506404

research-article2013
AOPXXX10.1177/1060028013506404Annals of PharmacotherapyHagopian et al

Research Report
Annals of Pharmacotherapy

Assessment of Bleeding Events Associated

47(12) 1641­–1648
© The Author(s) 2013
Reprints and permissions:
With Short-Duration Therapeutic sagepub.com/journalsPermissions.nav
DOI: 10.1177/1060028013506404

Enoxaparin Use in the Morbidly Obese aop.sagepub.com

Jennifer C. Hagopian, PharmD1, Jennifer N. Riney, PharmD1,

James M. Hollands, PharmD1, and Eli N. Deal, PharmD1

Abstract
Background: Enoxaparin dosing for patients with morbid obesity is uncertain, and therefore, an elevated incidence of
bleeding may exist in this group. Objective: To determine if increased bleeding events occur in patients with morbid
obesity (body mass index ≥ 40 kg/m2) compared with lower-weight patients with treatment doses of enoxaparin. Methods:
Patients at a single, academic medical center receiving enoxaparin for at least 24 hours from July to December 2009 were
retrospectively evaluated. Patients with morbid obesity were randomly selected among the total cohort and were matched
with lower-weight controls (1:2 ratio) based on the presence of renal dysfunction (serum creatinine >1.4 mg/dL). Bleeding
events, defined on the basis of laboratory changes, administration of blood products, or clinical data, occurring up to 24
hours after enoxaparin administration were evaluated. Independent risk factors for bleeding were assessed via multivariate
analysis. Results: The maximum single dose administered throughout the study was 150 mg, and the largest patient
enrolled weighed 175 kg. Final enoxaparin doses in morbidly obese (0.98 mg/kg) patients were lower compared with that
in controls (1.04 mg/kg, P < .01). The proportion of patients with bleeding events was 29% in the morbidly obese and 23.5%
in the control group (P = .30). Enoxaparin doses <0.9 mg/kg (adjusted odds ratio [AOR] = 2.35; 95% CI = 1.01-5.47; P =
.04), durations of therapy beyond 48 hours (AOR = 2.42; 95% CI = 1.35-4.33; P < .01), and female gender (AOR = 2.05;
95% CI = 1.12-3.73; P = .02) were associated with bleeding, whereas warfarin use (AOR = 0.46; 95% CI = 0.26-0.81; P <
.01) was associated with fewer bleeding events. Conclusions: Results suggest that dosing enoxaparin in morbidly obese
patients (up to 175 kg in weight) with doses capped at 150 mg was not associated with increased bleeding incidence.

Keywords
anticoagulation, enoxaparin, low-molecular-weight heparin, morbid obesity, obesity

Background In the recent updates to the CHEST guidelines, specific

The United States is experiencing an obesity epidemic. In
2011, 39 states had a prevalence rate of obesity equal to or
greater than 25%.1 The number of adults categorized as
morbidly obese (body mass index [BMI] >40 kg/m2)
increased by 70% between 2000 and 2010.2 It has been esti-
mated2 that in 2010, 6.6% of the US population (15.5 mil-
lion adults) had a BMI >40 kg/m2. The controversy
regarding dosing medications in this vulnerable patient
population remains. Enoxaparin, a low-molecular-weight
heparin (LMWH), is a medication that is normally dosed
based on actual body weight (ABW). However, data are
lacking on the safety and efficacy of dosing this medication
in morbidly obese patients. Using ABW to dose enoxaparin
may lead to over-anticoagulation and increased bleeding
risk because LMWH distributes into the intravascular com-
partment and not into tissues and body fat.3
dosing recommendations for patients with morbid obesity
are not addressed. The guidelines4 state that there are data to
dose enoxaparin using total body weight in patients weigh-
ing up to 144 kg and in patients with a BMI of 30 kg/m2.
Because specific recommendations regarding dosing in
obese patients are lacking, dosing among different practice
sites is likely to vary widely. A review article with the aim
of providing practical recommendations for the use of
LMWH in obese patients recommends dosing of enoxapa-
rin using total body weight and does not recommend the use
1
Barnes-Jewish Hospital, St Louis, MO, USA
Corresponding Author:
Jennifer C. Hagopian, Department of Pharmacy, Barnes-Jewish Hospital,
216 S Kingshighway Boulevard, St Louis, MO 63110, USA.
Email: jch2298@bjc.org

