Incidence, Management, and Clinical Outcomes of Procedure-Related Coronary Artery Perforation: Analysis of 13,888 Coronary Angioplasty Procedures

Original Article Acta Cardiol Sin 2008;24:80-5

Coronary Heart Disease Incidence, Management, and Clinical Outcomes of Procedure-Related Coronary Artery Perforation: Analysis of 13,888 Coronary Angioplasty Procedures

Su-Kiat Chua, Shih-Huang Lee, Kou-Gi Shyu, Huei-Fong Hung, Sheng-Chang Lin and Jun-Jack Cheng

Background: Coronary artery perforation is a rare but life-threatening complication of percutaneous coronary intervention. We report our experience of incidence, management, and clinical outcomes of procedure-related coronary artery perforation in 13,888 consecutive patients. Methods and Results: 13,888 patients underwent PCI for coronary artery disease from October 1992 to December 2006. During this period, 21 (0.15%) patients developed coronary artery perforation during PCI. Four patients with Ellis type I coronary artery perforation were treated successfully, including conservative treatment in 2 and device therapy in 2. Out of eight patients with Ellis type II coronary artery perforation, 2 patients received conservative treatment, 5 received device therapy and 1 received surgical ligation. One patient receiving device therapy had late cardiac tamponade and she expired due to aspiration pneumonia resulting from emergent endotracheal intubation. Among the seven patients with Ellis type III coronary artery perforation, 5 received device therapy and 2 received emergent surgical repairs. Additionally, four of them needed emergent pericardiocentesis for immediate cardiac tamponade. One of the 2 patients with Ellis type IV coronary artery perforation was treated with balloon inflation, but he expired due to concomitant pneumonia. The other patient received surgical repair. Conclusion: Management of coronary artery perforation can be tailored according to the classification of coronary artery perforation and the hemodynamic status. Most patients can be treated successfully with non-surgical therapies. However, surgical intervention should be provided promptly if non-surgical therapies fail. Furthermore, late cardiac tamponade might occur even in less advanced types of coronary artery perforation.

Key Words: Coronary artery perforation · Percutaneous coronary intervention · Stent · Surgical intervention

INTRODUCTION vention (PCI), and its incidence is between 0.2% and 0.8%.1-6 Procedure-related coronary artery perforation Coronary artery perforation is a rare but life-threat- may result in the development of cardiac tamponade, ening complication of percutaneous coronary inter- acute myocardial infarction (AMI) and mortality. The non-surgical managements of coronary artery perforation include reversal of heparin with protamine sulphate, Received: June 20, 2007 Accepted: January 30, 2008 prolonged balloon inflation, perfusion balloon inflation, Division of Cardiology, Shin Kong Wu Ho-Su Memorial Hospital and School of Medicine, Fu-Jen Catholic University, Taipei, Taiwan. conventional or covered stent implantation and coil or 2,3 Address correspondence and reprint requests to: Dr. Shih-Huang Lee, thrombus embolization. We report our experience of Division of Cardiology, Shin Kong Wu Ho-Su Memorial Hospital, incidence, managements and clinical outcomes of proce- No. 95, Wen Chang Rd., Shih-Lin District, Taipei, Taiwan, Tel: 886-2-2833-2211 ext. 2091; Fax: 886-2-2834-8910; E-mail: shlee88@ dure-related coronary artery perforation in 13,888 con- ms78.hinet.net secutive patients.

Acta Cardiol Sin 2008;24:80-5 80 Procedure-Related Coronary Artery Perforation

METHODS vasation; type II, pericardial or myocardial blush without contrast jet extravasation; type III, continuous jet-like From October 1992 to December 2006, 13,888 pa- dye extravasation resulting in hypotension and tampo- tients underwent PCI for coronary artery disease (CAD) nade; and Type IV, extravasation of contrast dye into an in this institution. All patients received the percutaneous anatomic cavity chamber.2,9 Successful management of transfemoral or transradial approach via an angiography coronary artery perforation was defined as no more or sheath, and standard angioplasty technique was used in minimal dye extravasation and/or no further clinical these patients.7,8 Each patient was pretreated with intra- hemodynamic instability. Late cardiac tamponade was venous heparin (100 units/kg) at the beginning of the defined as cardiac tamponade developed within 24 hours procedure. If necessary, an additional bolus of heparin after the patient left the cardiac catheterization room or was administered to maintain activated clotting time > operating room.2 300 sec. Angiographic stenosis was defined as a diame- ter stenosis of ³ 50%. The stenotic morphology of coronary arteries was based on the American College of Car- RESULTS diology/American Heart Association (ACC/AHA) lesion classification system.2,9 Procedure-related coronary ar- Patient characteristics tery perforation was classified according to the Ellis Totally 21 patients had procedure-related coronary classification: Type 1, extraluminal crater without extra- artery perforation, an incidence of 0.15 %. Table 1 shows

