J Thromb Thrombolysis (2009) 27:267–273

DOI 10.1007/s11239-008-0207-0

A descriptive evaluation of unfractionated heparin use during

pregnancy
Nathan P. Clark Æ Thomas Delate Æ Daniel M. Witt Æ
Suzanne Parker Æ Robert McDuffie

Published online: 8 March 2008

Ó Springer Science+Business Media, LLC 2008

Abstract Background The mainstay of oral anticoagulant Keywords Pregnancy
Heparin
Thromboembolism

therapy, warfarin sodium, crosses the placenta during Hemorrhage
pregnancy and may cause fetal complications. Unfrac-
tionated heparin (UFH) and low-molecular-weight heparin
(LMWH) do not cross the placenta and have demonstrated Introduction
utility in the prevention and treatment of thrombosis during
pregnancy. Objectives The purpose of this study was to Anticoagulation (AC) therapy is required occasionally
review treatment strategy, indication, and maternal and during pregnancy. Indication for AC includes prevention
fetal outcomes in anticoagulated pregnancies at Kaiser and treatment of venous thromboembolism (VTE) includ-
Permanente Colorado. Patients/Methods We identified 103 ing deep venous thrombosis (DVT) and pulmonary
pregnancies in 93 mothers prescribed an anticoagulant embolism (PE), prosthetic mechanical heart valve pro-
during a pregnancy occurring between January 1, 1998 and phylaxis, and prevention of pregnancy complications in the
March 31, 2005. Results The majority of patients were setting of thrombophilia (acquired or genetic disorder
treated with UFH (89.3%). Indications for anticoagulation predisposing to thrombosis and/or spontaneous loss of
included venous thromboembolism (VTE) prophylaxis pregnancy).
(53.4%), history of pregnancy loss (29.1%), acute VTE Recurrent or late pregnancy loss, severe recurrent pre-
(16.5%), and history of cerebral vascular accident (CVA) eclampsia, or intrauterine growth restriction (IUGR) occurs
(1.0%). There were no maternal deaths. Fetal demise in 1–5% of pregnancies [1, 2]. These obstetrical compli-
occurred in 8 pregnancies (7.8%) at a median 14 weeks cations may be related to impaired placental perfusion and
gestation (range 7–22 weeks). No fetal demise occurred in are 3- to 8-times more common in patients with thrombo-
pregnancies treated for acute VTE or history of CVA. philic disorders [2]. Women with multiple thrombophilic
There were two occurrences of pulmonary embolism disorders are estimated to have a 14-fold increased risk of
(1.9%) and two hemorrhagic events requiring transfusion pregnancy loss [2].
(1.9%). Conclusions Maternal and fetal adverse events The mainstay of oral AC therapy, warfarin sodium,
were infrequent in our population of anticoagulated preg- crosses the placenta and places the fetus at risk for com-
nancies. UFH remains a viable option among more plications including birth defects, bleeding, and minor
expensive LMWH products. neurological dysfunction following delivery [3–5].
Unfractionated heparin (UFH) and low-molecular-weight
heparin (LMWH) do not cross the placenta and have
N. P. Clark (&)
T. Delate
D. M. Witt
S. Parker
Clinical Pharmacy Anticoagulation Service, Kaiser Permanente demonstrated utility in the prevention and treatment of
Colorado, 280 Exempla Circle, Lafayette, CO 80026, USA thrombosis during pregnancy [6–10]. However, for many
e-mail: [email protected] indications, small, observational trials and case series
provide the only information available to guide UFH and
R. McDuffie
Department of Obstetrics and Gynecology, Colorado Permanente LMWH therapy during pregnancy. As a result, recom-
Medical Group, 2045 Franklin St., Denver, CO 80205, USA mendations for AC management during pregnancy are

123
268 N. P. Clark et al.

often extrapolated from studies of UFH and LMWH in Pharmacy claims were queried to identify from the above
non-pregnant populations [11]. identified patients, patients aged 18–45 years who had
Low-molecular-weight heparin is becoming the pre- purchased AC medications (UFH, LMWH, warfarin) during
ferred parenteral anticoagulant during pregnancy [12]. the 90 days prior and 300 days after the positive pregnancy
Potential benefits of LMWH over UFH include predictable test date. Chart reviews were conducted on those patients
anticoagulant response from weight based dosing, ease of who were identified in both groups. Patients were included
outpatient use, lower risk of osteopenia, and lower risk of in the analysis if the chart review determined that AC was
heparin-induced thrombocytopenia (HIT). Data clearly administered during the time of a pregnancy. Patients were
demonstrating the superiority of LMWH over UFH are excluded if they were less than 18 years of age at the esti-
lacking and emerging information indicates that outpatient mated time of conception, had an elective pregnancy
administration of UFH is a feasible, cost-effective alter- termination in the 1st trimester or an ectopic pregnancy,
native to LMWH [13]. were KPCO employees at the time of pregnancy, did not
The purpose of this study is to evaluate our experience have continuous membership in KPCO throughout the
with AC in pregnancy by describing treatment strategies, pregnancy, or received low concentration heparin for
indications for use, and maternal and fetal outcomes in intravenous line maintenance.
pregnant patients receiving AC medications monitored in a
centralized anticoagulation service. Study outcomes and data collection

