Abstract
Background:
Knee arthroscopy is the most commonly performed orthopaedic procedure in the United States and is usually considered to be a low-risk procedure. The purposes of this study were to describe the incidence of symptomatic deep venous thrombosis, symptomatic pulmonary embolism, and mortality after elective knee arthroscopy performed without thromboembolic prophylaxis, as well as to investigate the association of age, sex, procedure type, and oral contraceptive use with the odds of developing a venous thromboembolism.
Methods:
A retrospective cohort study of elective arthroscopic knee procedures during a twenty-seven-month period (January 1, 2006, through March 31, 2008) was performed with use of the administrative database of a large health maintenance organization. Use of ICD-9-CM (International Classification of Diseases, Ninth Revision, Clinical Modification) procedure codes for arthroscopic surgery identified 21,794 arthroscopic knee procedures. The occurrence of a symptomatic deep venous thrombosis or pulmonary embolism within ninety days after surgery was identified by reviewing administrative and electronic medical record data for inpatient, outpatient, urgent care, and emergency encounters. Mortality and the cause of death were captured with use of electronic medical records, Social Security Administration Death Master Files, and county death certificates. Patient charts were reviewed for confirmation of the deep venous thrombosis, pulmonary embolism, or death. Patients who had a history of a venous thromboembolism or who had received anticoagulation therapy within fourteen days prior to the index surgery were excluded.
Results:
The study cohort comprised 20,770 patients who met the inclusion criteria. Fifty-one patients (0.25%; 95% confidence interval, 0.18% to 0.31%) developed a deep venous thrombosis, and thirty-five (0.17%; 95% confidence interval, 0.11% to 0.22%) developed a pulmonary embolism. The incidence of venous thromboembolism was higher in patients who were fifty years of age or older (0.51% compared with 0.34% in younger patients), and the incidence in female patients was higher if they had been prescribed oral contraceptive medication (0.63% compared with 0.30% in female patients with no such prescription). No differences in the incidence of deep venous thrombosis or pulmonary embolism on the basis of sex or arthroscopic procedure code were noted. Nine patients (0.04%) died within ninety days of surgery, although only one death was confirmed to have resulted from a pulmonary embolism.
Conclusions:
The ninety-day incidence of symptomatic venous thromboembolism after elective knee arthroscopy was relatively low, with a 0.25% incidence of deep venous thrombosis and a 0.17% incidence of pulmonary embolism. The overall ninety-day mortality after arthroscopic knee surgery was 0.04%.
Level of Evidence:
Prognostic Level II. See Instructions for Authors for a complete description of levels of evidence.
The 2008 U.S. Surgeon General's report on venous thromboembolism (VTE) stated that 350,000 to 600,000 Americans will likely develop a deep venous thrombosis (DVT) or a pulmonary embolism (PE) each year and that 100,000 may die as a direct or indirect result of these disorders1. Within orthopaedic surgery, consensus standards for VTE prophylaxis during total joint replacement have been set by The Joint Commission and by the Centers for Medicare & Medicaid Services (CMS). In addition, the CMS has classified VTE after knee arthroplasty as a nonreimbursable hospital-acquired condition2. To date, no such standards for VTE prophylaxis have been set for other orthopaedic procedures. Understanding the risk associated with these other orthopaedic procedures is important so that any future federal recommendations may be based on sound scientific data.
The most commonly performed orthopaedic procedure in the United States is knee arthroscopy, with more than 1 million procedures performed annually3. Although knee arthroscopy is typically considered to be associated with a fairly low risk of complications such as DVT, PE, and death, some authors have recommended routine thromboprophylaxis for patients undergoing this procedure4-6. Prospective studies that employed the use of postoperative ultrasonography or venography to identify both symptomatic and asymptomatic DVTs after knee arthroscopy have reported incidences ranging from 1.2% to 17.9%7-12. A meta-analysis by Ilahi et al., which reviewed six prospective studies with a total of 684 patients, estimated an overall DVT incidence of 9.9%, but the incidence of proximal DVTs was only 2.1%13. Retrospective studies evaluating the occurrence of only symptomatic DVTs after knee arthroscopy performed without prophylaxis have indicated incidences ranging from 0% to 7.3%14-20. However, these studies included limited numbers of patients, which may explain the large variance among the estimates.
