- Email: [email protected]
Transarterial embolisation for congenital renal arteriovenous malformations improves clinical condition even with partial obliteration
Clinical Radiology 72 (2017) 1053e1059
Contents lists available at ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
Transarterial embolisation for congenital renal arteriovenous malformations improves clinical condition even with partial obliteration T.-W. Sheng, Y.-C. Wong, C.-H. Wu, L.-J. Wang* Department of Medical Imaging and Intervention, Linkou Chang Gung Memorial Hospital, Chang Gung University, No. 5, Fuxing St., Guishan Dist., Taoyuan City 33305, Taiwan, ROC
art icl e i nformat ion Article history: Received 31 December 2016 Received in revised form 23 June 2017 Accepted 10 July 2017
AIM: To evaluate the efficacy and safety of transarterial embolisation (TAE), used to treat congenital renal arteriovenous malformations (CRAVMs). MATERIALS AND METHODS: The medical records were searched retrospectively to identify patients who underwent TAE to treat CRAVM from January 2003 to August 2015. Patient demographics, clinical presentations, and imaging findings were reviewed. TAE outcomes, including complete or partial obliteration, clinical success, complications, renal function changes, and relapse of symptoms or signs after the final TAE, were assessed. RESULTS: Over the 12-year period, 16 patients (nine male, seven female; median age, 47 years) who underwent 21 TAE procedures to treat 16 CRAVMs were enrolled in the study. The most common clinical presentation was haematuria (81.3%). Thirteen patients (81.3%) had cirsoid and three (18.7%) had aneurysmal CRAVMs. Of the 16 CRAVMs, 11 (68.8%) were peripheral, four (25%) were central, and one (6.3%) was both peripheral and central. The complete obliteration rate was 56.3%. The clinical success rate was 87.5% over a median follow-up period of 398.5 days. Two (9.5%) major complications and 14 (66.7%) minor complications were encountered. No statistically significant change in renal function was evident after TAE. CONCLUSION: TAE was safe and effective when used to treat CRAVM; the complication profile was acceptable and renal function was preserved. TAE improved the clinical condition of CRAVM patients even when obliteration was only partial. Ó 2017 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
Introduction Renal arteriovenous malformations are abnormal communications between the renal arteries and veins anywhere upstream of the capillary level. Aetiologically, the condition
* Guarantor and correspondent: L.-J. Wang, Department of Medical Imaging and Intervention, Linkou Chang Gung Memorial Hospital, Chang Gung University, No. 5, Fuxing St., Guishan Dist., Taoyuan City 33305, Taiwan, Republic of China. Tel.: þ886-3-3281200x3786. E-mail address: [email protected] (L.-J. Wang).
is classified as either acquired or congenital. Acquired renal arteriovenous malformations account for approximately 75% of cases1 and are usually caused by trauma, malignancy, inflammation, surgery, or percutaneous kidney procedures. Congenital renal arteriovenous malformations (CRAVMs) are extremely rare (estimated prevalence, 0.04%)2 and are classified as cirsoid or aneurysmal. Cirsoid CRAVM is characterised by a tangle of dilated and tortuous arteriovenous communications, a so-called vascular nidus, supplied by multiple arteries. Aneurysmal CRAVM, also termed idiopathic or cavernous CRAVM, usually presents as an
http://dx.doi.org/10.1016/j.crad.2017.07.008 0009-9260/Ó 2017 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
1054
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059
arteriovenous fistula with a single arterial feeder. CRAVM may be asymptomatic or associated with gross haematuria, hypertension, or cardiac failure.3 The goal of treatment is to eradicate the vascular lesion completely, with maximal preservation of the renal parenchyma. Open surgery was the most common approach in the past; however, this is associated with a risk of injury to the renal parenchyma because of the peripelvic location of the vascular anomaly, which necessitates extensive branch artery dissection. Today, developments in endovascular techniques have rendered transarterial embolisation (TAE) the first-line treatment. As CRAVM is rare, only a few case series and case reports on TAE used to treat CRAVM have appeared. Therefore, the purpose of the present study was to review retrospectively the outcomes of patients undergoing TAE to treat CRAVM.
Materials and methods Patient population The medical records were searched retrospectively to identify patients who underwent TAE to treat CRAVM from January 2003 to August 2015 in Linkou Chang Gung Memorial Hospital. The institutional review board approved the review of patient medical data, including demographics, clinical presentations, and imaging findings. Informed consent was obtained from all patients prior to TAE. CRAVM diagnoses were based on clinical histories, computed tomography (CT), and angiography. No patient had a history of renal trauma, malignancy, inflammation, surgery, or any percutaneous procedure. Follow-up continued to the time of hospital discharge or the final documented outpatient visit; all follow-up durations were recorded.
TAE procedures Nine radiologists with 3e20 years of experience in interventional radiology performed the TAE procedures. Arterial access was obtained through the common femoral artery under local or general anaesthesia. Diagnostic renal angiography was performed to delineate the CRAVM vasculature (Fig 1). Subsequently, an angiocatheter was advanced to a point as close to the feeding artery as possible, prior to embolisation, to maximally preserve the renal parenchyma. A wide variety of embolic agents were used individually or in combination; these included gelatin sponges, coils, trisacryl gelatin microspheres (Embospheres), absolute ethanol, N-butyl-2-cyanoacrylate (NBCA), and ethylene vinyl alcohol copolymer (Onyx). The choice of embolic agents was based on the characteristics of the vasculature, the catheter position that was attainable, the availability of embolic agents, and the physician’s preference. Angiography was performed after embolisation to confirm successful occlusion. The number of TAE sessions per patient was recorded. Patient complaints made during TAE sessions were reviewed.
