Interventional Management of a Rare Combination of Nutcracker and Wilkie Syndromes

Nutcracker and Wilkie syndromes are rare mesoaortic compression entities, and their association is even less common. Data on interventional treatment of these pathologies are still scarce, but results from limited case series are encouraging. We report the case of a previously healthy 45-year-old woman diagnosed with nutcracker and Wilkie syndromes who presented with macroscopic hematuria, intermittent pain in the left flank and hypogastric region, postprandial nausea, and unexplained significant weight loss. A successful endovascular approach with stent implantation in the left renal vein was performed, but the stent migrated toward the left kidney, and this acute complication was managed through an interventional strategy as well. At the three-month follow-up, the patient described a marked improvement in all symptoms, except for the macroscopic hematuria. As it was our strong belief that the approach was efficient, we further investigated the “hematuria”, which eventually led to the diagnosis of endometrial carcinoma. A hysterectomy and bilateral adnexectomy were planned, and chemoradiotherapy was initiated with the goal of preoperative tumor reduction. To our knowledge, this is the first reported case in which both Wilkie and nutcracker syndromes were effectively treated by stent implantation in the left renal vein, complicated with very early stent migration due to inadequate apposition to the less compliant venous lumen. The treatment of the duodenal compression was indirectly included in the stenting of the left renal vein, as reclaiming the venous lumen widened the aortomesenteric angle. The aim of this review is to discuss our center’s transcatheter experience with these rare disorders and explore the literature in order to establish the benefits and limitations of such an approach.


Introduction
A reduced angulation of less than 22 degrees between the abdominal aorta and the superior mesenteric artery (SMA), with an aortomesenteric distance of less than 8 mm, is a very rare vascular alteration [1] which leads to a reduction in the aortomesenteric space and consequent compression of its structures, mainly of the left renal vein (nutcracker syndrome) and/or the duodenum (Wilkie syndrome) [2]. While each of these syndromes represent a rare finding, their association is even less frequent. Because the congenital form is not as common, several acquired etiologies have been described, including rapid weight loss, compression by adjacent lymphadenopathy or malignancy, severe lordosis, pregnancy, or intestinal malrotation [3]. Nutcracker and Wilkie syndromes share a common etiology, namely weight loss. reported by Schepper [14] in 1972. There are only a few reported cases in the literature that describe patients with radiological evidence of the compression of both the duodenum and the LRV, with even fewer patients reported to have both the radiological and clinical evidence for both disorders. Our PubMed/MEDLINE review of the literature found eight case reports describing concomitant nutcracker and Wilkie syndromes [15][16][17][18][19][20][21]. We hereby present our center's experience with a 45-year-old female patient diagnosed with both Wilkie and nutcracker syndrome, presented with typical symptoms, treated with an endovascular approach, with a surprising outcome. The aim of this review is to discuss the advances in the diagnosis and treatment of these rare syndromes in order to better understand its outcomes and limitations.

Imaging Diagnosis
Normally the SMA emerges from the abdominal aorta at a 90-degree angle. The normal aortomesenteric angle is reported to be 28 to 65 degrees, and the normal aortomesenteric distance ranges from 10 to 34 mm [2]. The LRV is situated anterior to the aorta in the fork between the SMA and abdominal aorta. In anterior nutcracker syndrome, the SMA arises from the aorta at an acute angle, compressing the LRV and causing left renal venous hypertension and/or final part of duodenum ( Figure 1). In posterior nutcracker syndrome, the LRV has a course posterior to the aorta and is compressed between it and the vertebral bones. In combined nutcracker syndrome, the anterior branch of the duplicated LRV is constricted between the aorta and the SMA, while the posterior is squeezed between the aorta and the vertebral bones.
reported by Schepper [14] in 1972. There are only a few reported cases in the lit that describe patients with radiological evidence of the compression of both the num and the LRV, with even fewer patients reported to have both the radiologi clinical evidence for both disorders. Our PubMed/MEDLINE review of the lit found eight case reports describing concomitant nutcracker and Wilkie syndrom 22]. We hereby present our center's experience with a 45-year-old female patien nosed with both Wilkie and nutcracker syndrome, presented with typical sym treated with an endovascular approach, with a surprising outcome. The aim of this is to discuss the advances in the diagnosis and treatment of these rare syndromes i to better understand its outcomes and limitations.