The primary study outcomes were occurrence of maternal

Methods hemorrhage, thromboembolism, death, and fetal demise.
Maternal hemorrhage was defined as any unscheduled visit
Setting due to bleeding or any hemorrhagic complication noted in
the patient progress report during pregnancy or within
This retrospective, longitudinal study was conducted at 6 weeks postpartum; including gastrointestinal, intracra-
Kaiser Permanente Colorado (KPCO), an integrated nial, intraabdominal, intraarticular, intraocular, vaginal, or
healthcare delivery system providing care to approximately cesarean section (C-section) incision bleeding. Maternal
450,000 patients in the Denver–Boulder metropolitan area. thromboembolism included diagnosis of DVT, PE, cerebral
The study was reviewed and approved by the KPCO vascular accident (CVA), or heart valve thrombosis during
Institutional Review Board. pregnancy or within 6 weeks postpartum. Fetal demise was
Anticoagulation services at KPCO are provided by the defined as loss of fetal heart beat at any time after confir-
centralized Clinical Pharmacy Anticoagulation Service mation of a viable intrauterine pregnancy. Secondary
(CPAS). The KPCO CPAS provides comprehensive ser- outcomes were limited to descriptive analyses of patient
vices for all KPCO patients requiring AC therapy and characteristics, including indication for AC, presence of
currently monitors more than 7,000 patients. Working thrombophilia, family history of thromboembolism, and
collaboratively with the referring physician, CPAS clinical gravidity/parity. Pregnancy complications assessed were
pharmacists initiate AC therapy, order relevant laboratory preeclampsia, oligohydraminos (measure by amniotic fluid
tests, adjust AC medications as necessary, and refill AC index [AFI]), IUGR, abruptio placentae, and blood loss
medications. The KPCO integrated, electronic medical, during delivery.
pharmacy, and laboratory records system and CPAS AC Details regarding the AC treatment strategy were also
database (Dawn-ACÒ; 4S Systems, Ltd., Cumbria, UK) collected, including medication type, dose, intensity, and
were utilized to identify patients, treatments, and outcomes monitoring. Unfractionated heparin dosing strategies were
for this study. categorized as adjusted dose or set dose. Adjusted dose
refers to treatment to a target anti-Xa level while set dose
Patients UFH included any strategy not adjusted based on anti-Xa
level. Low-molecular-weight heparin strategies were cate-
Patients were included if they purchased at least one AC gorized into therapeutic or prophylactic doses. Therapeutic
medication (warfarin, UFH or LMWH) from a KPCO doses were either weight-based (i.e. 200 IU/kg dalteparin
pharmacy during a pregnancy occurring between January 1, daily or in divided doses, or enoxaparin 1 mg/kg twice
1998 and March 31, 2005. Laboratory records were queried daily) or adjusted based on anti-Xa level. Estimated
to identify females who received routine pregnancy-related blood loss during delivery was categorized as \750cc,
tests (pregnancy, qual, serum; pregnancy, qual, urine; pre- 750–999cc, 1000–1499cc, 1500–1999cc, [2000cc, or
natal panel; prenatal partial; prenatal RPR) that were transfusion required. In addition, the type of anesthesia
reported as positive for pregnancy during the study period. used during labor and delivery was collected.