Downloaded from aop.sagepub.com at University of New England on June 10, 2015

1642 Annals of Pharmacotherapy 47(12)
of dose capping for venous thromboembolism.3 At the time
of this study, our institution did not have any formal proto-
cols for enoxaparin dosing in morbidly obese patients.
Because of this lack of standardization, our historical prac-
tices of dosing enoxaparin in the patient population has var-
ied widely with respect to dosing weight, dose capping, and
variations in prescribing habits among practitioners. The
available pharmacokinetic and clinical data with enoxapa-
rin use in patients with morbid obesity leaves us without
concrete data to make any firm conclusions on the safety
and efficacy of using total body weight to dose enoxaparin
in these patients.5-8 Although the studies seem to suggest
that using total body weight is safe and effective, clinical
experience is limited in using enoxaparin in patients weigh-
ing greater than 150 kg. The following research is needed to
help fill in the gaps in the literature by investigating bleed-
ing incidence with the therapeutic use of enoxaparin among
patients with a BMI ≥40 kg/m2.
Methods
Patients were considered for inclusion into the study if they
were hospitalized at Barnes-Jewish Hospital (St Louis, MO),
a 1200-bed academic medical center, between July 1 and
December 31, 2009. These dates were chosen because of the
availability of 1 consistent electronic database utilized during
the study period for the electronic medical record. To be
included, the patients had to receive treatment doses of
enoxaparin for at least 24 hours. Treatment doses of enoxapa-
rin were defined as a single dose greater than or equal to 0.85
mg/kg. Patients receiving all levels of care were included in
the study, including medical, surgical, and intensive care.
Patients who met these criteria were subcategorized into 4
weight classes based on BMI9: <18.5 kg/m2, 18.5-29.9 kg/
m2, 30-39.9 kg/m2, and ≥40 kg/m2. Patients classified in the
morbidly obese group (≥40 kg/m2) comprised the study
group; all others served as controls. Patients who did not
have a height, weight, or serum creatinine (SCr) value
recorded during the hospitalization were excluded. A total of
100 patients were selected using a randomization function in
Microsoft Excel from the entire cohort and matched with the
control group in a 1:2 fashion based on the presence of renal
dysfunction (defined as a SCr value ≥1.4 mg/dL) at any time
point during admission. In all, 100 patients were selected
based on the time frame available for data collection as well
as the ability to use this number with matched individuals in
smaller weight cohorts in a meaningful multivariate analysis
investigating the effect of weight on bleeding events. The
study was approved by the Washington University in Saint
Louis Human Studies Committee, and the requirement for
informed consent was waived.
Data were collected using electronic databases available
at our facility. Data collected for each patient included
demographics, indication for anticoagulation, comorbid
Downloaded from aop.sagepub.com at University of New England on June 10, 2015
medical conditions, medications or medical conditions that
could increase bleeding risk, level of care, dose (initial and
final) and duration of enoxaparin use, and laboratory param-
eters, including SCr, hemoglobin, hematocrit, and platelet
level. Patient weight (ABW) was recorded at the time of
enoxaparin initiation. Initial dose, final dose, and any dose
adjustments (for renal function or monitoring parameters)
were recorded. All doses were appropriately adjusted for
each patient’s renal function. Final enoxaparin dose was
defined as dose at discontinuation of medication or dis-
charge dose. If collected, antifactor Xa level and adminis-
tration of blood products during admission were
documented. Antifactor Xa levels were considered to be
steady-state peaks if drawn between 3 and 5 hours after at
least 3 administered doses of enoxaparin. Goal antifactor
Xa levels were 0.5 to 1.0 IU/mL for twice-daily dosing and
1.0 to 2.0 IU/mL for once-daily dosing.3 The development
of new thrombus at 30 days after discontinuation of therapy
was assessed for each patient as defined by diagnostic or
radiographic evidence or an admission for a new thrombus
(venous thromboembolism, pulmonary embolism, cerebro-
vascular accident, or myocardial infarction) at any institu-
tion within our hospital system.
The primary outcome measure was the development of
bleeding events while therapeutically anticoagulated with
enoxaparin as well as up to 24 hours after the discontinua-
tion of enoxaparin treatment. Bleeding events were defined
as any of the following: overt bleeding resulting in death,
retroperitoneal bleeding, intracranial hemorrhage, decrease
in hemoglobin concentration ≥2 g/dL, absolute decrease in
hematocrit concentration of at least 12% from baseline, or
administration of ≥2 units of packed red blood cells. A
bleeding event was captured if it occurred at any point dur-
ing enoxaparin administration as well as up to 24 hours
after discontinuation of enoxaparin treatment. If a bleeding
event occurred while enoxaparin was an active medication
on the medication profile but was being held for more than
24 hours, that bleeding event was not included as an event.
Secondary outcome measures included the characteriza-
tion of enoxaparin doses (mg/kg), development of thrombo-
embolic events at 30 days after discontinuation of
enoxaparin therapy, and hospital mortality differences
between cases and controls. All data needed for evaluation
were collected by one investigator (JCH) to maintain con-
sistency throughout the data collection process.
Student’s t test and the χ2 test were used to analyze baseline
