4(1):65–76. [15] Kitamura Y, Takayama Y, Nishie A, Asayama Y, Ushijima Y, Fujita N, et al. Inflammatory pseudotumor-like follicular dendritic cell tumor of the spleen: case report and review of the literature. Magn Reson Med Sci 2015;14(4):347–54. [16] Harisinghani MG, McLoud TC, Shepard JA, Ko JP, Shroff MM, Mueller PR. Tuberculosis from head to toe. Radiographics 2000;20(2):449–70. [17] Ozcan HN, Oguz B, Talim B, Ekinci S, Haliloglu M. Unusual splenic hemangioma of a pediatric patient: hypointense on T2-weighted image. Clin Imaging 2014;38(4):553–5. [18] Thacker C, Korn R, Millstine J, Harvin H, Van Lier Ribbink JA, Gotway MB. Sclerosing angiomatoid nodular transformation of the spleen: CT, MR, PET, and 99 (m)Tc-sulfur colloid SPECT CT findings with gross and histopathological correlation. Abdom Imaging 2010;35(6):683–9.
References [1] Ferrozzi F, Bova D, Draghi F, Garlaschi G. CT findings in primary vascular tumors of the spleen. Am J Roentgenol 1996;166:1097–101. [2] Mortele KJ. Imaging of tumoral conditions of the spleen. JBR-BTR 2000;83:213–5. [3] Abbott RM, Levy AD, Aguilera NS, Gorospe L, Thompson WM. From the archives of the AFIP: primary vascular neoplasma of the spleen: radiolgic-pathologic correlation. Radiographics 2004;24(4):1137–63. [4] Goerg C, Schwerk WB, Goerg K. Splenic lesions: sonographic patterns, follow up, differential diagnosis. Eur J Radiol 1991;13:59–66. [5] Stang A, Keles H, Hentschke S, von Seydewitz CU, Dahlke J, Malzfeldt E, et al. Differentiation of benign from malignant focal splenic lesions using sulfur hexafluoride–filled microbubble contrast-enhanced pulse-inversion sonography. Am J Roentgenol 2009;193:709–21. [6] Stang A, Keles H, Hentschke S, von Seydewitz CU, Dahlke J, Habermann C, et al. Incidentally detected splenic lesions in ultrasound: does contrast enhanced ultrasonography improve the differentiation of benign hemangioma/hamartoma from malignant lesions? Ultraschall Med 2011;32:582–92. [7] Yu X, Yu J, Liang P, Liu F. Real-time contrast-enhanced ultrasound in diagnosing of focal spleen lesions. Eur J Radiol 2012;81:430–6. [8] von Herbay A, Barreiros AP, Ignee A, Westendorff J, Gregor M, Galle PR, et al. Contrast-enhanced ultrasonography with SonoVue: differentiation between benign and malignant lesions of the spleen. J Ultrasound Med 2009;28:421–34. [9] Metser U, Even-Sapir E. The role of 18F-FDG PET/CT in the evaluation of solid
64