Transcatheter Arterial Embolisation of Renal Angiomyolipomas Using an Ethanol-Lipiodol Mixture

Hong Kong J Radiol 2024;27:Epub 15 November 2024

 

 

Department of Radiology, North District Hospital, Hong Kong SAR, China

 

  Full Article in PDF

Renal angiomyolipoma (AML) is a benign mesenchymal neoplasm of the kidney. It is broadly classified into two types, either sporadic (80%) or associated with tuberous sclerosis complex (TSC) [20%] which is an autosomal dominant disease with multisystem involvement.[1] [2] AMLs are composed of an angiomyogenic component (blood vessels and smooth muscle) and a lipomatous component (fat).[3] The diagnosis of renal AML is based on the presence of macroscopic fat.[4] [5] The differential diagnosis includes fat- and calcification-containing renal cell carcinomas, which are unusual.[6] [7] Therefore, a fat-containing renal lesion without calcifications or other atypical features can usually be diagnosed as AML based on its radiological features.

 

The major complication of AML is spontaneous tumoural rupture leading to retroperitoneal haemorrhage into the subcapsular and perirenal space (Wunderlich syndrome), which can be life-threatening.[8] Risk factors for tumour rupture include large size, multifocality, and aneurysm formation.[9] [10] Once the greatest tumour dimension is >4 cm, there is a greater incidence of symptoms which include bleeding and flank pain.[9] In our institution, transcatheter arterial embolisation (TAE) is employed for the treatment of AML in acute haemorrhage due to spontaneous rupture and high-risk AMLs that are considered suitable for prophylactic treatment (i.e., lesion dimension >4 cm and/or tumoural aneurysm ≥5 mm) with multidisciplinary agreement for selected cases. There is increasing use of TAE as prophylactic treatment for non-ruptured AMLs, giving its advantages of a low complication rate, avoiding general anaesthesia, less invasiveness, renal function preservation, and satisfactory outcome,[11] [12] [13] defined as absence of residual tumoural stain on digital subtraction angiography (DSA).[14] [15]

 

There are multiple embolisation agent options. In our institute, we use a 2:1 mixture of ethanol and Lipiodol. Our study aimed to identify the outcome of TAE of renal AML using this mixture, by documentation of the lesion volume reduction and any complications. We also aimed to identify any tumoural factors associated with volume reduction after embolisation.

 

We retrospectively identified all consecutive cases of renal AML TAE at North District Hospital in Hong Kong from January 2016 to December 2022 by reviewing the electronic records of the radiology department. Cases which ethanol and Lipiodol were not used as the embolic agent were excluded. Cases of urgent TAE due to presentation with acute haemorrhage were also excluded as the primary aim for this group of patients is haemostasis instead of volume reduction of the lesion.

 

Patient demographics and clinical data were reviewed in the electronic health records, including sex, age, presence or absence of TSC, presenting symptom, renal function tests (before and after TAE), procedure details, and hospitalisation record.

 

AML volume and percentage reduction were assessed from the latest computed tomography (CT) or ultrasound (US) before and after embolisation. The tissue components of each AML were determined on the latest pre-embolisation CT by consensus from three radiologists (with 12, 5, and 4 years of experience, respectively), who were blinded to clinical details and classified each AML as predominantly angiomyogenic (>50% soft tissue and/or blood vessel components) or predominantly lipomatous (>50% fat components). AML volume was calculated based on the x-, y-, and z-axis diameters (i.e., width, height, and length, respectively) on CT or US examination, with the equation to calculate the ellipsoid volume of v = xyzπ/6.

 

Figure 1 illustrates the embolisation procedure. TAE was performed through the common femoral artery with a 5-Fr angiographic catheter. Flush aortography and renal arteriography were first performed using DSA to identify the location and number of renal arteries and the arterial supply to the AML (Figure 1a). The arterial feeder to the AML was selectively catheterised (Figure 1b). TAE was performed (Figure 1c) by injecting the ethanol-Lipiodol mixture into the arterial feeder with a 1-mL syringe intermittently. Then, renal arteriography was performed to confirm that there were no residual arterial feeders to the lesion. Angiographic success was achieved when vascular stasis and the absence of visible arterial feeding vessels resulted (Figure 1d). The total volume of mixture injected depended on the endpoint of angiographic stasis and absence or occlusion of other feeders, and should not reach the maximum volume allowed according to the patient’s weight.

