Stent deformation and dislodgment into the aorta during mini-crush bifurcation PCI: a rare complication leading to emergency CABG Authors: Ezgi Çamlı Babayiğit, Alper Selim Kocaoğlu, Halit Emre Yalvaç, Erdi Babayiğit Published: 2026 ABSTRACT AbstractBackgroundStent dislodgment during percutaneous coronary intervention (PCI) has become exceedingly rare with modern stent designs; however, it may still occur in complex bifurcation interventions. When the event involves the left main coronary artery (LMCA), the consequences may be catastrophic.Case summaryA 65-year-old woman presented with anterolateral myocardial infarction. Coronary angiography revealed severe LAD–D1 bifurcation disease (LAD 90% after D1, D1 total occlusion) without significant calcification (Fig. 1). A mini-crush technique was planned using a 7 F JL4 guiding catheter. Following predilatation, a 2.75 × 13 mm DES was successfully deployed in D1, while a 2.75 × 18 mm DES had been pre-positioned (“parked”) in the LAD at the D1 level (Fig. 2). During withdrawal of the D1 stent balloon, the pre-positioned LAD stent became deformed, slipped from the balloon, and partially dislodged, hanging from the LMCA into the aorta (Fig. 3). Multiple retrieval attempts using LCB, EBU4, EBU3.5, and JL3.5 guide catheters and micro and macro snare systems were unsuccessful (Fig. 4). Heart team consultation recommended emergency coronary artery bypass grafting (CABG). The patient underwent LIMA–LAD, Ao–saphenous vein graft (SVG) to OM2, and Ao–SVG to D1. Intraoperative findings revealed that the slipped LAD stent was tightly embedded within the D1 stent and fractured during removal attempts. At 7-month follow-up, the patient remained asymptomatic with preserved LV function (EF 55%).DiscussionThis case highlights an important complication of bifurcation PCI. Pre-positioning (“parking”) of the main-vessel stent during mini-crush may carry an underrecognized risk of deformation and loss during side-branch balloon withdrawal, even in the absence of calcification or tortuosity.ConclusionOperators should be aware that stent parking in bifurcation PCI can predispose to balloon–stent interactions that may result in deformation, dislodgment, or embolization.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12872-026-05955-x. Introduction Stent dislodgment is an infrequent complication of PCI in the contemporary era, with reported incidences below 0.5% [1]. However, the risk increases in bifurcation lesions requiring complex two-stent techniques such as mini-crush, crush, culotte, or DK-crush. In particular, the manipulation of side-branch stents and balloons in close proximity to a parked main-vessel stent introduces unique mechanical interactions. LMCA involvement further increases the clinical risk, as stent deformation or embolization can lead to acute vessel compromise, hemodynamic instability, or the need for urgent surgery [2]. We report a case of LAD stent deformation during a mini-crush bifurcation PCI, resulting in unsuccessful percutaneous retrieval and emergency CABG. Case presentation A 65-year-old woman presented with an anterolateral myocardial infarction. She had a history of hypertension. She was on treatment with nebivolol 5 mg once daily and a fixed-dose combination of perindopril/amlodipine/indapamide 10/5/2.5 mg once daily. Baseline angiography revealed; LMCA: Normal, LAD: 90% severe stenosis after D1; D1 total occlusion, LCx: Normal, RCA: Proximal 30% plaque. Given the true LAD–D1 bifurcation lesion, a mini-crush strategy was selected (Fig. 1). There was no significant calcification or tortuosity. Baseline coronary angiography demonstrating a true LAD–D1 bifurcation lesion. A severe stenosis is present in the LAD distal to the first diagonal branch (red arrow), while the D1 branch is totally occluded (yellow arrow). The left main coronary artery (LMCA) is indicated by the blue arrow A 7 F sheath via femoral access and 7 F JL-4 guiding catheter (BOSTON SCIENTIFIC, Natick, MA, USA) were used. Intracoronary heparin (7,000 units) was administered. LAD and D1 were wired with floppy coronary guidewires (Shun S, Shunmei Medical Co). Predilatation was performed with a 2.0 × 15 mm semi-compliant balloon in D1 (Shun Loach, Shunmei Medical Co.) at 14 atm, restoring flow. Subsequently, a 2.0 × 15 mm semi-compliant balloon (Shun Loach, Shunmei Medical Co.) was inflated in the LAD at 16 atm. For the mini-crush strategy, a 2.75 × 18 mm DES (Firehawk, Microport) was first advanced and positioned across the LAD at the level of the D1 bifurcation and intentionally parked in the main vessel. A 2.75 × 13 mm DES (Firehawk, Microport) was then deployed in D1 at 14 atm (Fig. 2). During withdrawal of the D1 stent balloon, the balloon likely interacted