Deep Vein Thrombosis and Pulmonary Embolism After Spine Surgery: Incidence and Patient Risk Factors.

Abstract

Anticoagulation after spine surgery confers the unique risk of epidural hematoma. We sought to determine the incidence of and patient risk factors for deep vein thrombosis (DVT) and pulmonary embolism (PE) after spine surgery. We retrospectively reviewed the charts of 1485 patients who had spine surgery at a single tertiary-care center between 2002 and 2009. DVT and PE incidence were recorded along with pertinent patient history information. Univariate and multivariate analyses were performed on the data. VTE incidence was 1.1% (DVTs, 0.7%; PEs, 0.4%). Univariate analysis demonstrated that VTEs had 9 positive risk factors: active malignancy, prior DVT or PE, estrogen replacement therapy, discharge to a rehabilitation facility, hypertension, major depressive disorder, renal disease, congestive heart failure, and benign prostatic hyperplasia (P<.05). Multivariate analysis demonstrated 4 independent risk factors: prior DVT or PE, estrogen replacement therapy, discharge to a rehabilitation facility, and major depressive disorder (P>.05). Surgeons with an improved understanding of VTE after spine surgery can balance the risks and benefits of postoperative anticoagulation.

Full text

www.amjorthopedics.com June 2013 The American Journal of Orthopedics
®
267
An Original Study
Deep Vein Thrombosis and
Pulmonary Embolism After Spine Surgery:
Incidence and Patient Risk Factors
Leah M. Schulte, MD, Joseph R. O’Brien, MD, MPH, Matthew C. Bean, BS, Todd P. Pierce, BS,
Warren D. Yu, MD, and Clifton Meals, MD
V
enous thromboembolic events (VTEs), including deep
vein thrombosis (DVT) and pulmonary embolism
(PE), are serious and potentially life-threatening com-
plications of surgical procedures. DVTs and PEs are uncommon
after spine surgery. A recent meta-analysis found a DVT rate of
1.09% and a PE rate of 0.06%.
1
Management of postoperative
DVTs and PEs is difcult in spine surgery patients because of
the risk for spinal epidural hematoma (SEH).
SEH is a severe but rare complication. Its incidence after
spine surgery is 0.1% to 3%.
2
High-dose anticoagulation in
the postoperative spine surgery patient with a PE may increase
the risk for SEH. Thus, any treatment plan for a spine surgery
patient with a postoperative VTE must balance the risk for SEH
with the benet of anticoagulation. Most treatment algorithms
for DVT and PE call for high-dose boluses and therapeutic doses
of high-molecular-weight heparinoids. Bleeding complications
of anticoagulant use have a predictable dose-response curve.
3
Thus, management of a postoperative DVT or PE may pose an
SEH risk signicantly higher than that posed by routine low-
dose chemoprophylaxis. It is important to note that routine
use of high-molecular-weight heparinoids has been shown to
be safe in postoperative spine surgery patients.
4
Despite the regulatory
5
and specialty society
6
focus thrombo-
embolic events, there are no clear guidelines for identifying good
spine surgery candidates for chemoprophylaxis—because of
increased risk, whether from surgical factors or patient factors—
or for managing postoperative DVTs and PEs in spine surgery
patients. The American College of Chest Physicians guidelines
7
(Table I) are unclear on several points. In particular, patient
risk factors (ie, factors that put patients at high risk for VTE)
have not been studied in detail. Although VTE is rare in spine
surgery patients, signicant research is warranted to more clearly
delineate these issues.
We conducted a retrospective study to determine the spine
surgery patient risk factors for VTEs at an elective spine surgery
practice. Here we report on the incidence of and patient risk
factors for DVT and PE in a general, mostly nontraumatic spine
practice. All types of surgeries were included, and relative risk
(RR) factors were calculated. Our overall goal was to expand
the knowledge base regarding the type of patients at increased
risk for postoperative DVT and PE.