Outcome assessment Outcomes assessed after the final TAE session included complete or partial obliteration, clinical success, renal function changes, and relapse of symptoms or signs. Partial obliteration was defined as a <90% reduction in CRAVM volume. Clinical success was defined as the absence of any requirement for further intervention or surgery, as evidenced by the resolution of the initial presentations (e.g., gross haematuria) or an absence of residual CRAVM on postTAE images. Complications developing during each TAE session were classified using the definitions of the Society of Interventional Radiology.4 Major complications included those creating an unplanned need for more care, prolonged hospitalisation, permanent adverse sequelae, or death. Minor complications included those that had no sequelae and required only nominal observation or a short hospital stay. To assess changes in renal function, the serum creatinine (SCr) level was recorded, and the estimated glomerular filtration rate (eGFR) was calculated using the Modification of Diet in Renal Disease (MDRD) formula.5 The pre-TAE SCr level was the latest value obtained before the first TAE. The post-TAE SCr level was the first value obtained at least 1 month after the last TAE. SCr levels recorded within 1 month after the last TAE were excluded from consideration, to avoid any possible confounding effect of obstructive uropathy following haematuria and reversible contrastinduced nephropathy. Relapse during follow-up was recorded in terms of symptoms and signs.
Statistical analysis All statistical analyses were performed using Statistical Analysis System (SAS) software (ver. 9.3). Descriptive values for categorical and continuous variables are presented as numbers with percentages and as medians with interquartile range (IQR), respectively. The Wilcoxon signedrank sum test was used for two-tailed comparison of SCr and eGFR levels before and after TAE. The level of statistical significance was set at p<0.05.
Results Patient characteristics The demographics, TAE characteristics, and outcomes of the 16 patients are summarised in Table 1. Over the 12-year period, 16 patients (nine male, seven female; median age, 47 years; IQR: 35e62.5 years) underwent 21 TAE procedures to treat 16 CRAVMs. The median follow-up period after the final TAE was 398.5 days (IQR: 52.5e1,716.5 days). One patient had only one kidney because of a history of nephrectomy to treat renal cell carcinoma. The most common clinical presentation was haematuria (81.3%); 11 cases of gross haematuria and two of microscopic haematuria. Other symptoms included flank pain in five patients (31.3%), acute urinary retention in four (25%), and hypertension in three (18.8%). Three patients (18.8%) were asymptomatic and no patient exhibited cardiac failure. TAE was performed on
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059
1055
Figure 1 A 59-year-old man with an aneurysmal CRAVM underwent TAE with coil placement (patient no. 13). (a) A CT image shows abnormally dilated and tortuous vessels in the right kidney. (bec) Selective angiography reveals a vascular nidus composed of a dilated and tortuous arteriovenous fistula, with a single arterial feeder and an aneurysmally dilated draining vein. (d) After TAE with placement of coils, the vascular lesion was occluded completely and an inevitable focal renal infarction was evident.
asymptomatic patients on request. The patients were anxious and uncomfortable with conservative management; they were concerned about possible future gross haematuria, shock, and cardiac failure even after the low risks had been carefully explained to them.
TAE characteristics Angiographically, 13 patients (81.3%) had cirsoid CRAVMs and three (18.8%) had aneurysmal CRAVMs. Of the 16 kidney lesions, 11 (68.8%) were located peripherally, four (25%) centrally, and one (6.3%) extended broadly from the peripheral to the central region.
A total of 21 TAEs were performed in the 16 patients. Three underwent repeat TAE (two patients twice and one four times). Microcatheters were used in 13 of the 21 TAEs (61.9%). A wide variety of embolic agents were used either individually or in combination, including coils in 12 patients (57.1%), gelatin sponges in seven (33.3%), NBCA in three (14.3%), Embospheres in one (4.8%), absolute ethanol in one (4.8%), and Onyx in one (4.8%). A combination of coils and gelatin sponges was employed in three procedures (14.3%), while a combination of coils and NBCA was used in one case (4.8%). Liquid embolic agents (i.e., absolute ethanol, NBCA, and Onyx) were used in five procedures (23.8%).
1056
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059
Table 1 The demographics, features of transarterial embolisation (TAE), and outcomes of the 16 patients. Patient no./ age/sex
CRAVM type
TAE no.
CRAVM obliteration
Clinical success
Embolic agents used
Major complications
Minor complications
Follow-up (days)
1/91/F 2/69/M 3/46/M 4/48/F 5/34/M
Cirsoid Cirsoid Cirsoid Cirsoid Aneurysmal
1 1 1 1 1
C P Pa Pb P
Y Nc Y Y Nd
Coils Coils CoilsþGelfoam Embospheres Coils
N Y: PES Y: PES Y: PES N
39 113 15 3266 21
6/36/M 7/51/F 8/39/M
Cirsoid Cirsoid Cirsoid
2 1 4
P/Pb C P/C/C/C
Y Y Y
Y/Y: PES Y: PES Y/Y/Y/Y: PES
1/4 3325 230/109/820/ 556
9/46/F 10/54/F
Cirsoid Cirsoid
1 1
C Pb
Y Y
Gelfoam/Gelfoam CoilsþGelfoam Coils/ CoilsþGelfoam/ Gelfoam/ CoilsþNBCA Coils Absolute ethanol
N N N N Y: pulmonary embolisation N/N N N/N/N/N
N Y: PES
2159 2226
11/66/F 12/24/M 13/59/M 14/73/M 15/16/M 16/26/F
Cirsoid Cirsoid Aneurysmal Cirsoid Aneurysmal Cirsoid
1 2 1 1 1 1
C C/C C C C P
Y Y Y Y Y Y
Onyx Gelfoam/NBCA Coils Coils Coils NBCA
N Y: renal vein thrombosis N N/N N N N N
N N/Y: PES N Y: PES N Y: PES
1274 570 236 471 326 66
CRAVM, congenital renal arteriovenous malformation; C, complete obliteration; P, partial obliteration; Embospheres, trisacryl gelatin microspheres; Gelfoam, gelatin sponge; NBCA, N-butyl cyanoacrylate; Onyx, ethylene vinyl alcohol copolymer; Y, yes; N, no; PES, post-embolisation syndrome. a TAE was discontinued because the contrast medium loading limit was attained prematurely. b TAE was terminated early because of intractable pain. c Nephrectomy after TAE. d Ex-vivo resection of the vascular lesion and kidney autotransplantation after TAE.