The primary purpose of the present study was to describe the incidence of symptomatic DVT, symptomatic PE, and mortality after elective knee arthroscopy performed without thromboembolic prophylaxis. The secondary purpose was to investigate the association of age, sex, procedure type, and oral contraceptive use with the odds of a venous thromboembolic event in this population. Our hypothesis was that the incidence of DVT and PE would differ according to age, sex, and procedure type and that there would be an increased incidence in those female patients who used oral contraceptive medication.
A retrospective cohort study of knee arthroscopy procedures was performed with use of administrative data, electronic medical records, and pharmacy records from twenty-seven medical centers comprising the Kaiser Permanente health maintenance organization (HMO) in California. This HMO had an average annual membership of over 6.5 million patients during the study period, and over 10,000 knee arthroscopy procedures are performed annually in this population. During the study period (January 1, 2006, to March 31, 2008), 21,794 arthroscopic surgical procedures involving the knee were performed. After institutional review board approval, administrative discharge data and electronic medical records were utilized to determine the incidence of intraoperative and postoperative venous thromboembolic events and mortality within ninety days after the knee arthroscopy.
The integrated administrative and electronic medical records were examined, and patients who underwent knee arthroscopy were identified with use of the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) procedure codes (Table I). The patient discharge records for these procedures were used to obtain the principal procedure code, the principal diagnosis code, and all secondary diagnoses.
Patients who had a history of a previous DVT or PE identified by the presence of the ICD-9-CM code V12.50 as one of the discharge diagnoses were excluded from the analysis. Patients with a history of a DVT or PE are known to have a higher risk of a subsequent VTE, so this group was eliminated to study the true incidence of a new VTE in a group of patients whom orthopaedic surgeons would typically consider to be at low risk. Furthermore, pharmacy records were used to exclude patients who had received an order or a prescription for an anticoagulation medication within two weeks prior to or on the day of the surgery. This group would have received such medication because they either were considered by their surgeon to be at high risk or, more commonly, were receiving anticoagulation therapy for a coexisting condition. This group was excluded because the prophylaxis administered near the time of the surgery could have affected the likelihood of developing a VTE. Patients with a primary diagnosis of infection at the time of the surgery were excluded because the surgical procedure would not have been elective. Finally, patients with less than ninety days of follow-up were identified with use of the HMO membership status information in the electronic medical records and were excluded from the analysis.
Thromboembolic events recorded in the electronic medical record system and administrative databases were identified with use of the ICD-9-CM code algorithm developed for Patient Safety Indicator 12, postoperative pulmonary embolism and deep venous thrombosis, by the Agency for Healthcare Research and Quality (AHRQ) (Table II)21. All patient activity within our organization's integrated records (including inpatient stays or procedures, emergency room encounters, urgent care encounters, and outpatient visits) for ninety days after the knee arthroscopy procedure were reviewed with use of this algorithm. If an encounter was associated with any of these codes, the patient was flagged as having possibly experienced a VTE, and the charts were reviewed by the study coordinator and the senior author. This review ensured that false positive cases, which have been reported to represent a limitation of this algorithm, were reviewed and reclassified22-24. In addition to using the ICD-9-CM algorithm, we also reviewed all imaging and pharmacy records in an attempt to ensure that patients with a symptomatic VTE were not missed because of coding errors. We reasoned that although occasional coding errors may have occurred, we could use other records within the patient charts where a suspicion of the condition might be recorded. Thus, the records of patients who underwent ultrasonography of a lower extremity or who were prescribed an anticoagulation medication such as heparin, low-molecular-weight heparin, fondaparinux (Arixtra; GlaxoSmithKline, Research Triangle Park, North Carolina), or warfarin sodium during the ninety-day postoperative period were also reviewed for evidence of a VTE.
