Ann Nucl Med (2007) 21:585–592 DOI 10.1007/s12149-007-0066-3

ORIGINAL ARTICLE

Nonionic intravenous contrast agent does not cause clinically significant artifacts to 18F-FDG PET/CT in patients with lung cancer Young-Sil An · Seung S. Sheen · Y. J. Oh Sung C. Hwang · Joon-Kee Yoon

Received: 19 June 2007 / Accepted: 7 September 2007 © The Japanese Society of Nuclear Medicine 2007

Abstract Objective This study was performed to evaluate the effects of intravenous (i.v.) contrast agent on semiquantitative values and lymph node (LN) staging of 18 F-fluorodeoxyglucose positron emission tomography/ computed tomography (18F-FDG PET/CT) in patients with lung cancer. Methods Thirty-five patients with lung cancer were prospectively included. Whole-body PET and nonenhanced CT images were acquired 60 min following the i.v. injection of 370 MBq 18F-FDG and subsequently, enhancedCT images were acquired with the i.v. administration of 400 mg iodinated contrast agent without positional change. PET images were reconstructed with both nonenhanced and enhanced CTs, and the maximum and average standardized uptake values (SUVmax and SUVave) calculated from lung masses, LNs, metastatic lesions, and normal structures were compared. To evaluate the effects of the i.v. contrast agent on LN staging, we compared the LN status on the basis of SUVs (cut-offs; SUVmax = 3.5, SUVave = 3.0). Results The mean differences of SUVmax in normal structures between enhanced and nonenhanced PET/CT were 15.23% ± 13.19% for contralateral lung, 8.53% ±

6.11% for aorta, 5.85% ± 4.99% for liver, 5.47% ± 6.81% for muscle, and 2.81% ± 3.05% for bone marrow, and those of SUVave were 10.17% ± 9.00%, 10.51% ± 7.89%, 4.95% ± 3.89%, 5.66% ± 9.12%, and 2.49% ± 2.50%, respectively. The mean differences of SUVmax between enhanced and nonenhanced PET/CT were 5.89% ± 3.92% for lung lesions (n = 41), 6.27% ± 3.79% for LNs (n = 76), and 3.55% ± 3.38% for metastatic lesions (n = 35), and those of SUVave were 3.22% ± 3.01%, 2.86% ± 1.71%, and 2.33% ± 3.95%, respectively. Although one LN status changed from benign to malignant because of contrast-related artifact, there was no up- or downstaging in any of the patients after contrast enhancement. Conclusions An i.v. contrast agent may be used in PET/CT without producing any clinically significant artifact. Keywords Lung cancer · 18F-FDG PET/CT · Intravenous contrast · Contrast-related artifact · Nodal stage

Introduction 18

Y.-S. An · J.-K. Yoon (*) Department of Nuclear Medicine and Molecular Imaging, Ajou University School of Medicine, San 5, Wonchun-dong, Youngtong-gu, Suwon, Kyunggi-do 443-721, Korea e-mail: [email protected] S. S. Sheen · Y. J. Oh · S. C. Hwang Pulmonary and Critical Care Medicine, Ajou University School of Medicine, Kyunggi-do, Korea

F-fluorodeoxyglucose (FDG) positron emission tomography (PET) is widely used for staging, restaging, therapeutic response monitoring, and prognostication in patients with various cancers. Recently, combined PET/ computed tomography (CT) has gradually replaced PET in clinical oncology because it provides both metabolic and morphologic information of lesions within a single study, thus improving the accuracy in cancer staging [1–6]. Among various cancers, the use of PET/ CT instead of PET is needed most in lung cancer and

