SEARCH

SEARCH BY CITATION

Keywords:

  • venous thromboembolism;
  • pregnancy complication;
  • heparin;
  • low-molecular-weight;
  • post-partum haemorrhage;
  • pregnancy outcome

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References

Background

North American and European literature suggest that the incidence rate for pregnancy-related thromboembolism (VTE) ranges from 0.5 to 2 per 1000 pregnancies. However, there is a paucity of data regarding pregnancy-related VTE in Australia and New Zealand.

Aims

To define the epidemiology, management and adverse effects of pregnancy-related VTE in Australia and New Zealand.

Method

Retrospective chart review of pregnant patients with objectively diagnosed pregnancy-related VTE at Monash Medical Centre and the North Shore Hospital from January 2007 to March 2011.

Results

Sixty women with VTE were identified, 31 and 29 in the antepartum and post-partum period respectively. VTE occurred as early as 8 weeks of gestation. There was a trend towards higher proportion of PE in the postpartum period. Most antenatal patients were started on enoxaparin and dosed according to weight at diagnosis. A wide variability in maintenance dosing strategies was observed. Three (5%, 95% CI: 1% to 14%) patients suffered major bleeds, all occurring post-partum. Recurrences occurred in two post-partum patients who received a truncated course of enoxaparin for distal deep-vein thrombosis. Although more women had an induction of labour, this did not translate into an increased Caesarean section rate.

Conclusion

The epidemiology of pregnancy-related VTE is similar to that of other developed countries. All three bleeding events occurred in the immediate post-partum setting, highlighting the need for caution at this critical time. VTE recurrences occurred in those women with post-partum distal deep-vein thrombosis treated with an abbreviated course of enoxaparin.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References

When a woman falls pregnant, her risk of venous thromboembolism (VTE) increases by twofold to fivefold compared with the non-pregnant population of the same age with an overall absolute incidence of 0.5 to 2 per 1000 pregnancies.[1] Her risk of fatal pulmonary embolism (PE) is 0.79 to 1.94 per 100 000 pregnancies[2, 3]

Pregnancy-related VTE is one of the major causes of preventable maternal death in the developed world, ranking ahead of obstetric haemorrhage. The latest Confidential Enquiries into Maternal Deaths in the United Kingdom (CMACE 2011) showed a decline in maternal death, mainly due to a reduction in VTE death, and emphasised the importance of improved education and better implementation of thromboprophylaxis in pregnancy.

While there appears to be better recognition and understanding of the problem, there are still many areas of uncertainty. There is a paucity of evidence regarding the appropriate maintenance dose, monitoring and peri-delivery management of anticoagulants in pregnancy. Many of the recommendations are based on extrapolations from data in the non-pregnant population or from observational case-series.[4] Furthermore, an information void exists in Australasia. This retrospective audit aims to improve knowledge about local epidemiology, management and clinical outcome of pregnancy-related VTE.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References

A retrospective review of the practice of diagnosing and managing VTE in the pregnant population was performed at two major hospitals in Australia and New Zealand.

Monash Medical Centre provides health services to a population of approximately 750 000 people and conducts 8000 deliveries per year. Similarly, the North Shore Hospital (Waitemata District Health Board (WDHB)) is the largest district health board in New Zealand, with a catchment area of 550 000 people, and conducts approximately 6800 deliveries per year.

Patient selection

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References

At MMC, a preliminary list was generated from a broad search using International Classification of Diseases, 10th revision (ICD-10) codes: O08.7, O22.3, O22.3, O22.5, O22.9, O87.0 O87.1, O87.3, O87.9 and O88.2. It is known that the use of specific pregnancy ICD codes has low positive predictive value in identifying our target population.[5] Case notes were subsequently retrieved and searched to identify patients. At North Shore Hospital, patients were identified using a VTE database of consecutive VTE patients referred to this hospital and is likely to capture most events as most patients in this geographic area were referred there in order to access subsidised low-molecular-weight heparin (LMWH) therapy.

Patients with an objectively confirmed diagnosis of PE or deep-vein thrombosis (DVT) (calf and/or proximal DVT) during the period of January 2007 to March 2011 were included for review. Patients were excluded if there was a diagnosis of superficial thrombophlebitis, or if diagnostic imaging were normal for VTE in suspected cases.