123
A descriptive evaluation of unfractionated heparin use during pregnancy 269

Data were obtained by manual review of the included Table 1 Baseline characteristics
patients’ electronic medical records and a perinatal data- Characteristica Value
base using a standardized data abstraction tool. Pregnancy
complications were assessed by review of the standardized Mean age in yearsb (SD) 30.8 (4.9)
labor and delivery form used for all live births at KPCO Mean body mass index (SD) 28.8 (7.4)
and validated by two reviewers (NC, SP). Disagreements Anticoagulation indication
between the reviewers were resolved by a third reviewer VTE Prophylaxis (n, %) 55 (53.4)
(RM). History of pregnancy loss (n, %) 30 (29.1)
Acute VTE (n, %) 17 (16.5)
Statistical analysis History of CVAc (n, %) 1 (1.0)
Previous miscarriage
The analysis was primarily descriptive. Overall patient None (n, %) 36 (35.0)
characteristics are reported as means, medians, and stan- 1 (n, %) 31 (30.0)
dard deviations for interval- and ratio-level variables (e.g., C2 (n, %) 36 (35.0)
age, length of AC therapy) and proportions for categorical Personal history of VTE (n, % Yes) 59 (57.3)
variables. Bivariate analyses (non-parametric Wilcoxon Thrombophilia (n, % Yes) 49 (47.6)
rank sum test and Pearson’s chi-square) were used to Thrombophilia type (%, among those with thrombophilia n = 49)
describe and test proportions and mean/median values of Factor V leiden (n, %) 20 (40.8)
selected characteristics by cohort. Antiphospholipid antibody syndrome (n, %) 16 (32.7)
Protein S deficiency (n, %) 3 (6.1)
Protein C deficiency (n, %) 2 (4.1)
Results Prothrombin 20210 mutation (n, %) 2 (4.1)
Nephrotic syndrome (n, %) 1 (2.0)
A total of 17,439 positive pregnancy tests in 14,459 women Antithrombin deficiency (n, %) 1 (2.0)
were identified during the study period. Of these potential Elevated factor VIII activity (n, %) 1 (2.0)
pregnancies, 140 (0.8%) had a purchase of AC therapy Multiple thrombophilic traits (n, %) 3 (6.1)
within 90 days prior to and 300 days after the pregnancy
test. Chart review determined that patients for 25 women Abbreviations: SD—standard deviation; VTE—venous thromboem-
bolism; CVA—cerebral vascular accident
did not have continuous membership in KPCO throughout a
Total number of pregnancies, n = 103
the pregnancy and 12 pregnancies were receiving low b
As of conception
concentration heparin for intravenous line maintenance. c
Associated with illicit drug abuse
Thus, a total of 103 pregnancies were identified in 93
patients. One patient contributed three pregnancies and
eight patients contributed two pregnancies. Follow up was
complete for all patients. deviation 14.8 weeks). No patient was prescribed warfarin
Overall, patients were in their mid-childbearing years during pregnancy. In those patients on warfarin prior to
with ages ranging from 21 to 42 years, and had a high body conception (n = 14), nine (64%) continued warfarin until a
mass index at the time of pregnancy (Table 1). The positive pregnancy test, and then discontinued in favor of a
majority of pregnancies were anticoagulated for VTE heparin alternative. We found no evidence of birth defects
prophylaxis with either a personal history of VTE or in these nine cases. Five patients (36%) substituted heparin
known thrombophilia. No patient was anticoagulated to for warfarin prior to achieving pregnancy.
prevent thromboembolic complications associated with a The two intensities of adjusted dose UFH were 0.1–
prosthetic heart valve. Family history of VTE was found in 0.3 U/ml anti-Xa units and 0.3–0.7 U/ml anti-Xa units
8 (7.8%) of pregnancies and thrombophilia was present in drawn mid-interval (i.e. 6 h after the AM dose for every
47.6%. Three patients had more than one thrombophilic 12 h dosing). There were no patients monitored by an
trait. Two had protein S deficiency with a concomitant activated partial thromboplastin time. One patient received
heterozygous factor V Leiden mutation and the third a graduated set UFH dose where she received 5,000 units
patient had deficiencies of proteins C and S as well as twice daily in the first trimester, 7,500 units twice daily in
heterozygous factor V Leiden mutation. the second trimester, and 10,000 units twice daily in the
The vast majority of patients (89.3%) received UFH; third trimester. Her UFH dose is listed as ‘other’ in
others received LMWH or switched between the two dur- Table 2. Two patients were treated with therapeutic
ing pregnancy (Table 2). The mean duration of heparin intensity enoxaparin to a target peak anti-Xa level 0.5–
exposure (UFH or LMWH) was 24.8 weeks (standard 1.2 U/ml (drawn 4 h after the dose). Two patients were

123
270 N. P. Clark et al.

