Journal of Vascular Surgery
Volume 45, Issue 5 , Pages 998-1007, May 2007

Proximate versus nonproximate risk factor associated primary deep venous thrombosis: Clinical spectrum and outcomes

Presented at the Thirtieth Annual Meeting of the Midwestern Vascular Surgical Society, Cleveland, Ohio, Sep 7, 2006.

Section of Vascular Surgery, University of Michigan, School of Medicine, Ann Arbor, Mich

Received 19 September 2006; accepted 9 January 2007.

Article Outline

Objective

Although the treatment for acute deep vein thrombosis (DVT) is uniform, the circumstances under which it develops vary widely and may impact outcomes. This study compared clinical features and outcomes in patients who developed a primary DVT associated with a defined risk to those without any proximate risk factor.

Methods

Consecutive patients with a primary DVT and no past venous thromboembolism history from 2000 to 2002 were abstracted for demographics, risk factors, DVT anatomical characteristics, treatment, and outcomes of death and new pulmonary embolism. Comparison between patients with a proximate risk event within 30 days of DVT (Inpt) and those presenting with DVT with no defined proximate event (Outpt) was done by univariable and multivariable statistics. A validated survey was mailed to all living patients to assess long-term sequela.

Results

A total of 293 patients with a mean age of 55 years and 49% men had confirmed DVT by objective means (92% duplex) with a mean follow-up of 25 ± 21 months. Inpts were more likely to have recent surgery or blunt trauma, bilateral DVT, less use of low molecular weight heparin (LMWH), and new pulmonary emboli (all P <.05). Outpts with DVT were more likely to have a history of malignancy, tibial-popliteal DVT compared with iliofemoral DVT, higher use of LMWH, and coumadin. However, there was no difference in mortality. From the patient survey (21% response), Outpts were more likely than Inpts to develop later varicosities and have daily frustration related to their legs (P < .05), but no difference in edema or ulceration. Considering the entire group, independent factors associated with freedom from PE included ambulation (odds ratio [OR] = 2.3; 95% confidence interval [CI] = 1.1-5.0; P = .04) while bilateral DVT (OR = .26; 95% CI = .09-.76; P = .013) or subcutaneous heparin (OR = 22; 95% CI = .05-.98; P = .047) were associated with greater risk. Independent factors associated with survival included ambulation (OR = 3.0; 95% CI = 1.3-7.2; P = .02), Coumadin use (OR = 2.7; 95% CI = 1.2-6.1; P = .015), and tibiopopliteal DVT (OR = 2.4; 95% = 1.1-5.5; P = .03), while malignancy (OR = 0.1; 95% CI = .05-.24; P < .01) and myocardial infarction (OR = 0.12; 95% CI = .01-.92; P = .04) were associated with lower survival.

Conclusion

Patients who develop DVT related to a defined proximate risk event (Inpt) generally have more extensive DVT, an increased risk of PE, but less long-term functional morbidity and no difference in long-term mortality compared to those with no proximate risk.

 

The treatment for acute deep vein thrombosis (DVT) is relatively uniform and has well defined evidence based recommendations for prophylaxis and treatment.1, 2 With the advent of low molecular weight heparin (LMWH), graded compression hose, and accessible use of duplex ultrasonography, the consequences of DVT, such as pulmonary embolism (PE) and postthrombotic syndrome (PTS), should be lessened. However, recent epidemiologic data suggest that overall rates of venous thromboembolism (VTE) are similar to what they have been over the last half decade despite these new advancements3(and unpublished observations).

Patient risk factors for DVT are well documented and include malignancy, certain medications such as oral contraceptives, postsurgical or trauma state, genetic hypercoagulable disorders, or a combination thereof.3, 4 Patients who clinically manifest DVT probably have a two hit biologic phenomena.5 For example, a patient with an underlying genetic risk for DVT who then undergoes major surgery may manifest DVT that otherwise would not have occurred. Both medical illness and postsurgical states confer a similar significant acquired DVT risk3, 6, as these patients are often immobile and have other systemic inflammatory processes occurring. Two reviews from our own institution using an administrative database (unpublished data, submitted) suggest that medical patients may have slightly higher risk, perhaps because medical patients have been historically less likely to receive adequate DVT prophylaxis.7