Imaging Diagnosis
Normally the SMA emerges from the abdominal aorta at a 90-degree angle. T mal aortomesenteric angle is reported to be 28 to 65 degrees, and the normal aorto teric distance ranges from 10 to 34 mm [2]. The LRV is situated anterior to the aort fork between the SMA and abdominal aorta. In anterior nutcracker syndrome, th arises from the aorta at an acute angle, compressing the LRV and causing left renal hypertension and/or final part of duodenum ( Figure 1). In posterior nutcracker syn the LRV has a course posterior to the aorta and is compressed between it and the v bones. In combined nutcracker syndrome, the anterior branch of the duplicated constricted between the aorta and the SMA, while the posterior is squeezed betw aorta and the vertebral bones. The "gold standard" for diagnosis remains phlebography, intravascular p measurement, and intravascular ultrasound through which the venous pressure g between the LRV, the inferior vena cava, and the renal vein diameter can be acc The "gold standard" for diagnosis remains phlebography, intravascular pressure measurement, and intravascular ultrasound through which the venous pressure gradient between the LRV, the inferior vena cava, and the renal vein diameter can be accurately identified [22]. But all of these dedicated invasive investigations are being conducted on the basis of a high suspicion or a provisional diagnosis, which is always firstly based on a non-invasive imaging work-up. In a patient with an unremarkable workup for the common causes of hematuria and/or flank pain, a multimodality approach of a Doppler ultrasound followed by either CT or magnetic resonance (MR) venography will often suggest the diagnosis of an abnormal mesoaortic angle.
Ultrasonography can often reveal the compressed renal vein and allow for the measurement of the vein diameter and stenosis secondary to the compression. The long axis color Doppler view can exemplify the presence of a velocity gradient from the perihilar to mesoaortic region [22]. The presence of an increased flow through the collateral veins is further proof of renal venous hypertension. Although ultrasound can nicely display the stenosis and collaterals, studies have shown a poor correlation between the Doppler and the gold standard of the renocaval pullback invasive pressure measurements [23,24], which we consider to be the basis of the variation in the collateral decompression of the renal vein. For this reason, a direct measurement of the pressure between the left renal vein and inferior vena cava remains valuable for securing a diagnosis of nutcracker syndrome, with a pressure difference greater than 5 mmHg considered significant [22].
CT and MR angiography tests allow for the highlighting of the mesenteric artery origin from the abdominal aorta and the compression and stenosis of the LRV. Coronal and sagittal reconstructions also allow for the depiction of the left gonadal vein and collateral circulation with the lumbar veins. With sagittal reconstruction, it is possible to evaluate the incriminated angle, which if it is less than 35 degrees, is compatible with the diagnosis of the compression of the LRV and further clinical and physiological assessment should define if the physician is facing a syndrome or just a nutcracker phenomenon. Finally, an LRV diameter ratio (hilar to aortomesenteric ratio) of more than 4.9 has a positive predictive value of 100% [24].

Case Report
A 45-year-old female with no medical history who was recently diagnosed with nutcracker syndrome in the nephrology department was referred for further investigations and treatment strategy decision making. She complained of macroscopic hematuria, intermittent pain in the left flank and hypogastric region, and postprandial nausea and described a significant weight loss of 10 kg in the last 3 months. After the initial differential diagnosis of macroscopic hematuria, urologic and general nephrological causes were excluded (renal biopsy revealed no pathologic changes on light microscopy and immunofluorescence). The subsequent computed tomography (CT) angiography ( Figure 2) confirmed the presence of nutcracker syndrome with an aortomesenteric distance of 7.1 mm and reduced angulation of 16 degrees, with an important dilation of the left renal vein. In addition, severe distention of the stomach and intestines was noted.
