Long-term Low-Molecular-Weight Heparin and the Post-Thrombotic Syndrome: A Systematic Review

Transcription

1 CLINICAL RESEARCH STUDY Long-term Low-Molecular-Weight Heparin and the Post-Thrombotic Syndrome: A Systematic Review Russell D. Hull, MBBS, MSc, a Jane Liang, MSc, a Grace Townshend, MSc b a University of Calgary, Calgary, Alberta, Canada; b Watermeadow Medical, Witney, United Kingdom. ABSTRACT OBJECTIVE: Post-thrombotic syndrome causes considerable morbidity. The Home-LITE study showed a lower incidence of post-thrombotic syndrome and venous ulcers after 3 months of treating deep vein thrombosis with the low-molecular-weight heparin tinzaparin versus oral anticoagulation. This systematic review examined whether long-term treatment of deep vein thrombosis using low-molecular-weight heparin, rather than oral anticoagulation, reduces development of post-thrombotic syndrome. METHODS: We identified 9 articles comparing treatment of deep vein thrombosis using long-term lowmolecular-weight heparin with any comparator, which reported outcomes relevant to the post-thrombotic syndrome assessed 3 months post-deep vein thrombosis. RESULTS: Pooled analysis of 2 studies yielded an 87% risk reduction with low-molecular-weight heparin in the incidence of venous ulcers at 3 months (P.019). One study showed an overall odds ratio of 0.77 (P.001) favoring low-molecular-weight heparin for the presence of 8 patient-reported postthrombotic syndrome signs and symptoms. Pooled analysis of 5 studies showed a risk ratio for lowmolecular-weight heparin versus oral anticoagulation of 0.66 (P.0001) for complete recanalization of thrombosed veins. CONCLUSION: These results support the lower incidence of post-thrombotic syndrome and venous ulcers observed in Home-LITE. Long-term treatment with low-molecular-weight heparin rather than oral anticoagulation after a deep vein thrombosis may reduce or prevent development of signs and symptoms associated with post-thrombotic syndrome. Post-thrombotic syndrome and associated acute ulcers may develop more rapidly after deep vein thrombosis than previously recognized Elsevier Inc. All rights reserved. The American Journal of Medicine (2011) 124, KEYWORDS: Venous ulcer Deep vein thrombosis; Low-molecular-weight heparin; Post-thrombotic syndrome; Recanalization; Post-thrombotic syndrome is a common sequela of deep vein thrombosis, occurring in 20% to 50% of patients within 2 years after a deep vein thrombosis. 1,2 Current knowledge of post-thrombotic syndrome has been reviewed recently. 3 The pathophysiology of post-thrombotic syndrome is incompletely understood, but its development post-deep vein thrombosis is believed to be related to the presence of persistent venous obstruction and reflux, as well as calf muscle pump dysfunction 4 and inflammation. 3,5,6 Although standard anticoagulant treatment of deep vein thrombosis prevents thrombus extension and embolization to the pulmonary arteries, it does not directly lyse the acute thrombus, which may be only partially cleared. It is thought that the thrombus itself may not only obstruct venous flow but also Funding: LEO Pharma supported this study financially. Conflict of Interest: R. Hull has received grants/research support from Bayer Pharmaceuticals Corp, LEO Pharma Inc, and Sanofi-Aventis; been a consultant for Bayer Pharmaceuticals Corp, LEO Pharma, Inc, Pfizer Inc, GlaxoSmithKline, and Wyeth Pharmaceuticals; and sat on advisory boards for Bayer Pharmaceuticals Corp, Pfizer Inc, and Sanofi-Aventis. J. Liang has no conflicts of interest to declare. G. Townshend is an employee of Watermeadow Medical, which received payment from LEO Pharma for work on this manuscript. Editorial assistance was provided by Watermeadow Medical. Authorship: R. Hull had the original idea for the study, provided guidance and input into all drafts, and is the guarantor of the study. G. Townshend performed the literature searches, made the study selection, and wrote the first draft. J. Liang performed the statistical analyses. Requests for reprints should be addressed to Russell D. Hull, MBBS, MSc, Thrombosis Research Unit, 601 South Tower, Foothills Hospital, th Street NW, Calgary, Alberta, Canada, T2N 2T9. address: rdhull@ucalgary.ca /$ -see front matter 2011 Elsevier Inc. All rights reserved. doi: /j.amjmed

2 Hull et al Effect of Long-term Low-Molecular-Weight Heparin on Post-Thrombotic Syndrome 757 lead to damage of the delicate valves of the deep veins, probably in association with mediators of inflammation. Together, these processes result in venous hypertension and venous valvular reflux, which are considered to be the main pathophysiologic processes that lead to the clinical manifestations of post-thrombotic syndrome. 1,4,7 Patients at highest risk for post-thrombotic syndrome seem to be female, obese, and older, and have extensive or more proximal deep vein thrombosis, previous deep vein thrombosis (especially if ipsilateral), and signs of post-thrombotic syndrome at 1 month. 2 The most common symptoms of post-thrombotic syndrome are persistent or intermittent pain, heaviness, swelling, itching, and tingling or cramping in the limb, which are typically aggravated by standing or walking. Typical clinical signs of post-thrombotic syndrome include edema, venous ectasia, hyperpigmentation, eczema, varicose collateral veins, and, in severe cases, lipodermatosclerosis and venous ulceration (the most severe complication of postthrombotic syndrome). 