13

586

head-and-neck cancer [7]. It is well known that combined 18F-FDG PET/CT has a greater diagnostic accuracy when compared with CT alone, PET alone or the conventional visual correlation of PET and CT in assessing primary tumor, lymph node (LN), and metastasis of non-small-cell lung cancer. It also leads to a change of treatment more frequently when compared with PET alone or CT alone [8, 9]. The superiority of combined PET/CT as against PET alone has been confirmed in other cancers and in cancer screening [1, 3–6] as well. Numerous suggestions have been made for the optimal protocol of newly developed PET/CT. These relate to patient positioning, data acquisition, artifacts, CT-based attenuation correction, and CT contrast agents. Although the positive oral or intravenous (i.v.) contrast agents have advantages in improving the delineation of anatomic structures, the sensitivity in detecting pathologic lesions, and accuracy in lesion characterization, their routine use is controversial because of potential artifacts during CT-based attenuation correction. Highly concentrated focal radiodense contrast may result in an overestimation of FDG uptake in combined PET/CT [10–15]. However, when CT images of combined PET/CT are used for diagnostic purposes, those without i.v. contrast agents may be suboptimal because of similarity in CT densities of normal tissues [11]. Therefore, the possibility of misinterpretation by these contrast agents should be excluded before clinical application. The aim of this study was to evaluate the effects of a nonionic i.v. contrast agent on the interpretation of 18FFDG PET/CT in both technical and clinical aspects. For this purpose, in patients with lung cancer, we first measured the changes in standardized uptake value (SUV) caused by the i.v. contrast agent by comparing attenuation-corrected PET images by contrast-enhanced CT (CECT) with those by nonenhanced CT (NECT), and then evaluated its effect on LN stage by using SUV criteria.

Materials and methods Patient population This study included 35 patients (29 men, 6 women; mean age 62.8 years; range 39–83 years) with lung cancer (34 non-small-cell type and 1 small cell type) who were referred for 18F-FDG PET/CT. All patients signed an informed consent form, which details the use of i.v. contrast and its potential side effects, prior to participating in the study.

13

Ann Nucl Med (2007) 21:585–592

PET/CT acquisition and i.v. contrast injection After at least 4 h fasting, patients received 370 MBq 18FFDG via a peripheral i.v. line, which was changed to a saline lock for later injection of contrast agent. The glucose level was checked in patients prior to the study and also any history of diabetes mellitus was evaluated. All patients were instructed to rest comfortably for 60 min and empty the bladder before scanning. The PET/ CT scans were performed by a Discovery ST scanner (General Electric Medical Systems, Milwaukee, WI, USA). First, a routine nonenhanced CT scan from the base of the skull to the upper thigh (tube-rotation time of 1 s per revolution; 120 kV; 70 mA; 7.5 mm per rotation and acquisition time of 60.9 s for a scan length of 867 mm) was obtained. Subsequently, seven or eight frames (3 min per frame) of emission PET data were acquired in a two-dimensional mode, which finally was followed by a second contrast-enhanced CT scan. One hundred milliliters of iomeprol (Iomeron400, Bracco Diagnostics, Milan, Italy) containing 400 mg of iodine per milliliter was mixed with 50 ml of sterile saline and injected manually with a mean flow rate of 3 ml/s. CECT commenced at 20 s after starting i.v. injection on the basis of the routine enhancement protocol for lung cancer of our institute. The parameters of the CECT were the same as those of NECT. Emission PET images were reconstructed twice with both NECT and CECT using iterative reconstruction (software provided by General Electric Medical Systems, ordered-subsets expectation maximization with 2 iterations and 30 subsets, field of view = 600 mm, slice thickness = 3.27 mm). SUV was calculated for injected dosage and body weight. Analysis and interpretation of PET/CT images We evaluated five normal tissues, having a different range of CT densities, for each patient, and 152 pathologic tissues were assessed in 35 patients. A careful attention was paid to ensure anatomic consistency and avoid interference from adjacent activities of other tissues. Normal tissues consisted of (1) normal lung contralateral to primary lung mass lesion, (2) ascending aorta just above the carina level, (3) liver (segment VIII), (4) adductor muscle of right medial thigh, and (5) bone marrow of T12 vertebral body. Any lesions with benign changes such as calcification, inflammation, or fracture on CT or PET were excluded. In analyzing normal tissues, a circular region of interest (ROI) was placed on the identical axial location of the PET and both CT scans using a “copy-and-paste” tool built on Xeleris workstation (General Electric Medical Systems). For pathologic lesions, a manual ROI

1.47 ± 0.32 1.43 ± 0.32 2.01 ± 0.42 1.95 ± 0.40

<0.001

0.56 ± 0.12 0.53 ± 0.11 0.98 ± 0.22 0.94 ± 0.22

Adductor muscle of thigh Bone marrow (T12)

<0.001

1.95 ± 0.32 1.87 ± 0.32 2.53 ± 0.45 2.40 ± 0.44 Liver (segment 8)

