AJR:184, January 2005 113 AJR 2005;184:113–120 0361–803X/05/1841–113 © American Roentgen Ray Society Roy et al. MRI of Renal Neoplasms Significance of the Pseudocapsule on MRI of Renal Neoplasms and Its Potential Application for Local Staging: A Retrospective Study Catherine Roy, Sr. 1 Sofiane El Ghali 1 Xavier Buy 1 Veronique Lindner 1 Herve Lang 1 Christian Saussine 2 Didier Jacqmin 2 Received February 21, 2004; accepted after revision June 14, 2004. 1 University Hospital of Strasbourg, Radiology B Chirugie A, Hopital Civil CHU, 1, place de l’hopital BP 426, Strasbourg, Cedex, France 67091. 2 University Hospital of Strasbourg, Hopital Civil–Urology, Strasbourg, Cedex, France 67091. OBJECTIVE. The purpose of our study was to evaluate the role of MRI in showing a pseudocapsule for local staging of renal tumors, and its potential application to select patients for partial surgery. MATERIALS AND METHODS. Eighty tumors (73 renal cell carcinomas [RCCs] and seven oncocytomas) were preoperatively evaluated by MRI. MRI findings were assessed with a special focus on perinephric fat and pseudocapsule. Correlations were performed with patho- logic staging after surgery. RESULTS. At pathology, a pseudocapsule was recognized in 79 cases. Twenty-three RCC were staged pT3a (21 clear cell; two papillary). MR images exhibited a pseudocapsule in 90% of cases as a hypointense rim surrounding the tumor on T2-weighted images. MRI findings concerning isolated analysis of the pseudocapsule for differentiating stage T1/T2 from T3a were sensitivity: 86%, 50%; specificity: 95%, 92%; positive predictive value: 95%, 33%; neg- ative predictive value: 88%, 92%; and accuracy: 93%, 89%, for clear cell and papillary types, respectively. For stage T3a, with both abnormalities of the pseudocapsule and perirenal fat, re- sults were, for overall RCC sensitivity: 84%; specificity: 95%; positive predictive value: 91%; negative predictive value: 91%; and accuracy: 91%. CONCLUSION. The identification of the pseudocapsule offers an additional value for lo- cal staging by MRI. The presence of an intact pseudocapsule is a sign of lack of perinephric fat invasion. It is more likely to predict that the tumor can be removed by partial surgery. enal cell carcinomas (RCC) repre- sent 1–3% of all visceral neo- plasms. Due to both technical improvements and the increasing number of imaging examinations performed, many renal neoplasms of small size and early stage incidentally are detected for unrelated indications. These changes in the manifesta- tion of renal tumors have stimulated a grow- ing trend toward nephron-sparing surgical techniques [1], as current data show survival rates comparable to those associated with rad- ical nephrectomy [2]. Patients with RCC con- fined to the kidney have the best prognosis [3]. Although controversy remains regarding the use of partial nephrectomy for localized incidentally discovered small renal neo- plasms in patients with a normal contralateral kidney, nowadays in many centers, this type of surgery is performed in the clinical situa- tion to spare as much normal parenchyma as is possible [4]. However before conservative surgery is performed, an accurate preopera- tive imaging study is required to acquire all the needed anatomic information and to stage the neoplasm [5]. In the early reports of MRI, the presence of a hypointense rim surrounding RCCs corre- sponding to a so-called “pseudocapsule” was recognized [6–8]. It seems to be a pathologic feature frequently seen in the early stages. The identification of a pseudocapsule has been pro- posed as a sign for a localized intrarenal lesion so that partial nephrectomy could be indicated when an intact pseudocapsule is detected. However the meaning of this imaging feature has not yet been clearly established. The purpose of this study was to analyze the role of MRI in showing the presence of a pseudocapsule in renal tumors, and to deter- mine its significance and implication for the local staging of renal masses. Materials and Methods Patient Population Between February 2000 and June 2003, 77 patients (17 women, 60 men) with an age range of 23–75 years (mean ± 52 [SD] years), were re- R