with the parked LAD stent delivery system, catching the stent struts and causing deformation with loss of attachment from the delivery balloon. The stent subsequently migrated proximally, with its distal segment trapped beneath the D1 stent, while the body extended through the LMCA and protruded into the aortic root (Fig. 3, Supplementary Video 1). Intravascular imaging (IVUS/OCT) was not used because these modalities were not available in our catheterization laboratory at the time of the procedure. Percutaneous retrieval was attempted using multiple guiding catheters (LCB, EBU 4, EBU 3.5, and JL 3.5). Snare systems with loop sizes of 4–8 mm and 12–20 mm, advanced through a 7 F system, were used in repeated attempts to capture the dislodged stent; however, engagement of the severely deformed stent frame was unsuccessful (Fig. 4, Supplementary Video 2). Mild resistance during side-branch balloon withdrawal suggested balloon–stent interaction as the likely mechanism. Rewiring, balloon trapping, and stent crushing were not feasible due to severe deformation of the stent and its proximal extension into the LMCA and aortic root. The heart team recommended emergency CABG. The patient remained hemodynamically stable throughout the procedure, with no signs of hemodynamic compromise prior to referral for CABG. Surgical revascularization was performed, including LIMA–LAD and aorto–saphenous vein grafts to OM2 and D1. Intraoperative findings confirmed that the stripped LAD stent had become firmly embedded within the newly deployed D1 stent, explaining the failure of percutaneous retrieval. Importantly, this observation provides direct mechanistic support for the proposed balloon–stent interaction, whereby engagement of the parked stent struts during side-branch balloon withdrawal led to deformation, entrapment, and subsequent proximal migration. Procedural steps during bifurcation PCI. A Deployment of a 2.75 × 13 mm drug-eluting stent in the first diagonal branch (D1) after predilatation. B A 2.75 × 18 mm drug-eluting stent positioned in the LAD across the bifurcation and intentionally parked in the main vessel at the level of the D1 bifurcation Angiographic image demonstrating deformation and proximal migration of the parked LAD stent during withdrawal of the D1 balloon. The distal portion of the LAD stent is trapped beneath the D1 stent (red arrow), while the proximal segment extends through the left main coronary artery into the aortic root (yellow arrow). The dynamic sequence is shown in Supplementary Video 1 Percutaneous retrieval attempts of the dislodged LAD stent. A–D Multiple attempts were performed using snare systems introduced through different guiding catheters (LCB, EBU 4, EBU 3.5, and JL 3.5). The snare loop (yellow arrows) was advanced toward the dislodged stent (red arrows); however, engagement of the deformed stent frame was unsuccessful, likely due to deformation and partial entrapment beneath the previously deployed D1 stent. The retrieval attempts are also demonstrated in Supplementary Video 2 At 7-month follow-up, the patient remained stable and asymptomatic. Echocardiography showed EF 55%, mild septal hypokinesia, calcific aortic valve with mild AR, and mild MR. Discussion This report describes a rare instance of stent deformation with subsequent LMCA–aortic dislodgement during mini-crush bifurcation PCI in the absence of calcification, tortuosity, or device resistance. The classic crush technique involves intentional protrusion of the side-branch (SB) stent 4–5 mm into the main vessel (MV), later refined with kissing balloon inflation (KBI) to reduce restenosis [3]. The CACTUS trial demonstrated that classic crush yielded similar major adverse cardiac event rates compared with provisional stenting but did not show superiority; moreover, failure of KBI occurred in approximately 25% of cases and was associated with higher rates of myocardial infarction and target lesion revascularization [4]. To reduce the large amount of crushed metal created by the classic crush technique, the strategy evolved into the mini-crush technique, which targets approximately 2 mm of SB protrusion. With this modification, increasing operator experience, and the use of second-generation drug-eluting stents, KBI success rates have increased to as high as 98% [5]. Contemporary mini-crush techniques also incorporate proximal optimization and are typically performed using a 7-F system. In the present case, the LAD stent had been intentionally advanced and parked across the bifurcation at the level of D1 prior to side-branch stent deployment. During withdrawal of the D1 balloon, mechanical interaction likely occurred between the balloon and the parked LAD stent delivery system, resulting in longitudinal