2020 Volume 84 Issue 10 Pages 1746-1753
Background: Guide catheter-induced iatrogenic coronary artery dissection is a rare but feared complication. When it occurs, bailout stenting is widely performed; however, its prognosis and the impact of stent type remains unclear.
Methods and Results: The study population consisted of 77,257 consecutive patients (coronary angiography, 55,864; percutaneous coronary intervention, 21,393) between 2000 and 2015. We investigated the incidence, clinical outcomes, and angiographic results after bailout stenting and compared by stent type: bare-metal stent (BMS) and drug-eluting stent (DES). Iatrogenic coronary artery dissection occurred in 105 patients (incidence rate, 0.14%). All cases of iatrogenic coronary artery dissection that were recognized as requiring bailout procedure could be managed by stent implantation, and no patients died during bailout procedure. The 5-year cumulative incidences of cardiac death, target lesion revascularization, and major adverse cardiac events were 11.3%, 10.3%, and 21.0%, respectively. The binary restenosis rate was 10.4%, and it was not significantly different between BMS and DES implantation. In lesions with preprocedural stenosis, however, it was significantly lower in the DES group than in the BMS group. On the other hand, coronary artery dissection recurred in 8 patients, which was observed only after DES implantation.
Conclusions: The immediate and long-term outcomes of bailout stenting for iatrogenic coronary artery dissection were acceptable. Although DES may be favorable for stenotic lesions, coronary artery dissection can recur after DES implantation.
Diagnostic or guiding catheter-induced iatrogenic coronary artery dissection during cardiac catheterization is a rare but potentially devastating complication, with an associated risk of myocardial infarction (MI) and subsequent death.1 Iatrogenic left main coronary artery (LMCA) dissection has been reported to occur in <0.1% of cardiac catheterization procedures.2–4 Although emergency coronary artery bypass grafting (CABG) is a therapeutic option, prolonged periods of ischemia before the operation often lead to myocardial damage and death.5 In 1993, Garcia-Robles et al reported that prompt coronary stent implantation is an effective bail out procedure for coronary artery dissection,6 and since then, it has been the most widely used treatment option. However, the low incidence of iatrogenic coronary artery dissection means reports documenting bailout stenting are scarce and outcomes remain unclear.2–4,6–8 Currently, drug-eluting stents (DES) are the main stent type used in clinical practice because they reduce the occurrence of in-stent restenosis and target lesion revascularization (TLR) compared with bare-metal stent (BMS).9 Recent data, however, have raised concern that DES may be associated with very late stent thrombosis and late catch-up phenomenon,10 and their superiority to BMS remains unclear. In the current study, we aimed to describe the initial and long-term outcomes of bailout stenting for iatrogenic coronary artery dissection and the differences between BMS and DES.
This was a retrospective single-center study. The study population consisted of 77,257 patients who underwent either coronary angiography (CAG; 55,864) or percutaneous coronary intervention (PCI; 21,393) at Kurashiki Central Hospital between January 2000 and December 2015. Among them, iatrogenic coronary artery dissection caused by a diagnostic or guiding catheter occurred in 105 patients (during CAG, 11; during PCI, 94). In 12 patients, the dissection was small and did not worsen with time, so a bailout procedure was not performed. Therefore, the final study sample consisted of 93 patients undergoing bailout stenting using either BMS or DES. We investigated their immediate outcomes, 5-year clinical outcomes and angiographic follow-up results. We also compared them for BMS and DES implantation. The study flow is shown in Figure 1. The decision to perform bailout stenting and the choice of stent type were at the discretion of the operator. In Japan, BMS were available through the study period, but DES became available in August 2004. The types of DES used in this study were Cypher (sirolimus-eluting stent; Cordis, Johnson & Johnson, Miami, FL, USA), Taxus (paclitaxel-eluting stent; Boston Scientific, Natick, MA, USA), Endeavor (zotarolimus-eluting stent; Medtronic, Santa Rosa, CA, USA), Xience (everolimus-eluting stent; Abbott Vascular, Santa Clara, CA, USA), Nobori (biolimus-eluting stent; Terumo, Tokyo, Japan), Promus Element (everolimus-eluting stent; Boston Scientific), Resolute Integrity (zotarolimus-eluting stent; Medtronic), and Ultimaster (sirolimus-eluting stent; Terumo). The study was conducted in accordance with the provisions of the Declaration of Helsinki and the guidelines for epidemiological studies issued by the Ministry of Health, Labour and Welfare of Japan. All patients provided informed consent for the procedure and subsequent data collection.