characteristics for continuous and categorical data between
groups, as appropriate. To characterize the therapeutic doses of
enoxaparin prescribed, the dose (mg/kg) given for each patient
was calculated. This was compared between the case and con-
trol groups to see if a difference in dosing strategies among
BMI classes existed. Bleeding events between groups were
analyzed using χ2 analysis. Data were then separated into that
for patients who experienced a bleeding event and that for
Hagopian et al 1643
patients who did not. Univariate factors with a P value < .20
were placed into a multivariate logistic regression model in a
stepwise manner to determine independent predictors of bleed-
ing. Appropriateness of the model was evaluated using the
Hosmer and Lemeshow test. All statistical analyses were per-
formed with the aid of SPSS software (version 16.0; SPSS Inc,
Chicago, IL). A P value of .05 was used to determine statistical
significance.
Results
Patient Population
The original query returned a total of 1389 patients who
were on treatment doses of enoxaparin in the latter half of
2009. After those patients who met the exclusion criteria
were removed from the study, 948 patients remained. A
total of 167 patients had a BMI ≥40 kg/m2, and 781 patients
had a BMI <40 kg/m2.
A total of 300 patients were included in the final analy-
sis; 100 patients in the morbidly obese group and 200
patients in the control group. To further characterize the
control group, the patients were divided into 3 BMI groups:
<18.5, 18.5 to 29.9, and 30 to 39.9 kg/m2. The number of
patients that fell into these BMI groups was 3, 130, and 67
patients, respectively. The largest patient included in the
study weighed 175.5 kg, corresponding to a BMI of 66.4
kg/m2. The smallest patient included in the study weighed
48.3 kg, corresponding to a BMI of 18.3 kg/m2.
The majority of patients were female (62.3%), Caucasian
(65.7%), and admitted to the medical service (96.7%).
Patients were well matched between the case and control
groups regarding baseline characteristics, with the excep-
tion that patients with a BMI ≥40 kg/m2 were more likely to
be female, younger, and have diabetes or hypertension
(Table 1).
Enoxaparin Use
Enoxaparin use in cases and controls is summarized in
Table 2. The number of patients receiving dose changes (11
vs 17, P = .48) as a result of dose recalculation, changing
renal function, or antifactor Xa level did not differ between
groups. Both the initial and final doses (mg/kg) were sig-
nificantly lower in the morbidly obese group when com-
pared with controls. In both groups, the highest dose given
was 150 mg as a single dose. The enoxaparin dosing ranges
for cases and controls were 80 to 150 mg and 50 to 150 mg,
respectively.
Antifactor Xa levels were not routinely drawn in either
the morbidly obese or control patients. In total, 27 antifac-
tor Xa levels were drawn. Of those 27 levels, only 5 were
drawn appropriately, thus limiting the utility for
interpretation.
Downloaded from aop.sagepub.com at University of New England on June 10, 2015
Primary End Point
The primary outcome measure of bleeding events up to 24
hours after the discontinuation of enoxaparin treatment did
not differ between cases and controls. A total of 29 (29%)
and 47 patients (23.5%) experienced a bleeding event,
respectively (P = .30). A breakdown of the definitions that
were met that determined the bleeding outcomes in the
study is shown in Table 3. The definition that most com-
monly classified a bleeding event in the study was an abso-
lute decrease in hematocrit concentration of at least 12%
from baseline, with 28 total events in the study (5 events in
morbidly obese patients vs 23 events in controls). One
bleeding event in the morbidly obese group and 4 in the
control group occurred in patients whose enoxaparin was
held for more than 24 hours. These 5 events were included
in the analysis because they met criteria for bleeding events
at a different time point while enoxaparin was being
administered.
A post hoc analysis was conducted to evaluate if thera-
peutic use of enoxaparin was associated with increased
bleeding events when compared with patients of normal
weight (BMI group 18.5-29.9 kg/m2). A total of 29 mor-
bidly obese patients (29%) and 30 normal-weight patients
(23.1%) experienced a bleeding event; P = .43.
Secondary End Points
The development of new thromboembolic events did not
differ between cases and controls at 30 days after discon-
tinuation of enoxaparin therapy. Two thromboembolic
events occurred in the morbidly obese group compared with
7 in the control group; P = .72.
One patient (1%) in the morbidly obese group and 4
patients (2%) in the control group expired during the index
admission; P = .90. None of the deaths reported were a
result of the study medication or a bleeding event. The 1
death in the morbidly obese group was a result of respira-
tory failure. The 4 deaths in the control group were a result
of respiratory failure (2), underlying malignancy, and myo-
cardial infarction.
Independent risk factors for bleeding were assessed
using multivariate analysis. Univariate factors with a P
value <.20 that were included in the multivariate analysis
were the following: female gender, immunosuppression,
malignancy, warfarin use, final enoxaparin dose less than
0.9 mg/kg per dose, duration of enoxaparin use for more
than 48 hours, peak SCr >1.4 mg/dL, hematocrit less than
33%, and morbid obesity. Morbid obesity was forced into
the analysis to determine the impact of this weight category
on bleeding. The results of the multivariate analysis are
shown in Table 4. The final stepwise analysis included 4
variables: female gender, concomitant warfarin use, final
1644 Annals of Pharmacotherapy 47(12)

