Abstract
42/60 patients (70%) were electively treated. 27/42 patients (64%) had open and 15/42 (36%) endovascular aneurysm repair. There were no liver or bowel infarctions after elective treatment of hepatic or mesenteric artery aneurysms (n = 13) in CCDS/CEUS and in CTA. Treatment of patients with splenic or renal artery aneurysms led to partial or complete organ loss in 42% (8/19) after operative and in 50% (5/10) after endovascular treatment (p < 0.05).
Keywords
Introduction
Visceral artery aneurysms (VAAs) are rare vascular pathologies. They carry the risk of fatal haemorrhage because they are often asymptomatic until rupture and, consequently, underdiagnosed. VAAs include splanchnic and renal artery aneurysms.
The incidence of splanchnic artery aneurysms ranges from 0.1 to 2% [1, 16]. In autopsy studies, splanchnic artery aneurysms were found up to 10%, and even more frequent than abdominal aortic aneurysms. Accepted thresholds for treatment are aneurysms >2 cm in diameter, pregnant women, progressive enlargement, and all symptomatic aneurysms [12, 16]. Renal artery aneurysms occur in approximately 0.1% of the general population [16]. Accepted indications for renal artery aneurysm repair are aneurysms >1 cm in diameter in combination with risk factors as hypertension, ipsilateral or contralateral stenosis and childbearing age in women. Without risk factors, aneurysm size eligible for reconstruction is >2 cm [13].
Over many years, open surgical repair was the gold standard. Nowadays, endovascular techniques have established as therapeutic alternative [5, 18]. The endovascular approach allows the successful treatment of VAAs with a low periprocedural morbidity and mortality [22] and is associated with a decreased length of hospital stay [18]. Different treatment concepts for VAAs depend on patients’ presentation and location of the aneurysm, as well as on comorbidities and life expectance [5, 18]. Emergency cases on the one hand and elective cases on the other hand have a different approach. Immediate bleeding control has the upmost priority in emergency. Highest priority in elective cases has maintenance of the organ perfusion and effective aneurysm exclusion. So far, there are only few studies, which directly compare the technical success and the organ microvascularization after open surgical repair to the endovascular techniques [17–20].
Aim of this study was to focus on the organ microvascularization determined by CCDS/CEUS in correlation to CTA/MRI after operative or endovascular treatment of visceral artery aneurysms.
Material and methods
Patients and study protocol
Patients’ characteristics and presentations in emergency but especially in elective cases were consecutively reviewed as well as technical success and post-procedural organ microvascularization. A regular follow-up was not intended. The informed consent of all patients was obtained prior to the study. The study data were collected within the framework of an external quality control as a registry (Z-2016-0668-3) complying with the principles of the Helsinki/Edinburgh Declaration of 2002. Exclusion criteria were patient age <18 years or pregnancy. Contrast medium hypersensitivity and deterioration of renal function with creatinine clearance <30 ml/min were contraindications for application of iodine contrast medium.
During the period from April 1995 to January 2016, 168 patients (78 males, 90 females) were diagnosed with splanchnic or renal artery aneurysms at our hospital site. 106 patients had splenic (63%), 20 renal (12%), 19 hepatic (11%), 14 gastro-pancreatico-duodenal (8%), and 9 superior/inferior mesenteric artery aneurysms (5%). Most of the VAAs were accidentally discovered by CTA (156/168, 93%), some by MRI (7/168, 4%) or ultrasound (5/168, 3%). In selected patients with indication for operative or endovascular treatment, pre-operative diagnostics were complemented by digital subtraction angiography (DSA).
60/168 patients (36%) fulfilled treatment criteria and had either open (29/60, 48%) or endovascular (31/60, 52%) aneurysm repair. Table 1 shows the individual characteristics of all patients. Therapeutic success and organ microvascularization of the 60 treated patients were evaluated during hospital stay by CCDS/CEUS (n = 60), CT scan (n = 26), and MRI (n = 11).
