Abstract
The widespread use of cross-sectional imaging has resulted in an increase in the frequency of incidentally identified small renal masses (SRMs). With high cancer-specific survival rates following radical nephrectomy, there is an increasing patient cohort at risk of developing a metachronous contralateral tumour. Nephron-sparing surgery (NSS) has been shown to reduce progression of chronic kidney disease, being the impetus to utilize NSS for the management of SRMs in an anatomical or functional solitary kidney. At present, open partial nephrectomy is the gold standard treatment for an SRM in a solitary kidney; however, there are a number of other minimally invasive options, including laparoscopic partial nephrectomy, and in situ ablative procedures. In addition, variables such as warm and cold ischaemia and renal vascular clamping have been shown to affect outcomes. In this review, we summarize contemporary management options focusing specifically on oncological and functional outcomes.
Introduction
Renal cell carcinoma (RCC) is the third most common urological cancer accounting for ∼3–4% of all malignancies in Scotland. 1 Widespread cross-sectional imaging has resulted in a 2.3–4.3% annual increase in RCC detection, 2 with ∼48–66% of diagnoses being incidentally identified small renal masses (SRMs). 3 With cancer-specific survival rates following radical nephrectomy greater than 95% for T1a tumours (<4 cm), 4 there is an increasing cohort of patients surviving extended periods with a solitary kidney. Since the risk of a metachronous contralateral tumour does not decrease with time, 5 this cohort is at high risk for the development of RCC in a solitary kidney. These observations have driven the development and use of nephron-sparing surgery (NSS) for SRM management. NSS reduces the progression of chronic kidney disease (CKD), which has a higher baseline prevalence than previously thought and is associated with increased cardiovascular risk and mortality. 6
An absolute indication for NSS is an SRM in an anatomical or functional solitary kidney, to maximize functional preservation while maintaining oncological control. Open partial nephrectomy (OPN) is the gold standard treatment for an SRM in a solitary kidney. However, minimally invasive options including laparoscopic partial nephrectomy (LPN) and in situ ablative procedures such as cryotherapy, radiofrequency ablation (RFA), and high-intensity focused ultrasound (HIFU) have gained popularity. Technique-specific oncological control, complication and renal preservation rates are important considerations. Many SRMs are benign, or slow growing with low metastatic potential, and detected in elderly patients with significant co-morbidities. For these patients, a period of active surveillance is an attractive option as the risks of intervention may outweigh the benefits of oncological control. Renal mass biopsy may facilitate the clinical decision-making process.
At present, there are no established treatment guidelines for the choice of intervention. In this review, we summarize the contemporary management options, focusing specifically on oncological and functional outcomes. Details relating to the technical aspects of each intervention are beyond the scope of this review.
Active surveillance
Despite the increased incidence of SRMs, only a minority are high-grade or show other potentially aggressive features. In a large series (n = 2935), 30% of tumours below 4 cm in maximum dimension were benign and over 87% of those diagnosed as clear cell RCCs were lowgrade. 7 A landmark study followed 40 incidentally detected SRMs under 3.5 cm for a mean period of 3.25 years (range, 1.75–8.5). 8 Twenty-two of 26 tumours (84.6%) eventually excised after an average of 3.8 years (range, 1.8–8.5 years) were histologically confirmed to be RCCs. Tumour growth rates were variable, but the overall mean linear growth rate for all tumours was 0.36 cm/year (range, 0–1.1 cm/year). Nineteen tumors grew at a rate of less than 0.35 cm/year and no patient developed metastatic disease. Most SRMs had little or no change in size, and a small proportion with rapid growth rates were thought to be incurable despite early detection and treatment.
Other retrospective studies reported similar findings. Although follow-up periods were relatively short and patient numbers were limited, results consistently suggest that a large proportion of incidentally detected SRMs have a slow growth rate and indolent clinical course, 9 unlike larger masses. 10 However, a lack of growth per se does not indicate benign pathology. 11 Despite an increase in early intervention, the overall mortality rates for RCC have not fallen. 12 This could be due to lead time bias, but may also suggest that many SRMs are slow growing and have limited malignant potential.
