Optimal Dosing of Bemiparin As Prophylaxis Against Venous Thromboembolism in Surgery for Cancer: an Audit of Practice

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector International Journal of Surgery (2007) 5, 114e119

www.theijs.com

Optimal dosing of bemiparin as prophylaxis against venous thromboembolism in surgery for cancer: An audit of practice

J.L. Balibrea a,*, J. Altimiras b, I. Larruzea c,*,A.Go´mez-Outes d, J. Martı´nez-Gonza´lez d, E. Rocha e, on behalf of the Bemiparin Cooperative Study Group in Surgery for Cancerf

a Department of Surgery, School of Medicine, Universidad Complutense, Madrid, Spain b Department of Pharmacy, Corporacio´ Sanitaria Parc Taulı´, Sabadell, Barcelona, Spain c Department of Surgery, Hospital de San Eloy, Baracaldo, Vizcaya, Spain d Medical Department, Laboratorios Farmace´uticos Rovi, S.A., Madrid, Spain e Haematology Service, University Clinic of Navarra, Pamplona, Spain

KEYWORDS Abstract Aims: Low-molecular-weight heparins are drugs of ﬁrst choice for thromboprophy- Low-molecular-weight laxis in cancer surgery. We sought to determine the optimal use of bemiparin in cancer surgery heparin; in standard clinical practice. Bemiparin; Patients and methods: A retrospective, multicentre audit on the use of bemiparin in patients Venous undergoing cancer surgery and given prophylaxis with bemiparin was undertaken. Surgeons’ as- thromboembolism; sessment of venous thromboembolic (VTE) risk (moderate or high) was compared to the crite- Cancer surgery ria of current Consensus Guidelines for VTE management. We assessed the incidence of documented symptomatic VTE, bleeding events, thrombocytopenia, deaths and total events related to VTE or bemiparin prophylaxis (i.e. bleeding, thrombocytopenia). The potential economic impact of postoperative vs. preoperative bemiparin was also analysed. Results: Clinical records from 197 patients from 5 Spanish centres were checked. Prophylaxis was started postoperatively in 45 patients (22.8%). According to the surgeons’ criteria, 73 (37.1%) patients were at high VTE risk and received bemiparin 3500 IU/d. However, according to the criteria of current Guidelines, 189 (95.9%) patients were at high risk of VTE (heterogeneity P-value < 0.0001). Three (1.5%) patients, all of them receiving bemiparin 2500 IU/d, developed a symptomatic conﬁrmed VTE. There were 4 major and 5 minor bleeding events during bemiparin prophylaxis. A lower incidence of bleeding (2.2% vs. 5.3%; P Z 0.48) and total events (2.2% vs. 9.9%; P Z 0.11) was seen with bemiparin started postoperatively as compared

* Corresponding authors. Department of Surgery, School of Medicine, Universidad Complutense, C/Dr. Martı´n Lagos s/n, 28040 Madrid, Spain. Tel./fax: þ34 91 3303177 (J.L. Balibrea). E-mail address: [email protected] (J.L. Balibrea). f Members of the Bemiparin Cooperative Study Group in Surgery for Cancer are listed in the Appendix.

1743-9191/$ - see front matter ª 2006 Published by Elsevier Ltd on behalf of Surgical Associates Ltd. doi:10.1016/j.ijsu.2006.07.005 Bemiparin thromboprophylaxis in cancer surgery 115

to preoperative bemiparin. Bleeding rates did not signiﬁcantly differ between patients given low or high bemiparin prophylactic doses (4.0% vs. 5.5%; P Z 0.72). Two patients died due to cardio-respiratory failure and sepsis, respectively. Postoperative bemiparin provided net cost savings of V909 per patient compared to preoperative start of prophylaxis due to shorter hospital stays (9 vs. 11 days) and lower incidence of complications in the postoperative bemiparin group. Conclusions: Many cancer patients are still poorly assessed for risk of VTE. Bemiparin 3500 IU/d is associated with a lower incidence of VTE without signiﬁcant increase in complications as compared with bemiparin 2500 IU/d. Postoperative bemiparin prophylaxis seems to be as effective and safer than preoperative start of prophylaxis. Further prospective clinical studies are needed to fully address this issue. ª 2006 Published by Elsevier Ltd on behalf of Surgical Associates Ltd.

