|
Dermatan sulfate versus unfractionated heparin for the prevention of venous thromboembolism in patients undergoing surgery for cancer: a cost-effectiveness analysis |
Attanasio E, Russo P, Carunchio G, Caprino L |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology Dermatan sulfate (600mg intramuscularly on the second day before surgery, then 300mg once daily) for the prevention of postoperative venous thromboembolism in patients with cancer. Dermatan sulfate is a natural glycosaminoglycan that, unlike unfractionated heparin and low molecular weight heparin, selectively inhibits thrombin through catalysis of endogenous heparin cofactor II.
Economic study type Cost-effectiveness analysis.
Study population The study population consisted of patients aged over 40, scheduled for elective abdominal, thoracic, gynaecological or urological surgery for cancer resection. The planned aim of surgery was resection of primary neoplasms of a proven or presumed malignant nature or resection of pulmonary metastases. Criteria for exclusion were as follows: high bleeding risk (current or recent severe haemorrhage, prothrombin time 1NR of more than 1.3 platelet count lower than l00,000/microl, treatment with oral anticoagulants or ticlopidine in the week before surgery); history of adverse reactions to heparin; renal impairment (creatinine higher than 20 mg/dl); allergy to contrast media; venous thromboembolism in the past 6 months; and pregnancy.
Setting The setting was hospital. The economic analysis was carried out in Italy.
Dates to which data relate Clinical and resource use data were obtained from the Dermatan sulfate in Oncologic Surgery (DOS) study, published in 1999. The DOS patients were enrolled from January 1996 to July 1997. Some of the clinical probabilities incorporated into the model were obtained from the literature published between 1987 and 1997. The price year was 1998.
Source of effectiveness data The evidence for the final outcomes was based on a single study and a literature review.
Link between effectiveness and cost data Costing was conducted retrospectively on the same patient sample as that used in the effectiveness analysis.
Study sample Power calculations were performed to determine the sample size (on the basis of previous expedience in the prevention of postoperative DVT; dermatan sulphate was expected to be at least equally effective but less haemorrhagic than heparin; the expected rate of venous thromboembolism was 22.5% with either treatment; to accept equivalence, the upper 95% confidence limit of the absolute difference between treatments was not to exceed 7.5 percent points; this specification could be met with 80% probability by including 430 patients per group; the same sample would be sufficient to show a reduction from 10 to 5% in the bleeding complication rate, with a two-tailed alpha of 0.05 and 80% power). In total 842 patients were randomised to either the dermatan sulfate group (n=418) with a mean (SD) age of 63 (9.6) years or the UFH group (n=424) with a mean (SD) age of 63 (10.2) years. Of 857 patients randomised, 842 underwent surgery and were therefore analysed.
Study design The DOS study was a multicentre, randomised, open, controlled trial with blind central assessment of the efficacy outcome, performed in 27 Italian centres. An outpatient follow-up visit was planned 4 weeks after hospital discharge to check for the occurrence of symptomatic venous thromboembolism; information was collected by telephone when a patient could not attend the visit. Median treatment duration was 10 days and 8 days for dermatan sulfate and UFH, respectively. Regarding the number lost to follow-up, it was reported that efficacy was assessed in 521 patients (61.9% of those operated on) with adequate bilateral venography and/or confirmed pulmonary embolism (PE) (267 dermatan sulfate, 254 UFH); treatment groups assessed for efficacy did not differ significantly for patient characteristics. Randomisation was stratified by centre. Treatment allocation was performed by opening a sealed envelope on the third day before surgery (Day-3).
Analysis of effectiveness The principle used in the analysis of effectiveness appears to have been both intention to treat and treatment completers only. The health outcome measures were venographically proven deep vein thrombosis (DVT), symptomatic pulmonary embolism (PE) confirmed by objective testing and bleeding complications. Total mortality was also reported. Treatment groups were balanced for the baseline characteristics, except for an excess of male patients in the dermatan sulphate group. In a post-hoc descriptive analysis, factors associated with postoperative DVT risk were explored with a stepwise logistic procedure. Age, sex, type of surgery, duration of anaesthesia, thromboembolic risk factors, treatment compliance and use of elastic stockings were included in the model, in addition to treatment and centre.
Effectiveness results The effectiveness results were as follows:
Postoperative DVT was detected in 40 patients (15%; 95% CI: 10.7 - 19.3%) receiving dermatan sulfate versus 56 receiving UFH (22%; 95% CI: 16.9 - 27.1%) (p=0.033).
Relative risk reduction was 32.7% (95% CI: 3.1 - 53.2%).
Proximal DVT was present in 2 patients receiving dermatan sulfate (0.7%; 95% CI: 0 - 1.7%) versus 8 receiving UFH (3.1%; 95% CI: 1.0 - 5.2%) (p=0.0571).
Symptomatic nonfatal PE occurred in 1 patient per group. Both had DVT at venography.
