Abstract:
Venous thromboembolism (VTE) is a relatively rare complication of arthroscopic surgery of the lower extremity, but it does have the potential to result in significant morbidity and possible mortality. VTE has been reported to occur with knee arthroscopy, and guidelines for VTE prophylaxis before and after knee arthroscopy have been proposed. There are much fewer data regarding the incidence of VTE occurring after arthroscopy of the ankle and the hip. This article reviews the literature on the incidence, treatment, and prevention of VTE in association with arthroscopy of the lower extremity.Arthroscopy has long been perceived as low-risk surgery relative to other areas of orthopaedics, such as total joint replacement and trauma.1-3 Reported rates of venous thromboembolism (VTE) in association with lower extremity arthroscopy range from 0% to 17.9%. However, reported rates of pulmonary embolism (PE) and deep venous thrombosis (DVT) associated with total joint replacement in the absence of prophylaxis have approached 30% and 70%, respectively.2 Anticoagulation prophylaxis improves the incidence to approximately 0% for PE and 1.5% for DVT, but VTE still remains a major concern for every arthroscopic surgeon. Considerable controversy exists within the orthopaedic community regarding prophylaxis of VTE in patients undergoing total joint replacement. Multiple studies have been performed within this population, each with its own recommendations. The Seventh American College of Chest Physicians (ACCP) Conference on Antithrombotic and Thrombolytic Therapy performed an in-depth meta-analysis of these voluminous data and published guidelines for care.4 However, these data nevertheless remain controversial.5-7 The American Academy of Orthopaedic Surgeons (AAOS) has also established its own guidelines for the prevention of VTE in total joint arthroplasty of the hip and knee.8 VTE in arthroscopic surgery has not garnered the same level of concern, yet it remains an important cause of morbidity and even reported cases of mortality. This article reviews the relevant literature on VTE associated with lower extremity arthroscopy.
Key Words: Arthroscopy—Deep venous thrombosis—Lower extremity—Prophylaxis— Pulmonary embolism—Venous thromboembolism.
Brandon D. Bushnell, M.D., Adam W. Anz, M.D., and Jack M. Bert, M.D.
From the Departments of Orthopaedic Surgery, the University of North Carolina Hospitals (B.D.B.), Chapel Hill, and Wake Forest University (A.W.A.), Winston-Salem, North Carolina; the University of Minnesota School of Medicine, Minneapolis, and Summit Orthopedics, Ltd. (J.M.B.), St. Paul, Minnesota, U.S.A. The authors report no conflict of interest. Address correspondence and reprint requests to Brandon D. Bushnell, M.D., Department of Orthopaedic Surgery, University of North Carolina Hospitals, CB #7055, 101 Manning Drive, Chapel Hill, NC 27599, U.S.A. E-mail: bbushnel@unch.unc.edu
© 2008 by the Arthroscopy Association of North America
0749-8063/08/2405-7515$34.00/0
doi:10.1016/j.arthro.2007.11.010
Hip Arthroscopy
In a review article regarding hip arthroscopy published 20 years ago, Hawkins9 warned that “systemic complications, such as pulmonary emboli, must always be considered.” In multiple series of hip arthroscopy cases published in the past 2 decades, however, PE or DVT has simply not been reported. No currently published study has specifically focused on this issue, but a review of more than 5,500 cases in the literature has revealed a 0% rate of DVT and PE (Table 1).10-36
Table 1. Reported Incidence of Venous Thromboembolism in Hip Arthroscopy
| Author | Year | No. of Patients | DVT/PE Rate |
|---|---|---|---|
| Glick et al.10 | 1987 | 12 | 0% |
| Small11 | 1988 | 14 | 0% |
| Funke and Munzinger12 | 1996 | 19 | 0% |
| Byrd and Chern13 | 1997 | 11 | 0% |
| Baber et al.14 | 1999 | 328 | 0% |
| Griffin and Villar15 | 1999 | 640 | 0% |
| Byrd and Jones16 | 2000 | 38 | 0% |
| McCarthy et al.17 | 2001 | 436 | 0% |
| Dienst et al.18 | 2001 | 35 | 0% |
| O’Leary et al.19 | 2001 | 86 | 0% |
| Sampson20 | 2001 | 530 | 0% |
| Byrd21 | 2003 | 265 | 0% |