<0.001

1.58 ± 0.25 1.44 ± 0.26 <0.001 2.05 ± 0.27 1.90 ± 0.28 Ascending aorta

All data are presented as mean ± standard deviation; difference = SUV by CECT − SUV by NECT, difference (%) = difference/SUV of CECT × 100 (%) NECT nonenhanced CT, CECT contrast enhanced CT a Maximum SUV derived from non-enhanced CT b Maximum SUV derived from contrast-enhanced CT c Average SUV derived from non-enhanced CT d Average SUV derived from contrast-enhanced CT

<0.001

<0.001

<0.001

<0.001

<0.001

10.17% ± 9.00% (0%–40.00%) 10.51% ± 7.89% (0%–41.98%) 4.95% ± 3.89% (0%–16.99%) 5.66% ± 9.12% (0%–49.02%) 2.49% ± 2.50% (0%–12.50%) 0.04 ± 0.04 (0–0.20) 0.14 ± 0.08 (0–0.38) 0.09 ± 0.07 (0–0.26) 0.03 ± 0.05 (0–0.25) 0.03 ± 0.03 (0–0.16) 0.39 ± 0.13 0.36 ± 0.10

15.23% ± 13.19% (0%–50.00%) 8.53% ± 6.11% (0.53%–25.62%) 5.85% ± 4.99% (0%–20.00%) 5.47% ± 6.81% (0%–28.57%) 2.81% ± 3.05% (0%–15.10%) 0.08 ± 0.07 (0–0.31) 0.15 ± 0.10 (0.01–0.38) 0.13 ± 0.11 (0–0.47) 0.05 ± 0.05 (0–0.20) 0.06 ± 0.07 (0–0.37) 0.60 ± 0.19 0.52 ± 0.16 Lung

<0.001

SUVaved (CECT) P % Difference (range) Difference (range) SUVmaxb (CECT) SUVmaxa (NECT) Normal tissues

Table 1 Differences of standardized uptake values (SUVs) in normal tissues

587

SUVavec (NECT)

Difference (range)

% Difference (range)

P

Ann Nucl Med (2007) 21:585–592

was placed on each transaxial PET slice along the 70% threshold line of the maximum pixel count (Fig. 1). The maximum (SUVmax) and average SUVs (SUVave) were recorded, and a paired t test was used to reveal the difference of SUVs before and after contrast enhancement. The difference (%) was calculated as dividing the difference by the SUV by CECT. P values of less than 0.05 were considered to be statistically significant. To evaluate the effect of contrast agent on stage, all LNs were determined as positive or negative for malignancy on the basis of SUV criteria. The threshold of SUV to distinguish benign from malignant LN was 3.5 for SUVmax and 3.0 for SUVave. The cut-off value of SUVmax was decided referring to an earlier report [16] and that of SUVave was determined as 80% of 3.5. If a LN fulfilled both maximum and average criteria, it was considered to be positive for malignancy. The staging of LNs was based on tumor, node, metastasis (TNM) staging of lung cancer.

Results No patient complained of symptoms of contrast-related side effects during the study. Also there was no significant change of patient position between noncontrast and contrast-enhanced studies, which could exclude possible mismatches on CT and PET images. Three diabetic patients were included in our study population but their blood glucose levels were below 150 mg/dl. Pathologic tissues included 41 primary malignant lesions of lungs (5 of the 35 patients had more than one daughter nodule adjacent to the main mass), 76 LNs and 35 metastatic lesions (28 bones, 4 livers, and 3 adrenal glands). Effect of i.v. contrast on SUV Standardized uptake values of normal tissues were presented in Table 1. The mean differences of SUVmax in normal structures between enhanced and nonenhanced PET/CT images were 15.23% ± 13.19% for contralateral lung of primary lesion, 8.53% ± 6.11% for aorta, 5.85% ± 4.99% for liver, 5.47% ± 6.81% for muscle and 2.81% ± 3.05% for bone marrow, and those of SUVave were 10.17% ± 9.00%, 10.51% ± 7.89%, 4.95% ± 3.89%, 5.66% ± 9.12%, and 2.49% ± 2.50%, respectively. All normal tissue regions showed significant elevations of SUVmax and SUVave after contrast enhancement (P < 0.05) (Table 1, Fig. 2). For pathologic lesions, the mean differences of SUVmax between enhanced and nonenhanced PET/CT images were 5.89% ± 3.92% for lung lesion, 6.27% ± 3.79% for