The pathophysiology of DVT remains essentially consistent with what Virchow described in the 1800s, including hypercoagulability, stasis, and injury. Recently, the role of local and systemic inflammation has been recognized, both in the causation and resolution of DVT.8, 9 The biological characteristics of clinical DVT have not been well differentiated for several reasons, including the fact that most are treated in a similar therapeutic manner, a lack of vein-thrombus pathological specimens, and the inability to ethically study untreated or nonprophylaxed DVT patients. Further, the long-term outcomes of DVT have only shown differences in relation to thrombus burden and lysis times, but not necessarily related risk factors.10, 11, 12 For example, some patients with an iliofemoral DVT may have full resolution while some DVT never recannalize (as documented by serial duplex studies), despite presumably similar therapies, and, thus, a different propensity for PTS.11, 13, 14 The underlying reasons for these differences in DVT resolution may be determined by a number of factors including genetics, how rapidly the patient is treated with anticoagulation (and limb compression), or the circumstances by which the DVT develops.

This study evaluated clinical features and outcomes of patients who develop a primary DVT while hospitalized, compared to those who presented with a primary DVT without any defined antecedent risk event 30 days prior, presenting as outpatients. Factors related to occurrence of documented de novo PE are also examined.

Back to Article Outline

Methods 

All patients who were seen at the University of Michigan Hospitals via the emergency room or floor admission and had an ICD-9CM diagnosis code for DVT (440.41, .42, .43) were reviewed between 2000 and 2002. Follow-up data were obtained with the use of the electronic medical chart, as well as a validated quality of life survey (mailed twice if the patient did not respond).15, 16

Nonproximate event (Outpt) DVT were defined as patients that were diagnosed in the emergency room or in the clinic with primary DVT who had no documented antecedent direct surgical or traumatic event documented within 30 days of diagnosis. However, patients having a history of surgery or trauma beyond 30 days, with subsequent admission with a DVT were included in the Outpt group. Proximate event (Inpt) DVT were those patients who were diagnosed with a DVT while admitted on a medical or surgical service or having a surgery or hospitalization within 30 days of the diagnosis of DVT. A documented history of hypercoagulable state did not constitute a proximate event, as there is no time limit associated with this. Exclusions included those patients with upper extremity, soleal or gastrocnemius DVTs, and those with a documented prior history of VTE. Of note, gastrocnemius or soleal DVTs were not included given their low likelihood of contributing to significant PTS long term.13 Patients who were diagnosed with a primary pulmonary embolism (PE) but in whom a DVT scan was not performed were also excluded. After these exclusions, 293 of 430 patients were included for study.

Assessment of how the DVT diagnosis was made, including physical examination findings, duplex ultrasonography, venography, CT scan, or other means was documented. In those with duplex imaging, laterality (right or left) and region (iliofemoral or tibial-popliteal) was documented.

Patient demographics abstracted included age, tobacco use, comorbidities such as CHF (ejection fraction <30%) and obesity (BMI >40), oral contraceptive (OCP) or hormone replacement therapy (HRT) use, and family history of VTE. Known hypercoagulable states such as a deficiency of anticoagulant factors (protein C, S, and antithrombin), a positive antiphospholipid antibody titer, as well as Factor-V Leiden, and Prothrombin 20210A genetic mutations were documented. Recent proximate events such as orthopedic fractures or surgery, pregnancy, pneumonia, general surgery procedure, and amputation were documented in relation to greater or less than 30 days to DVT presentation. Postoperative events of acute myocardial infarction, bed rest status, and pneumonia were determined from chart review.

DVT prophylaxis was assessed for Inpts, including the categories of early ambulation, (within 24 hours of their procedure or illness), use of subcutaneous unfractionated heparin (uFH), sequential compression devices (SCD), Ted hose, LMWH, coumadin, or aspirin. DVT treatment assessed included uFH, LMWH, coumadin, aspirin, thrombolytic therapy, graded compression hose, and SCDs.

Outcomes included mortality and PE that occurred after the DVT diagnosis. Pulmonary embolism was documented with a positive imaging study such as pulmonary computed tomography angiography, pulmonary angiogram, or a high probability ventilation-perfusion scan. A modified questionnaire based on that validated by Comerota et al evaluated subjective and objective patients factors (fully listed in Appendix I).16 In brief, pain level, whether compression hose were worn, presence of ulcers, edema, amount of edema, and a Likert scale for their subjective assessment of lower extremity pain, daily frustration, and productivity were recorded.