Table 2 Antepartum anticoagulation therapy conjunction with UFH or LWMH until the target INR of 2–
Variable Value (%)
3 was achieved. Anticoagulation was continued on average
for 4–6 weeks. Thirteen patients continued postpartum AC
Heparin type (%, n = 103) with warfarin long term. These patients were either on
UFH (n, %) 92 (89.3) warfarin at the time of review or had continued warfarin
LMWH (n, %) 6 (5.8) until leaving KPCO. All patients anticoagulated for ante-
Combination of LMWH and UFH (n, %) 5 (4.9) partum DVT or PE received at least 6 weeks of postpartum
Type UFH during pregnancy (% among anticoagulation.
patients receiving UFH, n = 92) There were no maternal deaths. Fetal demise occurred in
Adjusted dose (n, %) 62 (67.4) 8 (7.8%) pregnancies (Table 4). No fetal demise occurred
Set dose (n, %) 20 (22.6) in those patients anticoagulated for acute VTE or history
LMWH dose (% among patients receiving LMWH, n = 6) of CVA. Fetal demise by indication was 16.7% (5/30) in
Therapeutic (n, %) 4 (66.7) pregnancies treated for a history of pregnancy loss and
Prophylactic (n, %) 2 (33.3) 5.5% (3/55) for those treated for VTE prophylaxis
UFH target anti-Xaa (% among patients with an (P = 0.091). Thrombophilia was present in 45.7% (43/94)
adjusted heparin dose, n = 62) of live births and 75% (6/8) of fetal deaths (P = 0.112).
Anti-Xa target 0.3–0.7 U/ml (n, %) 35 (56.5) Fetal demise occurred at a median 14 weeks (range
Anti-Xa target 0.1–0.3 U/ml (n, %) 27 (43.5) 7–22 weeks).
UFH dose (% among patients with a set heparin dose, n = 30) There were two episodes of maternal thromboembolism,
5,000 U BID (n, %) 17 (56.7) both of which were PE. The first event was a patient with
7,500 U BID (n, %) 11 (36.7) presumed PE recurrence 4 days after hospital discharge for
10,000 U BID (n, %) 1 (3.3) her initial antepartum DVT and PE managed with once
Other (n, %) 1 (3.3) daily dalteparin (200 units/kg) during her 12th week of
Abbreviations: LMWH—low molecular weight heparin; UFH— pregnancy. Diagnostic imaging was not repeated during the
unfractionated heparin; U—units; ml—milliliters; BID—twice daily second hospitalization to objectively confirm the diagnosis.
a
Mid-interval Xa level She remained hospitalized for 5 days on intravenous UFH.
At discharge she was transitioned to subcutaneous UFH,
treated with prophylactic intensity enoxaparin 40 mg daily three times daily, titrated to a mid-interval anti-Xa level of
throughout pregnancy for VTE prophylaxis. 0.3–0.7 units/ml for the remainder of the pregnancy and
Duration of antenatal anticoagulation varied according had no recurrent thromboembolism. The second PE
to indication. The median onset of acute VTE was occurred 5 days postpartum in a patient who delivered by
19 weeks gestation (range 7–34 weeks). Patients with C-section. She received dose adjusted UFH to target an
acute VTE were anticoagulated immediately upon diag- anti-Xa level of 0.3–0.7 units/ml during pregnancy for
nosis and continued on therapeutic LMWH or UFH VTE prophylaxis with known elevation of anticardiolipin
throughout the remainder of the pregnancy. No patients antibodies and a history of VTE. At the time of diagnosis,
with acute VTE required interruption of anticoagulation for she had been receiving warfarin and her international
antenatal bleeding events.
Most patients received postpartum warfarin (Table 3).
Warfarin was either initiated as monotherapy or in
Table 4 Patient complications
Table 3 Postpartum anticoagulation therapy Complicationa Value
a
Variable Value (%) Fetal demise (n, %) 8 (7.8)
b
UFH coverage to warfarin (n, %) 57 (60.6) Maternal hemorrhage (n, %) 20 (19.4)
None (n, %) 20 (21.3) Antepartum bleed (n, %) 8 (7.8)
b
Warfarin only (n, %) 11 (11.7) Postpartum bleed (n, %) 12 (11.7)
LMWH coverage to warfarinb (n, %) 3 (3.2) Bleed requiring transfusion (n, %) 2 (1.9)
UFH only (n, %) 2 (2.1) Venous thromboembolismb (n, %) 2 (1.9)
Unknown (n, %) 1 (1.1) Preeclampsiac (n, %) 6 (5.8)

Abbreviations: LMWH—low molecular weight heparin; UFH— Abbreviations: C-section—caesarean section

a
unfractionated heparin Among total number of pregnancies, n = 103
a b
Among successful pregnancies, n = 94 Pulmonary embolism
b c
INR goal 2–3 Mild (n = 3) or severe (n = 3) preeclampsia

