402lesions prospectively studied from January to Dec 2008.
98 head and neck pathologies
26 esophageal pathologies
142 lung pathologies
101 abdominal pathologies
42 pelvic pathologies
Of which 33 previously untreated patients with H& N , abdominal , lung and esophageal malignancies (25 males, 8 females, aged 47-69 years) which had lymphnodal metastasis were included in the study.
CT perfusion was performed with a 64 –slice MDCT. Tumor was localized and a 4-cm lesions region was selected independently for the dynamic study . Contrast bolus infusion at a rate of 50 mL at 5 mL/sec for 10 seconds, followed by a saline flush at 40 mL at 5mL/sec for 8 seconds. Total 30 dynamic acquisitions with inter - cycle interval 2sec and total scan time 60 seconds. Followed by routine contrast-enhanced scan. This scan was used for routine cancer diagnosis.
Data processed on Extended Brilliance_ Workstation and analyzed by using Brilliance perfusion 2.1.1 software. The artery input (ROI) was placed over the aorta/ respective main artery. ROI was repeated for each contiguous transverse level of the entire lesion. Global values of the entire lesion were calculated by taking the mean values of all individual sections.
We used maximum slope analytical model method, yielding five major kinetic parameters:
(1) Perfusion (measured in ml/min/ml); (2) Peak enhancement intensity (PEI, measured in HU); (3) Time to peak (TTP, measured in s); (4) Blood volume (BV, measured in ml/100 g); (5)Mean transit time (MTT) (sec),
Along with colour maps of the five kinetic parameters, time attenuation curves (TACs) for the input artery and tumour were generated.
The mean value of each parameter was calculated for every node separately.
Since all patients underwent neck dissection/ respective tumor resection, the results were compared with the histological analysis of resected nodes.