Notably, a tumor growth was described in the right ovary, possibly a hypertrophic corpus luteum, with a gynecological recommendation for periodic follow-up.
Selective left renal vein angiography was further performed and revealed the obstruction at the junction of the LRV with the inferior cava. Since the patient refused all surgical options, the interventional strategy consisting of LRV stenting was planned for the same procedure, with the aim of not only restoring normal LRV flow but also correcting the aortomesenteric angle, thereby reducing the compression of the duodenum. Bilateral femoral vein access was obtained and two wires were crossed in the LRV (a diagnostic super stiff Amplatz wire from the right and a coronary BMW wire from the left, as landmark). A balloon expandable stent measuring 9/40 mm was implanted at the obstruction site, but the stent migrated towards the renal side of the vein when post-dilation was tented. At this point, the guide catheter was replaced with a long-armed sheath and the stent lumen was recrossed with a super stiff wire, managing to retrieve the stent at the ostium vessel with a partially inflated balloon. Post-dilation was attempted once more, but, due to the trend towards migration, the procedure was aborted. As a post-procedural vascular ultrasound showed partial migration of the stent toward the left kidney and the patient complained of persistent flank pain, a second procedure was planned the next day, using the jugular vein as the access site, with the aim of repositioning the stent and thus optimizing the result. Echo-guided access to the jugular vein was obtained and the stent lumen was crossed with a super stiff wire. A 10/30 mm balloon was advanced into the stent, with 5 mm protrusion distally to the stent. After partial inflation of the balloon, the retraction of the stent at the lesion site was possible and a complete inflation at 16 atmospheres was performed, with a good apposition to the vessel wall being obtained, but with important deformation of the distal struts of the stent. As a result, a second balloon expandable to 9/60 mm was implanted at the ostium, which partially overlapped with the previous stent, and post-dilations with a 10/30 mm balloon were performed along the entire length of the stents. Final post-dilatation at the ostium was performed with a 12/40 mm balloon, with a good final angiographic result and a good and stable apposition being obtained, as the compression of the vein was no longer visualized. Notably, a tumor growth was described in the right ovary, possibly a hypertrophic corpus luteum, with a gynecological recommendation for periodic follow-up.
Selective left renal vein angiography was further performed and revealed the obstruction at the junction of the LRV with the inferior cava. Since the patient refused all surgical options, the interventional strategy consisting of LRV stenting was planned for the same procedure, with the aim of not only restoring normal LRV flow but also correcting the aortomesenteric angle, thereby reducing the compression of the duodenum. Bilateral femoral vein access was obtained and two wires were crossed in the LRV (a diagnostic super stiff Amplatz wire from the right and a coronary BMW wire from the left, as landmark). A balloon expandable stent measuring 9/40 mm was implanted at the obstruction site, but the stent migrated towards the renal side of the vein when post-dilation was tented. At this point, the guide catheter was replaced with a long-armed sheath and the stent lumen was recrossed with a super stiff wire, managing to retrieve the stent at the ostium vessel with a partially inflated balloon. Post-dilation was attempted once more, but, due to the trend towards migration, the procedure was aborted. As a post-procedural vascular ultrasound showed partial migration of the stent toward the left kidney and the patient complained of persistent flank pain, a second procedure was planned the next day, using the jugular vein as the access site, with the aim of repositioning the stent and thus optimizing the result. Echo-guided access to the jugular vein was obtained and the stent lumen was crossed with a super stiff wire. A 10/30 mm balloon was advanced into the stent, with 5 mm protrusion distally to the stent. After partial inflation of the