TAE outcomes The complete obliteration rate was 56.3% (9/16). Of the seven patients with partial obliterations (Fig 2), TAE was terminated before completion in four, because of intractable pain developing during embolisation (three cases) and concern about contrast medium overload (one case). The clinical success rate was 87.5% (14/16). Two patients exhibited clinical failure; one underwent nephrectomy and the other ex-vivo resection of the renal vascular lesion followed by autotransplantation. There were 14 (66.7%) instances of post-embolisation syndrome (PES: fever and flank pain, i.e., minor complications). Two major complications (9.5%) were recorded; one instance of pulmonary embolism and one of renal vein thrombosis. The pulmonary embolism was caused by coil displacement during TAE in a patient with an aneurysmal CRAVM. The patient did not develop any associated symptoms and no further intervention or management was required. Complete thrombosis of the main renal vein occurred in another patient who underwent TAE using absolute ethanol. Fortunately, the collateral venous drainage was rich and the thrombosis resolved spontaneously after conservative observation. The median pre- and post-TAE SCr levels were 0.9 mg/dl (IQR: 0.74e1.15 mg/dl) and 0.86 mg/dl (IQR: 0.66e1.1 mg/ dl), respectively. The median pre- and post-TAE eGFRs were 78.2 ml/min/1.73 m2 (IQR: 67.7e99.15) and 87 ml/min/1.73 m2 (IQR: 66.1e131.7), respectively. No significant change in
either the SCr level or eGFR was noted. No relapse, in terms of symptoms or signs, was evident.
Discussion Over a 12-year period, 16 patients underwent single- or multi-session TAE to treat CRAVMs. The complete obliteration rate was 56.3%, and the clinical success rate was 87.5% over a median follow-up period of 398.5 days. PubMed was searched for relevant English-language publications that appeared between 1998 and 2017; seven case series describing the use of TAE to treat CRAVM were identified (Table 2).6e12 Although it is difficult to compare studies because the definitions of clinical success varied, the results seem comparable to those reported previously.6e12 In the present study, three patients who suffered initial clinical failure underwent successful repeat TAE. After single- or multi-session TAE, all patients who underwent complete obliteration achieved clinical success; however, partial obliteration was often not associated with clinical failure. In some patients, haematuria ceased after only partial obliteration, which was presumably effective in reducing renal vascular pressure and occluding the bleeding foci.6,10 TAE for CRAVM may take several hours and causes severe pain, which may have contributed to incomplete obliteration. Three of the procedures were terminated early because of intractable pain. Therefore, adequate sedation and anaesthesia are essential in some patients to enable toleration of TAE.
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059
1057
Figure 2 A 26-year-old woman with a cirsoid CRAVM underwent TAE using N-butyl-2-cyanoacrylate (NBCA; patient 16). (a) Renal angiography reveals a tangle of abnormal vessels over the lower pole of the left kidney, with early venous drainage. (bec) Selective angiography reveals multiple tortuous feeders arising from the segmental and interlobar arteries, consistent with a cirsoid CRAVM. (d) Post-embolisation angiography reveals partial obliteration of the vascular lesion with a residual nidus, and feeders of uncertain origin that were difficult of access. Multifocal post-embolisation infarcts are evident in the interpolar region and the lower pole of the left kidney. Although only partial obliteration was achieved, the haematuria resolved without relapse.
Appropriate selection of an embolic agent is an important issue in successful endovascular treatment. Many embolic agents can be used for TAE of CRAVM. Coils were the most commonly used embolic agent in the present study, which are eminently controllable and can be positioned precisely. Nevertheless, coils occlude feeding arteries only at the locations where they are deployed; they do not directly occlude the nidus. Thus, liquid embolic agents, such as absolute ethanol, NBCA, and Onyx, have been recommended for endovascular treatment of CRAVM.1,7,9e11,13 Embolic liquids can penetrate and thus directly occlude the nidus; however, they are not as controllable as coils.
Based on the authors’ experience, complete obliteration is achievable with coils, especially in patients with aneurysmal CRAVMs or small focal cirsoid CRAVMs supplied by small numbers of segmental arteries; however, it can be rather challenging to successfully embolise cirsoid CRAVMs with multiple fine feeders arising at the hilar level, or from different segmental arteries. In such cases, embolic liquids may be better. A similar opinion has been voiced in the literature.11 Minor complications, such as PES, are common after TAE for CRAVM. Major complications, such as pulmonary embolism, renal vein thrombosis, and a large area of non-target
1058
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059
Table 2 Case series of transarterial embolisation (TAE) used to treat congenital renal arteriovenous malformations (CRAVMs). No.
Reference
No. of cases
Clinical successa
Follow-up period
Major complications
1
Takebayashi et al.10
30
100%
Mean 8 years (range: 4.1e15 years)
2 3
Zhang et al.11 Poh et al.9
6 6
100% 100%
4 5 6
Nassiri et al.8 Murata et al.7 Eom et al.6
12 12 24b
100% 100% 88%
Mean 22 months (range: 3e60 months) Mean 68 months (range: 12e150 months) Mean 55 months (range: 5e96 months) Mean 48 months (range: 7e92 months) Median 447 days (range: 4e2486 days)
1) Large non-target embolisation with reflux of Gelfoam rendering the kidney non-functional (one case) 2) Asymptomatic pulmonary embolism and renin-dependent hypertension after PVA embolisation (one case) No major complication N/A
7 8
Hwang et al.12 Current study
11 16
100% 88%
Mean 16.3 months (range: 6e42 months) Median 398.5 days (IQR: 52.5e1716.5 days)
No major complication No major complication Pulmonary embolism after NBCA embolisation with transient dyspnoea during the procedure (one case) No major complication 1) Asymptomatic pulmonary embolism after coil embolisation (one case) 2) Asymptomatic renal vein thrombosis (one case)
Gelfoam, gelatin sponge; N/A, not available; NBCA, N-butyl cyanoacrylate; PVA, polyvinyl alcohol; IQR, interquartile range. a Modified clinical success rates after single- or multi-session TAE. b Four of 24 cases in this study had acquired arteriovenous fistulae.