7,8 These symptoms are often chronic and lifestyle-limiting and present with different severities over time. CLINICAL SIGNIFICANCE In light of the negative impact of post-thrombotic syndrome, it is surprising that preventative and therapeutic options remain limited. Elastic compression stockings are recommended to prevent post-thrombotic syndrome in atrisk patients or to treat post-thrombotic syndrome without venous leg ulcers 8,9 but may be challenging to use in real life. Intermittent pneumatic compression can be used for treatment if ulcers are present. 8 No sound evidence as yet exists to recommend thrombolysis for post-thrombotic syndrome prevention, 5,7 and waiting until post-thrombotic syndrome is well established may not be in the best interest of patient morbidity. On the basis of the pathophysiology described, providing adequate anticoagulation after a deep vein thrombosis may reduce the development of postthrombotic syndrome 6,9,10 by preventing or reducing damage to the valves and microcirculation, and reducing inflammatory processes. The Home-LITE study 11 reported that in patients treated with tinzaparin for 3 months after a deep vein thrombosis, the occurrence of patient-reported leg ulcers and symptoms of post-thrombotic syndrome at 3 months was significantly lower than in patients treated with longterm warfarin. This observation has potential implications for clinical care. We performed a systematic review Long-term treatment of deep vein thrombosis with low-molecular-weight heparin rather than oral anticoagulation may improve recanalization of thrombosed veins and reduce or prevent development of signs and symptoms of post-thrombotic syndrome, including venous ulcers. Post-thrombotic syndrome and associated acute ulcers may develop more rapidly after a deep vein thrombosis than previously recognized. Patients who experience early symptoms and signs that could indicate deep vein thrombosis, recurrent deep vein thrombosis, or post-thrombotic syndrome should be treated with long-term low-molecular-weight heparin rather than oral anticoagulation. to determine whether similar findings have been reported. Our review was framed to answer the following question: Does long-term treatment of deep vein thrombosis using a low-molecular-weight heparin, rather than usual care, reduce the development of post-thrombotic syndrome or improve recanalization of thrombosed veins? MATERIALS AND METHODS We attempted to identify all trials published up to August 2009 that compared prospective, randomized treatment of deep vein thrombosis using long-term ( 3-month) treatment with low-molecularweight heparin versus any comparator group, and that reported outcomes relevant to post-thrombotic syndrome, recanalization of veins, or regression of thrombus size measured at least 3 months after the initial deep vein thrombosis. The methods are described in Supplementary Table 1. RESULTS We identified 9 studies that form the basis of this review (Table 1) (see Supplementary Table 1 online for details of study identification and selection). This article reports only findings relevant to post-thrombotic syndrome and recanalization, thrombus lysis, or reflux. Results concerning incidence of recurrent deep vein thrombosis, death, and bleeding have been reviewed in other publications. 8,12-16 Characteristics of Identified Studies The studies differed with respect to their inclusion criteria (Table 1). In all studies except 2, 17,18 the diagnosis of deep vein thrombosis was objectively confirmed. Two studies excluded recurrent deep vein thrombosis. 18,19 One study 18 included deep vein thrombosis of the upper extremity (4% of cases). Low-molecular-weight heparin treatment was with dalteparin (one study), enoxaparin (2 studies), nadroparin (2 studies), bemiparin (one study), or tinzaparin (3 studies) for 3 to 6 months. The comparator group in 7 studies consisted of initial unfractionated heparin or low-molecularweight heparin, during which time oral anticoagulation was commenced, followed by international normalized ratio adjusted oral anticoagulation only with warfarin or acenocoumarol for the duration of therapy. One study used a different low-molecular-weight heparin, parna-

3 Table 1 Reference Characteristics of Included Studies Patients 1. Das et al y with major DVT 2. Gonzalez-Fajardo et al 25 Gonzalez-Fajardo et al 21 Follow-up of Gonzalez-Fajardo et al López-Beret et al 26 Confirmed unilateral first episode of DVT, age 18 y Confirmed unilateral first episode of DVT, age 18 y Confirmed symptomatic DVT, 18 y 4. Kakkar et al 27 Confirmed DVT, 18 y 5. Daskalopoulos et al 22 Daskalopoulos et al 23 Follow-up of Daskalopoulos et al 22 Daskalopoulos et al 24 Follow-up of Daskalopoulos et al 22 Confirmed acute proximal DVT, 18 y Confirmed acute proximal DVT, 18 y Confirmed acute proximal DVT, 18 y 6. Hull et al 11 Confirmed proximal acute DVT, first or recurrent; 18 y 7. Romera et al 19 Symptomatic confirmed first episode of acute proximal DVT, 18y(n 122) Study Design Evaluation of Outcomes Intervention Comparator* Evaluation of venograms was Evaluation of venograms was Evaluation of recanalization was Evaluation of recanalization was Evaluation of outcomes was Evaluation of outcomes was Evaluation of outcomes was