Abstract
Background:
Patients with existing coronary artery stents are at increased risk for major adverse cardiac events (MACE) when undergoing noncardiac surgery (NCS). Although the use of antifibrinolytic (AF) therapy in NCS has significantly increased in the past decade, the relationship between perioperative AF use and its association MACE among patients with existing coronary artery stents has yet to be assessed. In this study, we aim to evaluate the association of MACE in patients with existing coronary artery stents who receive perioperative AF therapy during orthopedic surgery.
Methods:
A single-center retrospective cohort study was conducted in adult patients with existing coronary artery stents who underwent orthopedic surgery from 2008–2018. Two cohorts were established: patients with existing coronary artery stents who did not receive perioperative AF and patients with coronary artery stents who received perioperative AF. Associations between AF use and the primary outcome of MACE within 30-days postoperatively and the secondary outcomes of thrombotic complications, excessive surgical bleeding, and ICU admissions were analyzed using logistic regression models. Inverse probability of treatment weighting was used to control for confounding. Secondary analyses examining the association between coronary stent type/timing and the outcomes of interest were performed using unadjusted logistic regression models.
Results:
473 patients met study criteria, including 294 who did not receive AF and 179 patients who received AF. MACE occurred in 15 (5.1%) patients who did not receive AF and 1 (0.6%) who received AF (p=0.007). In weighted analyses, no significant difference was found in patients who received AF with regard to MACE (OR 0.13, 95% CI=0.01–1.74, p=0.12), thrombotic complications (OR 1.19, 95% CI=0.53–2.68, p=0.68), or excessive surgical bleeding (OR 0.13, 95% CI=0.01–2.23, p=0.16) compared to patients who did not receive AF.
Conclusion:
The results of this study are inconclusive whether an association exists between perioperative AF use in patients with coronary artery stents and the outcome of MACE compared to patients who did not receive perioperative AF therapy. The authors acknowledge that the imprecise confidence interval hinders the ability definitively determine whether an association exists in the study population. Further large prospective studies, powered to detect differences in MACE, are needed to assess the safety of perioperative AF in patients with existing coronary artery stents and to clarify the mechanism of perioperative MACE in this high-risk population.
Keywords: coronary artery stents, antifibrinolytic medication, noncardiac surgery, orthopedic surgery, major adverse cardiac events
INTRODUCTION
Patients with existing coronary artery stents undergoing noncardiac surgery (NCS) present perioperative challenges due to increased risk of major cardiac events (MACE) including stent thrombosis, myocardial infarction, need for repeat coronary revascularization, bleeding complications, and death.1,2 Cessation of dual antiplatelet therapy (DAPT) in the perioperative period increases the rate of stent thrombosis, while continuation of DAPT may increase the risk of excessive surgical bleeding.3,4
An increased emphasis on conservative red blood cell (RBC) transfusion triggers/targets has contributed to a surge in the use of perioperative antifibrinolytic (AF) therapy such as tranexamic (TXA) and aminocaproic acid (AA).5 Initial studies evaluating perioperative AF therapy reveal a clear decrease in mortality and/or blood loss in trauma, obstetric, and orthopedic surgeries.6–8 Subsequent studies have established efficacy in a wide spectrum of subspecialties including neurological, otolaryngological, urological, spine, and cardiac surgery.9–13
MACE among patients with previous percutaneous coronary intervention (PCI) and stenting, including in-stent thrombosis, have been anecdotally reported, but clinical studies are lacking.14,15 With an increasing prevalence of AF use in NCS and approximately 1.25 million coronary artery stents placed annually in the United States alone, there exists a strong need for research to help guide clinicians’ use of AF in this potentially high-risk patient population.16 The purpose of this study is Aim 1: assess the association of perioperative AF use in patients undergoing orthopedic surgery with existing coronary artery stents and the primary outcome of MACE and secondary outcomes of thrombotic complications and excessive surgical bleeding; and Aim 2: among patients with coronary artery stents, assess the relationship between stent type, quantity, and time between stent placement and surgery, and the aforementioned outcomes. The authors hypothesize that there will be no difference in MACE regardless of perioperative AF use.
METHODS
This study was approved by the Institutional Review Board (IRB) at Mayo Clinic, Rochester, Minnesota, and the requirement for written informed consent was waived for all patients who had previously granted permission to use their medical records for observational research (consistent with Minnesota Statute 144.295). The study cohort was identified through the Perioperative DataMart (Mayo Clinic, Rochester, MN), an institutional database that contains demographic, anesthetic, surgical, laboratory, and medication data in multiple perioperative care environments. Patients who had previous coronary stent deployment were identified through the Mayo Clinic PCI registry. Additional patient data not available in the Perioperative DataMart were extracted utilizing the Advanced Cohort Explorer, a resource used to search the electronic medical record for structured and unstructured inquiries from the inpatient and outpatient care environment. Both DataMart and Advanced Cohort Explorer have been comprehensively validated with continual monitoring of data quality.17,18 Patients who underwent NCS between January 1, 2008 and March 31, 2018 were identified and included in the study cohort. Patients who declined to participate in research, coronary stenting and surgical procedures occurring outside our institution, non-orthopedic procedures, orthopedic surgical procedures performed during the same hospitalization following cardiac surgery, subsequent cardiac catheterization procedures, patients with missing sex (excluded from multivariate analysis and weighting), missing American Society of Anesthesiologists (ASA) physical status classification score, and an ASA classification of 5 or 6 were excluded.