Materials and Methods
After obtaining institutional review board approval, we re-
viewed all 1485 cases of spine surgery performed at our ter-
tiary-care center between 2002 and 2009. The surgeries were
performed by 2 spine fellowship-trained surgeons (JRB and
WDY). These surgeries addressed the cervical, lumbar, and
Abstract
Anticoagulation after spine surgery confers the
unique risk of epidural hematoma.
We sought to determine the incidence of and
patient risk factors for deep vein thrombosis
(DVT) and pulmonary embolism (PE) after spine
surgery. We retrospectively reviewed the charts
of 1485 patients who had spine surgery at a
single tertiary-care center between 2002 and
2009. DVT and PE incidence were recorded
along with pertinent patient history informa-
tion. Univariate and multivariate analyses were
performed on the data.
VTE incidence was 1.1% (DVTs, 0.7%; PEs,
0.4%). Univariate analysis demonstrated that
VTEs had 9 positive risk factors: active malig-
nancy, prior DVT or PE, estrogen replacement
therapy, discharge to a rehabilitation facility,
hypertension, major depressive disorder, renal
disease, congestive heart failure, and benign
prostatic hyperplasia (P<.05). Multivariate analy-
sis demonstrated 4 independent risk factors:
prior DVT or PE, estrogen replacement therapy,
discharge to a rehabilitation facility, and major
depressive disorder (P>.05).
Surgeons with an improved understanding of
VTE after spine surgery can balance the risks
and benets of postoperative anticoagulation.
Authors’ Disclosure Statement: The authors report no actual or potential conict of interest in relation to this article.
AJO
DO NOT COPY

268 TheAmericanJournalofOrthopedics
®
June 2013 www.amjorthopedics.com
DVT and PE After Spine Surgery
L. M. Schulte et al
thoracic regions and used anterior and posterior approaches
(Table II). Charts were reviewed for pertinent data, includ-
ing age, body mass index (BMI), past medical history, medi-
cations, smoking history, specic spine procedure, type of
DVT prophylaxis, length of hospital stay, DVT or PE diagnosis,
and admission to a rehabilitation facility. At our institution,
routine prophylaxis involves having patients wear thigh-high
compression stockings with sequential compression devices
(SCDs) during and after surgery. We investigated all VTEs that
had been the subject of clinical suspicion during the hospital
stay or at routine follow-ups; there was no routine screening.
All VTEs were diagnosed with ultrasonography, ventilation-
perfusion quotient scan, or chest computed tomography (CT)
with contrast. DVT incidence and PE incidence were calculated
for this population group. Possible risk factors—including age,
sex, obesity, history of thromboembolism, active malignancy,
other comorbidities, prolonged hospital stay or rehabilitation,
and smoking—were evaluated for their association to VTEs.
Obesity was dened as BMI higher than 30 and active malig-
nancy was dened as a cancer undergoing treatment.
Descriptive characteristics for the cohort were compiled.
Means and standard deviations were calculated. Logistic
regression analyses were performed to determine the as-
sociation of comorbidities and other factors with DVT and
PE occurrence. Binomial univariate logistic regression was
used to analyze age, sex, BMI, prior VTE, comorbidities, and
smoking status for an association with VTE occurrence. Vari-
ables found to be potentially predictive of the outcome vari-
able from the univariate model (P<.2) were included in the
multivariate logistic regression models. Multivariate models
were used to assess the statistical signicance of each individual
variable while adjusting for other covariates. Variables found to
be nonsignicant (P>.05) were removed from the multivariate
model in sequential fashion starting with the variable with the
highest P. For each variable, RRs and 95% condence intervals
were calculated. SPSS Version 18.0 (SPSS, IBM, Chicago, Il-
linois) was used for all statistics.
Results
Of the 1485 patients who had spine surgery between 2002 and
2009 (Table II), 817 were women (55%) and 668 men (45%).
Table I. American College of Chest Physicians Recommendations for Spine Surgery,
7
February 2012
Patient Category Details Recommendations
Spine surgery patients — Mechanical prophylaxis, LDUH, or LMWH
High-risk patients undergoing
spine surgery
Malignant disease
Combined anterior-posterior approaches
Pharmacologic prophylaxis after adequate hemostasis
is achieved
Major trauma patients Spinal cord injury
Spine fracture requiring surgery
Pharmacologic plus mechanical prophylaxis
Contraindication to pharmacologic prophylaxis Mechanical prophylaxis
Add LDUH or LMWH when risk for bleeding normalizes
Do not use inferior vena cava filter for VTE prevention
Abbreviations: LDUH, low-dose unfractionated heparin; LMWH, low-molecular-weight heparin; VTE, venous thromboembolism.
