More on enoxaparin thromboprophylaxis in pregnancy: A review of 10 years' experience from King's College Hospital

Abstract

We read with interest the paper by Cox and colleagues1.Cox S. Eslick R. McLintock C. Effectiveness and safety of thromboprophylaxis with enoxaparin for prevention of pregnancy‐associated venous thromboembolism.J Thromb Haem. 2019; 17: 1160-1170Crossref PubMed Scopus (11) Google Scholar, describing the safety and effectiveness of thromboprophylaxis with enoxaparin during pregnancy at the National Women’s Hospital in New Zealand over a 15‐year period, and the follow‐up letters published by Tardy and colleagues and McLintock and colleagues.2.Tardy B. Buchmuller A. Bistervels I. et al.Thromboprophylaxis in pregnant women: For whom and which LMWH dosage?.J Thromb Haem. 2019; 17: 1401-1403Crossref Scopus (1) Google Scholar, 3.McLintock C. Cox S. Eslick R. Thromboprophylaxis in pregnant women: For whom and which LMWH dosage?.Reply. J Thromb Haem. 2019; 17: 1397-1398Crossref Scopus (1) Google Scholar These publications prompted us to review our own practice and outcomes. King’s College Hospital (KCH), a 900‐bed teaching hospital in South East London, has a busy obstetric unit, with approximately 6000 deliveries per year. At KCH all women requiring antenatal thromboprophylaxis are reviewed and managed in a specialist hematology clinic throughout pregnancy and the postpartum period. We retrospectively reviewed prescribing practice and outcomes of women who received LMWH (low molecular weight heparin) thromboprophylaxis in our clinic between 1 January 2009 and 31 December 2018. Women referred and commenced on LMWH thromboprophylaxis antenatally were identified from clinic letters and the hospital electronic patient record. Documented information from these sources was used to identify the specific risk factors warranting thromboprophylaxis, the dose of enoxaparin prescribed, as well as demographic information. Bleeding, thrombosis, and adverse events were identified and recorded. At KCH, weight‐based enoxaparin thromboprophylaxis is prescribed for both non‐pregnant and pregnant patients, as follows: weight < 50 kg: 20 mg od, 50—100 kg: 40 mg od, >100 kg: 80 mg od, >150 kg: 120 mg od. Additionally, delivery outcome data was obtained for a subset of these women (84 pregnancies) between 2015 and 2018, from the hospital maternity system, BadgerNet v2.9.1 (CleverMed). Descriptive statistics were used to analyze the data, and the project was classified as an audit, reference: KCH‐AUD224. Over the 10‐year period, 409 women had 502 pregnancies; mean age 34 (standard deviation [SD] ± 5.7), mean pre‐pregnancy weight 83.4 kg (SD ± 23.8), mode gravida 2; more than half of women were White (54%) and 36% were Black, reflecting our location in South East London. Women at risk for antenatal venous thromboembolism (VTE) were prescribed thromboprophylaxis in accordance with the Royal College of Obstetricians and Gynaecologists (RCOG) thromboprophylaxis guideline of the time.4.Royal College of Obstetricians and Gynaecologists. Thrombosis and Embolism during Pregnancy and the Puerperium, Reducing the Risk (Green‐top 37). 2009.Google Scholar, 5.Royal College of Obstetricians and GynaecologistsGreen‐top Guideline 37a: Reducing the Risk of Thrombosis and Embolism During Pregnancy and the Puerperium. RCOG, London2015Google Scholar The most prevalent risk factors were: presence of thrombophilia (43%), personal history of VTE (36%), body mass index> 30 kg/m2.Tardy B. Buchmuller A. Bistervels I. et al.Thromboprophylaxis in pregnant women: For whom and which LMWH dosage?.J Thromb Haem. 2019; 17: 1401-1403Crossref Scopus (1) Google Scholar (19%), family history of VTE (13%), age > 35 years (12%), and parity> 3 (11%). Most women (76%) were prescribed enoxaparin 40 mg daily dose in line with their weight. Bleeding was reported in 7.2%, with epistaxis being most common. All bleeding events were classed as minor and were reported at routine hematology clinic follow‐up. A significant number of women (17.3%) reported adverse effects, where bruising at the injection sites was the most common (49% of all adverse effects reported). Twelve (2.4%) women had thromboembolic events with 10 of these women having a prior history of VTE (see Table 1). It is important to note that 4 of these events occurred> 6 weeks postpartum, when thromboprophylaxis had ceased. We have reported all events encountered, even when outside the recognized risk period (during pregnancy and up to 6 weeks postpartum), as research suggests that in a small number of women, the period of risk can persist for up to 12 weeks.6.Kamel H. Navi B.B. Sriram N. et al.Risk