The perioperative outcomes of noncardiac surgery in patients who have received second-generation drug-eluting stents (DESs) have received limited study. We reviewed the medical records of 1,748 consecutive patients who received DES at our institution (1,789 procedures) from January 1, 2009, to July 1, 2012, to determine the outcomes of subsequent noncardiac surgery. During a median follow-up of 43 months, 221 patients underwent 345 noncardiac surgeries (138 low risk, 130 intermediate risk, and 77 high risk), of which 278 were in patients with previous second-generation DES implantation. The incidence of noncardiac surgery in patients with previous second-generation DES implantation was 4.5% at 1 year, 11.6% at 2 years, and 15.2% at 3 years. The mean time from stent implantation to surgery was 21 ± 12.9 months. Mean age was 66 ± 8 years, 99% were men, and 11% had a perioperative complication, including 5.8% major bleeding, 2.5% acute kidney injury, 2.2% major adverse cardiac event, and 1.4% stroke. Perioperative stent thrombosis occurred in 2 patients (0.7%, 95% confidence interval 0.2% to 2.6%): 1 patient had received a DES 14 months before surgery and had stent thrombosis on the day of surgery and the other had a DES implanted 21 months before surgery and developed stent thrombosis the day after surgery. In conclusion, the incidence of perioperative complications with noncardiac surgery after second-generation DES implantation was 11% and consisted mainly of bleeding (5.8%). The incidence of definite stent thrombosis was 0.7%.
Drug-eluting stents (DESs) significantly reduce the rates of in-stent restenosis compared with bare metal stent implantation. However, concerns emerged for increased risk of stent thrombosis, even many years after DES implantation. The perioperative period is a time of increased concern for stent thrombosis, as surgery causes a prothrombotic state and antiplatelet medications are often discontinued. Second-generation DESs are made of cobalt-chrome or platinum-chrome platforms and have thinner strut thickness and more biocompatible, durable polymer coatings compared with first-generation DESs. Second-generation DESs further improved the outcomes achieved with first-generation DESs by reducing the risk of restenosis, myocardial infarction (MI), and stent thrombosis. The impact of second-generation DESs on the incidence of perioperative stent thrombosis after noncardiac surgery has received limited evaluation and formed the focus of the present study.
Methods
We reviewed the records of 1,748 consecutive patients who underwent DES implantation at our institution (1,789 procedures) from January 1, 2009, to July 1, 2012, to determine whether they subsequently underwent noncardiac surgery and whether they developed any perioperative complications.
Stent thrombosis was defined according to the Academic Research Consortium criteria. If patients had multiple stent placements, time to surgery was recorded from the most recent DES placement. First-generation DESs included sirolimus- and paclitaxel-eluting stents (Cypher [Cordis, Warren, New Jersey] and Taxus [Boston Scientific, Natick, Massachusetts]). Second-generation DESs included everolimus-, zotarolimus-, and paclitaxel-eluting platinum chromium stents (Promus [Boston Scientific, Natick, Massachusetts], Xience [Abbott Vascular, Santa Clara, California], Endeavor and Resolute [Medtronic Vascular, Santa Rosa, California], and Ion [Boston Scientific, Natick, Massachusetts]). Patients were considered to have continued preoperative aspirin or ADP P2Y12 inhibitors if the medications were not discontinued or held ≥5 days before surgery. Major adverse cardiac events (MACEs) were defined as perioperative MI, coronary revascularization, and all-cause death. MI was defined as an increase in cardiac biomarkers (creatine kinase, creatine kinase myocardial band, or troponin) ≥3 times upper limit of normal, with at least one of the following: electrocardiographic changes suggestive of ischemia or patient report of chest pain lasting at least 20 minutes. Major bleeding was defined as any bleeding associated with hypotension, estimated blood loss >500 mL, or transfusion of at least 2 units of packed red blood cells. Acute renal failure was defined as a twofold increase in creatinine or decrease by 50% in glomerular filtration rate. Stroke was defined as pathological, imaging, or other objective evidence of cerebral, spinal, or retinal focal ischemic injury in a defined vascular distribution or clinical evidence of cerebral, spinal cord, or retinal focal ischemic injury based on symptoms lasting ≥24 hours or until death, with other etiologies excluded. An event was considered to be postoperative if it occurred within 30 days from noncardiac surgery.
