SPINE Volume 40, Number 9, pp 629-635 ©2015, Wolters Kluwer Health, Inc. All rights reserved.
CLINICAL CASE SERIES
Does Aspirin Administration Increase Perioperative Morbidity in Patients With Cardiac Stents Undergoing Spinal Surgery? Jason M. Cuellar, MD, PhD,* Anthony Petrizzo, DO,*† Ravi Vaswani, BS,‡ Jeffrey A. Goldstein, MD,*† and John A. Bendo, MD*†
Study Design. Cohort. Objective. To compare the perioperative morbidity of patients with cardiac stents after spine surgery who continue to take aspirin before and after the operation with a similar group of patients who preoperatively discontinued aspirin. Summary of Background Data. The preoperative discontinuation of anticoagulant therapy has been the standard of care for orthopedic surgical procedures. However, recent literature has demonstrated significant cardiac risk associated with aspirin withdrawal in patients with cardiac stents. Although it has recently been demonstrated that performing orthopedic surgery while continuing low-dose aspirin therapy seems to be safe, studies focused on spinal surgery have not yet been performed. Because of the risk of intraspinal bleeding and the serious consequences of subsequent epidural hematoma with associated spinal cord compression, spinal surgeons have been reluctant to operate on patients taking aspirin. Methods. This institutional review board–approved study included 200 patients. Preoperative parameters and postoperative outcome measures were analyzed for 100 patients who underwent spinal surgery after the discontinuation of anticoagulation therapy and 100 patients who continued to take daily aspirin through the perioperative period. The primary outcome measure was serious bleeding-related postoperative complications such as spinal epidural hematoma. The operative time, intraoperative estimated blood loss, hospital length of stay, transfusion of blood products, and 30-day hospital readmission rates were also recorded and compared. From the *Department of Orthopaedic Surgery and †Division of spine surgery, NYU Hospital for Joint Diseases, New York, NY; and ‡NYU School of Medicine, New York, NY. Acknowledgment date: March 26, 2014. Revision date: October 7, 2014. Acceptance date: October 29, 2014. The device(s)/drug(s) is/are FDA approved or approved by corresponding national agency for this indication. No funds were received in support of this work. Relevant financial activities outside the submitted work: board membership, consultancy, stocks, travel/accommodations/meeting expenses. Address correspondence and reprint requests to John A. Bendo, MD, Division of Spine Surgery, Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, 301 East 17th St, New York, NY 10003; E-mail: John.Bendo@ nyumc.org DOI: 10.1097/BRS.0000000000000695 Spine
Results. The patients who continued taking aspirin in the perioperative period had a shorter hospital length of stay on average (4.1 ± 2.7 vs. 6.2 ± 5.8; P < 0.005), as well as a reduced operative time (210 ± 136 vs. 266 ± 143; P < 0.01), whereas there was no significant difference in the estimated blood loss (642 ± 905 vs. 697 ± 1187), the amount of blood products transfused, overall intra- and postoperative complication rate (8% vs. 11%), or 30-day hospital readmission rate (5% vs. 5%). No clinically significant spinal epidural hematomas were observed in either of the study groups. Conclusion. The current study has observed no appreciable increase in bleeding-related complication rates in patients with cardiac stents undergoing spine surgery while continuing to take aspirin compared with patients who discontinued aspirin prior to surgery. Although very large studies will be needed to determine whether aspirin administration results in a small complication rate increase, the current study provides evidence that perioperative aspirin therapy is relatively safe in patients undergoing spinal surgery. Key words: aspirin, spine surgery, cardiac stent, perioperative anticoagulation, antiplatelet, spinal epidural hematoma, aspirin safety, bleeding complication. Level of Evidence: 2 Spine 2015;40:629–635
U
ntil recently, patients taking antiplatelet agents such as aspirin, clopidogrel, or nonsteroidal anti-inflammatory medications were required to discontinue all of these medications at least 5 days prior to spine surgery at our institution. There have been several reported episodes of postoperative death caused by thrombosis of indwelling cardiac drug-eluting or bare-metal cardiac stents in patients who preoperatively discontinued all anticoagulants, including aspirin. In addition, recent literature has demonstrated risks associated with aspirin withdrawal in patients with cardiac stents1–3 and potential benefits of its continuation.4,5 Surgeons are, therefore, beginning to operate more commonly on patients without preoperatively discontinuing aspirin administration, and in 2011, our institution initiated a policy mandating that aspirin be continued for all patients with indwelling cardiac stents, under the direction of their cardiologist. www.spinejournal.com
629
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. SPINEYIA1402.indd 629
3/30/15 7:18 PM
CLINICAL CASE SERIES It has recently been demonstrated that the practice of performing elective6 and urgent5,7 orthopedic surgery while continuing low-dose aspirin therapy seems to be safe, but no studies have been performed yet that focus on spine surgery patients.6 Because of the risk of intraspinal bleeding and the serious consequences of subsequent epidural hematoma with associated spinal cord compression,8 spinal surgeons have been reluctant to operate on patients taking aspirin.9 Therefore, the aim of this prospective cohort study was to determine whether this new practice adversely affects patients with cardiac stents undergoing spine surgery while taking aspirin. This cohort was compared with a similar but retrospective cohort of patients who underwent spinal surgery after antiplatelet therapy had been discontinued 5 days prior to surgery.