embolisation, are rare6,10,14,15; pulmonary embolism is the most common complication. One case of pulmonary embolism caused by coil displacement during treatment of an aneurysmal CRAVM was seen. This complication is most likely to occur during treatment of aneurysmal CRAVMs, because they are usually characterised by a single large arteriovenous communication. Pulmonary embolism caused by coil displacement is rare during treatment of cirsoid CRAVMs, which usually feature small nidal vessels.1,8 Besides coils, other embolic agents can potentially pass through the arteriovenous malformation into the pulmonary arteries. Pulmonary embolisation following NBCA embolisation was earlier associated with dyspnoea and desaturation developing during TAE.6,16e18 Furthermore, migration of ethanol or non-absorbable microparticles to the lung may be fatal.19e23 Although major complications are not common, TAE for CRAVM should be performed very carefully to avoid potential complications. Blood flow control via balloon occlusion can usefully prevent the migration of embolic materials.7,8,10,15 In addition, surgery is an alternative treatment for patients with large, high-flow CRAVMs, who are at high risk of pulmonary embolism during TAE.24,25 Although nephrectomy is sometimes unavoidable, nephron-sparing surgery, such as ex-vivo resection of the renal vascular lesion followed by autotransplantation, may be technically and clinically feasible, as in one of the present cases. The present work had certain limitations. First, this was a retrospective study, and as such had unavoidable, inherent limitations. Second, three patients had follow-up periods of <1 month (i.e., 4, 15, and 21 days), although the median follow-up period was approximately 1 year. Third, the small number of cases rendered statistical analysis futile, although to the authors’ knowledge, this is the third largest case series reported. For example, 11 of 11 patients (100%) treated using microcatheters experienced clinical success, in contrast to only three of five patients (60%) for whom microcatheters were not employed; however, this
difference did not attain statistical significance (p¼0.083, Fisher’s exact test), attributable to the low statistical power afforded by the small number of cases. In conclusion, TAE was a safe and effective treatment for CRAVM; the complication rate was acceptable and renal function was preserved. Although only partial obliteration was attained in some cases, this may be adequate to achieve clinical success.
References 1. Hatzidakis A, Rossi M, Mamoulakis C, et al. Management of renal arteriovenous malformations: a pictorial review. Insights Imaging 2014;5:523e30. 2. Cho KJ, Stanley JC. Non-neoplastic congenital and acquired renal arteriovenous malformations and fistulas. Radiology 1978;129:333e43. 3. Takaha M, Matsumoto A, Ochi K, et al. Intrarenal arteriovenous malformation. J Urol 1980;124:315e8. 4. Angle JF, Siddiqi NH, Wallace MJ, et al. Quality improvement guidelines for percutaneous transcatheter embolization: society of interventional radiology standards of practice committee. J Vasc Interv Radiol 2010;21:1479e86. 5. Levey AS, Bosch JP, Lewis JB, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461e70. 6. Eom HJ, Shin JH, Cho YJ, et al. Transarterial embolisation of renal arteriovenous malformation: safety and efficacy in 24 patients with followup. Clin Radiol 2015;70:1177e84. 7. Murata S, Onozawa S, Nakazawa K, et al. Endovascular embolization strategy for renal arteriovenous malformations. Acta Radiol 2014;55: 71e7. 8. Nassiri N, Dudiy Y, Carroccio A, et al. Transarterial treatment of congenital renal arteriovenous fistulas. J Vasc Surg 2013;58:1310e5. 9. Poh PG, Tan BS, Tham SC, et al. The use of n-butyl-2 cyanoacrylate as an embolic agent in the minimally invasive treatment of renal arteriovenous malformations. Ann Acad Med Singapore 2013;42:207e9. 10. Takebayashi S, Hosaka M, Kubota Y, et al. Transarterial embolization and ablation of renal arteriovenous malformations: efficacy and damages in 30 patients with long-term followup. J Urol 1998;159:696e701. 11. Zhang B, Jiang ZB, Huang MS, et al. The role of transarterial embolization in the management of haematuria secondary to congenital renal arteriovenous malformations. Urol Int 2013;91:285e90.
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059 12. Hwang JH, Do YS, Park KB, et al. Embolization of congenital renal arteriovenous malformations using ethanol and coil depending on angiographic types. J Vasc Interv Radiol 2017;28:64e70. 13. Cura M, Elmerhi F, Suri R, et al. Vascular malformations and arteriovenous fistulas of the kidney. Acta Radiol 2010;51:144e9. 14. Wiesinger B, Schober W, Tepe G, et al. Complication after embolization of a complex renal vascular malformation. Nephrol Dial Transplant 2005;20:1729e33. 15. Maruno M, Kiyosue H, Tanoue S, et al. Renal arteriovenous shunts: clinical features, imaging appearance, and transcatheter embolization based on angioarchitecture. RadioGraphics 2016;36:580e95. 16. Carapiet DA, Stevens JE. Pulmonary embolism following embolization of an arteriovenous malformation. Paediatr Anaesth 1996;6:491e4. 17. Harunarashid H, Lily S, Rozman Z, et al. Pulmonary embolism following histoacryl glue embolization for a large thigh arteriovenous malformation. Clin Ter 2012;163:393e5. 18. Kjellin IB, Boechat MI, Vinuela F, et al. Pulmonary emboli following therapeutic embolization of cerebral arteriovenous malformations in children. Pediatr Radiol 2000;30:279e83.
1059
19. Chapot R, Laurent A, Enjolras O, et al. Fatal cardiovascular collapse during ethanol sclerotherapy of a venous malformation. Interv Neuroradiol 2002;8:321e4. 20. Mitchell SE, Shah AM, Schwengel D. Pulmonary artery pressure changes during ethanol embolization procedures to treat vascular malformations: can cardiovascular collapse be predicted? J Vasc Interv Radiol 2006;17:253e62. 21. Yakes WF, Luethke JM, Parker SH, et al. Ethanol embolization of vascular malformations. RadioGraphics 1990;10:787e96. 22. Repa I, Moradian GP, Dehner LP, et al. Mortalities associated with use of a commercial suspension of polyvinyl alcohol. Radiology 1989;170: 395e9. 23. Wijeyaratne SM, Ubayasiri RA, Weerasinghe C. Fatal pulmonary embolism of polyvinyl alcohol particles following therapeutic embolisation of a peripheral arteriovenous malformation. BMJ Case Rep 2009;2009. 24. Obara H, Tanaka K, Matsubara K, et al. Large renal arteriovenous fistula with left-sided inferior vena cava. Ann Vasc Surg 2013;27. 1185.e9-12. 25. Tarmiz A, Jerbi S, Jaidane M, et al. An 18-cm-large renal arteriovenous fistula treated by nephrectomy. Ann Vasc Surg 2010;24. 551.e9-11.