Dalteparin sc 5000 anti-xa units OD (n 50) Enoxaparin sc 40 mg BD for 7 d, then 40 mg OD (n 93) Enoxaparin sc 40 mg BD for 7 d, then 40 mg OD (n 85) Nadroparin sc 0.1 ml/10 kg BD (n 81) Bemiparin sc 115 anti-xa IU/kg OD for 10 d, then 3500 anti-xa units (n 94) Tinzaparin sc 175 anti-xa IU/kg OD (n 50) Tinzaparin sc 175 anti-xa IU/kg OD (n not reported) Tinzaparin sc 175 anti-xa IU/kg OD (n not reported) Tinzaparin sc 175 anti-xa IU/kg OD (n 240) Tinzaparin sc 175 anti-xa IU/kg OD (n 119) Dalteparin sc 5000 anti-xa units OD for 10 d Warfarin 9 mg OD starting day 8 for 3 d, then INRadjusted (n 55) Initial UFH IV adjusted by APTT Coumarin 5 mg OD starting day 5, then INR-adjusted (n 92) Initial UFH IV adjusted by APTT Coumarin 5 mg OD starting day 5, then INR-adjusted (n 80) Nadroparin sc 0.1 ml/10 kg BD for 5 d Acenocoumarol 4 mg OD starting day 3, then INRadjusted (n 77) (A) Initial UFH IV adjusted by APTT; VKA 10 mg OD starting day 3 for 3 d, then INR-adjusted (n 98) (B) Bemiparin sc 115 anti-xa IU/kg OD for 7 d; VKA at starting dose of 10 mg OD starting day 3 for 3 d, then INRadjusted (n 105) Initial UFH IV for 5-7 d Acenocoumarol INR-adjusted starting day 3 (n 52) Initial UFH IV for 5-7 d Acenocoumarol INR-adjusted starting day 3 (n not reported) Initial UFH IV for 5-7 d Acenocoumarol INR-adjusted starting day 3 (n not reported) Tinzaparin sc 175 anti-xa IU/kg OD for 5 d, then warfarin INRadjusted, commenced within 24h(n 240) Tinzaparin sc 175 anti-xa IU/kg OD until INR 2 on 2 consecutive measurements, acenocoumarol 3 mg, and then INR-adjusted (n 122) Duration of Therapy (mo) Duration of Follow-up (mo) PTS-Related Outcomes Reported 3 3 Presence of recanalization Method of Assessment of Recurrent DVT and Recanalization Venography 3 3 Marder score Venography 3 3, 6, 12 mo 2, 3, 4, 5 y Marder score Presence of PTS at5y 3or6 3, 12 Recanalization Reflux 3 3 Marder score at 14 d Recanalization at 3 mo 6 3, 6, 12 Clinical signs of PTS including ulcer Marder score Reflux 6 Every 6 mo up to3y Recanalization Development of PTS Ulceration Reflux 6 42 Development of PTS Ulceration signs/symptoms of PTS including ulcer 6 1, 6, 12 Degree of recanalization Venous competence Venography Duplex scanning Venography Duplex ultrasonography Duplex ultrasonography As above VTE, ultrasonography Recanalization, not assessed Duplex ultrasound 758 The American Journal of Medicine, Vol 124, No 8, August 2011

4 Hull et al Effect of Long-term Low-Molecular-Weight Heparin on Post-Thrombotic Syndrome 759 Table 1 Continued Color Doppler ultrasound 3 or 6 1, 3, 6, 12 Recanalization Valve competence Parnaparin sc 6400 IU BD for 1 mo, then 6400 IU OD for 2 or 5 mo (n 51) Nadroparin sc 11,400 IU OD for weight 70 kg or 15,000 IU OD for weight 70 kg (n 40) throughout therapy pilot Article does not report evaluation 8. Bellosta et al 18 Unilateral first episode of DVT of the lower limb Venography Signs of PTS Vessel patency Median (for PTS and venous patency) Enoxaparin sc 1 mg/kg BD for 7-14 d, then enoxaparin at starting dose of 1.0 mg/kg BD (n 30)# Enoxaparin sc 1 mg/kg BD for 7-14 d, then enoxaparin at starting dose of 1.5 mg/kg OD (n 50) Nonrandomized open-label prospective Article does not report evaluation 9. Schobess et al 20 Symptomatic confirmed DVT, children aged 3moand 18 y BD twice daily; INR international normalized ratio; IV intravenously; OD once daily; PTS post-thrombotic syndrome; SC subcutaneously; UFH unfractionated heparin; VTE venous thromboembolism. *In each case, an oral anticoagulant was commenced during heparin therapy and then continued. Heparin therapy was discontinued when the target INR was observed on 2 consecutive measurements. Oral anticoagulant therapy was generally adjusted to give an INR value of Refers to follow-up for PTS-related observations. Longer therapy if a recurrent VTE occurred in first 3 mo. Dosage reported as anti-xa IU/mL in syringe, given at 0.1 ml/10 kg BD. Both treatments continued for 3 mo and were prolonged for 6 mo when the DVT affected the iliac or femoral veins and if the patient had persistent risk factors for thrombosis or an idiopathic DVT. Therapy was continued for 3 mo in cases of DVT with known causes and for 6 mo in cases of idiopathic DVT. #Treatment aimed to achieve a 2-4-h anti-xa level of IU/mL in the first 7-14 d, IU/mL at 4 h, and 0.1 IU/mL at 12 h. parin, as the comparator, 18 and one study used a different regimen of enoxaparin. 20 For methodological quality of the studies, see Supplementary Table 2, online. All studies but one 20 were randomized and prospective. Because of the nature of the treatments, all studies were open-label; 2 studies 18,20 did not use evaluation of outcomes and thus were excluded from our meta-analyses. Comparison of Outcomes We have compared results in groups according to the outcomes reported (Figure 1, Tables 2, 3 and 4, Supplementary Tables 3 and 4, online). We performed meta-analysis where the data was suitable; otherwise, we compared results qualitatively. Presence of Venous Ulcers. Two studies 20,21 included ulcer as a criterion for diagnosing post-thrombotic syndrome but did not specifically report on ulcer outcomes over time. Daskalopoulos et al 22 reported that venous ulcers were present at 1 year in 3 patients (5.8%) in the oral anticoagulation group and in none of the tinzaparin group (no P value reported). Daskalopoulos et al 