Pregnancy-related deaths in the United States have risen from a low of 7.2 per 100,000 live births in 1987 to a high of 17.3 per 100,000 live births in 2013, according to the Centers for Disease Control and Prevention (CDC) Pregnancy Mortality Surveillance System.
Pregnancy-related deaths in the United States have risen from a low of 7.2 per 100,000 live births in 1987 to a high of 17.3 per 100,000 live births in 2013, according to the Centers for Disease Control and Prevention (CDC) Pregnancy Mortality Surveillance System.1 Alarmingly, such deaths among African-American women are more than 3 times as common as among white women (40.4 vs 12.1 per 100,000).
Relative causes of maternal mortality have also changed. Between 1991 and 1999, pulmonary embolism (PE) was the leading cause of pregnancy-related deaths, accounting for 21% of cases.2 However, by 2013, cardiovascular disease and cardiomyopathies accounted for a quarter of all maternal deaths while PE accounted for only 9.2% of cases. Paradoxically, between 1994 and 2009, the rate of venous thromboembolism (VTE) among antepartum and postpartum hospitalizations increased by 17% and 47%, respectively, and the rate of maternal hospitalizations associated with PE increased 128%.3 While this increase was concordant with dramatic increases in major risk factors for VTE (including obesity, advanced maternal age at delivery, and medical comorbidities), it was inversely proportional to the actual maternal deaths ascribable to PE over that period. We can only speculate on explanations for this contradiction such as increased diagnoses of deep venous thrombi and small PEs with the advent of improved imaging techniques, increased clinical awareness and surveillance, more effective treatments, or simply better documentation. But lest we become complacent, we must acknowledge that any death from PE is at least theoretically preventable.
While it is too early to appreciate its full impact, there is evidence that the American College of Obstetricians and Gynecologists (ACOG) recommendation for universal use of peri- and postoperative pneumatic compression devices in all women undergoing cesarean delivery4 may be having an impact on PE-associated mortality. Clark and associates reported that there was an 86% decline in postoperative deaths due to PE from 7 per 458,097 cesarean births in 2006-2007 to 1 in 465,880 cesarean births in 2007–2012 following the introduction of such devices for all cesarean births beginning in 2007 in Hospital Corporation of America obstetrical facilities.5 But can we do better? A recent consensus statement from a patient safety group opines that we can.
The National Partnership for Maternal Safety (NPMS) was created to improve maternal health and safety in the United States and consists of multiple stakeholder organizations.6 This group recently proposed an aggressive patient safety bundle to reduce occurrence of VTE in pregnancy.7 The NPMS authors note that beyond universal use of pneumatic compression devices for cesarean delivery, current ACOG recommendations are rather limited in regard to use of prophylactic post-cesarean anticoagulation, stating it should be applied when “additional,” largely unspecified, VTE risks are present. Similarly, for hospitalized antepartum patients and patients who are having vaginal deliveries, ACOG recommends either prophylactic or therapeutic anticoagulation only when there is “significant risk” of VTE during pregnancy or the puerperium (ie, among patients with antiphospholipid antibody (APA) syndrome or inherited thrombophilias).4
In contrast, the Royal College of Obstetricians and Gynaecologists (RCOG) has taken a far more aggressive posture on the use of pharmacological thromboprophylaxis that has been accompanied by a corresponding temporal reduction in VTE deaths in the United Kingdom from 1.94 deaths per 100,000 births in 2003–2005 to 1.01 deaths per 100,000 births from 2001 to 2013. The 2015 RCOG guidelines recommend use of a rather complex scoring system with subsequent antenatal or postpartum thromboprophylaxis using low-molecular-weight heparin (LMWH) based on the score.8 The RCOG also recommends that practitioners consider thromboprophylaxis for all antepartum hospitalizations in the absence of specific contraindications and opines that all patients in labor who require a cesarean delivery and all patients with a BMI >40 kg/m2 receive 10 days of postnatal thromboprophylaxis.
The proposed NPMS VTE reduction bundle is comprised of 4 elements.7 The first is “readiness,” which essentially calls for every US obstetrical unit to develop a risk assessment strategy at 4 time points in pregnancy: 1) first prenatal visit; 2) all antenatal admissions; 3) immediate postpartum state; and 4) hospital discharge. The NPMS authors suggest using of one of 2 VTE risk scoring systems. The modified Caprini Risk Assessment Model quantifies VTE risks among perioperative patients assigning points for various risk factors including pregnancy and the first postpartum month.9 However, the difference in VTE occurrence between the highest and lowest Caprini categories is modest, 1.94% versus 0%, though the model did identify pregnancy and the puerperium to be a major VTE risk period (odds ratio 8.32; 95% CI:1.02–67.8).
