Abstract
Background:
Venous thromboembolism (VTE) accounts for approximately 9% of pregnancy-related deaths in the United States. National guidelines recommend postpartum risk stratification and pharmacologic prophylaxis in at-risk individuals. Knowledge on modern rates of postpartum pharmacologic thromboprophylaxis and its associated risks is limited.
Objective:
To describe the rate of, and factors associated with, initiation of postpartum VTE pharmacologic prophylaxis and to assess associated adverse outcomes.
Study Design:
Secondary analysis of a multicenter cohort of individuals delivering on randomly selected days at 17 U.S. hospitals (2019–20). Medical records were reviewed by trained and certified personnel. Those with an antepartum diagnosis of VTE, receiving antepartum anticoagulation, or known SARS-CoV-2 positive were excluded. The primary outcome was use of postpartum pharmacologic thromboprophylaxis. Secondary outcomes included bleeding complications, surgical site infection, hospital readmission, and VTE through 6 weeks postpartum. The rate of thromboprophylaxis administration was assessed by mode of delivery, institution, and continuance to the outpatient setting. Multivariable regression models were developed using k-fold cross-validation with stepwise backward elimination to evaluate factors associated with thromboprophylaxis administration. Univariable and multivariable logistic models with propensity score covariate adjustment were performed to assess the association between thromboprophylaxis administration and adverse outcomes.
Results:
Of 21,114 individuals in the analytic cohort, 11.9% (95% CI 11.4–12.3%) received postpartum pharmacologic thromboprophylaxis; the frequency of receipt was 29.8% (95% CI 28.7–30.9%) following cesarean and 3.5% (95% CI 3.2–3.8%) following vaginal delivery. Institutional rates of prophylaxis varied from 0.21% to 34.8%. Most individuals (83.3%) received thromboprophylaxis only as inpatients. In adjusted analysis, cesarean delivery (aOR 19.17, 95% CI 16.70–22.00), hysterectomy (aOR 15.70, 95% CI 4.35–56.65), and obesity (aOR 3.45, 95% CI 3.02–3.95) were the strongest factors associated with thromboprophylaxis administration. Thromboprophylaxis administration was not associated with surgical site infection (0.9% vs 0.6%; OR 1.48, 95% CI 0.80–2.74), bleeding complications (0.2% vs 0.1%; OR 2.60, 95% CI 0.99–6.80) or postpartum readmission (0.9% vs 0.3%; aOR 1.38, 95% CI 0.68–2.81). The overall rate of VTE was 0.06% (95% CI 0.03–0.10%) and was higher in those receiving prophylaxis (0.2%) compared to not (0.04%).
Conclusion:
Approximately one in 10 patients received postpartum pharmacologic thromboprophylaxis in this U.S. cohort. Rates of prophylaxis varied widely by institution. Cesarean delivery, hysterectomy, and obesity were predominant factors associated with postpartum thromboprophylaxis administration.
Keywords: bleeding complications, heparin-based prophylaxis, post-cesarean thromboprophylaxis, postpartum thromboembolism, rates of prophylaxis, wound complications
Introduction
Venous thromboembolism (VTE) accounts for approximately 9% of pregnancy-related deaths in the U.S., as well as significant maternal morbidity.1–5 Pregnancy and the puerperium are pro-thrombotic states that may be complicated by additional factors (e.g., obesity, cesarean delivery, hemorrhage) that further increase an individual’s VTE risk.4,6 Expert opinion and guidelines from national organizations (e.g., American College of Obstetricians and Gynecologists (ACOG), Society for Maternal-Fetal Medicine (SMFM), Royal College of Obstetricians and Gynaecologists (RCOG)) recommend risk-based stratification to determine if postpartum pharmacologic thromboprophylaxis is indicated.4,7–11
There is significant inter-institutional variation in postpartum pharmacologic prophylaxis, with rates of 1–85% reported at different institutions. The prophylaxis rate is highly dependent on the risk-stratification algorithm that is used.12–15 Some published algorithms only recommend risk stratification post-cesarean while others risk stratify after vaginal delivery as well.4,8,10 Further, the selected risk factors vary between guidelines, as do recommendations for the type and length of pharmacologic prophylaxis.12,16 Some experts warn against implementation of these protocols because of unproven efficacy of pharmacologic prophylaxis in pregnant and postpartum people and the potential risks of bleeding and wound complications.17–19
Current knowledge on national rates of postpartum pharmacologic thromboprophylaxis and its associated risks is limited. We aimed to describe the rate of postpartum pharmacologic prophylaxis and evaluate factors associated with administration in a U.S. cohort. Secondarily, we aimed to evaluate postpartum adverse outcomes among those receiving versus not receiving pharmacologic thromboprophylaxis.