Table 1. Clinical and angiographic characteristics, perforation classification, management and outcomes of study patients Age/ Underlying Procedural Antiplatelet/ Perforation ACC/AHA Device Ellis Initial Additional No CAD CTO Outcome sex disease indication anticoagulant site class responsible type treatment treatment 1 70/F DM AP A, H SVD RCA-D + B2 GW I Conserv - - 2 46/MHT/DM/LI AP A,H TVDRCA-P + C GW I Conserv - - 3 66/F - AP A, H TVD LAD-P + C Bl I Bl - - 4 72/M HT/LI AP A, H TVD LAD-P - B1 Bl I CS - - 551/M - AP A,H TVDLAD-Mi + B1 GW IIConserv - - 675/M - AP A,H TVDLAD-Mi + B2 Bl II Conserv - AMI 7 64/F HT/LI AP A, H SVD LAD-P + C Bl II Pr, Bl - - 8 72/F HT/LI AP A, H DVD LCX-Mi - B2 Bl II Pr, PB - - 9 HT PB LC; 82/F AMI A, Pl, H TVD LAD-P + B1 Bl II CS Death 10 74/M HT/DM AP A, H TVD LAD-Mi - B1 Bl II CS - - 11 72/M HT AMI A, Pl, H DVD LAD-Mi - C Bl II CoS - - 12 78/F HT AP A, H TVD LAD-P + B2 GW II Bl SL - 13 73/M HT AP A, H TVD LAD-D + B2 Bl III Pr, Bl - LC 14 65/F HT/LI AP A, H DVD LAD-Mi - A GW III Pr, Bl - - 15 66/M - AP A, H TVD LCX-Mi + B2 Bl III PC, Bl - - 1667/M DM AP A,H SVDLAD-Mi - C Stent IIIPC,PB - - 17 70/M HT AP A, H DVD LAD-D - B2 Bl III Bl PB AMI 1858/F LI AP A,H TVDLAD-Mi - B2 Stent III PC,Bl SR, - CABG 1961/M HT AP A,H SVDLAD-P + B1 Bl IIIPC,Bl SR - 20 78/M HT/LI AMI A, Pl, H TVD LCX-Mi + B2 Bl IV Bl - Death 21 70/F HT/DM AP A, H DVD LAD-Mi - C Stent IV CoS SR - A = aspirin; ACC/AHA class = American College of Cardiology/American Heart Association lesion classification; AMI = acute myocardial infarction; AP = angina pectoris; Bl = balloon inflation; CABG = coronary artery bypass graft; Conserv = conservative treatment; CS = conventional stent; CoS = covered stent; CTO = chronic total occlusion; D = distal; DM = diabetes mellitus; DVD = double-vessel disease; F = female; GW = guide wire; H = heparin; HT = hypertension; LAD = left anterior descending artery; LC = late cardiac tamponade; LCX = left circumflex artery; LI = dyslipidemia; M = male; Mi = middle; P = proximal; Pl = Plavix; Pr = protamine infusion; PB = perfusion balloon; PC = pericardiocentesis; RCA = right coronary artery; SL = surgical ligation; SR = surgical repair; SVD = single-vessel disease; TVD = triple-vessel disease.

81 Acta Cardiol Sin 2008;24:80-5 Su-Kiat Chua et al. clinical characteristics of the 21 patients with proce- Out of the seven patients with Ellis type III coronary dure-related coronary artery perforation. The mean age artery perforation, 4 patients received emergent peri- of the patients was 68 ± 8 years (range 46 to 82 years). cardiocentesis for immediate cardiac tamponade. Three Fourteen patients (67%) had hypertension, 7 (33%) had patients were treated with prolonged balloon inflation dyslipidemia, 5 (24%) had diabetes mellitus. Eighteen and 2 received perfusion balloon inflation. The remain- (86%) patients received PCI under the indication of an- ing two patients received emergent surgical repairs be- gina pectoris, and the remaining 3 (14%) patients had cause they did not respond to device therapy (patients AMI. Seventeen (81%) patients had more than one dis- no. 18 and 19). Patient no. 17 had postprocedural non- eased vessel; 24% were double-vessel disease and 57% ST elevation AMI. Furthermore, patient no. 13 had late were triple-vessel disease. The coronary artery perfora- cardiac tamponade within 24 hours, and he was treated tion occurred at the left anterior descending artery in 16 successfully with pericardiocentesis alone. (76%) patients, left circumflex artery in 3 (14%) and One of the 2 patients with Ellis type IV coronary ar- right coronary artery in 2 (10%). Coronary artery perfo- tery perforation was treated successfully with balloon in- ration developed during PCI for chronic total occlusion flation (patient no. 20); however, he expired due to his in 12 (57%) patients. According to the ACC/AHA lesion concomitant pneumonia at 17 days after the event of classification system, 1 (5%) patient had class A, 14 coronary artery perforation. The other patient received (67%) had class B and 6 (29%) had class C lesion mor- surgical repair because she did not respond to covered phology. Coronary artery perforation occurred in 5 (24%) patients during guide-wire manipulation, in 13 (62%) during coronary balloon angioplasty and in 3 (14%) during coronary stenting. The perforation classification of Ellis type was type I in 4 (19%), type II in 8 (38%), type III in 7 (33%) and type IV in 2 (10%) patients.