123
A descriptive evaluation of unfractionated heparin use during pregnancy 271

normalized ratio (INR) was therapeutic (INR = 2.3) the strategies. The results of this study provide a description of
day prior to onset of symptoms. the indications for AC that were most frequently found in
Maternal hemorrhage, by our definition, occurred in clinical practice. We found that during pregnancy, most
19% (20/103) of pregnancies. Hemorrhage was predomi- patients in our study received UFH prescribed primarily for
nantly vaginal (n = 13) or related to a cesarean incision VTE prophylaxis with either a personal history of VTE or
(n = 4). Most bleeding occurred in the postpartum period, known thrombophilia. Previous fetal demise was the next
including the two episodes requiring blood transfusion (1 most frequently encountered indication. The relative pro-
C-section and 1 vaginal bleed). Two patients suffered portion of these indications may change over time as
subchorionic placental hematoma and one patient suffered practice guideline recommendations regarding thrombo-
a subconjunctival hemorrhage. The fetus survived in each philia testing and anticoagulation management change. We
of these three cases. did not observe a change in prescribing preferences or
The first subchorionic hematoma occurred in a patient patterns over the duration of this study.
treated with moderate-intensity adjusted dose UFH for Comparing our results to clinical trial data is challeng-
VTE prophylaxis. The hematoma was diagnosed at ing given the heterogeneous nature of our population. All
15 weeks gestation. Unfractionated heparin was held for a anticoagulated mothers in our study survived pregnancy;
total of 8 weeks. The hematoma resolved and UFH was findings similar to previously reported results in a popu-
resumed. The second subchorionic hematoma occurred in a lation of pregnant women with a history of VTE receiving
patient treated with mini-dose UFH for history of preg- no antepartum AC [15]. We unexpectedly discovered two
nancy loss and was associated with vaginal bleeding. The patients who experienced pulmonary embolism while
hematoma was also diagnosed at 15 weeks gestation after receiving AC. The first event is possibly explained by the
which heparin was permanently discontinued. The sub- altered pharmacokinetic profile of LMWH observed during
conjunctival hemorrhage occurred in patient treated with pregnancy, causing a subtherapeutic level of AC [16]. It is
enoxaparin 40 mg daily for VTE prophylaxis. The bleed important to note in this case that repeat imaging was not
occurred at 32 weeks and was associated with no visual undertaken to objectively confirm the diagnosis. The sec-
impairment. Enoxaparin was continued through the ond event may have been related to insufficient overlap of
remainder of the pregnancy without additional complica- UFH when restarting warfarin postpartum. However, at the
tion. Most bleeding was minor and AC was often continued time of recurrence the patient was not anticoagulated for
(52.4%) or only temporarily interrupted (28.6%). acute VTE, so overlap was not deemed necessary. Despite
Blood loss at delivery was determined in 89 of 103 this, most of the patients received some degree of UFH or
pregnancies (eight fetal demise, five no data available). LMWH overlap when initiating warfarin postpartum.
Blood loss was reported as \750cc (83%, n = 74), 750– Comparing hemorrhage rates can be cumbersome due to
999cc (12%, n = 11), 1000–1499cc (4.5%, n = 4), no differing definitions of major hemorrhage. We did not dif-
patients had blood loss reported [1500cc or required ferentiate between major and minor hemorrhage. We
transfusion. included all bleeding noted upon chart review and did not
Pregnancy complications were encountered infre- attempt to discriminate between normal vaginal bleeding
quently. Preeclampsia developed in 5.8% of anticoagulated and hemorrhage related to AC treatment. The rate of all
pregnancies and was noted as severe in three cases (50%). hemorrhage observed in our analysis appears high relative to
There were three cases of oligohydraminos, one case of an other investigations possibly due to previous reports limit-
AFI \5 cm and two cases with an AFI \2 cm. No occur- ing their analysis to major hemorrhage [17, 18]. Much of the
rences of IUGR or abruptio placentae were found. Epidural hemorrhage identified was minor and most patients contin-
anesthesia was most frequently employed during labor and ued AC without interruption after the bleeding episode. Our
delivery (48%) followed by spinal anesthesia (22%). There low rate of maternal hemorrhage requiring transfusion of
were no documented cases of epidural or spinal hematoma. 1.9% (2/103) is similar to rates previously reported in pro-
spective evaluations of pregnant women receiving
prophylactic LWMH or UFH (0–2.9%) [9, 19–21].
Discussion The overall rate of fetal demise in this study was 8.7%,
which is lower than the rates reported in clinical trials, case
Anticoagulation during pregnancy is challenging due to the series, and cohort evaluations of prophylactic anticoagu-
lack of an oral agent safe for use in pregnancy and ran- lation for unexplained pregnancy loss in women with
domized, controlled trial data to guide therapy for many thrombophilia [19–25]. Our lower rate may be due to the
indications [14]. This study evaluated a heterogeneous inclusion of pregnancies that were anticoagulated for rea-
population of pregnant women from a large, centralized sons other than unexplained pregnancy loss. All patients
AC service who received a variety of AC treatment anticoagulated for acute VTE in our study had a successful

