European Journal of Radiology

Transcription

1 European Journal of Radiology 82 (2013) e255 e261 Contents lists available at SciVerse ScienceDirect European Journal of Radiology journal homepage: Response assessment of colorectal liver metastases with contrast enhanced CT/18F-FDG PET Ana María García Vicente a,, Esther Domínguez Ferreras b, Victoria Sánchez Pérez c, Víctor Manuel Poblete García a, José Carlos Villa Guzmán c, Fátima Jiménez Aragón b, Maria Dolores Pineda Pineda c, Ceferino Molino Trinidad b, Ángel Soriano Castrejón a a Department of Nuclear Medicine, University General Hospital, Ciudad Real, Spain b Department of Radiology, University General Hospital, Ciudad Real, Spain c Department of Oncology, University General Hospital, Ciudad Real, Spain article info abstract Article history: Received 20 November 2012 Received in revised form 24 December 2012 Accepted 26 December 2012 Keywords: Contrast enhanced CT PET FDG Liver metastases Response Colorectal cancer Chemotherapy Purpose: Evaluate the diagnostic performance of contrast enhanced CT/PET (cect/pet) in the response assessment of patients with colorectal cancer liver metastases. Methods: 33 ce CT/PET studies of 19 patients with colorectal liver metastases were prospectively evaluated. All of them, 13 (68.4%) were males and 6 (31.6%) females. Mean age and range were 63 [42 78]. All patients were treated with neoadjuvant chemotherapy. In all cases post-therapy diagnostic confirmation of liver lesions was obtained. A ce CT PET/was obtained 1 h after the injection of 370 MBq of 18F-FDG. Metabolic and morphologic studies were evaluated by two blinded nuclear physicians and radiologists respectively to assess the location, size and suspected diagnosis of lesions (benign or malignant). A combined assessment of both techniques was performed. The final diagnosis was established by histopathology or clinical/radiological follow-up greater than 6 months. Results: A total of 120 liver lesions were identified, 115 were malignant and 5 benign. From the malignant lesions, 105 were identified with the cect, 44 with the PET and 109 with cect/pet. All of the benign lesions were correctly classified with any of the three imaging techniques. The sensitivity of PET, cect and cect/pet were of 38%, 91% and 95% respectively and the specificity was 100% in all three of the diagnostic studies. Conclusion: Administration of intravenous contrast in the PET/CT is mandatory to evaluate treatment response rate of liver metastases due to the limitations of isolated metabolic images in these cases Elsevier Ireland Ltd. All rights reserved. 1. Introduction Colorectal cancer (CRC) is the second cause of cancer related death in the Western world, with a 5-year survival rate of 55%. Approximately 60% of patients with CRC develop metastatic disease being liver metastases the most common site of distant metastases [1]. The prognosis of metastatic colorectal cancer is poor, but there are considerable differences between patients with unresectable disease and those with resectable disease. In fact, the median survival of unresectable metastatic CRC does Corresponding author at: Nuclear Medicine Department, University General Hospital, C/Obispo Rafael Torija s/n 13005, Ciudad Real, Spain. Tel.: ; fax: address: angarvice@yahoo.es (A.M. García Vicente). not exceed 21 months, with a 5-year survival of <3% [2]. Hepatic resection is the only effective therapy for a subset of patients with CRC liver metastases, and is associated with 5-year survival rates ranging from 25 to 47% [3]. Unfortunately, surgical resection is feasible in less than 10% of these patients, and the recurrence rate after resection is high: 60 65% of the patients will develop recurrent liver disease after hepatic resection. This fact could indicate that these patients had undetected intra or extrahepatic tumor foci at the time of liver resection [4]. Moreover, several studies report unresectable disease in 40 70% of patients that undergo laparotomy for liver resection [5,6]. These data indicate that better patient selection is needed to avoid unnecessary operations. There are several potential ways of improving patient outcome such as, neoadjuvant chemotherapy followed by re-evaluation. This strategy let us a better preoperative staging. Neoadjuvant chemotherapy aims to down-stage non-resectable liver metastases X/$ see front matter 2013 Elsevier Ireland Ltd. All rights reserved.