 

Figure 1. Arteriogram from a 59-year-old female patient with renal angiomyolipoma (AML). (a) Marked vascularity of the AML in the superior pole of the left kidney (arrow). (b) Selective catheterisation to the arterial feeder of the AML (arrow). (c) Transcatheter arterial embolisation of the AML in the superior pole of the left kidney (arrow) performed by injecting a mixture containing ethanol and Lipiodol in a ratio of 2:1. (d) Post-embolisation angiogram demonstrates complete non-opacification of the AML (arrow) and stasis of contrast flow to the lesion representing technical success.

 

Figure 2 illustrates technical non-success in the first round of TAE, with technical success after repeated embolisations. The angiographic catheter was then removed, and haemostasis was achieved by manual compression. After haemostasis at the puncture site was achieved, patients were transferred to the urology ward for observation and were usually discharged the day after the procedure if no major complications occurred.

 

Figure 2. Arteriogram from a 37-year-old female patient with renal angiomyolipoma (AML). (a) Marked vascularity of the AML in the inferior pole of the left kidney (black arrow). (b) After the first round of embolisation, repeated angiography shows that there was still a persistent arterial feeder (white arrow) supplying the AML, indicating technical non-success. (c) After repeated embolisation with the ethanol-Lipiodol mixture, angiography demonstrates complete non-opacification of the AML (black arrow), indicating technical success.

 

Technical success, non-target embolisations, and major and minor complications were defined according to the quality improvement guidelines for TAE by the Society of Interventional Radiology (SIR) Standards of Practice Committee.[15] Technical success was defined as absence of residual tumoural stain on complete DSA. Major complications were defined as those events that resulted in prolonged hospitalisation, permanent renal damage, transfusion, or death. Minor complications included events that may have caused patient discomfort or some morbidity but did not meet the criteria for major adverse events.

 

Patients were followed up in the urology outpatient clinic of our hospital, with follow-up radiological examination including CT or US. Recurrence was defined as an increase in tumour volume on follow-up imaging and/or recurrent symptoms that required repeated TAE of the previously embolised tumour after 6 months of follow-up. Tumoural volume reduction of >50% after TAE was considered a good response, whereas reduction of ≤50% after TAE was considered a poor response.

 

Statistical analysis was performed using commercial software SPSS (Windows version 26.0; IBM Corp, Armonk [NY], United States). Categorical variables were expressed as frequencies and percentages, and quantitative data were expressed as mean ± standard deviation (SD). Fisher’s exact test was used to compare the categorical variables, and an independent sample t test was used to compare the continuous variables between groups. Pre- and post-procedure AML size and patient’s creatinine levels were compared using a paired t test. A p value of < 0.05 were considered to indicate statistical significance.

 

We identified 32 patients eligible for the study, with a mean age of 55.2 years (range, 30-73). Among these patients, 25 (78.1%) were female and seven (21.9%) were male. Two (6.3%) patients were diagnosed with TSC (Table 1).

 

Table 1. Clinical and lesion characteristics of all patients with good versus poor response to transcatheter arterial embolisation.

 

A mean volume of 3.61 mL (SD = 4.87) of the ethanol-Lipiodol mixture was injected for each AML. The average procedure time was 53.8 minutes (range, 22-143), with the starting time and ending time defined by the first and last fluoroscopy images. Clinical characteristics of the patients are summarised in Table 1.

 

All the cases (100%) were technically successful, which was greater than the suggested threshold of 90% according to the SIR reporting standards.[15] A paired t test was conducted to determine the effect of TAE on each patient’s serum creatinine level. The results indicated no significant difference between the creatinine level before and after TAE (mean ± SD = 63.67 ± 12.56 μmol/L vs. 64.8 ± 13.34 μmol/L; p = 0.342), suggesting no significant renal function impairment attributable to embolisation in this series.