| Clarke et al.22 | 2003 | 1,054 | 0% |
| Keeney et al.23 | 2004 | 102 | 0% |
| McCarthy24 | 2004 | 1,475 | 0% |
| Pasa et al.25 | 2005 | 24 | 0% |
| Yamamoto et al.26 | 2005 | 10 | 0% |
| Yamamoto et al.27 | 2005 | 30 | 0% |
| Awan and Murray28 | 2006 | 22 | 0% |
| Lo et al.29 | 2006 | 73 | 0% |
| Jerosh et al.30 | 2006 | 22 | 0% |
| Mullis and Dahners31 | 2006 | 36 | 0% |
| Owens and Busconi32 | 2006 | 11 | 0% |
| Bushnell et al.33 | 2007 | 156 | 0% |
| Kim et al.34 | 2007 | 43 | 0% |
| Philippon et al.35 | 2007 | 45 | 0% |
| Philippon et al.36 | 2007 | 37 | 0% |
| Total | — | 5,554 | 0% |
ANKLE ARTHROSCOPY
As in hip arthroscopy, ankle arthroscopy has a low rate of VTE. While not specific for ankle arthroscopy arthroscopy, a 0.2% to 4% incidence of VTE has been reported in association with all types of foot and ankle surgery.37-39 A review of recent literature specific for ankle arthroscopy confirms that this very low incidence rate can indeed be extrapolated to arthroscopic ankle surgery.In 15 studies involving more than 1,300 patients total, no DVT or PE was reported (Table 3).40-54 Three particular studies of VTE in ankle surgery in general merit discussion in light of their relevance to arthroscopic ankle surgery. Mizel et al.37 performed a prospective, multicenter analysis of the rate of VTE in 2,373 patients undergoing various procedures of the foot and ankle. They reported a 0.22% incidence of VTE, including a 0.15% rate of non-fatal pulmonary embolus. Patients were diagnosed by clinical examination alone. The authors found that postoperative immobilization and non–weight-bearing status statistically correlated with an increased risk of VTE.
They also performed a cost– benefit analysis of screening tests and prophylaxis for VTE, and they concluded that neither is justified in routine clinical care. Solis and Saxby38 published a prospective study in which 201 consecutive patients undergoing general foot and ankle surgery received duplex ultrasound testing. No prophylaxis for VTE was used in any of these patients. Deep calf vein thrombi were diagnosed in 7 patients, representing a 3.5% incidence rate. The authors concluded that postoperative immobilization, surgery of the hindfoot, increased tourniquet time, and advanced age all were risk factors for VTE. Hanslow et al.39 reported on a retrospective review of 602 patients undergoing general foot and ankle surgery utilizing VTE prophylaxis which varied depending on the treating surgeon. The diagnosis of VTE was made by clinical exam, and a 4% incidence of VTE was reported. A significantly increased risk was present in cases involving postoperative immobilization, previous DVT, rheumatoid arthritis, and recent air travel.
As in hip arthroscopy, screening patients for VTE risk in ankle arthroscopy involves the evaluation of general medical risk factors (Table 2). No specific studies of VTE in isolated arthroscopic ankle cases have been reported, but unique risk factors can be extrapolated from a small number of studies focusing on ankle surgery in general (Table 4). Orthopaedic surgeons should consider some sort of VTE prophylaxis when these risk factors are present, but no guidelines or studies have been reported regarding the choice of prophylaxis.
Table 2. General Risk Factors for Venous Thromboembolism
| Advanced age |
| Personal or family history of previous deep venous thrombosis or pulmonary embolism |
| Major systemic trauma |
| Lower extremity trauma |
| Prolonged immobilization |
| Full or partial paralysis of lower extremities |
| Central venous catheterization |
| Obesity |
| Tobacco use |
| Cancer |
| Pregnancy or postpartum status |
| Hormonal contraceptive use |
| Hormone replacement therapy |
| Treatment with selective estrogen receptor modulators |
| Known thrombophilic condition |
| Varicose veins |
| Acute medical illness |
| Heart failure |
| Respiratory failure |
| Inflammatory bowel disease |
| Myeloproliferative disorders |
| Nephrotic syndrome |
| Paroxysmal nocturnal hemoglobinuria |
Abbreviations. DVT, deep venous thrombosis; PE, pulmonary embolism.
