The role of transarterial embolization in the treatment of renal cell carcinoma




To review the role of transarterial renal embolization in our unit, assessing the indications, tolerability and efficacy of this technique for treating renal cell carcinoma (RCC).


Thirty patients undergoing transarterial embolization between 1991 and 1999 were identified and 25 case notes analysed retrospectively.


Most patients (14 of 25) presented with less advanced (stage I–III) RCC who were unable or unwilling to undergo radical surgery; the remainder (11) presented with advanced (stage IV) disease. The embolizing agent was ethanol, usually combined with stainless steel coils (85% of cases). Procedural pain and fever was controlled successfully. The median hospital stay associated with the procedure was 4 days. At the time of analysis six of 11 stage IV and 11 of 14 stage I–III patients were alive (median follow-up 27 and 39 months, respectively). Symptoms from the primary tumour were well controlled. Overall, 17 of 25 (68%) of patients reported no problems while three (12%) required brief hospital admission for treatment of persistent haematuria. Fourteen patients were subsequently re-staged; the primary tumour in two had increased, in seven remained unchanged and in five it decreased. No patients without metastases developed them and metastases in two patients regressed.


Transarterial embolization is associated with minimal morbidity and complications, and subsequent symptom control is good. The effect of palliative embolization on RCC progression is unknown and requires prospective investigation. Presently, there is no role for cytoreductive embolization; it should be included as a treatment option in clinical trials evaluating such options in patients with metastatic RCC.


Renal cancer accounts for ≈ 3% of all diagnosed malignancies and up to 95 000 deaths per year worldwide. The median age of incidence is 66 years [1]. Over 75% of all cases are classified as RCC, which is thought to arise from the proximal tubular epithelium. Symptoms may occur late and as a result up to 30% of patients have distant metastases at presentation, and a further 25% have local spread. The 5-year survival for patients with metastatic disease is < 20%[2]. The basis of treatment for localized disease is surgical resection, as the tumours are relatively resistant to both radiotherapy and chemotherapy [3]. Biological therapies including interleukin-2 and interferon are promising but objective response rates are only 10–15%[4,5].

Transarterial embolization (TAE) of renal tumours was first described in 1973 as a preoperative aid to resecting localized RCC and to palliate symptoms in metastatic disease. While preoperative embolization has been generally abandoned, its palliative role continues. We identified that TAE is also undertaken for patients with less advanced (stage I–III) disease, who are unable or unwilling to undergo radical surgery. In elderly patients with small tumours the morbidity and mortality of nephrectomy needs to be balanced against the risks posed by the tumour. The effect of embolization per se on the progression of RCC is unknown.

The aim of this retrospective study was to assess the role of TAE in our unit. We wished to establish the indications, methods and tolerability of the procedure, and its long-term symptom control and effect on disease progression


A retrospective study was undertaken of patients who had undergone TAE of RCC at the Royal Sussex County Hospital between 1991 and 1999. These patients were identified from radiological databases using procedure codes. Thirty patients were identified and 25 case-notes analysed; five sets of notes could not be retrieved or contained inadequate data for analysis. The variables recorded included patient age at diagnosis, presenting symptoms, methods of diagnosis and staging, tumour stage and the clinical indication for embolization. Medical and nursing notes were analysed to provide evidence of peri- and postoperative pain, fever and analgesic requirements. The outcome measures were procedural morbidity, symptom control, disease progression and survival.


The cohort of patients undergoing TAE was divided into two groups for analysis. The first comprised 11 (median age 73 years) with stage IV RCC (UICC TNM, 1997 [6]) including metastatic (M1) and advanced nodal (N2) disease (Table 1). Most of these patients (eight) underwent TAE for symptoms attributable to the primary tumour. One patient also had TAE before cytoreductive surgery of a vascular tumour; the surgery was abandoned and the patient subsequently received interleukin-2 immunotherapy. Two patients presented with metastases and small (T1) primary tumours, and TAE was used as prophylaxis against future local symptoms.