The Padua Prediction Score was developed to assess VTE risk among nonobstetrical Italian internal medicine patients.10 Again points were assigned for various thrombotic risk factors, none of which were pregnancy-related, and patients designated as either at high or low risk, among whom VTE occurred in 7.5% and 0.3%, respectively. These scoring paradigms’ predictive accuracy is compromised by the variable use of various mechanical and pharmacological thromboprophylaxis among the populations studied.
The second element of the NPMS bundle is “recognition” of at-risk patients using risk assessment strategies. The authors reviewed criteria from ACOG, the American College of Chest Physicians (ACCP), and that of the RCOG, described above.
The third element, “response,” emphasizes that at-risk patients receive appropriate thromboprophylaxis. The NPMS authors base recommendations on a selective mix of Caprini, Padua, ACOG, ACCP, and RCOG strategies that can be divided into 5 categories:
Outpatient antepartum patients
They suggest that antepartum patients who have had multiple prior VTE episodes, or 1 prior VTE event associated with either a co-existent high-risk inherited thrombophilia or APA syndrome, receive treatment doses of LMWH or unfractionated heparin (UFH). Prophylactic doses of LMWH or UFH are indicated in those with a prior idiopathic VTE, prior VTE associated with a low-risk inherited thrombophilia, prior VTE while pregnant or taking oral contraceptives (OCs), or in patients without prior personal VTE history who have a family history of VTE associated with a high-risk thrombophilia or APA syndrome.
Patients with antepartum hospitalizations
All patients not already receiving outpatient thromboprophylaxis, admitted to the hospital for at least 72 hours, and who are not at risk for bleeding should receive prophylactic LMWH or UFH. The latter is preferred if delivery is imminent. Hospitalized antepartum patients at risk for bleeding should receive pneumatic compression devices in lieu of heparin.
Patients with vaginal deliveries
Among such patients with a history of VTE or a thrombophilia (the NPMS authors do not discriminate between high- and low-risk types), pneumatic compression devices should be used in the intrapartum period followed by LMWH or UFH postpartum prophylaxis. Women at high risk for VTE based on RCOG or Padua criteria should also be considered for LMWH or UFH.
Patients with cesarean deliveries
All such patients not otherwise receiving LMWH or UFH should have pneumatic compression devices until fully ambulatory. The NPMS authors also recommend pharmacological thromboprophylaxis for women with risk factors based on either RCOG or Caprini criteria. However, they go further to state “Given the challenges in consistently identifying women with risk factors and issues related to poor compliance with mechanical devices, hospitals may choose a strategy in which all women undergoing cesarean birth receive postoperative thromboprophylaxis with unfractionated or low-molecular-weight heparin unless there is a specific contraindication.” This strategy would necessitate that more than a million US women each year receive post-cesarean heparin for some unspecified interval.
The NPMS authors recommend 6 weeks of treatment with LMWH or UFH therapy for postpartum patients with multiple prior VTE episodes or those with a prior VTE and either a high-risk inherited thrombophilia or APA syndrome. Six weeks of prophylactic LMWH or UFH therapy is recommended for women with a prior idiopathic or “provoked” VTE, one occurring while pregnant or on OCs or in the setting of a low-risk thrombophilia. For patients without a prior VTE, 6 weeks of prophylactic heparin therapy is also recommended for those with a family history of VTE associated with a high-risk thrombophilia, a family history of VTE when the patient herself has a low-risk inherited thrombophilia, and in any patient with a high-risk thrombophilia or APA syndrome.
The authors also address the thorny issue of heparin use in pregnant patients needing neuraxial anesthesia and recommend delaying anesthesia for 4 hours and preferably 6 hours after a prophylactic (5,000 U) dose of UFH, 6 hours after a dose of therapeutic UFH, 12 hours after a prophylactic LMWH dose, and 24 hours after a therapeutic LMWH dose. They provide recommendations for delays in postpartum epidural catheter removal or spinal needle placement after various doses of LMWH and UFH. They recommend stopping low-dose aspirin therapy for the prevention of pre-eclampsia at 35 to 36 weeks’ gestation.
The final element of the NPMS bundle is “Reporting and Systems Learning.” The authors opine that institutional and practitioner adherence to these policies be periodically audited and that each VTE event be reviewed to determine if care was optimal. Complications of thromboprophylaxis should also be reviewed and reported.
The proposed NPMS VTE bundle has been criticized as non-evidence-based and too costly. Sibai and Rouse have leveled 4 major critiques.11 The first is that the Padua and Caprini scoring systems advocated for use in pregnant patients were derived from low-quality evidence collected in generally older patients, many of whom had cancer, respiratory and cardiac failure, and other serious comorbidities rarely found in pregnant populations. Second, the antepartum NPMS thromboprophylaxis recommendations fly in the face of a recent Cochrane review that concluded insufficient evidence existed to justify such a policy. Indeed, this report concluded there was also insufficient evidence to justify post-cesarean prophylaxis at this time.12 Third, Sibai and Rouse criticize the NPMS authors for not stating how long thromboprophylaxis should be continued after vaginal delivery (2 days or 6 weeks?).