Imaging techniques
Ultrasound imaging
All examinations were performed by three experienced investigators. After operative or interventional treatment, B-mode ultrasound, CCDS and power Doppler were performed first before contrast imaging/CEUS [7] (since 2006) to visualize the morphology and microvascularization in the visceral aneurysms and related organs. Multi-frequency convex probes (1–4 MHz Logiq 9/GE, 1–4 MHz, Siemens Antares and S2000) were used. Digital data archiving allowed storage of cine-sequences. After informed oral and written consent, an intravenous bolus of 1.2–2.4 ml Sulphurhexafluorid (SonoVue®, Bracco Imaging Deutschland GmbH, Konstanz, Germany) followed by 10 ml NaCl was injected into the cubital vein for contrast enhancement. A real-time image documentation of at least 3 minutes from injection followed. Contrast harmonic imaging (CHI) subtraction technique was carried out simultaneously to a real-time ultrasound B-mode picture (true agent detection mode) to detect endoleaks and evaluate organ microvascularization with regard to partial and complete organ loss.
Computed tomography imaging
CTA was performed on a multislice-CT (Sensation 16 and 256, Siemens, Erlangen, Germany) using a collimation of 16×0.75 mm. Contrast medium (Accupaque 350®, GE Healthcare Buchler GmbH & Co. KG, Munich, Germany) was injected with a flow rate of 3-4 ml/s and a total volume of 100 ml, followed by injection of 50 ml of NaCl. Raw data, maximal intensity projections (MIP) and multiplanar reformations (MPR) were post-processed at the integrated working station. Reconstruction interval was indexed at 2 mm for MIP, 1 mm for MPR and 1 mm for axial slices. Endoleakage and organ microvascularization were classified according to a contrast enhancement with different Houndsfield units (HE) in the arterial phase (after 30–40 sec) and the late phase (after 90–120 sec). Additional coronary and parasagittal reconstructions were performed.
Magnetic resonance tomography imaging
All magnetic resonance angiography studies were performed on a Siemens 1.5 Tesla scanner (Magnetom Symphony, Siemens, Erlangen, Germany) using a flexible phased array coil. Bolus tracking technique was used for image acquisition while contrast medium was injected. Acquisition of CE-MRA was triggered after contrast was seen in the aortic arch. A power injector (MEDRAD® Spectris Solaris MR Injector) was exerted to administer 20 ml of Gadovist® (Gadolinium; Bayer Vital GmbH, Leverkusen, Germany). The following parameters were used: 3D Flash-sequence (TR 3.65 ms, TE 1.28 ms, flip angle 30, field of view 300 mm, slice thickness 0.8 mm, coronal panel). Voxel size was 1.1 × 0.7 × 0.8 mm. Maximum intensity projection (MIP) reconstructions were obtained at 15°steps.
Statistical analysis
Statistical calculations were made using SPSS (version 16.0 SPSS for Windows, Standard Version, LEAD technologies) and a Java Applet of the Department of medical computer science of University of Muenster, Germany for evaluation of sensitivity and specificity. Values of p < 0.05 were considered as statistically significant.
Results
For evaluation or organ microvascularization, patients were divided into emergency (bleeding) cases (18/60, 30%, Table 2) and elective cases (42/60, 70%, Table 3).
Emergency (bleeding) cases (n = 18)
18/60 patients (30%) presented with acute bleeding (two splenic, six hepatic, eight gastro-pancreatico-duodenal and two superior mesenteric artery aneurysms, Table 2). There was no patient with a ruptured renal artery aneurysm. Seventeen patients had pseudoaneurysms. A pancreatitis was found in 14/18 cases (78%). Other causes were one ruptured atherosclerotic, gastro-pancreatico-duodenal arcade aneurysm, two patients with aneurysms because of polytrauma after road accident, and one patient with polyarteritis nodosa.
16/18 patients (89%) were treated by endovascular means (Table 2). Overall technical success rate of the endovascular treatment concerning immediate bleeding control by transcatheter coil and/or Onyx® embolization (n = 13), stent-supported coiling (n = 2) or stentgraft exclusion (n = 1) was 94% (15/16). One patient with a splenic pseudoaneurysm due to necrotisising pancreatitis had to be converted to open surgery because of ongoing hemodynamic instability and failing endovascular bleeding control.