The charts for all patients with a potential VTE were reviewed to confirm the diagnosis. Diagnosis of a DVT required confirmation of the presence of a venous thrombus by duplex ultrasonography. DVT cases were classified as being in either a proximal vein (the popliteal vein or proximal to it) or the calf (any vein distal to the popliteal vein). Diagnosis of a PE required that a computed tomography (CT) scan of the chest or a ventilation perfusion scan indicate the presence of an embolism with moderate or high probability. Death was ascertained with use of the integrated electronic medical record system, which is regularly updated from county death records and the Social Security Administration Death Master Files.
Demographic data for all patients (including age, sex, and medical center location) were obtained from the electronic medical records. Oral contraceptive use in female patients was determined by a review of pharmacy records for a new or refilled prescription for an oral contraceptive medication within 120 days prior to the date of the knee arthroscopy.
Statistical Analysis
Univariate analyses were performed to examine the incidence of symptomatic DVT, symptomatic PE, and mortality events in patients who had undergone a knee arthroscopy procedure (Table III). Chi-square tests were performed to compare the DVT incidence, the PE incidence, and the overall incidence of thromboembolic events according to demographic categories and procedure type. Univariate conditional (fixed-effect) logistic regression models were created to estimate odds ratios. The 95% confidence interval (CI) was used to assess the association of patient age (less than fifty years old compared with fifty years or older), sex, procedure type (ICD-9-CM procedure code), and use of oral contraceptive medication with the occurrence of a symptomatic DVT or PE. A multivariate binary conditional logistic regression model was used to assess the age-adjusted association of oral contraceptive use in female patients with the occurrence of a thromboembolic event. All models accounted for hospital clustering and adjusted the standard error of the estimates accordingly. A p value of <0.05 (two-sided) was considered significant.
Source of Funding
No external financial support was received.
We identified 21,794 patients who underwent knee arthroscopy during the study period. Of these, 165 patients (0.8%) with a history of a VTE, 228 (1.0%) who had received some form of thromboembolic prophylaxis, and 180 (0.8%) who had a primary diagnosis of infection were excluded. An additional 451 patients (2.1%) who left the health plan within the ninety-day follow-up period and were lost to follow-up were also excluded. The study cohort thus included 20,770 patients with no documented history of a VTE who underwent elective knee arthroscopy without thromboprophylaxis. This cohort had a mean age (and standard deviation) of 44 ± 16 years; 57% (11,887) of the patients were male and 43% (8882) were female.
Fifty-one patients (0.25%; 95% CI, 0.18% to 0.31%) developed a symptomatic DVT within ninety days after surgery (Table III). Eleven of these events were in a vein in the calf and forty were in a proximal vein, resulting in a proximal DVT incidence of 0.19%. Two of the patients had a DVT in the contralateral limb and one developed a DVT in both limbs. A PE was identified in thirty-five patients (0.17%; 95% CI, 0.11% to 0.22%). Two patients had both a PE and a DVT. The overall incidence of VTE (DVT and/or PE) was 0.40% (95% CI, 0.32% to 0.49%).
The risk of developing a VTE depended on age. The overall VTE incidence was 0.34% (95% CI, 0.23% to 0.44%) in patients less than fifty years of age compared with 0.51% (95% CI, 0.36% to 0.66%) in patients fifty years of age or older (p = 0.052). The odds of a thromboembolic event were 1.54 times higher (95% CI, 1.00 to 2.37 times) in patients who were fifty years of age or older than in younger patients (p = 0.048), although the difference did not reach significance if DVT and PE odds were examined separately (Table IV). We did not find any significant association of sex or procedure code with thromboembolic events.