Table 1. The patient characteristics in the two groups
Patient no.Age, yearsPresenting symptomsStageASA gradeIndication for Embolization
  1. ASA, American Society of Anesthesiologists.

Group 1 (Stage IV)
 273Loin mass320IV2Loin pain
 370Loin pain420IV2Loin pain
 673Bone pain101IV1Prophylaxis
 769Bone pain201IV2Loin pain
 968Loin mass411IV2Loin pain
Group 2 (Stage I-III)
1372Incidental100I2Patient choice
1688Haematuria100I3Patient choice
1775Haematuria200II2Patient choice
1893Haematuria2X0II2Patient choice
1968Haematuria200II2Patient choice
2181Loin mass200II2Patient choice
2268Loin pain300III3Unfit
2384Loin pain310III2Patient choice
2459Haematuria300III2Patient choice

The second group comprised 14 patients with localized disease (stages I–III) who were unable or unwilling to undergo radical nephrectomy (Table 1); they were generally elderly (median age 80 years). Six patients were considered unfit for the procedure and eight declined surgery after full counselling. Of these, one had a progressive neurological condition with a poor prognosis. Another presented with multiple tumours throughout the remaining kidney having had previously undergone a renal resection on the contralateral side for RCC. Nephron-sparing surgery was not considered possible and the patient considered renal dialysis unacceptable.

In 21 of the 25 cases initial imaging of the tumour was with ultrasonography; subsequent staging investigations included chest X-ray in six, CT in 16 and MRI in eight.

The TAE was administered in a dedicated interventional radiology suite by one of three consultant radiologists (73% by one). In all cases the entire kidney was embolized, with no attempt to selectively infarct the tumour. Alcohol was used as an embolizing agent, 85% in combination with stainless steel coils, 8% in combination with sponges and 7% alone. Intravenous pethidine and midazolam was given peri-operatively to all patients. The mean (range) doses were pethidine 60 (25–150) mg and midazolam 8.3 (2–14) mg. One patient also received 75 mg of diclofenac intramuscularly. Antibiotics were given on five occasions as intravenous co-amoxiclav (three), gentamicin (one) and cefuroxime (one).

Postoperative analgesic use varied among patients (Table 2); the use of the more potent drugs declined rapidly in the days after the procedure (Fig. 1). Some patients were apyrexic after embolization, but 15 (60%) developed a fever which resolved in all but one within 3 days of surgery (Fig. 1). The median (range) hospital stay was 5 (1–15) days. In two patients nephrectomy was attempted at open surgery but the procedure was abandoned; excluding these the median stay was 4 days. One patient developed small bowel ileus after surgery, which resolved with conservative management, and he was discharged on day 14.

Table 2. Analgesic agents used after embolization (total drug use recorded; 25 patients 129 total inpatient-days)
AnalgesicUse (patient-days)
Parenteral opiate29
Morphine tablets  3
Liquid morphine  3
Tramadol  3
Diclofenac  6
Figure 1.

The incidence of fever (green bars) and opioid analgesic use (red stippled) after embolization in 25 patients.


All patients were reviewed periodically (at intervals of 1–7 months) in the department of urology at the Royal Sussex County Hospital. At the time of analysis six of 11 patients with advanced stage IV RCC remained alive (median follow-up 27 months) while five had died from the disease. Eleven of 14 patients with less advanced stage I–III disease remained alive (median follow-up 39 months) while three had died. One of these deaths was disease-specific, in a patient with multiple tumours in a single kidney. However, the other two deaths resulted from cardiovascular and neurological conditions unrelated to renal cancer.