Contents lists available at ScienceDirect
Clinical Radiology journal homepage: www.clinicalradiologyonline.net
Transarterial embolisation for congenital renal arteriovenous malformations improves clinical condition even with partial obliteration T.-W. Sheng, Y.-C. Wong, C.-H. Wu, L.-J. Wang* Department of Medical Imaging and Intervention, Linkou Chang Gung Memorial Hospital, Chang Gung University, No. 5, Fuxing St., Guishan Dist., Taoyuan City 33305, Taiwan, ROC
art icl e i nformat ion Article history: Received 31 December 2016 Received in revised form 23 June 2017 Accepted 10 July 2017
AIM: To evaluate the efficacy and safety of transarterial embolisation (TAE), used to treat congenital renal arteriovenous malformations (CRAVMs). MATERIALS AND METHODS: The medical records were searched retrospectively to identify patients who underwent TAE to treat CRAVM from January 2003 to August 2015. Patient demographics, clinical presentations, and imaging findings were reviewed. TAE outcomes, including complete or partial obliteration, clinical success, complications, renal function changes, and relapse of symptoms or signs after the final TAE, were assessed. RESULTS: Over the 12-year period, 16 patients (nine male, seven female; median age, 47 years) who underwent 21 TAE procedures to treat 16 CRAVMs were enrolled in the study. The most common clinical presentation was haematuria (81.3%). Thirteen patients (81.3%) had cirsoid and three (18.7%) had aneurysmal CRAVMs. Of the 16 CRAVMs, 11 (68.8%) were peripheral, four (25%) were central, and one (6.3%) was both peripheral and central. The complete obliteration rate was 56.3%. The clinical success rate was 87.5% over a median follow-up period of 398.5 days. Two (9.5%) major complications and 14 (66.7%) minor complications were encountered. No statistically significant change in renal function was evident after TAE. CONCLUSION: TAE was safe and effective when used to treat CRAVM; the complication profile was acceptable and renal function was preserved. TAE improved the clinical condition of CRAVM patients even when obliteration was only partial. Ó 2017 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
Introduction Renal arteriovenous malformations are abnormal communications between the renal arteries and veins anywhere upstream of the capillary level. Aetiologically, the condition
* Guarantor and correspondent: L.-J. Wang, Department of Medical Imaging and Intervention, Linkou Chang Gung Memorial Hospital, Chang Gung University, No. 5, Fuxing St., Guishan Dist., Taoyuan City 33305, Taiwan, Republic of China. Tel.: þ886-3-3281200x3786. E-mail address: [email protected] (L.-J. Wang).
is classified as either acquired or congenital. Acquired renal arteriovenous malformations account for approximately 75% of cases1 and are usually caused by trauma, malignancy, inflammation, surgery, or percutaneous kidney procedures. Congenital renal arteriovenous malformations (CRAVMs) are extremely rare (estimated prevalence, 0.04%)2 and are classified as cirsoid or aneurysmal. Cirsoid CRAVM is characterised by a tangle of dilated and tortuous arteriovenous communications, a so-called vascular nidus, supplied by multiple arteries. Aneurysmal CRAVM, also termed idiopathic or cavernous CRAVM, usually presents as an
http://dx.doi.org/10.1016/j.crad.2017.07.008 0009-9260/Ó 2017 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
1054
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059
arteriovenous fistula with a single arterial feeder. CRAVM may be asymptomatic or associated with gross haematuria, hypertension, or cardiac failure.3 The goal of treatment is to eradicate the vascular lesion completely, with maximal preservation of the renal parenchyma. Open surgery was the most common approach in the past; however, this is associated with a risk of injury to the renal parenchyma because of the peripelvic location of the vascular anomaly, which necessitates extensive branch artery dissection. Today, developments in endovascular techniques have rendered transarterial embolisation (TAE) the first-line treatment. As CRAVM is rare, only a few case series and case reports on TAE used to treat CRAVM have appeared. Therefore, the purpose of the present study was to review retrospectively the outcomes of patients undergoing TAE to treat CRAVM.
Materials and methods Patient population The medical records were searched retrospectively to identify patients who underwent TAE to treat CRAVM from January 2003 to August 2015 in Linkou Chang Gung Memorial Hospital. The institutional review board approved the review of patient medical data, including demographics, clinical presentations, and imaging findings. Informed consent was obtained from all patients prior to TAE. CRAVM diagnoses were based on clinical histories, computed tomography (CT), and angiography. No patient had a history of renal trauma, malignancy, inflammation, surgery, or any percutaneous procedure. Follow-up continued to the time of hospital discharge or the final documented outpatient visit; all follow-up durations were recorded.
TAE procedures Nine radiologists with 3e20 years of experience in interventional radiology performed the TAE procedures. Arterial access was obtained through the common femoral artery under local or general anaesthesia. Diagnostic renal angiography was performed to delineate the CRAVM vasculature (Fig 1). Subsequently, an angiocatheter was advanced to a point as close to the feeding artery as possible, prior to embolisation, to maximally preserve the renal parenchyma. A wide variety of embolic agents were used individually or in combination; these included gelatin sponges, coils, trisacryl gelatin microspheres (Embospheres), absolute ethanol, N-butyl-2-cyanoacrylate (NBCA), and ethylene vinyl alcohol copolymer (Onyx). The choice of embolic agents was based on the characteristics of the vasculature, the catheter position that was attainable, the availability of embolic agents, and the physician’s preference. Angiography was performed after embolisation to confirm successful occlusion. The number of TAE sessions per patient was recorded. Patient complaints made during TAE sessions were reviewed.
Outcome assessment Outcomes assessed after the final TAE session included complete or partial obliteration, clinical success, renal function changes, and relapse of symptoms or signs. Partial obliteration was defined as a <90% reduction in CRAVM volume. Clinical success was defined as the absence of any requirement for further intervention or surgery, as evidenced by the resolution of the initial presentations (e.g., gross haematuria) or an absence of residual CRAVM on postTAE images. Complications developing during each TAE session were classified using the definitions of the Society of Interventional Radiology.4 Major complications included those creating an unplanned need for more care, prolonged hospitalisation, permanent adverse sequelae, or death. Minor complications included those that had no sequelae and required only nominal observation or a short hospital stay. To assess changes in renal function, the serum creatinine (SCr) level was recorded, and the estimated glomerular filtration rate (eGFR) was calculated using the Modification of Diet in Renal Disease (MDRD) formula.5 The pre-TAE SCr level was the latest value obtained before the first TAE. The post-TAE SCr level was the first value obtained at least 1 month after the last TAE. SCr levels recorded within 1 month after the last TAE were excluded from consideration, to avoid any possible confounding effect of obstructive uropathy following haematuria and reversible contrastinduced nephropathy. Relapse during follow-up was recorded in terms of symptoms and signs.