We constructed a standardised case report form to collect variables and outcomes that were defined a priori.

Definitions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References

Trimesters were defined as follows:

  • First trimester: Weeks 1 to 13
  • Second trimester: Weeks 14 to 28
  • Third trimester: Weeks 29 to delivery
  • Post-partum period: 6-week period following delivery.

Bleeding was classified as major according to the International Society on Thrombosis and Haemostasis (ISTH) definition, clinically relevant non-major and minor.[6, 7] In the peri-delivery period, post-partum haemorrhage (PPH) was considered as a major bleed and was defined as 500 mL or more of blood loss after vaginal delivery or 750 mL or more after a Caesarean delivery.[8]

Study end-points

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References
  • Characteristics of women with pregnancy-related VTE
  • Incidence, chronological distribution and location of VTE in pregnancy
  • Incidence of thrombophilia
  • Use of imaging modalities
  • Anticoagulation choice, dosing strategy and anti-Xa monitoring utility
  • Methods of delivery and time off anticoagulation
  • Maternal bleeding complications and recurrence rate
  • Miscarriage, stillbirth, birthweight and Apgar score as a surrogate for neonatal outcome.

Statistical analysis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References

Descriptive statistics summarised baseline variables. Differences in frequencies/proportions between groups were compared using Fisher's exact or Chi-squared test (two-sided test, P < 0.05 for significance). The 95% confidence interval (CI) for proportion was calculated by the modified Wald's method.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References

Study population

Sixty pregnancy-related VTE were identified in the period of January 2007 to March 2011 (Fig. 1), giving an approximate VTE incidence rate ranging from 60 per 100 000 deliveries at WDHB to 130 per 100 000 at MMC. The antepartum VTE rates at both hospitals were similar (40 to 59 per 100 000 deliveries).

figure

Figure 1. Selection of patients. A total of 60 patients included, 42 patients' cases were excluded. DVT, deep-vein thrombosis; MMC, Monash Medical Centre; PE, pulmonary embolism; WDHB, Waitemata District Health Board.

Download figure to PowerPoint

Patient characteristics

Details of patient characteristics and ethnicity data are summarised in Table 1. The mean maternal age at diagnosis was 31.5 years with a mean weight of 76.6 kg. Six (10%) women weighed more than 100 kg. Fourteen (23%, CI: 14% to 36%) patients were of non-white ethnicity.

Table 1. Baseline patient characteristics
CharacteristicsNo. patients
  1. a

    Data available for 54/60 women only. Hx, history; SD, standard deviation; VTE, venous thromboembolism.

Mean age (SD) (years)31.5 (5.2)
Mean weight at diagnosis (SD) (kg)76.6 (15.3)a
Primigravida19 (32%)
Parity >47 (12%)
Thrombophilia12 (22%)
Past Hx of VTE3 (5%)
Ethnicity 
Anglo-Celtic40
East European – Croatian1
Southern European 
Greek24
Italian1
Spanish1
Asian 
Indian59
Filipino2
Vietnamese1
Cambodian1
Pacific islander 
Cook Island12
Samoan1
West Asian and North African 
Egyptian23
Turkish1
Not documented1

Chronology of VTE presentation

There were 31 antepartum VTE and 29 post-partum events. Overall, the rate of VTE was not statistically different across the three trimesters with 14, 7 and 10 VTE in the first, second and third trimester respectively (Fig. 2a). Of interest, the frequency of postpartum PE was twice that of the antenatal period (Fig. 2b). Three of the women with first trimester VTE had past history of unprovoked VTE and were not yet on recommended thromboprophylaxis by the time of the event.

figure

Figure 2. (a) Venous thromboembolism (VTE) distribution during pregnancy. (image) DVT, (image) PE, (image) PE + DVT. (b) (image) PE and (image) DVT distribution during pregnancy. DVT, deep-vein thrombosis; PE, pulmonary embolism.