Table 1.  Baseline Characteristics Between BMI Groups.

BMI ≥ 40 kg/m2 BMI < 40 kg/m2

(n = 100) (n = 200) P Value
Males, n (%) 25 (25) 88 (44) <.01
Caucasian, n (%) 62 (62) 135 (67.5) .34
Age, yearsa 50.3 ± 12.9 57.6 ± 16.9 <.01
Length of stay, daysa 9.8 ± 10.8 9.9 ± 12.2 .95
Admission to medicine unit, n (%) 97 (97) 193 (96.5) 1.00
Underlying comorbidities, n (%)
 Hypertension 63 (63) 94 (47) .01
  CAD or CHF 19 (19) 58 (29) .06
 Diabetes 39 (39) 42 (21) <.01
  Asthma or COPD 16 (16) 26 (13) .48
 IBS 3 (3) 9 (4.5) .76
  Immunocompromised state 21 (21) 44 (22) .84
 Malignancy 41 (41) 95 (47.5) .29
 Cirrhosis 0 1 (0.5) 1.00
 Other 69 (69) 156 (78) .09
Medications that increase bleeding risk, n (%)b
 Aspirin 27 (27) 75 (37.5) .70
 NSAIDs 16 (16) 21 (10.5) .17
 Clopidogrel 8 (8) 16 (8) 1.00
 Warfarin 50 (50) 89 (44.5) .37
 Steroids 20 (20) 45 (22.5) .62
 Otherc 4 (4) 10 (5) .78
Comorbid disease states that increase bleeding risk, n (%)
  Factor V deficiency 4 (4) 0 .01
Baseline laboratory valuesa
  Hemoglobin, g/dL 11.5 ± 2.4 11.1 ± 2.0 .17
  Hematocrit, % 34.1 ± 6.8 33.9 ± 5.8 .93
  Platelets, k/mm3 263.3 ± 121.0 258.8 ± 130.3 .78
  Baseline serum creatinine, mg/ 0.82 ± 0.26 0.84 ± 0.30 .54
dL
  Peak serum creatinine, mg/dL 0.95 ± 0.43 0.94 ± 0.38 .71
 INR 1.28 ± 0.29 1.29 ± 0.33 .64
  Renal dysfunction, n (%) 11 (11) 28 (14) .47

Abbreviations: BMI, body mass index; CAD, coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; IBS,
inflammatory bowel syndrome; NSAID, nonsteroidal anti-inflammatory drug; INR, international normalized ratio.
a
Mean and standard deviation, unless otherwise stated.
b
Medications being concomitantly administered with enoxaparin.
c
Including direct thrombin inhibitors (argatroban, bivalirudin), glycoprotein IIb/IIIa inhibitors (abciximab, eptifibatide, tirofiban), and prasugrel.

Table 2.  Enoxaparin Use for BMI Groups.

