Stent Loss and Retrieval During Percutaneous Coronary Interventions: A Systematic Review and Meta-Analysis

Abstract: Objective. We present a systematic review and meta-analysis of the frequency, consequences, and treatment of stent loss during percutaneous coronary intervention (PCI). Background. Stent loss during PCI has received limited study. Methods. We conducted a meta-analysis of 18 case series and 45 case reports published between 1991 and 2012 on stent loss during PCI. Data on the frequency of stent loss, treatment strategies, and clinical outcomes were collected. Results. A total of 1048 stents were lost in 968 PCIs. Stent loss occurred in 919 of 71,655 PCIs (1.3%; 95% confidence interval, 0.8%-2.8%). The incidence of stent loss in studies published before 2000, between 2000 and 2005, and after 2005 was 5%, 1%, and 0.3%, respectively (P<.001). Of the 1048 lost stents, 698 (66%) were successfully retrieved from the coronary circulation, 130 (12%) were deployed, 27 (3%) were crushed, and 28 (3%) were left untreated; treatment of 165 stents (16%) was not reported. A complication occurred in 171 patients (19%), of whom 98 (57%) had coronary artery bypass graft surgery, 31 (18%) myocardial infarction, 33 (19%) died, 10 (6%) had bleeding requiring transfusion, 5 (3%) had vascular access complications, and 1 patient (0.6%) had a cerebrovascular accident (some patients had more than one event). Conclusions. The incidence of stent loss during PCI is low and has been decreasing. Although the lost stents were successfully retrieved in most cases, stent loss was associated with high rates of complications, such as coronary artery bypass graft surgery, myocardial infarction, and death.

J INVASIVE CARDIOL 2013;25(12):637-641

Key words: percutaneous coronary intervention, complications, stent loss

_____________________________

Coronary stents have revolutionized the practice of interventional cardiology by reducing vessel recoil and the risk of restenosis during percutaneous coronary intervention (PCI).¹ In-stent restenosis and stent thrombosis are the most common complications after stent implantation.² Stent loss during at tempted deployment is a less common, but potentially life-threatening, PCI complication that has received limited study.³ We performed a systematic review and meta-analysis of the published literature on stent loss during PCI to determine its frequency, temporal trends, risk factors, treatment, and impact on clinical outcomes.

Methods

Search strategy and eligibility criteria. We performed a search of the Pubmed and Cochrane Library databases for manuscripts on coronary stent loss and stent retrieval. Bibliographies of the retrieved studies were searched by hand for other relevant studies. Human studies in English published between the years 1991 to 2012 were included if they reported the frequency of stent loss and associated treatment (including retrieval or deployment attempts) and clinical outcomes. Stent loss was defined as dislodgment of the stent from the stent delivery balloon in a location different that the intended area for deployment. Review articles were excluded, but case reports were included. A list of the included studies is shown in Supplemental Table 1. 

Data extraction. Data from included studies were extracted by two authors (MA and TM) and controversies were reviewed by both. In case of disagreement, a third reviewer (EB) was used to reach consensus. Extracted data included the total number of patients, the frequency of stent loss, the various treatment strategies (retrieval vs deployment vs crushing), age, sex, history of prior coronary artery bypass graft (CABG), target vessel, mechanisms of loss, clinical presentation, treatment including the use of retrieval devices/techniques, and occurrence of complications. The country and institute of origin, author, and enrollment period were reviewed to identify and exclude duplicate publications from the same cohort. The authors of 8 studies that reported incomplete information were contacted^4-11 and two of them^9,10 provided additional data for the present meta-analysis. To minimize the impact of publication bias inherent in case reports, only data reported in case series were used to estimate the frequency of stent loss and the resultant complications (ie, case reports were excluded from this part of the analysis).

Statistical analysis. The frequency of each evaluated outcome was abstracted from each study and presented as minimum, maximum, and cumulative rates. To assess heterogeneity across trials, we used the Cochrane Q statistic (heterogeneity P-value ≤.10 was considered significant) and I² statistic (25%, 50%, and 75% correlate with low, moderate, and high heterogeneity, respectively) for each outcome. Due to its conservative or “worst-case scenario” estimates, a random-effects model as described by DerSimonian and Laird was used to obtain a summary estimate and 95% confidence intervals (CIs).¹² The cumulative proportion and corresponding CIs are presented in Supplemental Table 2. Data collection, study selection, processing of the data, and reporting of the results were performed according to accepted principles related to systematic review and meta-analysis.^13-16 Publication bias for the frequency of stent loss and the occurrence of complications (death, emergency CABG, and composite major adverse cardiac events [MACE] endpoint) were estimated visually by funnel plots and are presented in Supplemental Figures 1-4. Proportions were compared using the chi-square test and P<.05 was considered statistically significant.