Materials & Methods
This was a secondary analysis of the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Maternal-Fetal Medicine Units (MFMU) Network Gestational Research Assessments for COVID-19 (GRAVID) study. GRAVID was a retrospective cohort study of pregnant and immediately postpartum individuals with singleton or twin gestations delivering on randomly selected days or with confirmed SARS-CoV-2 infection at one of 17 U.S. hospitals from March 2019 through December 2020. Medical records were reviewed by trained and certified personnel with abstraction of demographics, medical history, and obstetric outcomes through 6 weeks postpartum. Full methodology and findings of the GRAVID study have previously been published.20,21
For this analysis, we included only individuals delivered on randomly selected days, which included those who delivered on four randomly selected days per month from March through December 2019, eight randomly selected days per month from March through May 2020, and four randomly selected days per month from June through December 2020.20 Individuals with an antepartum diagnosis of VTE, receiving antepartum anticoagulation, or with unknown timing of anticoagulation use were excluded from the present analysis. Additionally, individuals with confirmed SARS-CoV-2 infection were excluded.
The primary outcome was use of postpartum pharmacologic thromboprophylaxis. Pharmacologic prophylaxis was defined as postpartum receipt of either unfractionated heparin or low molecular weight heparin. The specific pharmacologic thromboprophylaxis protocol used by each hospital was not available. Broadly, four institutions reported provider-level decision making around prophylaxis, one institution reported using prophylaxis for all cesarean deliveries, and the remaining institutions reported using a risk-based protocol. Mechanical prophylaxis was standardly used in individuals undergoing cesarean delivery at all included institutions.
For the secondary objective, adverse outcomes included bleeding complications requiring additional procedure (uterine packing, intrauterine balloon tamponade, uterine artery ligation, uterine compression sutures, laparotomy, evacuation of hematoma, arterial embolization, uterine evacuation, or hysterectomy), superficial or deep incisional surgical site infection, hospital readmission for vaginal bleeding or delayed postpartum hemorrhage (>24 hours after delivery), wound infection, or wound disruption, and VTE, including deep vein thrombosis or pulmonary embolism, inpatient or outpatient through 6 weeks postpartum.
The rates with 95% confidence intervals of thromboprophylaxis administration were calculated for the full cohort and by mode of delivery, institution, and continuance to the outpatient setting. Outpatient prophylaxis was further considered as administration for less than 2 weeks, or for 2–6 weeks postpartum.
Univariable comparisons of individual and clinical characteristics between those receiving and not receiving pharmacologic thromboprophylaxis were performed using chi-square, Fisher’s exact test, or Wilcoxon test as appropriate. Factors that were considered included age, race and ethnicity, insurance status, parity, date of delivery, delivery site, obesity (defined by body mass index >30 kg/m2), tobacco use, substance use (i.e., alcohol, marijuana, opiates, cocaine, amphetamines, hallucinogens, or benzodiazepines), use of assisted reproductive technology, aspirin use, maternal co-morbid conditions (i.e., cancer, autoimmune disease, pre-gestational diabetes mellitus, personal history of VTE, thrombophilia, chronic cardiovascular disease, pulmonary disease, chronic hypertension, chronic renal disease, chronic liver disease), hypertensive disorders of pregnancy, antepartum thrombocytopenia, gestational diabetes mellitus, gestational age at delivery, mode of delivery, postpartum hemorrhage, transfusion of blood products during delivery admission, stillbirth, peripartum infection, hysterectomy, additional procedure for bleeding at time of delivery, and medical therapy for bleeding at delivery.
