Frequency, Treatment, and Consequences of Device Loss and Entrapment in Contemporary Percutaneous Coronary Interventions

Abstract: Background. Device loss and entrapment are infrequent but potentially grave complications of percutaneous coronary interventions (PCI). There are limited contemporary data on the frequency, treatment, and consequences of these complications. Methods. We reviewed 2338 consecutive PCI cases performed between 1/2005 and 5/2010 at our institution to identify cases of device loss or entrapment. The angiograms and outcomes of these patients were reviewed. Results. During the study period, device loss occurred in 9 cases (0.38%; 95% confidence interval [CI], 0.18%-0.73%) and entrapment in 4 cases (0.17%; 95% CI, 0.05%-0.44%). The lost devices were stents (n = 5; 0.21%), a coronary balloon shaft (n = 1; 0.04%), a femoral arterial sheath (n = 1; 0.04%), an arterial catheter (n = 1; 0.04%), and an Ostial Pro catheter (Ostial Solutions) distal tip (n = 1; 0.04%). Entrapped devices included a coronary guidewire (n = 2; 0.08%), a Tornus catheter (Abbott Vascular; n = 1; 0.04%) and a Filterwire (Boston Scientific; n = 1; 0.04%). All patients with device loss were successfully managed percutaneously (1 patient experienced periprocedural myocardial infarction). Retrieval of the lost devices was attempted in 7 of 9 cases (78%) and was successful in 6 cases (86%). Retrieval was successful with the initial attempt in 2 patients but required >1 attempt in 4 patients. In contrast, 3 of 4 patients (75%) with device entrapment required emergency surgical removal and coronary artery bypass grafting. Conclusions. Device loss or entrapment is an infrequent complication of contemporary PCI. Device loss can be successfully managed percutaneously, whereas device entrapment often requires emergency cardiac surgery.

J INVASIVE CARDIOL 2012;24(5):215-221

Key words: stent loss, wire entrapment, percutaneous coronary intervention, stent retrieval, complication

____________________________________________

Although the safety of percutaneous coronary interventions (PCI) has significantly improved over time, complications may still occur.¹ One such infrequent, yet potentially serious complication is device loss or entrapment. There is limited published information on the frequency and consequences of this complication in the contemporary PCI era.^2-10 The goal of the present study was to review the frequency of device loss or entrapment in a contemporary tertiary care practice and describe its management and outcomes.

Methods

Study population. We retrospectively reviewed the procedural reports and coronary angiograms of 2338 consecutive PCI procedures performed at our institution between January 2005 and May 2010 to identify cases during which device loss or entrapment occurred. Information was abstracted on the type and possible causes of each complication, on procedural management, and patient outcomes. The procedural strategies, equipment used, and anticoagulation and antiplatelet regimens were at the discretion of the interventional cardiologist performing the procedure. Retrieval techniques and devices were also selected and employed at the discretion of the operator. The study was approved by the Institutional Review Board of our institution.

Definitions. Device loss was defined as dislodgement of a device inside the body at a location different than the intended delivery location. Device entrapment was defined as inability to remove the shaft of a device from the body. Retrieval success was defined as successful extraction of the lost or entrapped device.

Retrieval techniques and devices. Several retrieval techniques were utilized in our series.² The small-balloon technique^3-7 can be used when a stent is dislodged but guidewire access is maintained through the lost stent. A small balloon is advanced through the stent, inflated distal to the stent, and then withdrawn together with the lost stent.

Several loop snares, such as the Amplatz Goose Neck snare (ev3), the Microsnare Elite (Vascular Solutions) and the En Snare (Merit Medical) are available for facilitating retrieval efforts. These snares consist of a wire loop typically made of nitinol that is advanced through a microcatheter (or alternatively through a diagnostic or guide catheter), positioned around the lost device, and then pulled back, trapping the device against the catheter. The catheter/loop assembly is subsequently removed from the body, along with the lost device. The En Snare is a specially designed snare with 3 overlapping loops, designed to increase the likelihood of retrieving the lost device. Loop snares can also be fashioned in the catheterization laboratory inserting an exchange-length coronary guidewire through a multipurpose catheter and re-inserting the distal tip of the guidewire through the distal tip of the catheter.²

In the two-wire technique, two guidewires are advanced through the lost device, usually a stent, and rotated several fold in order to entangle their distal ends.^8,9 Once the wires are withdrawn, they may carry the lost device.