13

588

Ann Nucl Med (2007) 21:585–592

Fig. 1 A representative image of regions of interest (ROIs) drawn on a pathologic lesion. Transaxial positron emission tomography images of a metastatic lesion in liver derived from contrastenhanced (a) and nonenhanced computed tomography (CT) (b) were automatically changed to 10 scale maps by the embedded software using % threshold method (c, d), after which we drew ROIs along the 70% threshold lines (e, f)

LNs, and 3.55% ± 3.38% for metastatic lesions, and those of SUVave were 3.22% ± 3.01%, 2.86% ± 1.71% and 2.33 ± 3.95%, respectively. All pathologic lesions also revealed statistically significant elevation of SUVs after contrast-enhancement (P < 0.05) (Table 2, Fig. 3).

nodal status according to SUVave after i.v. contrast enhancement.

Results of LN staging

It is well known that 18F-FDG PET has benefits over CT for staging various cancers when assessing LN metastasis and distant metastasis [17, 18]. In addition, introduction of PET/CT has resulted in a lower noise emission scan, faster examination (thus, less motion artifact and higher patient throughput), greater topographic resolution, and more accurate localization and functional interpretation in a single setting [14, 15, 19, 20]. In interpreting CT images, i.v. contrast can give benefits in detection and characterization of lesions by increasing attenuation differences among anatomic structures [10, 21–25]. However, routine CT scanning in the PET/CT examination has generally been performed without i.v. administration of a contrast agent, because it might affect the CT measurements of attenuation correction. Therefore, CT in combined PET/CT is suboptimal for lesion delineation from vessels and other adjacent organs.

On the basis of SUV calculated from PET images reconstructed with noncontrast CT, median SUVmax, and SUVave of all 76 LNs were 5.18 (range 2.37–15.53) and 4.12 (range 2.13–7.76). The SUV criteria classified 76 LNs as 7 benign and 69 malignant LNs by SUVmax, and 4 benign and 72 malignant lesions by SUVave. Median SUVmax and SUVave increased to 5.56 (ranged 2.7–16.62) and 4.26 (2.23–7.86) after contrast enhancement, and only one LN status changed from benign to malignant because of contrast-related artifact. The SUVmax of this LN was elevated from 3.4 to 3.52. However, for LN stage, there was no up- or down-staging in any patients after contrast enhancement because one patient, whose LN status changed from benign to malignant, had another N3 node involvement, and thus, his nodal stage did not change. Moreover, there was no change of

13

Discussion

Ann Nucl Med (2007) 21:585–592

589

Fig. 2 Graphs of standardized uptake values (SUVmax) derived from nonenhanced CT (NECT) and contrast enhanced CT (CECT) in normal tissues. All normal tissues show increasing pattern after contrast enhancement. Lung (a), aorta (b), liver (c), muscle (d), and bone marrow (e)

Our interest was whether the use of i.v. contrast in PET/CT would distort the PET data and possibly affect staging in cancer patients. In our study, all of normal and pathologic tissues showed significantly elevated maximum SUV and average SUV following i.v. contrast enhancement. This result was different from that of a previous study done by Yat et al. [26]. They reported that physiologic tissues with a larger amount of i.v. contrast delivery (i.e., liver, spleen, and aorta) did produce a statistically significant elevation of maximum SUV, but less vascular or lowdensity tissues did not show this trend, and thus the mediastinal LNs were the only pathologic tissues that showed statistically significant change of maximum SUV. Meanwhile, in our study, liver and ascending aorta were less affected than lung (5.8%, 8.5% vs. 15.2%) by i.v. contrast, and all pathologic tissues showed significant elevation of maximum SUV. There were some dif-

ferences between the two study protocols, which appeared to cause these different results on pathologic and normal tissues. First of all, our study was performed with a larger homogeneous group of patients (152 lesions from 35 lung cancer patients), whereas the study of Yat et al. included many different types of malignancies (lung, nasopharynx, larynx, ovarian, uterine corpus, endometrial, retroperitoneal sarcoma, and colon cancer) and only 37 pathologic lesions were analyzed. In addition, we drew larger circular ROIs on normal tissues to minimize the effect of heterogeneity of FDG uptake, and thus our ROIs covered most area of a target organ within a given transaxial slice. The location of ROIs would be responsible for this difference. We chose the transaxial slices of lower lung field rather than upper lung field, because, in the Korean population, abnormal FDG uptake by stable or active tuberculosis is frequently observed in upper lung fields. Also, some other points