Comparison between Outpts and Inpts with DVT were compared by univariable and multivariable statistical analysis. Survey results were compared with univariable analysis only. Kaplan Meier Lifeplot analysis was also done. The statistical software SAS version 9.0 (Cary, NC) was used. The University of Michigan IRB approved this study.

Back to Article Outline

Results 

A total of 293 patients were evaluated (Inpt = 97; Outpts = 196), with a mean age of 55 +/- 16 years, and 49% were men. Mean follow up was 25 +/- 21 months. Deaths occurred in 45 patients, and new PE was diagnosed objectively in 38 (13%) patients after their initial DVT. Risk factors for the entire group are shown in Table I. A significant number of patients had a history of bedrest status, malignancy, as well as recent fractures, and surgical procedures (primarily orthopedic and general surgery) and use of OCP/HRT. Family history of DVT was rare, documented in 2% of patients, and those with a hypercoagulable state were 5.4%. Of those 13 patients with a documented hypercoagulable state, only one patient was in the Inpt group. There were no pregnant patients in either group, nor were any patients documented with prothrombin 20210A genetic variant. Baseline medications with hematologic effects included coumadin use in 13 patients and aspirin in 22 patients.

Table I. Venous thrombosis risk factors: Whole group
Factor%(N)
Nonambulatory30(99)
Malignancy28(82)
General surgery23(68)
Orthopedic Fx18(54)
Orthopedic surgery13(38)
OCP/HRT12(35)
Pneumonia7(20)
Myocardial infarction2(6)
CHF5(15)
Obesity (BMI > 40)5(14)
Amputation2(6)
FHx VTE2(6)
fV Leiden2(6)
Protein C/S2(6)

Fx, fracture; OCP, oral contraceptive; HRT, hormone replacement therapy; BMI, body mass index; FHx, family history; f, factor.

Diagnosis of DVT was confirmed by duplex ultrasound in 92% of all patients, with venography, CTV and MRI used for the remaining 8%. Physical examination corroborated this by positive limb findings in 142 patients (48%). Computed tomographic venography was positive in 10 patients, and venography was used rarely (1%). Medications used for DVT prophylaxis for Inpts included aspirin in 26 patients, coumadin in 13 patients, TEDs and SCD in eight patients, LMWH in three patients, and SubQ heparin in eight patients. No significant difference in prophylaxis regiments were noted between general and orthopedic surgery (P = .18). Therapy for DVT included LMWH or uFH in 97%, coumadin use in 75%, and thrombolytic therapy in 1% of patients. Of those not able to receive heparin anticoagulation, 3% underwent a vena cava filter placement.

Anatomical distribution of DVT included iliofemoral location in 53%, and slightly more left sided (52%) than right sided DVT were documented. Tibial popliteal DVT was present in 67%, while bilateral DVT were observed in 6% of patients.

Comparison of patient factors between Outpt and Inpt DVT by univariable analysis showed many major differences (Table II). Specifically, male gender, and surgery and trauma were more common in Inpts while medical illnesses such as CHF and malignancy were more common in Outpts. Positive physical exam limb findings were more often documented in Outpts with DVT, compared with Inpts. Bilaterality was more common whereas tibial-popliteal DVT trended, but did not quite reach significance comparing Inpts with Outpts (P = .06). Low molecular weight heparin and coumadin was more commonly used in Outpts compared with Inpts, while IV uFH was more commonly used in Inpts. Obesity, tobacco use, or hypercoagulable states were not significantly different between Inpts and Outpts. Pulmonary embolism was more commonly documented in Inpts but mortality was the same between the groups.

Table II. Comparison of outpatient and inpatient characteristics
FactorOutpt %(N)Inpt %(N)P
Men44(86)60(58).01
Ortho surg0(0)36(35)<.001
Ortho Fx8(16)23(22)<.001
General surg14(27)41(40)<.001
CHF9(18)3(3).02
Malignancy32(63)20(19).02
OCP/HRT17(33)1(1)<.001
Positive exam63(123)20(19)<.001
Bilateral DVT4(8)11(11).018
R sided DVT48(94)65(63).008
uFH25(49)37(36).04
LMWH79(155)45(44)<.001
Coumadin81(159)64(62).001
PE10(19)21(30).01
Death15(30)15(15).97

Fx, fracture; OCP, oral contraceptive; HRT, hormone replacement therapy; uFH, unfractionated heparin; PE, pulmonary embolism.