MATERIALS AND METHODS This was an institutional review board–approved study that was conducted on patients who underwent spine surgery between February 2009 and November 2013 at the NYU Langone Medical Center and the NYU Hospital for Joint Diseases. Two hundred patients were included in the study, which utilized 2 cohorts: a group of 100 patients who underwent spine surgery at least 5 days after discontinuing all antiplatelet medications (group 1) and a group of 100 patients who underwent spine surgery while still receiving perioperative aspirin therapy (group 2). All patients were identified by the use of a hospital billing code related to the retention of a previously inserted cardiac stent. Group 1 subjects were enrolled by including the first 100 consecutive eligible patients selected from a screening database, starting at the time of aspirin therapy policy change (November 2011) and working backward in time. Group 2 subjects were enrolled by including the first 100 consecutive eligible patients selected from a screening database, starting in 2011 and working prospectively. In total, 1200 patients with cardiac stents undergoing orthopedic surgery were screened to select 200 patients who underwent spine surgery and for whom information was adequately charted. Therefore, this study design enrolled all spine surgery patients with stents without exclusion and without matching surgical characteristics of one group to the other. The electronic medical record of each subject was reviewed for at least 60 days after spine surgery, and information was entered into a database, including patient's sex, age, body mass index, admission and discharge date, smoking status, medical and surgical history, number, type and location of cardiac stents, stent placement date, anticoagulant therapy usage, days prior to surgery the anticoagulant was discontinued if at all, current surgical diagnosis and procedure, operative time, estimated intraoperative blood loss, postoperative transfusion of blood products, hospitalization length of stay, 30-day hospital readmission and reason, and intraoperative and postoperative complications.
Inclusion and Exclusion Criteria All patients with cardiac stents undergoing spinal surgery were included. Only patients with known bleeding disorders were excluded from the study. 630
Does Aspirin Administration Increase Perioperative Morbidity • Cuellar et al
Statistical Analysis The primary goal of the present study was to investigate the effect of perioperative aspirin administration on the rate of clinically significant bleeding events, the most concerning of which would be spinal epidural hematoma (SEH). The overall rate of SEH after spine surgery has been reported to be 0.2%, with individual studies reporting rates as high as 1.0%.10 Therefore, we would expect a rate of postoperative SEH to be within the range of 0.2% and 1%. Post hoc sample size calculation using G-Power software11 (Dusseldorf, Germany) with a χ2 test whereby df = 1, α = 0.05, and β ≥ 0.8, a sample size of 100 per group, will detect a change from 0.2% to 4.5% (w = 0.202). Student t test was used to compare the average age, body mass index, mean hospital length of stay, mean operative time, and mean number of postoperative units of red blood cells, fresh frozen plasma, or platelets transfused in the 2 patient groups. Chi-square analysis was used to compare the proportion of female to male patients, patients with and without diabetes, patients with and without a cancer history, smokers, and nonsmokers between the 2 patient groups. A P value of less than 0.05 was predetermined to be a statistically significant difference. SPSS 19 (IBM, NY) and AcaStat and StatCalc software (Leesburg, VA) were used for statistical analyses.