123
272 N. P. Clark et al.

pregnancy. A recent report of thrombophilic patients the majority of pregnant anticoagulated patients are rep-
receiving thromboprophylaxis during pregnancy found no resented in our sample. We were unable to assess heparin
fetal demise in women without a history of previous fetal effects on bone mineral density. To the best of our
demise [26]. knowledge, no patient suffered an osteoporotic fracture
The introduction of LMWH has lead to diminishing use during study follow up.
of UFH in many clinical settings, including pregnancy
[12]. Potential benefits of LMWH include predictable
anticoagulant response from weight based dosing, ease of Conclusions
outpatient use, lower risk of osteopenia, and lower risk of
HIT. Outpatient management of acute DVT with unad- Analysis of a group of anticoagulated patients found that
justed, weight-based UFH has been demonstrated to be most patients received UFH as their primary AC therapy
feasible with efficacy similar to LMWH [13]. Our study during pregnancy. Maternal thromboembolism and fetal
demonstrated outpatient UFH administration is also feasi- demise were infrequent in this population. Our findings
ble during pregnancy. Osteoporosis is a concern as greater indicate that UFH remains a viable treatment option among
osteoblast inhibition has been demonstrated in vivo with more expensive LMWH products.
UFH [27]. However, clinical trials evaluating bone mineral
density in pregnant women exposed to either LWMH or Acknowledgements We would like to acknowledge the KPCO
Pharmacy Dept. and its Executive Director, Dennis Helling, PharmD,
UFH are equivocal and bone appears to remineralize after for their support of this study. This study was funded entirely by
delivery [9, 28–33]. HIT is very rare during pregnancy and Kaiser Permanente Colorado. This study partially was undertaken in
recent meta-analysis of VTE treatment found similar rates fulfillment of the requirements for Suzanne Parker’s American
of heparin-associated thrombocytopenia for UFH and Society of Health-Systems Pharmacy residency program.
LMWH treated patients, and concluded the rate of HIT was
too low to make an adequate comparison between groups
[34]. References
There are potential benefits to using UFH during preg-
nancy. A shorter half-life allows for more rapid clearance 1. Rey E, Kahn S, David M, Shrier I (2003) Thrombophilic disor-
ders and fetal loss: a meta-analysis. Lancet 361:901–908
in the setting abrupt onset of labor [35]. Protamine is
2. Kujovich J (2004) Thrombophilia and pregnancy complications.
available as a reversal agent in the setting of hemorrhage or Am J Obstet Gynecol 191:412–424
overdose [36, 37]. In addition, the anti-Xa assay reliably 3. Ginsberg J, Hirsh J, Turner C, Levine MN, Burrows R (1989)
measures UFH activity [11]. Great debate exists regarding Risks to the fetus of anticoagulant therapy during pregnancy.
Thromb Haemost 61:197–203
how to properly administer and monitor LWMH in popu- 4. Hall J, Paul R, Wilson K (1980) Maternal and fetal sequelae of
lations where the pharmacokinetics of LMWH are altered anticoagulation during pregnancy. Am J Med 68:122–140
[38]. Recent consensus guidelines have recommended 5. Wesseling J, Van Driel D, Heymans H, Rosendaal FR, Geven-
against the routine use of anti-Xa monitoring for LMWH Boere LM, Smrkovsky M, Touwen BC, Sauer PJ, Van der Veer E
(2001) Coumarins during pregnancy: long term effects on growth
which highlights the uncertainty regarding the relevance
and development in school age children. Thromb Haemost
and accuracy of this test [38–40]. 85:609–613
Our study is not without important limitations. The 6. Jacobsen A, Qvigstad E, Sandset P (2003) Low molecular weight
small number of patients did not allow for inferential heparin (dalteparin) for the treatment of venous thromboembo-
lism in pregnancy. BJOG 110:139–144
comparisons of the primary outcome between individual
7. Hunt B, Gatten M, Khamashta M, Nelson-Piercy C, Almeida A
indications or treatment strategies. Nevertheless, we were (2003) Thromboprophylaxis with unmonitored intermediate-dose
able to provide in-depth descriptions of various patient sub- low molecular weight heparin in pregnancies with a previous
groups. The observational study design required extracting arterial or venous thrombotic event. Blood Coagul Fibrinolysis
14:735–739
historical medical information from patient charts with 8. Ulander V, Lehtola A, Kaaja R (2003) Long-term outcome of
potentially incomplete and inaccurate documentation. Use deep venous thrombosis during pregnancy treated with unfrac-
of integrated, electronic records and a standardized chart tionated heparin or low molecular weight heparin. Thromb Res
abstraction form increased the likelihood of a thorough 111:239–242
9. Nelson-Piercy C, Letsky E, de Swiet M (1997) Low-molecular-
review of relevant information to identify and categorize
weight heparin for obstetric thromboprophylaxis: experience of
patients, exposures, and outcomes. The setting of this study sixty-nine pregnancies in sixty-one women at risk. Am J Obstet
in a centralized anticoagulation service within an integrated Gynecol 176:1062–1068
healthcare delivery system may limit the study’s general- 10. Blanco-Molina A, Trujillo-Santos J, Criado J (2007) Venous
thromboembolism during pregnancy or postpartum: findings from
izability. However, since patients were drawn from an
the RIETE registry. Thromb Haemost 97:186–190
anticoagulated population of over 7,000 patients in this 11. Bates S, Greer I, Hirsh J, Ginsberg J (2004) Use of antithrombotic
population-based study, there is a strong probability that agents during pregnancy. The Seventh ACCP Conference on