Table 3. Reported Incidence of Venous Thromboembolism in Ankle Arthroscopy
| Author | Year | No. of Patients | DVT/PE Rate |
|---|---|---|---|
| Van Dijk et al.40 | 1996 | 30 | 0% |
| Harrington et al.41 | 1996 | 24 | 0% |
| Acevedo et al.42 | 2000 | 29 | 0% |
| Rasmussen et al.43 | 2002 | 105 | 0% |
| Ono et al.44 | 2004 | 105 | 0% |
| Hintermann et al.45 | 2004 | 51 | 0% |
| Thomas et al.46 | 2005 | 50 | 0% |
| Nihal et al.47 | 2005 | 11 | 0% |
| Maiotti et al.48 | 2005 | 22 | 0% |
| Ferkel et al.49 | 2005 | 35 | 0% |
| Okuda et al.50 | 2005 | 30 | 0% |
| Lui et al.51 | 2006 | 5 | 0% |
| Savva et al.52 | 2007 | 808 | 0% |
| Collman et al.53 | 2007 | 39 | 0% |
| Hassan54 | 2007 | 23 | 0% |
| Total | — | 1,367 | 0% |
Abbreviations. DVT, deep venous thrombosis; PE, pulmonary embolism.
Table 4. Risk Factors for Venous Thromboembolism Associated Specifically with Foot and Ankle Surgery
| Postoperative immobilization |
| Nonweightbearing status |
| Hindfoot surgery |
| Increasing tourniquet time |
| Advancing age |
| Previous thromboembolic event |
| Rheumatoid arthritis |
| Recent air travel |
KNEE ARTHROSCOPY
The majority of the literature regarding VTE of lower extremity arthroscopy involves the knee. The two major groups of studies that exist are those determining the incidence of VTE without prophylaxis and those evaluating the prophylactic benefit of low–molecular weight heparin (LMWH). The remaining literature involves case reports and small series illustrating important considerations in medical management. Upon a thorough review of the literature, the incidence of VTE associated with knee arthroscopy ranges from 0% to 17.9% (Table 5).55-75
Incidence
Articles associating VTE and knee arthroscopy include those that have reported both the incidence of DVT and/or PE. Some of these studies published the incidence of thromboembolic events as their primary focus, while others reported it secondarily as part of a study with a different goal. The incidence of VTE, DVT, and PE differs significantly among various studies based upon multiple factors, including patient age, medical history, and type of surgical procedure performed. The literature and its reports of incidence reflect the changes in VTE
management over the last 3 decades. In the 1980s and 1990s, published studies regarding the incidence of VTE and knee arthroscopy reported the
diagnosis based upon clinical signs and symptoms— a method which suffers from low sensitivities and specificities. Recent articles have reported
that 40% to 50% of cases of DVT diagnosed by ultrasound or venogram were clinically asymptomatic, implying that the earlier studies may have
missed several cases of VTE.71 In more recently published articles, however, the diagnosis has been based upon the objective results of venography and/or ultrasound.
Perhaps the best overall analysis of VTE incidence after knee arthroscopy in patients without prophylaxis is a meta-analysis by Ilahi et al.76 which included a review of six studies and a total of 684 patients.77 This meta-analysis suggested an overall DVT rate of 9.9% and a proximal DVT rate of 2.1%. Criteria for inclusion in the meta-analysis were screening for DVT with ultrasound or venography and the exclusion of concomitant ligament repair or open knee surgical procedures.
Obviously, the general medical risk factors for VTE incidence also apply to knee arthroscopy (Table 2). A direct assessment of some of these factors has been performed in some studies specific to knee arthroscopy. Demers et al.70 performed an analysis which included age, sex, family history of VTE, previous personal history of VTE, home medications, anesthesia, duration of surgery, duration of immobilization, and tourniquet time.