Source of Funding The study was paid for completely by departmental funds.
RESULTS The preoperative parameters and postoperative outcome measures were analyzed for 100 patients who underwent spinal surgery after the discontinuation of anticoagulation therapy and 100 patients who continued to take aspirin daily through the perioperative period without hiatus. The preoperative clinical parameters recorded are presented in Table 1 and demonstrate that, with the exception of smoking status, there was minimal difference between the 2 groups of patients. The breakdown of surgical procedure performed is presented in Table 2. There was a greater proportion of cervical to lumbar procedures performed in group 1 patients (Table 2). In addition, there were a greater proportion of lumbar procedures involving the fusion of 3 to 4 segments in group 1. Other differences were not statistically significant. The data related to the type of cardiac stent, the anticoagulant medication taken by the patient, and when the medication was discontinued are presented in Table 3. A greater proportion of patients in group 2 were on aspirin alone, whereas more patients in group 1 were taking clopidogrel or warfarin alone. A greater proportion of patients in group 2 were taking 81 mg of aspirin. In addition, the aspirin dosage was unknown for significantly more patients in group 1. Intra- and postoperative parameters such as operative time, estimated blood loss, hospital length of stay, and postoperative transfusion of blood products are presented in Table 4. The patients who continued taking aspirin in the perioperative period (group 2) had a shorter hospital length of stay on average, as well as a reduced operative time, whereas there
www.spinejournal.com
May 2015
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. SPINEYIA1402.indd 630
3/30/15 7:18 PM
CLINICAL CASE SERIES
Does Aspirin Administration Increase Perioperative Morbidity • Cuellar et al
TABLE 1. Preoperative Parameters Group
P
1 (Off ASA)
2 (On ASA)
100
100
Age, yr
68 ± 9
69 ± 9
NS
Female:male
25:75
21:79
NS
Body mass index
30 ± 5.2
29.7 ± 5.2
NS
Diabetic patients
33
31
NS
Cancer history
20
15
NS
Smoking
22
5
CLINICAL CASE SERIES
Does Aspirin Administration Increase Perioperative Morbidity in Patients With Cardiac Stents Undergoing Spinal Surgery? Jason M. Cuellar, MD, PhD,* Anthony Petrizzo, DO,*† Ravi Vaswani, BS,‡ Jeffrey A. Goldstein, MD,*† and John A. Bendo, MD*†
Study Design. Cohort. Objective. To compare the perioperative morbidity of patients with cardiac stents after spine surgery who continue to take aspirin before and after the operation with a similar group of patients who preoperatively discontinued aspirin. Summary of Background Data. The preoperative discontinuation of anticoagulant therapy has been the standard of care for orthopedic surgical procedures. However, recent literature has demonstrated significant cardiac risk associated with aspirin withdrawal in patients with cardiac stents. Although it has recently been demonstrated that performing orthopedic surgery while continuing low-dose aspirin therapy seems to be safe, studies focused on spinal surgery have not yet been performed. Because of the risk of intraspinal bleeding and the serious consequences of subsequent epidural hematoma with associated spinal cord compression, spinal surgeons have been reluctant to operate on patients taking aspirin. Methods. This institutional review board–approved study included 200 patients. Preoperative parameters and postoperative outcome measures were analyzed for 100 patients who underwent spinal surgery after the discontinuation of anticoagulation therapy and 100 patients who continued to take daily aspirin through the perioperative period. The primary outcome measure was serious bleeding-related postoperative complications such as spinal epidural hematoma. The operative time, intraoperative estimated blood loss, hospital length of stay, transfusion of blood products, and 30-day hospital readmission rates were also recorded and compared. From the *Department of Orthopaedic Surgery and †Division of spine surgery, NYU Hospital for Joint Diseases, New York, NY; and ‡NYU School of Medicine, New York, NY. Acknowledgment date: March 26, 2014. Revision date: October 7, 2014. Acceptance date: October 29, 2014. The device(s)/drug(s) is/are FDA approved or approved by corresponding national agency for this indication. No funds were received in support of this work. Relevant financial activities outside the submitted work: board membership, consultancy, stocks, travel/accommodations/meeting expenses. Address correspondence and reprint requests to John A. Bendo, MD, Division of Spine Surgery, Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, 301 East 17th St, New York, NY 10003; E-mail: John.Bendo@ nyumc.org DOI: 10.1097/BRS.0000000000000695 Spine