Table II. Spine Procedures Performed
Type of Procedure n
Cervical 352
Posterior cervical decompression 3
Anterior cervical fusion/arthroplasty, 1 or 2 levels 216
Anterior cervical fusion, >2 levels 65
Posterior cervical fusion 53
Combined anterior/posterior cervical fusion 15
Thoracic 24
Thoracic fusion by thoracotomy 15
Posterior thoracic fusion 9
Lumbar 970
Diskectomy/1-level lumbar decompression 329
Multilevel lumbar decompression 25
Anterior lumbar fusion 35
Posterior lumbar fusion, 1 or 2 levels 388
Posterior lumbar fusion, 3-5 levels 146
Posterior thoracolumbar fusion, >5 levels 41
Lateral lumbar fusion 3
Combined anterior/posterior lumbar fusion 3
Other 139
Interspinous process insertion 67
Kyphoplasty 31
Miscellaneous 41
Total 1485
AJO
DO NOT COPY

DVT and PE After Spine Surgery
www.amjorthopedics.com June 2013 The American Journal of Orthopedics
®
269
L. M. Schulte et al
Mean (SD) age was 53 (14) years (range, 16-103 years). Sixteen
patients (1.1%) had at least 1 VTE. Eleven (0.7%) had a DVT
and 6 (0.4%) had a PE; 1 patient (0.1%) was formally diagnosed
with both. No patient died from a VTE.
Univariate RR calculations demonstrated that prior DVT or
PE, active malignancy, estrogen replacement therapy (ERT),
discharge to a rehabilitation facility, major depressive disorder
(MDD), hypertension, renal disease, congestive heart failure,
and benign prostatic hyperplasia were risk factors for postop-
erative VTEs (P≤.05) (Table III). Smoking, multiple procedures
within 30 days, obesity, sex, gastroesophageal reux disease,
hyperlipidemia, and sleep apnea were not signicant risk fac-
tors (P>.05). Multivariate analysis demonstrated that prior DVT
or PE, discharge to a rehabilitation facility, ERT, and MDD were
independent risk factors (P>.05) (Table IV).
Discussion
Venous stasis, endothelial injury, and hypercoagulable states
all contribute to the postoperative development of DVT. It is
thought that in total knee and hip arthroplasty, direct injury to
the large vessels in the leg contributes to the high rates of DVT
and PE in the absence of postoperative chemoprophylaxis.
8
In major spine surgery, prolonged immobilization and bone
work may contribute to increased risk for postoperative DVT,
but direct endothelial injury generally does not. Anterior spine
surgery may be an exception. Other investigators have analyzed
surgical risk factors that may play a role in DVT and PE after
spine surgery. A recent review of 14,932 cases found routine
use of chemoprophylaxis was not necessary in posterior-only
spine surgery.
8
It may be considered in anterior spine surgery,
which involves manipulation of the great vessels. Aside from
surgical risk factors, patient risk factors may also contribute to
risk for postoperative VTE.
Use of chemoprophylaxis in spine surgery is controversial.
In elective spine surgery, DVT prophylaxis must be balanced
against the risks for postoperative bleeding and epidural hema-
toma.
9
Epidural hematoma is a unique complication of spine
surgery.
8
The incidence of symptomatic epidural hematoma
causing neurologic compromise requiring surgery is between
0.1% and 3%.
8
The result can be a permanent neurologic decit,
such as paralysis.
2,8,10
Despite the apprehension toward chemi-
cal anticoagulation, Awad and colleagues
4
found no increased
risk for developing an epidural hematoma with use of chemical
prophylaxis in a retrospective review of 14,932 spine surgery
patients.
9
However, symptomatic epidural hematoma has been
reported with therapeutic-dose chemical anticoagulation.