Controlling for patient risk factors, de novo PE was more commonly observed in those patients with bilateral DVT, as well those receiving SubQ uFH, while freedom from PE was significantly increased with early ambulation (Table III).

Table III. Independent factors associated with freedom from pulmonary embolism
FactorOR95% CIP
Bilateral DVT0.260.09–.76.013
SubQ uFH0.220.05–0.98.047
Early ambulate2.291.06–4.97.036

Evaluating mortality by Kaplan Meier Lifetable analysis, there was no significant difference in survival over time comparing Inpt and Outpt groups, with overall survival ∼83% at 24 months (data not shown). As a group, factors independently associated with death included malignancy, acute myocardial infarction, and presence of bilateral DVT. Factors associated with lower mortality included tibial-popliteal DVT, coumadin use, and early ambulation (Table IV).

Table IV. Independent factors associated with freedom from mortality
FactorOR95% CIP
Malignancy0.100.05–.24<.0001
AMI0.120.01–.92.04
Bilateral DVT0.330.1–1.01.06
Tibiopopliteal DVT2.41.1–5.5.03
Coumadin2.71.2–6.1.015
Early ambulation3.01.3–7.2.02

AMI, acute myocardial infarction

Overall survey return of those alive was 53 patients (21%), and selected responses are shown in Table V. Venous varicosities, as well as median limb frustration score (1 = severe to 6 = none) was significantly greater in Outpts compared Inpts. Overall numbers were small and limb pain (37% to 39%) and edema was common (39% to 63%), but not significantly different between the two groups. The self reported amount of edema (mild = 0, to severe = 3) was judged equal, and leg pain scores were also similar. Finally, compression hose was documented in 26% of Outpts with only 6% of Inpts at follow-up.

Table V. Patient survey results
FactorOutpt %(N) N = 35Inpt %(N) N = 18P
Varicosities49(17)17(3).04
Ulcers11(4)6(1).65
Pain37(13)39(7)1.0
Compression hose26(9)6(1).14
Edema63(22)39(7).17
Amount edema⁎2 2 .93
Amount frustration⁎4 6 .05
Productive⁎5 5 .63

⁎Median Likert scale

Back to Article Outline

Discussion 

This study shows differences between those patients who develop a primary DVT with an associated proximate event compared with those with no proximate event. Most trials that have evaluated DVT incidence and therapy have been in the setting of specific operative procedures, illnesses, or those with known hypercoagulable states and have not considered the etiology per se. Indeed, the most comprehensive consensus guidelines do not differentiate therapy between DVT etiologies such as idiopathic DVT, vs a proximate event, except in delineating length of anticoagulant therapy as no evidence yet exists for differences.1 Similarly, the extent of DVT plays little role in dictating therapy, except in extreme limb threatening cases such as phlegmasia.

Classic work by the University of Washington group has shown that a large thrombus burden and specific anatomical segments of DVT may increase risk of PTS.11, 13 For example, the longer a thrombus takes to lyse and proximal multi-segmental compared with distal DVT both increase the long-term risk of PTS.10, 17 Rapid therapeutic anticoagulation is well established to decrease the incidence of PE and recurrent DVT,18, 19 but whether this decreases the risk of PTS is unknown. From our study, Outpts, presumably more delayed in receiving heparin anticoagulation, would have longer times of untreated DVT. The subjective patient response data suggest that later limb manifestations of PTS were indeed more severe in Outpts, although no quantitative evaluation of CEAP or venous clinical severity scores were able to be done. Greater numbers of patients and longer, more complete follow-up might have made these differences more significant. Few studies have focused on limb PTS and rapidity of anticoagulant therapy, although some studies have suggested more aggressive thrombolytic therapy to remove the thrombus burden as well as LMWH compared with uFH may decrease PTS.16, 20 Withholding treatment to answer this question would not be ethically feasible, and so far, the large multicenter trials of DVT prophylaxis21, 22, 23 have not provided patient data regarding the long-term PTS incidence in those patients who developed either symptomatic or asymptomatic DVT as defined venographically.