Download figure to PowerPoint

Location of DVT

Forty-four DVT were observed, including 29 (66%, CI: 51% to 78%) involving the left leg and 12 (27%, CI: 16% to 42%) involving the right leg. Three (7%, CI: 2% to 19%) DVT were bilateral. This left leg predisposition becomes more apparent when proximal iliofemoral site is considered (Fig. 3a). Iliofemoral DVT involved the left leg more commonly as compared with non-iliofemoral DVT (88% vs 56%, P = 0.048). On the other hand, distal DVTs (12 (27%), CI: 16% to 42%) were more common post-partum (three antepartum vs nine post-partum, P = 0.02) without left leg predominance (Fig. 3b).

figure

Figure 3. (a) Right and left distribution of iliofemoral versus non-iliofemoral deep-vein thrombosis (DVT). (image) Right DVT, (image) Left DVT. (b) Distal DVT distribution in antenatal and post-partum period. (image) Proximal DVT, (image) Distal DVT.

Download figure to PowerPoint

Thrombophilia

Thirteen (22%) patients had documented thrombophilia as detailed in Table 2. However, the actual number tested is unknown. Four patients showed the presence of lupus anticoagulant or anticardiolipin antibody.

Table 2. Frequency of thrombophilia
ThrombophiliaNo. of patients
  1. ACA, anti-cardiolipin antibodies; LA, lupus anticoagulant.

FV Leiden 
Heterozygous67
Homozygous1
Prothrombin gene – heterozygous1
Activated Protein C resistance1
Antiphospholipid antibodies 
LA24
High-titre ACA1
LA + ACA1
No documentation47
Diagnostic imaging

Doppler ultrasound was the primary investigation used to diagnose DVT. Ten (50%) patients were diagnosed with PE by single imaging modality – either ventilation-perfusion scan (V/Q) or computed tomography pulmonary angiography (CTPA). Four (20%) non-diagnostic V/Q scans resulted in further investigation, including CTPA or ultrasound to confirm PE or DVT (Table 3).

Table 3. Use of diagnostic modalities in venous thromboembolism diagnosis
Method(s) of diagnosisNo. of patients
  1. CTPA, computed tomography pulmonary angiography; U/S, whole leg ultrasound; V/Q, ventilation-perfusion scan.

U/S40
Single imaging 
CTPA or V/Q10
>1 imaging modality 
U/S, V/Q6
V/Q, CTPA2
U/S, V/Q, CTPA2

Management of VTE

Anticoagulant choice and dose

All patients were treated initially with either enoxaparin, dalteparin or intravenous (IV) unfractionated heparin (UFH) at diagnosis. Forty-six (77%) patients received twice-daily weight-adjusted enoxaparin, whereas only nine (15%) patients received daily dose of enoxaparin at diagnosis. Six of the latter experienced VTE in the post-partum setting. The remaining three were diagnosed with VTE in the first trimester, and subsequently, this daily enoxaparin dose was changed to twice daily when first reviewed in the haematology clinic. Post-partum patients with PE or proximal DVT were subsequently started and maintained on warfarin for at least 3 months.

Four patients were treated with UFH at diagnosis. One patient received UFH in the setting of foetal surgery for amniotic bands, and two patients presented with PE late in the third trimester and delivered soon afterward. No risk factors were identified in the last patient who suffered a post-partum PE and who received only one bolus dose of IV heparin in the emergency department prior to being subsequently discharged on daily enoxaparin and warfarin.

Antenatal enoxaparin dose adjustment

Initial dosing was based on the weight at diagnosis in all patients and maintained throughout the antenatal period except in nine (29%) patients who had at least a 25% dose reduction in their maintenance dose at a median of 9 weeks as illustrated in Table 4. No VTE recurrence was observed in those nine patients within 3 months of anticoagulant cessation.

Table 4. Antenatal patients with at least 25% reduction in maintenance enoxaparin dose
TrimesterVTE descriptionEnoxaparin (mg/kg)DoseDays till first dose changeMaintenance dose
  1. bd, twice a day; DVT, deep-vein thrombosis; L, left; PE, pulmonary embolism; R, right; VTE, venous thromboembolism.

1L iliofemoral DVT150 mg bd5280 mg daily
1L iliofemoral DVT180 mg bd124100 mg daily
1PE1.5100 mg daily8480 mg then 40 mg
1L distal DVT160 mg bd1450 mg then 40 mg
1L distal DVT1.5120 mg daily480 mg daily
1R proximal DVT1.5140 mg daily12780 mg daily
2PE170 mg bd6980 mg daily
3L iliofemoral DVT180 mg bd6360 mg daily
3L proximal DVT170 mg bd5680 mg daily
Monitoring with anti-Xa

Out of 31 antenatal patients, 19 patients underwent laboratory testing for anti-Xa levels (median anti-Xa testing = 2, range = 1–7). Overall, 31 anti-Xa levels were performed with 14, 7 and 10 in the first, second and third trimester respectively with none resulting in dose alteration. Only 16 anti-Xa results in 12 patients were available for review with a range of 0.49 to 1.19 U/mL.

