Stent Dislodgement During Coronary Angioplasty Complicated by Left Main Thrombosis: Application of VA-ECMO in The Catheterization Laboratory

Stent dislodgement is a serious complication during Percutaneous Coronary Interventions (PCI) and potentially responsible for cardiac arrest, a condition requiring a rapid intervention and restoring of coronary and systemic blood flow. Recently, Veno-Arterial ExtraCorporeal Membrane Oxygenation (VA-ECMO) has become readily applicable even in the setting of the catheterization laboratory. In the present paper we report a case of refractory cardiac arrest due to left main thrombosis complicating stent dislodgement, successfully treated with support of VA-ECMO. The patient gave the consent for publication of this anonymous case report.


pre-dilation with a 2.5 x 15 mm semi-compliant balloon (Trek;
Abbott Vascular), a 3.5 x 18mm Drug Eluting Stent (DES) (Xience Sierra; Abbott Vascular) was deployed into the LAD up to 15 atm, then post-dilated with a 3.5 x 15 mm non-compliant balloon (NC Trek). Subsequently, the same 0.014-inch BMW wire was removed from the LAD to be positioned in the distal tract of the OM1 and the proximal LCx was pre-dilated with the previous 2.5x15mm semicompliant balloon. Following, a 3.0x15mm DES (Xience Sierra) ( Figure 1) was successfully deployed in the proximal OM1, whereas a 3.5 x 15 mm DES (Xience Sierra) was implanted into the proximal LCx at 16 atm for 32 seconds. Post-deployment angiogram showed optimal stent expansion ( Figure 2). After complete left coronary revascularization, it was decided to remove the wire; however, during this maneuver, an unexpected resistance was encountered.     Considering that the wire tip could not be freed, and it proved impossible to advance a second guidewire, it was decided to forcefully pull back in order to retrieve the wire and the deformed stent into the guide catheter. Unfortunately, this attempt resulted in a wire fracture and proximal migration of the stent up to the distal LM.
Patient's hemodynamics was still stable; thus, it was decided to try to capture the dislodged stent by using a 5-mm Amplatz Gooseneck Snare (Medtronic, Minneapolis, MN, U.S.A.). Several attempts were done but without success. After a short while, the patient began to experience ischemic symptoms and hemodynamic instability.     After the procedure, the patient was transferred to the Intensive Care Unit (ICU). After 24 hours from the procedure, ECMO flows were firstly decreased to approximately 1L/min and then removed.
Inotropic agents were also gradually weaned off. On the 4th day after the procedure, extubation was performed. On the 5th day after the procedure, vital signs were stable without vasopressors support; therefore the patient was transferred to the general ward.
The patient had a full cardiopulmonary and neurological recovery and was discharged 10 days after the procedure on dual antiplatelet therapy, in good conditions, symptom free. The patient remained asymptomatic and free of angina at 3-month follow-up.
important issue in interventional cardiology, because potentially associated with intracoronary thrombosis and myocardial ischemia such as in our clinical case [2]. Several procedural mechanisms may contribute to the development of this complication, such as guide catheter compression of stents deployed in an ostial location, compression by post-dilation balloons, inadequate coaxial alignment of the guide catheter and secondary devices use [2,3].
Other anatomical and procedural risky features include extreme coronary angulation, severe tortuosity and/or calcification, diffuse disease, inadequate lesion pre-dilation or debulking, direct stenting, number and length of deployed stents, or previous bypass surgery [2,4]. In our case, a proper pre-dilation was done; however, the unfavorable angulation between the two implanted stents on LCx and OM1 may have contributed to the wire trapping and the stent dislodgement was favored by the wire forced pull-back performed by the operator with the addition of an inadequate coaxiality of the JL guiding catheter towards LCx due to the acute LM-LCx angle.
Finally, new-generation DES are manufactured with thin struts, and undoubtedly are easy to deliver and highly con¬formable.
However, this innovative design allows maintaining radial strength, but longitudinal strength may be reduced. Recently, cases of longitudinal compression of stents deployed in ostial position have been published [3].
A deformed/dislodged stent should be recovered when possible. Various retrieval techniques have been described, including the "balloon inflation technique", use of gooseneck snares, grasping forceps, and stent entrapment with the "twowires" technique or with basket devices [1][2][3][4][5][6][7]. However, all of them are time-consuming and if stent retrieval is not possible, particularly in hemodynamically unstable patients, the easier and faster treatment is to use a new stent to crush the dislodged stent against the vessel wall [4]. Although this technique is not widely accepted for LM and proximal LAD dislodgements because of an increased risk for stent thrombosis and restenosis, this approach was recently reported in a published case report, where a dislodged stent was safely crushed to the distal LM/ostial LAD vascular wall in a hemodynamically unstable patient [8]. In case of failure of all percutaneous techniques and according to the dislodgement coronary site (LM or proximal LAD), surgical intervention may represent the last option [5][6][7][8].