13

<0.05

<0.05

<0.01

<0.001

<0.001

5.50 ± 1.92

5.68 ± 1.96

4.36 ± 0.72

5.29 ± 2.61

5.32 ± 1.88

5.52 ± 1.91

4.25 ± 0.72

4.91 ± 2.59

a

3.92 ± 1.15 3.72 ± 1.09 <0.05

3.78 ± 0.43 3.70 ± 0.40 <0.05

4.20 ± 0.70 4.14 ± 0.73 <0.001

4.13 ± 0.71 4.05 ± 0.74 <0.001

4.44 ± 1.10 4.32 ± 1.06 <0.05 6.45 ± 3.04 6.08 ± 2.87

13

All data are presented as mean ± standard deviation; difference = SUV by CECT − SUV by NECT, difference (%) = difference/SUV of CECT × 100 (%) Maximum SUV derived from non-enhanced CT b Maximum SUV derived from contrast-enhanced CT c Average SUV derived from non-enhanced CT d Average SUV derived from contrast-enhanced CT

3.22% ± 3.01% (0.20%–13.74%) 2.86% ± 1.71% (0%–8.12%) 2.33% ± 3.95% (0%–17.27%) 2.03% ± 4.29% (0%–17.27%) 2.14% ± 1.13% (1.05%–3.67%) 5.37% ± 0.37% (4.95%–5.60%) 0.19 ± 0.17 (0.01–0.73) 0.12 ± 0.09 (0–0.43) 0.08 ± 0.11 (0–0.52) 0.07 ± 0.12 (0–0.52) 0.08 ± 0.05 (0.04–0.14) 0.20 ± 0.06 (0.15–0.27) 6.32 ± 2.09 6.12 ± 2.02 <0.001

5.89% ± 3.92% (0%–13.88%) 6.27% ± 3.79% (0%–19.55%) 3.55% ± 3.38% (0.21%–16.67%) 3.26% ± 3.40% (0%–16.67%) 2.80% ± 1.24% (1.37%–4.00%) 8.39% ± 4.27% (5.10%–11.41%) 11.94 ± 6.44 11.30 ± 6.04

Lung mass (n = 41) Lymph nodes (n = 76) All metastatic lesions (n = 35) Bones (n = 28) Liver (n = 4) Adrenal glands (n = 3)

0.64 ± 0.60 (0–2.74) 0.38 ± 0.29 (0–1.36) 0.17 ± 0.14 (0.01–0.56) 0.17 ± 0.14 (0.01–0.56) 0.12 ± 0.05 (0.06–0.17) 0.28 ± 0.09 (0.21–0.40)

% Difference (range) Difference (range) SUVaved (CECT) SUVavec (NECT) P % Difference (range) Difference (range) SUVmaxb (CECT) SUVmaxa (NECT) Pathologic tissues