Study flow chart. CAG, coronary angiography; PCI, percutaneous coronary intervention.
Clinical information was obtained by review of the medical records or by telephone contact with the patients, family members, or primary care physicians. Angiographic follow-up was routinely scheduled at 6–8 months and 18–20 months after the procedure. The follow-up angiogram was obtained earlier if clinically indicated. Follow-up angiography performed within 12 months was considered early follow-up, and that performed between 12 and 24 months was considered late follow-up. We investigated the mortality rate associated with iatrogenic coronary artery dissection, use of mechanical circulatory support such as intra-aortic balloon pumping (IABP) or venoarterial extracorporeal membrane oxygenation (VA-ECMO), and periprocedural myocardial injury as immediate outcomes. Periprocedural myocardial injury was defined as elevated creatine kinase-myocardial band (CK-MB) levels 3-fold greater than the upper reference limit (URL). Clinical outcomes were evaluated as cardiac death, TLR, spontaneous MI, definite or probable stent thrombosis and major adverse cardiac events (MACE). MACE consisted of cardiac death, nonfatal spontaneous MI, and TLR. TLR was defined as either repeated PCI or CABG for restenosis or thrombosis of the target lesion within the stent or within 5 mm proximal or distal to the stent on CAG. MI was defined as elevated CK-MB levels 3-fold greater than the URL and associated with clinical signs or symptoms of ischemia. Definite or probable stent thrombosis was defined according to the Academic Research Consortium definition.11 Binary restenosis at follow-up was defined as a stenosis occupying >50% of the diameter. Late restenosis was defined as diameter stenosis ≥50% at late follow-up in lesions with <50% diameter stenosis at early follow-up.
Statistical AnalysisCategorical variables were compared using the chi-square test or Fisher’s exact test. Continuous variables are expressed as mean (standard deviation) and were compared using Student’s t-test or the Mann-Whitney U test based on the distributions. The cumulative incidence was estimated using the Kaplan-Meier method, and differences were assessed using the log-rank test. All P values <0.05 were considered significant. Statistical analyses were performed using SPSS version 23 (International Business Machines, Armonk, NY, USA).
The incidence rate of iatrogenic coronary artery dissection was 0.14% and significantly higher during PCI than during CAG (0.44% vs. 0.02%; P<0.001). Table 1 shows the patient and procedural characteristics with iatrogenic coronary artery dissection. Dissection Types A and B by the National Heart, Lung, and Blood Institute classification,12 which are generally clinically benign, accounted for 41.0%, Types C through F, which portend significant morbidity and mortality if untreated, accounted for 59.0%, and Type F, which represents total occlusion of the coronary lumen without distal antegrade flow, accounted for 17.1%. Table 2 shows the patient and procedural characteristics with bailout stenting. BMS were more frequently used for ST-segment elevation acute coronary syndrome (STE-ACS) patients than DES, whereas DES was more frequently used for LMCA and preprocedural stenotic lesion than BMS. Although the stent size was significantly smaller in the DES group than in the BMS group, there was no significant difference in the use rate of 3.5-mm or larger stents. Intravascular ultrasound was more frequently used in the DES group than in the BMS group.