Finally, the novel “hairpin-trap” technique was used to retrieve a lost stent in our series, as recently described.¹⁰ In this procedure, a polymer jacketed guidewire tip is bent into a hairpin loop and inserted through the touhy valve until it exits through the guiding catheter’s distal end. The hairpin is then advanced through the lost stent and withdrawn, “hooking” the stent through the now-open hairpin. The free tip of the guidewire is reinserted into the guiding catheter and a small balloon is inflated within the guide catheter tip, adjacent to the re-inserted guidewire. The entire system is then withdrawn, removing the lost stent.

Results

Patient characteristics. Device loss or entrapment occurred in 13 of 2338 PCIs performed during the study period (0.56%; 95% confidence interval [CI], 0.30%-0.95%). Device loss occurred in 9 cases (0.38%; 95% CI, 0.18%-0.73%) and entrapment in 4 cases (0.17%; 95% CI, 0.05%-0.44%). The baseline characteristics of the 13 device loss or entrapment cases are shown in Table 1. Mean age was 64 ± 10 years and 92% of the patients were men. The patients had a high prevalence of diabetes, hyperlipidemia, hypertension, prior myocardial infarction, and prior PCI. The mean fluoroscopy time and contrast volume used in the device loss/entrapment series was 46 ± 23 minutes and 324 ± 196 mL, respectively.

Device loss. The lost devices (Table 2) were stents (n = 5; 0.21%; 95% CI, 0.07%-0.50%), a coronary balloon shaft (n = 1; 0.04%; 95% CI, 0.071%-0.24%), a femoral arterial sheath (n = 1; 0.04%; 95% CI, 0.071%-0.24%), an arterial line (n = 1; 0.04%; 95% CI, 0.071%-0.24%), and an Ostial Pro catheter distal tip (n = 1; 0.04%; 95% CI, 0.071%-0.24%). All stents were pre-mounted drug-eluting stents. All patients with device loss were successfully managed percutaneously and only 1 patient experienced periprocedural myocardial infarction. Retrieval of the lost devices was attempted in 7 of 9 cases (78%) and was successful in 6 cases (86%). Retrieval was successful with the initial attempt in 2 patients but required >1 attempt in 4 patients.

Of the 5 stent loss cases, the stent was retrieved in 2 cases and deployed in the other 3 (Table 2). Stent loss was due to tortuosity (1 case), tortuosity and calcification (1 case), and attempts to deliver a stent through previously deployed stents (3 cases). Retrieval was attempted in 3 cases; in 1 case, the small-balloon technique was utilized without success, followed by stent deployment; in a second case, the small-balloon technique and a Microsnare Elite failed, followed by successful retrieval using the “hairpin trap” technique;10 in a third case, snaring with a loop snare failed, but retrieval was successful with an En Snare (Figure 1). In the remaining 2 cases of stent loss, the stents were deployed in the target vessel without attempting to retrieve them.

In addition to the 5 stents, a portion of a femoral sheath, an arterial catheter, a broken balloon shaft, and an Ostial Pro catheter were lost and percutaneously retrieved.

At the end of a procedure, a 7 Fr 45-cm long femoral arterial sheath unraveled during attempts to exchange it for a shorter sheath. After contralateral femoral access was obtained, a 10 mm Gooseneck snare was used to snare and externalize the J-tipped 0.035˝ wire, over which the sheath had been inserted. Advancement of a 10 mm Gooseneck snare and an En Snare retrogradely over the externalized guidewire was not possible due to acute aortic bifurcation angle. A snare was created using an Ironman wire (Abbott Vascular) and a diagnostic catheter, allowing successful snaring of the sheath. The unraveled sheath segment was cut and the avulsed sheath segment was successfully retrieved through the contralateral femoral artery. Bilateral iliofemoral angiography did not reveal any vessel injury.

In the second case, a balloon shaft fractured during PCI of a right coronary artery lesion. The distal balloon shaft fragment remained partially in the vessel and partially in the guide catheter and was successfully retrieved using a Gooseneck snare  (Figure 2).