Introduction Patients received open label prophylaxis with bemiparin (Hibor, Laboratorios Farmaceúticos Rovi, Madrid, Spain) at Low-molecular-weight heparins (LMWH) are effective and the daily subcutaneous dose of 2500 IU or 3500 IU, depend- safe for prophylaxis of venous thromboembolism (VTE) in ing on the degree of risk according to surgeons’ criteria. 1,2 An additional stratification of thromboembolic risk was per- surgical patients, and represent a thromboprophylactic 3 3,4 formed based on the ACCP Consensus Conference and the method of first choice. 4 Cancer patients have an increased risk of both post- International Consensus Statement. According to these lat- operative VTE5 and bleeding during anticoagulant therapy6 ter criteria, major cancer surgery in patients over 40 years compared to cancer-free patients. Without prophylaxis, the of age was considered high-risk surgery. Major cancer sur- overall prevalence of systematically detected deep vein gery in patients under 40 years or minor cancer surgery in thrombosis (DVT) estimated by the fibrinogen uptake test patients over 40 years of age was considered as moderate in cancer surgery is 35%.7 Compared to placebo, LMWH risk surgery. Risk assessment by the surgeons and risk factor prophylaxis provides relative risk reductions of 81% in veno- assessment based on current VTE Consensus Guidelines graphic DVT, 71% in clinical VTE, 68% in fatal PE and 60% in were compared. Prophylaxis was maintained based on death by any cause.7e10 By contrast, major bleeding events investigator’s criteria, during the risk period or until full mo- increase from 1.4% with placebo to approximately 3.0% in bilization of the patient. Data were recorded in 2 periods: patients receiving an LMWH,7e10 and to 8.1% with unfractio- the first covered the hospitalisation time, and the second nated heparin (UFH) in cancer surgery.2 However, bleeding period comprised the time from hospital discharge to the usually follows a benign clinical course, and it seems rea- first outpatient visit. We defined 2 types of cohorts: depend- sonable to give priority to pharmacological prophylaxis, ing on the timing of first bemiparin administration (post- vs. since mortality rates are significantly reduced. Therefore, preoperative) and depending on the bemiparin dose used the use of new strategies to reduce bleeding rates in cancer (2500 IU vs. 3500 IU). Data on the dose and timing of first surgery should be encouraged. bemiparin administration were extracted from the medica- Bemiparin (Hibor, Ivor, Zibor, and Badyket) is a new tion orders and medication administration records, which ‘‘second generation’’11 LMWH with a low bleeding tendency were available from patients’ hospital medical records. that can be initiated pre- or postoperatively,11,12 without The efficacy and safety outcomes were calculated in differ- compromising efficacy.11e13 Postoperative start of bemi- ent cohorts (post- vs. preoperative, and 3500 IU vs. 2500 IU). parin prophylaxis should theoretically minimise the risk of Finally, a potential correlation of the type of surgery or con- spinal haematoma and the perioperative bleeding rates. comitant medication on bleeding rates was analysed. Thus, it could be of great interest to compare pre- and Concomitant medications, including drugs that could postoperative start of bemiparin prophylaxis, particularly alter haemostasis (analgesics or anti-platelet drugs), and as regards the relative incidence of VTE and bleeding. use of non-pharmacological treatments, such as elastic This study was designed to describe the patterns of use bandages or intermittent pneumatic compression devices, of bemiparin in cancer surgery in standard clinical practice. were extracted from clinical records. We also analysed the influence of bemiparin dose and The outcomes collected were as follows: symptomatic timing of first administration on clinical and economic VTE (DVT and/or PE) confirmed by objective testing (Dopp- outcomes. ler ultrasound or ascending contrast venography for DVT and high-probability lung scanning, pulmonary angiography for PE, or necropsy in fatal cases); major bleeding (defined as Study design and methods overt bleeding associated with a decrease in haemoglobin levels by 2 g per decilitre or more or requiring transfusion of This study was designed as a retrospective, multicentre, 2 or more units of packed red cells or whole blood, bleeding cohort study. We checked clinical records from patients requiring reoperation, fatal, retroperitoneal or intracranial who underwent surgery for cancer and had received bleeding, or bleeding requiring treatment discontinuation); thromboprophylaxis with bemiparin. Since this was a retro- minor bleeding (if it was overt