10
Blood loss and wound complications are also associated with
chemical prophylaxis.
11
Mechanical prophylaxis methods, spe-
cically use of SCDs alone, have effectively reduced DVT and
PE incidence.
12
Experience with VTEs has led some adult spine
deformity surgeons to routinely use chemoprophylaxis after
surgery. However, most spine surgeons use only mechanical
prophylaxis methods in elective spine surgery.
13,14
It is clear
that more work is needed to delineate which surgical and
patient factors increase the risk for VTE in the postoperative
spine surgery patient.
In this study, we identied some patient risk factors for
VTE after spine surgery. The statistically signicant risk factors
for DVT were prior DVT or PE, ERT, and MDD. Discharge to
a rehabilitation facility was a statistically signicant variable.
Other studies have implicated hypercoagulable states as a risk
factor for DVT,
4
so it is not surprising that prior DVT or PE was
a risk factor in this study. It is important to note that our center
does not treat patients with metastatic cancer on a regular basis.
Thus, this signicant risk factor may be excluded from our
results largely because of institutional selection bias. ERT has
also been implicated as a risk factor for DVT and PE in other
studies.
15
Discharge to a rehabilitation facility as a risk factor
is notable but may be more of a reection on the size of the
surgical procedure than on patient characteristics. For instance,
larger spinal fusion procedures with slow patient mobilization
may have led to prolonged recumbency and the need for more
rehabilitation. Thus, this risk factor may be a reection of
venous stasis. MDD as a risk factor may also reect prolonged
recumbency. It may also reect the use of various antidepres-
sants. Further investigation along those lines is warranted.
This study had a few limitations. It was retrospective and
did not use routine screening to identify patients with VTE.
Thus, the 1% incidence likely underestimates the actual in-
Table III. Univariate Analysis of Patient Risk Factors
Risk Factor RR CI P<
Active malignancy 8.9 2.1-37.0 .001
Prior DVT or PE 8.3 1.3-46.3 .018
Estrogen replacement therapy 6.2 1.4-26.1 .01
Discharge to rehabilitation facility 20.8 1.3-5.2 .018
Hypertension 7.7 1.2-48.2 .023
Major depressive disorder 9.5 2.5-35.9 .001
Renal disease 10.2 1.7-55.5 .007
Congestive heart failure 10.2 1.7-55.5 .007
Benign prostatic hypertrophy 6.6 1.1-37.7 .041
Abbreviations: CI, confidence interval; DVT, deep vein thrombosis; PE, pulmo-
nary embolism RR; relative risk.
Table IV. Multivariate Analysis of Patient Risk Factors
Risk Factor RR CI P<
Prior DVT or PE 2.4 1.8-60.7 .026
Estrogen replacement therapy 3.1 3.5-128.8 .007
Discharge to rehabilitation facility 2.9 2.4-40.3 .002
Major depressive disorder 1.7 1.1-24.7 .040
Abbreviations: CI, confidence interval; DVT, deep vein thrombosis; PE, pulmo-
nary embolism; RR, relative risk.
AJO
DO NOT COPY

DVT and PE After Spine Surgery
cidence that might be found with prospective screening of
postoperative spine surgery patients. Furthermore, the hospital
patient population used in this study did not include many
patients with spinal trauma or metastatic spine tumors. Thus,
our conclusions are limited by institutional selection bias.
More work at major spinal trauma centers and cancer centers
is needed to delineate patient risk factors for VTE.
We have described patient risk factors for postoperative
VTEs. Further work is needed to more clearly delineate patient
risk factors for VTEs after spine surgery.
Dr. Schulte and Dr. O’Brien are Assistant Professors, Mr. Bean and
Mr. Pierce are Medical Students, Dr. Yu is Associate Professor, and
Dr. Meals is Resident Surgeon, Department of Orthopaedic Surgery,
George Washington University, Washington, District of Columbia.
Address correspondence to: Leah M. Schulte, MD, Department of
Orthopaedic Surgery, George Washington University, 2150 Pennsyl-
vania Ave NW, 7th Floor, Washington, DC 20037 (tel, 216-287-114;
fax, 202-741-3313; e-mail, leahmschulte@gmail.com).