BMI ≥ 40 kg/m2 BMI < 40 kg/m2

(n = 100) (n = 200) P Value
Initial enoxaparin dose, mg/kg 0.96 ± 0.11 1.05 ± 0.19 <.01
Final enoxaparin dose, mg/kg 0.98 ± 0.11 1.04 ± 0.17 <.01
Twice-daily dosing, n (%) 97 (97) 183 (91.5) .07
Number of doses, mean 6.41 ± 5.90 6.90 ± 6.82 .54
Enoxaparin indication
  ACS, n (%) 8 (8) 18 (9) .77
  DVT, n (%) 59 (59) 102 (51) .19
  PE, n (%) 34 (34) 63 (31.5) .66
  Valve replacement, n (%) 4 (4) 14 (7) .30
  Atrial fibrillation, n (%) 9 (9) 24 (12) .43
  Other, n (%) 5 (5) 6 (3) .52

Abbreviations: BMI, body mass index; ACS, acute coronary syndrome; DVT, deep vein thrombosis; PE, pulmonary embolism.

Downloaded from aop.sagepub.com at University of New England on June 10, 2015

Hagopian et al 1645

Table 3.  Breakdown of Bleeding Event Definitions.

BMI ≥ 40 kg/m2 BMI < 40 kg/m2

Bleeding Event Definition Met, n (%) (n = 100) (n = 200)
Absolute decrease in hematocrit concentration of at least 12% from baseline only 5 (5) 23 (11.5)
Decrease in hemoglobin concentration ≥2 g/dL only 1 (1) 0
Administration of ≥2 units of packed red blood cells only 10 (10) 6 (3)
Retroperitoneal bleeding and administration of ≥2 units of packed red blood cells 1 (1) 0
Decrease in hemoglobin concentration ≥2 g/dL and absolute decrease in hematocrit 6 (6) 8 (4)
concentration of at least 12% from baseline
Decrease in hemoglobin concentration ≥2 g/dL and administration of ≥2 units of 2 (2) 0
packed red blood cells
Absolute decrease in hematocrit concentration of at least 12% from baseline and 2 (2) 6 (3)
administration of ≥2 units of packed red blood cells
Decrease in hemoglobin concentration ≥2 g/dL and absolute decrease in hematocrit 2 (2) 4 (2)
concentration of at least 12% from baseline and administration of ≥2 units of
packed red blood cells

Table 4.  Multivariate Analysis of Independent Risk Factors for Bleeding.a

Factor Bleed Versus No Bleed Adjusted Odds Ratio 95% CI P Value

Female gender, n (%) 56 (73.7) vs 131 (58.5) 2.05 1.12-3.73 .02
Warfarin use, n (%) 26 (34.2) vs 113 (50.4) 0.46 0.26-0.81 <.01
Final enoxaparin dose <0.9 mg/kg, n (%) 13 (17.1) vs 16 (7.1) 2.35 1.01-5.47 .04
Duration of enoxaparin use >48 hours, n (%) 53 (69.7) vs 107 (47.8) 2.42 1.35-4.33 <.01
a
Overall model P value <.001; Hosmer and Lemeshow test goodness of fit = 0.918; Nagelkerke R2 = 0.13.

enoxaparin dose <0.9 mg/kg per dose, and duration of
enoxaparin use of more than 48 hours.
Enoxaparin doses <0.9 mg/kg (adjusted odds ratio
[AOR] = 2.35; 95% CI = 1.01-5.47; P = .04), durations of
therapy beyond 48 hours (AOR = 2.42; 95% CI = 1.35-
4.33; P < .01), and female gender (AOR = 2.05; 95% CI =
1.12-3.73; P = .02) were associated with bleeding, whereas
warfarin use (AOR = 0.46; 95% CI = 0.26-0.81; P < .01)
was associated with fewer bleeding events. Morbid obesity
was not associated with increased bleeding events.
Discussion
In this retrospective review, hospitalized patients with mor-
bid obesity were not found to have increased bleeding
events associated with therapeutic enoxaparin use.
Independent risk factors for bleeding included the follow-
ing: lower final enoxaparin doses, longer durations of
enoxaparin, and female gender.
To our knowledge, there are no data that confirm that the
risk factors found in this study are risk factors in other stud-
ies as well. We found that patients who received an increased
number of doses of enoxaparin were at a higher bleeding
risk. This correlation is plausible because the greater the
number of doses of an anticoagulant to which a patient is
exposed, the higher the risk of bleeding. An alternative
explanation is that patients who are on enoxaparin for lon-
ger durations of time could experience a lab-defined bleed-
ing event because of drops in hemoglobin or hematocrit
from prolonged hospital stays or hemodilution from fluid
administration. In the multivariate analysis, those patients
who were administered an average of 9.3 doses of enoxapa-
rin versus 5.9 doses of enoxaparin were at an increased risk
of bleeding. In addition to increased enoxaparin exposure,
female gender was also found to increase the risk of bleed-
ing. We are unaware of a likely reason for why women
would be at a higher risk of bleeding. An interesting finding
from the multivariate analysis was that lower final enoxapa-
rin doses (<0.9 mg/kg/dose) were associated with increased
bleeding events. This outcome may have been a result of
bleeding or perceived bleeding risk. For example, providers
may have either empirically lowered or actually lowered
doses because of a confirmed bleeding event. Increased SCr
and renal dysfunction have been well characterized in the
literature as risk factors for bleeding because of drug accu-
mulation because enoxaparin is renally excreted.3,10 Despite
being well characterized in the literature, renal dysfunction
was not found to be an independent risk factor for bleeding
in this retrospective review. Of note, patients were matched
based on the presence of renal dysfunction (SCr ≥ 1.4 mg/
dL) at any time during their index hospitalization. However,
the SCr levels assessed in the study were those measured

