136 POSTER Two central veins comparative study for totally implantable venous access devices in 1201 cancer patients

Poster session: Oesophageal and gastric cancer agus, of which contact angle thoracic aorta are over 90 degrees, and had undergone operation subsequently. The diagnosis of squamous carcinoma was confirmed histologically from biopsy specimens by endoscopy. Forty seconds after intravenous injection of iodinated contrast material, ECG-Gated 16-low MDCT imaging was acquired with simultaneous recording of ECG signals. CT scanning was performed with a 1.25-mm slice thickness, helical pitch 6.00 (Light Speed Ultra 16, GE, Milwaukee, WI). The R-to-R interval on ECG was equally divided into 10 phases (From 0% portion to 90%portion). Images were reconstructed every 10% of the Rto-R interval Using serial 10 volumetric data, and 4D movies was acquired in the workstation (Advantage Workstation 4.1, GE, Milwaukee, WI). We observed the movement of EC and aorta. Results: In 15 cases, we could observe a low density band, which is the separable space between EC and aorta. In other 5 cases, the low density band was not detected, but we could observe that tumor contact angle changes over 30 degrees during the pulsation in these 5 cases. All 20 subjects underwent operation with radical esophagectomy, and all revealed no EC invasion of aorta in histological results (pT3). Limitation: If we diagnose EC as cT4 to aorta, there is no indication for surgery and unable to compare CT findings with histological results. Therefore in this study, only a negative predictive value could be evaluated, as that, if a low density band is observed or the tumor contact angle changes, the tumor will be T3 cancer. Conclusion: ECG-gated MDCT can eliminate the influence of aortic pulsation. If a low density band is visualized between EC and aorta or the tumor contact angle to aorta changes during the aortic pulsation it the 4D movie, the tumor is not T4 to aorta but T3 because of the movement between them. In limitation, if we diagnose EC as cT4 to aorta, there is no indication for surgery and unable to compare CT findings with histological results. Therefore in this study, only a negative predictive value could be evaluated, as that, if a low density band is observed or the tumor contact angle changes, the tumor will be T3 cancer. This noninvasive modality may become a useful imaging technique to evaluate EC invasion of thoracic aorta. 135

POSTER

Prediction of the response to chemoradiation therapy (CRT) in esophageal squamous cell carcinoma by Perfusion CT K. Hayano 1 , S. Okazumi 1 , K. Shuto 1 , R. Mochizuki 1 , T. Aoyagi 1 , K. Narushima 1 , A. Sato 1 , T. Kazama 2 , N. Yanagawa 2 , T. Ochiai 1 . 1 Chiba University Graduate School of Medicine, Department of Frontier Surgery, Chiba, Japan; 2 Chiba University Graduate School of Medicine, Department of Radiology, Chiba, Japan Background: The ability to predict and assess the response to chemoradiationtherapy (CRT) by contrast-enhanced CT would be valuable for managing esophageal squamous cell carcinoma. The purpose of this study is to evaluate the usefulness of Perfusion CT to predict the response to CRT in patients with esophageal squamous cell carcinoma. Methods and materials: Twenty-four consecutive patients with esophageal squamous cell carcinoma underwent Perfusion CT before and after CRT, using the commercially available CT Perfusion 3 software (GE Medical Systems, Milwaukee, WI). We retrospectively investigated the correlations between the Perfusion parameters and the response to CRT. Results: There were 17 responders and 7 non-responders according to the clinical response criteria proposed by the Japanese Society for Esophageal Disease. Responsive tumors had significantly higher pre-CRT blood flow (P = 0.0009), significantly higher pre-CRT blood volume (P = 0.02), and significantly shorter pre-CRT mean transit time (P = 0.02) than non-responsive tumors. Conclusion: Perfusion CT may help to identify patients with advanced esophageal squamous cell carcinoma who will benefit from CRT. In the near future, we may be able to determine who should undergo surgery without CRT, and who should under go CRT before surgery. To obtain more conclusive results, a larger patient population should be studied, but we hope that our results will provide important insight into selecting the optimal therapeutic strategies for the treatment of esophageal squamous cell carcinoma.