Demographic and preoperative data was extracted, including coronary angiographic data (date of stenting, indication, type of stent [bare metal or drug-eluting], location of stenting, and number of stents deployed), maintenance of antiplatelet therapy (none, aspirin alone, P2Y12 inhibitor alone, or DAPT), anticoagulant use preoperatively, risk factors for coronary artery disease (CAD), history of thrombotic disease (deep venous thrombosis [DVT], stroke, pulmonary embolism [PE]) and preoperative laboratory testing. Intraoperative data was extracted including date of surgery, type of surgical procedure, urgency of surgery (emergent or elective), ASA classification, anesthesia type, type and dose of AF medication administered, intraoperative fluid balance, estimated blood loss (EBL), and perioperative blood product administration. Postoperative data including MACE, thrombosis within 30 days of surgery (DVT, PE, stroke, visual loss), blood product transfusion in the first 48 hours postoperatively, reoperation for bleeding, postoperative anticoagulation, hospital and ICU length of stay (LOS), 30-day post-surgery all-cause mortality, and 1 year post-surgery all-cause mortality were reviewed.
Definitions
The primary outcome was MACE, defined as a patient experiencing death, acute myocardial infarction (MI), repeat coronary revascularization, or in-stent thrombosis within 30 days of surgical date. Secondary outcomes included thrombotic complications (defined as acute DVT, PE, stroke, or vision loss within 30 days postoperatively), excessive surgical bleeding19 (defined as EBL ≥1000 mL, fatal bleeding, symptomatic bleeding, bleeding in a closed space, transfusion of ≥2 blood products intraoperatively or within 48 hours postoperatively, bleeding that required subsequent surgical exploration, or bleeding that caused hemodynamic instability), and postoperative ICU admission. Perioperative MI was defined as elevated cardiac biomarker(s) (troponin or creatinine kinase myocardial band) in addition to one of the following: electrographic changes suggestive of ischemia, echocardiographic evidence of new regional wall motion abnormalities, ischemic symptoms, or angiographic evidence of intracoronary thrombus.20 Postoperative screening for MACE with serum cardiac biomarkers occurred at the discretion of individual providers. Perioperative blood products (within 48 hours postoperatively) included RBC, platelets, fresh frozen plasma, cryoprecipitate, autologous RBC salvage, or factor concentrates.
Statistical Analysis
This is a retrospective cohort study, consisting of orthopedic surgical patients with coronary stents who received or did not receive perioperative AF therapy. Given the small number of patients with the primary outcome of MACE (N=16), there was insufficient power to adjust for all a priori identified confounding variables using simple covariate adjustment. The authors used inverse probability of treatment weighting to account for imbalances in a priori identified confounding variables.21 The administration of AF therapy served as the dependent variable and patient demographics served as predictors (Table 1) and was modeled using logistic regression. Variables contributing to the propensity included the following a priori identified potential confounders: age, sex, body mass index (BMI), ASA classification, Charlson Comorbidity Index (CCI), emergency surgery status, history of PE, history of DVT, aspirin use, antiplatelet use, revised cardiac risk index (RCRI), preoperative warfarin use, and preoperative hemoglobin. Absolute standardized differences <10% were considered balanced between patients with and without AF therapy when compared before and after weighting. Data were descriptively summarized using frequencies and percentages for categorical data and means and standard deviations for continuous data. Data distributions across AF use were compared using Chi-square/Fisher exact tests for categorical variables (where appropriate) and Wilcoxon rank-sum tests for continuous data.
Table 1.
Demographics and Patient Characteristics
| Unweighted | Weighted | ||||||
|---|---|---|---|---|---|---|---|
| Outcome | Stents − AF N=294 |
Stents + AF N=179 |
p-value | Stand. Diff. (%) |
Stents − AF | Stents + AF | Stand. Diff. (%) |
| Age | 71.8 (11.2) | 72.2 (9.8) | 0.65 † | 0.04 | 71.9 (11.2) | 72.2 (9.8) | 0.03 |
| Male Sex | 195 (66.3%) | 129 (72.1%) | 0.19 ‡ | 0.12 | 68.5% | 68.7% | <.01 |
| BMI | 30.6 (6.1) | 31.3 (5.8) | 0.22 † | 0.12 | 30.9 (6.1) | 30.9 (5.8) | <.01 |
| ASA | 0.85 § | 0.01 | 0.03 | ||||
| 1 | 1 (0.3%) | 0 (0.0%) | 0.3% | 0.0% | |||
| 2 | 59 (20.1%) | 35 (19.6%) | 19.9% | 20.6% | |||
| 3 | 207 (70.4%) | 131 (73.2%) | 70.8% | 68.8% | |||
| 4 | 27 (9.2%) | 13 (7.3%) | 9.0% | 10.6% | |||
| Charlson Comorbidity Index | 6.6 (3.2) | 6.1 (3.0) | 0.07 † | 0.17 | 6.4 (3.2) | 6.4 (3.0) | 0.01 |
| Emergency Surgery | 16 (5.4%) | 1 (0.6%) | 0.004 § | 0.29 | 3.6% | 2.2% | 0.09 |
| History of PE | 26 (8.8%) | 15 (8.4%) | 0.86 ‡ | 0.02 | 8.8% | 9.2% | 0.01 |
| History of DVT | 43 (14.6%) | 26 (14.5%) | 0.98 ‡ | <.01 | 14.6% | 14.6% | <.01 |
| Preop Aspirin | 258 (87.8%) | 154 (86.0%) | 0.59 ‡ | 0.05 | 87.1% | 87.2% | <.01 |
| Preop DAPT | 126 (42.9%) | 61 (34.1%) | 0.06 ‡ | 0.18 | 39.6% | 39.4% | <.01 |
| Preop Warfarin | 2 (0.7%) | 1 (0.6%) | 1.00 § | <.01 | 0.6% | 0.5% | <.01 |
| Preop Hgb | 12.6 (2.1) | 13.3 (1.8) | 0.001 † | 0.31 | 12.9 (2.1) | 12.9 (1.8) | 0.03 |
Numbers indicate N (%) and Mean (SD)
Equal variance T-Test
Chi-square
Fisher exact
Abbreviations: AF, antifibrinolytic medications; ASA, American Society of Anesthesiologists physical status classification; BMI, body mass index; DAPT, dual antiplatelet therapy; Diff, difference; DVT, deep venous thrombosis; Hgb, hemoglobin; MACE, major adverse cardiac events; PE, pulmonary embolism; Preop, preoperative; Stand, standard.