Results. The patients who continued taking aspirin in the perioperative period had a shorter hospital length of stay on average (4.1 ± 2.7 vs. 6.2 ± 5.8; P < 0.005), as well as a reduced operative time (210 ± 136 vs. 266 ± 143; P < 0.01), whereas there was no significant difference in the estimated blood loss (642 ± 905 vs. 697 ± 1187), the amount of blood products transfused, overall intra- and postoperative complication rate (8% vs. 11%), or 30-day hospital readmission rate (5% vs. 5%). No clinically significant spinal epidural hematomas were observed in either of the study groups. Conclusion. The current study has observed no appreciable increase in bleeding-related complication rates in patients with cardiac stents undergoing spine surgery while continuing to take aspirin compared with patients who discontinued aspirin prior to surgery. Although very large studies will be needed to determine whether aspirin administration results in a small complication rate increase, the current study provides evidence that perioperative aspirin therapy is relatively safe in patients undergoing spinal surgery. Key words: aspirin, spine surgery, cardiac stent, perioperative anticoagulation, antiplatelet, spinal epidural hematoma, aspirin safety, bleeding complication. Level of Evidence: 2 Spine 2015;40:629–635
U
ntil recently, patients taking antiplatelet agents such as aspirin, clopidogrel, or nonsteroidal anti-inflammatory medications were required to discontinue all of these medications at least 5 days prior to spine surgery at our institution. There have been several reported episodes of postoperative death caused by thrombosis of indwelling cardiac drug-eluting or bare-metal cardiac stents in patients who preoperatively discontinued all anticoagulants, including aspirin. In addition, recent literature has demonstrated risks associated with aspirin withdrawal in patients with cardiac stents1–3 and potential benefits of its continuation.4,5 Surgeons are, therefore, beginning to operate more commonly on patients without preoperatively discontinuing aspirin administration, and in 2011, our institution initiated a policy mandating that aspirin be continued for all patients with indwelling cardiac stents, under the direction of their cardiologist. www.spinejournal.com
629
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. SPINEYIA1402.indd 629
3/30/15 7:18 PM
CLINICAL CASE SERIES It has recently been demonstrated that the practice of performing elective6 and urgent5,7 orthopedic surgery while continuing low-dose aspirin therapy seems to be safe, but no studies have been performed yet that focus on spine surgery patients.6 Because of the risk of intraspinal bleeding and the serious consequences of subsequent epidural hematoma with associated spinal cord compression,8 spinal surgeons have been reluctant to operate on patients taking aspirin.9 Therefore, the aim of this prospective cohort study was to determine whether this new practice adversely affects patients with cardiac stents undergoing spine surgery while taking aspirin. This cohort was compared with a similar but retrospective cohort of patients who underwent spinal surgery after antiplatelet therapy had been discontinued 5 days prior to surgery.
MATERIALS AND METHODS This was an institutional review board–approved study that was conducted on patients who underwent spine surgery between February 2009 and November 2013 at the NYU Langone Medical Center and the NYU Hospital for Joint Diseases. Two hundred patients were included in the study, which utilized 2 cohorts: a group of 100 patients who underwent spine surgery at least 5 days after discontinuing all antiplatelet medications (group 1) and a group of 100 patients who underwent spine surgery while still receiving perioperative aspirin therapy (group 2). All patients were identified by the use of a hospital billing code related to the retention of a previously inserted cardiac stent. Group 1 subjects were enrolled by including the first 100 consecutive eligible patients selected from a screening database, starting at the time of aspirin therapy policy change (November 2011) and working backward in time. Group 2 subjects were enrolled by including the first 100 consecutive eligible patients selected from a screening database, starting in 2011 and working prospectively. In total, 1200 patients with cardiac stents undergoing orthopedic surgery were screened to select 200 patients who underwent spine surgery and for whom information was adequately charted. Therefore, this study design enrolled all spine surgery patients with stents without exclusion and without matching surgical characteristics of one group to the other. The electronic medical record of each subject was reviewed for at least 60 days after spine surgery, and information was entered into a database, including patient's sex, age, body mass index, admission and discharge date, smoking status, medical and surgical history, number, type and location of cardiac stents, stent placement date, anticoagulant therapy usage, days prior to surgery the anticoagulant was discontinued if at all, current surgical diagnosis and procedure, operative time, estimated intraoperative blood loss, postoperative transfusion of blood products, hospitalization length of stay, 30-day hospital readmission and reason, and intraoperative and postoperative complications.