The rate of bleeding complications in all operated patients was 29 of 418 (6.9%) receiving dermatan sulfate and 32 of 424 (7.5%) receiving UFH (95% CI for the absolute risk difference: 4.1 - 2.9%).
Total mortality was similar in both groups (7 of 418 receiving dermatan sulfate and 9 of 424 receiving UFH).
Clinical conclusions In oncologic surgery, dermatan sulphate prevents venous thromboembolism more effectively than heparin without increasing bleeding complications.
Modelling A predictive decision analysis (created with Vanguard Decision Pro 3.0, 1998) was used to estimate the costs and effects associated with each strategy. The target population of the model was a population size of 60,000 patients with a diagnosis of cancer (in Italy, 536,000 patients underwent surgery in public or private hospitals in 1994; of these, 61,000 had a diagnosis of cancer).
Outcomes assessed in the review The probabilities incorporated in the model were as follows: prophylaxis success (absence of postoperative DVT) in patients with cancer, proximal DVT rate (out of total DVT), major bleeding during prophylaxis, progression of distal DVT to proximal DVT, proximal DVT becoming symptomatic, distal DVT becoming symptomatic, PE after asymptomatic proximal DVT, sudden death from undetected PE, PE becoming symptomatic, death after symptomatic DVT, death after symptomatic PE, death after asymptomatic PE, and rate of fatal bleeding during the therapy of thromboembolic events.
Study designs and other criteria for inclusion in the review The inclusion criteria were not reported. It was only reported that the base-case values were the most representative values reported in the literature.
Sources searched to identify primary studies Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Number of primary studies included A total of 11 studies were included in the review.
Methods of combining primary studies Investigation of differences between primary studies Results of the review The outcome values were as follows:
prophylaxis success (absence of postoperative DVT) in patients with cancer (best and worst value), 0.850 (0.802-0.891) for dermatan sulfate and 0.780 (0.723-0.864) for UFH;
proximal DVT rate (out of total DVT) (best and worst value), 0.05 (0.006-0.169) for dermatan sulfate and 0.143 (0.040-0.262) for UFH;
major bleeding during prophylaxis, (best and worst value), 3.1% (1.9-4.3) for dermatan sulfate and 3.1% (1.9-4.3) for UFH;
progression of distal DVT to proximal DVT, 0.200 (range: 0.200 - 0.345);
proximal DVT becoming symptomatic, 0.400 (range: 0.162 - 0.400);
distal DVT becoming symptomatic, 0.050 (range: 0.037 - 0.062);
PE after asymptomatic proximal DVT, 0.500 (range: 0.115 - 0.613);
sudden death from undetected PE, 0.110 (range: 0.082 - 0.137);
PE becoming symptomatic, 0.290 (range: 0.217 - 0.362); death after symptomatic DVT, 0.006 (range: 0.00 - 0.006);
death after symptomatic PE, 0.027 (range: 0.013 - 0.080);
death after asymptomatic PE, 0.300 (range: 0.225 - 0.375); and
rate of fatal bleeding during the therapy of thromboembolic events, 0.39%.
Measure of benefits used in the economic analysis The measure of benefits used was the number of deaths prevented (or lives saved).
Direct costs Costs were not discounted due to the short time frame of the cost analysis (the time horizon of postoperative DVT or PE does not usually exceed 4 months). Some quantities were reported separately from the costs and cost items were reported separately. The cost analysis covered the costs of anticoagulant prophylaxis (drugs, consumables, nursing time spent in drug administration, prophylactic drug monitoring and management of major bleeding complications) and thromboembolic events (hospitalisation, diagnostic procedure, drug acquisition and administration, and management of major bleeding complications). The perspective adopted in the cost analysis was that of the hospital. The cost analysis was a retrospective study using the resource use pattern observed in the DOS randomised trial. The procedures used to diagnose venous thromboembolism were derived from current literature. Values incorporated into costs represented tariffs rather than the actual value of the consumed resources. The price year was 1998.
Indirect Costs Indirect costs were not considered.
Currency Euros (EUR). The costs were calculated in Italian currency and converted into euros using the mean value of September 2000; EUR0.872114 = $US1.00
Sensitivity analysis A series of one-way sensitivity analyses was performed on the parameters incorporated in the model.
Estimated benefits used in the economic analysis The potential mortality from postoperative venous thromboembolism was 0.34% for dermatan sulfate and 0.653% for UFH. With respect to 60,000 patients considered in the model, 204 and 392 potential deaths would have occurred with dermatan sulfate and UFH prophylaxis as a consequence of 9,000 versus 13,200 DVTs, respectively. Therefore the potential lives saved with dermatan sulfate would be 188 per year.
Cost results In the base case (based on a population of 60,000), the cost per patient was EUR154 for dermatan sulfate and EUR185 for UFH, and hence a hospital would save EUR31 per patient by the dermatan sulfate option.