but did not meet the other spective audit of practice, there were no pre-established criteria for major bleeding); moderate thrombocytopenia exclusion criteria. (platelet count lower than 100,000 platelets per cubic 116 J.L. Balibrea et al. millilitre); severe thrombocytopenia (platelet count of less Table 1 Demographic data and baseline characteristics than 50,000 per cubic millilitre); and deaths. We also analysed total events related to VTE or bemiparin prophylaxis Characteristic Total (N Z 197) (i.e. bleeding, thrombocytopenia) in each cohort. Age e yr. A minimisation cost analysis of preoperative versus post- Median 70 operative start of bemiparin prophylaxis was performed by Range 15e95 analysing the summary costs for treatment and complica- Sex e male/female 111/86 tions derived from the study. Unit costs for treatments and e complications were adopted from the average wholesale Risk factor no. (%) price of treatments14 and from published medical resource Age > 75 yr. 65 (33%) cost data.15,16 The results were expressed in terms of costs History of venous 2 (1%) e and incremental cost-differences. The study setting was thromboembolism no. (%) the Spanish National Health System. Thus, only direct med- Obesity 32 (16.2%) ical costs were included. The time horizon chosen was 40 Varicose veins 8 (4.1%) days. Costs were given in euros for the year 2002, according Chronic heart failure 11 (5.6%) to the latest Spanish clinical and administrative data sources Chronic obstructive 7 (3.6%) available. Univariate sensitivity analyses were used to iden- lung disease tify important model uncertainties. Hormone-replacement therapy 0 (0%) 2 Risk factors e no. (%) 21 (10.7%) Statistical analysis Chemotherapy 34 (17.3%) Radiotherapy 11 (5.6%) Absolute risks of VTE, death, bleeding events and other Major surgery e n (%) 195 (99.0%) safety outcomes were shown. Subgroups were compared Type of surgery e n (%) depending on the dosage schedule (pre- vs. postoperative) Gastrointestinal tract 134 (68%) and the dose used (2500 IU vs. 3500 IU). A two-sided c2 test Female reproductive organs 21 (10.7%) or a Fisher exact test was used for qualitative variables, Genitourinary tract 27 (13.7%) and a Student’s test was used for quantitative variables. Lungs and/or bronchial tree 13 (6.6%) A P-value of less than 0.05 was considered significant for Other 7 (3.6%) all statistical tests. These calculations were performed 2 Sites 17 (8.6%) using SPSS software, version 7.5. Palliative surgery e no. (%) 12/190 (6.3%) Results Duration of surgery Median 2 h 50 min Range 20 min to 6 h Study population Type of anaesthesia A total of 197 patients undergoing cancer surgery were General 182 (93.3%) retrospectively included in the study in 5 Spanish centres Neuraxial 13 (6.7%) (see Appendix). Hospital stay, 10 (7, 14.5) Demographic characteristics are shown in Table 1. Major days e median (1st, 3rd quartile) cancer surgery was performed in 189 patients (95.9%). The most commonly involved organs were colon (50 cases), Prophylaxis information rectum (39 cases), sigmoid (33 cases), stomach (16 cases), Start of prophylaxis preoperative 152 (96/56) lungs (13 cases), breasts (12 cases), urinary bladder and/or (2500 IU/3500 IU), n ureters (13 cases), prostate (10 cases) and uterus and/ Start of prophylaxis postoperative 45 (28/17) or ovaries (6 cases). (2500 IU/3500 IU), n All the 197 patients were administered open label Prophylactic dose used 124/73 prophylaxis with bemiparin (Table 1). Hospital stay was 2 2500/3500 IU, n/n days shorter in patients given postoperative bemiparin as Duration of prophylaxis, 12 (2, 51) compared to those receiving the preoperative schedule median (range) (9.0 vs. 11.0 days; median). Patients receiving NSAIDs, n (%) 104 (52.8%) In the surgeons’ opinion, 124 (63%) patients had NSAID Z non-steroidal anti-inflammatory drug. a moderate risk of VTE and were therefore given bemiparin 2500 IU/d, while 73 (37.1%) patients were at high risk and received bemiparin 3500 IU/d. However, according to the criteria of current VTE Consensus Guide- lines, 189 (95.9%) patients were at high risk of VTE Guidelines was statistically significant (heterogeneity P- (>40 years undergoing major cancer surgery) and 8 value < 0.0001), resulting in bemiparin under-dosage in (4.1%) patients were at moderate risk of VTE (6 pa- 116 (59%) patients. tients < 40 years and 2 patients undergoing minor cancer One hundred and four (52.8%) of the 197 patients were surgery). Heterogeneity between investigators’ risk as- given an analgesic agent for postoperative pain relief. sessment and the criteria of current VTE Consensus Metamizole was the most commonly used analgesic. Bemiparin thromboprophylaxis in cancer surgery 117