Table 2 Differences of SUVs in pathologic lesions

<0.001

Ann Nucl Med (2007) 21:585–592

P

590

such as the amount and flow rate of contrast or the interval from i.v. injection to CT scan might contribute to the results. An earlier study by Mawlawi et al. [27] also evaluated the potential clinical effect of i.v. contrast on PET/CT. They enrolled 12 tumor regions in nine patients with malignancy in lung, esophagus and pancreas. According to their results, the largest increase in SUVmax was 18.5% in malignant lesions after using i.v. contrast, but this increase was not significant in oncologic staging when these regions had background 18F-FDG uptake. However, there were still concerns on contrast-induced artifact, which could alter diagnostic decisions, and therefore we compared the LN stage on attenuation corrected PET images between before and after contrast enhancement. Although contrast induced artifact changed one LN status from benign (SUVmax = 3.4) to malignant (SUVmax = 3.52), it could not alter the LN stage of any of 35 patients. Therefore, according to our observation, i.v. contrast-related artifact does not effect significantly the clinical evaluation of cancers. However, the degree of artifact could vary according to the concentration of contrast agent. From this point of view, delay time between contrast injection and CT scan is important in artifact generation of PET/CT. As arterial phase CT concentrates more contrast medium than venous phase CT, it might induce a clinically significant artifact on PET images. Therefore, although i.v. contrast agent did not cause clinically significant artifacts in lung cancer, it would not be safe in other types of cancer. For routine use of i.v. contrast in clinical PET/CT, an improvement in staging accuracy by CECT is crucial. In our experience, i.v. contrast increased the accuracy of staging with PET by discriminating the primary mass from obstructive pneumonia or metastatic nodes from adjacent vascular structures. In addition, it helped detect small hepatic metastasis with less prominent FDG uptake, which was not discernible on low-dose noncontrast CT. However, we did not deal with that topic because the objective of this study was to evaluate the potential effect of contrast-related artifact on SUVbased staging. Whether CECT provides additional information in TNM staging should be investigated in a later study. In our study, a total of 21 of the 35 patients underwent surgical resection. The other 14 lung masses were pathologically confirmed by fine needle aspiration and biopsy. Only 35 of the 76 nodes were removed by surgical resection or mediastinoscopic biopsy, which was insufficient for pathologic correlation. This was a limitation of our study. Another limitation was a possible technical error induced by manual contrast injection. At the opening

Ann Nucl Med (2007) 21:585–592

591

Fig. 3 Graphs of SUVmax derived from NECT and CECT in pathologic tissues. All pathologic lesions show increasing pattern after contrast enhancement. Lung mass (a), lymph nodes (b), bone (c), liver (d), and adrenal gland metastasis (e)

part of the study, we were not sufficiently experienced with manual contrast injection, and so we could not inject the full volume of contrast within a limited time. A few earlier images, which were not correctly enhanced, were excluded from the study. To minimize the error, injection was performed by two physicians (J.K.Y and Y.S.A) and not by others. We also made an effort to control other factors related to SUVs, such as injection dosage of FDG, uptake time, hydration state and blood glucose level. There were three diabetic patients in our study population. We checked their serum glucose levels before the study and all were below 150 mg/dl.

in lung cancer. All the normal and pathologic lesions showed statistically significant elevation of SUVs after contrast-enhancement, although there was no up or down nodal staging in all patients. Our data suggest that nonionic i.v. contrast media can be used in PET/CT without clinically significant artifact in patients with lung cancer. Acknowledgment We appreciate the technical support of GyooSeul Shin, a nuclear medicine technologist, who helped us establish the protocol of i.v. contrast use.

References Conclusions We assessed the effects of i.v. contrast agent on semiquantitative values and LN stage of 18F-FDG PET/CT

1. Bar-Shalom R, Guralnik L, Tsalic M, Leiderman M, Frenkel A, Gaitini D, et al. The additional value of PET/CT over PET in FDG imaging of oesophageal cancer. Eur J Nucl Med Mol Imaging 2005;32:918–24.

13

592 2. Beyer T. Townsend DW, Brun T, Kinahan PE, Charron M, Roddy R, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med 2000;41:1369–79. 3. Chen YK, Ding HJ, Su CT, Shen YY, Chen LK, Liao AC, et al. Application of PET and PET/CT imaging for cancer screening. Anticancer Res 2004;24:4103–8. 4. Kim JH, Czernin J, Allen-Auerbach MS, Fueger BJ, Hecht JR, Ratib O, et al. Comparison between 18F-FDG PET, in-line PET/CT, and software fusion for restaging of recurrent colorectal cancer. J Nucl Med 2005;46:587–95. 5. Nanni C, Rubello D, Fanti S, Farsad M, Ambrosini V, Rampin L, et al. Role of 18F-FDG-PET and PET/CT imaging in thyroid cancer. Biomed Pharmacother 2006;60:409–13. 6. Yuan S, Yu Y, Chao KS, Fu Z, Yin Y, Liu T, et al. Additional value of PET/CT over PET in assessment of locoregional lymph nodes in thoracic esophageal squamous cell cancer. J Nucl Med 2006;47:1255–9. 7. Goerres GW, von Schulthess GK, Steinert HC. Why most PET of lung and head-and-neck cancer will be PET/CT. J Nucl Med 2004;45:66S–71S. 8. Antoch G, Stattaus J, Nemat AT, Marnitz S, Beyer T, Kuehl H, et al. Non-small cell lung cancer: dual-modality PET/CT in preoperative staging. Radiology 2003;229:526–33. 9. Lardinois D, Weder W, Hany TF, Kamel EM, Korom S, Seifert B, et al. Staging of non-small-cell lung cancer with integrated positron-emission tomography and computed tomography. N Engl J Med 2003;348:2500–7. 10. Albertyn LE. Rationales for use of intravenous contrast medium in computed tomography. Australas Radiol 1989;33: 29–33. 11. Antoch G, Freudenberg LS, Beyer T, Bockisch A, Debatin JF. To enhance or not to enhance? 18F-FDG and CT contrast agents in dual-modality 18F-FDG PET/CT. J Nucl Med 2004;45:56S–65S. 12. Anotoch G, Freudenberg LS, Egelhof T, Stattaus J, Jentzen W, Debatin JF, et al. Focal tracer uptake: a potential artifact in contrast-enhanced dual-modality PET/CT scans. J Nucl Med 2002;43:1339–42. 13. Anotoch G, Freudenberg LS, Stattaus J, Jentzen W, Mueller SP, Debatin JF, et al. Whole-body positron emission tomography-CT: optimized CT using oral and IV contrast materials. AJR Am J Roentgenol 2002;179:1555–60. 14. Beyer T, Antoch G, Muller S, Egelhof T, Freudenberg LS, Debatin J, et al. Acquisition protocol considerations for combined PET/CT imaging. J Nucl Med 2004;45:25S–35S.