Surgeries were classified as low, intermediate, or high risk according to the American College of Cardiology/American Heart Association guidelines. Low-risk surgeries included endoscopic, superficial, cataract, breast, and ambulatory procedures; intermediate-risk surgeries included intraperitoneal, intrathoracic, carotid endarterectomy, head and neck, and prostate procedures; high-risk surgeries included aortic and other major vascular surgery, peripheral vascular surgery, and emergent surgeries. Simple epidural injections and nerve blocks were excluded from the study.
Continuous parameters were reported as mean ± SD and compared using the Wilcoxon rank-sum test. Discrete parameters were reported as percentages and compared using the chi-square or the Fisher’s exact test, as appropriate. A 2-sided p value of <0.05 was considered statistically significant. All analyses were done using JMP 11.0 (SAS Institute, Cary, North Carolina). The study was approved by the Institutional Review Board.
Results
During a median follow-up of 43 months, 221 of the 1,748 DES patients underwent 345 noncardiac surgeries. Of those 345 noncardiac surgeries, 278 occurred in patients who had received a second-generation DES, 58 in patients who had received a first-generation DES, and 9 in patients who had received both a first- and a second-generation DES. The mean time from DES implantation to surgery was 21 ± 12.9 months for second-generation DES and 22 ± 11.3 months for first-generation stents (p = 0.138). The incidence of noncardiac surgery for second-generation DES was 4.5% at 1 year, 11.6% at 2 years, and 15.2% at 3 years. The incidence of noncardiac surgery for first-generation DES was 3.8% at 1 year, 13.3% at 2 years, and 22.5% at 3 years. The characteristics of the study patients are listed in Table 1 . The most common types of surgery were (1) general (n = 77, 22.3%), (2) vascular (n = 77, 22.3%), and (3) ophthalmologic (n = 72, 20.9%). The least common surgeries were podiatric (n = 2, 0.6%), neurosurgical, dermatologic, and dental (each n = 5, 1.5%). There were no significant differences in the type of surgery among patients who had undergone first- or second-generation DES (p = 0.630).
| Variable | All (221 Patients, 345 Noncardiac Procedures) | Second-Generation DES (180 Patients, 278 Procedures) | First-Generation DES (35 Patients, 58 Procedures) | Both (6 Patients, 9 Procedures) | p |
|---|---|---|---|---|---|
| Age (years) | 65 ± 8 | 66 ± 8 | 65 ± 6 | 67 ± 7 | 0.871 |
| Men | 219 (99.1%) | 178 (98.9%) | 35 (100%) | 6 (100%) | 0.662 |
| Previous myocardial infarction | 78 (35%) | 65 (36%) | 8 (23%) | 5 (83%) | 0.015 |
| Previous coronary bypass graft surgery | 64 (29%) | 54 (30%) | 8 (23%) | 2 (33%) | 0.666 |
| Current smoker | 66 (33%) | 57 (36%) | 6 (19%) | 3 (50%) | 0.107 |