136

S41 POSTER

Two central veins comparative study for totally implantable venous access devices in 1201 cancer patients C. Araújo, J.L. Fougo, P. Antunes, J.P. Silva, J.M. Fernandes, C. Dias, H. Pereira, T. Dias, H. Silva. Portuguese Institute of Oncology, Department of Surgical Oncology, Porto, Portugal Introduction and aims: Subclavian vein has been traditionally the vein of choice to central venous catheterization by general surgeons. Alternative settings for the introduction of totally implantable venous access devices (TIVAD) and the search for lower rates of morbidity led to the choice of other central veins. This study compares two different venous accesses, subclavian (SC) versus internal jugular (IJ), in terms of early and late morbidity. Patients and methods: This is a prospective, non-randomized, observational, uni-institutional (tertiary cancer centre) study. From March 2003 to March 2006, 1231 TIVAD were placed (1201 patients), in an ambulatory operating room, under vital signs and EKG monitoring, using local anaesthesia and without perioperative radiological control. Patient chart records were analysed in order to verify long term outcomes and morbidity. For statistical analysis, non-parametric tests, Student’s t-test and Logistic Regression were used (SPSS 14.0, SPSS Inc.). Results: Median patients age was 56 years (range:15-83); 63.6% of the patients were female. Most frequent tumours implied were colorectal (38.4%) and breast (33.9%); therapeutic intention was adjuvant in 46% and palliative in 33%. Of the 1231 TIVAD, 617 were inserted via SC and 614 via IJ vein. The two groups (SC vs. IJ) were comparable as to patient general characteristics. Median dwell time was 363 days (3-1132) for SC and 244 (3-853) for IJ catheters. Immediate complications were more frequent in SC than in IJ approach (respectively, 5.0% vs. 1.6%; p=0.001). Catheter malposition occurred in 2.3% when using SC vein and in 0.2% for IJ (p=0.001). There was no significant relationship for pneumothorax or arterial puncture. Long term morbidity was also more frequent in SC than in IJ group (respectively, 15.8%, 87/551, vs. 7.6%, 39/512; p<0.001). Venous thrombosis developed in 2.0% (11/551) patients with an SC TIVAD as compared to 0.6% (3/512) with IJ TIVAD (p=0.044) [OR= 0.28 (95%CI 0.08-1.04)]. Catheter malfunction was significantly dependent on the vein used: SC – 9.4% (52/551) vs. IJ – 4.3% (22/512) (p=0.001) [OR= 0.43 (95%CI 0.260.72(]. There were no differences in sepsis or dislocated catheter tip rates. Conclusions: Our results support the preferential use of Internal Jugular vein for the insertion of TIVAD. 137

POSTER

Pattern and predictors of recurrence in resected oesophageal cancer A.A. Ayantunde 1 , J.P. Duffy 2 , N.T. Welch 1 , S.L. Parsons 1 . 1 Nottingham University Hospitals, Department of Surgery, Nottingham, United Kingdom; 2 Nottingham University Hospitals, Department of Thoracic Surgery, Nottingham, United Kingdom Introduction: The prognosis for oesophageal cancer is poor and significant numbers of the patients still experience recurrence not too long after radical resection. We evaluated the pattern of recurrence after curative oesophagectomy for cancer and seek to identify factors predictive of recurrent disease. Patients and Methods: A total of 310 consecutive patients who underwent potentially curative resection for oesophageal cancer between January 2001 and December 2004 were followed up for evidence of recurrent disease. Physical examination was complimented by ultrasonography, computed tomography, magnetic resonance imaging, isotope bone scan, endoscopy and laparoscopy where indicated. Univariate and multivariate survival analyses were performed using the Kaplan-Meier survival function and Cox proportional hazard model with SPSS® version 13.0 and factors achieving p-value <0.05 considered significant. Results: There were 237 adenocarcinomas and 73 squamous cell carcinomas with median follow up period of 23 (1-61) months. The perioperative mortality was 8.4%. The median age was 68 (40-89) years and male to female ratio was 3.4:1. 144 (46.45%) patients developed recurrent disease with 80% (116/144) recurring within the first two years after surgery. The median time to recurrence was 13 (1-55) months and me-