deformation and loss of balloon–stent attachment. This interaction ultimately led to proximal migration of the stent into the LMCA and aortic root. Such device–device interactions may be underestimated during routine bifurcation PCI. Mechanisms of device–device interaction in bifurcation PCI In the present case, the most likely mechanism was balloon–stent interaction during withdrawal of the side-branch (SB) balloon, which likely engaged the struts of the parked LAD stent, resulting in longitudinal deformation, loss of balloon–stent attachment, and subsequent proximal migration into the LMCA and aortic root. The potential mechanisms underlying this complication can be grouped into three main categories: Procedural factors Procedural aspects play a central role in device–device interaction. Excessive protrusion of the SB stent beyond the intended 1–2 mm in mini-crush may increase the metallic burden within the main vessel, predisposing to entrapment during balloon withdrawal [4]. Studies comparing classic and mini-crush techniques demonstrated that larger protrusion volumes correlate with higher procedural complexity and greater risk of device–device interference [6]. In addition, residual balloon inflation—either due to incomplete negative suction, asymmetric balloon folding, or balloon compliance—can create small protrusions capable of catching stent struts during withdrawal [7]. This mechanism is often underestimated, as angiographic images do not reliably visualize subtle balloon asymmetry. Several case reports have shown that even a minimal amount of residual balloon volume can markedly increase friction against metallic scaffolding, particularly in bifurcation anatomies [8]. Device-related factors Stent design characteristics, particularly reduced longitudinal strength or suboptimal crimping, may increase susceptibility to deformation under traction forces [9]. Although modern DES platforms have improved structural integrity, longitudinal deformation and stent loss remain recognized complications [10]. Anatomical and imaging-related factors In the absence of intravascular imaging, subtle malapposition, carinal shift, or misalignment between devices may go undetected. Even in angiographically simple lesions, such micro-level misalignment can create an “interlocking” effect during device manipulation [11]. Intravascular imaging could help optimize stent positioning and reduce the risk of such interactions. Taken together, these factors illustrate that device–device mechanical interactions in bifurcation PCI may arise from a combination of procedural, device-related, and anatomical elements, even in apparently straightforward cases. Role of intravascular imaging The absence of intravascular imaging represents an important limitation in the present case and may have contributed to the complication. IVUS or OCT could have provided detailed assessment of side-branch stent protrusion, allowing more precise control of the intended 1–2 mm extension during the mini-crush technique. In addition, intravascular imaging may have identified subtle malapposition, stent misalignment, or carinal shift that are not apparent on angiography alone but may predispose to device–device interaction. Furthermore, real-time imaging guidance could have facilitated optimization of stent positioning and spatial relationships between devices, potentially reducing the risk of balloon–stent entrapment during withdrawal. Therefore, this case highlights that intravascular imaging is not only a tool for lesion assessment but also a critical adjunct for procedural safety, particularly in complex bifurcation interventions. Overall significance Taken together, these factors illustrate that device–device mechanical interactions in bifurcation PCI can occur even when angiographic anatomy appears straightforward and no calcification or tortuosity is present. Recognizing these mechanisms may help reduce the incidence of such rare but severe complications. Attention to side-branch stent protrusion length, meticulous confirmation of complete balloon deflation prior to withdrawal, selection of stents with favorable longitudinal strength, optimization of device alignment with intravascular imaging when available, and consideration of alternative strategies such as balloon parking represent practical measures to enhance procedural safety. In our case, these findings are consistent with a balloon–stent interaction mechanism during withdrawal of the side-branch balloon, whereby engagement of the parked LAD stent struts led to longitudinal deformation, loss of balloon–stent attachment, and subsequent proximal migration into the LMCA and aortic root. From a practical perspective, this case underscores that even in the absence of