Factors that were statistically significant in the univariable analysis were considered in model selection. Weighted analyses were performed to account for the sampling of individuals who delivered on randomly selected dates. K-fold cross-validation with stepwise backward elimination was performed to evaluate and determine factors associated with postpartum pharmacologic thromboprophylaxis. Unadjusted and adjusted odds ratios with 95% confidence intervals for receiving postpartum pharmacologic thromboprophylaxis are reported. Additionally, for each risk factor the rate of prophylaxis by factor is reported with unadjusted and adjusted odds ratios with 95% confidence intervals.
The association between postpartum pharmacologic thromboprophylaxis administration and adverse outcomes was assessed using multivariable logistic models with propensity score covariate adjustment. Propensity scores were calculated based on factors that were determined to be associated with postpartum pharmacologic thromboprophylaxis administration. Maternal race and ethnicity were excluded from model selection as there was no biologic plausibility that race was related to any of the outcomes. Adjusted models also included covariates for year of delivery and delivery site. Unadjusted and adjusted odds ratios with 95% confidence intervals for secondary outcomes are reported.
Multivariable modeling was not performed for low outcome event rates. P values <0.05 were considered statistically significant. As this was a secondary analysis, all reported outcomes are considered exploratory. No adjustment was made for multiple comparisons. All analyses were completed using SAS 9.4 (Cary, NC). The GRAVID study was approved by the institutional review board (IRB) of each participating institution and completed under a waiver of consent.20 This analysis included only deidentified data and was considered exempt from IRB approval. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines were followed.22
Results
The GRAVID cohort study included 21,842 individuals delivering on randomly selected days from March 2019 through December 2020. Following exclusion of 728 individuals who had been diagnosed with an antepartum VTE, received antepartum anticoagulation or were SARS-CoV-2 positive, 21,114 remained for this analysis (Figure 1). Of these, 2,502 individuals (11.9%, 95% CI 11.4–12.3%) received postpartum pharmacologic thromboprophylaxis; the frequency of receipt was 29.8% (95% CI 28.7–30.9%) following cesarean and 3.5% (95% CI 3.2–3.8%) following vaginal delivery. Rates of postpartum pharmacologic thromboprophylaxis across institutions ranged from 0.21% to 34.8% (median 14.6%, interquartile range 2.0–18.9%). Most individuals (83.3%) received pharmacologic thromboprophylaxis only as inpatients. Of those receiving outpatient pharmacologic thromboprophylaxis, only a small proportion (8.4%) received it greater than 2 weeks.
Figure 1.
Study population
*Individual may meet more than one exclusion.
Individuals receiving postpartum pharmacologic thromboprophylaxis as compared to those not receiving thromboprophylaxis were more likely to be of advanced maternal age (32.5% vs 20.6%, p<0.001), be obese (78.6% vs 54.4%, p<0.001), deliver by cesarean (79.8% vs 25.3%, p<0.001), and have a co-morbid condition (33.9% vs 20.5%, p<0.001; Table 1). In unadjusted and adjusted analyses, cesarean delivery, hysterectomy, and obesity were the factors most strongly associated with postpartum pharmacologic thromboprophylaxis administration (Table 2; Table 3).
Table 1.