Although there is no good quality evidence for active surveillance of SRMs in solitary kidneys, an initial strategy of observation for tumours up to 3–4 cm may be appropriate for the elderly or those with significant co-morbidities. It should be noted, however, that 1.4% of tumours 1 cm or smaller present with metastatic disease, and 3.5% of these patients die of RCC within five years of diagnosis. 13 These observations suggest that many clinically localized tumours larger than 1 cm have micrometastases at presentation.
Open partial nephrectomy
The evidence for OPN in the management of unilateral SRMs (<5 cm) in patients with a normal contralateral kidney is well established. OPN is now also considered the standard of care for renal tumours in a solitary kidney.14–16 An early large series (n = 76) with 28 year follow-up reported high early complication rates (24%), reflecting the technically challenging nature of NSS. Thirteen percent developed acute kidney injury (AKI), with temporary renal replacement therapy (RRT) in only one case and proteinuria in 16% of cases. Chronic renal failure (CRF) (serum creatinine >2 mg/dL) was observed in 13% cases and thought to be due to a larger tumour size (40% cases ≥5 cm). Local recurrence rates (11%) were higher than published rates for bilateral tumours and were thought to be due to the imperative nature of the procedure in the solitary kidney. Reported cancer-specific survival rates were 81% and 64% at five and 10 years, respectively.
The largest series of OPN in the solitary kidney (n = 400) reported high preoperative renal insufficiency (46%). 17 Although a postoperative creatinine rise was seen in 25% of cases, only 4% required immediate RRT; of these only two patients needed long-term RRT. Factors affecting short-term renal function included renal ischaemia time and hypothermia, the percentage of kidney tissue remaining, patient age and the timing of the contralateral nephrectomy. However, in the long term, it was only the percentage of renal tissue remaining and patient age that affected renal function as a result of increased susceptibility to hyperperfusion and glomerulosclerosis, respectively. High rates of long-term functional preservation (95.5%) were achieved, and only 5% of patients developed end stage renal failure (ESRF) requiring RRT at a mean of 3.6 years after surgery. Cancer-specific survival rates were 89% and 82% at five and 10 years, respectively.
Other smaller series with variable follow-up (range, 5–13 years) reported similar functional and oncological outcomes.15,18,19 Observed AKI rates were variable (range, 0–9%), with temporary RRT required only in a few cases (range, 0–4%) and low rates of long-term RRT for ESRF (range, 0–0.7%). In a report with higher rates of postoperative CKD (76%), this was attributed to a higher rate of preoperative CKD (55.3%) and larger tumour bulk. 19 Positive margin rates of up to 15% were observed in one series; 15 however, five-year overall (range, 59.6–68%) and cancer-specific (range, 77.5–88%) survival were comparable to initial series. In a study specifically reporting outcomes of repeated OPN procedures on a solitary kidney (n = 25; median, 4 tumours) 20 perioperative complication rates were high (52%). Although three patients required postoperative RRT and a significant decrease in renal function was observed within three months of surgery, this was restored at one year. Eight patients required further surgical intervention de novo tumour recurrence at a median time of 36 months. The average metastasis-free survival rate was 95% at 57 months (median, 50; range, 3–196).
The effect of intraoperative renal ischaemia
In addition to a loss of functional renal tissue following NSS, intraoperative renal vascular clamping and subsequent ischaemic insult to the solitary kidney has been shown to affect functional outcomes. In a study of ischaemia during OPNs in solitary kidneys (n = 89), the mean immediate postoperative decrease in glomerular filtration rate (GFR) with no ischaemia, warm ischaemia (mean, 12 minutes) and cold ischaemia (mean, 33 minutes) was 29%, 37% and 45%, respectively (P < 0.01). 16 However, there was no statistically significant difference in late GFR drops between groups. Although cold ischaemia and vascular risk factors were associated with immediate GFR decline, no single factor affected late GFR values. Local recurrence (18%) was associated with positive margins and tumour stage and was the single most important factor for ESRF due to completion nephrectomy (6%). The high recurrence rate was attributed to a large number of non-organ-confined (T3) tumours (23%). Three patients (4%) developed ESRF without local recurrence.
A comparison of cold ischaemia (n = 300) or warm ischaemia (n = 360) 21 revealed significantly longer cold ischaemia times (44 minutes cold versus 22 minutes warm; P < 0.001), without significant difference in median GFR decrease at three months. Longer ischaemia time, lower baseline GFR and less parenchymal preservation were risk factors for AKI. Increasing age and tumour size, lower preoperative GFR and longer ischaemia time were significantly related with a long-term decrease in GFR. However, ischaemia time was no longer significantly related to postoperative GFR in a multivariate analysis including the percentage of parenchyma spared.