2 e256 A.M. García Vicente et al. / European Journal of Radiology 82 (2013) e255 e261 and improve outcome following hepatic resection of resectable liver metastases. Positron emission tomography (PET) with the glucose analog 18-fluoro-2-deoxy-d-glucose (FDG-PET) is a sensitive diagnostic tool to diagnosis colorectal metastases. Approximately 25% of patients carry new intra- or extrahepatic tumors on FDG-PET performed after standard imaging. Screening with FDG-PET before hepatic resection in CRC metastases significantly improves the survival rates of resected patients, probably due to a better patients selection [7]. On the other hand, computed tomography (CT) using a contrast enhanced (ce) protocol is the principal preoperative imaging technique in the assessment of CRC metastases, specially liver metastases, due to be a non-invasive technique and its wide availability, with a high sensitivity and positive predictive values [8]. Integrated FDG-PET and CT involves the intravenous injection of a tagged biologically active compound, FDG, which is selectively taken up by metabolically active tissues followed by a combined assessment with CT and PET. It has therefore a biochemical and anatomical basis, with metastases appearing as glucose avid lesions. FDG-PET/CT is useful not only for detection of distant metastases in preoperative planning; it is also used to assess the extent of hepatic colorectal metastases [9]. However, there are limitations of using PET/CT for CRC liver metastases. In particular, PET/CT cannot detect lesions less than 5 mm. Furthermore, the use of chemotherapy less than 1 month before the scan can result in high false negative results [10,11]. In these cases, cect has been demonstrated to be superior to the PET to correctly evaluate liver metastases after neadjuvant chemotherapy [12]. The aim of our study was to evaluate the usefulness of adding a portal-phase cect protocol to a standard PET acquisition for the detection of CRC liver metastases after neoadjuvant chemotherapy. 2. Materials and methods This prospective multicentre study, enrolling 3 centers, was approved by the local ethics committee of our institution and Investigation Board. This study is supported by a FIS grant PI-11/ Patients Written informed consent was obtained from all patients. Other information as previous history of liver resection, chemoembolization or radiofrequency was collected. CeCT/PET was performed in a reference center, after 2 months-4 cycles of chemotherapy treatment until achieving resectability. Thirty three studies from 19 patients with hepatic metastases from CRC were included. There were 13 (68.4%) men and 6 (31.6%) women, with a mean age of 63 years (range years). In Table 1 are described all the patients characteristics CeCT/PET acquisition The patients were asked to fast for at least 4 h before undergoing cect/pet. They all had glucose levels under 160 mg/ml. All the patients received an intravenous injection of 370 MBq (10 mci) of FDG. Data acquisitions by an integrated PET/CT system (DSTE 16 s; GE Medical Systems) were performed 1 hour after FDG injection from the head to the proximal legs. Previous to PET acquisition and after the injection of 100 ml of iodinated contrast agent (non ionic Iomeprol), a portal venous phase (60 s) delayed CT scan (120 kv and modulated < 440 ma, 1.2 mm section thickness) was performed. The contrast was delivered through the same peripheral cannula used for administration of FDG using an automatized Table 1 Patients characteristics. Characteristics Total patients 19 Median age [range] 61 [38 76] Sex Male 13 Female 6 Tumour site Colon 5 Rectum 11 Sigma 3 Unknown 3 Stage at diagnosis III-B 3 III-C 1 IV 15 Liver metastasis Diagnosis 12 Recurrence 4 Diagnosis and postsurgery recurrence 3 Local liver treatment n Time Surgery) 5 20 weeks Radioembolization 1 16 months None 13 Previous chemotherapy n Time (weeks) XELOX 1 4 XELOX-Bevacizumab FOLFOX-Cetuximab 6 4 FOLFOX-Bevacizumab 1 13 FOLFIRI-Cetuximab FOLFIRI-Bevacizumab 4 2 XELIRI-Bevacizumab 1 3 TOMUDEX 1 3 XELODA-RT 1 13 n: number of patients, time: time to cect/pet (median2), injector. Immediately after CT scanning, a 3D PET emission scan that covered the identical transverse field of view was obtained. Acquisition time was 3 min per table