After emergency treatment, two patients showed a segmental liver malperfusion in CEUS and CTA, which could be treated conservatively. In one case of a gastro-pancreatico-duodenal aneurysm, secondary segmental small bowel resection had to be performed one day after primary surgery because of infarction (Table 2). Only two patients (11%) with a bleeding gastro-pancreatico-duodenal arcade aneurysm were treated by open arterial ligature (Table 2). No patient died in the post-procedural period.
Elective cases (n = 42)
42/60 patients (70%) were treated electively (fifteen splenic, fourteen renal, six hepatic, two gastro-pancreatico-duodenal and five superior mesenteric artery aneurysms, Table 3). Aneurysm formation was due to atherosclerosis in the majority of the cases (n = 35), fibromuscular dysplasia (n = 1), mycotic embolization (n = 1), pancreatitis (n = 4), and liver transplantation (n = 1).
27/42 patients (64%) were treated by open aneurysm repair (Table 3). Standard procedure was aneurysm resection and arterial reconstruction or patch angioplasty (21/27, 78%). One inferior mesenteric artery aneurysm was treated by arterial ligature because of sufficient collateralization. In four out of seven patients (57%) with splenic artery aneurysm, splenectomy was necessary for successful aneurysm exclusion. In one patient unilateral nephrectomy has to be performed due to distal location of the renal artery aneurysm. Segmental renal infarction without deterioration of renal function was noticed in three patients (Table 3). In summary, five complete and three partial organ loss (8/19, 42%) had to be observed after elective splenic or renal operative aneurysm repair. There were no liver or bowel infarction. Figure 1 shows an example of a successful open renal artery aneurysm repair with regular postoperative organ microvascularization.
15/42 patients (36%, Table 3) were treated by transcatheter coil and/or Onyx® embolization (10/15, 67%), stent-supported coiling (3/15, 20%) or stentgraft (2/15, 13%). Exclusion of the aneurysm was successful in all cases. Out of the 15 endovascular treated patients, segmental splenic malperfusion occurred in one patient. In two patients, secondary splenectomy had to be performed because of symptomatic, total splenic malperfusion, post-interventionally (3/8, 38%, Table 3). All two patients treated interventionally for renal artery aneurysm suffered from partial renal infarction (2/2, 100%, Table 3, Fig. 2). Again, no liver or bowel infarction was seen after elective, endovascular treatment (Fig. 3). One occlusion of a hepatic artery stentgraft remained asymptomatic. All together, partial or complete organ loss occurred in 5/10 patients (50%) electively treated for splenic or renal artery aneurysms (p < 0.05).
Discussion
Best treatment concept for patients presenting with ruptured, bleeding visceral artery aneurysms is the endovascular first policy. With the upmost priority on immediate bleeding control, endovascular therapy has several advantages. First, angiography does not only confirm pathology and localisation of the aneurysms, but also allows rapid bleeding control and definitive aneurysm exclusion [17]. Technical success rate is high [4, 22] and conversion to surgery has to be done rarely (in our study, only 1/16 interventionally treated patients with a ruptured VAA). Second, ruptured aneurysms are, in the majority of cases, pseudoaneurysms due to infection, chronic inflammation or trauma [3, 22]. In our series, just one patient presented with a ruptured, atherosclerotic, gastro-pancreatico-duodenal arcade aneurysm. And third, ruptured pseudoaneurysms occur often in anatomically difficult locations and/or in abdomens linked by adhesions [17, 22]. Consequently, in our series, ruptured aneurysms were more often treated by endovascular techniques (16/18, 89%) than open surgery (2/18, 11%). Only three out of the 18 patients (17%) treated in case of emergency showed partial or complete organ loss. Reason might be that ruptured pseudoaneurysms often occur in vascular regions, namely hepatic or intestinal, with sufficient collateralization.