of a Thrombotic Event after the 6‐Week Postpartum Period.N Engl J Med. 2014; 370: 1307-1315Crossref PubMed Scopus (3) Google Scholar In our cohort, during the recognized risk period, there were three objectively diagnosed VTE events (0.6%), of which one patient had reported adherence issues. Additionally, there were four possible VTE events, for which the treating clinician initiated treatment dose enoxaparin, based on clinical judgement. Collectively, this gives a VTE event rate of 1.4% in our cohort during the antenatal and 6‐week postnatal period, when thromboprophylaxis was prescribed.TABLE 1Details of women who experienced thromboembolic eventsCaseLMWHRisk factor(s)First reviewed in clinicTrimesterinitiatedEventTime of event1Dalteparin 5000 units OD‐shortage of enoxaparinPrevious pregnancy related left DVT, high BMI, parity ≥ 3Feb 20181st trimester(5 weeks)Proximal DVTPatient presented with 1‐week history of left calf pain and behind the knee. Diagnosed with femoral and popliteal vein thrombosis – treatment dose enoxaparin commenced.8/402Enoxaparin 40 mg ODProvoked DVT (COCP and ankle fracture)May 20171st trimester(9 weeks)Superficial thrombophlebitisMiscarried at 11 weeks and had an evacuation of retained products of conception (ERPC). Patient advised to continue on enoxaparin for 10 days after ERPC. In clinic, patient reported a painful left arm since the ERPC, with pain and tenderness related to her cannulation site. A scan showed thrombus in the left arm cephalic vein, extending from the wrist into the median cubital vein at the level of the antecubital fossa and into the cephalic vein to the mid upper arm level. Patient prescribed a further 4 weeks to complete 6‐week of prophylactic‐dose enoxaparin as per hospital protocol for superficial thrombophlebitis.10 days after ERPC3Enoxaparin 80 mg ODPrevious VTE – provoked by long haul flight, high BMI, parity ≥ 3Jan 20152nd trimester(17 wk)Possible DVT and PEFor 2–3 weeks, she had become increasingly breathless and was having difficulty walking upstairs. Her oxygen saturation on air was 100%, pulse 110 bpm. Although there was no pleuritic chest pain, it was difficult to entirely exclude PE clinically. Duplex ultrasound of the left leg was suspicious of left sided popliteal DVT, although it was difficult to know whether this episode of DVT was new or old. Enoxaparin changed to 120 mg BD for 2 weeks then 180 mg OD.20/404Enoxaparin 40 mg ODPrevious provoked DVT (COCP)Jan 20171st trimester(9 weeks)Transient ischaemic attackMagnetic resonance imaging (MRI) showed small cortical infarcts; Doppler showed 40% occlusive thrombus in left carotid‐‐‐underwent endarterectomy. Enoxaparin dose increased to 80 mg BD and low dose aspirin added.25/405Enoxaparin40 mg ODPE postpartum (2005), factor V Leiden heterozygoteFeb 20112nd trimester(13 weeks)Possible PEAt routine antenatal care follow‐up mentioned she had shortness of breath on exertion. Chest x‐ray was normal; duplex ultrasound of both legs was normal. VQ scan revealed reduced perfusion at the left lower zone, although this abnormality appeared smaller than it was compared to the original VQ scan of 2005. Dose of enoxaparin increased to 120 mg OD.30/406Enoxaparin 40 mg ODAge> 35, high BMI, parity ≥ 3, varicose veinsSep 20153rd trimester(28 weeks)PEAt 31 weeks she developed increased breathlessness, and VQ was suggestive of a focal mismatch in the right lung anteriorly consistent with pulmonary embolism. Enoxaparin increased to 80 mg BD.31/407Enoxaparin60 mg BDPrevious pregnancy‐related PE and DVT, high BMI, family history of VTE, parity ≥ 3May 20102nd trimester(26 weeks)Possible PEPresented to AE with shortness of breath and left thigh pain. Doppler ultra‐sound negative and VQ reported as low probability for PE. Dose of enoxaparin increased to 120 mg BD–symptoms resolved.32/408Enoxaparin 40 mg ODPrevious left DVT and post‐thrombotic syndromeDec 20172nd trimester (15 weeks)Possible recurrent proximal DVTPatient complained of a 3‐day history of left leg pain and swelling; duplex ultrasound scan revealed old thrombus from popliteal to common femoral vein, and no obvious acute DVT; however, it was difficult to fully exclude a small amount of new thrombus.Enoxaparin dose has increased to 120 mg OD with resolution of symptoms.36/409Enoxaparin 80 mg ODProvoked left DVT postpartum, high BMI, parity ≥ 3May 20151st trimester(6 weeks)Proximal DVTAt 33/40 patient underwent an emergency C‐section due to vaginal bleeding.3 weeks later, she then developed a red swollen, painful right leg and was found to have an extensive right proximal DVT. Patient known to have medication adherence issues. She was managed on rivaroxaban (not breastfeeding).3 weeks postpartum10Enoxaparin40 mg ODVenous malformation of the upper inner right thigh. Superficial thrombophlebitis early on during index pregnancyJun 20112nd trimester(13 weeks)Superficial thrombophlebitis.One week after stopping prophylactic enoxaparin she developed recurrent pain in her right thigh with D‐dimer> 8000. Duplex ultrasound revealed partial thrombus at the LSV aneurysm (as previously seen), extending into the LSV proximally to mid‐thigh and to the tortuous veins in the muscle of the posterior thigh at the site of pain but without extension into the femoral vein. Restarted on treatment dose enoxaparin for 6 weeks and surgical options re‐explored.7 weeks postpartum(off enoxaparin)11Enoxaparin 60 mg ODPrevious superficial vein thrombosis and recurrent left DVT, factor V Leiden heterozygoteJun 20111st trimester(4 weeks)Possible recurrent DVTAfter completing 6 weeks thromboprophylaxis following delivery, at follow‐up clinic, patient reports some discomfort around her varicose veins in the left lower leg, and a repeat duplex ultrasound showed likely old proximal DVT. There was also thrombus below the knees which could not be accurately determined as old or new on ultrasound. Enoxaparin 150 mg OD started.7 weeks postpartum(off enoxaparin)12Enoxaparin 80 mg ODProvoked PE (COCP), high BMIJul 20122nd trimester(15 weeks)PEDeveloped allergic (skin) reaction to enoxaparin. Was switched to dalteparin 10 000 IU OD. Completed thromboprophylaxis for 6 weeks postpartum and then stopped. At 3 months postpartum, she was admitted to hospital with history of acute shortness of breath and pleuritic chest pain. Diagnosed (CT pulmonary angiogram) with bilateral pulmonary emboli in lobar, segmental and subsegmental pulmonary arteries.3 months postpartum(off dalteparin)Abbreviations: BD, bis die (twice daily); BMI, body mass index; COCP, combined oral contraceptive pill; CT, computed tomography; DVT, deep vein thrombosis; LMWH, low molecular weight heparin; LSV, lesser saphenous vein; OD, omni die (daily); PE, pulmonary embolism; VQ, ventilation/perfusion. Open table in a new tab Abbreviations: BD, bis die (twice daily); BMI, body mass index; COCP, combined oral contraceptive pill; CT, computed tomography; DVT, deep vein thrombosis; LMWH, low molecular weight heparin; LSV, lesser saphenous vein; OD, omni die (daily); PE, pulmonary embolism; VQ, ventilation/perfusion. For 84 (17%) pregnancies (between 1 January 2015 and 31 December 2018), we were able to retrieve information on the delivery outcomes. Their demographic information mirrored the demographics of the population of the total cohort. The live birth rate was 95.2%. The mean infant weight was 3390 g (SD ± 520), with 51% being male infants. Two thirds of the cohort (67.1%) had vaginal deliveries, with the remainder caesarean sections (17.7% were elective, 7.6% were emergency, 7.6% not clear from the documentation). Two thirds of women (66.6%) had pain relief through neuraxial analgesia. Mean recorded blood loss was 571mL (SD ± 434), with 35.9% having ≥500 mL estimated blood loss, and 14.1% having estimated blood loss of ≥1000 mL. Our experience suggests that although weight‐based thromboprophylaxis with enoxaparin is effective, a number of women will experience bruising at the injection sites and minor bleeding. Although classed as minor according to International Society on Thrombosis and Haemostasis definitions, we agree with Tardy and colleagues2.Tardy B. Buchmuller A. Bistervels I. et al.Thromboprophylaxis in pregnant women: For whom and which LMWH dosage?.J Thromb Haem. 2019; 17: 1401-1403Crossref Scopus (1) Google Scholar that for the women themselves, it is not necessarily a minor issue and it is an important reminder for clinicians, including midwives, to check for these and provide appropriate support. In our cohort, two thirds of women had access to neuraxial analgesia; higher than that reported in the Auckland cohort.1.Cox S. Eslick R. McLintock C. Effectiveness and safety of thromboprophylaxis with enoxaparin for prevention of pregnancy‐associated venous thromboembolism.J Thromb Haem. 2019; 17: 1160-1170Crossref PubMed Scopus (11) Google Scholar In a recently published study from 199 women in Israel, 121 (60.8%) underwent planned induction of labor and 78 (39.2%) were admitted following spontaneous onset of labor. Eighty percent of women were on enoxaparin 40mg daily and the remaining 20% on 60mg daily. The authors of this work report that among those with spontaneous onset of labor, 69 (88.5%) were eligible for neuraxial analgesia.7.Rottenstreich A. Zacks N. Kleinstern G. et al.Planned induction versus spontaneous delivery