position. PET image data sets were reconstructed iteratively by applying the CT data for attenuation correction, and co-registered images were displayed on a workstation Image interpretation Metabolic images (FDG-PET) were evaluated by two senior nuclear physicians. Visual and quantitative SUV analysis of PET findings was done: area of marked focal FDG accumulation greater than the background activity of the liver was interpreted as site of malignant disease. Maximum SUV (SUVmax) was calculated for all liver lesions. The SUVmax of each lesion was obtained by placing a circular ROI (region of interest) manually at the site of the maximum FDG uptake on the transaxial images. Results of cect images were reported by two senior radiologists with a special interest in hepatic imaging, assessing the number, size and location of the lesions attending to liver segments, and classifying the lesions as suspicious of being malignant or benign using radiological criteria. Any hypoenhancing lesion with or without peripheral enhancement with no typical imaging characteristics of benign lesions was classified as malignant. A lesion was classified as benign if it showed a typically benign enhancement pattern and showed stability with respect to studies previous to the cancer diagnosis if available. Both diagnostic reports were compared. In case of discordances in location or characterization of lesions a nuclear physician and a radiologist revised and evaluated in consensus the studies. A combined assessment of cect/pet was obtained. A lesion was classified as malignant when considered malignant by either of the two techniques was considered malignant. On the contrary a n

3 A.M. García Vicente et al. / European Journal of Radiology 82 (2013) e255 e261 e257 Fig. 1. Axial PET (A) and cect images (B) of a patient diagnosed of colon cancer with 4 malignant liver lesions in a diagnostic cect/pet. In a cect/pet performed 3 weeks after 4 cicles of chemotherapy, axial cect images (C) shows persistence of malignant lesions and axial PET images (D) complete response. Patient underwent surgery and 4 malignant lesions were resected. lesion was classified as benign if it was considered benign on both techniques 2.4. Final diagnosis Final diagnosis was obtained by histological confirmation when patients underwent liver resection and by clinical and radiological follow up of at least 6 months by a multidisciplinary team consisting of radiologist, hepatobiliary surgeons, medical oncologists and nuclear physician in the remaining. In the latter, a lesion was classified as malignant when it maintained or increased its size and/or metabolism in later studies or decreased metabolism and/or size with chemotherapy. On the contrary a lesion was considered benign when, without further treatment, its size decreased or stabilized in the follow up and it showed no detectable metabolism Data analysis Concordant and discordant findings of cect and FDG-PET/CT were compared in a lesion-based analysis The concordance between the imaging studies and the final diagnosis according to the status of a lesion (benign or malignant) was assessed with Cohen s kappa classifying the results in: poor (<0.20), weak ( ), moderate ( ), good ( ) and very good ( ). Values of sensitivity, specificity, positive predictive value, negative predictive value and accuracy of FDG-PET and cect in the detection of viable liver metastases were calculated compared to the histopathology of the lesions or the follow up. A ROC curve comparative analysis of the diagnostic performance of PET, cect and cect/pet was performed. Moreover a regression analysis was performed to determine association of liver metastases detection with FDG-PET and cect to the size of the liver metastases in the cect images, and to the time between the conclusion of chemotherapy and the imaging studies. The existence of a correlation was confirmed with a Pearson correlation. We performed a stratified analysis of the PET diagnostic performance for lesions smaller than 10 mm and lesions of Table 2 Statistical diagnostic parameters for PET, cect and cect/pet. PET cect cect/pet Value (%) CI Value (%) CI Value (%) IC Se Sp PPV NPV Se: sensitivity, Sp: specificity, PPV: positive predictive value, NPV: negative predictive value, CI: confidence interval.