| Ever smoked | 164 (82%) | 136 (84%) | 22 (69%) | 6 (100%) | 0.042 |
| Average smoking (pack-years) | 41 ± 35 | 43 ± 36 | 26 ± 25 | 46 ± 27 | 0.079 |
| Hyperlipidemia ∗ | 197 (89%) | 160 (89%) | 32 (91%) | 5 (83%) | 0.820 |
| Hypertension † | 197 (89%) | 160 (89%) | 31 (89%) | 6 (100%) | 0.496 |
| Diabetes mellitus | 122 (55%) | 95 (53%) | 26 (74%) | 1 (17%) | 0.008 |
| Diabetes mellitus on insulin | 53 (24%) | 40 (22%) | 13 (37%) | 0 (0%) | 0.037 |
| Peripheral arterial disease | 48 (22%) | 42 (23%) | 6 (17%) | 0 (0%) | 0.160 |
| Glomerular filtration rate (ml/min) | 75 ± 33 | 75 ± 31 | 71 ± 41 | 90 ± 26 | 0.253 |
| Chronic renal insufficiency | 40 (18%) | 32 (18%) | 7 (20%) | 1 (17%) | 0.949 |
| End stage renal disease | 14 (6%) | 10 (6%) | 4 (11%) | 0 (0%) | 0.328 |
| Left ventricular ejection fraction (%) | 51 ± 12 | 52 ± 12 | 49 ± 11 | 52 ± 10 | 0.418 |
| Number of coronary vessels stented | 0.046 | ||||
| 1 | 163 (74%) | 138 (77%) | 22 (63%) | 3 (50%) | |
| 2 | 49 (22%) | 34 (19%) | 13 (37%) | 2 (33%) | |
| 3 | 9 (4%) | 8 (4%) | 0 (0%) | 1 (17%) | |
| Type of coronary stents implanted | <0.001 | ||||
| Paclitaxel | 10 (5%) | 9 (26%) | 1 (17%) | ||
| Sirolimus | 28 (13%) | 26 (74%) | 2 (33%) | ||
| Everolimus | 104 (47%) | 101 (56%) | 3 (50%) | ||
| Zotarolimus | 77 (35%) | 77 (43%) | 0 (0%) | ||
| ION paclitaxel | 2 (1%) | 2 (1%) | 0 (0%) | ||
| Indication for PCI | 0.020 | ||||
| Elective | 106 (48%) | 94 (52%) | 11 (31%) | 1 (17%) | |
| Acute coronary syndrome | 115 (52%) | 86 (48%) | 24 (69%) | 5 (83%) | |
| Medications used before surgery | |||||
| Aspirin | 309 (90%) | 249 (90%) | 52 (90%) | 8 (89%) | 1.00 |
| Clopidogrel | 200 (58%) | 167 (60%) | 28 (48%) | 5 (56%) | 0.255 |
| Prasugrel | 11 (3%) | 10 (4%) | 1 (2%) | 0 (0%) | 0.546 |
| Ticagrelor | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | N/A |
| Beta-blocker | 299 (87%) | 244 (88%) | 48 (83%) | 7 (78%) | 0.463 |
| Angiotensin-converting enzyme inhibitor | 195 (57%) | 161 (58%) | 26 (45%) | 8 (89%) | 0.019 |
| Angiotensin receptor blocker | 56 (16%) | 44 (16%) | 12 (21%) | 0 (0%) | 0.134 |
| Calcium channel blocker | 88 (26%) | 79 (28%) | 8 (14%) | 1 (11%) | 0.028 |
| Statin | 303 (88%) | 240 (86%) | 54 (93%) | 9 (100%) | 0.098 |
| Preoperative aspirin | 229 (66%) | 188 (68%) | 39 (67%) | 2 (22%) | 0.022 |
| Preoperative thienopyridine | 100 (29%) | 85 (31%) | 13 (22%) | 2 (22%) | 0.392 |
| Preoperative dual antiplatelet therapy | 88 (26%) | 76 (27%) | 11 (19%) | 1 (11%) | 0.213 |
| No preoperative antiplatelet therapy | 103 (30%) | 80 (29%) | 17 (29%) | 6 (67%) | 0.070 |
∗ Hyperlipidemia was defined as low-density cholesterol >100 mg/ml or use of antilipidemic medications.
† Hypertension was defined as blood pressure >140/90 mm Hg or use of antihypertensive medications.