Baseline characteristics of study population
| Characteristic | Postpartum prophylaxis (n=2502) | No postpartum prophylaxis (n=18612) | p-value |
|---|---|---|---|
| Age (years) | 31.5 ± 5.95 | 29.6 ± 5.77 | <0.001 |
| Advanced maternal age (≥ 35 years) | 812 (32.5) | 3841 (20.6) | <0.001 |
| Race and ethnicity | <0.001 | ||
| Non-Hispanic Black | 730 (29.2) | 3786 (20.3) | |
| Non-Hispanic White | 892 (35.7) | 8494 (45.6) | |
| Hispanic | 574 (22.9) | 4232 (22.7) | |
| Other | 306 (12.2) | 2100 (11.3) | |
| Private insurance | 1324 (53.2) | 10054 (54.5) | 0.21 |
| Obesity (BMI ≥ 30 kg/m2) | 1925 (78.6) | 9796 (54.4) | <0.001 |
| Tobacco use | 208 (8.3) | 1358 (7.3) | 0.07 |
| Substance use | 205 (8.2) | 1583 (8.5) | 0.60 |
| Assisted reproductive technology | 164 (6.6) | 640 (3.5) | <0.001 |
| Aspirin use | 584 (23.4) | 1891 (10.2) | <0.001 |
| Co-morbid conditions | 848 (33.9) | 3814 (20.5) | <0.001 |
| Cancer | 31 (1.2) | 160 (0.9) | 0.06 |
| Autoimmune disease | 43 (1.7) | 338 (1.8) | 0.73 |
| Pre-gestational diabetes | 133 (5.3) | 349 (1.9) | <0.001 |
| History of venous thromboembolism | 26 (1.0) | 30 (0.2) | <0.001 |
| Thrombophilia | 19 (0.8) | 74 (0.4) | 0.01 |
| Chronic cardiovascular disease* | 58 (2.3) | 194 (1.0) | <0.001 |
| Pulmonary disease | 438 (17.5) | 2259 (12.1) | <0.001 |
| Chronic hypertension | 325 (13.0) | 765 (4.1) | <0.001 |
| Chronic renal disease† | 20 (0.8) | 75 (0.4) | 0.005 |
| Chronic liver disease‡ | 28 (1.1) | 160 (0.9) | 0.19 |
| Pregnancy induced hypertension | 923 (36.9) | 3985 (21.4) | <0.001 |
| Antepartum thrombocytopenia | 41 (1.6) | 192 (1.0) | 0.006 |
| Gestational diabetes | 294 (11.8) | 1348 (7.3) | <0.001 |
| Gestational age at delivery (weeks) | 38.9 (37.0 – 39.4) | 39.1 (38.0 – 39.7) | <0.001 |
| Prolonged labor (≥ 24 hours) | 486 (19.7) | 2615 (14.3) | <0.001 |
| Cesarean delivery | 1996 (79.8) | 4701 (25.3) | <0.001 |
| Postpartum hemorrhage (>1000 mL) | 416 (16.6) | 795 (4.3) | <0.001 |
| Transfusion of blood products | 145 (5.8) | 392 (2.1) | <0.001 |
| Procedure for bleeding at delivery§ | 88 (3.5) | 323 (1.7) | <0.001 |
| Medical therapy for bleeding at delivery|| | 477 (19.1) | 2730 (14.7) | <0.001 |
| Hysterectomy | 31 (1.2) | 16 (0.1) | <0.001 |
| Peripartum infection¶ | 209 (8.4) | 867 (4.7) | <0.001 |
| Multifetal gestation | 125 (5.0) | 349 (1.9) | <0.001 |
| Stillbirth (≥ 20 weeks) | 12 (0.5) | 128 (0.7) | 0.23 |
Data are mean (± standard deviation), median (interquartile range), or n (%) unless otherwise specified.
BMI, body mass index.
Including history of myocardial infarction, cardiomyopathy, heart failure, heart arrhythmias
Including chronic renal failure, glomerulonephritis, etc., or designated by proteinuria >500 mg on 24-hour urine collection, or elevated serum creatinine >1.2 mg/dL
Including Hepatitis B, Hepatitis C, autoimmune hepatitis, primary sclerosing cholangitis, Wilson’s disease, etc., but excluding non-alcoholic fatty liver disease
Including at least one of the following: uterine packing, intrauterine balloon, uterine artery ligation, uterine compression sutures, laparotomy, evacuation of hematoma, arterial embolization, uterine evacuation
Including at least one of the following: tranexamic acid, methergine, hemabate, prostaglandins
Including presence of at least one of the following: chorioamnionitis, endometritis, pelvic abscess, pyelonephritis, pneumonia, bacteremia, sepsis
Number of missing values: maternal age (n=3), insurance status (n=177), obstetric history (n=18), body mass index (n=645), assisted reproductive technology (n=309), aspirin use (n=43), history of cancer (n=12), gestational diabetes (n=27), estimated date of delivery (n=9), length of labor (n=326), transfusion of blood products (n=1), procedure for bleeding (n=3), hysterectomy (n=2).
Table 2.