Inferior vena cava (IVC) filter and thrombolysis use

Two women underwent IVC filter insertion after presenting with PEs late in the third trimester, at 37 and 38 weeks respectively. Both women delivered soon after and had uncomplicated filter removals within 6 weeks of insertion. No pregnant women received thrombolysis

Use of class II compression stocking to prevent post-thrombotic syndrome (PTS)

The use of a class II compression stocking was documented in 22 (50%) patients with DVT. Clinical features, severity of the post-thrombotic syndrome, compliance and length of treatment were poorly documented.

Delivery management of antenatal VTE patients

Delivery mode

Complete obstetric data were available for 25 of the 31 antenatal patients. Sixteen patients were electively induced. Fourteen patients underwent normal vaginal delivery and two with instrumentation. Nine (30%) patients underwent Caesarean section with five of these deliveries performed in an emergency setting.

Epidural use

Epidural use was planned in 12 patients. However, two patients did not receive epidural analgesia as a result of recent anticoagulant use in the context of a quick onset of labour at induction and a spontaneous vaginal labour respectively.

Enoxaparin was ceased a median of 28 h prior to the epidural needle insertion with a range of 20.5 to 54 h (interquartile range (IQR): 22 to 43.5 h, n = 9 available for analysis). The two patients receiving UFH had their infusion ceased approximately 6 h prior to epidural insertion. No epidural haematoma was observed.

Peri-delivery enoxaparin management

Timing and adjustment of the enoxaparin dose were poorly documented. The last dose of enoxaparin was reduced to an intermediate dose in eight patients. In general, enoxaparin was ceased a median of 24 h with a range of 8 to 96 h (IQR: 24 to 38.5 h, n = 21 available for analysis) prior to delivery and restarted at a median of 12 h with a range of 6 to 48 h (IQR: 8 to 17 h, n = 20 available for analysis) post-delivery. Three patients had a prolonged labour and experienced a long time off anticoagulation. The restarting dose was prophylactic in 12 women who were all diagnosed with VTE in either the first or early second trimester.

Outcome

Maternal and foetal/neonatal outcomes are summarised in Table 5.

Table 5. Maternal and foetal/neonatal outcome
 Antenatal VTE group (n = 31)95% CIPost-partum VTE group (n = 29)95% CI
  1. CI, confidence interval; IQR, interquartile range; N/A, not applicable; VTE, venous thromboembolism.

Maternal outcome    
VTE recurrence00% to 13%2 (7%)1% to 23%
Bleeding3 (10%)3% to 26%3 (10%)3% to 27%
Major bleed2 (6%)1% to 22%1 (3%)0% to 29%
Clinically relevant non-major1 (3%)0% to 18%1 (3%)0% to 29%
Minor bleed00% to 13%1 (3%)0% to 29%
Median length of stay/days (IQR)6 (4–8)N/AN/AN/A
Foetal/neonatal outcome    
Median birthweight/g (IQR)3140 g (2793–3665 g)N/A2635 g (1087–3421 g)N/A
Low birthweight (<2500g)3N/A1N/A
Very low birthweight (<1500g)1N/A5N/A
Apgar score (<7 at 1 minute)3N/A6N/A
Foetal death/stillbirth1 (3%)0% to 18%2 (7%)1% to 23%
Overall maternal outcome

There were no antepartum recurrences of VTE (0/31, CI: 0% to 13%). In contrast, there were two recurrences in the post-partum period (2/29 (7%), CI: 1% to 23%); both women were diagnosed with distal DVT and received only 2 weeks of anticoagulant therapy.

Overall, there were six bleeding events, including three (5%, CI: 1% to 14%) major bleeding events, all occurring in the post-partum period (Table 6). Three bleeding events, two major and one minor, occurred in women diagnosed with antenatal VTE. The first was a primary PPH (600 mL) in a woman who was haemodynamically stable and observed without the need of surgical intervention or blood products. Another woman experienced a large rectus sheath haematoma at the Caesarean section site, requiring surgical evacuation and also required blood product support for a primary PPH (800 mL). One minor vaginal (PV) bleed occurred in a woman who was simply observed without the need for intervention.