Results

Search results. Of the 1459 studies retrieved through electronic and manual search, a total of 63 studies (18 case series and 45 case reports) met the inclusion criteria and were included in the final pooled analysis (Figure 1). A full listing of all 63 articles is provided in Supplemental Table 1. 

Incidence of stent loss and procedural complications. The 18 published case series reported stent loss in 919 of 71,655 patients who underwent PCI. The incidence of stent loss ranged between 0.21%-8.4% with a pooled estimate of 1.3% (95% CI, 0.8%-2.3%; Table 1). 

Mean age of the study cohort was 63.5 years and most patients were men with high frequency of coronary artery disease risk factors and prior coronary events (Table 2). Most patients presented with stable angina. Stent loss most commonly occurred in the right coronary artery, followed by the left anterior descending artery. Table 3 outlines the angiographic characteristics of the lesions in which stent loss occurred, demonstrating high lesion complexity. 

The reported MACE rate among the 919 stent loss patients included in the 18 case series ranged between 0%-33% with a pooled estimate of 16.9% (95% CI, 10.6%-27%). Most of the MACE consisted of CABG, with a pooled estimate rate of 9.4% and a range of 5.7%-15.5%. There were 33 deaths in the cohort with a reported range from 0%-15%, yielding a pooled estimate of 5.8%. MI was less common, with a rate of 5.1% in pooled analyses. Inspection of funnel plots demonstrated no evidence for publication bias for stent loss rates, but a possibility for publication bias for MACE, CABG, and death (Supplementary Figures 1-4).

Mechanism of stent loss. Tortuosity and/or calcification were the most common angiographic characteristics associated with stent loss (n = 347; 36%), followed by failed stent retraction into guide catheter (n = 276; 28.3%), failure to cross the lesion (n = 93; 9.6%), failure of the stent to expand (n = 12; 1.2%), stent dislodgment during crossing of a previously implanted stent (n = 10; 1%), dissection after balloon dilatation (n = 10; 1%), angulation/poor guide support (n = 8; 0.8%), stent deformity (n = 1; 0.1%), or other causes (n = 18; 2%). The cause of stent loss was not reported in 193 PCIs (20%).

Stent retrieval. Of the 1048 lost stents reported in the case series and case reports, a total of 698 (66%) were successfully retrieved from the coronary circulation, of which 381 stents (55%) were subsequently removed from the body and 317 stents (45%) were lost again to the peripheral circulation; 130 stents (12%) were deployed, 27 stents (3%) were crushed, and 28 stents (3%) were left undeployed/uncrushed. In some cases, multiple retrieval devices and strategies were used to retrieve the lost stents (Table 4). Snares were the most commonly used devices (used in 33.7% of stent retrieval attempts). 

Clinical outcomes. In most PCIs (748/919; 81%) reported in case series of stent loss, the patient had an uneventful recovery. A complication occurred in 171 patients (19%), of whom 98 (57%) had CABG surgery, 31 (18%) had MI, 33 (19%) died, 10 (6%) had bleeding requiring transfusion, 5 (3%) had vascular access complications, and 1 (0.6%) had a cerebrovascular accident (some patients had more than one event).

Temporal trends. To evaluate changes over time, studies were divided into three groups based on publication year. Most studies were published within the past 10 years. The frequency of stent loss significantly decreased over time from 5.3% in studies published between 1991 and 1999 to 0.38% in studies published between 2005 and 2012 (P<.001) (Figure 2). The complication rates in cases in which stent loss occurred was 14.6% in studies published between 1991 to 1999, 19.4% between 2000 to 2004, and 12.3% in studies published between 2005 to 2012 (P=.09).

Discussion

Our meta-analysis is the first to evaluate the pooled frequency, treatment, and outcomes of stent loss during PCI. The main findings were that: (1) stent loss occurs infrequently and its frequency has been decreasing over time; (2) lost stents are successfully retrieved in the majority of cases; and (3) although most patients have an uneventful recovery, there is high frequency of procedural complications, such as emergency CABG surgery, MI, or death. 