Clinical characteristics associated with postpartum thromboprophylaxis administration
| Characteristic | Postpartum prophylaxis (n=2502) | No postpartum prophylaxis (n=18612) | OR (95% CI) | Adjusted OR* (95% CI) |
|---|---|---|---|---|
| Obesity (BMI > 30 kg/m2) | 1925 (78.6) | 9796 (54.4) | 3.08 (2.79 – 3.41) | 3.45 (3.02 – 3.95) |
| Cesarean delivery | 1996 (79.8) | 4701 (25.3) | 11.67 (10.53 – 12.94) | 19.17 (16.70 – 22.00) |
| Aspirin use | 584 (23.4) | 1891 (10.2) | 2.70 (2.43 – 2.99) | 1.78 (1.52 – 2.08) |
| Maternal co-morbid conditions† | 848 (33.9) | 3814 (20.5) | 1.99 (1.82 – 2.18) | 1.40 (1.22 – 1.61) |
| Hypertensive disorders of pregnancy | 923 (36.9) | 3985 (21.4) | 2.15 (1.96 – 2.34) | 1.87 (1.64 – 2.14) |
| Gestational diabetes | 294 (11.8) | 1348 (7.3) | 1.71 (1.49 – 1.95) | 1.42 (1.15 – 1.75) |
| Gestational age at delivery (weeks) | 38.9 (37.0 – 39.4) | 39.1 (38.0 – 39.7) | 0.93 (0.92 – 0.94) | 0.95 (0.93 – 0.96) |
| Prolonged labor (≥ 24 hours) | 486 (19.7) | 2615 (14.3) | 1.47 (1.32 – 1.64) | 1.41 (1.21 – 1.66) |
| Hysterectomy | 31 (1.2) | 16 (0.1) | 14.58 (7.96 – 26.69) | 15.70 (4.35 – 56.65) |
| Peripartum infection‡ | 209 (8.4) | 867 (4.7) | 1.87 (1.59 – 2.18) | 1.55 (1.22 – 1.96) |
| Postpartum hemorrhage (>1000 mL) | 416 (16.6) | 795 (4.3) | 4.47 (3.94 – 5.07) | 1.50 (1.21 – 1.84) |
| Race and ethnicity | ||||
| Non-Hispanic Black | 730 (29.2) | 3786 (20.3) | 1.88 (1.68 – 2.09) | 1.43 (1.21 – 1.68) |
| Non-Hispanic White | 892 (35.7) | 8494 (45.6) | Reference | Reference |
| Hispanic | 574 (22.9) | 4232 (22.7) | 1.30 (1.16 – 1.45) | 0.96 (0.81 – 1.14) |
| Other | 306 (12.2) | 2100 (11.3) | 1.33 (1.15 – 1.54) | 1.14 (0.94 – 1.38) |
Data are median (interquartile range), or n (%) unless otherwise specified.
OR, odds ratio; CI, confidence interval; BMI, body mass index.
Model adjusted for obesity, cesarean delivery, year of delivery, aspirin use, maternal comorbid condition, hypertensive disorders of pregnancy, gestational diabetes, gestational age at delivery, prolonged labor, hysterectomy, peripartum infection, postpartum hemorrhage, race-ethnicity, and delivery site.
Including presence of at least one of the following: cancer, autoimmune disease, pre-gestational diabetes, personal history of venous thromboembolism, thrombophilia, chronic cardiovascular disease, pulmonary disease, chronic hypertension, chronic renal disease, chronic liver disease
Including presence of at least one of the following: chorioamnionitis, endometritis, pelvic abscess, pyelonephritis, pneumonia, bacteremia, or sepsis
Table 3.