Table 6. Description of bleeding events
VTE diagnosisTime of diagnosisDelivery modeBleeding severityTiming of bleedingBleed detailTransfusion
  1. CS, Caesarean section; ED, emergency department; LMWH, low-molecular-weight heparin; PPH, post-partum haemorrhage; PV, per vaginam; VKA, vitamin K antagonist; VTE, venous thromboembolism.

Ante31 week gestationCSMajorPost-partum, day 2 full LMWHRectus sheath haematoma requiring evacuationYes
Ante37 week gestationNormal Vaginal1 PPHDeliveryPPH = 600 mLNo
Ante8 week gestationNormal VaginalMinorAntenatal, week 9 gestationPV spotting ED attendanceNo
PostDay 10 post-partumNormal VaginalMajorPost-partum day 3 full LMWHPV bleed = 4 LYes
PostDay 7 post-partumNormal VaginalClinically relevant non-majorPost-partum, day 2 full LMWHVulval bleeding + sutureNo
PostDay 8 post-partumNormal VaginalMinorPost-partum, week1 post-LMWH/VKAPV spotting ED attendanceNo

In the post-partum VTE group, there was one major bleed, one clinically relevant non-major bleed and one minor bleed. The first woman had severe secondary PPH requiring massive transfusion (4 L) 2 weeks after vaginal delivery. The second had prolonged bleeding post-vaginal delivery from an episiotomy site, lengthening her hospitalisation. The third woman had mild PV bleeding. The latter two women did not require blood products.

All major bleeds occurred in the post-partum period between day 1 to 3 following resumption of full-dose enoxaparin in the setting of recent VTE diagnoses.

Foetal/neonatal outcome of antenatal group

One spontaneous miscarriage was documented at week 8 of pregnancy in a woman who suffered from a distal DVT 2 weeks prior. Two preterm deliveries occurred at week 30 and 32. The median birthweight of the antenatal group was 3060 g, with four neonates having low to very low birthweight. Three neonates had a low Apgar score (<7) at 1 minute. However, only one required transfer to the neonatal intensive care unit after birth.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References

The epidemiological findings in this retrospective pregnancy-related VTE audit are consistent with those of Europe and North America.[9-12] First, the antenatal VTE rate at both centres appears to be similar and consistent with the range of 50–200 per 100 000 pregnancies reported in the literature. Second, the 50% antenatal and 50% post-partum VTE distribution, and the previously well-described left to right predominance of iliofemoral DVT were again observed.[13, 14]

Our data also confirm the clinical impression that distal DVT is uncommon in the antepartum period. However, distal DVT in the post-partum period was not infrequent. This partly reflects the Australasian practice of whole-leg ultrasound scanning with routine examination of the calf veins. It also suggests that once the baby is delivered, the pathogenesis of post-partum DVT is more akin to the non-pregnant state, given that the compression effects of the gravid uterus on the left iliac vein quickly abates. However, consistent with the observation that the post-partum period is still a thrombotic state[15, 16] was the unexpected observation of two episodes of DVT recurrence, both in women with distal DVT who had only 2 weeks of LMWH therapy – a practice commonly used for non-pregnant patients in Australia and New Zealand.

Pregnancy-related VTE recurrence can occur at any time during pregnancy, and therefore, if prophylaxis is advocated, it should begin as soon as a woman falls pregnant. Three patients in our study exemplified this; each had a prior history of VTE, but had not yet started prophylaxis. All suffered their recurrence as early as the 8th to 11th weeks of gestation. This further reinforces the importance of starting prophylaxis early when it is indicated.[17]

In contrast to antenatal VTE, about half of the VTE in the post-partum period were PE, highlighting two important issues in the post-partum period. First, VTE in the post-partum period has the potential for higher morbidity and mortality given the higher proportion of PE, and secondly, the post-partum period is the period of highest VTE risk per day in pregnancy, estimated to be increased by 15- to 30-fold when compared with age-matched non-pregnant control.[18]

Ethnicity

VTE has long been regarded as a disease that affects predominantly Caucasians. It is becoming clearer that VTE also affects the non-Caucasian population and deserves greater attention.[19-21] Our retrospective review showed that a significant proportion of non-Caucasian pregnant women (22%) were affected by VTE and highlighted the importance of applying current thromboprophylaxis guidelines to all pregnant women.