4 e258 A.M. García Vicente et al. / European Journal of Radiology 82 (2013) e255 e261 Fig. 2. cect/pet of a patient with rectal cancer with synchronous liver metastases. After liver resection and chemotherapy, a liver recurrence was diagnosed. Axial cect/pet 3 weeks after 10 cycles of chemotherapy. In axial cect images (A) steatosis is observed, that causes difficulties in the detection of liver lesions. In axial PET images (B) 3 lesions were detected. In the follow up the liver disease progressed. 10 mm or bigger. For both stratification groups sensibility, specificity, PPV, NPV and AUC on the ROC curve analysis were calculated. 3. Results After chemotherapy a total of 120 liver lesions were identified. 115 were malignant and 5 benign. 13 malignant lesions (10.8%) were diagnosed by histology and the rest in the clinical follow-up (89.2%). From the malignant lesions, 105 were identified with the cect, 44 with the PET and 109 with any of both techniques. Fig. 1 shows an example of lesions identified only by cect. Out of the 10 malignant lesions not identified with cect, 4 were correctly diagnosed by PET and corresponded to a liver with extreme post-therapy density changes such as highly irregular patchy steatosis or grade III steatosis (Fig. 2). All benign lesions were correctly classified with both PET and cect. The agreement between imaging diagnostic techniques and the final diagnosis was very low for the PET, with a Cohen s kappa coefficient of 0.05 and moderate for the cect and cect/pet with values of 0.47 and 0.60 respectively. In Table 2 shows the statistical diagnostic parameters for cect, PET and cect/pet. In ROC analysis, the area under curve had values of (p = 0.149), (p = 0.001) and (p = 0.000) for the PET, cect and cect/pet respectively. Thus the diagnostic value of cect and cect/pet was statistically significant while PET alone showed no statistically significant diagnostic value (Fig. 3). The size of malignant lesions oscillated between 5 and 150 mm. The mean size of malignant liver lesions was 18 ± 20 mm. For the malignant lesions detected by PET the mean size was 27 ± 31 mm. The mean SUVmax of malignant lesions detected by PET was of 4.6 ± 2.2. Time between last cycle of chemotherapy and cect/pet was 2 13 weeks. The regression analysis of both, cect and PET performance showed correlation to the size of the lesion (p = 0.02 and p = 0.03) but not to the time elapsed between the last cycle of chemotherapy (p = 0.45 p = 0.09) and the imaging study. Pearson correlation shows significant association between PET diagnostic performance and lesion size (p = 0.001) and a less strong Fig. 3. Title: ROC curve comparative analysis of the diagnostic performance of PET, cect and cect/pet.

5 A.M. García Vicente et al. / European Journal of Radiology 82 (2013) e255 e261 e259 Table 3 Statistical diagnostic parameters for PET attending to lesion size. Lesion size < 10 mm Lesion size 10 mm value CI value CI Se (%) Sp (%) PPV (%) NPV (%) AUC Se: sensitivity, Sp: specificity, PPV: positive predictive value, NPV: negative predictive value, AUC: area under curve; CI: confidence interval. but nonetheless significant association between cect diagnostic performance and lesion size (p = 0.020) Since the cect detected and correctly classified most of the lesions, no further analysis was performed. On the other hand, PET detected 38.2% of the malignant lesions and since the regression analysis showed correlation of the diagnostic performance with lesion size, we performed a stratified analysis of the PET diagnostic performance for lesions of less than 10 mm (30 lesions) and lesions of 10 mm or greater (90). For both stratification groups sensibility, specificity, PPV, PNV and UCA on the ROC curve analysis were calculated (Table 3). The stratified analysis showed a lightly better diagnostic performance for bigger lesions but in none of the stratification groups the diagnostic value, in the ROC curve analysis, showed statistically significant results (p = 0.29 for bigger lesions and p = 0.53 for the smaller ones). Thus isolated PET is not an accurate enough diagnostic technique to be reliable on its own, globally or after size stratification. 