Overall, 17 of 25 (68%) patients reported no further problems from their primary tumour. Only three (12%) required hospital admission for treatment of these symptoms (Fig 2). Re-staging investigations were conducted on an ad-hoc basis but overall, 14 of 25 (64%) patients received additional imaging of the abdomen and thorax (CT, ultrasonography and chest X-ray). Only four of 11 of patients with stage IV disease were re-imaged; one showed no change in primary or metastatic tumour volume, another had a reduced volume in the primary tumour alone, while two had apparent regression of metastases (one of whom was also receiving immunotherapy). Ten of 14 patients with less advanced disease were re-imaged; no metastases were identified. In two cases the primary tumour volume had increased, in four it was unchanged and in four it had reduced.

Figure 2.

Figure 2.

The control of symptoms from primary tumour for stage IV (a) and stage I–III (b) RCC. In a, of 11 embolized patients, nine presented with symptoms and two were asymptomatic before embolization. Inb, of 14 patients who declined radical surgery, eight presented with symptoms from the primary RCC and six were asymptomatic before embolization.

Figure 2.

Figure 2.

The control of symptoms from primary tumour for stage IV (a) and stage I–III (b) RCC. In a, of 11 embolized patients, nine presented with symptoms and two were asymptomatic before embolization. Inb, of 14 patients who declined radical surgery, eight presented with symptoms from the primary RCC and six were asymptomatic before embolization.


The two indications for TAE in this study were controlling symptoms attributable to the primary tumour in patients with metastatic RCC, and providing an alternative to radical surgery in patients who would otherwise receive no treatment in less advanced disease.

The role of embolization in managing RCC has always been controversial. Almgard et al.[6] were the first to describe renal embolization in humans in 1973; they suggested the technique would facilitate a subsequent nephrectomy, control troublesome haematuria in patients with localized disease and might improve the disease course in patients with metastases. Kalman and Varenhorst [7] systematically reviewed published series (1973–97) and reported 3225 case histories. Preoperative embolization accounted for 2503 (78%) of these, used to reduce the size and vascularity of renal tumours, and thus provide a mechanical advantage in a subsequent nephrectomy. Despite initial encouraging reports [8,9] other investigators found no benefits for subsequent renal surgery [10,11]. It was also proposed that tumour necrosis in the delay between embolization and nephrectomy would create an autovaccine that would up-regulate the natural antitumour activity of the immune system in patients [12–14]. Subsequent studies found no survival benefit for patients with metastatic disease undergoing embolization and nephrectomy [15,16]. There is a lack of good-quality evidence to determine the efficacy of embolization-nephrectomy, and its use has declined. A postal survey of UK urologists in 1983 identified 60% of respondents who practised embolization-nephrectomy [17], while a similar survey in 1992 recorded < 0.2% of respondents using embolization routinely in managing RCC [11].

Palliative embolization to control local symptoms in metastatic disease accounts for 19% of patients from published case series reviewed by Kalman and Varenhorst [7]. The use of palliative embolization has declined in the UK, approximately halving between 1983 and 1992. Only 35% of UK urologists surveyed in 1992 believed it was a useful technique in this role [11,17]. However, reasonable symptom control has been reported by a variety of retrospective studies; Nurmi et al.[18] reported that 11 of 14 patients with gross haematuria and three of six with severe loin pain were successfully palliated. Another series of 13 patients showed resolution of all haematuria, although three had a relapse [19]. In the present patients symptom control was excellent, with only three of 16 patients who presented with haematuria or loin pain and undergoing embolization requiring hospital admission to control these symptoms during the follow-up. Recently, Onishi et al.[20] compared the outcome of 54 patients with metastatic RCC, 24 undergoing embolization and 30 not; there was a significant survival advantage for those treated.

Embolization for symptom control is not limited to metastatic disease; a few case reports have been published where it has been successfully used to palliate symptoms of RCC in a solitary kidney [21,22]. However there are few published data about symptom control in those who forego radical surgery for other reasons, e.g. advanced age and/or comorbidity. This is surprising, as in the present experience eight of 25 (32%) of all procedures were undertaken for this group of patients. Also, six (24%) of all patients had embolization for asymptomatic localized RCC. There appear to be no published data recording its effectiveness in long-term tumour control for this group of patients.