These studies suggest that although warm and cold ischaemia affect immediate postoperative renal function, there are no long-term effects on GFR decline,16,17,21,22 contradicting a large study on the renal effects of vascular clamping in patients with solitary kidneys.23,24 Warm and cold ischaemia greater than 20 and 35 minutes, respectively, were associated with an increase in immediate postoperative complications as well as a higher incidence of AKI and CRF (P = 0.002 and 0.003, respectively). Warm ischaemia greater than 20 minutes was associated with an increased risk of CRF (41% versus 19%; P = 0.008), increase in creatinine greater than 0.5 g/dL (42% versus 15%; P < 0.001) and permanent RRT (10% versus 4%; P = 0.145). These observations are supported by data from 104 partial nephrectomies on solitary kidneys, both with (n = 29) and without (n = 75) renal hilar clamping.25,26 Most (97%) hilar clamping was carried out under cold ischaemia (median, 25 minutes), and an absence of hilar clamping resulted in a significantly smaller fall in early postoperative GFR (11.8% versus 27.7%) and late GFR (60.9% versus 17.7%). Although hilar clamping was associated with a higher positive margin rate, local recurrence rate and five-year cancer-specific survival was similar between the two groups.
Laparoscopic partial nephrectomy
A minimally invasive alternative to OPN is LPN, offering advantages including shorter hospital stay and quicker convalescence, with equivalent functional and oncological outcomes. One of the first published LPN series reported overall and cancer-specific five-year survival rates of 86% and 100%, respectively, 27 which is comparable to the OPN series. The procedure is technically demanding, and a decline in complication rates in the latter series reflects the operative learning curve. 28 The first largest series of 22 cases of LPN in a solitary kidney 29 reported a mean tumour size of 3.6 cm (range, 1.4–8.3); mean warm ischaemia time was 29 minutes (range, 14–55). A 27% fall in GFR appeared proportionate to the median amount of kidney parenchyma excised, and only one patient required temporary RRT. At a median follow-up of 2.5 years (range, 0.5–4.5 years) cancer-specific and overall survival was 100% and 91%, respectively.
Recently, studies have compared OPN and LPN in a solitary kidney, including the effects of ischaemia. In a study exploring the effect of hilar clamping in LPN (n = 43) and OPN (n = 319) in a solitary kidney, 24 the median warm ischaemia time was 21 minutes (range, 4–55). A longer warm ischaemia time was associated with AKI (19%) and CKD, and a threshold of 25 minutes was found to be the best discriminator of renal functional outcomes. This significant result remained when a multivariable analysis was carried out adjusting for preoperative GFR, tumour size and type of partial nephrectomy. In a further study of LPN (or robot-assisted LPN), which included 47 solitary kidneys, 30 there was no difference in preoperative GFR or the prevalence of a solitary kidney between the clamped and unclamped groups. Specifically, for the solitary kidneys, vascular clamping was associated with a median decrease in GFR (21% versus 4.4%).
Probe ablative procedures: RFA and cryoablation
Renal cryoablation and RFA, delivered using percutaneous, open or laparoscopic techniques, allow in situ SRM treatment without concerns for renal ischaemia and advanced surgical/laparoscopic skills. Furthermore, patients with multiple tumours due to inherited tumour syndromes, solitary kidneys and/or serious medical co-morbidities are suitable candidates. Long-term outcome data are lacking for solitary kidneys, however, published initial results and intermediate follow-up data are encouraging with cancer-specific survival rate ranging from 98% to 100%.31,32
The largest series reported RFA for 55 tumours in 30 patients with solitary kidneys 33 using a percutaneous or intraoperative open approach. The average tumour size was 2.0 cm (range, 1.2–5.4). No major complications were reported with good technical success rates. Radiological local tumour control was observed in 26 patients (50 tumours) over an average 25 months (3–47 months). There was no difference in pre- and postoperative calculated creatinine clearance (P = 0.072), systolic (P = 0.102) and diastolic (P = 0.790) blood pressure. Others have reported comparable results,34–36 albeit with slightly higher complication rates 34 and less successfully functional outcomes. However, no patients have been rendered dialysis dependant, and tumour response rates of up to 69% have been observed, although recurrent disease was detected in up to 25% of cases. 35 Using selective embolization and RFA, 10 larger tumours in 13 patients with solitary kidneys have been treated with promising results. 36 There was no recurrence of the treated tumours, although one patient developed a new tumour 18 months after RFA, which was successfully treated with further RFA.