Am J Orthop. 2013;42(6):267-270. Copyright Frontline Medical Com-
munications Inc. 2013. All rights reserved.
References
1. Sansone JM, del Rio AM, Anderson PA. The prevalence of and specic
risk factors for venous thromboembolic disease following elective spine
surgery. J Bone Joint Surg Am. 2010;92(2):304-313.
2. Kou J, Fischgrund J, Biddinger A, Herkowitz A. Risks factors for spinal
epidural hematoma after spinal surgery. Spine. 2002;27(15):1670-1673.
3. Kasapis C, Gurm HS, Chetcuti SJ, et al. Dening the optimal degree
of heparin anticoagulation for peripheral vascular interventions: insight
from a large, regional, multicenter registry. Circ Cardiovasc Interv.
2010;3(6):593-601.
4. Awad JN, Kebaish KM, Donigan J, et al. Analysis of the risk factors for
the development of post-operative spinal epidural hematoma. J Bone
Joint Surg Br. 2005;87(9):1248-1252.
5. National Quality Forum. National Voluntary Consensus Standards for
Prevention and Care of Venous Thromboembolism: Additional Perfor-
mance Measures. http://www.qualityforum.org/Publications/2008/10/
National_Voluntary_Consensus_Standards_for_Prevention_and_Care_
of_Venous_Thromboembolism__Additional_Performance_Measures.
aspx. Published 2008. Accessed April 5, 2010.
6. American Academy of Orthopaedic Surgeons. Deep Vein Thrombosis.
http://www5.aaos.org/dvt/main.cfm. Accessed April 5, 2010.
7. Guyatt GH, Akl EA, Crowther M, Gutterman DD, Schuünemann HJ;
American College of Chest Physicians Antithrombotic Therapy and
Prevention of Thrombosis Panel. Executive summary: Antithrombotic
Therapy and Prevention of Thrombosis, 9th ed: American College of
Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest.
2012;141(2 suppl):7S-47S.
8. Shepard A, Mills C. Fatal pulmonary embolism following hip and knee re-
placement. A study of 2153 cases using routine mechanical prophylaxis
and selective chemoprophylaxis. Hip Int. 2006;16(1):53-56.
9. Rodriguez y Baena R, Gaetani P, Tancioni F, Tartara F. Spinal epidural
hematoma during anticoagulant therapy. A case report and review of the
literature. J Neurosurg Sci. 1995;39(1):87-94.
10. Morse K, Weight M, Molinari R. Extensive postoperative epidural
hematoma after full anticoagulation: a case report and full review of the
literature. J Spinal Cord Med. 2007;30(3):282-287.
11. Glotzbecker MP, Bono CM, Wood KB, Harris MB. Thromboembolic dis-
ease in spinal surgery: a systematic review. Spine. 2009;34(3):291-303.
12. Epstein NM. Efcacy of pneumatic compression stocking prophylaxis in
the prevention of deep venous thrombosis and pulmonuary embolism
following 139 lumbar laminectomies with instrumented fusions. J Spinal
Disord Tech. 2006;19(1):28-31.
13. Ploumis A, Ponnappan RK, Sarbello J, et al. Thromboprophylaxis in trau-
matic and elective spinal surgery: analysis of questionnaire response and
current practice of spine trauma surgeons. Spine. 2010;35(3):323-329.
14. Glotzbecker MP, Bono CM, Harris MB, Brick G, Heary RF, Wood KB.
Surgeon practices regarding postoperative thromboembolic prophylaxis
after high-risk spinal surgery. Spine. 2008;33(26):2915-2921.
15. Lowe G, Woodward M, Vessey M, Rumley A, Gough P, Daly E. Throm-
botic variables and risk of idiopathic venous thromboembolism in women
aged 45-64 years. Relationships to hormone replacement therapy.
Thromb Haemost. 2000;83(4):530-535.
Join the discussion.
Share your opinion.
Leave your comment!
www.amjorthopedics.com/b log/default.aspx
AJO
DO NOT COPY