In a third case, an 18-gauge femoral arterial catheter was incised during attempts to exchange the catheter for a sheath. The catheter fragment embolized to the right peroneal artery and was removed using a Microsnare Elite (Figure 3).

In the fourth case, an Ostial Pro stent positioning system was used in an attempt to stent an ostial saphenous vein graft lesion. During attempts to position the device, its distal portion fractured. The fractured fragment was successfully snared using a loop snare and removed without complications.

Device entrapment. In contrast to device loss, 3 of 4 patients (75%) with device entrapment required emergency surgical removal and coronary artery bypass grafting. The entrapped devices were coronary guidewires (n = 2), a Filterwire (Boston Scientific), and a Tornus catheter (Abbott Vascular).

In the first entrapment case, a patient underwent stenting of an unprotected left main coronary artery. Both the left anterior descending and the circumflex artery were wired and the left main was stented with a 3.0 x 18 mm everolimus-eluting stent into the left anterior descending artery, jailing the circumflex guidewire. During attempts to remove the jailed circumflex guidewire, its distal trip became entangled with the left main stent, forming a knot (Figure 4D). Multiple retrieval attempts failed and the patient underwent surgical wire retrieval followed by coronary artery bypass graft surgery (Figure 4).

In the second case, wire loss occurred when the tip of a Cross-It 100 XT wire (Abbott Vascular) fractured and became entrapped during attempts to cross a chronic total occlusion of the right coronary artery. Attempts to cross the lesion with a Miracle 3 guidewire (Abbott Vascular) resulted in dissection, so the procedure was aborted. Comparison of the retrieved guidewire fragment with an intact guidewire demonstrated that the avulsed segments was <3 cm in length. The device remained embedded in the lesion without other complications.

In the third case, a Filterwire was used during attempts to stent a saphenous vein graft to mid left anterior descending artery. After stenting, the Filterwire could not be retrieved in spite of multiple attempts, requiring surgical removal.¹¹

In the fourth patient, a Tornus catheter (Abbott Vascular) became entrapped during attempts to cross a mid right coronary artery chronic total occlusion. Attempts to advance a second guidewire next to the entrapped catheter failed and the patient required surgical removal of the catheter and coronary artery bypass grafting, followed by an uneventful recovery.

Discussion

Our case series demonstrates that device loss or entrapment is an infrequent complication of contemporary PCI. Device loss could be successfully managed percutaneously in all cases, whereas device entrapment required coronary artery bypass graft surgery in most cases.

Stents were the most commonly lost devices. The reported incidence of stent loss is <1% in most contemporary PCI series (Table 3). The incidence of stent loss in our series (0.21%) is the lowest ever reported, and is likely a reflection of the improvements in stent design, adjunctive equipment, and stenting techniques. The most common cause of stent loss in our series was attempts to deliver a stent through a previously deployed stent.^12-14 Similarly, Kozman et al reported that 6 of 23 stent dislodgements (26%) occurred while passing through a previously placed stent.¹³ Stenting distally to proximally may prevent such a complication, although occasionally the need for distal stenting does not arise until after a proximal stent is deployed, for example in cases of distal edge dissection. Sometimes stent deformation can occur during attempts to cross the lesion and then the stent is stripped off while attempting to withdraw it into the guide catheter. Stent delivery through a deeply engaged “mother and child” guide catheter, a Guideliner catheter, or a Proxis catheter may facilitate stent delivery and prevent stent loss.³ Other factors contributing to stent loss are tortuosity^3,12,15 and calcification.^12,13,16 Careful preparation of challenging lesions before stent delivery (with predilation, atherectomy, etc.)15 could also optimize stent delivery and lower the risk of stent loss.

Another important observation from ours and other series is that stent retrieval is not always necessary; in the majority of stent loss cases in our series, the stent was deployed (or crushed with another stent) either because retrieval failed (in 1 case) or as the default strategy (in 2 cases). Indeed, stent deployment in a coronary segment that is unlikely to be significantly affected by stenting may be the most time-efficient and low-risk strategy, as stent retrieval attempts can prolong the procedure, increase radiation exposure, and result in distal stent embolization or target vessel injury. Occasionally, the lost stents may be difficult to visualize, especially thin strut stents in obese patients with calcified or previously stented coronary arteries. In such cases, intravascular ultrasonography may facilitate localization of the stent (Figure 1).