We used logistic regression to assess the association between perioperative AF therapy and the binary outcomes of MACE, thrombotic complications, excessive surgical bleeding, and postoperative ICU admission. Inverse probability of treatment weighting was used to control for confounding variables. We also performed secondary analyses to examine the association between perioperative AF therapy with the time from PCI to surgical procedure (days) and coronary stent type (i.e. bare metal stent [BMS], drug-eluting stent [DES]) and the outcomes of interest. All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC). A two-sided p-value < 0.05 was considered statistically significant. No a priori power calculation was used to estimate sample size. The rate of MACE in patients with stents who did not receive AF was 5.1%. It was determined that a clinically meaningful difference was an absolute difference of 4.0%. Given these assumptions, our study had 76% power to detect a statistically significant difference larger than 4% (e.g. 5.1% in patients without AF vs. 1.1% in patients with AF) at the 0.05 level.
RESULTS
A total of 473 patients underwent orthopedic surgery between January 1, 2008 and March 31, 2018 and were incorporated in this review, including 294 patients with Stents - AF (62.2%) and 179 patients with Stents + AF (37.8%) (Figure 1).
Figure 1:
Consort Flow Diagram.
Abbreviations: AF, antifibrinolytics; ASA, American Society of Anesthesiologists.
Demographic and patient characteristics are displayed in Table 1. Standardized differences before and after weighting are presented in Table 1 and Figure 2. All variables were balanced following weighting (absolute standardized difference < 10). Among the 473 patients, 16 (3.4%) experienced MACE, 38 (8.0%) experienced thrombotic complications, and 12 (2.5%) experienced excessive surgical bleeding (Table 2). One or more MACE was identified in 16 patients including 5 with acute coronary syndrome (ACS) without subsequent revascularization, 4 with ACS who underwent subsequent revascularization, 3 with ACS who died within 30 days postoperatively, one with ACS who underwent revascularization and died within 30 days postoperatively, and 3 patients who died within 30 days postoperatively with no documented cause of death. After weighting, the proportion of MACE was 4.7% in Stents - AF patients and 0.6% in Stents + AF patients (OR 0.13, 95% CI=0.01–1.74, p=0.12). Associations between AF use and outcomes with and without weighting are presented in Table 2.
Figure 2:
Standardized Mean Differences Between Unmatched and Propensity-Matched Cohorts.
Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index; DVT, deep vein thrombosis; Hb, hemoglobin; proc, procedure; RCRI, revised cardiac risk index; PE, pulmonary embolism.
Table 2.
Associations by Antifibrinolytic use and Outcomes using Inverse Probability Weighting
| Unweighted | Weighted | ||||||
|---|---|---|---|---|---|---|---|
| Outcome | Stents − AF N=294 |
Stents + AF N=179 |
p-value | Stents − AF | Stents + AF | Odds Ratio § (95% CI) |
p-value § |
| MACE | 15 (5.1%) | 1 (0.6%) | 0.007 † | 4.7% | 0.6% | 0.13 (0.01, 1.74) | 0.12 |
| Thrombotic Complications | 22 (7.5%) | 16 (8.9%) | 0.57 ‡ | 7.4% | 8.7% | 1.19 (0.53, 2.68) | 0.68 |
| Excessive Surgical Bleeding | 12 (4.1%) | 0 (0.0%) | 0.006 ‡ | 3.8% | 0.0% | 0.13 (0.01, 2.21) | 0.16 |
| ICU Admission | 58 (19.7%) | 16 (8.9%) | 0.002 ‡ | 18.2% | 12.3% | 0.63 (0.32, 1.23) | 0.18 |
Numbers indicate N (%) and Odds Ratio (95% CI)
Fisher exact
Chi-square
Weighted odds ratio and p-value estimated using a logistic regression model with the inverse probability weighting. The referent group is No AF.
Abbreviations: AF, antifibrinolytic medications; MACE, major adverse cardiac events; N, number.
In the secondary analysis, Stents – AF patients had a higher rate of prior BMS deployment (23.1% vs. 14.0%, p-value = 0.02), presented for NCS sooner (median of 21 months vs. 35 months, p-value < 0.001), and were more likely to undergo emergency surgery (5.4% vs. 0.6%, p-value=0.004) than Stents +AF patients (Table 1, Table 3). In unadjusted comparisons, the rate of MACE was highest when NCS was performed 0–180 days after PCI (0–90 days, 16.1%; 91–180 days, 0.0%; 181–356 days, 3.6%; 366+ days, 2.5%; overall p-value=0.005; Table 4). Patients who underwent NCS ≥366 days following PCI had lower odds of MACE compared to patients who underwent NCS 0–90 days following PCI (OR 0.13, 95% CI=0.04–0.40, p-value<0.001; Table 4). The rate of MACE in patients with BMS was 3.4% and 3.2% in patients with DES. There were no differences in thrombotic complications between groups (Stents - AF 7.5% vs. Stents + AF 8.9%, OR 1.22, 95% CI=0.63–2.38, p=0.56). Excessive surgical bleeding was highest when NCS was performed 0–180 days after PCI (0–90 days, 9.7%; 91–180, 14.3%; 181–365, 1.8%; 366+, 1.4%; overall p-value=0.001). Patients who underwent NCS ≥366 days following PCI had lower odds of excessive surgical bleeding compared to patients who underwent NCS 0–90 days following PCI (OR 0.12, 95% CI=0.03–0.51, p-value= 0.004).