23,24 also reported superiority of tinzaparin over oral anticoagulation with respect to the incidence of ulcers at 18, 24, 30, 36, and 42 months (data not reported). In the Home-LITE study, 11 patients were asked to report whether an ulcer was present in the skin above their ankle on the affected leg (wording chosen to ensure that patients would report only genuine venous ulcers). The skin ulcers were acute, detected during follow-up at 3 months, and not preexisting. Ulcers were present in 1 patient treated with tinzaparin (0.5%) and 8 patients treated with oral anticoagulation (4.1%; P.02). Relative risk ratios for the presence of ulcer after tinzaparin therapy for 3 or 6 months or usual care based on these 2 studies are shown in Figure 1A. The overall risk ratio was 0.13 (favoring tinzaparin) (95% CI, ; P.019). Presence of Post-Thrombotic Syndrome Based on Clinical Signs and Symptoms. Four studies assessed the presence of post-thrombotic syndrome based on clinical signs and symptoms at follow-up periods from 3 months to 5 years (Table 2). These studies do not lend themselves to meta-analysis given the different assessment methods and definitions used and widely varying follow-up periods. One study evaluated clinical signs of post-thrombotic syndrome at 1 year, and the treatment groups did not differ significantly: Recanalization is reported separately below. 22 At longer follow-up periods, superiority of tinzaparin in preventing post-thrombotic syndrome was noted in an abstract and editorial (data not available to us). 23,24 At 6- to 12-month follow-up of children treated with enoxaparin for 3 months (median), Schobess et al 20 observed a post-thrombotic syndrome incidence of 6.7% in the once-daily group and 6.0% in the twice-daily group, based on objective signs of an increase in calf or ankle

5 760 The American Journal of Medicine, Vol 124, No 8, August 2011 Figure 1 Risk ratios for the (A) presence of venous ulcers and (B) absence of complete recanalization of thrombosed veins or complete lysis of thrombus, after treatment with long-term low-molecular-weight heparin or oral anticoagulation. A ratio of 1 favors tinzaparin. Both were calculated using a fixed-effect model. For ulcers: Daskalopoulos et al s trial: 24 6-mo therapy with tinzaparin and follow-up at 1 y; Hull et al s study: 11 3-mo therapy with tinzaparin and follow-up at 3 mo. For recanalization: Values are those reported for follow-up at 6 mo for Romera et al 19 and 3 mo for all the other studies. For López-Beret et al, 26 values reported for the popliteal vein were used. Values for the common and superficial femoral veins would have favored lowmolecular-weight heparin to an even greater extent (Table 3). CI confidence interval; LMWH low-molecular-weight heparin; OA oral anticoagulant. circumference of 2 to 4 cm, dark pigmentation of the skin, venous telangiectasia, varicose veins, or open ulcer. At 5-year follow-up, González-Fajardo et al 21 reported a post-thrombotic syndrome incidence of 60.7% in patients treated with enoxaparin for 3 months versus 70.5% in those treated with oral anticoagulation (not significantly different). The authors comment that the small sample size probably made it impossible to show a significant difference. In Home-LITE, patients completed a questionnaire at 3 months, indicating the presence or absence of 8 symptoms and signs commonly used in combination to diagnose postthrombotic syndrome (Supplementary Figure 1, online). 11 For all 8 symptoms and signs there was a trend favoring low-molecular-weight heparin over usual care. The overall odds ratio of 0.77 (95% CI, ) favored low-molecular-weight heparin (P.001). Presence and Degree of Recanalization. Four studies 19,23-26 reported significantly more patients with recanalization and lysis of thrombi with low-molecular-weight heparin than with oral anticoagulation at follow-up periods from 3 months to 1 year (Table 3). In 2 other studies, 17,27

6 Hull et al Effect of Long-term Low-Molecular-Weight Heparin on Post-Thrombotic Syndrome 761 Table 2 Presence of Post-Thrombotic Syndrome Based on Clinical Signs and Symptoms (% of Patients) LMWH Regimen (Reference) Tinzaparin 6 mo (Daskalopoulos et al 22 ) Method of Determination Clinical signs (other than ulcer; for ulcer, Figure 1A) Follow-up Period LMWH 1y n 50 Edema: 32% Local tenderness/pain: 18% Skin changes: 14% Enoxaparin, median 5 mo (Schobess et al 20 ) Objective signs 6-12 mo n % of children overall Enoxaparin 3mo Villalta scale 5 y n 56 (González-Fajardo et al 21 ) 60.7% (19.6% severe) Occurred in 65% of patients overall (both groups) Tinzaparin 3 mo (Hull et al 11 ) Patient questionnaire (signs other than ulcer; for ulcer, Figure 1A) LMWH low-molecular-weight heparin; OA oral anticoagulant. 