Rate of pharmacologic thromboprophylaxis among individuals with risk factors with unadjusted and adjusted odds ratio estimates
| Characteristic | n / N | Rate (%) | OR (95% CI) | aOR (95% CI) |
|---|---|---|---|---|
| Obesity (BMI ≥ 30 kg/m2) | ||||
| Yes | 1925 / 11721 | 16.4 | 3.08 (2.79–3.41) | 3.45 (3.02–3.95) |
| No | 524 / 8748 | 6.0 | 1.0 | 1.0 |
| Cesarean delivery | ||||
| Yes | 1996 / 6697 | 29.8 | 11.67 (10.53–12.94) | 19.17 (16.70–22.00) |
| No | 506 / 14417 | 3.5 | 1.0 | 1.0 |
| Aspirin use | ||||
| Yes | 584 / 2475 | 23.6 | 2.70 (2.43–2.99) | 1.78 (1.52–2.08) |
| No | 1911 / 18596 | 10.3 | 1.0 | 1.0 |
| Maternal co-morbid conditions† | ||||
| Yes | 848 / 4662 | 18.2 | 1.99 (1.82–2.18) | 1.40 (1.22–1.61) |
| No | 1654 / 16452 | 10.1 | 1.0 | 1.0 |
| Hypertensive disorders of pregnancy | ||||
| Yes | 923 / 4908 | 18.8 | 2.15 (1.96–2.34) | 1.87 (1.64–2.14) |
| No | 1579 / 16206 | 9.7 | 1.0 | 1.0 |
| Gestational diabetes | ||||
| Yes | 294 / 1642 | 17.9 | 1.71 (1.49–1.95) | 1.42 (1.15–1.75) |
| No | 2203 / 19445 | 11.3 | 1.0 | 1.0 |
| Gestational age at delivery | ||||
| <37w0d | 552 / 2745 | 20.1 | 2.19 (1.94–2.47) | 1.66 (1.39–2.00) |
| 37w0d - 38w6d | 720 / 5941 | 12.1 | 1.21 (1.09–1.35) | 1.08 (0.93–1.25) |
| 39w0d - 39w6d | 844 / 8064 | 10.5 | 1.0 | 1.0 |
| ≥40w0d | 385 / 4355 | 8.8 | 0.80 (0.71–0.92) | 1.01 (0.85–1.19) |
| Prolonged labor (≥ 24 hours) | ||||
| Yes | 486 / 3101 | 15.7 | 1.47 (1.32–1.64) | 1.41 (1.21–1.66) |
| No | 1986 / 17687 | 11.2 | 1.0 | 1.0 |
| Hysterectomy | ||||
| Yes | 31 / 47 | 66.0 | 14.58 (7.96–26.69) | 15.70 (4.35–56.65) |
| No | 2471 / 21065 | 11.7 | 1.0 | 1.0 |
| Peripartum infection‡ | ||||
| Yes | 209 / 1076 | 19.4 | 1.87 (1.59–2.18) | 1.55 (1.22–1.96) |
| No | 2293 / 22038 | 11.4 | 1.0 | 1.0 |
| Postpartum hemorrhage (>1000 mL) | ||||
| Yes | 416 / 1211 | 34.4 | 4.47 (3.94–5.07) | 1.50 (1.21–1.84) |
| No | 2086 / 19903 | 10.5 | 1.0 | 1.0 |
| Race and ethnicity | ||||
| Non-Hispanic Black | 730 / 4516 | 16.2 | 1.88 (1.68–2.09) | 1.43 (1.21–1.68) |
| Non-Hispanic White | 892 / 9386 | 9.5 | 1.0 | 1.0 |
| Hispanic | 574 / 4806 | 11.9 | 1.30 (1.16–1.45) | 0.96 (0.81–1.14) |
| Other | 306 / 2406 | 12.7 | 1.33 (1.15–1.54) | 1.14 (0.94–1.38) |
In unadjusted models, postpartum pharmacologic thromboprophylaxis administration was associated with an increased odds of postpartum readmission but not surgical site infection or bleeding complications (Table 4). In adjusted models, pharmacologic thromboprophylaxis was no longer associated with readmission. The overall rate of VTE in the study population was 0.06% (95% CI 0.03–0.10%) and was higher in those receiving postpartum prophylaxis (0.2% receiving prophylaxis vs 0.04% not receiving prophylaxis; OR 4.87, 95% CI 1.59–14.90).
Table 4.