| n=105 | |
|---|---|
| Age, years | 69.2±10.5 |
| Male | 75 (71.4) |
| Hypertension | 75 (71.4) |
| Diabetes mellitus | 37 (35.2) |
| Insulin-treated diabetes | 11 (10.5) |
| Dyslipidemia | 61 (58.1) |
| Hemodialysis | 6 (5.7) |
| Previous MI | 42 (40.0) |
| Clinical diagnosis | |
| Stable angina | 78 (74.3) |
| NSTE-ACS | 19 (18.1) |
| STE-ACS | 8 (7.6) |
| Type of CAD | |
| None | 2 (1.9) |
| 1-vessel disease | 45 (42.9) |
| 2-vessel disease | 36 (34.3) |
| 3-vessel disease | 22 (21.0) |
| LMCA | 12 (11.4) |
| Procedure type | |
| Diagnostic CAG | 11 (10.5) |
| PCI | 94 (89.5) |
| Vascular access | |
| Radial or brachial | 58 (55.2) |
| Femoral | 47 (44.8) |
| Dissection site | |
| RCA | 68 (64.8) |
| LMCA | 37 (35.2) |
| Culprit catheter (guide/diagnostic) | |
| RCA (n=68) | |
| Amplatz left | 37 (54.4) |
| Judkins right | 22 (32.4) |
| KIMNY Mini | 3 (4.4) |
| Mitsudo | 6 (8.8) |
| LMCA (n=37) | |
| Amplatz left | 3 (8.1) |
| Judkins left | 26 (70.3) |
| Extra backup | 3 (8.1) |
| KIMNY Mini | 1 (2.7) |
| Mitsudo | 4 (10.8) |
| Culprit catheter size | |
| 5F | 11 (10.5) |
| 6F | 25 (23.8) |
| 7F | 68 (64.8) |
| 8F | 1 (1.0) |
| Dissection type (NHLBI classification) | |
| A | 12 (11.4) |
| B | 31 (29.5) |
| C | 23 (21.9) |
| D | 16 (15.2) |
| E | 5 (4.8) |
| F | 18 (17.1) |
Data are expressed as number (%) or mean (standard deviation). CAD, coronary artery disease; CAG, coronary angiography; LMCA, left main coronary artery; MI, myocardial infarction; NHLBI, National Heart, Lung, and Blood Institute; NSTE-ACS, non-ST-segment elevation acute coronary syndrome; PCI, percutaneous coronary intervention; RCA, right coronary artery; STE-ACS, ST-segment elevation acute coronary syndrome.
| Overall (n=93) |
BMS (n=41) |
DES (n=52) |
P value | |
|---|---|---|---|---|
| Age, years | 69.5±10.4 | 69.2±10.8 | 69.7±10.2 | 0.80 |
| Male | 67 (72.0) | 31 (75.6) | 36 (69.2) | 0.64 |
| Hypertension | 64 (68.8) | 25 (61.0) | 39 (75.0) | 0.18 |
| Diabetes mellitus | 35 (37.6) | 11 (26.8) | 24 (46.2) | 0.08 |
| Insulin-treated diabetes | 11 (11.8) | 4 (9.8) | 7 (13.5) | 0.75 |
| Dyslipidemia | 52 (55.9) | 16 (39.0) | 36 (69.2) | 0.01 |
| Hemodialysis | 5 (5.4) | 1 (2.4) | 4 (7.7) | 0.38 |
| Previous MI | 35 (37.6) | 13 (31.7) | 22 (42.3) | 0.39 |
| Clinical diagnosis | 0.03 | |||
| Stable angina | 67 (72.0) | 26 (63.4) | 41 (78.8) | |
| NSTE-ACS | 18 (19.4) | 8 (19.5) | 10 (19.2) | |
| STE-ACS | 8 (8.6) | 7 (17.1) | 1 (1.9) | |
| Type of CAD | 0.16 | |||
| None | 2 (2.2) | 2 (4.9) | 0 (0) | |
| 1-vessel disease | 38 (40.9) | 20 (48.8) | 18 (34.6) | |
| 2-vessel disease | 32 (34.4) | 12 (29.3) | 20 (38.5) | |
| 3-vessel disease | 21 (22.6) | 7 (17.1) | 14 (23.1) | |
| LMCA | 12 (12.9) | 6 (14.6) | 6 (11.5) | 0.76 |
| Procedure type | 0.70 | |||
| Diagnostic CAG | 7 (7.5) | 4 (9.8) | 3 (5.8) | |
| PCI | 86 (92.5) | 37 (90.2) | 49 (94.2) | |
| Dissection site | 0.03 | |||
| RCA | 61 (65.6) | 32 (78.0) | 29 (55.8) | |
| LMCA | 32 (34.4) | 9 (22.0) | 23 (44.2) | |