balloon, the retraction of the stent at the lesion site was possible and a complete inflation at 16 atmospheres was performed, with a good apposition to the vessel wall being obtained, but with important deformation of the distal struts of the stent. As a result, a second balloon expandable to 9/60 mm was implanted at the ostium, which partially overlapped with the previous stent, and post-dilations with a 10/30 mm balloon were performed along the entire length of the stents. Final post-dilatation at the ostium was performed with a 12/40 mm balloon, with a good final angiographic result and a good and stable apposition being obtained, as the compression of the vein was no longer visualized. Figures 3 and 4 illustrate the main steps of the endovascular procedures. After the procedure, no complications occurred and the patient reported the relief of their pain, but with persistent hematuria. Dual antiplatelet therapy was initiated (Aspirin and Clopidogrel) for 3 months, as well as Sulodexide therapy. The patient was discharged the next day and clinical and imagistic follow-up was scheduled. As per the Wilkie syndrome, the patient was encouraged to regain weight by hypercaloric intake and enriched fluids (if tolerated), and adopt an antalgic position that has been described to aid enteral feeding (such as the prone position, left lateral decubitus position, and the knee-chest maneuver). Moreover, the treatment of the duodenal compression was indirectly included in LRV stenting treatment as reclaiming the venous lumen widens the aortomesenteric angle (the renal vein being superior in relation to the duodenum, closer to the corner of the angle).
A follow-up was conducted after 3 months. At this point, the patient described an important improvement of symptoms, with complete relief of their pain in the left flank, no postprandial nausea, and a slight weight gain (1.5 kg), but with persistent macroscopic hematuria. However, the patient also described frequent episodes of metrorrhagia, which raised an important question: is the hematuria described by the patient real or is it more? Thus, further investigations into the ovarian tumor were recommended and, after magnetic resonance imaging and another gynecological examination were performed, the diagnosis of endometrial cancer was established and chemoradiotherapy was initiated with the aim of pre-operative tumor reduction. An abdominal ultrasound showed the permeable stent in the correct position in the LRV, without restenosis or compression, and regular flow in the LRV ( Figure 5). Another clinical and imaging follow-up was scheduled after the planned surgery (a hysterectomy and bilateral adnexectomy). At the last telephone follow-up, the patient was noted to be continuing her chemoradiotherapy, with the indication for surgical resection depending on the progressive oncologic status, and that the hematuria had ceased and the urine was subjectively clear.

Stent Migration
We described the case of a 45-year-old woman with symptomatic nutcracker and Wilkie syndromes diagnosed by CT angiography, without any other known comorbidities. Because the patient refused surgery, an interventional approach was planned, which consisted of a stent implantation in the LRV. Although nutcracker syndrome is a wellcharacterized pathological condition, there is no clear consensus on the optimal treatment approach. Endovascular treatment represents a valid alternative for surgery, as reported by several authors [26,27]. However, this strategy is associated with a non-negligible risk of stent migration, as shown by several case reports and studies [28,29]. Wu Z et al. found a stent migration rate of 6.7% in a cohort of 75 patients, with no significant difference in the preoperative anteroposterior diameter and peak velocity of the aortomesenteric portion, and renal hilum of the LRV on a duplex ultrasound [29]. The authors concluded that the stent selection and its accurate deployment are important factors in preventing stent migration. Another general recommendation to avoid this complication refers to oversizing the stent by 20% and placing it in the first branch of the renal vein [30]. These aspects must be carefully calculated in advance, as there have been cases of stent migration into the right atrium that required open-heart surgery [27].

