Background and objectiveBladder cancer is the most common malignancy in urinary system. In clinicalsettings, bladder cancer is classified into non-muscular invasive type (Tisã€Taã€T1) andmuscular invasive type (T2-T4) on the basis of T staging. Non-muscular invasivebladder cancer can be treated by transurethral resection of bladder tumor (TURBT),while muscular invasive bladder cancer is treated by curative cystectomy withsimultaneous pelvic lymphadenectomy. The choice of different treatment for bladdercancer is decided by T staging. Thus, accurate T staging before operation is greatimportance. MRI is the most accurate imaging method to determine the muscularinvasive of bladder cancer. Functional magnetic resonance imaging (fMRI), such asdiffusion weighted imaging (DWI) and dynamic contrast-enhanced MRI (DCE-MRI),achieves a comprehensive study of morphology and function. DWI and DCE-MRIcould be useful to increase the diagnostic performance for T staging of bladder cancer.The purpose of this study is to evaluate the application of conventional MRI, DWIand DCE-MRI for the diagnosis of muscular invasion of bladder cancer and toinvestigate the value of DWI and DCE-MRI comparing with conventional MRI. Materials and methodsFrom October2012to February2014,65consecutive patients with bladdercancer were enrolled in this study. All patients underwent conventional MRI sequence,DWI sequence and DCE-MRI sequence. Two experienced radiologists (R1and R2)independently evaluated the muscular invasion of bladder cancer on conventionalMRI, on conventional MRI+DWI, and on conventional MRI+DWI+DCE-MRI in ablind manner. Receiver operator characteristic (ROC) curve analyses were performedto evaluate the diagnostic performance for conventional MRI only, conventionalMRI+DWI and conventional MRI+DWI+DCE-MRI. Measure the apparent diffusioncoefficient (ADC) value of local lesions adjacent to muscular layer of bladder wall,analyze the ADC value of these lesions by the ROC curve, and calculate thesensitivity and specificity for diagnosis of muscular invasion.Results40cases were finally diagnosed as muscular invasive bladder cancer;25caseswere non-muscular invasive bladder cancer. R1achieved an area under theROC-curve (AUC) of0.859, accuracy75.0%,sensitivity77.5%and specificity72.0%on conventional MRI versus0.907,83.1%,85.0%and80.0%on conventionalMRI+DWI and0.944,92.3%,95.0%and88.0%on conventional MRI+DWI+DCE-MRI. The AUC difference of three methods was statistically significant (p=0.031).For R2these figures were0.861,73.8%,75.0%and72.0%on conventional MRIversus0.917,83.1%,82.5%and84.0%on conventional MRI+DWI and0.936,90.8%,90.0%and92.0%on conventional MRI+DWI+DCE-MRI. The AUCdifference of three methods also was statistically significant (p=0.026). The ADCvalue of muscular layer in muscular invasive bladder cancer was1.52±0.15×10-3mm2/s, the non-muscular invasive bladder cancer was2.21±0.39×10-3mm2/s. WithADC value of1.61×10-3mm2/s as the threshold the sensitivity and specificity of ADCfor the diagnosis of muscular invasion were91.2%and90.4%. ConclusionConventional MRI combined with DWI can improve the diagnostic performanceof muscular invasion of bladder cancer than conventional MRI alone. However, it isfurther improved when combining conventional MRI, DWI and DCE-MRI together.Quantitative measurement of ADC value in muscular layer adjacent to lesions is ofhigher accuracy in diagnosing muscular invasion of bladder wall. When setting ADCvalue of1.61×10-3mm2/s as the threshold, the sensitivity and specificity of ADC forthe diagnosis of muscular invasion are relatively higher. |