among women using prophylactic anticoagulation therapy.BJOG. 2020; 127: 1241-1248Crossref PubMed Scopus (4) Google Scholar In the subset of women with delivery outcomes from our cohort, the enoxaparin thromboprophylaxis prescription did not preclude neuraxial analgesia for many women. While we understand the concern that higher doses of enoxaparin3.McLintock C. Cox S. Eslick R. Thromboprophylaxis in pregnant women: For whom and which LMWH dosage?.Reply. J Thromb Haem. 2019; 17: 1397-1398Crossref Scopus (1) Google Scholar may limit access to neuraxial analgesia, we feel these concerns are not justified. LMWH clearance does increase with body weight; provided the dose increase is proportionate to a 40mg enoxaparin dose (for example, as per the KCH protocol), then the 12‐hour rule would still apply. Anecdotally we are aware that many women discontinue thromboprophylaxis at the first signs of labor and await further guidance from the managing obstetric team. In our previous enoxaparin in pregnancy study8.Patel J.P. Auyeung V. Patel R.K. et al.Women’s views on and adherence to enoxaparin during pregnancy and the puerperium.J Thromb Haem. 2012; 10: 2526-2534Crossref PubMed Scopus (27) Google Scholar, women reported that a worry they did have with injecting LMWH during pregnancy was not being able to get access to their preferred mode of analgesia at the time of delivery. Clinicians reviewing women in the weeks leading up to delivery should reassure women prescribed thromboprophylaxis regarding this, and ensure, as much as possible, there is a plan in place. The estimated blood loss reported at the time of delivery in our cohort was similar to that reported by Cox and colleagues.1.Cox S. Eslick R. McLintock C. Effectiveness and safety of thromboprophylaxis with enoxaparin for prevention of pregnancy‐associated venous thromboembolism.J Thromb Haem. 2019; 17: 1160-1170Crossref PubMed Scopus (11) Google Scholar Without a control arm, it is difficult to draw any firm conclusions from our findings and what, if any, role the antenatally injected LMWH had in this. We are unaware of any work which has measured intrapartum anti‐Xa activity in women who have been injecting antenatal thromboprophylaxis and whether a relationship exists with bleeding at the time of delivery. This, in our opinion, would be a worthwhile study to perform. The women in our cohort appear to be at a lower VTE risk, as a whole, compared to the Auckland cohort. We report 12 events (2.4% overall), although 3 of these events occurred in women who were >6 weeks postpartum and had stopped LMWH thromboprophylaxis. A further woman was a superficial thrombophlebitis event post evacuation of retained products of conception (ERPC) and one was a transient ischaemic attack (TIA). Excluding these, we had 7 events (1.4%), including 4 women who had possible events. Outside of the pregnancy setting, LMWH thromboprophylaxis is not 100% effective and does raise an important question about the optimal thromboprophylaxis strategy during pregnancy and the puerperium. We agree with McLintock et al that there is a fundamental knowledge gap; further research is required in the risk assessment models used, to help decide which women should and should not receive antenatal and postnatal thromboprophylaxis, when, and for how long. We share a common belief with the authors of both the New Zealand and Dutch groups that a personalized VTE risk assessment should aim to avoid unnecessary LMWH, particularly in the setting of lower‐risk conditions. In recent years, there has been significant debate about which women should receive thromboprophylaxis, particularly during the postnatal period.9.Arya R. Postpartum venous thromboembolism prophylaxis: harm versus benefit.BJOG. 2018; 125: 1117Crossref PubMed Scopus (5) Google Scholar Some attempt has been made in the past by Eckmana and colleagues to suggest including women’s values10.Eckmana M.H. Alonso‐Coellob P. Guyattb G.H. et al.Women's values and preferences for thromboprophylaxis during Pregnancy: A comparison of direct‐choice and decision analysis using patient specific utilities.Thromb Res. 2015; 136: 341-347Abstract Full Text Full Text PDF Scopus (17) Google Scholar in the decision‐making process. This, in our opinion, is worth building on. From our experience weight‐based prophylactic enoxaparin regimen is serving the women we manage well. We congratulate the investigators of the HiLow study on reaching their recruitment targets recently and we very much look forward to reviewing the results and the further discussions which will ensue as a result. The authors report no relevant conflicts of interest in relation to this work.