Table 5. Reported Incidence of Venous Thromboembolism in Knee Arthroscopy
| Author | Year | DVT | PE |
|---|---|---|---|
| Jackson and Abe55 | 1972 | 1.4% | 0.5% |
| McGinty et al.56 | 1977 | 7.3% | 0% |
| Carson57 | 1979 | 4.9% | 1.6% |
| Guhl58 | 1979 | 1.0% | 0% |
| Lysholm et al.59 | 1981 | 1.0% | 0% |
| Dandy and O’Carroll60 | 1982 | 0.3% | 0% |
| Mulhollan61 | 1982 | 0.3% | 0.02% |
| DeLee62 | 1985 | 0.1% | 0.03% |
| Rand63 | 1985 | 3.4% | 0% |
| Coudane et al.64 | 1986 | 0.75% | 0.25% |
| Collins65 | 1989 | 0.3% | 0.06% |
| Stringer et al.66 | 1989 | 4.2% | 0% |
| Williams et al.67 | 1995 | 3.5% | 0% |
| Cullison et al.68 | 1996 | 1.5% | 0% |
| Durica et al.69 | 1997 | 3.2% | 0% |
| Demers et al.70 | 1998 | 17.9% | 0% |
| Delis et al.71 | 2001 | 7.8% | 0% |
| Bergqvist and Lowe72 | 2002 | 4.2% | 0% |
| Ng et al.73 | 2005 | 1.2% | 0% |
| Hoppener et al.74 | 2006 | 5.7% | 0.003% |
| Reigstad and Grimsgaard75 | 2006 | 0% | 0.17% |
Despite a review of all of these
risk factors, only a tourniquet time of greater than 60
minutes was found to be a statistically significant
factor associated with the development of VTE. Delis
et al.71 studied the following criteria: age greater than
65 years, body mass index over 30, smoking, hormone
replacement or contraceptive use, history of chronic
venous insufficiency, and history of previous VTE. In
this series, the only statistically significant single risk
factor was a history of previous VTE. The authors also
concluded that two or more general risk factors for
hypercoaguability increased the risk of VTE after
knee arthroscopy.71
Prophylaxis
The second major group of articles regarding VTE
in knee arthroscopy involves a series of studies regarding
prophylactic options. There are three published
randomized, controlled trials and three cohort
studies that have assessed prophylaxis with LMWH
after knee arthroscopy. While results have varied
slightly between these studies, the overall consensus
seems to be that LMWH is effective at reducing VTE
risk.
Wirth et al.78 followed 239 total patients and randomized
them into matched groups: 117 receiving
reviparin for 7 to 10 days after surgery and 122
receiving no treatment. DVT was diagnosed by colorcoded
sonography. There was a VTE incidence of
0.85% in the reviparin group and 4.1% in the control
group, correlating with a relative risk reduction of
80%. The single patient with a DVT in the reviparin
group was found to have low levels of protein C and
protein S, indicating a possible coagulopathy. There
were no patients with major bleeding episodes in the
reviparin group.78
Michot et al.79 published a similar study involving
130 patients randomized into two groups: 66 treated
with dalteparin and 64 without prophylaxis. The first
dose of LMWH was given just before surgery and
continued daily for 4 weeks. DVT was diagnosed by
compression ultrasonography. There was a VTE incidence
of 15.6% without prophylaxis and 1.5% with
prophylaxis. This represented a statistically significant
risk reduction for VTE (P .004). The authors reported
that 80% of all DVTs were diagnosed within
the first 2 weeks after surgery.79
Marlovits et al.80 reported on 175 patients undergoing
arthroscopic anterior cruciate ligament repair. All
patients received enoxaparin preoperatively and during
their 3- to 8-day postoperative hospitalization.
Patients were then randomized into two groups: 87
patients receiving enoxaparin and 88 receiving placebo
for 3 weeks after discharge. This study employed
magnetic resonance venography to evaluate for DVT
and lung scans to evaluate for PE. There was a 2.8%
incidence of DVT in patients treated with enoxaparin,
compared to a 41.2% incidence of DVT of those
receiving placebo. This represented a statistically significant
reduction in risk of DVT with enoxaparin (P
.001). There was no statistically significant difference
in bleeding events between the two groups, and
there were no cases of PE. After an analysis of risk
factors, this study concluded that age greater than 30 years and immobilization before surgery were statistically
significant for increased risk of VTE.
Three other studies focusing on VTE prophylaxis
did not involve a randomized, controlled design but
still merit discussion here. Schippinger et al.81 and
Obernosterer et al.82 followed a cohort of 101 patients
who received dalteparin after knee arthroscopy. Ultrasound,
phlebography, and lung scans were used to
screen for DVT and PE. This study reported a 12%
rate of VTE. Three patients had DVT, 4 patients
developed PE, and 5 patients had both DVT and PE.
This study reported an increased rate of VTE in spite
of LMWH prophylaxis, which is especially concerning
in that 4 of the 8 DVTs and 8 of the 9 PEs were
clinically silent. There was no correlation with tourniquet
time, anesthetic type, or duration of surgery
with the incidence of DVT.
Holland and Schain83 reviewed 102 patients who
underwent postoperative prophylactic treatment with
nadroparin for 5 to 6 days after knee arthroscopy.
There was a 4.9% rate of symptomatic DVT in spite of
prophylaxis with LMWH. Eighty percent of the DVTs
occurred after prophylactic treatment was terminated.