The rate of postsurgical DVT/PE from a validated prospectively maintained national database is ∼1.0% (Henke PK, et al, JVS 2007; in press). Not surprisingly, few of the Inpts who developed DVT had documented mechanical or pharmacological prophylaxis, despite well published consensus guidelines.1 The reasons for nonprophylaxis were not specifically delineated in this study and were directed by the primary service team. This group of patients may have been judged to be at elevated bleeding risk, although the rates of clinically significant bleeding with mechanical and LMWH prophylaxis in many trials is low.1, 2, 22 Others have reviewed multiple causes and proposed several solutions regarding the problem of noncompliance with VTE prophylaxis.24 This is also a selected group, and we do not have the full data regarding concurrent hospitalized and surgical patients during this same time. It is likely that the majority of patients admitted during this time did have appropriate prophylaxis, and is why they did not develop a clinical DVT.

The factors associated independently with de novo PE were interesting. Early ambulation was historically thought to increase the risk of PE, although that is clearly not the case from this or other studies1, 25 Bilateral DVT were independently associated with decreased freedom from PE by ∼80%. This suggests a greater thrombus burden and perhaps impaired natural lytic mechanisms contribute to PE. A recent registry study also confirms this association.26 Conversely, it is possible these patients were not treated as rapidly with heparin anticoagulation as the patients with unilateral DVT or these patients had an idiopathic increased clotting potential. Prior studies have shown that rapid therapeutic anticoagulation significantly decreases PE in those with DVT.1, 19 Unfortunately, we did not have access to the exact timeline of diagnosis of DVT, administration of heparin, and subsequent PE occurrence. Less clear is why the use of prophylactic subcutaneous uFH, but not LMWH was associated with an increased risk of PE. One possibility is that prophylaxis of SubQ uFH was not given frequently enough. It is supposed to be administered three times a day for efficacy in high risk patients,1 but in practice it is often just two times a day. Alternatively, LMWH may have greater anti-inflammatory properties, more effectively decreased thrombus formation, or promoted DVT resolution, possibly via cell adhesion molecule inhibition.9, 27

Mortality in this group of patients was primarily associated with their underlying medical illness such as malignancy, or complications such as myocardial infarction. More Outpts than Inpts had a current history of malignancy, possibly because many patients receive their chemotherapy as outpatients, as well as the fact that DVT is often the first presentation of a cancer.28 Much speculation exists regarding the pathophysiology of malignancy induced hypercoagulability, and the type of malignancy and chemotherapy plays a role as well.29, 30, 31 Interestingly, the occurrence of myocardial infarction has been associated with DVT12, although the overall patient numbers were small in this study.

Bilateral DVT approached, but did not reach, significance in association to increased mortality. Consistently, however, a significantly lower risk of death was observed with more distal tibiopopliteal DVT. These findings suggest that overall thrombus burden likely plays a role in mortality, although probably not directly from PE, as no independent relationship was found between PE and mortality in this group of patients. It may be that rapid and comprehensive medical supportive therapy prevented mortality. The magnitude of the PE, whether segmental or subsegmental, was not specifically delineated, and large studies confirm mortality of PE at 10% to 20%.32, 33, 34 Early ambulation and coumadin use was associated with lower mortality, but these are probably surrogate marker for patients (a less ill group of patients who were able to mobilize) and/or able to tolerate this form of long-term anticoagulation. Coumadin has not been shown to decrease mortality in patients with malignancy compared with LMWH in larger and longer duration studies.35 Consistently, no difference in mortality was observed over time by KM analysis between Inpts and Outpts.

Limitations of the current study include that it is retrospective and relies on chart abstraction of which documentation is sometimes incomplete. Similarly, we did not directly interview patients and may have missed some proximate risk events in Outpts that were not recorded in the chart. Similarly, some of the Outpts had proximate events prior to 30 days, but were chosen from the standard perioperative complication time frame and from prior DVT prophylaxis recommendations.21 As few of these patients were seen by vascular surgeons in clinic, no assessment of their objective limb status for CEAP class was possible. The low survey response rate was unfortunate and may bias the results in that the nonresponders may have either been more or less disabled with PTS manifestations. However, the proportion of Outpts and Inpts respondents was similar to the overall distribution of these two groups.

In conclusion, patients who develop a DVT as Inpts compared with Outpts shared some expected and unexpected differences. No specific changes in current therapy are suggested from this review, except to emphasize VTE prophylaxis for Inpts, increased compliance with compression hose, and early ambulation when possible. A long-term prospective patient study is underway to better delineate the natural history of DVT by serum biomarker analysis and duplex ultrasound, and the development of PE and PTS.