Female patients who had received a new prescription or refilled a prescription for an oral contraceptive medication within the four months prior to the knee arthroscopy had a higher DVT incidence (0.47%) compared with female patients who had not recently received such a prescription (0.19%, p = 0.035). The overall VTE incidence was also higher in the group that had been prescribed oral contraceptives (0.63% compared with 0.30%, p = 0.035). However, the PE incidence in this group was not significantly higher (0.16% compared with 0.13%, p = 0.728). The odds of developing a DVT, adjusted for age, were 2.54 times higher (95% CI, 1.08 to 5.97 times; p = 0.033) in female patients who had been prescribed an oral contraceptive medication than in female patients who had not, and the overall odds of developing a VTE were 2.15 times higher (95% CI, 1.05 to 4.40 times; p = 0.036).
Nine patients (0.04%) died during the ninety-day follow-up period. The mean age of the patients who died was significantly greater than that of the patients who did not (sixty compared with forty-four years, p = 0.014). The cause of death in one patient was confirmed by an autopsy to have been a PE. This patient had undergone an initial arthroscopy followed by an open osteochondral transplantation procedure. The cause of death of one twenty-five-year-old patient was indeterminate but an autopsy showed no evidence of a PE. Autopsy confirmation of the cause of death was not available for the remaining seven patients, and the cause of death was determined by review of the hospital records and the county death certificates. The causes of death included pneumonia, septicemia, cerebral vascular accident (two), and drug overdose (two). The cause of death could not be determined in one of these seven patients, and PE therefore could not be ruled out as a cause. Thus, the incidence of death due to a PE was in the range of 0.0048% to 0.0096%.
Because the incidence of VTE after total knee or hip replacement performed without thromboprophylaxis may be as high as 50%25, patients undergoing hip or knee replacement are routinely provided with some form of thromboprophylaxis. In contrast, the risk of thromboembolism associated with knee arthroscopy has been considered to be low, and thromboprophylaxis has not been the standard of care25. Understanding the incidence of VTE after knee arthroscopy is important as regulatory agencies play a greater role in dictating the practice patterns of surgeons.
The incidence of symptomatic VTE after elective knee arthroscopy in this patient cohort, 0.40%, was relatively low. The incidence of symptomatic DVT was 0.25% and that of symptomatic PE was 0.17%. Patients fifty years of age or older had a higher incidence of VTE compared with younger patients. Also, female patients who had been prescribed an oral contraceptive medication within four months prior to surgery had a higher incidence of DVT and a higher overall incidence of VTE compared with female patients who had not received such a prescription. Our initial hypothesis was thus partially confirmed, as older patients and those female patients who had received a prescription for oral contraceptives had an elevated incidence of VTE, but no differences on the basis of sex or procedure type were found.
The incidence of DVT after knee arthroscopy in previous reports varies between 0% and 17.9%26. The incidence of symptomatic DVT in the present study was similar to that in the largest previous study, by Jaureguito et al., in which 2050 patients who underwent arthroscopic knee surgery were evaluated and 0.24% were noted to have a clinically detectable DVT confirmed by ultrasonography27. The results were also very similar to those in a study by Felcher et al. of 7264 patients undergoing podiatric surgery, in which the incidences of DVT and PE were 0.16% and 0.14%, respectively, resulting in an overall VTE incidence of 0.3%28. Those authors also noted an increased incidence in female patients receiving oral contraceptive medication.
Other authors have used ultrasonographic evaluation of all knee arthroscopy patients to prospectively compare the incidence of both symptomatic and asymptomatic DVT in those patients receiving low-molecular-weight heparin and in a control group receiving no prophylaxis. Studies by Michot et al.29 and Wirth et al.30 found the VTE incidence to be higher in the control group (15.6% compared with 1.5% and 4.1% compared with 0.85%, respectively), but in both studies all of the DVTs in the control group were in calf veins and such occurrences are often subclinical. Marlovits et al. found a significantly lower DVT incidence in knee arthroscopy patients treated with subcutaneous enoxaparin after discharge compared with patients treated with a placebo injection, but 42% of the DVTs were in calf veins31. In the largest study to date, Camporese et al. randomized 1761 patients to thromboprophylaxis with either low-molecular-weight heparin or graduated compression stockings and found the DVT incidence to be 3.2% in the compression stocking group compared with 0.9% in the heparin group. More than half of the DVTs were in calf veins, and the incidence of clinically relevant bleeding was higher in the heparin group4. These studies suggest that thromboprophylaxis can decrease the incidence of postoperative DVT but that many of the DVTs are subclinical and may not require treatment. The added expense of treatment and the potential for bleeding complications must be considered.