3mo n 480 in total Odds ratio for LMWH vs OA: 0.77 ( ) (not reported as % of patients) Oral Anticoagulation P Value n % 15.4% 19.2% Not applicable n % (29.5% severe) NS.001 Not applicable NS differences in recanalization were not significant between treatment groups (Table 3). We performed a meta-analysis on the 5 studies that reported total recanalization or lysis of the thrombus in terms that could be compared (Figure 1B). The risk ratio was 0.66 (95% CI, ; P.0001) in favor of long-term low-molecular-weight heparin over usual care. In 2 of the studies, 22,25 which reported the extent of thrombus lysis in terms of Marder scores at 3 months to 1 year, Marder scores were significantly better in patients Table 3 Presence and Degree of Recanalization of Thrombosed Veins/Lysis of Thrombus as Number (Percentage) of Patients LMWH Regimen (Reference) Outcome and Method Follow-up Period LMWH n/n (%) Oral Anticoagulation n/n (%) Between-Group P Value Dalteparin 3 mo (Das et al 17 ) Recanalization; venography 3 mo 35/44 (80%) 35/42 (83%) NS Enoxaparin 3mo (Gonzalez-Fajardo et al 25 ) Lysis of thrombus; venography 3 mo 40/84 (47.6) substantial or complete lysis 14/80 (17.5%) substantial or complete lysis.001 for overall fate of thrombus Nadroparin 3-6 mo (López- Beret et al 26 ) Nadroparin 3-6 mo (López- Beret et al 26 ) Bemiparin 3 mo vs (A) UFH then VKA (B) Short-term bemiparin then VKA (Kakkar et al 27 ) Tinzaparin 6 mo (Daskalopoulos et al 23,24 ) Tinzaparin 6 mo (Romera et al 19 ) Tinzaparin 6 mo (Romera et al 19 ) Enoxaparin 6-12 mo (Schobess et al 20 ) (values estimated from bar graph) Nadroparin or parnaparin 3-6 mo (Bellosta et al 18 ) Nadroparin or parnaparin 3-6 mo (Bellosta et al 18 ) Nadroparin or parnaparin 3-6 mo (Bellosta et al 18 ) Recanalization scored by own scoring system; duplex scanning Recanalization scored by own scoring system; duplex scanning Partial or complete recanalization; venography or Doppler ultrasound 3 mo CFV 13/38 (34.2%) SFV 13/45 (28.9%) PV 20/53 (37.7%) 12 mo CFV 23/36 (63.9%) SFV 28/42 (66.7%) PV 31/50 (62.0%) 1/34 (2.9%) 1/46 (2.1%) 7/58 (12.1%) 6/31 (19.3%) 6/41 (14.6%) 14/51 (27.5%) 3 mo 66/81 (81.5%) (A) 64/85 (75.3%) (B) 71/89 (79.8%) NS Recanalization; duplex scanning Up to 42 mo Stated as superior, but no data provided Not reported Not reported Total recanalization; duplex 6 mo 87/119 (73.1%) 58/122 (47.6%).001 ultrasonography Total recanalization; duplex 1 y 107/117 (91.5%) 83/120 (69.2%).001 ultrasonography Vessel patency, radiologic evaluation 6 mo Complete patency: Not applicable NS 44% enoxaparin BD, 56% enoxaparin OD Partial patency: 50% enoxaparin BD, 33% enoxaparin OD Recanalization scored by own scoring 3 mo 9 (22.5%) nadroparin, Not applicable NS system; color Doppler ultrasound 14 (27.5%) parnaparin Recanalization scored by own scoring 6 mo 11 (27.5%) nadroparin, Not applicable NS system; color Doppler ultrasound 23 (45.1%) parnaparin Recanalization scored by own scoring 1 y 20 (50%) nadroparin, Not applicable NS system; color Doppler ultrasound 31 (60.8%) parnaparin CFV common femoral vein; LMWH low-molecular-weight heparin; PV popliteal vein; SFV superficial femoral vein.

7 762 The American Journal of Medicine, Vol 124, No 8, August 2011 Table 4 Presence and Degree of Recanalization of Thrombosed Veins/Lysis of Thrombus as Measured by the Marder Score* LMWH Regimen (Reference) Enoxaparin 3 mo (Gonzalez-Fajardo et al 25 ) Tinzaparin 6 mo (Daskalopoulos et al 22 ) Tinzaparin 6 mo (Daskalopoulos et al 22 ) Tinzaparin 6 mo (Daskalopoulos et al 22 ) Method LMWH low-molecular-weight heparin. *Lower Marder score indicates less thrombosis. Follow-up Period LMWH Marder Score Venography 3 mo Baseline End Duplex scanning Modified Marder score; shown as median, interquartile range Duplex scanning Modified Marder score; shown as median, interquartile range Duplex scanning Modified Marder score; shown as median, interquartile range Oral Anticoagulation Marder Score Baseline End Between-Group P Value mo 9 (7-10) 10 (8-15) mo 6.5 (4-9) 8 (6-12) y 5 (3-7) 7 (5-10).011 treated with low-molecular-weight heparin versus oral anticoagulation (Table 4). Degree of Reflux or Venous Competence. Two studies 22,26 examined the percentage of patients with venous reflux at 6 to 12 months (Supplementary Table 3). In each study, outcomes did not differ significantly, except for better reflux in the communicating veins in one study 26 when using low-molecular-weight heparin. The 1-year results from Daskalopoulos et al 22 suggest a trend toward better venous reflux results with low-molecular-weight heparin. The results of these 2 studies could not be pooled because of different reporting methods. Romera et al 19 reported results in terms of venous competence rather than reflux; in this study, 59% of the lowmolecular-weight heparin group exhibited venous competence at 1-year follow-up, significantly more than with oral anticoagulation (19%; P.001) (Supplementary Table 4). Bellosta et al 18 reported venous competence rates of 67% and 65% after long-term treatment with nadroparin or parnaparin, respectively (Supplementary Table 4). DISCUSSION Our systematic review shows that long-term low-molecularweight heparin seems to be more beneficial than oral anticoagulation in preventing development of post-thrombotic syndrome according to all the different post-thrombotic syndrome-related end points analyzed. Furthermore, our findings suggest that patients who experience early symptoms and signs that could indicate deep vein thrombosis, recurrent deep vein thrombosis, or early manifestation of post-thrombotic syndrome should be treated with long-term low-molecular-weight heparin rather than oral anticoagulation, and that this could potentially limit or prevent further development of post-thrombotic syndrome. Research on post-thrombotic syndrome has been limited by the lack of a gold standard objective test to confirm its diagnosis and grade its severity. The recent International Society on Thrombosis and Haemostasis Scientific and Standardization Committee statement on standardizing the definition of post-thrombotic syndrome is an attempt to remedy this lack. 28 Before 2009, no large study of anticoagulant regimens had evaluated post-thrombotic syndrome as a prespecified study end point. 