We greatly appreciate the statistical assistance of Mary Proctor, MS and the technical assistance of Susan Blackburn, RN and Brett Almond, MD.

Back to Article Outline

Author contributions 


Conception and design: PH, DW

Analysis and interpretation: PH, EF, TW

Data collection: EF, MV, BD, DW

Writing the article: PH, TW

Critical revision of the article: PH, EF, MV, BD, TW, DW

Final approval of the article: PH, EF, MV, BD, TW, DW

Statistical analysis: PH

Obtained funding: PH

Overall responsibility: PH

Back to Article Outline

Appendix 

Back to Article Outline

References 

  1. Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR, Colwell CW, et al. Prevention of venous thromboembolism: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest. 2004;126:338S–400S
  2. Mismetti P, Laporte S, Darmon JY, Buchmuller A, Decousus H. Meta-analysis of low molecular weight heparin in the prevention of venous thromboembolism in general surgery. Br J Surg. 2001;88:913–930
  3. Heit JA. Venous thromboembolism: disease burden, outcomes and risk factors. J Thromb Haemost. 2005;3:1611–1617
  4. Henke PK, Schmaier A, Wakefield TW. Vascular thrombosis due to hypercoagulable states. In:  Rutherford RB editors. Vascular Surgery. 6th ed.. Philadelphia: Elsevier Saunders; 2005;p. 726–732
  5. Brouwer JL, Veeger NJ, Kluin-Nelemans HC, van der Meer J. The pathogenesis of venous thromboembolism: evidence for multiple interrelated causes. Ann Intern Med. 2006;145:807–815
  6. Goldhaber SZ, Tapson VF. A prospective registry of 5451 patients with ultrasound-confirmed deep vein thrombosis. Am J Cardiol. 2004;93:259–262
  7. Goldhaber SZ, Turpie AG. Prevention of venous thromboembolism among hospitalized medical patients. Circulation. 2005;111:e1–e3
  8. Wagner DD. New links between inflammation and thrombosis. Arterioscler Thromb Vasc Biol. 2005;25:1321–1324
  9. Wakefield TW, Henke PK. The role of inflammation in early and late venous thrombosis: Are there clinical implications?. Semin Vasc Surg. 2005;18:118–129
  10. Hull RD, Marder VJ, Mah AF, Biel RK, Brant RF. Quantitative assessment of thrombus burden predicts the outcome of treatment for venous thrombosis: a systematic review. Am J Med. 2005;118:456–464
  11. Meissner MH, Manzo RA, Bergelin RO, Markel A, Strandness DE. Deep venous insufficiency: the relationship between lysis and subsequent reflux. J Vasc Surg. 1993;18:596;605; discussion 606-8
  12. Kikura M, Takada T, Sato S. Preexisting morbidity as an independent risk factor for perioperative acute thromboembolism syndrome. Arch Surg. 2005;140:1210–1217discussion 1218
  13. Johnson BF, Manzo RA, Bergelin RO, Strandness DEJ. Relationship between changes in the deep venous system and the development of the postthrombotic syndrome after an acute episode of lower limb deep vein thrombosis: a one- to six-year follow-up. J Vasc Surg. 1995;21:307–313
  14. Haenen JH, Janssen MC, Wollersheim H, Van’t Hof MA, de Rooij MJ, van Langen H, et al. The development of postthrombotic syndrome in relationship to venous reflux and calf muscle pump dysfunction at 2 years after the onset of deep venous thrombosis. J Vasc Surg. 2002;35:1184–1189
  15. Mathias S, Prebil L, Putterman C, Chmiel J, Throm R, Comerota AJ. A health-related quality of life measure in patients with deep vein thrombosis: a validation study. Drug Information Journal. 1999;33:1173–1187
  16. Comerota AJ. Quality-of-life improvement using thrombolytic therapy for iliofemoral deep venous thrombosis. Rev Cardiovasc Med. 2002;3(Suppl 2):S61–S67
  17. Gabriel F, Labios M, Portoles O, Guillen M, Corella D, Frances F, et al. Incidence of postthrombotic syndrome and its association with various risk factors in a cohort of Spanish patients after one year of follow-up following acute deep venous thrombosis. Thromb Haemost. 2004;92:328–336