Stent Migration
We described the case of a 45-year-old woman with symptomatic nutcracker and Wilkie syndromes diagnosed by CT angiography, without any other known comorbidities. Because the patient refused surgery, an interventional approach was planned, which consisted of a stent implantation in the LRV. Although nutcracker syndrome is a wellcharacterized pathological condition, there is no clear consensus on the optimal treatment approach. Endovascular treatment represents a valid alternative for surgery, as reported by several authors [25,26]. However, this strategy is associated with a non-negligible risk of stent migration, as shown by several case reports and studies [27,28]. Wu Z et al. found a stent migration rate of 6.7% in a cohort of 75 patients, with no significant difference in the preoperative anteroposterior diameter and peak velocity of the aortomesenteric portion, and renal hilum of the LRV on a duplex ultrasound [28]. The authors concluded that the stent selection and its accurate deployment are important factors in preventing stent migration. Another general recommendation to avoid this complication refers to oversizing the stent by 20% and placing it in the first branch of the renal vein [29]. These aspects must be carefully calculated in advance, as there have been cases of stent migration into the right atrium that required open-heart surgery [26].
In our case, we chose the largest available stent in our clinic, which was a 9/40 mm Zeus CC balloon expandable stent (Rontis AG, Zug, Switzerland) and inflated it at high pressure, which provided a 10-15% oversizing, but a stent migration toward the left kidney still occurred. The complication was managed by retrieving the stent using a partially inflated balloon and implanting a second longer stent which was partially overlapped with the previous stent, significantly increasing the stability. The short-term follow-up demonstrated the preserved patency and correct position of the stent; thus, it is fair to assume that not only a diameter oversizing but also a length oversizing are important factors to be considered in order to prevent stent migration.
Moreover, as all of the symptoms except for the "hematuria" were significantly improved, it is our strong belief that the intervention was efficient, and the bleeding might actually be metrorrhagia, determined by the endometrial cancer. Furthermore, the oncological disease was most likely the main trigger of the weight loss, which could have determined the reduction in the aortomesenteric space and consequently in the compression of its structures.
Regarding the post-procedural treatment, the data are still scarce and, taking into consideration the active bleeding as well as the high ischemic risk (a complex procedure followed by a complication, and the long stent length), we arbitrarily decided to treat this patient with 3 months of dual antiplatelet therapy. The role of antiplatelet and anticoagulant agents in the prevention of venous in-stent restenosis and thrombosis requires further investigation, with particular attention on a unique subset of patients (after thrombosis vs. nonthrombotic vein lesions) [30,31].
Finally, the disappearance of the metrorrhagia during chemoradiotherapy supports our theory of treatment efficacy in terms of complete symptom relief and renal vein decompression. The uterine tumor complicated this case by "double jeopardy", perhaps causing the tapering of the meso-aortic angle and further mimicking the hematuria.

Surgical or Interventional Management
Patients with mild symptoms can be treated conservatively, with emphasis on weight gain that increases retroperitoneal adipose tissue, resulting in a change in the position of the left kidney with a decrease in tension on the LRV; this approach has been shown to relieve symptoms of nutcracker in 30% of patients [22]. This is highly encouraged in young individuals (those <18 years) as body growth releases the LRV from the arterial fork.
As per open surgical treatment, various techniques have been employed in an attempt to either decrease venous hypertension or alleviate the predominant symptoms of hematuria or pelvic congestion. The severity of symptoms, patient demographic, and the level of understanding of the available local expertise techniques can guide the clinician as to when and how to intervene. The first reported case of treated nutcracker was in 1974 [32]. Since then, a number of surgical approaches have been described. These include direct reimplantation of the left renal vein [33], SMA transposition [34], nephropexy [35], autotransplantation of the left kidney [36], isolated nephrectomy [37], and external stenting or "shielding" of the LRV [38]. Most of these approaches were developed to lower venous hypertension. Some techniques, such as left gonadal vein ligation, gonadocaval bypass, splenorenal venous bypass, and/or embolization and sclerosis of pelvic varices, are more directed to treat pelvic congestion symptoms [39]. The ligation of the collateral vessels, coil embolization of the gonadal veins, and ablation of the varices have been established together with renocaval pressure gradient relieving procedures in cases where pelvic congestion resulted from nutcracker syndrome [40].