Certainly, the most important issue during the acute management of this complication is represented by patient's hemodynamic conditions as well as coronary flow distal to the lost stent. If the patient is unstable with impaired coronary flow, it is crucial to stabilize his/her hemodynamic conditions first and promptly reestablish the coronary flow as quickly as possible [8].
In the present case, initially the patient was asymptomatic, with a preserved distal flow, thus several attempts were firstly carried out to retrieve the dislodged stent. When the condition deteriorated, the priority was to stabilize hemodynamics with VA-ECMO support and then to restore coronary flow with the fastest and most effective way. In fact, survival is low for both in-hospital (15-20%) and outof-hospital (10%) cardiac arrest managed with conventional CPR [9]. Thus, several factors contributed to the complete recovery of our patient. Even though he had several comorbidities, he was relatively young, CPR maneuvers were promptly initiated, VA-ECMO support was carried out quickly, and the condition causing cardiac arrest was corrected as soon as possible.
Over the past decade, use of temporary Percutaneous Mechanical Circulatory Support (pMCS) devices for cardiac arrest management is increased [10,11]. An ideal pMCS should be readily available, inserted rapidly, associated with low complication rates and cost-effective [10]. The two main pMCS suitable for cardiac arrest management in the catheterization laboratory are VA-ECMO and Impella [10]. The Impella, compared with VA-ECMO, has the advantage of using smaller vascular access and unloads the left ventricle; on the other side, it does not oxygenate the blood, it provides a lower level of hemodynamic support, and does not assist the right ventricle, unless even a right sided Impella is placed [10]. ECMO is usually established by cannulation of a femoral vein and artery circuit and consists of a centrifugal blood pump, an oxygenator and a heparin-coated circuit [12]. VA-ECMO provides both blood oxygenation and bi-ventricular hemodynamic support in patients with cardiopulmonary failure, thereby providing adequate temporary perfusion (non-pulsatile flow) and oxygenation to the organs with subsequent rapid restoration of hemodynamic conditions [12,13].
Notably, the beneficial effects of emergent ECMO implantation in the catheterization laboratory, are more prominent in patients without previous heart failure who suffer refractory cardiac arrest due to a potentially treatable and therefore reversible cause of myocardial ischemia [13]. Of note, in the present case, early ECMO insertion was essential to perform a subsequent successful revascularization.
Obviously, since conventional CPR was initiated immediately after onset of ventricular fibrillation, it also played a critical bridging role until ECMO insertion.
Despite the clear benefit of mechanical support, there is a paucity of data regarding emergency use of VA-ECMO in the catheterization laboratory. Few case reports showed that ECMO is becoming increasing available also in this setting and may represent an effective temporizing measure/bridge to the final therapeutic intervention, such as PCI or surgery, that could solve the reversible cause of cardiac arrest (1) [13][14][15]. In 14 cases of refractory cardiac arrest during PCI or TAVI (Transcatheter Aortic Valve Implantation), ECPR using percutaneous VA-ECMO allowed to restore circulation in all patients, with 50% of survival at hospital discharge [11]. Moreover, surgical ECMO implantation by vessel exposure has been gradually replaced by percutaneous techniques. Indeed, it has been demonstrated that fluoroscopy-guided emergency percutaneous implantation of femoral VA-ECMO can be effectively and safely done by an interventional cardiologist in the catheterization laboratory [14]. Unfortunately, ECMO is still associated with a broad  [11,12]. Also, of note, unlike the other pMCS devices, ECMO increases LV afterload, inducing left ventricular distension [11][12][13][14][15].

Conclusion
Stent deformation and displacement is a rare complication of PCI associated with high mortality. With increased complexity of PCI-treated coronary lesions, interventional cardiologists can deal with this serious and potentially fatal complication. In this article, we have reported the successful treatment of stent dislodgement by crushing technique in a patient with refractory cardiac arrest assisted with VA-ECMO. VA-ECMO provides an effective temporary circulatory support in the catheterization laboratory in case of refractory cardiac arrest and its early initiation can be lifesaving.

Conflict of Interest
The authors have no conflicts of interest to declare.