Managements and clinical outcomes of coronary artery perforation Four patients with Ellis type I coronary artery perforation were treated successfully, including conservative treatment in 2, balloon inflation in 1 and conventional stent implantation in 1. Pericardiocentesis and surgical intervention were not needed in these 4 patients. Among the eight patients with Ellis type II coronary artery perforation, 2 patients received conservative treatment, 3 were managed with prolonged balloon inflation and 2 were treated with stent implantation, including conventional stent in 1 and polytetrafluoroethylene- covered stents in 1 (Patient no. 11, Figure 1). The remaining one patient (no. 12) received surgical ligation because the coronary artery perforation could not heal after 30-minute balloon inflation. Patient no. 9 with

AMI, however, who was treated with conventional stent- Figure 1. Panel A. Coronary angiography showed significant stenosis ing followed by perfusion balloon inflation, had late car- (white arrow) of the left anterior descending coronary artery in a diac tamponade. She expired 20 days later due to aspira- 72-year-old man. Panel B. Balloon angioplasty at the site of stenosis. tion pneumonia resulting from emergent endotracheal Panel C. Coronary artery perforation developed after balloon angioplasty. Panel D. The patients underwent polytetrafluoroethylene-covered intubation. Patient no. 6 had postprocedural non-ST ele- stent implantation, and later angiography showed no extravasations of vation AMI. contrast dye.

Acta Cardiol Sin 2008;24:80-5 82 Procedure-Related Coronary Artery Perforation stent implantation. tion was successfully managed in only 2 of their 5 pa- The clinical events after procedure-related coronary tients with Ellis type II or III coronary artery perfora- artery perforation are summarized in Table 2. Cardiac tion.13 In the present study, out of the 3 patients treated tamponade occurred in 6 (29%) patients with coronary with conventional stent implantation, 2 were successes. artery perforation. While 4 patients with immediate car- The remaining patient (no. 9) failed and received perfu- diac tamponade had Ellis type III coronary artery perfo- sion balloon inflation as additional treatment. However, ration and needed emergent pericardiocentesis, the re- late cardiac tamponade occurred within 24 hours in this maining two patients with Ellis type II and III, respec- patient. Despite perfusion balloon inflation, conven- tively, had late cardiac tamponade within 24 hours. Post- tional stent implantation still carries with it a risk of late procedural AMI developed in 1 patient with Ellis type II cardiac tamponade. In recent years, covered stents have and 1 with Ellis type III coronary artery perforation. become an alternative treatment when conservative ap- Three of the 4 patients receiving emergent surgery had proaches fail. Similar to perfusion balloons, covered Ellis type III/IV coronary artery perforation. Two pa- stents must be implanted with caution in regions of sig- tients with Ellis type II and IV coronary artery perfora- nificant side branching. tion, respectively, had in-hospital mortality. Previous studies have shown that coil or thrombus embolization could treat coronary artery perforation.4,9,19,20 Embolization can be used if the bleeding site is in a DISCUSSION small distal branch or in the setting of chronic coronary artery occlusion.9,18 Coronary artery perforation during PCI is infre- Previous reports showed that some patients needed quent, but this complication is associated with high mor- to receive surgery when conservative treatment and de- bidity and mortality. The incidence of coronary artery vice therapy could not correct hemodynamic instability.1 perforation during PCI in our study was 0.15%. Previous Kihara et al. also reported that coronary artery perfora- studies showed that coronary artery perforation increased tion could be managed successfully by emergent surgical significantly in the patients receiving laser ablation, rota- intervention without cardiopulmonary bypass.15 In the tional and extraction atherectomies compared with those present study, 4 patients receiving emergent surgery in- receiving balloon angioplasty alone.2,3,10-12 Coronary ar- tervention had immediate excellent outcomes. tery perforation most commonly occurred in patients with coronary balloon angioplasty in the present study. Coronary artery perforation type-specific therapy Therapeutic options for procedure-related The options for management may be chosen accord- coronary artery perforation ing to the severity of perforation and the hemodynamic Previous studies showed that patients with coronary status of the patients. Previous studies and the present artery perforation could be managed successfully with study have shown that patients with Ellis type I and II conventional stent or covered stent implantation.13,14 coronary artery perforation rarely developed adverse Briguori et al. reported that conventional stent implanta- cardiac events such as cardiac tamponade and severe