Hospitalisation period did not develop further thrombotic or haemorrhagic complications. One patient with gastric cancer, who had Three of 197 patients developed a documented symptomatic received bemiparin 2500 IU before surgery, died 24 days VTE: 2 patients had distal DVT, and the remaining patient after urgery due to cardio-respiratory failure during had proximal DVT complicated with non-fatal PE (Table 2). In hospitalisation. all 3 cases of symptomatic VTE, patients were at high risk but Postoperative administration showed a trend to a lower received the 2500 IU bemiparin dose. There were no cases of incidence of total events related to VTE or bemiparin fatal PE. The incidence of thromboembolic events was lower prophylaxis compared to preoperative administration in the postoperative bemiparin group, but the difference [2.2% (1 of 45) vs. 9.9% (15 of 152); difference, 7.7%; was not statistically significant (0.0% vs. 2.0%; difference, 95% CI, 14% to 2.3%; P Z 0.11]. In addition, the number of 2.0%, 95% CI, 6.2% to 5.7%, P Z 0.39). The incidence of total events was also lower in patients treated with the thromboembolic events was lower among the patients re- 3500 IU bemiparin dose compared to 2500 IU dose [4.1% ceiving bemiparin 3500 IU as compared to 2500 IU, but the (3 of 73) vs. 10.5% (13 of 124); difference, 6.4%; 95% CI, difference was not statistically significant (0% vs. 2.4%; dif- 13.9% to 2.1%; P Z 0.12]. ference, 2.4%; 95% CI, 7.2% to 3%; P Z 0.20). Nine (4.6%) of 197 patients experienced a bleeding event during bemiparin prophylaxis (95% CI, 2.1e8.5%). All such Outpatient period events were related to the surgical wound or occurred intraoperatively. Four events (2.0%), all of them occurring The median period from surgery to the first outpatient visit e after preoperative bemiparin dosing, were major (Table 2). was 40 days (range: 20 95). There were no cases of Major bleeding cases were 3 intra-abdominal bleedings symptomatic VTE or bleeding events during the period after colon surgery that required repeat surgery, and 1 from hospital discharge on day 10 (median) to the first surgical wound bleeding after rectal surgery requiring outpatient visit on day 40 (median). One patient operated transfusion of 3 units of red packed cells. There were no for an oesophageal neoplasm who was administered bemi- cases of fatal, retroperitoneal or intracranial bleeding. parin 2500 IU before surgery died because of pneumonia Postoperative administration showed a trend to lower and sepsis 20 days after hospital discharge. bleeding rates compared to preoperative start of prophylaxis, but the differences were not statistically significant Pharmacoeconomic analysis (Pre- vs. [2.2% (1 of 45) vs. 5.3% (8 of 152); difference, 3.1%; 95% CI, 8.4% to 6.8%; P Z 0.47]. Bleeding rates did not signif- postoperative start of prophylaxis) icantly differ between 3500 and 2500 IU bemiparin doses [5.5% (4 of 73) vs. 4.0% (5 of 124); difference, 1.5%; 95% The rates of clinical complications for post- and preoper- CI, 5% to 9.6%; P Z 0.72]. ative bemiparin are summarised in Table 2. Bleeding risk was higher, but not statistically significant, In order to calculate the cost of treatment with LMWH, in gastrointestinal tract surgery as compared to other types hospital stay costs, pharmacy costs and complications costs of surgery (RR, 3.8; P Z 0.19). Treatment with analgesics were added. and/or anti-platelet drugs was not a risk factor favouring Pharmacoeconomic results are given as costs and cost- bleeding (RR, 0.45; P Z 0.26). Only 13 patients underwent differences per patient administered prophylaxis for VTE epidural anaesthesia. None of these patients experienced with either post- or preoperative bemiparin (Table 3). Post- bleeding events. operative bemiparin provided net cost savings of V909 per One patient with colon cancer and hepatic cirrhosis patient compared to preoperative bemiparin when costs de- developed moderate thrombocytopenia (56,000 platelets rived from treatment and complications during the 40-day per cubic millilitre) while receiving bemiparin 2500 IU for observation period were considered (V3297 vs. V4206). 13 days. Treatment was not discontinued and patient Net cost savings were mainly attributable to shorter hospital