Table 3:
Coronary Stent Demographics
| Characteristic | Stents − AF N=294 |
Stents + AF N=179 |
p-value |
|---|---|---|---|
| Coronary Stent Variables | |||
| Months from PCI to procedure | 21 (11, 42) | 35 (17, 60) | <0.001 † |
| Days from PCI to procedure | <0.001 ‡ | ||
| 0–90 | 28 (9.5%) | 3 (1.7%) | |
| 91–180 | 15 (5.1%) | 6 (3.4%) | |
| 181–365 | 45 (15.3%) | 11 (6.1%) | |
| 366+ | 206 (70.1%) | 159 (88.8%) | |
| Patients with BMS | 68 (23.1%) | 25 (14.0%) | 0.02 § |
| BMS number | 0.052 § | ||
| 0 | 226 (76.9%) | 154 (86.0%) | |
| 1 | 49 (16.7%) | 18 (10.1%) | |
| 2+ | 19 (6.5%) | 7 (3.9%) | |
| Patients with DES | 228 (77.6%) | 156 (87.2%) | <0.01 § |
| DES number | 0.03 § | ||
| 0 | 66 (22.4%) | 23 (12.8%) | |
| 1 | 141 (48.0%) | 99 (55.3%) | |
| 2+ | 87 (29.6%) | 57 (31.8%) |
Numbers indicate N (%) unless otherwise noted.
Wilcoxon Rank-Sum
Fisher exact
Chi-square
Abbreviations: AF, antifibrinolytic medications; BMS, bare metal stent; DAPT, dual antiplatelet therapy; DES, drug-eluting stent; PCI, percutaneous coronary intervention.
Table 4:
Unadjusted Outcomes Associated with Coronary Stent Demographics Unadjusted
| Unadjusted | |||||
|---|---|---|---|---|---|
| Outcome | Total N=473 |
Event | Estimate (95% CI) |
p-value | Overall p-value |
| MACE | Odds Ratio | ||||
| Months from PCI to procedure, per 1 unit | 473 | 24.7 (28.5) | 0.99 (0.97, 1.01) | 0.23 | 0.23 |
| Days from PCI to procedure | 0.005 | ||||
| 0–90 | 31 | 5 (16.1%) | 1.00 (ref) | - | |
| 91–180 | 21 | 0 (0.0%) | 0.11 (0.01, 2.29) | 0.16 | |
| 181–365 | 56 | 2 (3.6%) | 0.22 (0.05, 1.08) | 0.06 | |
| 366+ | 365 | 9 (2.5%) | 0.13 (0.04, 0.40) *** | <.001 | |
| BMS | 0.93 | ||||
| No | 380 | 13 (3.4%) | 1.00 (ref) | - | |
| Yes | 93 | 3 (3.2%) | 1.05 (0.32, 3.51) | 0.93 | |
| BMS Number | 0.29 | ||||
| 0 | 380 | 13 (3.4%) | 1.00 (ref) | - | |
| 1 | 67 | 1 (1.5%) | 0.61 (0.11, 3.42) | 0.58 | |
| 2+ | 26 | 2 (7.7%) | 2.78 (0.66, 11.65) | 0.16 | |
| DES | 0.86 | ||||
| No | 89 | 3 (3.4%) | 1.00 (ref) | - | |
| Yes | 384 | 13 (3.4%) | 0.90 (0.27, 3.00) | 0.86 | |
| DES number | 0.98 | ||||
| 0 | 89 | 3 (3.4%) | 1.00 (ref) | - | |
| 1 | 240 | 8 (3.3%) | 0.90 (0.25, 3.23) | 0.88 | |
| 2+ | 144 | 5 (3.5%) | 0.97 (0.25, 3.85) | 0.97 | |
| Thrombotic complications | Odds Ratio | ||||
| Months from PCI to procedure, per 1 unit | 473 | 30.6 (24.7) | 1.00 (0.98, 1.01) | 0.56 | 0.56 |
| Days from PCI to procedure | 0.83 | ||||
| 0–90 | 31 | 1 (3.2%) | 1.00 (ref) | - | |
| 91–180 | 21 | 2 (9.5%) | 2.61 (0.31, 22.21) | 0.38 | |
| 181–365 | 56 | 5 (8.9%) | 2.17 (0.33, 14.33) | 0.42 | |
| 366+ | 365 | 30 (8.2%) | 1.85 (0.34, 10.18) | 0.48 | |
| BMS | 0.25 | ||||
| No | 380 | 28 (7.4%) | 1.00 (ref) | - | |
| Yes | 93 | 10 (10.8%) | 1.56 (0.73, 3.29) | 0.25 | |
| BMS number | 0.19 | ||||
| 0 | 380 | 28 (7.4%) | 1.00 (ref) | - | |
| 1 | 67 | 9 (13.4%) | 2.01 (0.91, 4.42) | 0.08 | |
| 2+ | 26 | 1 (3.8%) | 0.73 (0.13, 4.07) | 0.72 | |
| DES | 0.64 | ||||
| No | 89 | 8 (9.0%) | 1.00 (ref) | - | |
| Yes | 384 | 30 (7.8%) | 0.83 (0.37, 1.84) | 0.64 | |
| DES number | 0.86 | ||||
| 0 | 89 | 8 (9.0%) | 1.00 (ref) | - | |
| 1 | 240 | 18 (7.5%) | 0.80 (0.34, 1.87) | 0.60 | |
| 2+ | 144 | 12 (8.3%) | 0.90 (0.36, 2.27) | 0.83 | |
| Excessive Surgical Bleeding | Odds Ratio | ||||
| Months from PCI to procedure, per 1 unit | 473 | 14.0 (20.4) | 0.95 (0.91, 0.99) * | 0.02 | 0.02 |
| Days from PCI to procedure | 0.001 | ||||
| 0–90 | 31 | 3 (9.7%) | 1.00 (ref) | - | |
| 91–180 | 21 | 3 (14.3%) | 1.54 (0.30, 7.82) | 0.60 | |
| 181–365 | 56 | 1 (1.8%) | 0.22 (0.03, 1.61) | 0.14 | |
| 366+ | 365 | 5 (1.4%) | 0.12 (0.03, 0.51) ** | 0.004 | |
| BMS | 0.19 | ||||
| No | 380 | 8 (2.1%) | 1.00 (ref) | - | |
| Yes | 93 | 4 (4.3%) | 2.20 (0.68, 7.10) | 0.19 | |
| BMS number | 0.31 | ||||
| 0 | 380 | 8 (2.1%) | 1.00 (ref) | - | |
| 1 | 67 | 3 (4.5%) | 2.38 (0.66, 8.55) | 0.19 | |
| 2+ | 26 | 1 (3.8%) | 2.58 (0.42, 15.77) | 0.31 | |
| DES | 0.16 | ||||
| No | 89 | 4 (4.5%) | 1.00 (ref) | - | |
| Yes | 384 | 8 (2.1%) | 0.43 (0.13, 1.38) | 0.16 | |
| DES number | 0.36 | ||||
| 0 | 89 | 4 (4.5%) | 1.00 (ref) | - | |
| 1 | 240 | 6 (2.5%) | 0.53 (0.15, 1.81) | 0.31 | |
| 2+ | 144 | 2 (1.4%) | 0.33 (0.07, 1.61) | 0.17 | |