Frequency of stent loss. The pooled incidence of stent loss in our study was 1.3% and has been decreasing over time (Figure 2). This is likely the result of improved stent design and stent delivery systems. The introduction of premounted stents contributed greatly to stent loss rate decline, which occurred in spite of treating increasingly complex groups of patients and lesions. Moreover, the stent cross-profile has decreased and the stent delivery systems have significantly improved.¹⁷

Treatment of stent loss. An attempt to retrieve the lost stent was made in approximately two-thirds of the stent loss patients. In most cases, the lost stents were retrieved using a snare, which consists of a loop advanced through a delivery catheter. The lost stent is captured by retrieving the loop into the delivery catheter, resulting in “capture” of the stent between the loop and the delivery catheter tip. There are currently several types of snares (single loop or multi-loop, and in several diameters ranging from 2-3 mm to 45 mm) enabling effective stent (or other device) retrieval.³

The second most commonly used retrieval technique was the small-balloon technique, which is only feasible if guidewire position is maintained through the lost stent. This is a simple technique, in which a small balloon is advanced through the lost stent, inflated distally, and withdrawn, displacing the lost stent into the guiding catheter.^18-21 

Other retrieval devices or techniques, such as forceps,²² basket retrieval devices,⁴ the Cook retained fragment retriever (Cook),²² embolic protection devices, such as filters,²³ the two-wire technique,^24,25 and the hairpin-wire technique²⁶ were infrequently used. Occasionally, the lost stent could be retrieved from the coronary circulation, but because of deformation it could not be removed from the body and required vascular surgery to be performed. However, stents lost in the peripheral circulation do not appear to cause complications. 

Although stent deployment or crushing was infrequently utilized (in 15% of patients), it may be a particularly attractive option for patients in whom stent loss occurs in a location in which stenting is unlikely to cause complications, since stent crushing/deployment can be achieved expeditiously. Rahman et al²⁷ described a unique technique where the embolized stent was pushed into the target lesion using conventional angioplasty wires and angioplasty balloon catheters and eventually deployed. This option can be quite challenging, but could be an alternative to crushing/deploying the embolized stents, if feasible.

Clinical consequences of stent loss. Our systematic review demonstrates that most patients in whom stent loss occurs are successfully treated without additional complications. However, some patients may require emergency CABG. In the series by Colkesen et al,²⁸ emergency CABG was performed in 3 of 14 patients (43%) due to jeopardized coronary flow by the embolized stent in 2 patients and due to hemodynamic deterioration requiring cardiopulmonary resuscitation in 1 patient. Three of 20 patients (15%) in whom stent loss occurred in the study by Eggebrecht et al¹⁸ died following emergency CABG; two patients had the surgery after a failed retrieval attempt and 1 patient underwent surgery because of inability to precisely localize the intracoronary lost stent. Since stent loss occurs rarely, prevention of complications related to stent loss relies heavily on familiarity of the operator with stent retrieval techniques and with alternative treatment options, such as crushing or deploying the stent,³ which might occasionally be safer than attempting to retrieve the lost stent.²⁹ In the majority of stent loss cases in the series by Iturbe et al,²⁹ the stent was deployed or crushed either because of a failed retrieval attempt (in 1 patient) or as the primary strategy (in 2 patients). 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, since stent retrieval attempts can prolong the procedure, increase radiation exposure, and result in distal stent embolization or target vessel injury.²⁹

Study limitations. There are several important limitations of our study. First, the number of studies reporting stent loss was relatively small. Second, reporting of certain complications, such as bleeding and dissection, was not performed by all studies. Third, publication bias is always possible in meta-analyses. Visual assessment of the funnel plot for stent loss rates appeared symmetric and unbiased; however, the funnel plot for MACE, CABG, and death suggested that publication bias was possible. Finally, while a patient-level meta-analysis would be ideal, such data were not available for the present review. 

Conclusion

Our systematic review and meta-analysis demonstrates that stent loss occurs infrequently and its incidence has decreased over time. Stents are successfully retrieved in most cases, and most patients have an uneventful recovery, although some patients (approximately 1 in 5) experience additional complications, such as need for emergency CABG, acute MI, or death, emphasizing the need for continued efforts to prevent and appropriately treat this complication.  