Table 2 Overall incidence of clinical events, baseline probabilities for postoperative vs. preoperative bemiparin and ranges for clinical outcomesa Outcome Total (N Z 197) Post-op. Bemiparin Pre-op. Bemiparin Difference (95% CI)b n (%) (%) (%) (%) Venous thromboembolism 3 (1.5)a 0.0 2.0 2.0 (6.2 to 5.7) Pulmonary embolism 1 (0.5) 0.0 0.7 0.7 (7.6 to 3.7) Deep vein thrombosis 3 (1.5) 0.0 2.0 2.0 (6.2 to 5.7) Total bleeding 9 (4.6) 2.2 5.3 3.1 (8.4 to 6.8) Major bleeding 4 (2.0) 0.0 2.6 2.6 (6.6 to 6.0) Minor bleeding 5 (2.5) 2.2 2.6 0.4 (9.2 to 5.0) Thrombocytopenia 1 (0.5) 0.0 0.7 0.7 (7.6 to 3.7) Death from any cause 2 (1.0) 0.0 1.3 1.3 (6.8 to 4.7) a The patient with non-fatal PE had also proximal DVT, thus the composite VTE is 3 and not 4. b A negative difference favours postoperative bemiparin. 118 J.L. Balibrea et al.

optimal measure to minimise the risk of bleeding in cancer Table 3 Results of the cost-difference analysis in euros surgical patients, without compromising efficacy. (2002 values) Postoperative bemiparin was cost-saving, and shortened Variable Post-op. Pre-op. Differencea hospital stay by 2 days (mean), compared to preoperative bemiparin bemiparin bemiparin cohort. This is a result of great interest for major 21 Treatment costs ambulatory surgery. Hospital stay costs 3195 3898 703 Although the timing of the first administration of 22 Pharmacy costs 94 118 24 thromboprophylaxis is controversial, it is recognised that bemiparin administration started 6 h postoperatively Complications costs does not compromise efficacy, it is safe and compatible Pulmonary embolism 0 25 25 with neuraxial anaesthesia techniques.13 On the contrary, Deep vein thrombosis 0 37 37 patients on preoperative LMWH thromboprophylaxis can Total bleeding 8 32 24 be assumed to have altered coagulation. In these patients, Thrombocytopenia 0 2 2 needle placement should occur at least 10e12 h after the Death from any cause 0 94 94 LMWH dose.19,20 Thromboprophylaxis started less than 6 h Total costs, V 3297 4206 909 after surgery may increase the risk of bleeding without im- proved efficacy.22 Initiation 12e24 h postoperatively may a A negative difference favours postoperative bemiparin pro- be less effective than initiation at 6 h.22 phylaxis, representing cost savings. This study has several methodological limitations as follows: its retrospective nature, the relatively small sam- stays (9.0 vs. 11.0 days; median) and to a lower incidence of ple size and low incidence of clinical events, the absence complications in the postoperative bemiparin group. of a comparative group and the selection on the timing and Univariate sensitivity analyses showed that postopera- dose of bemiparin, which was decided by the surgeons tive bemiparin was cost-saving compared to preoperative according to standard clinical practice. However, the results bemiparin in all the ranges tested for complications and of this observational study should be interpreted in the V V costs. Cost savings ranged from 475 to 1237 in the worst setting of normal clinical practice. These types of studies and best case, respectively. add information that is not available in randomised controlled trials where exclusion criteria, limited sample size Discussion and use in specific clinical situation lead to significant loss of useful pragmatic data.23 Therefore, we were able to Most patients in our study were given lower bemiparin doses assess the ‘‘real life’’ effect and use of bemiparin prophylaxis than recommended by current VTE Consensus Guidelines.3,4 in cancer patients. Our observation is consistent with the results from a recent In conclusion, many cancer patients are still poorly report showing that the risk of VTE in cancer patients is assessed for risk of VTE. It may be simpler in this group to underestimated.17 Thus, physicians are recommended to fol- consider all to be at high risk given the low incidence and low current VTE Consensus Guidelines,3,4 