Synthesis of costs and benefits Costs and benefits were not combined since the use of dermatan sulfate was the dominant strategy. In the sensitivity analysis performed, the additional cost per potential life saved was considered as the cost-effectiveness measure. It was reported that final costs and effects were mainly sensitive to variations in the rates of DVT and pulmonary embolism, and to the possible need for 1 extra day of hospitalisation because of the earlier preoperative initiation of dermatan sulfate prophylaxis.
Authors' conclusions Dermatan sulfate is more cost-effective than UFH for the prevention of postoperative venous thromboembolism in patients with cancer. If the hospital stay needs to be prolonged, then the dermatan sulfate option may involve a small additional cost (EUR47) per potential life saved.
CRD COMMENTARY - Selection of comparators A justification was given for the choice of the comparator. It is currently recommended as the prophylactic agent of choice in major general surgery (another agent of choice was reported to be low molecular weight heparins (LMWHs)). You, as a database user, should consider whether this strategy is a widely used health technology in your own setting.
Validity of estimate of measure of effectiveness The internal validity of the effectiveness results is likely to be high due to the randomised nature of the DOS study design, the power calculations performed, and the comparability of the study groups in terms of baseline characteristics (except for the gender composition). Furthermore, It was reported that the DOS trial was adequately sized and controlled, and was one of the very few conducted so far in oncologic surgery. However, the efficacy rates were calculated based on the treatment completers sample rather than on the intention to treat sample. The study sample appears to have been representative of the study population. Regarding the probabilities adopted from the literature, it is worth noting that no systematic literature review (including sources searched, the critical assessment of the primary studies included in the review, the methods used to combine the outcomes of different studies, and differences between studies) was performed to establish a sound review for the probabilities obtained from the literature.
Validity of estimate of measure of benefit The estimate of the benefit measure was modelled, using a decision analytic model, which appears to be appropriate. No quality-of-life considerations were taken into account; it was, however, noted that the expected quality of life after surgery for cancer is likely to be related to the underlying disease, type of surgery performed and subsequent anticancer therapies rather than to the type of perioperative antithrombotic prophylaxis.
Validity of estimate of costs The following features were positive aspects of the cost analysis, likely to enhance its validity: details of methods of cost estimation were given; the price year, conversion rate, and perspective adopted in the cost analysis were reported; sensitivity analysis was performed on some of the cost data. However, the validity of the cost analysis may have been weakened by the following features: the cost data were based on tariffs rather than on true costs; costing was conducted retrospectively; statistical analyses were not performed on resource use and cost data; the effects of alternative treatment strategies on indirect costs were not addressed.
Other issues The authors' conclusions appear to be justified given the uncertainties in the data. The issue of generalisability to other settings or countries was not addressed; it is not clear whether the ranges considered in the sensitivity analyses covered the issue of generalisability. Some comparisons were made with other studies. The degree to which the study sample was representative of the study population was not addressed in the authors' comments. One of the caveats made by the authors was that possible fatal bleeding due to prophylaxis was not considered in the model, since in the DOS study this was an extremely rare event with either treatment. It was deemed that the inclusion of the estimate of 0.39% as the rate of fatal bleeding (as reported in the literature) in the model would have resulted in a small increase in mortality favouring dermatan sulfate (4 additional deaths as compared with 8 with UFH).
Implications of the study The authors mentioned that this was the first economic analysis performed in the specific field of antithrombotic prophylaxis in surgery for cancer, and therefore, more studies may be needed to assess the robustness of the economic results in different settings and countries.
Source of funding Supported by Mediolanum Farmaceutici S.p.a..
Bibliographic details Attanasio E, Russo P, Carunchio G, Caprino L. Dermatan sulfate versus unfractionated heparin for the prevention of venous thromboembolism in patients undergoing surgery for cancer: a cost-effectiveness analysis. PharmacoEconomics 2001; 19(1): 57-68 Other publications of related interest Di Carlo V, Agnelli G, Prandoni P, Coccheri S, Gensini G F, Gianese F, et al. Dermatan sulphate for the prevention of postoperative venous thromboembolism in patients with cancer. Thrombosis and Haemostasis 1999;82:30-34.
Indexing Status Subject indexing assigned by NLM MeSH Adult; Anticoagulants /economics /therapeutic use; Cost-Benefit Analysis; Decision Trees; Dermatan Sulfate /economics /therapeutic use; Health Care Costs; Heparin /economics /therapeutic use; Humans; Multicenter Studies as Topic; Neoplasms /surgery; Postoperative Complications /economics /prevention & Randomized Controlled Trials as Topic; Retrospective Studies; Venous Thrombosis /economics /prevention & control; control AccessionNumber 22001008065 Date bibliographic record published 30/11/2001 Date abstract record published 30/11/2001 |
|
|
|