123
A descriptive evaluation of unfractionated heparin use during pregnancy 273

Antithrombotic and Thrombolytic Therapy. CHEST 126:627S– 26. Folkeringa N, Brouwer JL, Kortweg F, Veeger NJ, Erwich JJ,
644S Holm JP, van der Meer J (2007) Reduction of high fetal loss rate
12. Copplestone J, Pavord S, Hunt B (2004) Anticoagulation in preg- by anticoagulant treatment in antithrombin, protein C or protein S
nancy: a survey of current practice. Br J Haemotol 124:123–125 deficient women. Br J Heamatol 136:656–661
13. Kearon C, Ginsberg J, Julian J, Douketis J, Solymoss S, Ockel- 27. Matziolis G, Perka C, Disch A, Zippel H (2003) Effects of
ford P, Jackson S, Turpie AG, Mackinnon B, Hirsh J, Gent M fondaparinux compared with dalteparin, enoxaparin and unfrac-
(2006) Comparison of fixed-dose weight-adjusted unfractionated tionated heparin on human osteoblasts. Calcif Tissue Int 73:370–
heparin and low-molecular-weight heparin for acute treatment of 379
venous thromboembolism. JAMA 296:935–942 28. Douketis J, Ginsberg J, Burrows R, Duku EK, Webber CE, Brill-
14. Lockwood C (2002) Inherited thrombophilias in pregnant patients: Edwards P (1996) The effects of long-term heparin therapy
detection and treatment paradigm. Obstet Gynecol 99:333–341 during pregnancy on bone density. Thromb Haemost 75:254–257
15. Brill-Edwards P, Ginsberg J, Gent M, Hirsh J, Burrows R, Kearon 29. Backos M, Rai R, Thomas E, Murphy M, Dore C, Regan L
C, Geerts W, Kovacs M, Weitz J, Robinson KS, Whittom R, (1999) Bone density changes in pregnant women treated with
Couture G (2000) Safety of withholding heparin in pregnant heparin: a prospective, longitudinal study. Hum Reprod 14:2876–
women with a history of venous thromboembolism. NEJM 2880
343:1439–1444 30. Casele H, Haney E, James A, Rosene-Montella K, Carson M
16. Barbour L, Oja J, Schultz L (2004) A prospective trial that (2006) Bone density changes in women who receive thrombo-
demonstrates that dalteparin requirements increase in pregnancy prophylaxis in pregnancy. Am J Obstet Gynecol 195:1109–1113
to maintain therapeutic levels of anticoagulation. Am J Obstet 31. Pettila V, Leinonen P, Markkola A, Hillesmaa V, Kaaja R (2002)
Gynecol 191:1024–1029 Postpartum bone mineral density in women treated for throm-
17. De Carolis S, Ferrazzani S, De Stefano V, Garofalo S, Fatigante boprophylaxis with unfractionated heparin or LMW heparin.
G, Rossi E, Leone G, Caruso A (2006) Inherited thrombophilia: Thromb Haemost 87:182–186
treatment during pregnancy. Fetal Diagn Ther 21:281–286 32. Dahlman T, Sjoberg H, Ringertz H (1994) Bone mineral density
18. Kupferminc M, Fait G, Many A, Lessing JB, Yair D, Bar-Am A, during long-term prophylaxis with heparin in pregnancy. Am J
Eldor A (2001) Low-molecular weight heparin for the prevention Obstet Gynecol 170:1315–1320
of obstetric complications in women with thrombophilia. Hy- 33. Barbour L, Kick S, Steiner J, LoVerde M, Heddleston L, Lear J,
pertens Pregnancy 20:34–44 Baron A, Barton P (1994) A prospective study of heparin-induced
19. Triolo G, Ferrante A, Ciccia F, Accardo-Palumbo A, Perino A, osteoporosis in pregnancy using bone densitometry. Am J Obstet