4. Discussion In CRC patients, staging and surveillance are usually performed with cect scans due to high availability and relative low cost. PET is performed to identify extrahepatic disease and, occasionally, to better characterize marginal liver lesions [13]. In the pre-neoadjuvant setting, some works using FDG-PET have demonstrated a higher accuracy than cect images to detect hepatic metastases in patients with CRC [14,15]. A previous meta-analysis study demonstrated that FDG PET was the most accurate diagnostic imaging technique to detect colorectal liver metastases, with an overall sensitivity of 95% per-patient basis and 76% per-lesion basis [16]. However, other studies have shown that the sensitivity of FDG PET detecting hepatic metastases on a per-lesion basis was only 54 65%, suggesting limited detection ability for small-sized lesions or low metabolic lesions [2,17]. These results could be explained by the relatively high FDG uptake in normal liver and the variable accumulation of FDG in metastatic lesions [2]. Ruers et al. [17] found that for lesions 1 cm, spiral CT was more sensitive than FDG-PET. Of 22 lesions, FDG-PET detected 14% and CT 64%. For lesions greater than 1.5 cm diameter, the detection rate of FDG-PET and CT was similar. Therefore the real role of PET and cect is controversial and some authors argue that cect appears to enough for the detection of intrahepatic disease [18]. The development of combined modalities of CT and PET imaging, thereby presenting overlays of anatomic (CT) and functional (PET) information, may, however, lead to significant improvement of preoperative liver staging and preoperative judgment on resectability [5]. A proper evaluation and location of liver lesions is critical to make an optimal diagnosis and surgery planning. In conventional PET/CT the CT is used for attenuation correction and a basic location. The milliampere second setting is low (30 50 mas) to reduce the radiation dose to the patient. In contrast, a much higher settings (>120 mas) and intravenous contrast administration are needed to acquire the high quality images necessary to enable diagnosis. Neoadjuvant chemotherapy may impair lesion detection and underestimate lesion size, as a result of the occurrence of intraparenchymal changes [19,20]. Therefore, accurate imaging of the liver following neoadjuvant chemotherapy is crucial to carry out an optimal patients selection eligible for surgical resection. Scientific evidence on the accuracy of the various imaging modalities for preoperative imaging of colorectal liver metastases after neoadjuvant chemotherapy is limited and ambiguous. A recent meta-analysis revealed a pooled sensitivity estimates of 85.7% ( %) for MRI, 69.9% ( %) for CT, 54.5% ( %) for FDG-PET, and 51.7% ( %) for PET-CT [21]. Therefore this meta-analysis showed that in the absence of MRI, CT is the best alternative with a pooled sensitivity of 69.9% while both FDG-PET and PET-CT, which perform rather well in chemonaive liver metastases, have a low diagnostic performance in the neoadjuvant setting. Attending to our results, the sensitivity of PET was lower that the previously reported. This fact could be explained by the reduced time between last chemotherapy treatment and cect/pet (within 4 weeks in 29/33 studies). The mean size of the residual liver lesions cannot explain the low performance of PET compared to the cect in our results. Therefore other circumstances must be taken into account. There are several possible causes, which could explain the decreased sensitivity of FDG-PET/CT in the detection of colorectal metastases following neoadjuvant therapy such as: 4.1. Size of the lesion The sensitivity of the FDG-PET in detecting colorectal metastases has been reported as being directly related to the size of the lesions. In the present work, the average size of the metastases following chemotherapy treatment was of 19 mm and although it was demonstrated a significant association between PET diagnostic performance and lesion size, we can assume that this fact was not the main reason for the decreased sensitivity of FDG-PET following chemotherapy is most of our lesions, as other authors previously reported [22] Chemotherapy and metabolic shutdown The sensitivity of PET has been reported to be decreased when performed within 4 weeks of chemotherapy, likely due to a temporary metabolic shutdown [23,24]. Chemotherapy reduces metabolic activity in cancer cells, in particular the activity of the glycolytic hexokinase enzyme (GLUT-1 transporter) that collects FDG [25]. In our sample, the 89% of the malignant lesions underwent chemotherapy within 4 weeks previous to cect/pet. This short time between last chemotherapy treatment and cect/pet could affect metabolic detection of the metastatic lesions, probably due to molecular changes although our results were not significant Necrosis Other reason behind the chemotherapy-induced decrease in diagnostic performance of FDG-PET and PET-CT may include induced necrosis, which may give initially solid metastases a more