The strengths of the present study include the large sample size, the self-contained and therefore low-attrition population encompassing twenty-seven medical centers, and the availability of a comprehensive and integrated electronic medical record system to capture VTE events. The advantage of this integrated medical record system is that it allows the identification of not only the population with ICD-9-CM codes of interest but all activity related to those codes in the patient records. Thus, in addition to searching records of hospital visits, as is done in most large database studies, we were also able to review emergency room, urgent care, and office visits with use of our algorithm. This allowed for the accurate identification of the knee arthroscopies performed during the time period and of the symptomatic thromboembolic events that occurred. The integrated medical record system also provided the opportunity to cross-check for thromboembolic events with use of the radiology and pharmacy databases.
The limitations of this study include its retrospective nature and our inability to perform stratification according to potential risk factors other than the reported factors that were readily available in the administrative database. The impact of potential risk factors such as length of surgery, tourniquet usage, occurrence of a prior VTE, type of anesthesia, family history of a bleeding disorder, body mass index, and smoking history was not evaluated as such data were not available in the database. Aspirin use around the time of the surgery may also have had an effect on the VTE incidence, but we were unable to evaluate the use of aspirin in this cohort because aspirin is an over-the-counter medication whose use cannot be tracked with our available databases.
An additional inherent limitation of using our data source is the reliance on the diagnosis codes, medication usage, and ordering of imaging tests to identify patients with a VTE. In an attempt to minimize misclassification of our patients as false negatives as a result of improper coding, the charts of all patients who had ultrasonography or anticoagulation medication ordered within the ninety-day postoperative period were also reviewed to determine whether a VTE had been noted. However, patients in whom an asymptomatic DVT or PE occurred after the arthroscopy were not identified because a confirmatory test would only have been performed if a clinical suspicion of a VTE existed. It is possible that younger patients have a greater ventilatory reserve and therefore would not manifest symptoms of a small PE. Although venography might be a more accurate confirmatory test for a DVT than duplex ultrasonography, in which the results are operator-dependent, it would not be feasible to use venography in a study of this size.
Some of the patients in the study cohort underwent another procedure in addition to the knee arthroscopy. This occurred in the one patient whose death was confirmed to be the result of a PE; this patient underwent both arthroscopy and a subsequent open osteochondral allograft procedure. Two patients who developed a DVT also underwent an open procedure after the arthroscopy; one of these patients underwent an open synovectomy, and the other underwent a medial patellofemoral ligament reconstruction. The reported VTE incidence in our study would have been even lower if these patients who underwent additional surgery had been excluded.
Although we noted an increased risk in female patients who had received a prescription for an oral contraceptive medication within the four months prior to surgery, we cannot be sure that those patients actually took the medication. We excluded patients with a documented history of DVT or PE and therefore cannot comment about the incidence in this higher-risk group. Finally, it could be argued that the observed differences according to age and oral contraceptive usage, although statistically significant, are not clinically important because the incidence was still relatively low in both instances.
The current recommendations in the American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition) state that unless patients have thromboembolic risk factors, no thromboprophylaxis other than early mobilization is necessary for knee arthroscopy25. The current literature is still inconclusive regarding the overall benefit of thromboprophylaxis after arthroscopic knee surgery, as many of the DVTs that have been identified may have been clinically unimportant. We found the incidence of symptomatic VTE after elective knee arthroscopy to be 0.40% with use of a large health maintenance organization database, which would suggest that the current recommendations by the American College of Chest Physicians are appropriate for elective knee arthroscopy in low-risk patients.