 

Complications were classified as major and minor.[15] Only one patient had a major complication, with a small dissection (1.5 cm in length and 0.2 cm in thickness) at the middle part of the left renal artery identified on postprocedure angiogram (Figure 3a). A total of 4000 IU intra-arterial heparin was injected immediately. Follow-up CT showed the small dissection of the left renal artery (Figure 3b). After discussion with the urologist and in view of the small size of the dissection, conservative management was decided upon. On repeated follow-up CT 15 months after the procedure, the dissection remained static (Figure 3c). The overall major complication rate was 3.1%, which was lower than the suggested threshold of 6% according to the SIR reporting standards.[15] Four (12.5%) patients had minor complications which were all cases of post-embolisation syndrome[16] characterised by fever and pain.

 

Figure 3. A 68-year-old female patient with left renal transcatheter arterial embolisation complicated by renal artery dissection. (a) Post-embolisation angiogram showing a small dissection (1.5 cm long and 0.2 cm wide) in the left main renal artery (arrow). Conservative management was decided upon. Sagittal reformat of follow-up computed tomography the next day (b) and after 15 months (c) showed that the dissection remained unchanged (arrow in [c]).

 

The mean hospitalisation days were 1.7 (SD = 1.6) after embolisation. Four (12.5%) patients had hospitalisations >2 days, including the day of admission to the ward before doing the procedure. Among these four patients, one case was complicated by renal artery dissection as discussed above, and three cases were complicated by post-embolisation syndrome, which responded to analgesics.

 

All 32 patients had follow-up CT or US with a mean imaging follow-up duration of 23.8 months (SD = 18.7). All AMLs showed reduction in volume with a mean of 59.8% (SD = 25%) observed (Figure 4). A paired t test indicated that there was a statistically significant difference between AML size before and after TAE (mean ± SD = 148.45 ± 332.96 cm³ vs. 62.63 ± 147.43 cm³; p = 0.015).

 

Figure 4. Angiomyolipoma volume before and after transcatheter arterial embolisation. Every single line represents each case of renal angiomyolipoma, demonstrating the volume change after embolisation.

 

The clinical and lesion characteristics were analysed to explore the potential association with treatment response (i.e., percentage of AML volume reduction). Table 1 summarises the clinical and lesion characteristics with good versus poor response to TAE. Statistical analysis revealed that a predominant angiomyogenic component of AML was significantly associated with good response to TAE (p = 0.002). Although the mean of pre-TAE AML volume of the good response group was greater than that of the poor response group, it did not reach statistical significance (186.89 cm³ vs. 92.27 cm³; p = 0.358). Similarly, the mean volume of ethanol-Lipiodol mixture injected was higher in the good response group; however, this did not reach statistical significance (4.54 mL vs. 2.25 mL; p = 0.130). There were no significant differences in patient age (p = 0.533), sex (p = 0.403), or TSC incidence (p = 1.000) [Table 1].

 

None of the cases required further urgent TAE for the same AML. Two (6.3%) patients had repeated elective TAE for the same AML after multidisciplinary meeting discussion. One was due to residual large size of the lesion, while the other one was due to residual large size as well as occasional flank pain suspected to be caused by intermittent bleeding despite no active contrast extravasation on serial CT. Details of the two cases are summarised in Table 2.

 

Table 2. Details of two cases requiring elective repeated embolisation.

 

The study shows that TAE with an ethanol-Lipiodol mixture is effective and safe in the prophylactic treatment of renal AML, with a high technical success rate in our centre (100%) and a low major complication rate (3.1%).

 

Oesterling et al[9] reported the correlation of symptoms and AML size, showing that 82% of AML with diameter >4 cm were symptomatic. Further evidence from Yamakado et al[17] supported these findings, with a lesion size >4 cm predicting the risk of rupture with a sensitivity of 100% and specificity of 38%, and an aneurysm size of 5 mm predicting the risk of rupture with a sensitivity of 100% and specificity of 86%. In our institution, we adopted these criteria (AML dimension >4 cm or pseudo-aneurysm size ≥5 mm) for prophylactic AML embolisation, and emergency embolisation for those presenting with acute haemorrhage.