| Dissection type (NHLBI classification) | 0.29 | |||
| A and B | 35 (37.6) | 18 (43.9) | 17 (32.7) | |
| C, D, E, and F | 58 (62.4) | 23 (56.1) | 35 (67.3) | |
| Stenosis (%DS ≥50%) at dissection site | 28 (30.1) | 7 (17.1) | 21 (40.4) | 0.02 |
| Stent size, mm | 3.48±0.35 | 3.64±0.39 | 3.35±0.25 | 0.01 |
| Stent size ≥3.5 mm | 70 (75.3) | 33 (80.5) | 37 (71.2) | 0.34 |
| Stent length, mm | 19.9±6.6 | 17.2±6.1 | 22.0±6.2 | 0.32 |
| IVUS use | 54 (58.1) | 13 (31.7) | 41 (78.8) | 0.01 |
| DES type | ||||
| Cypher stent | – | – | 12 (23.1) | |
| Taxus stent | – | – | 2 (3.8) | |
| Endeavor stent | – | – | 2 (3.8) | |
| Xience stent | – | – | 2 (3.8) | |
| Nobori stent | – | – | 31 (59.6) | |
| Promus Element stent | – | – | 1 (1.9) | |
| Resolute Integrity stent | – | – | 1 (1.9) | |
| Ultimaster stent | – | – | 1 (1.9) | |
Data are expressed as number (%) or mean±standard deviation. BMS, bare metal stent; DES, drug-eluting stent; DS, diameter stenosis; IVUS, intravascular ultrasound. Other abbreviations as in Table 1.
All cases of iatrogenic coronary artery dissection that were recognized as requiring bailout procedure could be managed by stent implantation, and no patients died during bailout procedure. Although 2 patients manifested as hemodynamic collapse after the occurrence of Type F dissection of the LMCA and required mechanical circulatory support by IABP and VA-ECMO, they were weaned from them after intensive therapy. Periprocedural myocardial injury occurred in 16 (20.5%) of 78 stable angina patients, all of whom had a Type C through F dissection. CK-MB levels 10-fold greater than the URL occurred in 5 patients (6.4%), all of whom had a Type F dissection. In 15 patients with a Type F dissection that occurred during elective procedures, the CK-MB level was 3-fold greater than the URL in 10 patients (66.7%) and 10-fold greater than the URL in 5 patients (33.3%).
Long-Term Clinical OutcomesThe median follow-up period was 51 months (interquartile range [Q1–Q3], 24–80 months). Figure 2 and Figure 3 show the clinical outcomes. The 5-year cumulative incidences of cardiac death, TLR and MACE were 11.3%, 10.3%, and 21.0%, respectively. TLR beyond 1 year was performed in only 2 patients. There were no significant differences in the clinical outcomes between the BMS and DES groups (cardiac death, 11.4% vs. 12.2%, P=0.82; TLR 13.3% vs. 8.8%, P=0.25; and MACE, 20.9% vs. 23.4%, P=0.63), although the 1-year TLR rate tended to be lower in the DES group than in the BMS group (13.3% vs. 2.0%). Spontaneous MI occurred in 3 patients (BMS, 1; DES, 2). Definite stent thrombosis was not observed, but probable stent thrombosis occurred in 1 patient 43 months after DES implantation.
Cumulative incidence of cardiac death, target lesion revascularization (TLR), and major adverse cardiac events (MACE).
Clinical Outcomes compared between bare-metal stents (BMS) and drug-eluting stents (DES). (A) Cardiac death, (B) Target LESION revascularization and (C) major adverse cardiac events.