Besides effective aneurysm exclusion, maintenance of the organ perfusion has the upmost priority in elective cases. So far, endovascular management has been shown to be a good alternative to open surgery in many cases [4, 22], which is associated with a decreased length of hospital stay [18]. Technical success rates for endovascular management range between 73 to 98% [4, 22]. For hepatic, especially intra-hepatic, and the much less common gastro-pancreatico-duodenal arcade and inferior mesenteric artery aneurysms, endovascular therapy has become the method of choice because open repair might be difficult and arterial reconstruction is not necessarily required due to extended possibility of collateralization in highly connected vascular regions [8, 18]. Concerning superior mesenteric or celiac artery aneurysms, treatment concepts depend on collateral circulation, which has to be determined, pre-operatively. If collateralization is sufficient, both endovascular therapy and arterial ligature are valuable therapeutic options. Otherwise, arterial reconstruction is required [17]. In our elective treatment group, none of the patients either treated operatively or interventionally showed hepatic or bowel infarctions in the post-procedural period.
For the most common visceral artery aneurysms, splenic and renal artery aneurysms, treatment concepts are more complex according to end organ perfusion without collateralization. In these cases, detailed evaluation of the characteristics of the aneurysm as well as of the anatomy of the affected artery is important to avoid splenic or renal infarctions or the post-embolization syndrome [3, 17]. A careful, post-procedural examination is crucial to detect malperfusion of the affected organs. Until now, CTA is the most accepted and widely performed imaging in vascular pathologies like aneurysms although the cumulative radiation dose and potential nephrotoxicity limit the use, especially in older patients and after invasive angiography with contrast agent. In contrast, duplex ultrasound, especially CEUS, is a developing technique strongly recommended and evaluated by comparative studies with CTA after endovascular aortic aneurysm repair (EVAR) [15]. At present, restricted technical and personnel resources hinder rapid and wide application despite the evident advantages [2, 24]. MRI can not always be considered as an alternative due to nephrotoxic contrast agent and non appropriate steel stentgrafts after EVAR. In these cases, CEUS and DSA with immediate treatment option might be a solution. However, exclusion of the visceral aneurysm as well as the organ microvascularization can be evaluated immediately after the endovascular or operative procedure by the use of CEUS.
Splenic infarction can occur in 25% to 40% after aneurysm embolization [8, 22]. Improvement has been made using endovascular stentgraft placement. Unlike coil embolization, endografts exclude the aneurysm and preserve organ microvascularization. [8–10, 23]. But if anatomy of the vessels and configuration of the aneurysms are not suitable for endovascular treatment, conventional surgery should be performed, also including splenectomy or nephrectomy in highly selected cases [3, 22]. In our study, five out of ten patients with splenic and renal artery aneurysms treated interventionally showed partial or complete organ infarction (50%) in contrast to 8 out of 19 (42%) surgically treated (p < 0.05).
Limitations of the study are the retrospective, single centre data analysis and the short-term post-procedural follow-up as far as three months. CEUS was started since 2006. Prior to this, CCDS or power Doppler was used to improve visualization of organ microvascularization. However, its strength is the number of 60 patients investigated by CCDS/CEUS and CTA/MRA for organ microvascularization after operative compared to endovascular treatment of VAAs. The ultrasound examinations were performed by only three experienced examiners over the 20 year-period.
Conclusion
First therapeutic option in case of emergency is the endovascular approach with the upmost priority on immediate bleeding control. In elective cases, the endovascular therapy is the method of choice for hepatic, gastro-pancreatico-duodenal arcade or inferior mesenteric artery aneurysms according to the connected vascular regions with the possibility of reperfusion. If collateralization is sufficient, both endovascular therapy and arterial ligature are valuable therapeutic options for elective superior mesenteric or celiac artery aneurysms. Otherwise, arterial reconstruction is required. Patients with elective splenic or renal artery aneurysm repair need to be evaluated very carefully due to end-organ perfusion. Endovascular aneurysm exclusion should only be done, if anatomy of the vessels and configuration of the aneurysms are suitable. Apart from that or in case of a hilum aneurysm, conventional surgery is the method of choice to avoid end-organ infarction and the post-embolization syndrome. Comorbidities, cancer, life expectance, and preference of the patient have also to be taken into consideration for definitive decision.
Conflict of interest
PM Kasprzak, EM Jung, R. Müller-Wille, W. Wohlgemuth, and R. Kopp declare no conflicts of interest.
W Schierling received travel costs from Bracco Imaging Deutschland GmbH.
K Pfister received travel costs and speaker’s fee from Bracco Imaging Deutschland GmbH.