A. Santamaría; S. Juárez; A. Reche; A. Gómez-Outes; J. Martínez-González; J. Fontcuberta 

Int J Clin Pract.  2006;60(5):518-525.  ?2006 Blackwell Publishing

Summary and Introduction

Summary

The objective of this study is to assess the clinical and economic outcomes associated with outpatient treatment and secondary prophylaxis of acute venous thromboembolism (VTE) with a low-molecular-weight heparin, bemiparin.

This study was designed as an open-label, multicentre, prospective, cohort study in standard clinical practice.

Sixty-three investigators from 54 Spanish centres participated in the study.

Five hundred eighty-three patients (434 outpatients and 149 inpatients) with acute VTE were followed up for 98 days (median).

Outcome measures were costs and adverse events during initial VTE treatment with bemiparin (outpatient vs. inpatient cohorts) and long-term treatment [bemiparin (BEM) vs. vitamin K antagonists (VKA) cohorts].

Mean total costs per patient were lower in the outpatient cohort as compared with those in the inpatient cohort (1206 vs. ?5191; difference = -?3985; p < 0.001), with similar rates of adverse events (5.1 outpatient vs. 7.4% inpatient; p = 0.196) over 98 days. Mean total costs per patient were similar in the BEM/BEM and BEM/VKA cohorts (3616 vs. ?3831; difference = -?215; p = 0.412), but patients on long-term bemiparin treatment had lower rates of major bleeding (0.4 vs. 1.7%; p = 0.047), minor bleeding (1.8 vs. 6%; p = 0.032) and total adverse events (2.9 vs. 9.5%; p = 0.007) than patients in the BEM/VKA cohort.

Outpatient management of VTE with bemiparin in selected patients resulted in significant cost-savings compared to inpatient treatment, while maintaining effectiveness and safety. Bemiparin may be a safer and cost-neutral alternative to VKA for long-term treatment of VTE.

Introduction

Venous thromboembolism (VTE), a disease that encompasses both deep vein thrombosis (DVT) and pulmonary embolism (PE),^[1] is a common cardiovascular condition resulting in substantial morbidity and mortality. Treatment approaches for DVT and PE are similar.^[2] Patients admitted to hospital with VTE may receive initial treatment with either subcutaneous (SC) low-molecular-weight heparin (LMWH) or intravenous unfractionated heparin (UFH).^[1,2] The main advantages of LMWHs include a better bioavailability and a more predictable anticoagulant activity as compared to UFH, a convenient SC administration and no need for monitoring of their anticoagulant effect on most patients.^[3] This allows for treatment of patients at their homes.^[4-6]

According to current recommendations, initial treatment with UFH or LMWH should be followed by oral anticoagulant therapy with a vitamin K antagonist (VKA; i.e. warfarin, acenocoumarol) for at least 3 months.^[1,2] However, when oral anticoagulation is either contraindicated or inappropriate, long-term treatment with LMWH should be used.^[1,2] While routine laboratory monitoring of antithrombotic activity is not required in patients receiving long-term treatment with LMWHs, VKA treatment requires frequent monitoring of prothrombin time [PT; International Normalised Ratio (INR)].^[1]

Bemiparin sodium (Hibor^?, Zibor^?, Ivor^? and Badyket^?) is a new, second-generation LMWH^[7] that has been shown in clinical trials to be more effective than UFH for reducing thrombus size^[8] and to be cost-effective for acute inpatient treatment of DVT.^[9] Long-term bemiparin was as effective and safe as, and a cost-neutral alternative to VKA for long-term treatment (secondary prophylaxis after initial treatment) of recurrent VTE for 3 months in clinical trials.^[8,9]

There are limited data documenting common strategies used for initial outpatient or inpatient treatment of VTE and for long-term treatment, or describing the clinical and economic outcomes with each of these regimens. A prospective study was performed to evaluate the clinical and economic outcomes associated with acute outpatient vs. inpatient treatment and with long-term treatment of VTE with bemiparin in standard clinical practice.

Methods

This study was designed as an observational, open-label, multicentre, prospective, cohort study to assess the effectiveness, safety and pharmacoeconomics of bemiparin in the outpatient vs. inpatient management of VTE. Patients were admitted to hospital if they had a concomitant condition requiring hospitalisation, severe VTE symptoms, high bleeding risk^[6,10-12] or any other reason for inpatient treatment, according to the physician's criterion. The decision on the type of long-term treatment to be used was left to the investigator, according to standard clinical practice. Sixty-three investigators from 54 Spanish centres participated in the study.

Study Population

Patients of both sexes aged 18 years or older, receiving treatment with bemiparin for an acute VTE event (DVT with or without concomitant clinically stable PE), documented by objective methods, were recruited into the study. Exclusion criteria included (i) hypersensitivity to heparin or other pig-derived substances, (ii) history of documented or suspected immune-mediated heparin-induced thrombocytopaenia (HIT), (iii) active haemorrhage or increased risk of bleeding due to impaired haemostasis or organ lesion (e.g. active peptic ulcer, haemorrhagic stroke, cerebral aneurysm or cerebral neoplasm), (iv) severe impairment of liver and pancreas function, (v) injuries to and surgery on the central nervous system, eyes and ears in the past 2 months, (vi) disseminated intravascular coagulation attributable to HIT, (vii) acute or subacute endocarditis, (viii) impossibility for the patient to be adequately followed-up, (ix) pregnancy, (x) end-stage disease or a life expectancy of less than 3 months and (xi) participation in another study in the past month.

Treatments

All patients received bemiparin 115 IU/kg SC once daily (Hibor^?, Laboratorios Rovi, Madrid, Spain) based on the patient's weight (<50 kg: 5000 IU/day; 50-70 kg: 7500 IU/day; >70- 100 kg: 10,000 IU/day; >100 kg: 115 IU/kg/day) for 7-10 days. This was followed by oral VKA, in a dose adjusted to achieve an INR ranging from 2 to 3, or SC bemiparin once daily, based on the investigators' criterion, for 3 months. Patients scheduled for outpatient treatment and/or secondary prophylaxis (long-term treatment) with bemiparin were trained in subcutaneous self-injection techniques to ensure the safe administration of bemiparin.

Adverse Events

Adverse events included symptomatic recurrent DVT or PE documented by objective methods (see Related Procedures), major and minor bleeding complications, thrombocytopaenia and death from any cause. Proportions of patients in each of the cohorts experiencing adverse events were calculated. Major bleeding was defined as clinically overt bleeding associated with a fall in haemoglobin of at least 2 g/dl or requirement of a transfusion of 2 or more blood units, bleeding from a critical site (retroperitoneal, intracranial or spinal bleeding), fatal bleeding or any bleeding requiring treatment discontinuation. Minor bleeding was defined as any other overt bleeding not meeting major bleeding criteria. All patients were monitored for adverse events on day 7 ? 2 and 3 months afterwards.

Related Procedures

Procedures and diagnostics identified as related to VTE included those required for objective documentation of VTE: Doppler ultrasound and venogram for DVT, and ventilation/perfusion (V/P) lung scan, helical computed tomographic (CT) scan or pulmonary angiogram for PE. In addition, PT/INR, aPTT or anti-Xa were measured as related to anticoagulant monitoring. The frequencies of these procedures and diagnostics for patients in each of the cohorts were determined.

Pharmacoeconomic Analysis

In the primary pharmacoeconomic analysis, total costs in the outpatient and inpatient cohorts treated with bemiparin were compared. In the secondary pharmacoeconomic analysis, total costs were compared in the secondary prophylaxis cohorts [long-term bemiparin (BEM/BEM) and BEM/VKA] for 3 months. As the study perspective was that of the Spanish National Health System, only direct medical costs were included. Indirect medical or nonmedical costs (worker's lost wages, home care by family members, travel expenses, telephone calls to physician's office, etc.) were not considered in this analysis. The time horizon was 3 months. The costs provided in the data set were calculated by multiplying the units of resources used during the study (i.e. hospitalisation days, number and type of diagnostic procedures and laboratory tests performed, number of doses administered, number and type of clinical complications, number of outpatient medical visits, etc.) by unit costs, as shown in Table 1 . Estimates of these unit costs were derived using the costs to charges function from the Spanish Medicare Reimbursement Rate and the actual costs of medications.^[13-17] Mean costs were calculated for each treatment group, and the results of selected groups were compared.

Median lengths of stay (LOS) and rates of early discharges within 3 days were calculated for total inpatients and for each treatment cohort (BEM/BEM and BEM/VKA).

In our sensitivity analyses, total costs in each cohort and costs differences were recalculated based on a percent change in unit costs ( Table 1 ) of ?50%.

Patient and Physician Satisfaction

Patient and physician satisfaction with outpatient treatment was assessed using a questionnaire based on a 5-point scale: excellent, very good, good, fair or poor.

Statistical Analyses

Differences in the means of continuous variables were evaluated by using an anova test, and differences in proportions were evaluated using χ²-test and Fisher exact test. For cost data, both the t-test, assuming an underlying normal distribution of data, and a nonparametric Wilcoxon rank sum test, which has no underlying distribution assumptions, were used. The reported p-values are for a two-tailed test statistic. No p-value is reported for which sample sizes were not large enough for a meaningful measure of difference. Statistical analyses were performed by using the statistical software package spss for Windows (SPSS version 13.0, Chicago, IL, USA).

Results

Study Populations and Treatments

Of the 601 patients from 54 Spanish centres screened for enrolment between October 2002 and November 2004, 18 patients were excluded from the analysis, because they did not have at least one outcome measurement. Thus, 583 patients were analysed. Of these, 434 patients (74.4%) were treated on an outpatient basis and 149 patients (25.6%) were admitted to hospital ( Table 2 ). Reasons for admission in the 149 inpatients included concomitant disease requiring hospitalisation (27%), high risk of recurrent VTE (20.7%), symptom severity (19.4%), social patient condition making outpatient management difficult (16%), instruction in treatment and control of anticoagulant therapy (4.2%), high bleeding risk (2.5%), need for additional procedures (1.7%), postoperative period (1.7%), risk of poor treatment compliance (1.3%) and others (5.5%). Fifty-nine patients had 2 or more reasons for admission.

Mean age and sex were similar in the outpatient and inpatient cohorts, and also in the BEM/BEM and BEM/VKA cohorts ( Table 2 ). Acute DVT was located in the lower limbs (572 patients), upper limbs (seven patients), jugular veins (three patients) and pelvic veins (one patient). DVT was associated with PE in 19 patients (3.3%) ( Table 2 ). Tests used to document baseline DVT included Doppler ultrasound (89.6%), venography (8.2%) and combined Doppler ultrasound plus venography (2.2%). Confirmatory tests in patients with baseline PE included V/P lung scan (47.4%), helical CT (42.1%) and pulmonary angiogram (10.5%).

At baseline, patients in the inpatient group more frequently had proximal DVT, concomitant PE and a history of cancer, congestive heart failure or cerebral-vascular disease ( Table 2 ). Five of 19 patients with concomitant clinically stable PE at baseline and 45 of 76 patients with cancer were treated on an outpatient basis.

Bemiparin dosage most commonly used for initial treatment was 7500 IU/day (51.6% of patients) for 9 days (median). The median bemiparin daily dose used for long-term treatment was 5000 IU (1st and 3rd quartile, 5000 and 7500 IU) for 89 days (median). VKA was started on the second day (median) of bemiparin initial treatment and continued for 97 days (median) until day 98 (median observation period). Acenocoumarol was given to 115 of the 116 patients administered VKA, while warfarin was used in the remaining patient. At the final visit, 61.1% of INRs were within 2 and 3.

Adverse Events

Adverse event rates in each cohort are shown in Table 3 . Recurrent VTE was uncommon, only occurring in three episodes during the whole observation period (two in the outpatient cohort and one in the inpatient cohort). Adverse event rates were similar in the outpatient and inpatient cohorts.

Long-term bemiparin treatment (BEM/BEM cohort) was associated with lower rates of adverse events of major and minor bleeding as compared to the BEM/VKA cohort during the 98-day observation period ( Table 3 ). If the long-term treatment period (from day 10 to day 98) is considered only, patients on bemiparin also had lower rates of major bleeding (0.4 vs. 1.7%; p = 0.047), total bleeding (1.4 vs. 5.2%; p = 0.041) and total adverse events (2.2 vs. 6%; p = 0.043) as compared with patients treated with VKA.

There were eight deaths during the study. Of these, one in the inpatient cohort and one in the outpatient cohort were related to fatal PE and major bleeding, respectively. The remaining six deaths were not VTE- or drug-related. There were four cases of mild (range from 100,000 to 150,000/mm³) and transient thrombocytopaenia that did not require treatment discontinuation: two cases occurred during bemiparin treatment and two during VKA treatment. There were no cases of severe HIT.

Pharmacoeconomic Analysis

Outpatient management was associated with lower total costs as compared to inpatient treatment ( Table 4 ), resulting in net cost-savings of ?3985 per treated patient (p < 0.001). Treatment with BEM/BEM was associated with higher medication costs than BEM/VKA that were totally offset by lower hospitalisation costs, lower monitoring costs and less medical visits during follow-up in the BEM/BEM cohort ( Table 4 ).

Median LOS for total inpatients, BEM/BEM and BEM/VKA inpatients are shown in Figure 1. Early discharge within the first 3 days of hospitalisation was significantly more common in the BEM/BEM cohort than in the BEM/VKA cohort (43.3 vs. 22.6%; p = 0.041).

Figure 1. 

Distribution of length of stay for VTE admissions by treatment groups. VTE, venous thromboembolism; LOS, length of stay; BEM, bemiparin; VKA, vitamin K antagonist

     

Sensitivity Analyses

In the univariate sensitivity analyses, outpatient treatment remained significantly cost-saving as compared to inpatient treatment assuming a percent change in unit costs of ?50% ( Table 5 ). Long-term treatment with bemiparin remained cost-neutral when drug purchase costs were increased by 50%.

Patient and Physician Satisfaction

Outpatient treatment was reported to be 'excellent', 'very good' or 'good' by 95.4% of patients and 98.8% of physicians. Only 4.6% of patients and 1.2% of doctors rated outpatient treatment as 'fair' or 'poor'.

Discussion

Our study was performed to investigate the costs and complications associated with initial outpatient vs. inpatient management of VTE, and also with long-term treatment either with bemiparin or with VKA. Outpatient treatment resulted in significantly lower costs and similar rates of clinical complications than inpatient treatment during a 98-day observation period after an acute VTE episode. Sensitivity analyses supported the robustness of the results, with bemiparin outpatient treatment resulting in cost-savings in all the ranges tested for unit costs.

In our study, approximately three of every four patients could be treated on an outpatient basis. Several exclusion criteria have been used during outpatient DVT treatment programmes,^6,10,11] including concurrent diseases requiring hospital admission, active bleeding or gastrointestinal bleeding within 6 months, iliofemoral thrombosis or concomitant PE, severe hepatic or renal insufficiency, pregnancy or lactation, age <18 years, patients unable for self-treatment at home or likelihood of poor compliance. However, a less restrictive approach could increase the proportion of eligible patients while maintaining efficacy and safety.^[12] Preliminary data suggest that LMWHs may be used for out-of-hospital treatment in patients with nonmassive PE,^[18,19] and in cancer patients,^[20] but further research is needed to assess the efficacy and safety of this treatment approach. Recently, it has been proposed as a simple prognostic model that could help identify patients with PE who are at low risk of adverse medical outcomes and are candidates for outpatient treatment.^[21] Nevertheless, before this prediction rule can be considered for use in clinical practice, it should be validated in a prospective study.

One of the main disadvantages of conventional treatment of VTE with UFH is the need for hospitalisation and its impact on patient well-being and social life. Outpatient treatment of VTE with LMWH may improve patient's quality-of-life and satisfaction.^[22] In our study, over 95% of patients and physicians were satisfied with the outpatient treatment of VTE with bemiparin.

In agreement with a previous pharmacoeconomic analysis,^[9] long-term treatment with bemiparin resulted, in our study, in similar rates of clinical complications and costs compared to treatment with bemiparin plus VKA. While pharmacy costs of long-term bemiparin treatment are higher, routine laboratory monitoring of its anticoagulant effect is not required in most patients. Moreover, inpatients treated with BEM/BEM in our study had lower hospital costs than those treated with BEM/VKA, due to a significantly higher rate of early discharge within the first 3 days. Use of long-term LMWH may allow for earlier discharge in inpatients with acute VTE and outpatient treatment, which is consistent with evolving practices with regard to anticoagulant management in order to reduce hospital utilisation.

Median bemiparin daily dose used during long-term treatment was 5000 IU, instead of the bemiparin 3500 IU daily dose used for 3 months after the initial treatment during clinical trials.^[8,9] Nevertheless, patients given bemiparin for long-term treatment had lower bleeding and total adverse event rates than those treated with VKA. A recent meta-analysis showed that LMWHs appear at least as effective and probably safer than VKA for long-term treatment of VTE.^[23] Clinical trials evaluating different LMWHs for long-term treatment have used different LMWH doses, ranging from a high-risk prophylaxis dose after the initial treatment to full therapeutic doses for 3 or 6 months, according to various protocols.^[23-25] Therefore, head-to-head comparative trials are required in long-term VTE treatment.^[24]

This analysis has several limitations. The main limitation is the potentialdifferences between patients treated with different regimens that may confound comparisons, because the decision to treat patients on an outpatient or inpatient basis or the type of long-term treatment to be used was left to the investigator, according to standard clinical practice. For example, patients treated on an outpatient basis had a more favourable clinical prognosis as regards less frequency of proximal DVT or concomitant PE and less frequency of concomitant diseases at baseline. In addition, a higher frequency of proximal DVT at baseline and male sex was found in patients treated with VKA as compared to those receiving long-term bemiparin, which might be related to a higher risk of VTE recurrence. However, there were no VTE recurrences during long-term treatment in the bemiparin or VKA cohorts. Randomised clinical trials would be needed to address this issue fully, and such confounding potential should be taken into account when interpreting these findings.

More widespread use of LMWHs could allow for shorter hospital stays and increased outpatient management for initial treatment of VTE, resulting in significant cost-savings as compared to inpatient VTE treatment, while maintaining efficacy and safety. The LMWH, bemiparin, may be a safer and cost-neutral alternative to VKA for long-term treatment of VTE.

CLICK HERE for subscription information about this journal.