complex lesion characteristics, procedural factors alone may create substantial mechanical risk. Operators should therefore remain particularly vigilant during balloon withdrawal in the presence of a parked stent and consider modifying the procedural strategy, especially when intravascular imaging is not available. Challenges in retrieval and bailout strategies Stent loss in the LMCA is associated with significant hemodynamic risk and requires prompt decision-making. In the present case, retrieval attempts failed most likely due to severe stent deformation and partial anchoring within the D1 stent, which markedly limited stent mobility. Snare techniques were unsuccessful, likely because the deformed stent provided no stable segment for capture. Balloon trapping or re-engagement was not feasible due to proximal extension of the stent into the aorta and loss of its tubular structure, preventing safe rewiring. Guideliner-assisted retrieval was also unlikely to succeed given the combination of deformation, LM involvement, and partial embedding. From a practical standpoint, bailout strategies for stent loss can be approached in a stepwise manner: Attempt snare retrieval if the stent is mobile and accessible, Attempt snare retrieval if the stent is mobile and accessible, Consider balloon trapping or small balloon re-engagement if the stent can be rewired, Consider balloon trapping or small balloon re-engagement if the stent can be rewired, Use guide extension systems to improve support and alignment when feasible, Use guide extension systems to improve support and alignment when feasible, Consider crushing the stent in situ if anatomically suitable, and Consider crushing the stent in situ if anatomically suitable, and Proceed to emergency CABG when percutaneous options fail or are deemed unsafe. Proceed to emergency CABG when percutaneous options fail or are deemed unsafe. In our case, the combination of stent deformation, proximal extension into the aorta, and anchoring within the D1 stent made percutaneous retrieval ineffective, and emergency CABG represented the safest definitive treatment. Conclusion Stent dislodgment during bifurcation PCI is uncommon but may occur even in anatomically uncomplicated settings when mechanical interaction between devices is underestimated. This case highlights that stent parking during mini-crush may carry unrecognized mechanical risks, particularly during side-branch balloon withdrawal, even in the absence of calcification or tortuosity. Although stent parking is a recognized component of the mini-crush technique, balloon parking may represent a reasonable alternative in selected cases, particularly when intravascular imaging is not available. Operators should remain vigilant during mini-crush procedures, especially during withdrawal of SB balloons near a parked MV stent. Familiarity with bailout strategies is essential, and emergency CABG remains a lifesaving option when percutaneous retrieval fails. Supplementary Information Supplementary Material 1. Supplementary Material 2. Acknowledgements The authors declare that there are no individuals or organizations to acknowledge. Abbreviations Percutaneous coronary intervention Left anterior descending artery First diagonal branch Left main coronary artery Coronary artery bypass grafting Drug-eluting stent Intravascular ultrasound Optical coherence tomography Left internal mammary artery Saphenous vein graft Kissing balloon inflation Side branch Main vessel Authors’ contributions Ezgi Çamlı Babayiğit: Interventional procedure, Conceptualization, data collection, manuscript drafting, and approved the final manuscript. Alper Selim Kocaoğlu: Surgical management, clinical management of the patient, and critical revision of the manuscript. Halit Emre Yalvaç: Data interpretation, literature review, and manuscript editing. Erdi Babayiğit: Interventional procedure, operative data contribution, and critical revision of the manuscript. Funding This study received no funding from any public, commercial, or nonprofit organization. Data availability All data generated or analyzed in this study are available in the published article. For any additional questions, please contact the corresponding author. Declarations Ethics approval and consent to participate Ethics committee approval was not required for this case report according to institutional policy. Consent for publication Written informed consent was obtained from the patient for publication of all clinical details and associated images. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Data Availability Statement All data generated or analyzed in this study are available in the published article. For any additional questions, please contact the corresponding author.