28 Home- LITE has now filled this gap; the presence of postthrombotic syndrome, as recorded by a patient questionnaire, using many of the patient symptoms included in the Villalta scale, was a prospective secondary end point in Home-LITE. 11 Our literature search also retrieved several smaller studies that reported post-thrombotic syndrome-related outcomes after long-term anticoagulation. Ideally, future studies should include development of post-thrombotic syndrome as a prespecified end point. We were obliged to analyze our results in groups, according to the different end points reported. We will now consider each end point in turn. Ulcers Pooled analysis of 2 studies, reporting the incidence of venous ulcers at 3 or 12 months of follow-up, yielded a significant risk reduction of 87% using low-molecularweight heparin versus oral anticoagulation (P.019). The incidence of venous ulcers at 3 months in the Home- LITE study is of particular interest. 11 The Scientific and Standardization Committee recommends that post-thrombotic syndrome should not be diagnosed before 3 months to avoid inappropriately attributing acute symptoms and signs to post-thrombotic syndrome. 28 It is clearly difficult to distinguish whether early signs and symptoms after a deep vein thrombosis are acute manifestations of deep vein thrombosis or due to a recurrent deep vein thrombosis or evolving post-thrombotic syndrome. In the Home-LITE study, however, rates of recurrent venous thromboembolism were identical at 3 months (3.3% in each treatment group),

8 Hull et al Effect of Long-term Low-Molecular-Weight Heparin on Post-Thrombotic Syndrome 763 whereas rates of patient-reported venous ulcer were clearly different (0.5% with low-molecular-weight heparin and 4.1% with oral anticoagulation, P.019). 11 Because a venous ulcer indicates severe post-thrombotic syndrome, regardless of other signs and symptoms, in various scoring systems, 9,28,29 this finding suggests that post-thrombotic syndrome occurs earlier than previously recognized, and that it can be modified by using low-molecular-weight heparin. We suggest that patients who develop any such signs and symptoms of post-thrombotic syndrome within the first 6 months after the initial deep vein thrombosis should be treated with long-term low-molecular-weight heparin, regardless of whether the specific diagnosis can be confirmed, to limit further development of post-thrombotic syndrome, as well as treating any recurrent deep vein thrombosis that may be present. Presence of Signs and Symptoms of Post- Thrombotic Syndrome Three studies reported nonsignificant differences between treatment groups in the incidence of signs and symptoms of post-thrombotic syndrome In the questionnaire used in Home-LITE, however, all 8 patient-reported assessments of the presence of symptoms of post-thrombotic syndrome showed a trend favoring low-molecularweight heparin over oral anticoagulation, and the overall odds ratio of 0.77 (95% CI, ) favored lowmolecular-weight heparin (P.001). 11 The difference between low-molecular-weight heparin and usual care could conceivably widen if the questions were asked again after longer follow-up. Rodger et al 30 showed that assessments of post-thrombotic syndrome based on combinations of physical findings have good to excellent interobserver reliability, and the Scientific and Standardization Committee has recommended the Villalta scale for defining the post-thrombotic syndrome in clinical studies because it has been shown to be reliable, valid, acceptable, and responsive for measuring post-thrombotic syndrome in patients with previous, objectively confirmed deep vein thrombosis, 28 a recommendation that we support. In clinical practice, if scoring using the Villalta scale is not practical, Hull s 8-point patient questionnaire could provide a more useful tool for patient or health care provider monitoring of changes in individual patients symptoms over time. Recanalization Our pooled analysis of 5 studies showed a risk ratio of 0.66 (95% CI, ; P.0001) in favor of low-molecularweight heparin versus usual care for the likelihood of complete recanalization of thrombosed veins. Since we completed our study, Romera-Villegas et al 16 have published an evaluation of long-term treatment of venous thromboembolism with lowmolecular-weight heparins or oral anticoagulation. They reported pooled results for thrombus regression at 3 to 6 months as follows: relative risk of 0.49 (95% CI, ) for 3 studies using full-dose low-molecular-weight heparin and relative risk of 0.73 (95% CI, ) for 3 studies using prophylactic doses of low-molecular-weight heparin. Although we could not reproduce their analyses exactly, because we did not have all the original data, the results, showing a substantial difference in recanalization in favor of low-molecular-weight heparins, are in agreement with our analysis. A reduction in thrombus size of 30% has been used in several trials as a sign of clinical benefit for the individual patient. 