Statistical analysis All statistical analyses were performed using Statistical Analysis System (SAS) software (ver. 9.3). Descriptive values for categorical and continuous variables are presented as numbers with percentages and as medians with interquartile range (IQR), respectively. The Wilcoxon signedrank sum test was used for two-tailed comparison of SCr and eGFR levels before and after TAE. The level of statistical significance was set at p<0.05.
Results Patient characteristics The demographics, TAE characteristics, and outcomes of the 16 patients are summarised in Table 1. Over the 12-year period, 16 patients (nine male, seven female; median age, 47 years; IQR: 35e62.5 years) underwent 21 TAE procedures to treat 16 CRAVMs. The median follow-up period after the final TAE was 398.5 days (IQR: 52.5e1,716.5 days). One patient had only one kidney because of a history of nephrectomy to treat renal cell carcinoma. The most common clinical presentation was haematuria (81.3%); 11 cases of gross haematuria and two of microscopic haematuria. Other symptoms included flank pain in five patients (31.3%), acute urinary retention in four (25%), and hypertension in three (18.8%). Three patients (18.8%) were asymptomatic and no patient exhibited cardiac failure. TAE was performed on
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059
1055
Figure 1 A 59-year-old man with an aneurysmal CRAVM underwent TAE with coil placement (patient no. 13). (a) A CT image shows abnormally dilated and tortuous vessels in the right kidney. (bec) Selective angiography reveals a vascular nidus composed of a dilated and tortuous arteriovenous fistula, with a single arterial feeder and an aneurysmally dilated draining vein. (d) After TAE with placement of coils, the vascular lesion was occluded completely and an inevitable focal renal infarction was evident.
asymptomatic patients on request. The patients were anxious and uncomfortable with conservative management; they were concerned about possible future gross haematuria, shock, and cardiac failure even after the low risks had been carefully explained to them.
TAE characteristics Angiographically, 13 patients (81.3%) had cirsoid CRAVMs and three (18.8%) had aneurysmal CRAVMs. Of the 16 kidney lesions, 11 (68.8%) were located peripherally, four (25%) centrally, and one (6.3%) extended broadly from the peripheral to the central region.
A total of 21 TAEs were performed in the 16 patients. Three underwent repeat TAE (two patients twice and one four times). Microcatheters were used in 13 of the 21 TAEs (61.9%). A wide variety of embolic agents were used either individually or in combination, including coils in 12 patients (57.1%), gelatin sponges in seven (33.3%), NBCA in three (14.3%), Embospheres in one (4.8%), absolute ethanol in one (4.8%), and Onyx in one (4.8%). A combination of coils and gelatin sponges was employed in three procedures (14.3%), while a combination of coils and NBCA was used in one case (4.8%). Liquid embolic agents (i.e., absolute ethanol, NBCA, and Onyx) were used in five procedures (23.8%).
1056
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059
Table 1 The demographics, features of transarterial embolisation (TAE), and outcomes of the 16 patients. Patient no./ age/sex
CRAVM type
TAE no.
CRAVM obliteration
Clinical success
Embolic agents used
Major complications
Minor complications
Follow-up (days)
1/91/F 2/69/M 3/46/M 4/48/F 5/34/M
Cirsoid Cirsoid Cirsoid Cirsoid Aneurysmal
1 1 1 1 1
C P Pa Pb P
Y Nc Y Y Nd
Coils Coils CoilsþGelfoam Embospheres Coils
N Y: PES Y: PES Y: PES N
39 113 15 3266 21
6/36/M 7/51/F 8/39/M
Cirsoid Cirsoid Cirsoid
2 1 4
P/Pb C P/C/C/C
Y Y Y
Y/Y: PES Y: PES Y/Y/Y/Y: PES
1/4 3325 230/109/820/ 556
9/46/F 10/54/F
Cirsoid Cirsoid
1 1
C Pb
Y Y
Gelfoam/Gelfoam CoilsþGelfoam Coils/ CoilsþGelfoam/ Gelfoam/ CoilsþNBCA Coils Absolute ethanol
N N N N Y: pulmonary embolisation N/N N N/N/N/N
N Y: PES
2159 2226
11/66/F 12/24/M 13/59/M 14/73/M 15/16/M 16/26/F
Cirsoid Cirsoid Aneurysmal Cirsoid Aneurysmal Cirsoid
1 2 1 1 1 1
C C/C C C C P
Y Y Y Y Y Y
Onyx Gelfoam/NBCA Coils Coils Coils NBCA
N Y: renal vein thrombosis N N/N N N N N
N N/Y: PES N Y: PES N Y: PES
1274 570 236 471 326 66
CRAVM, congenital renal arteriovenous malformation; C, complete obliteration; P, partial obliteration; Embospheres, trisacryl gelatin microspheres; Gelfoam, gelatin sponge; NBCA, N-butyl cyanoacrylate; Onyx, ethylene vinyl alcohol copolymer; Y, yes; N, no; PES, post-embolisation syndrome. a TAE was discontinued because the contrast medium loading limit was attained prematurely. b TAE was terminated early because of intractable pain. c Nephrectomy after TAE. d Ex-vivo resection of the vascular lesion and kidney autotransplantation after TAE.