  18. Hull RD, Raskob GE, Rosenbloom D, Pineo GF, Lerner RG, Gafni A, et al. Treatment of proximal vein thrombosis with subcutaneous low molecular-weight heparin vs intravenous heparin (An economic perspective). Arch Intern Med. 1997;157:289–294
  19. Weinmann EE, Salzman EW. Deep-vein thrombosis. N Engl J Med. 1994;331:1630–1641
  20. Hull RD, Pineo GF, Mah AF, Brant R. A randomized trial evaluating long-term low molecular-weight heparin therapy out-of-hospital vs warfarin sodium comparing the postphlebitic outcomes at three months. 2001;98:447A.
  21. Bergqvist D, Agnelli G, Cohen AT, Eldor A, Nilsson PE, Le Moigne-Amrani A, et al. Duration of prophylaxis against venous thromboembolism with enoxaparin after surgery for cancer. N Engl J Med. 2002;346:975–980
  22. Samama MM, Cohen AT, Darmon JY, Desjardins L, Eldor A, Janbon C, et al. A comparison of enoxaparin with placebo for the prevention of venous thromboembolism in acutely ill medical patients (Prophylaxis in Medical Patients with Enoxaparin Study Group). N Engl J Med. 1999;341:793–800
  23. Breddin HK, Hach-Wunderle V, Nakov R, Kakkar VV. Effects of a low-molecular-weight heparin on thrombus regression and recurrent thromboembolism in patients with deep-vein thrombosis. N Engl J Med. 2001;344:626–631
  24. Tooher R, Middleton P, Pham C, Fitridge R, Rowe S, Babidge W, et al. A systematic review of strategies to improve prophylaxis for venous thromboembolism in hospitals. Ann Surg. 2005;241:397–415
  25. Aschwanden M, Labs KH, Engel H, Schwob A, Jeanneret C, Mueller-Brand J, et al. Acute deep vein thrombosis: early mobilization does not increase the frequency of pulmonary embolism. Thromb Haemost. 2001;85:42–46
  26. Trujillo-Santos J, Herrera S, Page MA, Soto MJ, Raventos A, Sanchez R, et al. Predicting adverse outcome in outpatients with acute deep vein thrombosis (Findings from the RIETE Registry). J Vasc Surg. 2006;44:789–793
  27. Myers DD, Henke PK, Wrobleski SK, Hawley AE, Farris DM, Chapman AM, et al. P selectin inhibition enhances thrombus resolution and decreases vein wall fibrosis in a rat model. J Vasc Surg. 2002;36:928–938
  28. White RH, Chew HK, Zhou H, Parikh-Patel A, Harris D, Harvey D, et al. Incidence of venous thromboembolism in the year before the diagnosis of cancer in 528,693 adults. Arch Intern Med. 2005;165:1782–1787
  29. Sallah S, Wan JY, Nguyen NP. Venous thrombosis in patients with solid tumors: determination of frequency and characteristics. Thromb Haemost. 2002;87:575–579
  30. Lin J, Wakefield TW, Henke PK. Risk factors associated with venous thromboembolic events in patients with malignancy. Blood Coagul Fibrinolysis. 2006;17:265–270
  31. Rickles FR, Levine M, Edwards RL. Hemostatic alterations in cancer patients. Cancer Metastasis Rev. 1992;11:237–248
  32. Horlander KT, Mannino DM, Leeper KV. Pulmonary embolism mortality in the United States, 1979-1998: an analysis using multiple-cause mortality data. Arch Intern Med. 2003;163:1711–1717
  33. Carson JL, Kelley MA, Duff A, Weg JG, Fulkerson WJ, Palevsky HI, et al. The clinical course of pulmonary embolism. N Engl J Med. 1992;326:1240–1245
  34. Kniffin WD, Baron JA, Barrett J, Birkmeyer JD, Anderson FA. The epidemiology of diagnosed pulmonary embolism and deep venous thrombosis in the elderly. Arch Intern Med. 1994;154:861–866
  35. Lee AY, Levine MN, Baker RI, Bowden C, Kakkar AK, Prins M, et al. Low-molecular weight heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl J Med. 2003;349:146–153

 Competition of interest: none.

PII: S0741-5214(07)00058-4

doi:10.1016/j.jvs.2007.01.042

Journal of Vascular Surgery
Volume 45, Issue 5 , Pages 998-1007, May 2007

Article Tools