13

Ann Nucl Med (2007) 21:585–592 15. Nakamoto Y, Chin BB, Kraitchman DL, Lawler LP, Marshall LT, Wahl RL. Effects of nonionic IV contrast agents at PET/CT imaging: phantom and canine studies. Radiology 2003;227:817–24. 16. Kim BT, Lee KS, Shim SS, Choi JY, Kwon OJ, Kim H, et al. Stage T1 Non-small cell lung cancer: preoperative mediastinal nodal staging with integrated FDG PET/CT – a prospective study. Radiology 2006;241:501–9. 17. Gambhir SS, Czernin J, Schwimmer J, Silverman DH, Coleman RE, Phelps ME. A tabulated summary of the FDG literature. J Nucl Med 2001;42:1S–93S. 18. van Tinteren H, Hoekstra OS, Smit EF, van den Bergh JH, Schreurs AJ, Stallaert RA, et al. Effectiveness of positron emission tomography in the preoperative assessment of patients with suspected non-small-call lung cancer: the PLUS multicentre randomized trial. Lancet 2002;359:1388– 93. 19. Czernin J, Schelbert H. PET/CT imaging: facts, opinions, hopes, and questions. J Nucl Med 2004;45:1S–3S. 20. Townsend DW, Beyer T, Kinahan PE, Brun T, Roddy R, Nutt R, et al. The SMART scanner: a combined PET/CT tomography for clinical oncology. IEEE Nucl Sci Symp Conf Rec 1998;2:1170–4. 21. Alfidi RJ, Haaga J, Meaney TF, MacIntyre WJ, Gonzalez L, Tarar R, et al. Computed tomography of the thorax and abdomen: a preliminary report. Radiology 1975;117:257–64. 22. Beurgener FA, Hamlin DJ. Intravenous contrast enhancement in computed tomography of pelvic malignancies. Rofo 1981;134:656–61. 23. Kormano MJ, Goske MJ, Hamlin DJ. Attenuation and contrast enhancement of gynecologic organs and tumors in CT. Eur J Radiol 1981;1:307–11. 24. Violante MR, Dean PB. Improved detectability of VX2 carcinoma in the rabbit liver with contrast enhancement in computed tomography. Radiology 1980;134:237–9. 25. Garrett PR, Meshkov SL, Perlmutter GS. Oral contrast agents in CT of the abdomen. Radiology 1984;153:545–6. 26. Yat YY, Chan WS, Tam YM, Vernon P, Wong S, Coel M, et al. Application of intravenous contrast in PET/CT: does it really introduce significant attenuation correction error? J Nucl Med 2005;46:283–91. 27. Mawlawi O, Erasmus JJ, Munden RF, Pan T, Knight AE, Macapinlac HA, et al. Quantifying the effect of IV contrast media on integrated PET/CT: clinical evaluation. AJR Am J Roentgenol 2006;186:308–19.