TAE outcomes The complete obliteration rate was 56.3% (9/16). Of the seven patients with partial obliterations (Fig 2), TAE was terminated before completion in four, because of intractable pain developing during embolisation (three cases) and concern about contrast medium overload (one case). The clinical success rate was 87.5% (14/16). Two patients exhibited clinical failure; one underwent nephrectomy and the other ex-vivo resection of the renal vascular lesion followed by autotransplantation. There were 14 (66.7%) instances of post-embolisation syndrome (PES: fever and flank pain, i.e., minor complications). Two major complications (9.5%) were recorded; one instance of pulmonary embolism and one of renal vein thrombosis. The pulmonary embolism was caused by coil displacement during TAE in a patient with an aneurysmal CRAVM. The patient did not develop any associated symptoms and no further intervention or management was required. Complete thrombosis of the main renal vein occurred in another patient who underwent TAE using absolute ethanol. Fortunately, the collateral venous drainage was rich and the thrombosis resolved spontaneously after conservative observation. The median pre- and post-TAE SCr levels were 0.9 mg/dl (IQR: 0.74e1.15 mg/dl) and 0.86 mg/dl (IQR: 0.66e1.1 mg/ dl), respectively. The median pre- and post-TAE eGFRs were 78.2 ml/min/1.73 m2 (IQR: 67.7e99.15) and 87 ml/min/1.73 m2 (IQR: 66.1e131.7), respectively. No significant change in
either the SCr level or eGFR was noted. No relapse, in terms of symptoms or signs, was evident.
Discussion Over a 12-year period, 16 patients underwent single- or multi-session TAE to treat CRAVMs. The complete obliteration rate was 56.3%, and the clinical success rate was 87.5% over a median follow-up period of 398.5 days. PubMed was searched for relevant English-language publications that appeared between 1998 and 2017; seven case series describing the use of TAE to treat CRAVM were identified (Table 2).6e12 Although it is difficult to compare studies because the definitions of clinical success varied, the results seem comparable to those reported previously.6e12 In the present study, three patients who suffered initial clinical failure underwent successful repeat TAE. After single- or multi-session TAE, all patients who underwent complete obliteration achieved clinical success; however, partial obliteration was often not associated with clinical failure. In some patients, haematuria ceased after only partial obliteration, which was presumably effective in reducing renal vascular pressure and occluding the bleeding foci.6,10 TAE for CRAVM may take several hours and causes severe pain, which may have contributed to incomplete obliteration. Three of the procedures were terminated early because of intractable pain. Therefore, adequate sedation and anaesthesia are essential in some patients to enable toleration of TAE.
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059
1057
Figure 2 A 26-year-old woman with a cirsoid CRAVM underwent TAE using N-butyl-2-cyanoacrylate (NBCA; patient 16). (a) Renal angiography reveals a tangle of abnormal vessels over the lower pole of the left kidney, with early venous drainage. (bec) Selective angiography reveals multiple tortuous feeders arising from the segmental and interlobar arteries, consistent with a cirsoid CRAVM. (d) Post-embolisation angiography reveals partial obliteration of the vascular lesion with a residual nidus, and feeders of uncertain origin that were difficult of access. Multifocal post-embolisation infarcts are evident in the interpolar region and the lower pole of the left kidney. Although only partial obliteration was achieved, the haematuria resolved without relapse.
Appropriate selection of an embolic agent is an important issue in successful endovascular treatment. Many embolic agents can be used for TAE of CRAVM. Coils were the most commonly used embolic agent in the present study, which are eminently controllable and can be positioned precisely. Nevertheless, coils occlude feeding arteries only at the locations where they are deployed; they do not directly occlude the nidus. Thus, liquid embolic agents, such as absolute ethanol, NBCA, and Onyx, have been recommended for endovascular treatment of CRAVM.1,7,9e11,13 Embolic liquids can penetrate and thus directly occlude the nidus; however, they are not as controllable as coils.
Based on the authors’ experience, complete obliteration is achievable with coils, especially in patients with aneurysmal CRAVMs or small focal cirsoid CRAVMs supplied by small numbers of segmental arteries; however, it can be rather challenging to successfully embolise cirsoid CRAVMs with multiple fine feeders arising at the hilar level, or from different segmental arteries. In such cases, embolic liquids may be better. A similar opinion has been voiced in the literature.11 Minor complications, such as PES, are common after TAE for CRAVM. Major complications, such as pulmonary embolism, renal vein thrombosis, and a large area of non-target
1058
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059
Table 2 Case series of transarterial embolisation (TAE) used to treat congenital renal arteriovenous malformations (CRAVMs). No.
Reference
No. of cases
Clinical successa
Follow-up period
Major complications
1
Takebayashi et al.10
30
100%
Mean 8 years (range: 4.1e15 years)
2 3
Zhang et al.11 Poh et al.9
6 6
100% 100%
4 5 6
Nassiri et al.8 Murata et al.7 Eom et al.6
12 12 24b
100% 100% 88%
Mean 22 months (range: 3e60 months) Mean 68 months (range: 12e150 months) Mean 55 months (range: 5e96 months) Mean 48 months (range: 7e92 months) Median 447 days (range: 4e2486 days)
1) Large non-target embolisation with reflux of Gelfoam rendering the kidney non-functional (one case) 2) Asymptomatic pulmonary embolism and renin-dependent hypertension after PVA embolisation (one case) No major complication N/A
7 8
Hwang et al.12 Current study
11 16
100% 88%
Mean 16.3 months (range: 6e42 months) Median 398.5 days (IQR: 52.5e1716.5 days)
No major complication No major complication Pulmonary embolism after NBCA embolisation with transient dyspnoea during the procedure (one case) No major complication 1) Asymptomatic pulmonary embolism after coil embolisation (one case) 2) Asymptomatic renal vein thrombosis (one case)
Gelfoam, gelatin sponge; N/A, not available; NBCA, N-butyl cyanoacrylate; PVA, polyvinyl alcohol; IQR, interquartile range. a Modified clinical success rates after single- or multi-session TAE. b Four of 24 cases in this study had acquired arteriovenous fistulae.