There are many reports of significant morbidity and mortality associated with TAE. A post-infarction syndrome of fever, pain and nausea for ≈ 36 h is well described, and can be considered a part of the process [23]. However erroneous infarction of bowel, contralateral kidney and spine were also widely reported, with a mortality of 3.3% and serious morbidity of 9.9% recorded by Lammer et al.[24] in a series of 121 procedures. The introduction of ethanol as an embolic agent to replace gel-foam in the early 1980s was associated with a decrease in the complication rate and symptoms of post-infarction syndrome [11,12]. There was minimal morbidity associated with the procedure in the present series and no serious complications. Similarly the post-infarction syndrome was well controlled. Reassuringly, in accordance with another recent study [20], we found the technique safe, and associated with minimal morbidity and a brief hospital stay.

Preoperative embolization to reduce RCC disease progression has been discredited [16] but the effect of embolization per se in prolonging survival (irrespective of symptom control) has not been rigorously assessed. In the present patients most were undergoing embolization for stage I–III disease and might be expected to have a longer lifespan than those with metastatic disease (11 of 14 with stage I–III RCC remain alive at 39 months median follow-up). For these patients, any antitumour effect is welcome in addition to symptom palliation. We identified encouraging disease stability and even apparent regression in most patients re-staged. We are aware that only about half of the patients were re-staged and this retrospective evidence requires validation. Other reports of disease progression after palliative embolization are also of small retrospective studies and are contradictory [19,25,26].

The natural history of small renal masses that are untreated has been examined by Rendon et al.[27]; 13 radiologically detected lesions were imaged for a median of 42 months. Two progressed quickly and five were ultimately resected but the remainder grew slowly (1.32 cm3/year) with no symptoms. This suggests that a proportion of patients undergoing embolization for early-stage disease may not gain additional benefit from this procedure. However, until such patients can be reliably identified, withholding treatment seems unwise.

Contemporary chemo/immunotherapy for metastatic RCC may produce partial or occasionally complete responses in selected groups of patients. The role of cytoreductive nephrectomy in the presence of metastases is controversial. In 2000 The South-western Oncology group reported on a prospective trial to assess the effect on survival of using radical nephrectomy before interferon therapy for metastatic RCC. They showed an overall survival benefit of 4.4 months (12.5 vs 8.1 months) for those undergoing surgery [28]. For patients with a short life expectancy the morbidity of a large procedure may offset the modest survival benefit. One specialist centre reported that only 7% of those with metastatic RCC were suitable for cytoreductive surgery [29]. Thus, there may be a role for cytoreductive embolization for some patients with metastatic RCC, perhaps for those of lower performance status. There is anecdotal evidence of regression of metastatic RCC using embolization and interferon-α[30].

Recent developments in minimal-access surgery are radically changing RCC treatment options and this must be acknowledged in evaluating the contemporary role of embolization. Renal tumours may be treated with laparoscopic nephrectomy (radical and partial), percutaneous cryosurgery and percutaneous ultrasound ablation in those unable or unwilling to undergo open surgery for stage I–III RCC. Selective tumour destruction may be preferable to whole-organ embolization for small lesions, although this issue is unresolved. All these techniques require evaluation as cytoreductive adjuncts to chemo/immunotherapy in metastatic RCC.

In conclusion, contemporary renal embolization has a role in palliating symptoms derived from the primary tumour in patients with advanced (stage IV) disease and for those with less advanced disease who do not receive radical surgery. Embolization is associated with minimal morbidity and complications. Symptom control is good, with 68% requiring no further intervention during the follow-up. The role of cytoreductive embolization is not established and should be included as a treatment option to nephrectomy in clinical trials evaluating treatment options in patients with metastatic RCC.


transarterial embolization.