In summary, the incidence of symptomatic VTE after elective knee arthroscopy in this study was relatively low. The ninety-day DVT incidence was 0.25% and the PE incidence was 0.17%. The odds of developing a DVT were 2.54 times higher in a woman who had been prescribed an oral contraceptive medication than in a woman who had not. The odds of a thromboembolic event were 1.54 times higher in a patient who was fifty years of age or older than in a younger patient. The overall ninety-day mortality was 0.04% after elective arthroscopic knee surgery, although only one patient (0.0048%) died as a result of a confirmed PE.
U.S. Department of Health & Human Services: Office of the Surgeon General. The Surgeon General's call to action to prevent deep vein thrombosis and pulmonary embolism. .
Centers for Medicare and Medicaid Services. Hospital-acquired conditions. .
National Center for Health Statistics. National health statistics reports. .
Camporese
G;
Bernardi
E;
Prandoni
P;
Noventa
F;
Verlato
F;
Simioni
P;
Ntita
K;
Salmistraro
G;
Frangos
C;
Rossi
F;
Cordova
R;
Franz
F;
Zucchetta
P;
Kontothanassis
D;
Andreozzi
GM; KANT (Knee Arthroscopy Nadroparin Thromboprophylaxis) Study Group. Low-molecular-weight heparin versus compression stockings for thromboprophylaxis after knee arthroscopy: a randomized trial. Ann Intern Med.
2008;149:73-82.
Eynon
AM;
James
S;
Leach
P. Thromboembolic events after arthroscopic knee surgery. Arthroscopy.
2004;20:23-4.
Schippinger
G;
Wirnsberger
GH;
Obernosterer
A;
Babinski
K. Thromboembolic complications after arthroscopic knee surgery. Incidence and risk factors in 101 patients. Acta Orthop Scand.
1998;69:144-6.
Delis
KT;
Hunt
N;
Strachan
RK;
Nicolaides
AN. Incidence, natural history and risk factors of deep vein thrombosis in elective knee arthroscopy. Thromb Haemost.
2001;86:817-21.
Demers
C;
Marcoux
S;
Ginsberg
JS;
Laroche
F;
Cloutier
R;
Poulin
J. Incidence of venographically proved deep vein thrombosis after knee arthroscopy. Arch Intern Med.
1998;158:47-50.
Hoppener
MR;
Ettema
HB;
Henny
CP;
Verheyen
CC;
Büller
HR. Low incidence of deep vein thrombosis after knee arthroscopy without thromboprophylaxis: a prospective cohort study of 335 patients. Acta Orthop.
2006;77:767-71.
Ng
WM;
Chan
KY;
Lim
AB;
Gan
EC. The incidence of deep venous thrombosis following arthroscopic knee surgery. Med J Malaysia.
2005;60:14-6.
Stringer
MD;
Steadman
CA;
Hedges
AR;
Thomas
EM;
Morley
TR;
Kakkar
VV. Deep vein thrombosis after elective knee surgery. An incidence study in 312 patients. J Bone Joint Surg Br.
1989;71:492-7.
Williams
JS
Jr;
Hulstyn
MJ;
Fadale
PD;
Lindy
PB;
Ehrlich
MG;
Cronan
J;
Dorfman
G. Incidence of deep vein thrombosis after arthroscopic knee surgery: a prospective study. Arthroscopy.
1995;11:701-5.
Ilahi
OA;
Reddy
J;
Ahmad
I. Deep venous thrombosis after knee arthroscopy: a meta-analysis. Arthroscopy.
2005;21:727-30.
Carson
RW. Arthroscopic meniscectomy. Orthop Clin North Am.
1979;10:619-27.
Collins
JJ. Knee-joint arthroscopy—early complications. Med J Aust.
