Renal tumor size: comparison between computed tomography and surgical measurements

OBJECTIVE: We studied the agreement between renal tumor size as assessed on computed tomography (CT) before surgery and that measured during histopathological examination on the radical nephrectomy specimen. METHODS: We retrospectively analyzed the records of 100 consecutive patients treated with radical nephrectomy for a renal tumor. The tumor size was determined in all patients by the largest diameter shown within the month before surgery on contrast-enhanced CT and as measured postoperatively by the pathologist. A possible influence of the clinical and pathological parameters was assessed in a multivariate analysis. RESULTS: CT estimate and surgical measurement of tumor size were highly correlated (r = 0.9; p<0.001). Median (range) tumor size was 70.0 mm (13-180) and 60.0 mm (10-180) as measured, respectively, on CT and in the specimen, with a significant difference (p = 0.005). Multiple regression did not reveal any significant influence of tumor side, location, type, nuclear grade as well as patient gender, body mass index and radiological center (p>0.3 in all cases). The extent of difference between CT and surgical measurements was significantly influenced by the surgical size of the tumor (p = 0.03): the smaller the tumor, the more the CT overestimated the tumor size. If nephron-sparing surgery had been planned for tumors equal to or less than 40 mm, 24 patients would have been selected following the CT estimate, while 27 patients would have met this criterion on the surgical measurement. CONCLUSION: Renal tumors were statistically smaller than the estimate from CT, although this was not systematically the case. This should be kept in mind when issuing recommendations on the optimal cutoff size value under which nephron-sparing surgery is considered equivalent to radical nephrectomy.