The rates of stent thrombosis, MACE, major bleeding, acute renal failure, syncope, and stroke up to 30 days after noncardiac surgery are listed in Table 2 . Perioperative stent thrombosis occurred in 2 patients (0.6% of all patients, 95% confidence interval 0.2% to 2.1%, and 0.7% of second-generation DES patients, 95% confidence interval 0.2% to 2.6%). Overall, there was a trend toward lower incidence of perioperative complications in patients who had received a second-generation DES compared with those who had received a first-generation DES before surgery.
| Outcome | All Procedures (n = 345) | Procedures in Patients with Second-Generation DES (n = 278) | Procedures in Patients With First-Generation DES (n = 58) | p |
|---|---|---|---|---|
| Any Complication ∗ | 38 (11%) | 31 (11%) | 8 (14%) | 0.281 |
| Major bleeding | 21 (6.1%) | 16 (5.8%) | 5 (8.6%) | 0.414 |
| MACE | 10 (2.9%) | 6 (2.2%) | 4 (6.9%) | 0.172 |
| Acute renal failure | 9 (2.6%) | 7 (2.5%) | 2 (3.5%) | 0.729 |
| Stroke | 4 (1.1%) | 4 (1.4%) | 0 (0%) | 0.419 |
| Stent thrombosis | 2 (0.6%) | 2 (0.7%) | 0 (0%) | 0.648 |
| Syncope | 0 (0%) | 0 (0%) | 0 (0%) | N/A |
∗ Any complication includes any patient who had perioperative major bleeding, MACE, acute renal failure, stroke, stent thrombosis, or syncope.
The first patient who developed perioperative stent thrombosis was a 62-year-old man who underwent elective transurethral resection of the prostate for gross refractory hematuria 14 months after implantation of an Endeavor zotarolimus-eluting stent into the right coronary artery for stable angina. Aspirin and prasugrel were held before surgery, and eptifibatide was administered. Three hours after surgery and before eptifibatide could be restarted, he developed severe chest pain, bradycardia, and lateral ST-segment elevation. Emergent coronary angiography confirmed thrombosis of the proximal right coronary artery stent, which was successfully treated with percutaneous coronary intervention followed by uneventful recovery.
The second patient who developed perioperative stent thrombosis was a 53-year-old man who presented for elective open sigmoidectomy with diverting loop ileostomy for repeated episodes of diverticulitis 21 months after stenting of the first obtuse marginal artery for non-ST-segment elevation MI with an Endeavor zotarolimus-eluting stent. Aspirin was continued during the perioperative period, but clopidogrel was discontinued 7 days before surgery. Thirty hours after surgery, the patient became dyspneic, diaphoretic, and tachycardic and developed ST-segment depressions in leads V2 to V3. He had no chest pain. Cardiac biomarkers were markedly elevated (creatine kinase myocardial band, >300 ng/ml and troponin I >100 ng/ml). Coronary angiography confirmed first obtuse marginal stent thrombosis, and percutaneous coronary intervention was performed. The patient developed severe hypotension. Despite insertion of an intra-aortic balloon pump and multiple vasopressors, he died on the second postoperative day.
Of 345 cases of noncardiac surgery, 138 were low-risk, 130 were intermediate-risk, and 77 were high-risk procedures. The risk of MACE (p = 0.295) and major bleeding (p = 0.029) was lower in low-risk surgeries compared with intermediate- and high-risk surgeries ( Figure 1 ). The risk of any perioperative complication (5.8% vs 15.4% vs 13.0%, respectively, p = 0.028) was likewise lower in low-risk surgeries compared with intermediate- and high-risk surgeries. Among the 278 noncardiac surgeries in patients with previous second-generation DES, the risk of MACE (p = 0.406), major bleeding (p = 0.099), and any perioperative complication (6.0% vs 15.5% vs 11.9%, respectively, p = 0.070) was lower in low-risk surgeries compared with intermediate- and high-risk surgeries.