Castelli A, Giarratano A, Licata G (2003) Randomized study of Gynecol 170:862–869
subcutaneous low molecular weight heparin plus aspirin versus 34. Morris T, Castrejon S, Devendra G, Gamst A (2007) No differ-
intravenous immunoglobulin in the treatment of recurrent fetal ence in risk for thrombocytopenia during treatment of pulmonary
loss associated with antiphospholipid antibodies. Arthritis Rheum embolism and deep venous throbosis with either low-molecular-
48:728–731 weight heparin or unfractionated heparin. A meta-analysis.
20. Gris J, Mercier E, Quere I, Lavigne-Lissalde G, Cochery-Nou- CHEST 132:1131–1139
vellon E, Hoffet M, Ripart-Niveu S, Tailland M, Dauzat M, 35. Handeland GF, Abidgaard GF, Holm U, Arneses KE (1990) Dose
Mares P (2004) Low-molecular-weight heparin versus low-dose adjusted heparin treatment of deep venous thrombosis: a com-
aspirin in women with one fetal loss and a constitutional parison of unfractionated and low molecular weight heparin. Eur
thrombophilic disorder. Blood 103:3695–3699 J Clin Pharmacol 39:107–112
21. Rai R, Cohen H, Dave M, Regan L (1997) Randomised con- 36. Wolzt M, Weltermann A, Nieszpauer-Los M, Schneider B,
trolled trial of aspirin and aspirin plus heparin in pregnant women Fassolt A, Lechner K, Eichler HG, Kyrle P (1995) Studies on the
with recurrent miscarriage associated with phospholipid anti- neutralizing effects of protamine on unfractionated and low
bodies (or antiphospholipid antibodies). BMJ 314:253–257 molecular weight heparin (Fragmin) at the site of activation of the
22. Grandone E, Brancaccio V, Colaizzo D, Scianname N, Pavone G, coagulation system in man. Thromb Haemost 73:439–443
Di Minno G, Margaglione M (2002) Preventing adverse obstetric 37. Beng CJ. Berstad A. Talstad I (1991) Incomplete reversal of
outcomes in women with genetic thrombophilia. Fertil Steril enoxaparin-induced bleeding by protamine sulfate. Haemostasis
78:371–375 21:155–160
23. Brenner B, Hoffman R, Blumenfeld Z, Weiner Z, Younis J (2000) 38. Shojania A (2004) More one: is laboratory monitoring of low-
Gestational outcome in thrombophilic women with recurrent molecular-weight heparin necessary? Thromb Haemost 2:2276–
pregnancy loss treated by enoxaparin. Thromb Haemost 83:693– 2277
697 39. Buller H, Agnelli G, Hull R, Hyers TM, Prins H, Raskob GE
24. Backos M, Rai R, Baxter N, Chilcott IT, Cohen H, Regan L (2004) Antithrombotic therapy for venous thromboembolic dis-
(1999) Pregnancy complications in women with recurrent mis- ease. The seventh ACCP conference on antithrombotic and
carriage associated with antiphospholipid antibodies treated with thrombolytic therapy. CHEST 126:401S–428S
low dose aspirin and heparin. Br J Obstet Gynaecol 106:102–107 40. Kitchen S, Iampietro R, Woolley A, Preston F (1999) Anti Xa
25. Farquharson R, Quenby S, Greaves M (2002) Antiphospholipid monitoring during treatment with low-molecular-weight heparin
syndrome in pregnancy: a randomized, controlled trial of treat- or danaparoid: inter-assay variability. Thromb Haemost 82:1289–
ment. Obstet Gynecol 100:408–413 1293

123