References

1. Sigwart U, Puel J, Mirkovitch V, Joffre F, Kappenberger L. Intravascular stents to prevent occlusion and restenosis after transluminal angioplasty. N Engl J Med. 1987;316(12):701-706.
2. Pullara A, Longo G, Gonella A, et al. Incidence and very long-term outcomes of stent thrombosis after bare-metal or drug-eluting stent implantation: a retrospective analysis. Minerva cardioangiol 2013;61(1):1-9.
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. Douard H, Besse P, Broustet JP. Successful retrieval of a lost coronary stent from the descending aorta using a loop basket intravascular retriever set. Cathet Cardiovasc Diagn. 1998;44(2):224-226.
5. Gomes WJ, Palma JH, Carvalho AC, Buffolo E. Surgical retrieval of undeployed intracoronary stent without cardiopulmonary bypass. Eur J Cardiothorac Surg. 1997;12(4):666-667.
6. Cicek D, Pekdemir H. A rare and avoidable complication of percutaneous coronary intervention: stent trapped in the left main coronary artery and its unusual treatment. Hellenic J Cardiol. 2011;52(4):367-370.
7. Kobayashi Y, Nonogi H, Miyazaki S, Daikoku S, Yamamoto Y, Takamiya M. Successful retrieval of unexpanded Palmaz-Schatz stent from left main coronary artery. Cathet Cardiovasc Diagn. 1996;38(4):402-404.
8. 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.
9. Webb JG, Solankhi N, Carere RG. Facilitation of stent retention and retrieval with an emboli containment device. Catheter Cardiovasc Interv. 2000;50(2):215-217.
10. Ngow H. Stent dislodgement in left anterior descending coronary artery and successful deployment with two parallel wires and a small balloon. Kardiol Pol. 2009;67:1040-1042; discussion p. 1043.
11. Bruschi G, Colombo P, Piccalo G, Trunfio S, Colombo T, Vitali E. Successful emergent surgical revascularization and retrieval of entrapped drug eluting stent. J Cardiovasc Med (Hagerstown). 2008;9(2):182-183.
12. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177-188.
13. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis of observational studies in epidemiology (MOOSE) group. JAMA. 2000;283(15):2008-2012.
14. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol. 2009;62(10):e1-e34.
15. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of reporting of meta-analyses. Lancet. 1999;354(9193):1896-1900.
16. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535.
17. Meredith IT, Whitbourn R, Scott D, et al. PLATINUM QCA: a prospective, multicentre study assessing clinical, angiographic, and intravascular ultrasound outcomes with the novel platinum chromium thin-strut PROMUS Element everolimus-eluting stent in de novo coronary stenoses. EuroIntervention. 2011;7(1):84-90.
18. Eggebrecht H, Haude M, von Birgelen C, et al. Nonsurgical retrieval of embolized coronary stents. Catheter Cardiovasc Interv. 2000;51(4):432-440.
19. 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.
20. 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.
21. 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.
22. Foster-Smith KW, Garratt KN, Higano ST, Holmes DR Jr. Retrieval techniques for managing flexible intracoronary stent misplacement. Cathet Cardiovasc Diagn. 1993;30:63-68.
23. Khattab AA, Geist V, Toelg R, Richardt G. The AngioGuard: a simplified snare? Int J Cardiovasc Intervent. 2004;6(3-4):153-155.
24. Wong PH. Retrieval of undeployed intracoronary Palmaz-Schatz stents. Cathet Cardiovasc Diagn. 1995;35(3):218-223.
25. 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.
26. Brilakis ES, Abdel-Karim AR, Banerjee S. Hairpin-trap: a novel stent retrieval technique. Catheter Cardiovasc Interv 2011;77(2):213-216.
27. Rahman A, Chaudhury H, Ali M. Embolized stent in the coronary circulation “without riding the wire” and its novel management strategy. J Invasive Cardiol. 2009;21(10):E201-E203.
28. Colkesen AY, Baltali M, Acil T, et al. Coronary and systemic stent embolization during percutaneous coronary interventions: a single center experience. Int Heart J. 2007;48(2):129-136.
29. Iturbe JM, Abdel-Karim AR, Papayannis A, et al. Frequency, treatment, and consequences of device loss and entrapment in contemporary percutaneous coronary interventions. J Invasive Cardiol. 2012;24(5):215-221.