Diagnosis and medical imaging

V/Q scan was the most common first-line modality used for the diagnosis of PE and was only non-diagnostic with the requirement of a CTPA in approximately 20% of cases, which is similar to the reported 25% non-diagnostic rate of V/Q scan.[22] This reinforces that V/Q is a suitable investigation in the younger pregnant population, who generally have normal lungs and in whom the use of V/Q scan minimises radiation exposure to the proliferating breast tissue and thyroid.

Anticoagulant use

The LMWH, enoxaparin, was used in the majority (93%) of pregnancy-related VTE cases in accordance to local and international guidelines,[4] and was dosed according to the weight at diagnosis. In fact, most women in our study were maintained on this dose with no episodes on recurrence. Anti-Xa monitoring was of low clinical utility in our cohort. Almost 2/3 of woman had anti-Xa levels measured, but interestingly, none of the results led to LMWH dose modification, as the results were within the suggested therapeutic range.

Nine antepartum patients receiving an initial twice-daily enoxaparin were changed to a reduced (at least 25%) once-daily maintenance dose at a median of 9 weeks from diagnosis. None of these woman suffered VTE recurrences. This dose reduction strategy is both effective and safe in cancer-related VTE and challenges the presumption that full anticoagulant doses of LMWH are required for the duration of pregnancy.[23]

Peri-delivery management in antenatal women

The rate of induction in this cohort is higher than the Victorian rate in 2008 (64% vs 24.3%), suggesting a preference of physicians/obstetricians for a controlled environment in the setting of anticoagulation use.[24] On the other hand, this did not translate to a higher rate of Caesarean section when compared with local data in Victoria (36% vs 30.4%). Moreover, anticoagulation did not appear to deter epidural/spinal analgesia use when compared with its use in the Victorian pregnant population (32% vs 26.8%).

The median time from enoxaparin cessation to delivery was 27 h. Of concern was the interval between cessation of therapy and its recommencement in the post-partum period – some women were off therapy for up to 72 h. The alternatives to avoid this, such as the use of either prophylactic subcutaneous or intravenous infusion of heparin (ceasing 6 h prior to anticipated delivery and reversible with protamine), appear to be poorly adopted.

We observed three major bleeds (3/60 (5.0%), CI: 1.2% to 14.3%) in our cohort, which is significant but not inconsistent with the bleeding rate (2/105 (1.9%), CI: 1.0% to 7.1%) reported in the literature.[25] Importantly, all three major bleeding events occurred in the immediate post-partum setting.

Limitations of study

The major limitation of retrospective chart reviews is missing documentation. There is a possibility that some post-partum patients at WDHB were missed, given that these patients could have sought medical attention elsewhere. Moreover, data were obtained from a relatively small group of patients (n = 60), implying a lack of precision in estimating the frequency of rare events, such as major bleeding and recurrence. We used standard data collection forms and agreed definitions for all of the study outcomes in order to improve the quality and reliability of the data gathered. Overall, our results appear to be consistent with the literature.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References

The incidence of pregnancy-related VTE in Australia and New Zealand appears similar to other developed countries. While our data confirm the efficacy of LMWH in pregnancy-related VTE, the peri-delivery major bleeding events highlight the need for caution and vigilance at this specific period. A planned induction for labour delivery did not appear to increase the need for Caesarean section. This approach may explain the similar rate of neuraxial analgesia use in our cohort when compared with the local non-VTE pregnant women. A unique observation was the higher rate of post-partum distal DVT and the two observed recurrences in such patients when treated with only 2 weeks of enoxaparin.

Prospective studies are required to clarify uncertainties, especially in regards to appropriate LMWH dosing, monitoring and whether full therapeutic dose is required for the entire length of pregnancy.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References

We thank Ms Corinne Ng, Haematology Research Unit, Monash Medical Centre, for data entry and Ms Jeanette Gamgee, Haematology Research Unit, Monash Medical Centre, for proof reading.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Patient selection
  6. Definitions
  7. Study end-points
  8. Statistical analysis
  9. Results
  10. Discussion
  11. Conclusion
  12. Acknowledgements
  13. References