Abbreviations: AF, antifibrinolytic medications; BMS, bare metal stent; CI, confidence interval; DES, drug-eluting stent; MACE, major adverse cardiac events; N, number; PCI, percutaneous coronary intervention; ref, reference.
DISCUSSION
In patients presenting for NCS with prior coronary artery stents, the decision to as to whether it is safe to administer to AF therapy is not well described. Clinicians are therefore left to weigh the risk/benefit of possible MACE or thrombotic complications with those of excessive surgical bleeding. The aim of this study was to evaluate the association of AF therapy and MACE in patients with existing coronary artery stents who underwent orthopedic surgery. Unfortunately, the results of this study are inconclusive as to whether an association exists between perioperative AF use in patients with coronary artery stents and the outcome of MACE compared to patients who did not receive perioperative AF therapy.
Perioperative AF Therapy
Current evidence supporting the use of perioperative AF in NCS includes several large, randomized controlled trials involving trauma, obstetric, and orthopedic surgical patients.6–8 In the CRASH-2 trial, adult trauma patients who received TXA infusions demonstrated decreased all-cause mortality, decreased risk of death due to bleeding, and no significant difference in vascular occlusive events.6 The WOMAN trial consisted of women who experienced post-partum hemorrhage after vaginal or cesarean birth and demonstrated a reduction in death due to bleeding, especially when women received TXA within 3 hours of childbirth.7 In patients undergoing total knee arthroplasty, a meta-analysis of 14 studies involving TXA showed a significant reduction in blood transfusion among those who received AF compared to placebo.8
Despite the above evidence, the use of perioperative AF in NCS has remained highly variable and provider dependent. A recent survey evaluating the use of perioperative TXA revealed high variability among anesthesiologists: 67% administered TXA in routine lower extremity arthroplasty, 49% in trauma surgery, 40% for routine spine surgery, while 36% withheld TXA for vascular/abdominal/pelvic/thoracic surgeries.22 Of those surveyed, 57% agreed that there is uncertainty about the relative risks and benefits of perioperative TXA in NCS.
MACE and Thrombotic Complications
Patients with existing coronary artery stents are at increased risk of MACE after NCS, especially in the setting of urgent or emergent surgery, history of MI within the previous 6 months, and a RCRI >2.1 No previous studies have evaluated the risk of MACE in patients with coronary stents who receive perioperative AF therapy during NCS. Isolated case reports note the risk of MACE, including stent thrombosis. A recent review by Kaptein cited 6 case reports of perioperative MI, including 2 cases of in-stent thrombosis, following TXA administration.14,23 Both cases of in-stent thrombosis occurred in patients with overlapping and bifurcated coronary artery stents.
In the current study, the rate of MACE was not significant different between groups (4.7% in Stents –AF patients and 0.6% in Stents +AF patients; OR 0.13, 95% CI=0.01–1.74, p=0.12). We applied inverse probability of treatment weighting to account for imbalances in a priori identified confounding variables (i.e. age, sex, BMI, ASA classification, CCI, emergency surgery status, history of PE, history of DVT, aspirin use, DAPT, RCRI, preoperative warfarin use, and preoperative hemoglobin) in an effort to mitigate potential confounders and evaluate the association of AF use between groups. The authors acknowledge that the sample size and event infrequency impaired the ability to definitively determine whether an association exists in the study population.
Ultimately, at our institution, the decision to administer AF was at the discretion of the anesthesiologist and surgeon, whose pattern of complex medical decision making is difficult to quantify with retrospective cohort study variables. For example, a general hesitancy among providers to administer AF to patients with coronary artery stents may have led to fewer patients in the Stents + AF group. If a higher overall “sickness” among Stents –AF patients was present, it was not detected by our study given the well-matched demographics present in Table 1.