Downloaded from aop.sagepub.com at University of New England on June 10, 2015

1646 Annals of Pharmacotherapy 47(12)
during the time of enoxaparin use, accounting for differ-
ences between the 2 groups despite the matching process to
minimize these differences.
The current study did not show that morbid obesity was
a risk factor for bleeding but does support enoxaparin dos-
ing based on ABW, with use of dose capping at 150 mg in
patients weighing up to 175.5 kg. It is unclear from previ-
ous pharmacokinetic and clinical data5-8 if this dosing prac-
tice is supported from a safety perspective.
One of the earliest studies that assessed the pharmacoki-
netics of enoxaparin in obese patients was performed by
Sanderink et al.5 This study was performed in 24 obese
patients with BMIs between 30 and 40 kg/m2 and 24 matched
nonobese patients, with BMIs ranging from 18 to 25 kg/m2.
The results of this study show that the mean antifactor Xa
level was greater in obese volunteers compared with non-
obese volunteers (1.563 IU/mL vs 1.488 IU/mL), but this dif-
ference was not statistically different.5 In accordance with
similar mean antifactor Xa levels, no hemorrhagic complica-
tions were seen in this study. Therefore, study investigators
concluded that enoxaparin was well tolerated, by means of
mortality and major bleeding events, in both patient popula-
tions, and therefore, there is no need for dose modification
for obese patients with BMI up to 40 kg/m2.
Bazinet et al6 performed a pharmacokinetic study to
assess how antifactor Xa levels vary in response to patient
weight. In this study, 51 obese patients (BMI >30 kg/m2)
were compared with 69 normal-weight patients (BMI =
18-30 kg/m2). For those patients receiving once-daily dos-
ing of enoxaparin, the mean antifactor Xa level was 1.15
IU/mL in obese patients compared with 1.13 IU/mL in nor-
mal-weight patients.6 Similarly, for patients receiving
twice-daily dosing of enoxaparin, the mean antifactor Xa
level was 1.17 IU/mL in obese patients compared with 1.12
IU/mL in normal-weight patients.6 After adjusting for age,
gender, and renal function, it was found that antifactor Xa
levels increased by 0.01 IU/mL for every increase of 1 kg/
m2 (95% CI = 0.002-0.017).6 The safety analysis found that
antifactor Xa levels did not correlate with bleeding events.
Based on these results, it was suggested that dosing adjust-
ments would not be required in obese patients.
A retrospective cohort study from the American College
of Cardiology American Heart Association Guidelines
(CRUSADE) database aimed to determine if there was an
association between enoxaparin dose and major bleeding
across multiple BMI classes. A total of 19 061 patients were
eligible for analysis. Results show that in obese patients,
clinicians tended to use less than the recommended 1 mg/kg
dose. In addition, patients who were classified as obese
(BMI ≥30 kg/m2) received a statically significantly lower
dose of enoxaparin compared with nonobese patients.
However, when the recommended enoxaparin dose was
used compared with a lower dose, it was associated with a
higher bleeding risk in those patients who weighed more
Downloaded from aop.sagepub.com at University of New England on June 10, 2015
than 150 kg (AOR = 2.42; 95% CI = 0.70-8.37).7 In contrast
to the above-mentioned studies, these data showed that in
larger-weight patients, therapeutic enoxaparin use is associ-
ated with increased bleeding risk.
A small case series performed at our institution by Deal
et al8 investigated antifactor Xa levels in morbidly obese
patients treated with enoxaparin. In this retrospective
review, 26 patients with morbid obesity (BMI > 40 kg/m2)
were included. Results found that in the majority of the
patients included, dosing below the recommended dose of 1
mg/kg every 12 hours resulted in antifactor Xa levels either
at or above goal.8 In addition, bleeding events were more
frequent in those patients with supratherapeutic antifactor
Xa levels.8
Although the data are conflicting, these previous studies
allude to a potential increased bleeding risk when using
therapeutic doses of enoxaparin in the morbidly obese
patient population. The current study showed that compared
with controls, morbidly obese patients received a statisti-