The first large series of percutaneous cryoablation of SRMs in solitary kidneys examined outcomes of 15 tumours (mean diameter, 3.1 cm) in 14 patients. 37 No complications were observed during treatment and no significant change in serum creatinine was observed postoperatively. The mean follow-up was 17 months (range, 2–30 months) for 12 patients, with three patients requiring re-treatment for incomplete tumour ablation, with no radiographic recurrence at the end of the follow-up. In a study of 35 percutaneous procedures performed on 38 tumours on 31 patients with solitary kidneys, 38 mean maximum tumour diameter was 3 cm (range, 1.7–7.3 cm) and mean follow-up was 14 months (range, <1–42 months). Overall, postoperative GFR fell significantly after ablation, but was not significant in 24 patients treated for a single tumour; no patients required dialysis. In 25 patients with three month postoperative follow-up, a decrease in renal function was seen in 15 patients, of which 10 patients had a history of previous renal ablation or partial nephrectomy of the solitary kidney. Hence, an important factor in predicting outcome appears to be previous intervention. The local tumour control rate was 92%.
A comparative study of cryoablation (n = 29) and RFA (n = 36) for SRMs in solitary kidney reported median 15.1- and 38.8-month follow-up, respectively. 39 There were three treatment failures in total; however, patients were re-treated successfully with RFA. No significant clinical impact on renal function was demonstrated postoperatively. Overall, there were more recurrences following RFA (n = 14) than cryoablation (n = 3). Overall cancer-specific, recurrence-free and metastasis-free survival rates for cryotherapy versus RFA were 89% versus 93%, 100% versus 96%, 69% versus 58% and 86% versus 91%, respectively.
A recent meta-analysis of 47 studies representing 1375 kidney lesions treated with cryoablation or RFA 11 demonstrated a higher repeat ablation rate (8.5% versus 1.3%; P < 0.0001) and local tumour progression rate (12.9% versus 5.2%; P < 0.0001) for RFA compared with cryoablation. The higher incidence of local tumour progression correlated significantly with treatment by RFA on univariate (P = 0.001) and multivariate regression analyses (P = 0.003). Metastases were reported less frequently in cryoablation (1.0%) than in RFA (2.5%; P = 0.06). Although long-term data were lacking in the studies analysed, the data suggest that cryoablation results in fewer re-treatments and improved local tumour control as compared with RFA. It also suggests that it may be associated with a lower risk of metastatic progression when compared with RFA. A theoretical advantage of cryoablation over RFA is the reliable management of lesions adjacent to the collecting system and reduced risk of urine leak and fistula formation in animal models. 32
Which NSS technique is the best?
Despite a number of NSS techniques for SRMs, it is unclear which provides the best functional and oncological outcome in solitary kidneys. In a review of 89 patients with 98 renal tumours managed by RFA or OPN under cold ischaemia over a period of 18.1 and 30.0 months, respectively, 40 median tumour size was greater in the OPN group (3.9 versus 2.8 cm; P = 0.001), while the median preoperative GFR was lower in the RFA group. Compared with RFA, patients in the OPN group had a greater decline in GFR soon after the procedure (15.8% versus 7.1%), 12 months after surgery (24.5% versus 10.4%) and at the last follow-up (28.6% versus 11.4%; all P < 0.001). Within the RFA group, patients had recurrence either outside (n = 3) or within (n = 2) the primary ablation zone. Within the OPN group, three patients had local recurrences and two others had positive surgical margins.
A review of RFA (n = 21), percutaneous cryoablation (n = 29) and partial nephrectomy with variable ischaemia times (n = 62) over a six year period 41 did not reveal any significant differences in postoperative GFR between groups (P = 0.91) or change in CKD stage (P = 0.87). Similarly, a comparison of laparoscopic (n = 8), open (n = 4) and percutaneous cryoablation (n = 11) with LPN (n = 2) and OPN (n = 13) in solitary kidneys revealed comparable functional and oncological results. 42 However, patients in the nephrectomy group had a larger mean tumour size (3.4 versus 2.5 cm), higher mean blood loss, a higher rate of perioperative complications and, interestingly, a significantly longer duration of hospital stay (5.8 versus 1.8 days).
