The usefulness of preoperative CT in colon cancer staging: impact of radiologists' experience.

PURPOSE: To assess the agreement between radiologists in the detection of specific features related to T- and N-stage and evaluate accuracy in colon cancer staging. METHODS: Patients who underwent contrast-enhanced computed tomography (CECT) before surgery were enrolled and evaluated by three radiologists with different experience. Pathological data were used as the reference standard. Tumor location, presence of tumor bulging, fat stranding, lateroconal fascia invasion, enlarged vessels, axial short diameter of the biggest node, shape, enhancement pattern, intranodal necrosis, and cluster were collected. Reliability analysis was performed with κ statistics and intraclass correlation coefficients (ICCs). Logistic regressions were used to determine independent predictors associated with staging. RESULTS: Seventy five patients were evaluated. The reliability analysis was moderate to substantial for tumor location (κ = 0.853), T staging (κ = 0.531), tumor bulging (κ = 0.478), fat stranding (κ = 0.490), lateroconal fascia invasion (κ = 0.436), enlarged vessels (κ = 0.401), the axial short diameter of the biggest node (ICC = 0.732), shape (κ = 0.484), enhancement pattern (κ = 0.431), intranodal necrosis (κ = 0.606), and cluster (κ = 0.358). For all readers, sensitivity was higher for T3 lesions (60-69%) and specificity was higher for T2 and T4 lesions (84.5-90.0% and 82.4-85.1%). The diagnostic accuracy was acceptable for all lesions and among all readers (50.7-92.1%). The lateroconal fascia invasion and enlarged vessels resulted as independent predictor factors (OR = 3.292 and OR = 2.651) for T staging, while nodes' cluster and dimension as independent predictor factors of N staging [OR = 3.798 and OR = 1.083]. CONCLUSION: Reader's experience is one of the most important factors associated with the correct classification of colon cancer. Moreover, CECT can help depict radiological features independently associated with the T and N stages.

When the diameter of the abdominal aorta should be considered as abnormal? A new ultrasonographic index using the wrist circumference as a body build reference.

PURPOSE: To use US to evaluate the normal values of aortic diameter (AD), stratifying the population by age, gender and body build, as measured using wrist circumference (WC). MATERIALS AND METHODS: Between April 2010 and February 2012, consecutive patients ≥ 30 years of age, without history of abdominal aortic aneurysm (AAA) were prospectively enrolled. They underwent an abdominal ultrasonography for reasons other than aorta evaluation. AD was measured at the infrarenal (AD1), intermediate (AD2), and iliac bifurcation (AD3) levels: a diameter ≥ 3 cm was considered as an aneurysm. The maximal aortic diameter (AD(max)) was measured for AAA patients. WC was measured; AD/WC ratio was calculated and presented in percentage: the range of normal values was obtained excluding AAA cases and calculated as mean ± 1.96 × standard deviation. Pearson correlation coefficient was used. RESULTS: We recruited 1200 patients, 15 (1.25%; age range=64-86 years) had AAA. AD ranges of the other patients were: AD1=0.74-1.84 cm, AD2=0.78-1.85 cm, and AD3=0.68-1.76 cm for females; AD1=0.86-2.02 cm, AD2=0.91-2.08 cm, and AD3=0.84-1.95 cm for males. AD2/WC ratio of non-AAA patients range was 4-15%, with only one outlier at 18%, while AD(max)/WC ratio of AAA patients range was 15-35% (p<0.001). ADs were significantly correlated with WC (r=0.253, p<0.001 for AD1, r=0.318, p<0.001 for AD2 and r=0.280, p<0.001 for AD3). CONCLUSION: The definition of normal AD should consider body build. An AD2/WC ratio of 15% may be regarded as a threshold to differentiate AAA- from non-AAA patients. Patients with AD2/WC values comprised between 12% and 15% may be at risk for AAA.