Discussion
More than 50% of patients with RCC are diagnosed incidentally by abdominal ultrasound or non-enhanced CT for other medical reasons (10,11). Radiological T stage of a renal cancer is a major factor in predicting prognosis and survival in these patients. Some studies have reported that multi-phasic CT or MRI for the diagnosis of RCC have accuracy of up to 90% (12-14). In this study, using a large sample size, we also showed that there was a substantial concordance between radiological and pathological staging in Turkey. In addition, the results showed that CT (75.8%) is being used three times more commonly than MRI (24.2%) for the diagnosis and staging of RCC. The reason why urologists prefer CT rather than MRI for renal imaging may be due to the fact that CT is less time consuming and cheaper than MRI, and the urologists are more familiar with CT images rather than MRI images.
Although the accuracy of CT and MRI in the diagnosis and staging of RCC is generally high, the sensitivity and specificity values are found to be lower in T3 cases compared to other stages (15). Renard et al., retrospectively, investigated the diagnostic accuracy of CT in predicting pT3a RCC in 96 cases (15). Renal sinus fat infiltration, peri-nephric fat infiltration and renal venous wall involvement were assessed by two radiologists specialized in urological imaging and compared with the histopathologic staging. The authors found that assessment of renal tumor extension into perinephric fat remained a difficult task, leading to reduced accuracy in T3a staging. Similarly, the results of our study showed that the diagnostic accuracy of CT or MRI in stage III RCC was lower than other stages. The importance of these findings is that in all RCC types prognosis worsens with stage (2), and this is also true for stage III cases compared to those with stage I and II tumors. Chevinsky et al. reported pathological stage T3a as a poor  prognostic factor in RCC regardless of tumor size and also demonstrated that there was an increased rate of risk of recurrence with perinephric fat invasion compared to those with pT1/T2 tumors (16). Therefore, radiologic under-staging in pT3a cases, will underestimate the risk of cancer recurrence and survival rates, and the patient will be misinformed regarding prognosis of his/her tumor during patient counselling before surgery.
Although, both renal vein invasion and perirenal fat invasion are classified as T3a disease, it was reported in recent studies that patients with pT3aN0M0 RCC with renal vein invasion have a significantly poorer prognosis than those with fat invasion (17). In TNM sub-group analysis of Stage III, we found that the sensitivities of perirenal fat and renal vein invasions were 15.4% and 11.3%, respectively. Although these values are very low compared to values reported in other studies (15), other studies also showed that peri-nephric fat and renal vein invasion in RCC are difficult to evaluate radiologically (5). By using the Surveillance Epidemiology and End Results registries Srivastava et al. reported that from the patients undergoing partial nephrectomy, the estimated proportion up-staged to pT3a was 9.5%, and 19.5% for cT1b, and cT2, respectively (5). In our study, incidence of up-staging from localized stages to stage III was 17.5%, which is consistent with the results reported in literature. Therefore, preoperative imaging in patients with stage III RCC has to be improved. Advanced MRI techniques such as diffusion weighted and perfusion-weighted imaging are being explored for renal mass assessment and staging (18).
Presence of PSM on final pathology creates uncertainty in terms of further management options. Some have performed an immediate or delayed nephrectomy whereas others followed patients without complete nephrectomy (19,20). The incidence of PSM ranges from 0–10.7% in literature and the rate of PSM may be influenced by tumour stage, fat invasion and tumor grade (19,20). Bansal et al, by looking at the partial nephrectomy patients included in the Canadian Kidney Cancer information system database, reported that higher stage (≥T3) and grade were associated with a higher risk of PSM (19). In our study, PSM rate was 8.4% in patients up-staged from localized tumor to pathologically stage III and 12.4% in radiologically stage III cases with concordant pathology (p=0.08). As there is no statistically significant difference between these two rates, during surgery one should also be as cautious as possible in radiologically localized disease in order not to have a PSM.
The present study is limited by its retrospective nature. In addition, central pathological and radiological review could not be performed. Patients were included from different centres and therefore the quality of radiologic and pathologic evaluation is probably variable. However, the aim of our study was to evaluate the accuracy of radiological staging of RCC in daily routine urology practice, rather than assessment of radiological techniques or surgical procedures. So, we did not perform any comparison between radiological techniques, surgical procedures or centers.
There was a substantial concordance between radiological (CT and/or MRI) and pathological T staging in RCC. However, this is not true for stage T3 cases. The reason is that, it is difficult to evaluate peri-nephric fat and renal vein invasion radiologically. Therefore, the sensitivity of preoperative radiological imaging in patients with pT3a tumors is insufficient and lower than the other stages. Consequently, preoperative imaging in patients with T3 RCC has to be improved, in order to better inform the patients regarding prognosis of their disease.