Early follow-up angiography was performed for 77 patients (82.8%). The binary restenosis rate was 10.4%, and there was no significant difference between the BMS and DES groups (16.7% vs. 4.9%, P=0.14). In lesions with preprocedural stenosis, however, the binary restenosis rate was significantly lower in the DES group than in the BMS group (57.1% vs. 12.5%, P=0.045), and that in the LMCA tended to be lower in the DES group than in the BMS group (37.5% vs. 6.3%, P=0.091) (Figure 4). On the other hand, coronary artery dissection recurred in 8 patients, all in the DES group. The incidence of recurrence of coronary artery dissection was significantly higher in the DES group than in the BMS group (0.0% vs. 19.5%, P=0.006). A representative case of recurrent coronary artery dissection is shown in Figure 5. Late follow-up angiography was performed for 60 (84.5%) of the remaining 71 patients, excluding those undergoing TLR. Late restenosis did not occur in either group.
Binary restenosis rate at early follow-up angiography. BMS, bare-metal stent; DES, drug-eluting stent; LMCA, left main coronary artery; RCA, right coronary artery.
Incidence and representative case of recurrent coronary artery dissection. (A) Iatrogenic coronary dissection caused by a guiding catheter (arrow). (B,C) Nobori stent placed in the dissection site (dotted line). The dissection had healed on final angiography. (D) At 8-month follow-up angiography, the dissection had reappeared (arrow).
Table 3 lists the 8 cases of recurrent coronary artery dissection: 4 after Cypher stent implantation and 4 after Nobori stent implantation. The incidence of recurrent coronary artery dissection was 50.0% after Cypher stent implantation and 16.0% after Nobori stent implantation in patients undergoing follow-up angiography (P=0.07). There was no recurrent coronary artery dissection in patients treated with other types of stent, although the number of patients who underwent follow-up angiography was as small as 8 patients. Of the 8 patients with recurrent coronary artery dissection, 1 died of acute MI 43 months after the index procedure. It was a probable case of very late stent thrombosis and she had ceased antiplatelet therapy. Another patient died suddenly of unknown cause. Late follow-up angiography was performed for the remaining 5 patients, and the dissection was apparent in all of them.
| Case no. |
Age (years) |
Sex | Coronary risk factors |
Dissection site |
Dissection type (NHLBI class) |
Stent type/ stent size |
F/U duration (months) |
Clinical event |
|---|---|---|---|---|---|---|---|---|
| 1 | 74 | F | HT, DL | RCA | B | Cypher/3.0 mm | 60 | None |
| 2 | 73 | F | HT, DM | RCA | F | Cypher/3.5 mm | 43 | Died of AMI 43 months after procedure |
| 3 | 67 | F | HT | RCA | B | Cypher/3.0 mm | 60 | TLR at 8 months after procedure |
| 4 | 81 | F | HT, DL | RCA | F | Cypher/3.5 mm | 60 | None |
| 5 | 66 | F | HT, DL | LMCA | F | Nobori/3.5 mm | 57 | Sudden death 57 months after procedure |
| 6 | 61 | F | HT, DL | RCA | C | Nobori/3.5 mm | 35 | None |
| 7 | 81 | M | HT, DM | LMCA | D | Nobori/3.5 mm | 9 | Died of pneumonia 9 months after procedure |
| 8 | 70 | M | HT | RCA | D | Nobori/3.5 mm | 50 | None |
AMI, acute myocardial infarction; DL, dyslipidemia; DM, diabetes mellitus; F, female; F/U, follow-up; HT, hypertension; M, male; TLR, target lesion revascularization.
The main findings of this study were as follows: (1) immediate and long-term outcomes with bailout stenting for iatrogenic coronary artery dissection were acceptable; (2) there were no significant differences in clinical outcomes between BMS and DES implantation; and (3) coronary artery dissection can recur after DES implantation, although DES may be favorable for stenotic lesions.