 

Various embolic agents have been used for embolisation of AMLs, including pure ethanol, ethanol-Lipiodol mixture, polyvinyl alcohol particles, absorbable gelatin powder (Gelfoam; Pfizer Inc, New York [NY], United States), N-butyl cyanoacrylate, and coils.[18] [19] [20] In our centre, the ethanol-Lipiodol mixture is the preferred agent for embolisation. Ethanol is a potent liquid embolic agent that can cause permanent arterial thrombosis and endothelial damage resultant in necrosis of viable tissue,[21] whereas Lipiodol is iodised poppyseed oil, being radiopaque under fluoroscopy, providing better control of injection and preventing reflux.[22] The optimal ratio of the ethanol-Lipiodol mixture has been discussed in different literature, ranging from 2:1 to 4:1.[22] [23] [24] [25] [26] In an animal study by Gao et al27, the effect and safety of TAE with various volume ratios of ethanol and Lipiodol in a rabbit VX2 tumour model were investigated, showing that the volume ratios of ethanol to Lipiodol from 1:2 to 4:1 were equally effective. The information gleaned from these results could provide insight to future research on the optimal ratio of ethanol-Lipiodol mixture in TAE of human renal AML. In our centre, we adopted a ratio of 2:1 ethanol to Lipiodol after balancing and optimising the therapeutic embolic effect and fluoroscopic visualisation.

 

The drawback of using ethanol is the potential risk of non-targeted embolisation and alcohol toxicity, which include central nervous system depression, haemolysis, and cardiac arrest. Monitoring for systemic toxicity is crucial when the dose of ethanol >1 mL/kg.[24] In our centre, we do not inject ethanol >1 mL/kg (i.e., 60 mL of ethanol for a 60 kg adult). In our study, we injected 1 mL to 26.0 mL (mean = 3.61) of 2:1 ethanol-Lipiodol mixture, which is below the limit.

 

The study also identified AML volume reduction percentage associated with the dominant angiomyogenic component, which was compatible with published literature.[28] [29] This also correlates with the theory that ethanol causes permanent arterial damage, and hence, tumoural necrosis in the angiomyogenic component of AML, and therefore the larger the angiogenic component of the lesion, the better treatment response. This is demonstrated in Figure 5 showing pre- and post-embolisation CT, with Lipiodol predominantly staining the angiomyogenic component of the AML.

 

Figure 5. (a) Computed tomography (CT) of a 59-year-old woman with an exophytic fat-containing lesion (arrow) in the upper pole of the left kidney, diagnosed as renal angiomyolipoma (AML). (b) Follow-up CT of the same patient 11 months after transcatheter arterial embolisation of the AML, with Lipiodol stain (dashed arrow) in the angiomyogenic component. Note the lipomatous component (arrow) is relatively free of Lipiodol stain. There was a significant lesion size reduction, particularly of the angiomyogenic component.

 

Previous literature also reported initial AML volume correlates with post-embolisation volume reduction.[28] In our study, we identified that the pre-embolisation AML volume of the good response group was greater than that of the poor response group; however, it did not reach statistical significance (Table 1).

 

TSC-associated AML is known to develop at a younger age and tends to exhibit a much faster growth rate over time than sporadic AML. Multiple studies have demonstrated that, in contrast to sporadic AMLs, TSC-associated AMLs tend to regrow after TAE, with a recurrence rate up to 60%.[30] [31] [32] Furthermore, TSC-related AMLs tend to develop in both kidneys; therefore, medical therapy would be required. In cases of asymptomatic TSC-associated AMLs >3 cm in size, mammalian target of rapamycin inhibitors are considered as the first-line treatment.[33] However, in our study, we observed only two unilateral cases of TSC, and none of the TSC-associated AMLs required re-embolisation or showed an increase in size during the follow-up imaging at 7 months and 49 months post-embolisation, respectively. Nevertheless, further extended follow-up is necessary to conclusively ascertain the absence of AML recurrence or size escalation.

 

We acknowledge this study’s limitations, such as its retrospective nature and the small and heterogeneous population. The imaging follow-up periods were variable, ranging from 3 to 77 months (mean = 23.8). Also, the post-embolisation AML volume assessment was based on CT or US, which could result in measurement differences between the two imaging modalities.

 

TAE of renal AML using a mixture of ethanol and Lipiodol in 2:1 ratio is an effective and safe treatment option that significantly reduces the AML volume. A predominant angiomyogenic component of AML predicts significant AML volume reduction after embolisation.