cally significantly lower dose of enoxaparin (mg/kg), but
doses were still comparable to the recommended 1 mg/kg,
twice-daily dosing regimen. As such, the lower dose given
to morbidly obese patients compared with controls is
unlikely to be clinically significant. It is noted, however,
that dose capping was utilized in the morbidly obese group.
The highest single dose given in the study was 150 mg, yet
the largest patient in the morbidly obese group weighed
175.5 kg. At a minimum, in morbidly obese patients, doses
of enoxaparin were 0.85 mg/kg, which qualified as appro-
priate therapeutic doses according to our study criteria.
From this study, we cannot assess the bleeding risk associ-
ated with enoxaparin dosing in patients weighing greater
than 150 kg who receive a single dose of enoxaparin greater
than 150 mg, highlighting the need for a prospective evalu-
ation of this study question. Despite the limitation that a
small number of patients (n = 9, data not shown) received
empirical dose capping, there was no difference in bleeding
rates between the 2 patient populations.
Other important limitations regarding this study should
be noted. Exposure to enoxaparin was of short duration in
the majority of patients. It cannot be concluded from this
study if longer enoxaparin use would result in increased
number of bleeding events and, thus, limits extrapolation to
patients being treated with enoxaparin for an extended
duration of time. The retrospective nature of the study pre-
sented challenges in capturing a true bleeding event, and
multiple comparisons utilized in this small cohort increase
the chances of a type 1 error. Despite reviewing each
patient’s electronic chart, the majority of the time, an actual
bleeding event was not specifically noted. To account for
the first limitation, a surrogate definition of bleeding was
used. The development of a bleeding event was a compila-
tion of bleeding definitions used in the historical studies
reviewed above. The specific bleeding definitions chosen
Hagopian et al 1647
from these studies were meant to capture bleeding events
that we deemed to be clinically meaningful and of both
minor and major severities. We also chose these bleeding
definitions to capture bleeding events that could be easily
ascertained in a retrospective review without relying on
specific providers to note bleeding events based on signs
and symptoms of bleeding in written charts. The decision
was made not to use the Thrombolysis in Myocardial
Infarction (TIMI) or Global Utilization of Streptokinase
and t-PA for Occluded Coronary Arteries (GUSTO) defini-
tions of bleeding events in this study. The reason these defi-
nitions were not utilized is that they are intended to be used
in prospective investigations where ongoing surveillance
for adverse events can be ensured. The downside of using
the definitions in this study is that there is a high likelihood
that more bleeding events were captured than would be
clinically considered a bleeding event in daily practice.
Considering the high bleeding rates seen in the study (20%-
30%), it is likely that this occurred. The retrospective design
of this study may also contribute to the study results by the
presence of confounders that were not identified during
data collection. For example, factors such as uncontrolled
hypertension were unable to be accounted for in this analy-
sis. Another limitation of this study is the fact that at our
institution, antifactor Xa levels are not routinely obtained,
and drug exposure as measured by this method was not
available for analysis.
A final limitation that should be noted is our use of SCr
as a measure of renal dysfunction versus CrCl. Because our
main aim was to provide insight into the impact of dosing of
enoxaparin in patients with morbid obesity, we used a SCr
cutoff rather than estimated renal function because the
method for estimating CrCl in extreme obesity is controver-
sial, and comparisons of estimated CrCl among various
weight categories are difficult to standardize. Despite the
limitation of using this cutoff for matching patients, it
should be noted that very few patients (39 out of 300) expe-
rienced a SCr >1.4 mg/dL during their enoxaparin treat-
ment. Furthermore, the mean peak SCr during treatment
was close to 1 mg/dL in both study groups, with less than