Inclusion and Exclusion Criteria All patients with cardiac stents undergoing spinal surgery were included. Only patients with known bleeding disorders were excluded from the study. 630
Does Aspirin Administration Increase Perioperative Morbidity • Cuellar et al
Statistical Analysis The primary goal of the present study was to investigate the effect of perioperative aspirin administration on the rate of clinically significant bleeding events, the most concerning of which would be spinal epidural hematoma (SEH). The overall rate of SEH after spine surgery has been reported to be 0.2%, with individual studies reporting rates as high as 1.0%.10 Therefore, we would expect a rate of postoperative SEH to be within the range of 0.2% and 1%. Post hoc sample size calculation using G-Power software11 (Dusseldorf, Germany) with a χ2 test whereby df = 1, α = 0.05, and β ≥ 0.8, a sample size of 100 per group, will detect a change from 0.2% to 4.5% (w = 0.202). Student t test was used to compare the average age, body mass index, mean hospital length of stay, mean operative time, and mean number of postoperative units of red blood cells, fresh frozen plasma, or platelets transfused in the 2 patient groups. Chi-square analysis was used to compare the proportion of female to male patients, patients with and without diabetes, patients with and without a cancer history, smokers, and nonsmokers between the 2 patient groups. A P value of less than 0.05 was predetermined to be a statistically significant difference. SPSS 19 (IBM, NY) and AcaStat and StatCalc software (Leesburg, VA) were used for statistical analyses.
Source of Funding The study was paid for completely by departmental funds.
RESULTS The preoperative parameters and postoperative outcome measures were analyzed for 100 patients who underwent spinal surgery after the discontinuation of anticoagulation therapy and 100 patients who continued to take aspirin daily through the perioperative period without hiatus. The preoperative clinical parameters recorded are presented in Table 1 and demonstrate that, with the exception of smoking status, there was minimal difference between the 2 groups of patients. The breakdown of surgical procedure performed is presented in Table 2. There was a greater proportion of cervical to lumbar procedures performed in group 1 patients (Table 2). In addition, there were a greater proportion of lumbar procedures involving the fusion of 3 to 4 segments in group 1. Other differences were not statistically significant. The data related to the type of cardiac stent, the anticoagulant medication taken by the patient, and when the medication was discontinued are presented in Table 3. A greater proportion of patients in group 2 were on aspirin alone, whereas more patients in group 1 were taking clopidogrel or warfarin alone. A greater proportion of patients in group 2 were taking 81 mg of aspirin. In addition, the aspirin dosage was unknown for significantly more patients in group 1. Intra- and postoperative parameters such as operative time, estimated blood loss, hospital length of stay, and postoperative transfusion of blood products are presented in Table 4. The patients who continued taking aspirin in the perioperative period (group 2) had a shorter hospital length of stay on average, as well as a reduced operative time, whereas there
www.spinejournal.com
May 2015
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. SPINEYIA1402.indd 630
3/30/15 7:18 PM
CLINICAL CASE SERIES
Does Aspirin Administration Increase Perioperative Morbidity • Cuellar et al
TABLE 1. Preoperative Parameters Group
P
1 (Off ASA)
2 (On ASA)
100
100
Age, yr
68 ± 9
69 ± 9
NS
Female:male
25:75
21:79
NS
Body mass index
30 ± 5.2
29.7 ± 5.2
NS
Diabetic patients
33
31
NS
Cancer history
20
15
NS
Smoking
22
5