in order to estab- generally benign nature of bleeding complications. Bemi- lish the extent of VTE risk in their patients, and to parin 3500 IU/d is associated with a lower incidence of VTE administer the appropriate LMWH dose (for moderate or without significant increase in complications as compared high risk) according to each LMWH specifications. Bemiparin with bemiparin 2500 IU/d. Postoperative bemiparin prophy- 3500 IU/d administered to patients at high risk of VTE fully laxis seems to be as effective and safer than preoperative prevented documented symptomatic VTE, whereas 3 high- start of prophylaxis and it also seems to be cost-saving. Fur- risk patients experienced a VTE event while receiving ther prospective clinical studies are needed to fully address 2500 IU/d bemiparin dose, recommended for patients with this issue. a moderate risk of VTE. In addition, bleeding rates did not significantly differ between low and high bemiparin prophylac- Ethical statement tic doses. Thus, no evidence supporting administration of bemiparin 2500 IU to patients at high risk of VTE was found. This study is a retrospective audit of practice, therefore, no There are several factors related to prophylaxis (e.g. informed consent or approval by an Ethics Committee was pre- vs. postoperative start of prophylaxis) and anaesthetic required. procedures (general vs. neuraxial anaesthesia) that could alter bleeding rates in patients undergoing surgery. As regards relationship between type of anaesthesia and Acknowledgement surgical bleeding, in a previous study with bemiparin in orthopaedic surgery, neuraxial anaesthesia significantly Supported by a research grant of Laboratorios Farmaceúti- decreased operative blood loss compared to general anaes- cos Rovi, S.A., Spain. thesia.18 Most patients in our study did not receive neuraxial anaesthesia, although this practice should be considered in order to reduce blood loss in these patients. Postoperative Appendix start of bemiparin prophylaxis 6 h after surgery is consistent with current recommendations to minimise the risk of The members of the Bemiparin Cooperative Study Group spinal haematoma when neuraxial anaesthesia and throm- in Surgery for Cancer (all in Spain) were as follows: boprophylaxis are used together19,20 and could be an Dr. J. Altimiras, Hospital General de Catalunã, Sant Cugat Bemiparin thromboprophylaxis in cancer surgery 119 del Valle´s, Barcelona (currently working for Corporacio´ 11. Plane`s A. Review on bemiparin sodium e a new second gener- Sanitaria Parc Taulı´, Sabadell, Barcelona); Dr. I. Larruzea, ation low-molecular-weight heparin e and its applications in Dr. F. Ramos, Hospital de San Eloy, Barakaldo, Vizcaya; Dr. I. venous thromboembolism. Expert Opin Pharmacother 2003;4: e Portal, Hospital Mutua de Terrassa, Barcelona; Dr. F. Ara- 1551 61. 12. Chapman TM, Goa KL. Bemiparin: a review of its use in the pre- nzadi, Hospital del Bidasoa, Iruń, Guipu´zcoa; Dr. I. de Diego, vention of venous thromboembolism and treatment of deep Clınica Santa Marıa de la Asuncioń, Tolosa, Guipuzcoa. ´ ´ ´ vein thrombosis. Drugs 2003;63:2357e77. 13. Navarro-Quilis A, Castellet E, Rocha E, Paz-Jimeńez J, Plane`s A. Efficacy and safety of bemiparin compared with References enoxaparin in the prevention of venous thromboembolism after total knee arthroplasty. A randomized double-blind 1. Nurmohamed MT, Rosendaal FR, Buller HR, Dekker E, clinical trial. J Thromb Haemost 2003;1:425e32. Hommes DW, Vandenbroucke JP, et al. Low molecular weight 14. General Council of Official Associations of Pharmacists. Cata- heparin versus standard heparin in general and orthopaedic logue of pharmaceutical products. Madrid: General Council surgery: a meta-analysis. Lancet 1992;340:152e6. of Official Associations of Pharmacists; 2002. 