In an effort to evaluation the association of perioperative AF and the primary outcome of MACE, it is prudent to mention the unclear mechanism of perioperative MI.24 Coronary artery supply-demand mismatch from perioperative bleeding and the pro-inflammatory/prothrombotic state of the perioperative period have been cited as possible mechanisms for perioperative MI.3,25 A potential benefit of fibrinolysis inhibition by AF medications is that plasmin and plasminogen are inflammatory mediators that promote cellular binding, proteolytic activity, and production of inflammatory mediators.26 It is unclear the extent that AF medications modulate inflammation by limiting plasmin’s proteolytic activity but this may warrant further research in patients at risk for perioperative MACE.
Consideration of AF-associated thrombotic complications is of substantial importance to the perioperative provider. Several studies (i.e. CRASH-2 trial, WOMAN trial, and a 2010 large Cochrane Review) have demonstrated that perioperative AF is not associated with vascular occlusive or thrombotic events including stroke, DVT, or PE. 6,7,27 In the current study, there were no statistically significant differences in thrombotic complications (DVT, PE, stroke, or vision loss within 30 days postoperatively) amongst groups following weighted analysis (Table 2). These data are consistent with aforementioned studies involving TXA in NCS patients and are important to note given providers concern over the potential for thrombotic complications in the perioperative period, particularly in high-risk patients.
Excessive Surgical Bleeding
Patients with coronary stents frequently are maintained on DAPT or monotherapy in the perioperative period and present increased risk for excessive surgical bleeding events. Howell, et al 28 reported that surgical patients on DAPT were at increased risk of both MACE and bleeding when compared to patients on monotherapy. Benenati, et al 29 reported that limiting the duration of DAPT following PCI to 1–3 months demonstrated no increased odds of ischemic complications while reducing the odds of major and any bleeding. These benefits of reduced bleeding are weighed against the risk of MACE as described by previous studies which show reduced MACE rates when DAPT is continued >12 months post-PCI.30
In the current study, no significant difference was found between perioperative AF therapy and excessive surgical bleeding (OR 0.13, 95% CI=0.01–2.21, p=0.16). Certainly, multiple factors contribute to excessive bleeding in patients with and without coronary artery stents such as history of antiplatelet or anticoagulant use, timeline of DAPT interruption, baseline hematologic disease, and surgical procedure type. The authors attempted to overcome these potential confounders by including these and other relevant variables in our inverse probability of treatment weighting model. An inverse probability of treatment weighting model was chosen due to the limited power of the study and the inability to properly adjust for all variables. Furthermore, the primary aim of this study was not to assess the efficacy of AF in reducing bleeding in patients undergoing NCS as these studies already exist.
Timing of NCS and MACE
Timing of NCS after PCI with stenting is of critical importance to patients and providers as MACE is highest when NCS is performed within the first 180 days following DES implantation.2 Current ACC/AHA guidelines recommend delaying elective NCS for 30 days after BMS implantation and optimally 6 months after DES implantation (Class 1 recommendation, Level of Evidence B-NR).4 A previous study conducted by Hawn, et al 1 which consisted of 28,029 patients with coronary artery stents (47.6% DES, 52.4% BMS) undergoing NCS which demonstrated a stepwise decrease in MACE as time from PCI increased (11.6% at <6 weeks, 6.4% at 6 weeks to 6 months, 4.2% at 6–12 months, and 3.5% at >12 months).
Previous studies at our institution have provided data on the influence of stent type and timing of PCI to NCS on the outcome of MACE. Among 899 patients with existing BMS, the frequency of MACE was 10.5% when NCS was performed <30 days after PCI and 2.8% when NCS was performed >90 days after BMS implantation.31 In this cohort, bleeding events were not associated with time between PCI with BMS and NCS or with the use of antiplatelet therapy in the week before NCS. A study of 520 patients undergoing NCS with existing first-generation DES also revealed decreasing MACE rates of 6.4%, 5.7%, 5.9%, and 3.3% at 0–90, 91–180, 181–365, and 366–730 days between PCI and NCS respectively, although not statistically significant (p-value = 0.727).32 Regarding second generation DES, a retrospective cohort of 282 patients demonstrated a MACE frequency of 17.1% when NCS was performed 0–90 days after PCI, 10.0% when NCS was performed 91–180 days after PCI, 0.0% when NCS was performed 181–365 days after PCI, and 3.1% when NCS was performed ≥366 days after PCI.2 Although patients who were not on antiplatelet therapy had lower bleeding rates, there was no statistically significant reduction in excessive surgical bleeding between patients on aspirin monotherapy vs. DAPT.
In the current study, the optimal timing of AF administration following PCI with stenting was not analyzed due to a limited number of patients in the study sample. Further, large studies are needed to assess the safety of perioperative AF in patients at various time intervals.
Limitations
This study has limitations intrinsic to a large, retrospective cohort analysis with inaccuracies in chart review and missing data. While this study was completed at a large, tertiary referral center with >12,000 orthopedic surgical procedures performed annually during the study period, the low incidence of MACE, thrombotic complications, and excessive surgical bleeding made statistical analysis of the outcomes of interest difficult due to limited statistical power. The authors elected to use inverse probability weighting in an effort to maximize the power of the study and best report the associations between AF use and the outcomes of interest.
Certain components of MACE are influenced by patient symptomatology which is affected by a patient’s level of consciousness and analgesia following NCS.33 Given the absence of routine postoperative ECG and cardiac biomarkers, less clinically obvious MACE (i.e. NSTEMI) may have been underdiagnosed. Additionally, stent complications are known to be dynamic with a plateau between 1–5 years post-PCI as measured by a composite of cardiac death, MI, ischemia-driven target lesion revascularization, and target lesion failure.34 Our study does not have the longitudinal follow-up design to uniformly assess these data which may predispose to latency bias.