Endovascular stenting of the left renal vein has recently been documented, with multiple studies demonstrating symptoms relief [25][26][27]. In fact, endovascular treatment could also solve the problem of duodenal compression ("2 in 1"), a great advantage over surgical reimplantation. The first stent was implanted in 1994 by Neste et al. [40] and since then, transcatheter treatment has gained popularity and notoriety due to its high success rate and few complications. The most relevant complication (for incidence and potentially damage) is stent migration, while in-stent restenosis and venous occlusion resulting from fibromuscular hyperplasia or thrombosis rarely occur. Thrombosis rather occurs before stenting, due to stasis in the dilated vein [41]. This detail is not negligible as the thrombus can embolize and cause acute pulmonary embolism [42]. There is no consensus on the postinterventional antithrombotic regimen but a "defensive" management would be 2-3 months of initial anticoagulation (until the stent endothelialization occurs) and prescribing aspirin long-term or dual antiplatelet therapy for at least 2 months [22,24]. The only limitation with endovascular stenting would be the lack of long-term data, although there is a signal of good long-term follow-up after 2-3 years [43]. The most pertinent reported cohorts of stented nutcracker syndrome patients are summarized in Table 1; at first sight, variability can be observed in the study population and follow-up. In one of the largest available registries to present, Chen et al. [26] retrospectively evaluated 61 patients with nutcracker syndrome treated by an endovascular approach and at long-term follow-up (66 months), there were only 2 patients without a significant improvement in the symptoms of hematuria, proteinuria, and flank pain, while the rest of the patients experienced clinical improvement at various periods of time (most of them after 6 months). Based on these observations, the authors recommended this approach as the primary option for nutcracker syndrome. The stenting techniques were mostly "borrowed" from the percutaneous experience in superior vena cava syndromes or May-Thurner syndrome (chronic compression of the left iliac vein against lumbar vertebrae by the overlying right common iliac artery) [44]. Due to the risk of migration, an oversized auto-expandable stent should always be preferred; they have more radial force and prevent recoil compared to dedicated venous stents. There is only a single published study that has implanted a specifically designed venous stent (Zilver Vena; Cook Medical, Bloomington, Ind) with relatively good outcomes, but in a small cohort (20 patients) with a short follow-up (10-122 days) [45]. Notably, a recent study from the University of Pittsburgh, USA used IVUS in 61% of their cases and had no stent migration, which highlights the importance of accurate stent sizing [46]. Interestingly, a group from China managed to build customized stents for patients with nutcracker syndrome by 3D printing; for the first step, they printed the entire kidney model based on CT images exported in DICOM format, then, through surgical planning, they finally printed out the stents using titanic alloy powder [47]. The only downside to this innovative idea is that the stent needs to be implanted surgically, as it is already expanded [47]. With time and more intensive imaging screening, it is likely that the prevalence of these syndromes will increase, which will inevitably refine our experience with these percutaneous techniques. The principle is the same as in coronary interventions: multidisciplinary teams, intraprocedural imaging, and proof of ischemia are encouraged [48]. In a patient who has a thin or asthenic body habitus, nutritional support can defer or replace surgical treatment. The final decision depends on the severity of symptoms. Certainly, the cause of weight loss must also be sought, because the nutritional vicious circle could have another origin.

Conclusions
Nutcracker and Wilkie syndromes are very rare findings, and their association is even less common. Interventional treatment of these pathologies is described in limited case series, and even if data are scarce, the results are encouraging and the latest techniques solve the problem of stent migration. To our knowledge, this is the first reported case in which both Wilkie and nutcracker syndromes were effectively treated by stent implantation in the left renal vein, complicated with acute stent migration that could also efficiently be treated by an endovascular approach. In contrast to surgical treatment, renal vein stenting can also relieve duodenal compression by opening the meso-aortic angle (because of the vein being located higher up, within the angle). On the basis of our case and several other case series reports, we believe that a tailored procedure consisting of a careful selection of a self-expandable stent oversized in length and diameter and close monitoring is a safe and effective strategy.