Our series shows that apart from stents, other devices can be lost or entrapped during PCI. These devices range from coronary guidewires to femoral arterial sheaths. Often these devices are bulkier than the stents, which may facilitate retrieval, but may also result in compromised antegrade flow and severe ischemia.

The incidence of wire entrapment is under-reported. When a guidewire fractures, the distal spring coil may unravel, creating a metal “bird’s nest” that could predispose to thrombosis (Figure 4F and 4G). In cases of guidewire fracture, it is important to perform intravascular ultrasonography to confirm that no wire coil unraveling has occurred.¹⁷ In such cases, covering the guidewire with a stent may suffice,¹⁸ although successful percutaneous wire fragment removal techniques have been described.¹⁷

Device loss or entrapment can be a severe complication leading to emergency surgery or death. Because it is infrequent, most operators may not have large experience on how to approach such patients. Moreover, some cardiac catheterization laboratories may not stock retrieval equipment or may have limited familiarity with its use. That is why regular review of the various retrieval techniques and equipment is important. Training in retrieval techniques is similar to training on how to extinguish a fire or how to respond to an earthquake. It is possible that one may never experience such an event, but if one does, knowing how to best respond can make the difference between recovery, emergency surgery, or death.

We commonly use snares through a guiding catheter, because the snare delivery catheters are often too short. For retrieval of intracoronary lost devices (usually stents) the MicroElite snare (which is 0.014˝ in diameter, has loop size of 2-7 mm, is 180 cm in length and does not require a delivery catheter) or the 2 mm En Snare are commonly utilized. For retrieving objects in the aorta (for example, retrogradely externalized guidewires), an 18-30 mm Ensnare through a JR4 or multipurpose diagnostic or guide catheter is our snare of choice, because of the 3-loop design that facilitates object retrieval. However, it should be used with caution as the snare wires may cause vessel injury.

Study limitations. Our study has important limitations. It was a retrospective analysis from a single center with extensive experience in treating complex patients. The device types lost or entrapped were heterogeneous, limiting the ability to provide broad generalizations of the findings, but providing a comprehensive report of what device loss or entrapment may occur in a contemporary cardiac catheterization laboratory. The equipment used in our patients may not be available in all cardiac catheterization laboratories. Almost all patients were men, although it is unknown whether gender differences exist in device loss or retrieval.

Conclusion

In contemporary PCIs, device loss and entrapment is infrequent but can occur with a variety of devices. Device loss most often can be treated percutaneously, yet device entrapment requires emergency cardiac surgery in most cases.