The decision to administer or withhold perioperative AF is complex and difficult to measure in a retrospective study. Despite statistically similar patient demographics, co-morbidities, and medications, Stents – AF patients had more preexisting BMSs, presented for NCS sooner, were more likely to be on DAPT at the time of surgery, and were more likely to undergo emergency NCS. Furthermore, variables that influence thrombotic risk including, but not limited to, AF dosing and renal function cannot be controlled in a retrospective analysis and require further investigation in this high risk patient population. Ultimately, the decision to administer AF was at the discretion of individual providers in this study.
While this study attempts to answer an important clinical question to help inform clinicians, the authors acknowledge that the potential for type-II error exists due to the underpowered nature of the study. While the study did not show a statistical difference between perioperative AF therapy and MACE in patients with coronary artery stents undergoing orthopedic surgery, the imprecise confidence interval impairs the ability to determine true association or lack thereof.
Conclusions
Patients with existing coronary artery stents undergoing NCS are a high-risk surgical population. The results of this study are inconclusive as to whether an association exists between perioperative AF therapy and MACE in patients with coronary artery stents undergoing orthopedic surgery. Further large prospective studies, powered to detect differences in MACE, are needed to assess the safety of perioperative AF in patients with existing coronary artery stents and to clarify the mechanism of perioperative MACE in this high-risk population.
Key Points:
Question:
Are patients with coronary artery stents who receive perioperative antifibrinolytic (AF) therapy for noncardiac, orthopedic surgery at an increased risk of major adverse cardiac events?
Findings:
In patients with coronary artery stents, there was no significant difference in MACE between patients who received and did not receive perioperative AF therapy; however, given the imprecise confidence interval, definitively concluding lack of association between AF use and MACE is not possible.
Meaning:
Further prospective studies are needed to assess the true safety of perioperative AF in patients with existing coronary artery stents and to clarify the mechanism of perioperative MACE in this high-risk population.
ACKNOWLEDGMENTS
The authors acknowledge Nicole Androsevick, Bradley Lewis, and Joshua Slesser for their contributions.
FUNDING
This project was supported by CTSA Grant Number UL1 TR002377 from the National Center for Advancing Translational Science (NCATS). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.
ABBREVIATIONS
- AF
antifibrinolytic
- ASA
American Society of Anesthesiologists
- DAPT
dual antiplatelet therapy
- DVT
deep venous thrombosis
- LOS
length of stay
- MACE
major adverse cardiac events
- NCS
noncardiac surgery
- PCI
percutaneous coronary intervention
- PE
pulmonary embolism
- TXA
tranexamic acid
Footnotes
Conflicts of Interest: None
Financial Disclosures: None
REFERENCES
- 1.Hawn MT, Graham LA, Richman JS et al. Risk of major adverse cardiac events following noncardiac surgery in patients with coronary stents. JAMA 2013;310:1462–72. [DOI] [PubMed] [Google Scholar]
- 2.Smith BB, Warner MA, Warner NS et al. Cardiac Risk of Noncardiac Surgery After Percutaneous Coronary Intervention With Second-Generation Drug-Eluting Stents. Anesth Analg 2019;128:621–8. [DOI] [PubMed] [Google Scholar]
- 3.Albaladejo P, Marret E, Samama CM et al. Non-cardiac surgery in patients with coronary stents: the RECO study. Heart 2011;97:1566–72. [DOI] [PubMed] [Google Scholar]
- 4.Levine GN, Bates ER, Bittl JA et al. 2016 ACC/AHA Guideline Focused Update on Duration of Dual Antiplatelet Therapy in Patients With Coronary Artery Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines: An Update of the 2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention, 2011 ACCF/AHA Guideline for Coronary Artery Bypass Graft Surgery, 2012 ACC/AHA/ACP/AATS/PCNA/SCAI/STS Guideline for the Diagnosis and Management of Patients With Stable Ischemic Heart Disease, 2013 ACCF/AHA Guideline for the Management of ST-Elevation Myocardial Infarction, 2014 AHA/ACC Guideline for the Management of Patients With Non-ST-Elevation Acute Coronary Syndromes, and 2014 ACC/AHA Guideline on Perioperative Cardiovascular Evaluation and Management of Patients Undergoing Noncardiac Surgery. Circulation 2016;134:e123–55. [DOI] [PubMed] [Google Scholar]
- 5.Carson JL, Guyatt G, Heddle NM et al. Clinical Practice Guidelines From the AABB: Red Blood Cell Transfusion Thresholds and Storage. JAMA 2016;316:2025–35. [DOI] [PubMed] [Google Scholar]
- 6.Lancet. Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. The Lancet 2010;376:23–32. [DOI] [PubMed] [Google Scholar]
- 7.Shakur H, Roberts I, Fawole B et al. Effect of early tranexamic acid administration on mortality, hysterectomy, and other morbidities in women with post-partum haemorrhage (WOMAN): an international, randomised, double-blind, placebo-controlled trial. The Lancet 2017;389:2105–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Alshryda S, Sarda P, Sukeik M et al. Tranexamic acid in total knee replacement. Journal of Bone and Joint Surgery 2011;93-B:1577–85. [DOI] [PubMed] [Google Scholar]