2. Mismetti P, Laporte S, Darmon JY, Buchmuller A, Decousus H. 15. Order of 26th December 2001 of the General Section of the Meta-analysis of low molecular weight heparin in the preven- National Healthcare Institute, on the revision of the prices tion of venous thromboembolism in general surgery. Br J Surg that healthcare centres must apply for the services rendered 2001;88:913e30. in the cases when the sums must be requested from third 3. Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR, parties who are obliged to pay or from users not entitled Colwell CW, et al. Prevention of venous thromboembolism: to healthcare assistance form the National Health System, 5. the Seventh ACCP Conference on antithrombotic and thrombo- BOE. p. 637e9 [in Spanish]. Available from: http://www. lytic therapy. Chest 2004;126(Suppl. 3):338Se400S. boe.es; 2001 [accessed 31.10.03]. 4. Nicolaides AN, Breddin HK, Fareed J, Goldhaber S, Haas S, 16. Foral Order 83/2002, of 15th May, by the Healthcare Advisor, Hull R, et al. Prevention of venous thromboembolism. Interna- pursuant to which the tariffs for services rendered in the tional Consensus Statement. Guidelines compiled in accor- Healthcare Sentences of Servicio Navarro de Salud-Osasunbi- dance with the scientific evidence. Int Angiol 2001;20:1e37. dea are approve, 91. Boletıń Oficial de Navarra. p. 6530e41 5. Prandoni P, Piccioli A, Girolami A. Cancer and venous thrombo- [in Spanish]. Available from: http://www.cfnavarra.es; 2002 embolism: an overview. Haematologica 1999;84:437e45. [accessed 31.10.03]. 6. Prandoni P, Lensing AWA, Piccioli A, Bernardi E, Simioni P, 17. Kirwan CC, Nath E, Byrne GJ, McCollum CN. Prophylaxis for Girolami B, et al. Recurrent venous thromboembolism and venous thromboembolism during treatment for cancer: ques- bleeding complications during anticoagulant treatment in pa- tionnaire survey. BMJ 2003;327:597e8. tients with cancer and venous thrombosis. Blood 2002;100: 18. Kakkar VV, Howes J, Sharma V, Kadziola Z. A comparative, dou- 3484e8. ble-blind, randomized trial of a new-second generation LMWH 7. Marassi A, Balzano G, Mari G, D’Angelo SV, Della Valle P, (bemiparin) and UFH in the prevention of post-operative Di Carlo V, et al. Prevention of postoperative deep vein throm- venous thromboembolism in patients undergoing hip arthro- bosis in cancer patients. A randomized trial with low molecular plasty. Thromb Haemost 2000;83:523e9. weight heparin (CY 216). Int Surg 1993;78:166e70. 19. Tryba M. European practice guidelines: thromboembolism pro- 8. Ho YH, Seow-Choen F, Leong A, Eu KW, Nyam D, Teoh MK. phylaxis and regional anesthesia. Reg Anesth Pain Med 1998; Randomized, controlled trial of low molecular weight heparin 23(Suppl. 2):178e82. vs. no deep vein thrombosis prophylaxis for major colon and 20. Horlocker TT, Wedel DJ, Benzon H, Brown DL, Enneking FK, rectal surgery in Asian patients. Dis Colon Rectum 1999;42: Heit JA, et al. Regional anesthesia in the anticoagulated 196e203. patient: defining the risks. Reg Anesth Pain Med 2004; 9. Ockelford PA, Patterson J, Johns AS. A double-blind random- 29(Suppl. 2):1e12. ized placebo controlled trial of thromboprophylaxis in major 21. Raich-Brufau M, Martıńez-Gonza´lez J, Bustos-Molina F. elective general surgery using once daily injections of a low Encuesta nacional sobre la prevencioń de la enfermedad trom- molecular weight heparin fragment (Fragmin). Thromb Hae- boembo´lica venosa en cirugıá mayor ambulatoria. Cir May Amb most 1989;62:1046e9. 2004;9:31e6. 10. Pezzuoli G, Neri Serneri GG, Settembrini P, Coggi G, Olivari N, 22. Raskob GE, Hirsh J. Controversies in timing of the first dose of Buzzetti G, et al. Prophylaxis of fatal pulmonary embolism in anticoagulant prophylaxis against venous thromboembolism general surgery using low-molecular weight heparin CY 216: after major orthopaedic surgery. Chest 2003;124:379Se85S. a multicentre, double-blind, randomized, controlled, clinical 23. Black N. Why we need observational studies to evaluate the trial versus placebo (STEP). Int Surg 1989;74:205e10. effectiveness of health care. BMJ 2001;312:1215e8.