Guide catheter-induced iatrogenic coronary artery dissection during cardiac catheterization is a feared complication with potentially fatal outcome,1,13 and its incidence has been reported to be <0.1% in studies that were limited to the LMCA.2–4 This study showed an incidence of iatrogenic coronary artery dissection due to a diagnostic or guiding catheter, including both the LMCA and right coronary artery, of 0.14%. Approximately 90% of cases occurred during PCI, and it was approximately 20-fold more likely to occur during PCI compared with CAG.
When dissection does occur, conservative therapy, bailout stenting, and CABG are regarded as potential strategies. Eshtehardi et al reported that MACE-free survival at 5 years after bailout procedure was approximately 60%, and that there were no significant differences in 5-year clinical outcomes between bailout stenting and CABG for iatrogenic LMCA dissection, although their study population was as small at 31 patients (bailout stenting, 14 patients; CABG, 17 patients).4 In contrast, Curtis et al reported that prolonged periods of ischemia during the waiting time for CABG often culminated in unfortunate results even when CABG was successfully performed.5 In the present study, all patients except those for whom conservative therapy was selected were successfully treated with bailout stenting. No patients died during bailout procedures. Although 2 patients required mechanical circulatory support with IABP and VA-ECMO after the occurrence of iatrogenic coronary artery dissection, they were weaned from them after intensive therapy. Moreover, their long-term outcomes compared favorably with the previous study.4 Therefore, prompt stent implantation is an effective procedure in patients with iatrogenic coronary artery dissection, and their immediate and long-term outcomes are considered acceptable. However, approximately 20% of the patients had myocardial injury, and periprocedural myocardial injury is reportedly associated with a higher risk of death and adverse cardiac events.14–17 Therefore, careful postprocedural management is necessary even when the bailout procedure is successfully performed.
Comparison of the long-term clinical outcomes on the basis of stent type revealed no significant differences between the BMS and DES groups. TLR beyond 1 year occurred for 2 patients in the DES group. TLR at 1 year tended to be lower in the DES group than in the BMS group; however, the 5-year cumulative incidence was similar in both groups. Although the binary restenosis rate at early follow-up angiography was similar in both groups, in terms of stenotic lesions and the LMCA, it tended to be lower in the DES group than in the BMS group. On the other hand, coronary artery dissection recurred after DES implantation; 1 patient died of acute MI 43 months after the procedure, which was a probable case of very late stent thrombosis. The mechanism of recurrence of coronary artery dissection is unclear because additional assessments by intravascular ultrasound and optical coherence tomography were not performed. Further studies are required to clarify the mechanism of this phenomenon. On the basis of the angiographic results, we speculate that DES is more favorable for treating iatrogenic coronary artery dissection than BMS in lesions with preprocedural stenosis and in the LMCA because DES is more suitable for suppressing in-stent restenosis. However, it is important to recognize that coronary artery dissection can recur after DES implantation, although its incidence may be lower in the newer-generation DES era.
Study LimitationsTwo major limitations of our study need to be considered. First, this was a retrospective and observational study. Selection bias was potentially present, especially in the comparison of stent types because they were not randomized and could reflect physician’s preference. Second, the study population was relatively small because iatrogenic coronary artery dissection is a rare complication. Notwithstanding, our study represents the largest population published to date describing the long-term outcomes with bailout stenting for iatrogenic coronary artery dissection.
All cases of iatrogenic coronary artery dissection that were recognized as requiring bailout procedure could be managed by stent implantation, and the outcomes after bailout stenting were acceptable. However, periprocedural myocardial injury can occur, and coronary artery dissection can recur after DES implantation, although DES may be favorable for stenotic lesions. Therefore, close postprocedural management is necessary even when a bailout procedure is successfully performed.
The authors appreciate the staff of the cardiac catheterization laboratory, and Miho Kobayashi, Makiko Kanaike and Yoshimi Sano for their assistance with the manuscript.
The authors have no conflicts of interest to disclose.
Name of the ethics committee: Kurashiki Central Hospital Medical Ethics Committee. Reference number: 3307.
