Zheng Zhu1, Xinming Zhao2, Yanfeng Zhao1, Lei Yang1, Jing Zhao1, Jingrui Dai1, Chunwu Zhou1. 1. Department of Imaging Diagnosis, Cancer Institute & Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100021, China. 2. Department of Imaging Diagnosis, Cancer Institute & Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100021, China. Email: xinmingzh@sina.com.
Abstract
BACKGROUND: Benign and malignant primary retroperitoneal tumors (RT) have different pathological manifestations, and overlapping imaging characteristics. This study aimed to evaluate the value of computed tomography (CT) for differentiating benign from malignant RT. METHODS: One hundred and ninety-four patients with clinical and radiographic data were evaluated retrospectively following surgical resection of primary RT. There were 38.1% (n = 74) benign lesions and 61.9% (n = 120) malignant lesions. Categorical variables were tested with a chi-square test or Fisher's exact test for the diagnostic indexes and sensitivity and specificity of CT characteristics. RESULTS: In univariate analysis, the differences in ill-defined margins, irregular surfaces, long diameter >6.75 cm, short diameter >6.25 cm, and solid or mixed texture had statistical significance; the sensitivity and specificity were 44.2% and 91.9%, 70.0% and 62.2%, 68.8% and 60.7%, 59.7% and 74.2%, 87.5% and 35.1%, respectively. In multivariate analysis, a combination of all the above indexes was the best model for differentiating malignant tumors, resulting in the most accurate diagnosis of malignancies with a sensitivity of 77.2% and a specificity of 81.1% (P < 0.0001) when the score was 4. The differences in other findings including CT attenuation, number, and calcification had no statistical significance. The unique characteristics included the spotted enhancement (mottled high density) in schwannoma (P < 0.0001), adipose tissue in liposarcoma (P < 0.0001) and paravertebral location in neurogenic tumors (P < 0.0001). CONCLUSIONS: More accurate differential diagnosis of primary RT can be made through comprehensive analysis of the combined diagnostic indexes of CT. Some specific characteristics of CT can assist in preoperative planning.
BACKGROUND: Benign and malignant primary retroperitoneal tumors (RT) have different pathological manifestations, and overlapping imaging characteristics. This study aimed to evaluate the value of computed tomography (CT) for differentiating benign from malignant RT. METHODS: One hundred and ninety-four patients with clinical and radiographic data were evaluated retrospectively following surgical resection of primary RT. There were 38.1% (n = 74) benign lesions and 61.9% (n = 120) malignant lesions. Categorical variables were tested with a chi-square test or Fisher's exact test for the diagnostic indexes and sensitivity and specificity of CT characteristics. RESULTS: In univariate analysis, the differences in ill-defined margins, irregular surfaces, long diameter >6.75 cm, short diameter >6.25 cm, and solid or mixed texture had statistical significance; the sensitivity and specificity were 44.2% and 91.9%, 70.0% and 62.2%, 68.8% and 60.7%, 59.7% and 74.2%, 87.5% and 35.1%, respectively. In multivariate analysis, a combination of all the above indexes was the best model for differentiating malignant tumors, resulting in the most accurate diagnosis of malignancies with a sensitivity of 77.2% and a specificity of 81.1% (P < 0.0001) when the score was 4. The differences in other findings including CT attenuation, number, and calcification had no statistical significance. The unique characteristics included the spotted enhancement (mottled high density) in schwannoma (P < 0.0001), adipose tissue in liposarcoma (P < 0.0001) and paravertebral location in neurogenic tumors (P < 0.0001). CONCLUSIONS: More accurate differential diagnosis of primary RT can be made through comprehensive analysis of the combined diagnostic indexes of CT. Some specific characteristics of CT can assist in preoperative planning.