References

1. Prasad A, Herrmann J. Myocardial infarction due to percutaneous coronary intervention. N Engl J Med. 2011;364(5):453-464.
2. Stephen G. Ellis DRH. Strategic Approaches in Coronary Intervention, 3rd edition. Lippincott Williams & Wilkins: 2005.
3. Brilakis ES, Best PJ, Elesber AA, et al. Incidence, retrieval methods, and outcomes of stent loss during percutaneous coronary intervention: a large single-center experience. Catheter Cardiovasc Interv. 2005;66(3):333-340.
4. Eggebrecht H, Haude M, von Birgelen C, et al. Nonsurgical retrieval of embolized coronary stents. Catheter Cardiovasc Interv. 2000;51(4):432-440.
5. Farshid A, Pitney MR. Intracoronary embolization and retrieval of radio-opaque ring marker on the ACS Multi-Link stent sheath. Cathet Cardiovasc Diagn. 1998;43(3):306-307.
6. Qiao S, Gao R, Chen J, et al. Successful retrieval of intracoronary lost balloon-mounted stent using a small balloon. Chin Med J (Engl). 2000;113(1):93-94.
7. Rozenman Y, Burstein M, Hasin Y, Gotsman MS. Retrieval of occluding unexpanded Palmaz-Schatz stent from a saphenous aorto-coronary vein graft. Cathet Cardiovasc Diagn. 1995;34(2):159-161.
8. Bogart DB, Jung SC. Dislodged stent: a simple retrieval technique. Catheter Cardiovasc Interv. 1999;47(3):323-324.
9. Veldhuijzen FL, Bonnier HJ, Michels HR, el Gamal MI, van Gelder BM. Retrieval of undeployed stents from the right coronary artery: report of two cases. Cathet Cardiovasc Diagn. 1993;30(3):245-248.
10. Brilakis ES, Abdel-Karim AR, Banerjee S. Hairpin-trap: a novel stent retrieval technique. Catheter Cardiovasc Interv. 2011;77:213-216.
11. Badhey N, Lichtenwalter C, de Lemos JA, et al. Contemporary use of embolic protection devices in saphenous vein graft intervention: insights from the stenting of saphenous vein grafts trial. Catheter Cardiovasc Interv. 2010;6(2):263-269.
12. Elsner M, Peifer A, Kasper W. Intracoronary loss of balloon-mounted stents: successful retrieval with a 2 mm “Microsnare” device. Cathet Cardiovasc Diagn. 1996;39(3):271-276.
13. Kozman H, Wiseman AH, Cook JR. Long-term outcome following coronary stent embolization or misdeployment. Am J Cardiol. 2001;88(6):630-634.
14. Lohavanichbutr K, Webb JG, Carere RG, et al. Mechanisms, management, and outcome of failure of delivery of coronary stents. Am J Cardiol. 1999;83(5):779-781, A9.
15. Ijsselmuiden AJ, Tangelder GJ, Cotton JM, et al. Direct coronary stenting compared with stenting after predilatation is feasible, safe, and more cost-effective in selected patients: evidence to date indicating similar late outcomes. Int J Cardiovasc Intervent. 2003;5(3):143-150.
16. Holmes DR Jr, Garratt KN, Popma J. Stent complications. J Invasive Cardiol. 1998;10(7):385-395.
17. Sianos G, Papafaklis MI. Septal wire entrapment during recanalisation of a chronic total occlusion with the retrograde approach. Hellenic J Cardiol. 2011;52(1):79-83.
18. Marti V, Markarian L. [Angioplasty guidewire entrapment affer stent implantation: report of two cases and review of the literature]. Arch Cardiol Mex. 2007;77(1):54-57.
19. Foster-Smith KW, Garratt KN, Higano ST, Holmes DR Jr. Retrieval techniques for managing flexible intracoronary stent misplacement. Cathet Cardiovasc Diagn. 1993;30(1):63-68.
20. Alfonso F, Martinez D, Hernandez R, et al. Stent embolization during intracoronary stenting. Am J Cardiol. 1996;78(7):833-835.

21. Cantor WJ, Lazzam C, Cohen EA, et al. Failed coronary stent deployment. Am Heart J. 1998;136(6):1088-1095.
22. Dunning DW, Kahn JK, O’Neill WW. The long-term consequences of lost intracoronary stents. J Interv Cardiol. 2002;15(5):345-348.
23. Nikolsky E, Gruberg L, Pechersky S, et al. Stent deployment failure: reasons, implications, and short- and long-term outcomes. Catheter Cardiovasc Interv. 2003;59(3):324-328.
24. Roffi M, Luscher TF, Sutsch G, et al. Failure to retrieve undeployed paclitaxel-eluting coronary stents. Am J Cardiol. 2006;97(4):502-505.

____________________________________________

From the VA North Texas Healthcare System and University of Texas Southwestern Medical Center, Dallas, Texas.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. S. Banerjee has received speaker honoraria from St. Jude Medical, Cordis, and Medtronic and research support from Boston Scientific. ES Brilakis has received speaker honoraria from St. Jude Medical and Terumo, research support from Abbott Vascular and his spouse is an employee of Medtronic.
Manuscript submitted November 29, 2011, provisional acceptance given January 2, 2012, final version accepted January 16, 2012.
Address for correspondence: Emmanouil S. Brilakis, MD, PhD, FSCAI, Dallas VA Medical Center (111A), 4500 South Lancaster Road, Dallas, TX 75216. Email: esbrilakis@yahoo.com