Finally, they note that the RCOG criteria for thromboprophylaxis after cesarean delivery would apply to half of US cesarean delivery patients (1.2 million a year) and cost up to $52 million for a 4-day course and $130 million for a 10-day course, just for the drugs. Universal application would double these costs. Moreover, additional costs would accrue from the hemorrhagic complications of such therapy. Sibai and Rouse also point out that the NPMS authors’ concern about patient and hospital staff non-adherence with pneumatic compression device use flies in the face of the aforementioned findings by Clark and associates5 and they advocate increased use of this modality as an alternative to the proposed NPMS VTE bundle.
Our rising national maternal mortality rates should concern us all. Clearly, intensive study must be brought to bear on all such causes, especially on the rapid rise in cardiovascular-related deaths, which also have far broader public health implications (see my editorial “Ob/gyns should join the fight against quiet killer,” ContemporaryOBGYN.net/quiet-killer). Moreover, although fatal PEs represent a substantially lower proportion of total maternal deaths compared with 2 decades ago, as noted previously, virtually all deaths from PE are at least theoretically preventable. Thus, the question at hand is: Just how far are we willing to go in preventing pregnancy-related VTE both in terms of dollars spent and complications accepted per life saved?
I believe ACOG should carefully consider the NPMS recommendations and modestly liberalize its indications for pharmacological thromboprophylaxis. In the interim, we should all work hard to ensure that at-risk patients are adherent with pneumatic compression devices and that nurses and physicians are equally vigilant in ensuring such adherence. I suggest joining your postpartum charge nurse and rounding on all your facilities’ post-cesarean delivery patients. You may be astounded by how few of these women actually have pneumatic compression devices applied to their legs!
1. Centers for Disease Control and Prevention. Pregnancy Mortality Surveillance System. https://www.cdc.gov/reproductivehealth/maternalinfanthealth/pmss.html.
2. Centers for Disease Control and Prevention. Pregnancy-Related Mortality Surveillance-United States, 1991–1999. Causes of pregnancy-related death, by outcome of pregnancy and pregnancy-related mortality ratios (PRMR), United States, 1991–1999. https://www.cdc.gov/mmwr/preview/mmwrhtml/ss5202a1.htm#tab3.
3. Ghaji N, Boulet SL, Tepper N, Hooper WC. Trends in venous thromboembolism among pregnancy-related hospitalizations, United States, 1994-2009. Am J Obstet Gynecol. 2013;209(5):433.e1-8.
4. James A; Committee on Practice Bulletins-Obstetrics Practice Bulletin No. 123: thromboembolism in pregnancy. Obstet Gynecol. 2011;118(3):718-729.
5. Clark SL, Christmas JT, Frye DR, Meyers JA, Perlin JB. Maternal mortality in the United States: predictability and the impact of protocols on fatal postcesarean pulmonary embolism and hypertension-related intracranial hemorrhage. Am J Obstet Gynecol. 2014;211(1):32.e1-9.
6. DÊ¼Alton ME, Main EK, Menard MK, Levy BS. The National Partnership for Maternal Safety. Obstet Gynecol. 2014;123(5):973-977.
7. DÊ¼Alton ME, Friedman AM, Smiley RM, Montgomery DM, Paidas MJ, DÊ¼Oria R, Frost JL, Hameed AB, Karsnitz D, Levy BS, Clark SL. National Partnership for Maternal Safety: Consensus Bundle on Venous Thromboembolism. Obstet Gynecol. 2016;128(4):688-698.
9. Bahl V, Hu HM, Henke PK, Wakefield TW, Campbell DA Jr, Caprini JA. A validation study of a retrospective venous thromboembolism risk scoring method. Ann Surg. 2010;251(2):344-350.
10. Barbar S, Noventa F, Rossetto V, et al. A risk assessment model for the identification of hospitalized medical patients at risk for venous thromboembolism: the Padua Prediction Score. J Thromb Haemost. 2010;8(11):2450-2457.
11. Sibai BM, Rouse DJ. Pharmacologic thromboprophylaxis in obstetrics: broader use demands better data. Obstet Gynecol. 2016;128(4):681-684.
12. Bain E, Wilson A, Tooher R, Gates S, Davis LJ, Middleton P. Prophylaxis for venous thromboembolic disease in pregnancy and the early postnatal period. Cochrane Database Syst Rev. 2014;(2):CD001689.