1989;150:702-3
Complications of arthroscopy and arthroscopic surgery: results of a national survey. Committee on Complications of Arthroscopy Association of North America. Arthroscopy.
1985;1:214-20.
Jackson
RW;
Abe
I. The role of arthroscopy in the management of disorders of the knee. An analysis of 200 consecutive examinations. J Bone Joint Surg Br.
1972;54:310-22.
McGinity
JB;
Geuss
LF;
Marvin
RA. Partial or total meniscectomy: a comparative analysis. J Bone Joint Surg Am.
1977;59:763-6.
Rand
JA. Arthroscopic management of degenerative meniscus tears in patients with degenerative arthritis. Arthroscopy.
1985;1:253-6.
Reigstad
O;
Grimsgaard
C. Complications in knee arthroscopy. Knee Surg Sports Traumatol Arthrosc.
2006;14:473-7.
Agency for Healthcare Research and Quality. Patient safety indicators technical specifications, version 4.3, June 2001. .
Henderson
KE;
Recktenwald
A;
Reichley
RM;
Bailey
TC;
Waterman
BM;
Diekemper
RL;
Storey
PE;
Ireland
BK;
Dunagan
WC. Clinical validation of the AHRQ postoperative venous thromboembolism patient safety indicator. Jt Comm J Qual Patient Saf.
2009;35:370-6.
White
RH;
Sadeghi
B;
Tancredi
DJ;
Zrelak
P;
Cuny
J;
Sama
P;
Utter
GH;
Geppert
JJ;
Romano
PS. How valid is the ICD-9-CM based AHRQ patient safety indicator for postoperative venous thromboembolism?Med Care.
2009;47:1237-43.
Zhan
C;
Battles
J;
Chiang
YP;
Hunt
D. The validity of ICD-9-CM codes in identifying postoperative deep vein thrombosis and pulmonary embolism. Jt Comm J Qual Patient Saf.
2007;33:326-31.
Geerts
WH;
Bergqvist
D;
Pineo
GF;
Heit
JA;
Samama
CM;
Lassen
MR;
Colwell
CW. Prevention of venous thromboembolism: American College of Chest Physicians evidence-based clinical practice guidelines (8th edition). Chest.
2008;133(
6 Suppl):381S-453S.
Bushnell
BD;
Anz
AW;
Bert
JM. Venous thromboembolism in lower extremity arthroscopy. Arthroscopy.
2008;24:604-11.
Jaureguito
JW;
Greenwald
AE;
Wilcox
JF;
Paulos
LE;
Rosenberg
TD. The incidence of deep venous thrombosis after arthroscopic knee surgery. Am J Sports Med.
1999;27:707-10.
Felcher
AH;
Mularski
RA;
Mosen
DM;
Kimes
TM;
DeLoughery
TG;
Laxson
SE. Incidence and risk factors for venous thromboembolic disease in podiatric surgery. Chest.
2009;135:917-22.
Michot
M;
Conen
D;
Holtz
D;
Erni
D;
Zumstein
MD;
Ruflin
GB;
Renner
N. Prevention of deep-vein thrombosis in ambulatory arthroscopic knee surgery: a randomized trial of prophylaxis with low—molecular weight heparin. Arthroscopy.
2002;18:257-63.
Wirth
T;
Schneider
B;
Misselwitz
F;
Lomb
M;
Tüylü
H;
Egbring
R;
Griss
P. Prevention of venous thromboembolism after knee arthroscopy with low-molecular weight heparin (reviparin): results of a randomized controlled trial. Arthroscopy.
2001;17:393-9.
Marlovits
S;
Striessnig
G;
Schuster
R;
Stocker
R;
Luxl
M;
Trattnig
S;
Vécsei
V. Extended-duration thromboprophylaxis with enoxaparin after arthroscopic surgery of the anterior cruciate ligament: a prospective, randomized, placebo-controlled study. Arthroscopy.
2007;23:696-702.
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.