21 Residual venous thrombosis in patients with previous deep vein thrombosis is a predictive risk factor for post-thrombotic syndrome, and greater thrombus regression is associated with lower development of the post-thrombotic syndrome. 21 González-Fajardo et al 21 suggest that residual thrombosis probably reflects an underlying hypercoagulable state or reduced thrombolytic capacity that increases patients risk for recurrent events and the postthrombotic syndrome. Indeed, a study by Breddin et al 31 of coagulation parameters and venographic outcomes after treatment of deep vein thrombosis suggested that a 30% decrease in Marder score is consistent with a reduction in hypercoagulation. Hull et al 32 conducted a meta-analysis of trials that compared short-term therapy of deep vein thrombosis with lowmolecular-weight heparin with unfractionated heparin. They showed that the greater improvement in clot burden (assessed by venography before and after initial treatment) obtained with low-molecular-weight heparin was associated with a lower rate of recurrent venous thromboembolism on follow-up. Recurrent venous thromboembolism is an important predictor of post-thrombotic syndrome. 6,9,33 The apparent superiority of long-term low-molecularweight heparin over oral anticoagulation in preventing development of different post-thrombotic syndrome-related end points suggests that effects of low-molecular-weight heparins beyond their anticoagulant activity may be involved in reducing the risk of post-thrombotic syndrome. Low-molecularweight heparins are known to stimulate the release of tissue factor pathway inhibitor, a molecule with anti-inflammatory and anti-angiogenic effects, 34 and may reduce the degree of inflammation and subsequent damage to the veins. Degree of Reflux/Venous Competence Two studies showed possible trends for improvement in reflux/venous competence with low-molecular-weight heparin. 22,26 Another study reported that significantly more patients exhibited venous competence with tinzaparin versus oral anticoagulation (P.001). 19 The link between reflux and post-thrombotic syndrome remains unclear. Venous valvular incompetence has low specificity for post-thrombotic syndrome; however, the presence of valvular incompetence helps to confirm the diagnosis of post-thrombotic syndrome in symptomatic patients. 5

9 764 The American Journal of Medicine, Vol 124, No 8, August 2011 CONCLUSIONS The main shortcoming of our study has been mentioned: Post-thrombotic syndrome-related end points were diagnosed and reported differently by different authors, and full data were not always included. Therefore, meta-analysis was possible only across small numbers of studies, and some results were compared qualitatively. Nevertheless, long-term treatment was significantly better, or tended to be better, with low-molecular-weight heparin than with oral anticoagulation for every end point that we considered. The analyses of underlying pathophysiologic changes also are consistent with the clinical findings of the different studies. In particular, they provide a rationale for the lower incidence of post-thrombotic syndrome and venous ulcers with long-term low-molecular-weight heparin versus oral anticoagulation that was reported in Home-LITE. 11 Our findings suggest that long-term treatment with low-molecular-weight heparin rather than oral anticoagulation after a deep vein thrombosis may reduce or possibly even prevent development of post-thrombotic syndrome. Given the scarcity of options for successfully treating post-thrombotic syndrome, this finding could be of considerable clinical and economic benefit. We suggest that long-term low-molecular-weight heparin therapy be considered in particular for any patients presenting with severe post-thrombotic syndrome symptoms (because greater post-thrombotic severity at 1 month has been shown to predict higher post-thrombotic syndrome scores over time). 2 Furthermore, the suggestion that post-thrombotic syndrome may develop more rapidly after a deep vein thrombosis than has been recognized represents a paradigm shift in current understanding. Early signs and symptoms should not automatically be assumed to be acute signs and symptoms of the index event or of a recurrent deep vein thrombosis, and patients exhibiting such symptoms should be managed with therapeutic doses of long-term low-molecular-weight heparin. References 1. Van Dongen CJJ, Prandoni P, Frulla M, et al. Relation between quality of antithrombotic treatment and the development of the postthrombotic syndrome. J Thromb Haemost. 2005;3: Kahn SR, Shrier I, Julian JA. Determinants and time course of the postthrombotic syndrome after acute deep venous thrombosis. Ann Intern Med. 2008;149: Prandoni P, Kahn S. Post-thrombotic syndrome: prevalence, prognostication and need for progress. Br J Haematol. 2009;145: Kolbach DN, Sandbrink MWC, Hamulyak K, et al. Non-pharmaceutical measures for prevention of post-thrombotic syndrome. Cochrane Database Syst Rev. 2003, Issue 3. Art. No.: CD DOI: / CD pub2. 5. Kahn SR, Ginsberg J. Relationship between deep venous thrombosis and the postthrombotic syndrome. Arch Intern Med. 2004;164: Kahn SR. How I treat postthrombotic syndrome. Blood. 2009;114: Kahn SR, Ginsberg J; Thrombosis interest Group of Canada. TIGC Guideline on the Post-Thrombotic Syndrome, August Available at: Accessed December 11, Kearon C, Kahn S, Agnelli G, et al. Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edn). Chest. 2008;133; 454S-545S. 9. Prandoni P, Lensing AW, Prins MH, et al. Below-knee elastic compression stockings to prevent the post-thrombotic syndrome: a randomized, controlled trial. Ann Intern Med. 2004;141: Vedantham S. Interventional approaches to acute venous thromboembolism. Semin Respir Crit Care Med. 2008;29: Hull RD, Pineo GF, Brant RF, et al for the LITE Trial Investigators. Home therapy of venous thrombosis with long-term LMWH versus usual care: patient satisfaction and post-thrombotic syndrome. Am J Med. 2009;122: Akl EA, Barba M, Rohilla S, et al. Anticoagulation for the long term treatment of venous thromboembolism in patients with cancer. Cochrane Database Syst Rev. 2008, Issue 2. Art. No.: CD DOI: / CD pub Ferretti G, Bria E, Giannarelli D, et al. Is recurrent venous thromboembolism after therapy reduced by low-molecular-weight heparin compared with oral anticoagulants? Chest. 2006;130: Segal JB, Streiff MB, Hofmann LV, et al. Management of venous thromboembolism: a systematic review for a practice guideline. Ann Intern Med. 2007;146: Van der Heijden JF, Hutten BA, Büller HR, Prins MH. Vitamin K antagonists or low-molecular-weight heparin for the long term treatment of symptomatic venous thromboembolism. Cochrane Database Syst Rev. 2001, Issue 3. Art. No.: CD DOI: / CD Romera-Villegas A, Cairols-Castellote MA, Vila-Coll R, et al. Longterm use of different doses of low-molecular-weight heparin versus vitamin K antagonists in the treatment of venous thromboembolism. Ann Vasc Surg. 2010;24: Das SK, Cohen AT, Edmondson RA, et al. Low-molecular-weight heparin versus warfarin for prevention of recurrent venous thromboembolism: a randomized trial. World J Surg. 1996;20: Bellosta R, Ferrari P, Luzzani L, et al. Home therapy with LMWH in deep vein thrombosis: randomized study comparing single and double daily administrations. Angiology. 2007;58: Romera A, Cairols MA, Vila-Coll R, et al. A randomised open-label trial comparing long-term sub-cutaneous low-molecular-weight heparin compared with oral-anticoagulant therapy in the treatment of deep venous thrombosis. Eur J Vasc Endovasc Surg. 2009;37: Schobess R, Düring C, Bidlingmaier C. Long-term safety and efficacy data on childhood venous thrombosis treated with a low molecular weight heparin: an open-label pilot study of once-daily versus twicedaily enoxaparin administration. Haematologica. 2006;91: González-Fajardo JA, Martin-Pedrosa M, Castrodeza J, et al. Effect of the anticoagulant therapy in the incidence of post-thrombotic syndrome and recurrent thromboembolism: Comparative study of enoxaparin versus coumarin. J Vasc Surg. 2008;48: Daskalopoulos ME, Daskalopoulou SS, Tzortzis E, et al. Long-term treatment of deep venous thrombosis with a low molecular weight heparin (tinzaparin): a prospective randomized trial. Eur J Vasc Endovasc Surg. 2005;29: Daskalopoulos ME, Daskalopoulou SS, Prezas L, et al. Three-year follow-up of long-term deep venous thrombosis treatment with the low molecular weight heparin tinzaparin. Angiology. 2006;25(Suppl 1): Daskalopoulos ME, Daskalopoulou SS, Liapis CD. Tinzaparin in longterm treatment of deep venous thrombosis. Eur J Vasc Endovasc Surg. 2007;34: Gonzalez-Fajardo JA, Arreba E, Castrodeza J, et al. Venographic comparison of subcutaneous low-molecular weight heparin with oral anticoagulant therapy in the long-term treatment of deep venous thrombosis. J Vasc Surg. 1999;30:

10 Hull et al Effect of Long-term Low-Molecular-Weight Heparin on Post-Thrombotic Syndrome López-Beret P, Orgaz A, Fontcuberta J, et al. Low molecular weight heparin versus oral anticoagulants in the long-term treatment of deep venous thrombosis. J Vasc Surg. 2001;33: Kakkar VV, Gebska M, Kadziola Z, et al. Bemiparin Investigators. Low-molecular-weight heparin in the acute and long-term treatment of deep vein thrombosis. Thromb Haemost. 2003;89: Kahn SR, Partsch H, Vedantham S et al.; Subcommittee on Control of Anticoagulation of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Definition of post-thrombotic syndrome of the leg for use in clinical investigations: a recommendation for standardization. J Thromb Haemost. 2009;7: Brandjes DP, Büller HR, Heijboer H, et al. Randomised trial of effect of compression stockings in patients with symptomatic proximal-vein thrombosis. Lancet. 1997;349: Rodger MA, Kahn SR, Le Gal G, et al. Inter-observer reliability of measures to assess the post-thrombotic syndrome. Thromb Haemost. 2008;100: Breddin HK, Kadziola Z, Scully M, et al. Risk factors and coagulation parameters in relationship to phlebographic response and clinical outcome in the treatment of acute deep vein thrombosis. Thromb Haemost. 2003;89: Hull RD, Marder VJ, Mah AF, et al. Quantitative assessment of thrombus burden predicts the outcome of treatment for venous thrombosis: a systematic review. Am J Med. 2005;118: Prandoni P, Lensing AW, Cogo A, et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med. 1996;125: Mousa SA, Bozarth J, Barrett JS. Pharmacodynamic properties of the low molecular weight heparin, tinzaparin: effect of molecular weight distribution on plasma tissue factor pathway inhibitor in healthy human subjects. J Clin Pharmacol. 2003;43: Appendix SUPPLEMENTARY DATA Supplementary data associated with this article can be found, in the online version, at doi: /j.amjmed