- 9.Lancet. Effects of tranexamic acid on death, disability, vascular occlusive events and other morbidities in patients with acute traumatic brain injury (CRASH-3): a randomised, placebo-controlled trial. The Lancet 2019;394:1713–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ping WD, Zhao QM, Sun HF, Lu HS, Li F. Role of tranexamic acid in nasal surgery: A systemic review and meta-analysis of randomized control trial. Medicine (Baltimore) 2019;98:e15202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mina SH, Garcia-Perdomo HA. Effectiveness of tranexamic acid for decreasing bleeding in prostate surgery: a systematic review and meta-analysis. Cent European J Urol 2018;71:72–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Winter SF, Santaguida C, Wong J, Fehlings MG. Systemic and Topical Use of Tranexamic Acid in Spinal Surgery: A Systematic Review. Global Spine J 2016;6:284–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ker K, Edwards P, Perel P, Shakur H, Roberts I. Effect of tranexamic acid on surgical bleeding: systematic review and cumulative meta-analysis. BMJ 2012;344:e3054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kaptein YE. Acute ST-elevation myocardial infarction due to in-stent thrombosis after administering tranexamic acid in a high cardiac risk patient. BMJ Case Rep 2019;12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Garg J, Pinnamaneni S, Aronow W, Ahmad H. ST Elevation Myocardial Infarction After Tranexamic Acid: First Reported Case in the United States. American Journal of Therapeutics 2014;21:e221–e4. [DOI] [PubMed] [Google Scholar]
- 16.Turgeon RD, Koshman SL, Youngson E et al. Association of Ticagrelor vs Clopidogrel With Major Adverse Coronary Events in Patients With Acute Coronary Syndrome Undergoing Percutaneous Coronary Intervention. JAMA Intern Med 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chute CG, Beck SA, Fisk TB, Mohr DN. The Enterprise Data Trust at Mayo Clinic: a semantically integrated warehouse of biomedical data. J Am Med Inform Assoc 2010;17:131–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Singh B, Singh A, Ahmed A et al. Derivation and validation of automated electronic search strategies to extract Charlson comorbidities from electronic medical records. Mayo Clin Proc 2012;87:817–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Schulman S, Angeras U, Bergqvist D et al. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in surgical patients. J Thromb Haemost 2010;8:202–4. [DOI] [PubMed] [Google Scholar]
- 20.Thygesen K, Alpert JS, Jaffe AS et al. Fourth Universal Definition of Myocardial Infarction (2018). J Am Coll Cardiol 2018;72:2231–64. [DOI] [PubMed] [Google Scholar]
- 21.McCaffrey DF, Griffin BA, Almirall D et al. A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Stat Med 2013;32:3388–414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Painter TW, McIlroy D, Myles PS, Leslie K. A survey of anaesthetists’ use of tranexamic acid in noncardiac surgery. Anaesth Intensive Care 2019;47:76–84. [DOI] [PubMed] [Google Scholar]
- 23.Bridges KH, Wilson SH. Acute Coronary Artery Thrombus After Tranexamic Acid During Total Shoulder Arthroplasty in a Patient With Coronary Stents: A Case Report. AA Pract 2018;10:212–4. [DOI] [PubMed] [Google Scholar]
- 24.Devereaux PJ, Eikelboom J. Insights into myocardial infarction after noncardiac surgery in patients with a prior coronary artery stent. Br J Anaesth 2016;116:584–6. [DOI] [PubMed] [Google Scholar]
- 25.Wasowicz M, Syed S, Wijeysundera DN et al. Effectiveness of platelet inhibition on major adverse cardiac events in non-cardiac surgery after percutaneous coronary intervention: a prospective cohort study. Br J Anaesth 2016;116:493–500. [DOI] [PubMed] [Google Scholar]
- 26.Levy JH, Koster A, Quinones QJ, Milling TJ, Key NS. Antifibrinolytic Therapy and Perioperative Considerations. Anesthesiology 2018;128:657–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Henry DA, Carless PA, Moxey AJ et al. Anti-fibrinolytic use for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 2011:CD001886. [DOI] [PubMed] [Google Scholar]
- 28.Howell SJ, Hoeks SE, West RM et al. Prospective observational cohort study of the association between antiplatelet therapy, bleeding and thrombosis in patients with coronary stents undergoing noncardiac surgery. Br J Anaesth 2019;122:170–9. [DOI] [PubMed] [Google Scholar]
- 29.Benenati S, Galli M, De Marzo V et al. Very Short vs. Long Dual Antiplatelet Therapy after Second Generation Drug-eluting Stents in 35,785 Patients undergoing Percutaneous Coronary Interventions: a Meta-analysis of Randomised Controlled Trials. Eur Heart J Cardiovasc Pharmacother 2020. [DOI] [PubMed] [Google Scholar]
- 30.Cho S, Kim JS, Kang TS et al. Long-Term Efficacy of Extended Dual Antiplatelet Therapy After Left Main Coronary Artery Bifurcation Stenting. Am J Cardiol 2020;125:320–7. [DOI] [PubMed] [Google Scholar]
- 31.Nuttall G, Brown M, Stombaugh J et al. Time and Cardiac Risk of Surgery after Bare-metal Stent Percutaneous Coronary Intervention. Anesthesiology 2008;109:588–95. [DOI] [PubMed] [Google Scholar]
- 32.Rabbitts J, Nuttall G, Brown M et al. Cardiac Risk of Noncardiac Surgery after Percutaneous Coronary Intervention with Drug-eluting Stents. Anesthesiology 2008;109:596–604. [DOI] [PubMed] [Google Scholar]
- 33.Devereaux P, Xavier D, Pogue J et al. Characteristics and Short-Term Prognosis of Perioperative Myocardial Infarction in Patients Undergoing Noncardiac Surgery. Annals of Internal Medicine 2011;154:523–8. [DOI] [PubMed] [Google Scholar]
- 34.Madhavan MV, Kirtane AJ, Redfors B et al. Stent-Related Adverse Events >1 Year After Percutaneous Coronary Intervention. J Am Coll Cardiol 2020;75:590–604. [DOI] [PubMed] [Google Scholar]