embolisation, are rare6,10,14,15; pulmonary embolism is the most common complication. One case of pulmonary embolism caused by coil displacement during treatment of an aneurysmal CRAVM was seen. This complication is most likely to occur during treatment of aneurysmal CRAVMs, because they are usually characterised by a single large arteriovenous communication. Pulmonary embolism caused by coil displacement is rare during treatment of cirsoid CRAVMs, which usually feature small nidal vessels.1,8 Besides coils, other embolic agents can potentially pass through the arteriovenous malformation into the pulmonary arteries. Pulmonary embolisation following NBCA embolisation was earlier associated with dyspnoea and desaturation developing during TAE.6,16e18 Furthermore, migration of ethanol or non-absorbable microparticles to the lung may be fatal.19e23 Although major complications are not common, TAE for CRAVM should be performed very carefully to avoid potential complications. Blood flow control via balloon occlusion can usefully prevent the migration of embolic materials.7,8,10,15 In addition, surgery is an alternative treatment for patients with large, high-flow CRAVMs, who are at high risk of pulmonary embolism during TAE.24,25 Although nephrectomy is sometimes unavoidable, nephron-sparing surgery, such as ex-vivo resection of the renal vascular lesion followed by autotransplantation, may be technically and clinically feasible, as in one of the present cases. The present work had certain limitations. First, this was a retrospective study, and as such had unavoidable, inherent limitations. Second, three patients had follow-up periods of <1 month (i.e., 4, 15, and 21 days), although the median follow-up period was approximately 1 year. Third, the small number of cases rendered statistical analysis futile, although to the authors’ knowledge, this is the third largest case series reported. For example, 11 of 11 patients (100%) treated using microcatheters experienced clinical success, in contrast to only three of five patients (60%) for whom microcatheters were not employed; however, this
difference did not attain statistical significance (p¼0.083, Fisher’s exact test), attributable to the low statistical power afforded by the small number of cases. In conclusion, TAE was a safe and effective treatment for CRAVM; the complication rate was acceptable and renal function was preserved. Although only partial obliteration was attained in some cases, this may be adequate to achieve clinical success.
References 1. Hatzidakis A, Rossi M, Mamoulakis C, et al. Management of renal arteriovenous malformations: a pictorial review. Insights Imaging 2014;5:523e30. 2. Cho KJ, Stanley JC. Non-neoplastic congenital and acquired renal arteriovenous malformations and fistulas. Radiology 1978;129:333e43. 3. Takaha M, Matsumoto A, Ochi K, et al. Intrarenal arteriovenous malformation. J Urol 1980;124:315e8. 4. Angle JF, Siddiqi NH, Wallace MJ, et al. Quality improvement guidelines for percutaneous transcatheter embolization: society of interventional radiology standards of practice committee. J Vasc Interv Radiol 2010;21:1479e86. 5. Levey AS, Bosch JP, Lewis JB, et al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130:461e70. 6. Eom HJ, Shin JH, Cho YJ, et al. Transarterial embolisation of renal arteriovenous malformation: safety and efficacy in 24 patients with followup. Clin Radiol 2015;70:1177e84. 7. Murata S, Onozawa S, Nakazawa K, et al. Endovascular embolization strategy for renal arteriovenous malformations. Acta Radiol 2014;55: 71e7. 8. Nassiri N, Dudiy Y, Carroccio A, et al. Transarterial treatment of congenital renal arteriovenous fistulas. J Vasc Surg 2013;58:1310e5. 9. Poh PG, Tan BS, Tham SC, et al. The use of n-butyl-2 cyanoacrylate as an embolic agent in the minimally invasive treatment of renal arteriovenous malformations. Ann Acad Med Singapore 2013;42:207e9. 10. Takebayashi S, Hosaka M, Kubota Y, et al. Transarterial embolization and ablation of renal arteriovenous malformations: efficacy and damages in 30 patients with long-term followup. J Urol 1998;159:696e701. 11. Zhang B, Jiang ZB, Huang MS, et al. The role of transarterial embolization in the management of haematuria secondary to congenital renal arteriovenous malformations. Urol Int 2013;91:285e90.
T.-W. Sheng et al. / Clinical Radiology 72 (2017) 1053e1059 12. Hwang JH, Do YS, Park KB, et al. Embolization of congenital renal arteriovenous malformations using ethanol and coil depending on angiographic types. J Vasc Interv Radiol 2017;28:64e70. 13. Cura M, Elmerhi F, Suri R, et al. Vascular malformations and arteriovenous fistulas of the kidney. Acta Radiol 2010;51:144e9. 14. Wiesinger B, Schober W, Tepe G, et al. Complication after embolization of a complex renal vascular malformation. Nephrol Dial Transplant 2005;20:1729e33. 15. Maruno M, Kiyosue H, Tanoue S, et al. Renal arteriovenous shunts: clinical features, imaging appearance, and transcatheter embolization based on angioarchitecture. RadioGraphics 2016;36:580e95. 16. Carapiet DA, Stevens JE. Pulmonary embolism following embolization of an arteriovenous malformation. Paediatr Anaesth 1996;6:491e4. 17. Harunarashid H, Lily S, Rozman Z, et al. Pulmonary embolism following histoacryl glue embolization for a large thigh arteriovenous malformation. Clin Ter 2012;163:393e5. 18. Kjellin IB, Boechat MI, Vinuela F, et al. Pulmonary emboli following therapeutic embolization of cerebral arteriovenous malformations in children. Pediatr Radiol 2000;30:279e83.
1059
19. Chapot R, Laurent A, Enjolras O, et al. Fatal cardiovascular collapse during ethanol sclerotherapy of a venous malformation. Interv Neuroradiol 2002;8:321e4. 20. Mitchell SE, Shah AM, Schwengel D. Pulmonary artery pressure changes during ethanol embolization procedures to treat vascular malformations: can cardiovascular collapse be predicted? J Vasc Interv Radiol 2006;17:253e62. 21. Yakes WF, Luethke JM, Parker SH, et al. Ethanol embolization of vascular malformations. RadioGraphics 1990;10:787e96. 22. Repa I, Moradian GP, Dehner LP, et al. Mortalities associated with use of a commercial suspension of polyvinyl alcohol. Radiology 1989;170: 395e9. 23. Wijeyaratne SM, Ubayasiri RA, Weerasinghe C. Fatal pulmonary embolism of polyvinyl alcohol particles following therapeutic embolisation of a peripheral arteriovenous malformation. BMJ Case Rep 2009;2009. 24. Obara H, Tanaka K, Matsubara K, et al. Large renal arteriovenous fistula with left-sided inferior vena cava. Ann Vasc Surg 2013;27. 1185.e9-12. 25. Tarmiz A, Jerbi S, Jaidane M, et al. An 18-cm-large renal arteriovenous fistula treated by nephrectomy. Ann Vasc Surg 2010;24. 551.e9-11.