A study with 10-year follow-up compared LPN (n = 48) and laparoscopic cryoablation (n = 30). 43 Tumours were larger in the LPN group (3.2 versus 2.6 cm), there was greater blood loss and more postoperative complications (22.9% versus 6.7%), and postoperative temporary RRT was required in three patients. Within a mean follow-up of 42.7 and 60.2 months for LPN and cryoablation, respectively, local recurrence was detected in the cryoablation group only (n = 4). However, the LPN group had overall better cancer-specific and recurrence free survival rates at three and five years (P < 0.05).
The only comparative study of LPN (n = 36), cryoablation (n = 36) and RFA (n = 29) for SRMs in a solitary kidney 44 revealed cancer-specific and overall survival at two years of 100% and 91.2%, 88.5% and 88.5% and 83.9% and 83.9% for each technique, respectively. Disease free survival was significantly better for LPN than for cryoablation and RFA (100% versus 69.6% and 33.2%, respectively; P = 0.0001). The mean estimated postoperative GFR decrease for LPN, cryoablation and RFA was 26%, 6% and 13% (P = 0.0016). These data suggest that intermediateterm oncological outcomes are superior for LPN despite poorer renal function outcomes than probe ablative procedures.
Renal mass biopsy
Although histological diagnosis aids clinical decision-making for individual patients, renal mass biopsy has been under-utilized due to diagnostic inaccuracy in early studies. Differentiating RCC variants and oncocytomas can be difficult within a limited tissue specimen, and failure rates are high in small (<3 cm) or large (>6 cm) lesions. 45 Hence, biopsy has been reserved to rule out renal metastasis, abscess or lymphoma and to confirm the RCC diagnosis in disseminated malignancy or unresectable tumours.
In more contemporary series, overall positive predictive and negative predictive values of 95.7% and 82.0%, respectively, have been reported for renal mass biopsy, which are higher than the renal imaging performance characteristics. 46 Serious complications are rare, the minor complication rate is less than 5% and no recent cases of tumour seeding are reported. 47 Failure rates due to insufficient material or inaccurate pathological findings are less than 5% in recent studies. 46 Overall, renal mass biopsy has been reported to be over 90% accurate in determining RCC subtypes 46 and may be useful at the extremes of age in determining choice of treatment. Newer molecular techniques may aid histopathological identification of tumour subtypes.
Conclusions
The management of an SRM in a solitary kidney poses a clinical challenge to the urologist, as it remains an imperative indication for NSS. At present, there are no established treatment guidelines for renal mass biopsy and NSS for SRMs in a solitary kidney. A greater understanding of the biology and natural history of renal masses together with advances in minimally invasive techniques offers more management options. Microwave ablation therapy and HIFU are also methods of thermal energy delivery undergoing clinical evaluation.48,49 With a lack of good quality, long-term outcome data for probe ablative procedures, these techniques may be reserved for elderly patients with poor baseline renal function. These techniques need further evaluation as part of a welldesigned prospective study with good quality clinical follow-up data.
Partial nephrectomy offers the best oncological outcomes at the expense of long-term renal function, although functional outcomes appear comparable between open and laparoscopic techniques. The duration of renal ischaemia has an impact on renal functional outcomes, and renal hypothermia ‘buys’ ischaemia time and is largely determined by the technical aspects of surgery. However, LPN can be performed efficaciously and safely with adequate intravenous hydration, pharmacological renal protection, minimizing the ischaemic insult and advanced laparoscopic experience and expertise. 29
Comparative studies of NSS for SRMs in solitary kidneys have been confounded by a number of factors relating to tumour biology and study design. The accurate diagnosis of RCC itself is a challenge due to biological and tissue heterogeneity, 50 and in the absence of a tissue diagnosis, oncological outcomes from probe ablative procedures may be based on radiological features alone. With an increase in the use of renal mass biopsy, a better understanding of the natural history of SRMs will help to tailor evidence-based surveillance strategies, and determine the extent to which treatment alters the natural history of the disease. Further prospective comparative studies, accounting for a heterogeneous patient cohort with differing disease pathologies, will determine the optimal strategy for oncological control while maintaining renal function.