6 e260 A.M. García Vicente et al. / European Journal of Radiology 82 (2013) e255 e261 cystic appearance. However, there is no FDG uptake in necrotic areas, and therefore these lesions are not visualized. In our work, necrosis was not analyzed in cect. PET demonstrated false negative results in 11 lesions 20 mm but was able to detect 20 lesions with equal characteristics attending to the size. Due to that, it can be expected that necrosis might affect the metabolic detection although the fact that most of the lesions underwent chemotherapy within 4 weeks had also a impact in the metabolic behavior, being difficult to assess the isolated effect of each circumstance. Therefore attending to different reasons, partial response to therapy may cause a decreased FDG uptake in metastatic lesions, in comparison to the physiological background uptake of FDG in the liver, making them undetectable for FDG-PET. False positive results can be related with patients who had undergone a previous hepatic resection, for which follow-up FDG-PET detected uptake in the same location of the resected metastasis. FDG uptake in the tumor bed following a previous liver resection is not specific for tumor recurrence [22]. In our sample none false positive was detected with any diagnostic technique. Neoadjuvant chemotherapy induces important changes on the liver parenchyma, such as steatosis (irinotecan and 5-FU) or sinusoidal obstruction (Oxaliplatin) resulting in a lower density of the liver parenchyma and less contrast enhancement, leading to a decreased liver-to-lesion contrast, thereby hindering the detection, characterization, and delineation of lesions that can affect the diagnostic performance or CT [20,19,26]. In our work, only10 false negatives were found for cect. Out of these, 4 were detected by PET and corresponded to a patient with liver steatosis and another with post-surgery recurrence with changes in liver density that affected the correct evaluation on morphologic image. The reduced sample of patients and the limited number of histological confirmations and the very small number of benign lesions in the sample could affect our results On the other hand, although histological examination is considered the primary reference standard, the natural history of the liver metastasis make difficult the resection of lesions. Therefore follow up imaging has been used to confirm the presence of liver disease in the nonoperable patients by assessing lesion growth over time [27,28]. On the other hand, to the best of our knowledge no previous work has reported their experience in the combined assessment of cect and PET obtained from an integrated acquisition protocol. Although metabolic alterations in tumor cells may occur before changes in tumor size, the diagnostic performance of PET was low for assessing response. If the reduced metabolism has a prognostic value, should be demonstrated in future works. Correct detection and monitoring of colorectal liver metastases have a major influence on the treatment strategy and prognosis of CRC patients. Therefore detailed information of the presence, location, size and number of metastases, in the pre-chemotherapy as well as the post-chemotherapy setting, is very important because of the therapeutic and prognostic connotations. Furthermore the combined assessment of metabolic and morphologic imaging is basic to establish a correct location and size of lesions in order to avoid misinterpretations. 5. Conclusion The performance of cect/pet in this work was optimal for the diagnosis of residual liver disease. This study reveals the importance of the integration of both diagnostic modalities with the maximum diagnostic capability of CT overcoming the limitations of PET. References [1] Penna C, Nordlinger B. Colorectal metastases (liver and lung). Surgical Clinics of North America 2002;82: [2] Jonker DJ, Maroun JA, Kocha W. Survival benefit of chemotherapy in metastatic colorectal cancer: a meta-analysis of randomized controlled trials. British Journal of Cancer 2000;82: [3] Cardona K, Mastrodomenico P, D Amico F, Shia J, Gönen M, Weiser MR, et al. Detailed pathologic characteristics of the primary colorectal tumor independently predict outcome after hepatectomy for metastases. Annals of Surgical Oncology 2013;20: [4] Scheele J, Altendorf-Hofmann A. Resection of colorectal liver metastases. Surgery 1991;11: [5] Gibbs JF, Weber TK, Rodríguez-Bigas MA, Driscoll DL, Petrelli NJ. Intraoperative determinants of unresectability for patients with colorrectal hepatic metastases. Cancer 1982;82: [6] Steele G, Bleday R, Mayer RJ. A prospective evaluation of hepatic resection for colorectal carcinoma metastases to the liver: gastrointestinal tumor study group protocol Journal of Clinical Oncology 1991;9: [7] Fernández F, Drebin JA, Linehan DC, Dehdashti F, Siegel BA, Strasberg SM. Five years survival after resection of hepatic metastases from colorectal cancer in patients screened by positron emission