Montebugnoli et al.84 reported no proximal DVT or
other thromboembolic events in a study of 509 patients
receiving parnaparin after minor arthroscopic
knee surgery. The authors recommended treatment
with the LMWH for 10 days after surgery. There were
8 adverse events related to the parnaparin, however,
raising some concern about the risk– benefit ratio of
treatment.84
Future research is required to determine if the benefits
of prophylaxis for VTE after knee arthroscopy
outweigh the risks and costs. However, the randomized,
controlled trials of LMWH prophylaxis seem to
support its use in decreasing the risk of VTE after
knee arthroscopy. Yet the cohort studies illustrate that
multiple issues must be resolved before definitive
prophylactic treatment recommendations can be
made. A recent Cochrane Database review7 highlighted
the inability of the current literature to definitively
support thromboprophylaxis of any type, and
recommended that future studies perform more organized
stratification of patients by risk factor and type
of procedure. The type of LMWH, duration of therapy,
timing of therapy (preoperative and postoperative),
method of DVT/PE diagnosis, and screening for
other risk factors may all play a role in the analysis of
LMWH efficacy and also must be considered. Although
future studies will be required to determine the
appropriate methodology of using LMWH, this treatment
currently represents the best literature-supported option for prophylaxis.4,78-81,83 Specific indications for
LMWH use thus remain to be more clearly defined. In
addition, while the use of other methods such as
aspirin, warfarin, heparin, and multimodal prophylaxis
have all been well described in relation to total
joint arthroplasty, no studies have focused on their use
specifically for arthroscopy.
Fatal Pulmonary Embolism
As evidenced by the literature discussed earlier, PE
appears to be a very rare occurrence in association
with knee arthroscopy, and most cases of PE are
clinically silent or respond to treatment.85-88 Several
case reports of fatal PE exist in which one or more risk
factors were present for VTE.89-92 In some cases,
however, the only risk factor may have been the
arthroscopy itself (i.e., lower extremity surgery as a
risk factor). For example, a fatal PE was reported by
Navarro-Sanz and Fernandez-Ortega91 in a 46-year old
man who had no known risk factors other than his
orthopaedic procedure, which involved a tourniquet
time of only 35 minutes. This case serves as a reminder
of the importance of always considering the
potential risk of VTE in every surgical case, appropriately
discussing the possibility of this complication
with the patient preoperatively, and documenting that
discussion in the medical record.
Another recent case report by Janssen and Sala92
involved a 17-year-old female soccer player who suffered
a fatal PE after an anterior cruciate ligament
reconstruction. This patient had identifiable risk factors
of oral contraceptive use and family history of
coagulopathy. This case raises the question of whether
or not to discontinue oral contraceptives before and
after knee arthroscopy. Berg and Montanarella93 recommended
that oral contraceptive pills be continued
with the addition of chemoprophylaxis in patients
undergoing knee arthroscopy—arguing that the risk of
pregnancy-related death as a consequence of unprotected
sex is higher than the risk of death as a complication
of oral contraceptive use.
Conclusions
VTE remains one of the most feared potential complications
of arthroscopy of the lower extremity.
There have been no reports of cases of DVT or PE
associated with hip or ankle arthroscopy to date, and
thus the surgeon should base prophylaxis decisions
before these procedures on the existing medical factors
associated with each individual patient. The vast majority of existing data about VTE prophylaxis in
lower extremity arthroscopy pertains to the knee, including
specific guidelines and recommendations. The
incidence of VTE associated with knee arthroscopy
appears to be approximately 10% and can be reduced
with the use of prophylactic LMWH. Presently, however,
the guidelines of the Seventh ACCP Conference
on Antithrombotic and Thrombolytic Therapy remain
the only current official recommendations that are
specific to knee arthroscopy. The guidelines recommend
prophylaxis with LMWH only when patients
have pre-existing risk factors for hypercoaguability or
after a complicated/prolonged procedure.
4 Further research
is required to evaluate thromboembolic disease
in lower extremity arthroscopy, with attention to specifics
such as indications, risks, benefits, dosages,
intervals, costs, and other features of various prophylactic
options.
The above publication appeared in: Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 24, No 5 (May), 2008: pp 604-611
For additional information on ACL knee injuries, or to learn more about what is involved during ACL reconstruction surgery, please contact the office of orthopedic knee surgeon, Dr. Adam Anz, serving the greater Pensacola, Gulf Breeze, and Gulf Coast communities.