Table 2. Clinical events following coronary artery perforation Coronary artery perforation Ellis type Overall (n = 21) Type I (n = 4) Type II (n = 8) Type III (n = 7) Type IV (n = 2) Emergent pericardiocentesis 0 (0) 0 (0)0 4(57)0(0)4(19) Late cardiac tamponade 0 (0) 1 (13) 1 (14) 0 (0) 2 (10) Postprocedural AMI 0 (0) 1 (13) 1 (14) 0 (0) 2 (10) Emergent surgery 0 (0) 1 (13) 2 (29) 01(50) 4(19) In-hospital death 0 (0) 1 (13) 0 (0)001(50) 2(10) Values presented as number (%). AMI = acute myocardial infarction.

83 Acta Cardiol Sin 2008;24:80-5 Su-Kiat Chua et al. ischemia and could be treated successfully with conser- monia. The remaining patient (no. 9) developed late car- vative treatment and/or device therapy in most cases.2-4 diac tamponade after device therapy. She expired due to Patients with Ellis type III coronary artery perforation development of aspiration pneumonia resulting from usually have severe cardiac complications such as car- emergent endotracheal intubation during late cardiac diac tamponade, and this complication may result in the tamponade. Patient no. 9 had received conventional need for urgent surgical intervention and death.2-4 Fas- stenting, followed by perfusion balloon inflation as addi- seas et al. and the present study have reported that all pa- tional treatment. However, she suffered from late cardiac tients receiving emergent pericardiocentesis had Ellis tamponade. It is possible that more aggressive use of type III coronary artery perforation.3 About half of the antiplatelet agents and anticoagulants in AMI patients patients with Ellis type III coronary artery perforation in have complicated the management of coronary artery the present study needed emergent pericardiocentesis. perforation. Prompt reversal of anticoagulation with suf- Immediate balloon reinflation can minimize pericardial ficient amount of intravenous protamine may be helpful effusion, and bedside echocardiography may facilitate in these patients. In addition, other alternative options pericardiocentesis if hemodynamic status allows. In ad- for device therapy should be considered. dition, prompt reversal of anticoagulation with intravenous protamine is valuable. The other managements in- Study limitations clude perfusion balloon, conventional or covered stent The number of the patients in the present study was implantation, embolization and emergent surgical repair. small; we could not determine if the usage of antiplatelet Patients with Ellis type IV coronary artery perforation and/or anticoagulation therapies increased the risk of had more favorable presentation compared to those with coronary artery perforation. Secondly, patients were Ellis type III. Previous reports had shown that the vascu- treated according to the experiences and judgments of lar communication might even close spontaneously dur- their interventionists, so the treatment algorithms might ing follow-up.17 However, the present study showed that be inconsistent.16,17 monitoring, intensive care and option of surgical repair are necessary for some patients with Ellis type IV coronary artery perforation. CONCLUSION

Clinical outcome after initial management of Management of coronary artery perforation can be coronary artery perforation tailored according to the classification of the coronary Ellis et al. reported that two (13%) of 15 instances of artery perforation and the hemodynamic status. Most pa- cardiac tamponade in their series were delayed.2 In the tients with coronary artery perforation can be treated present study, two (33%) of 6 patients with cardiac tam- successfully with non-surgical management; however, ponade, including Ellis type II and III coronary artery surgical intervention should be provided promptly with- perforation respectively, had late cardiac tamponade out delay if coronary artery perforation does not respond within 24 hours after dismissal from the cardiac cathe- to the non-surgical therapies. Furthermore, patients terization room. Thus, the interventionist should be should be monitored carefully because late cardiac tam- aware of this information after the initial management of ponade and postprocedural AMI might occur even in coronary artery perforation. Patients should be moni- less advanced types of coronary artery perforation. tored carefully for at least 24 hours since even less advanced types of coronary artery perforation may have late cardiac tamponade. REFERENCES Among the three AMI patients in this study, coro- 1. Gunning MG, Williams IL, Jewitt DE, et al. Coronary artery per- nary artery perforation was treated successfully in 2 pa- foration during percutaneous intervention: incidence and out- tients, including covered stenting in patient no. 11 and come. Heart 2002;88:495-8. prolonged balloon inflation in patient no. 20. However, 2. Ellis SG, Ajluni S, Arnold AZ, et al. Increased coronary perfora- patient no. 20 expired because of his concomintant pneu- tion in the new device era. Incidence, classification, management,