tomography with 18-Ffluorodeoxyglucose (FDG-PET). Annals of Surgery 2004;240: [8] Tan CH, Iyer R. Use of computed tomography in the management of colorectal cancer. World Journal of Radiology 2010;2: [9] Wiering B, Krabbe PFM, Dekker HM, Oyen WJG, Ruers TJM. The Role of FDG-PET in the Selection of Patients with Colorectal Liver Metastases. Annals of Surgical Oncology 2007;14: [10] Selzner M, Hany TF, Wildbrett P, McCormack L, Kadry Z, Clavien PA. Does the novel PET/CT imaging modality impact on the treatment of patients with metastatic colorectal cancer of the liver? Annals of Surgery 2004;240: [11] Adie S, Yip C, Chu F, Morris DL. Resection of liver metastases from colorectal cancer: does preoperative chemotherapy affect the accuracy of PET in preoperative planning? ANZ Journal of Surgery 2009;79: [12] Spatz J, Holl G, Sciuk J, Anthuber M, Arnholdt HM, Märkl B. Neoadjuvant chemotherapy affects staging of colorectal liver metastases. A comparison of PET, CT and intraoperative ultrasound. International Journal of Colorectal Disease 2011;26: [13] Frankel TL, Gian RK, Jarnagin WR. Preoperative imaging for hepatic resection of colorectal cancer metastasis. Journal of Gastrointestinal Oncology 2012;3:11 8. [14] Rappeport ED, Loft A, Bethelsen AK, von der Recke P, Larsen PN, Mogensen AM, et al. Contrast-enhanced FDG-PET/CT vs. SPIO-enhanced MRI vs. FDG- PET vs. CT in patients with liver metastases from colorectal cancer: a prospective study with intraoperative confirmation. Acta Radiologica 2007;48: [15] Kim BI. Clinical application of F-18 FDG PET (PET/CT) in colorectal and anal cancer. Nuclear medicine and Molecular Imaging 2008;42(suppl.):52 9. [16] Bipat S, van Leeuwen MS, Comans EF, Pijl ME, Bossuyt PM, Zwinderman AH, et al. Colorectal liver metastases: CT, MR imaging, and PET for diagnosis. Metaanalysis. Radiology 2005;237: [17] Ruers TJ, Langenhoff BS, Neeleman N, Jager GJ, Strijk S, Wobbes T, et al. Value of positron emission tomography with [F-18] fluorodeoxyglucose in patients with colorectal liver metastases: a prospective study. Journal of Clinical Oncology 2002;2: [18] Patel S, McCall M, Ohinmaa A, Bigam D, Dryden DM. Positron emission tomography/computed tomographic scans compared to computed tomography scans for detecting colorectal liver metastases. a systematic review. Annals of Surgery 2011;253: [19] Robinson PJ. The effects of cancer chemotherapy on liver imaging. European Radiology 2009;19: [20] Vauthey JN, Pawlik TM, Ribero D, Wu TT, Zorzi D, Hoff PM, et al. Chemotherapy regimen predicts steatohepatitis and an increase in 90 days mortality after surgery for hepatic colorectal metastases. Journal of Clinical Oncology 2006;24: [21] van Kessel CS, Buckens CF, van den Bosch MA, van Leeuwen MS, van Hillegersberg R, Verkooijen HM. Preoperative imaging of colorectal liver metastases after neoadjuvant chemotherapy: a meta-analysis. Annals of Surgical Oncology 2012;19: [22] Lubezky N, Metser U, Geva R, Nakache R, Shmueli E, Klausner JM, et al. The role and limitations of 18-fluoro-2-deoxy-D-glucose Positron Emission Tomography (FDG-PET) scan and Computerized Tomography (CT) in restaging patients with hepatic colorectal metastases following neoadjuvant chemotherapy: comparison with operative and pathological findings. Journal of Gastrointestinal Surgery 2007;11: [23] Kostakoglu L, Goldsmith SJ. 18F-FDG-PET evaluation of the response to therapy for lymphoma and for breast, lung and colorectal carcinoma. Journal of Nuclear Medicine 2003;44: [24] Glazer ES, Beaty K, Abdalla EK, Vauthey JN, Curley SA. Effectiveness of positron emission tomography for predicting chemotherapy response in colorectal cancer liver metastases. Archives of Surgery 2010;145: [25] Akhurst T, Kates TJ, Mazumdar M, Yeung H, Riedel ER, Burt BM, et al. Recent chemotherapy reduces the sensitivity of [18F] fluorodeoxyglucose positron emission tomography in the detection of colorectal metastases. Journal of Clinical Oncology 2005;23:

7 A.M. García Vicente et al. / European Journal of Radiology 82 (2013) e255 e261 e261 [26] Angliviel B, Benoist S, Penna C, El Hajjam M, Chagnon S, Julie C, et al. Impact of chemotherapy on the accuracy of computed tomography scan for the evaluation of colorectal liver metastases. Annals of Surgical Oncology 2009;16: [27] van Kessel CS, van Leeuwen MS, van den Bosch MA, Borel RIH, Mali WP, Westers P, et al. Accuracy of multislice liver CT and MRI for preoperative assessment of colorectal liver metastases after neoadjuvant chemotherapy. Digestive Surgery 2011;28: [28] Bacigalupo L, Aufort S, Eberle MC, Assenat E, Ychou M, Gallix B. Assessment of liver metastases from colorectal adenocarcinoma following chemotherapy: SPIO-MRI versus FDG-PET/CT. La Radiologia Medica 2010;115: