VALUE IN HEALTH 14 (2011) S89 –S92

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/jval

Short-Term Therapy with Enoxaparin or Unfractionated Heparin for

Venous Thromboembolism in Hospitalized Patients: Utilization Study
and Cost-Minimization Analysis
Catia Argenta, MSc,1 Maria Angélica Pires Ferreira, MSc,2 Guilherme Becker Sander, PhD,2
Leila Beltrami Moreira, PhD1,2,3,*
1
Graduate Program in Medical Sciences, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil; 2Pharmacy and Therapeutics Committee,
Hospital de Clínicas de Porto Alegre, Porto Alegre, Brazil; 3Pharmacology Department, Universidade Federal do Rio Grande do Sul; National Institute of
Science and Technology for Health Technology Assessment–CNPq, Porto Alegre, Rio Grande do Sul, Brazil

A B S T R A C T

Objectives: To evaluate the direct costs of venous thromboembolism
(VTE) treatment with unfractionated heparin (UFH) and low-molecular
weight heparin, from the institutional perspective. Methods: This is a
real-world cohort study that included inpatients treated with UFH or
enoxaparin for deep venous thromboembolism or pulmonary embo-
lism in a tertiary public hospital. To estimate medical costs we com-
puted the acquisition costs of drugs, supplies for administration, labo-
ratory tests, and hospitalization cost according to the patient ward.
Results: One hundred sixty-seven patients aged 18 to 92 years were
studied (50 treated with UFH and 117 with enoxaparin). The median of
days in use of heparin was the same in both groups. Activated partial
thromboplastin time was monitored in 98% of patients using UFH and
56.4% using enoxaparin. Nonstatistically significant differences were
observed between groups in the number of bleeding events (10.0% and
9.4%; P ⫽ 1.00); blood transfusion (2.0% and 2.6%; P ⫽ 1.00); death (8.0%
and 3.4%; P ⫽ 0.24); and recurrent VTE, bleeding, or death (20.0% and
14.5%; P ⫽ 0.38). Daily mean cost per patient was US$12.63 ⫾ $4.01 for
UFH and US$9.87 ⫾ $2.44 for enoxaparin (P ⬍ 0.001). The total costs
considering the mean time of use were US$88.39 and US$69.11.
Conclusion: The treatment of VTE with enoxaparin provided cost sav-
ings in a large teaching hospital located in southern Brazil.
Keywords: heparin, deep venous thrombosis, utilization study, cost
analysis.
Copyright © 2011, International Society for Pharmacoeconomics and
Outcomes Research (ISPOR). Published by Elsevier Inc.

which could result in different antithrombotic effects, but there is

Introduction
Deep venous thrombosis (DVT) in the lower extremities is the
most frequent manifestation of venous thromboembolism (VTE),
with an incidence of 0.48 to 1.6 cases/1000 persons-year among
community residents [1–3]. It is a common condition affecting
mainly inpatients and rates increase with age. The most life-
threatening manifestation is pulmonary embolism (PE), affecting
0.23 to 0.69 cases/1000 person-years [4]. The treatment of VTE rec-
ommended by the American College of Chest Physicians [5] in-
volves short-term low-molecular-weight-heparin (LMWH) or un-
fractionated heparin (UFH) therapy plus long-term oral warfarin
therapy. Anticoagulant therapy with UFH followed by warfarin
prevents thrombus extension, reduces the risk of recurrent
thrombosis, and prevents death in patients with VTE [6,7]. Subcu-
taneous LMWH is as effective and safe as conventional UFH ther-
apy, but does not require laboratory monitoring and is less likely to
cause bleeding, immune thrombocytopenia, and osteoporosis [8 –
10]. LMWH preparartions differ considerably in composition,
no evidence that any LMWH preparation is better or worse than
another in terms of efficacy or safety outcomes [11].
Enoxaparin is among the most widely studied treatments for
VTE [12,13]. Despite LMWH having a greater acquisition cost,
previous pharmacoeconomic analyses have shown that LMWH
is more cost-effective than UFH [14 –16]. It has been calculated
that outpatient treatment with LMWH may save $1641 per pa-
tient in comparison to UFH hospital treatment [17]. This eco-
nomic benefit of outpatient treatment of VTE seems to be pres-
ent in different health systems of developed countries,
although the same can not be extrapolated to developing na-
tions. Economic evaluations in developing countries are desir-
able to estimate VTE hospital treatment cost with UFH and
LMWH, because people with low income do not have access to
outpatient treatment with LMWH. We conducted a cohort study
to evaluate the direct costs of short-term heparin anticoagula-
tion treatment for VTE in a large teaching hospital located in
southern Brazil, from the institutional perspective.

Conflicts of interest: The authors have indicated that they have no conflicts of interest with regard to the content of this
article.
* Address correspondence to: Leila Beltrami Moreira, Farmacologia Clínica sala 947, Hospital de Clínicas de Porto Alegre, Ramiro Barcelos,
2350, 90.035-903, Porto Alegre, RS, Brazil.
E-mail : lbmoreira@hcpa.ufrgs.br.
1098-3015/$36.00 – see front matter Copyright © 2011, International Society for Pharmacoeconomics and Outcomes Research (ISPOR).
Published by Elsevier Inc.
doi:10.1016/j.jval.2011.05.017
S90 VALUE IN HEALTH 14 (2011) S89 –S92
Methods
This was a cohort study that prospectively included patients
hospitalized from March 2005 through January 2007 in a univer-
sity-affiliated, general tertiary teaching hospital with 749 beds
in southern Brazil. These patients had been treated with intra-
venous UFH or subcutaneous LWMH for suspected or confirmed
DVT or PE. The study was approved by the institutional review
board.
Patients receiving UFH or LWMH were identified through the
institution’s computerized prescription system and had their
clinical records revised to be included in the study. All patients
receiving an anticoagulation dose of heparin to treat VTE were
potentially eligible. The LMWH included on the hospital formu-
lary was enoxaparin, based in the lowest price acquisition pol-
icy of the institution. Patients identified with DVT or PE were
prospectively followed-up until the end of the heparin antico-
agulation period. The only exclusion criterion was age younger
than 18 years. Data about diagnosis confirmation, anticoagula-
tion regimen, duration of treatment, laboratory monitoring
with activated partial thromboplastin time (aPTT), bleeding,
thrombocytopenia, blood transfusion, and protamine prescrip-
tion were recorded for all included patients. Adverse events
potentially related to DVT or its treatment were defined as com-
bined endpoint and included in-hospital death, bleeding, or re-
current VTE.
Costs were assessed directly from the hospital’s records on
prices paid for each one of the elements involved in patients’ as-
sistance, considering the prices during their period of hospitaliza-
tion. This information was provided directly by the hospital, so
retrospective costs did not have to be estimated based on any
other indirect information. Total medication costs were calculated
for each patient, considering the prospectively collected data on
medicines, supplies, and laboratory tests. To estimate direct med-
ical costs we computed the acquisition costs of drugs and supplies
associated with UFH treatment, laboratory tests, and hospitaliza-
tion cost according to the patient ward. The cost of use of auto-
matic pump for UFH administration was not computed because
they were not hospital property and its charge was covered by the
cost of pump-specific infusion equipment required. The costs
were converted into US dollars considering the mean exchange
rate from April to May 2007.
Data were analyzed with PASW Statistics version 18.0 (2009,
SPSS Inc., Chicago, IL) and a level of significance of 0.05 was set.
Chi-square statistics were used in the comparison of categorical
variables and Student t or Mann-Whitney tests were applied to
compare continuous variables. Although costs and time of treat-
ment had small skewness deviation, t test and Mann-Whitney
results were similar and t test value was provided to compare
means. Logistic regression modeling was applied to analyze the
association of heparin form and composite clinical outcomes, to
take into account potential confounders identified in crude anal-
yses. Sensitivity analyses were performed during June 2010, con-
sidering the actual cost of drugs to account for acquisition prices
variation. Propensity score for LWMH prescription was computed
to adjust for indication bias.
Results
From the 200 patients included, 33 were excluded because they
had been treated for arterial thrombosis. Twelve patients re-
ceived both heparins and were classified in the group of UFH
(n ⫽ 7) or LMWH (n ⫽ 5) according to the first drug prescribed.
The change from the first prescribed drug to the other was more
frequent in the UFH group (14.0% vs. 4.3% P ⫽ 0.044). The char-
acteristics of the groups are similar when excluding the 12 pa-
tients that received both heparins (Table 1 in Supplemental Ma-
terials found at: doi:10.1016/j.jval.2011.05.017). Patients were 18
to 92 years old; UFH was first prescribed for 50 (29.9%) patients
and LMWH for 117 (70.1%) patients. DVT was treated in 107
cases, PE in 43 cases, and both conditions in 17 cases. Diabetes,
surgery, infectious disease, and renal failure were significantly
more frequent in the group that began anticoagulation with
UFH. Previous PE was more frequent in LMWH group (P ⫽ 0.015).
Only general surgery had crude association with combined end-
point (death, bleeding, or recurrent VTE). Among nine critically
ill patients, six were treated with UFH (12%) and three with
LMWH (2,6%) (chi-square P ⫽ 0.036). The DVT and PE diagnoses
were confirmed in 93.0% and in 51.7% of patients, respectively.
The median of days in use of heparin was the same in both
groups and warfarin was initiated at the first day in 28.1% of 32
patients on the UFH group and in 24.2% of 89 patients on LMWH
group treated with oral anticoagulant. aPTT was monitored in
98% of patients receiving UFH and in 56.4% of patients receiving
LWMH. Half of the UFH group had aPTT measured less than
once a day (Table 2 in Supplemental Materials found at: doi:
10.1016/j.jval.2011.05.017). Nonstatistically significant differ-
ences were observed in the number of bleeding events (10.0% in
UFH and 9.4% in LMWH; chi-square P ⫽ 1.00), blood transfusion
(2.0% and 2.6%; chi-square P ⫽ 1.00), and death (8.0% and 3.4%;
chi-square P ⫽ 0.242). Only one patient in the UFH group evolved
to PE. Combined endpoint consisting of recurrent VTE, bleeding,
or in-hospital death occurred in 20.0% and 14.5% of patients in
the UFH and LMWH groups, respectively (chi-square P ⫽ 0.379).
In a logistic regression model, adjusted for propensity score to
receive UFH and for the hospital ward (clinical or obstetric unit,
surgery unit, or intensive care unit), LMWH group members
showed no significant risk reduction of combined outcome
(odds ratio 0.87; 95% confidence interval 0.32–2.41; P ⫽ 0.79).
We calculated the drug cost per day of treatment taking into
account the actual price for the institution (Table 3 in Supplemen-
tal Materials found at: doi:10.1016/j.jval.2011.05.017). We excluded
patients who were treated initially with one form of heparin and
then changed to other on the follow-up (12 individuals). Daily
mean cost per patient was US$12.63 ⫾ $4.01 for UFH and US$9.87 ⫾
$2.44 for LWMH (t test P ⬍ 0,001). The total cost of short-term
heparin treatment considering the mean length of use was
US$88.39 and US$69.11, respectively, representing a cost saving of
US$19.28 per heparin treatment. The medication itself is the main
component of the LMWH group costs (92.88%), whereas it repre-
sents only 5.25% of costs in the UFH group (Fig. 1 in Supplemental
Materials found at: doi:10.1016/j.jval.2011.05.017). Sensitivity anal-
ysis was performed changing the drug cost according to the actual
price of acquisition in 2010 converted to US dollars (exchange rate
in June 26, 2010). The daily mean cost became US$81.48 ⫾ $55.52
for UFH and US$14.23 ⫾ $8.42 for LWMH (t test P ⬍ 0,001), and the
total unadjusted cost of short-term heparin anticoagulation was
respectively US$757.31 ⫾ $359.64 and US$570.94 ⫾ $201.76 (t test
P ⫽ 0.002), representing a cost saving of US$186.37 per heparin
treatment.
Discussion
This was a pharmacoeconomic analysis of heparin use to treat
VTE conducted in a developing country, through cost analysis
from a public health perspective. This kind of analysis is justified
based on the therapeutic equivalence between initial treatment of
DVT and PE with UFH and LMWH [5]. One characteristic that dis-
tinguishes our study is the evaluation of a real cohort rather then
a hypothetical one, concomitantly including groups of patients
receiving different treatments. This reflects the usual care of these
patients and makes it possible to estimate the actual costs of treat-
ment in a defined setting.
 VALUE IN HEALTH 14 (2011) S89 –S92
Patients included in the study were younger then those of hy-
pothetical cohorts [17,18] and similar to those of other pharmaco-
economic analyses that collected data from individual patients
[14,19]. Sex is not a risk factor for VTE and is in accordance with the
nonsignificant predominance of women in the sample [20]. The
18.6% prevalence of confirmed PE in the studied sample is similar
to the 19% in a clinical trial of VTE [12]. In the same trial the major
or minor bleeding rates were lower then those recorded in our
cohort but they were not different between the treatment groups
either.
There was no difference on clinical outcomes between the two
groups. This is in accordance with others studies [12,13], but clin-
ical trials are not all consistent about equivalence between enoxa-
parin and UFH. A meta-analysis [21] demonstrated a statistically
significant reduction in clinical outcomes with LMWH when com-
bining all trials, but benefit persisted only for reduction of the
thrombus size when the 11 studies with adequate concealment of
allocation before randomization were considered. Comparison
with literature is difficult because the conditions affecting en-
rolled patients in clinical trials are quite different and the small
sample size is a limitation that precludes a definitive conclusion of
our study.
We did not expect that LMWH prescription was greater than
UFH prescription because the institutional policy during the
study period reinforced the use of UFH for VTE treatment based
on equal efficacy and lower acquisition costs. The preferred use
of UFH in renal failure is recommended in view of the plasmatic
accumulation due to delayed elimination and risk of excessive
anticoagulant effect [22]. In surgical patients, the preference
could be related to the potential for reversion of anticoagulant
effect with protamine and the shorter half-life then LMWH in
case of bleeding. On the other hand, the preference for using
LMWH in face of previous PE is not evidence-based. The cross-
over from one drug to another was not expected either, but we
did not investigate the reasons for it. The greater proportion of
infected, critically ill, and diabetic patients in the UFH group
suggests they were at higher risk and may be responsible for the
statistically nonsignificant increase in death rate in this group.
The propensity score was constructed to adjust for these risk
factors and was included in the logistic regression model that
showed no independent association of heparin group with the
composite endpoint. But the nonsignificant relative risk reduc-
tion of 17% favoring LMWH may be explained by the small sam-
ple size and lack of power.
The total costs of initial treatment for VTE were less expensive
with enoxaparin than with UFH, providing an economic advantage
of 21.0%. It provided cost savings regardless of the abuse in order-
ing monitoring laboratory tests when LMWH was prescribed. If we
had computed the costs of automatic infusion pump use this eco-
nomic advantage would be even greater. The economic advan-
tages are in agreement with those found in studies that compared
the costs of UFH and LMWH for VTE treatment [14,15,16,18,23,24]
despite the differences in cost components like home care [14,23],
outpatient treatment [14], and the health care providers’ perspec-
tive [18]. It must be emphasized that the institution is a tertiary
university-affiliated hospital that acquires great quantities of the
drug. Therefore, it is a favored buyer that can purchase LMWH at
prices significantly lower than the average wholesale price. The
costs saving associated to LMWH may be explained mainly by the
supplies associated with UFH treatment. Besides of the same du-
ration of UFH and LMWH use, computing the costs of hospitaliza-
tion must implicate in greater economic differences because UFH
seems to be mainly indicated for more severely ill patients. It must
be noted that total cost, including hospitalization cost was not
adjusted for the clinical disadvantages of patients with UFH. The
sensitivity analysis was performed computing actual drugs’ prices
in June 2010. The economic LMWH advantage became even
S91
greater as a result. Because of evidence of a drug contaminant
associated with greater incidence of adverse reactions UFH had its
commercialization stopped [25]. When commercialization re-
started, there was a considerable price increase for UFH, inflating
costs still further.
Conclusions
Treatment of VTE with enoxaparin provided costs savings in a
large teaching hospital located in southern Brazil, from the insti-
tutional perspective. Despite differences in the settings where
studies of economic evaluation were conducted, the findings
agree in regard to the economic advantages of use of LMWH. Im-
plementation of a critical pathway for anticoagulation is desirable
to promote rational use of heparins and to save costs.
Source of financial support: This study was funded by Fundo de
Incentivo à Pesquisa/Hospital de Clínicas de Porto Alegre.
Supplemental Materials
Supplemental material accompanying this article can be found in
the online version as a hyperlink at doi:10.1016/j.jval.2011.05.017,
or if hard copy of article, at www.valueinhealthjournal.com/issues
(select volume, issue, and article).
REFERENCES
[1] Anderson FA Jr, Wheeler HB, Goldberg RJ, et al. A population-based
perspective of the incidence and case-fatality rates of deep vein
thrombosis and pulmonary embolism. The Worcester DVT Study.
Arch Intern Med 1991;151:933– 8.
[2] Nordström M, Lindblad B, Bergqvist D, et al. A prospective study of the
incidence of deep-vein thrombosis within a defined urban population.
J Intern Med 1992;232:155– 60.
[3] Naess IA, Christiansen SC, Romundstad P, et al. Incidence and
mortality of venous thrombosis: a population-based study. Thromb
Haemost 2007;5:692–9.
[4] Silvestein MD, Heit JA, Mohr DN, et al. Trends in the incidence of deep
vein thrombosis and pulmonary embolism: a 25-year population-
based study. Arch Intern Med 1998;158:585–93.
[5] Hirsh J, Raschke R. Heparin and low-molecular-weight heparin: the
seventh ACCP conference on antithrombotic and thrombolytic
therapy. Chest 2004;126:188s–203s.
[6] Brandjes DBM, Heijboer H, Buller HR, et al. Acenocoumarol and
heparin compared with acenocoumarol alone in the initial treatment
of proximal-vein thrombosis. N Engl J Med 1992;327:1485–59.
[7] McRae S, Eikelboom JM. Latest medical treatment strategies for
venous thromboembolism. Expert opinion on pharmacotherapy
2007;8:1221–33.
[8] Majerus PW, Tollefsen DM. Bood coagulation and anticoagulant,
thrombolitic, and antiplatelet drugs. In: Brunton LL, Lazo JS, Parker KL,
(ed.). The Pharmacological Basis of Therapeutics. (11th ed.). New York:
Brunton LL, Lazo JS, Parker KL, 2005.
[9] Snow V, Qassem A, Barry P, et al. Management of venous
thromboembolism: a clinical practice guideline from the american
college of physicians and the american academy of famy physicians.
Ann Inten Med 2007;146:204 –10.
[10] Quinlan DJ, McQuillan A, Eikelboom JW. Low-molecular-weight
heparin compared with intravenous unfractioned heparin for
treatment of pulmonary embolism: a meta-analysis of randomized,
controlled trials. Ann Intern Med 2004;140:175– 83.
[11] Kearson C, Ginsberg JS, Julian JA, et al. Comparison of fixed-dose
weight-ajusted unfractionated heparin and low-molecular-weight
heparin for acute treatment of venous thromboembolism. JAMA 2006;
296:23–30.
[12] Merli G, Spiro TE, Olsson CG, et al. Subcutaneous enoxaparin once or
twice daily compared with intravenous unfractionated heparin for
treatment of venous thromboembolic disease. Ann Intern Med 2001;
134:191–202.
[13] Mismetti P, Quenet S, Levine M, et al. Enoxaparin in the treatment of
deep vein thrombosis with or without pulmonary embolism. Chest
2005;128:2203–10.
S92 VALUE IN HEALTH 14 (2011) S89 –S92
[14] Boucher M, Rodger M, Johnson JA, et al. Shifting from inpatient to
outpatient treatment of deep vein thrombosis in a tertiary care center.
A cost-minimization analysis. Pharmacoterapy 2003; 23:301–9.
[15] Spyropoulos AC, Hurley JS, Ciesla GN, et al. Management of acute
proximal deep vein thrombosis: phamacoeconomic evaluation of
outpatient treatment with enoxaparin vs inpatient treatment with
unfractioned heparin. Chest 2002;122:108 –14.
[16] Gould MK, Dembitzer AD, Sanders GD, et al. Low-molecular-weight
heparins compared with unfractioned heparin for treatment of acute
deep venous thrombosis: a cost-effectiveness analysis. Ann Intern
Med 1999;130:789 –99.
[17] Holzheimer RG. Low-molecular-weight heparin (LMWH) in the
treatment of thrombosis. Eur J Med Res 2004;9:225–39.
[18] Aujesky D, Smith K, Cornuz J, et al. Cost-effectiveness of low-
molecular-weight heparin for treatment of pulmonary embolism.
Chest 2005;128:1601–10.
[19] Huse DM, Cummins G, Taylor DC, et al. Outpatient treatment of
venous thromboembolism with low-molecular-weight heparin: an
economic evaluation. Am J Manag Care 2002;(89 Suppl.):S10 – 6.
[20] Hureta CJS, Wallander MA, Garcia RLA. Risk factors an short-term
mortality of venous thromboembolism diagnosed in the primary care
setting in the United Kingdom. Arch Intern Med 2007;167:935– 43.
[21] Dongen CJJ, Belt AGM, Prins MH, et al. Fixed dose subcutaneous low
molecular weight heparins versus adjusted dose unfractionated
heparin for venous thromboembolism. Cochrane Database Syst Rev
2004;18:CD001100.
[22] Lim W, Dental F, Eikelboom JW, et al. meta-analysis: low-molecular-
weight heparin and bleeding in patients with severe renal
insufficiency. Ann Intern Med 2006;144:673– 84.
[23] Spyropoulos AC, Frost FJ, Hurley JS, et al. Costs and clinical outcomes
associated uit low-molecular-weight heparin vs unfractioned heparin
for perioperative bridging in patient receiving long-term oral
anticoagulat therapy. Chest 2004;125:1642–50.
[24] Knight KK, Wong J, Hauch O, et al. Economic and utilization outcomes
associated with choice of treatment for venous thromboembolism in
hospitalized patients. Value Health 2005;8:191–200.
[25] Kishimoto TK, Viswanathan K, Ganguly T, et al. Contaminated
heparin associated with adverse clinical events and activation of the
contact system. N Engl J Med 2008;358:2457– 67.