15% of the patients with a baseline SCr >1.1 mg/dL.
Additionally, all patients were appropriately dosed accord-
ing to their estimated renal function (using CrCl with the
Cockcroft-Gault method), and only 20/300 patients received
a once-daily dosing strategy. Differences in age and gender
among the weight categories also may have had an impact
on renal function estimates, but the true effects on our study
population cannot be determined.
Despite the limitations mentioned, this study does lend
support to our institution’s current enoxaparin dosing prac-
tices in the morbidly obese. Dosing enoxaparin in morbidly
obese patients (up to a weight of 175 kg) using total body
weight with doses capped at 150 mg does not pose a signifi-
cant safety risk regarding increased bleeding events. Of note,
Downloaded from aop.sagepub.com at University of New England on June 10, 2015
the mean duration of use among these patients was short, at
approximately 3 days. The high number of morbidly obese
patients included in the study allowed the study question to be
evaluated in a large number of patients that exceeded the
number of morbidly obese patients included in previous stud-
ies. Future studies should include a prospective evaluation of
the study question to address the limitations noted above.
Conclusion
Current practices of therapeutic enoxaparin dosing are not
associated with increased bleeding events in the morbidly
obese patient population. Results suggest that dosing enoxa-
parin in morbidly obese patients (up to weight of 175 kg)
with doses capped at 150 mg does not pose a significant
safety risk regarding bleeding events. To further define
bleeding risk among morbidly obese patients, prospective
research following recommended dosing using total body
weight is needed in this special patient population.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to
the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research,
authorship, and/or publication of this article.
Authors’ Note
These results were presented as an abstract and a poster at the
American College of Clinical Pharmacy Annual Meeting; October
16-19, 2011; Pittsburgh, PA.
References
1. Centers for Disease Control and Prevention. US adult obesity
facts. http://www.cdc.gov/obesity/data/adult.html. Accessed
October 11, 2012.
2. Sturm R, Hattori A. Morbid obesity rates continue to rise rap-
idly in the United States. Int J Obes (Lond). 2012;37:889-891.
doi:10.1038/ijo.2012.159.
3. Nutescu EA, Spinler SA, Wittkowsky A, Dager WE. Low-
molecular weight heparins in renal impairment and obesity:
available evidence and clinical practice recommendations
across medical and surgical settings. Ann Pharmacother.
2009;43:1064-1083. doi:10.1345/aph.1L194.
4. Garcia DA, Baglin TP, Weitz JI, Samama MM. Parenteral anti-
coagulants. Chest. 2012;141(suppl):e24S-e43S. doi:10.3178/
chest.11-2291.
5. Sanderink GJ, Liboux AL, Jariwala N, et al. The pharmacoki-
netics and pharmacodynamics of enoxaparin in obese volun-
teers. Clin Pharmacol Ther. 2002;72:308-318. doi:10.1067/
mcp.2002.127114.
6. Bazinet A, Almanric K, Brunet C, et al. Dosage of enoxapa-
rin among obese and renal impairment patients. Thromb Res.
2005;116:41-50. doi:10.1016/j.thromres.2004.10.004.
1648 Annals of Pharmacotherapy 47(12)

7. Spinler SA, Ou FS, Roe MT, et al. Weight-based dosing of
enoxaparin in obese patients with non-ST-segment elevation
acute coronary syndromes: results from the CRUSADE ini-
tiative. Pharmacotherapy. 2009;29:631-638. doi:10.1592/
phco.29.6.631.
8. Deal EN, Hollands JM, Riney JN, Skrupky LP, Smith JR,
Reichley RM. Evaluation of therapeutic anticoagulation with
enoxaparin and associated anti-Xa monitoring in patients
with morbid obesity: a case series. J Thromb Thrombolysis.
2011;32:188-194. doi:10.1007/s11239-011-0584-7.
9. Centers for Disease Control and Prevention. About BMI for
adults. http://www.cdc.gov/healthyweight/assessing/bmi/
adult_bmi/index.html. Accessed March 10, 2013.
10. Lim W, Dentali F, Eikelboom JW, Crowther MA. Meta-analysis:
low-molecular-weight heparin and bleeding in patients with
severe renal insufficiency. Ann Intern Med. 2006;144:673-684.
doi:10.7326/0003-4819-144-9-200605020-00011.

Downloaded from aop.sagepub.com at University of New England on June 10, 2015