Acta Cardiol Sin 2008;24:80-5 84 Procedure-Related Coronary Artery Perforation

and outcome. Circulation 1994;90:2725-30. 12. Woodfield SL, Lopez A, Heuser RR, et al. Fracture of coronary 3. Fasseas P, Orford JL, Panetta CJ, et al. Incidence, correlates, guidewire during rotational atherectomy with coronary perfora- management, and clinical outcome of coronary perforation: an- tion and tamponade. Cathet Cardiovasc Diagn 1998;44:220-3. alysis of 16,298 procedures. Am Heart J 2004;147:140-5. 13. Briguori C, Nishida T, Anzuini A, et al. Emergency polytetra- 4. Javaid A, Buch AN, Satler LF, et al. Management and outcomes fluoroethylene-covered stent implantation to treat coronary rup- of coronary artery perforation during percutaneous coronary in- tures. Circulation 2000;102:3028-31. tervention. Am J Cardiol 2006;98:911-4. 14. Subraya RG, Tannenbaum AK. Successful sealing of perforation 5. Fukutomi T, Suzuki T, Popma JJ, et al. Early and late clinical out- of saphenous vein graft by coronary stent. Cathet Cardiovasc comes following coronary perforation in patients undergoing per- Intervent 2000;50:460-2. cutaneous coronary intervention. Circ J 2002;66:349-56. 15. Shimakura KS, Tanaka T, et al. Surgical angioplasty of ruptured 6. Ajluni SC, Glazier S, Blankenship L, et al. Perforations after per- left anterior descending coronary artery without cardiopulmonary cutaneous coronary interventions: clinical, angiographic, and bypass. Jap J Thorac Cardiovasc Surg 2000;48:326-8. therapeutic observations. Cathet Cardiovasc Diagn 1994;32: 16. Satler LF. A revised algorithm for coronary perforation. Cathet 206-12. Cardiovasc Intervent 2002;57:215-6. 7. Lee RJ, Lee SH, Shyu KG, et al. Immediate and long-term out- 17. Dippel EJ, Kereiakes DJ, Tramuta DA, et al. Coronary perfora- comes of stent implantation for unprotected left main coronary ar- tion during percutaneous coronary intervention in the era of ab- tery disease. Int J Cardiol 2001;80:173-7 ciximab platelet glycoprotein IIb/IIIa blockade: an algorithm for 8. Lee RJ, Shih KN, Lee SH, et al. Predictors of long-term outcomes percutaneous management. Cathet Cardiovasc Intervent 2001; in patients after elective stent implantation for unprotected left 52:279-86. main coronary artery disease. Heart Vessels 2007;22(2):99-103 18. Mahmud E, Douglas JS Jr. Coil embolization for successful treat- 9. Baim DS. Grossman’s Cardiac Catheterization, Angiography, and ment of perforation of chronically occluded proximal coronary Intervention. 7th ed. Philadelphia: Lippincott, Williams & Wil- artery. Cathet Cardiovasc Intervent 2001;53(4):549-52. kins, 2006:50-1,446-513. 19. Assali AR, Moustapha A, Sdringola S, et al. Successful treatment 10. Holmes DR, Reeder FS, Ghazzal ZM, et al. Coronary perforation of coronary artery perforation in an abciximab-treated patient by after excimer laser coronary angioplasty: The Excimer Laser microcoil embolization. Cathet Cardiovasc Intervent 2000; Coronary Angioplasty Registry experience. J Am Coll Cardiol 51(4):487-9. 1994;23:330-5. 20. Fischell TA, Carter AJ, Ashraf K, et al. Coronary artery rupture 11. Cohen BM, Weber VJ, Reisman M, et al. Coronary perforation during balloon angioplasty, rescued with localized thrombin in- complicating rotational ablation: the U.S. multicenter experience. jection and coil embolization. Cathet Cardiovasc Intervent 2006; Cathet Cardiovasc Diagn 1996;3(Suppl):55-9. 68(2):254-7.

85 Acta Cardiol Sin 2008;24:80-5