Postpartum pharmacologic thromboprophylaxis administration and adverse outcomes
| Postpartum prophylaxis (n=2502) | No postpartum prophylaxis (n=18612) | OR (95% CI) | Adjusted OR* (95% CI) | |
|---|---|---|---|---|
| Bleeding complication† | 6 (0.2) | 19 (0.1) | 2.60 (0.99 – 6.80) | – |
| Postpartum readmission | 23 (0.9) | 49 (0.3) | 3.57 (2.13 – 6.01) | 1.38 (0.68 – 2.81) |
| Vaginal bleeding/hemorrhage | 2 (0.1) | 27 (0.1) | ||
| Wound disruption | 7 (0.3) | 5 (0.03) | ||
| Wound infection | 14 (0.6) | 17 (0.1) | ||
| Surgical site infection‡ | 18 / 1996 (0.9) | 26 / 4701 (0.6) | 1.48 (0.80 – 2.74) | – |
| Venous thromboembolism | 5 (0.2) | 8 (0.04) | 4.87 (1.59 – 14.90) | – |
Data are n (%) unless otherwise specified.
OR, odds ratio; CI, confidence interval
Model adjusted for propensity score, year of delivery, and delivery site. Propensity score based on factors associated with postpartum prophylaxis: obesity, cesarean delivery, aspirin use, co-morbid conditions, pregnancy induced hypertension, gestational diabetes, gestational age at delivery, prolonged labor, hysterectomy, peripartum infection, and postpartum hemorrhage.
Including at least one of the following within 6 weeks postpartum: uterine packing, intrauterine balloon tamponade, uterine artery ligation, uterine compression sutures, laparotomy, evacuation of hematoma, arterial embolization, uterine evacuation, or hysterectomy.
Denominator reduced to cesarean deliveries only. Including superficial (involving only skin and subcutaneous tissue of incision with presence of purulent drainage, organism identified from incisional specimen, or incision deliberately opened by care provider, and presence of localized pain, localized swelling, erythema, or heat) and deep (involving deep soft tissues of the incision with presence of purulent drainage, or spontaneous dehiscence or deliberate opening with organism identified from incisional specimen and presence of fever or localized pain, or abscess).
Comment
Principal Findings
In this secondary analysis of a U.S. cohort, approximately one in 10 patients received postpartum pharmacologic thromboprophylaxis. The factors most strongly associated with thromboprophylaxis administration were cesarean delivery, hysterectomy, and obesity. Use of postpartum pharmacologic thromboprophylaxis was not associated with surgical site infections or bleeding complications. Use of postpartum pharmacologic thromboprophylaxis was associated with a higher odds of VTE, likely reflecting confounding by indication.
Results in the Context of What is Known
Studies from population and administrative databases demonstrate low rates of postpartum pharmacologic thromboprophylaxis. Using the Premier Healthcare database, Friedman et al found 1.3% of individuals (n=16,639/1,263,205) received pharmacologic thromboprophylaxis after cesarean delivery from 2003 to 2010, while less than 1% received pharmacologic thromboprophylaxis after vaginal delivery from 2006 to 2012.13,14 Following more widespread publication of national postpartum thromboprophylaxis guidelines in the mid-2010s, analysis using the MarketScan Commercial Research Database found rates of prophylaxis increased (0.37% in 2008 to 0.53% in 2017; p<0.001 for test of trend) although absolute rates remained low.23 We found higher rates of postpartum pharmacologic thromboprophylaxis (11.9%) in this U.S. cohort. These findings may suggest prior under ascertainment of prophylaxis in large databases or reflect increased uptake of postpartum prophylaxis among the institutions in this study, most of which were academic medical centers. Findings may also reflect pandemic-related changes in clinical practice.
Palmerola and colleagues found rates of postpartum pharmacologic thromboprophylaxis at the institutional level vary widely depending on the risk-stratification algorithm selected with rates of prophylaxis estimated at 1.0% following the ACOG, 34.8% following the American College of Chest Physicians, and 85.0% following the RCOG guidelines.4,8,9, The specific protocol for each institution was not known but most institutions endorsed using a risk-based approach. We found a wide range of prophylaxis rates (0.2% to 34.8%) across institutions, likely reflecting multiple factors, including institutional differences in thromboprophylaxis protocol used. As no thromboprophylaxis approach has been clearly found to be superior, institutions should consider surveillance of outcomes with thromboprophylaxis protocol use to further inform practice.
Clinical Implications
Clinical factors with the highest odds ratios for administration of postpartum pharmacologic thromboprophylaxis in this cohort included cesarean delivery (aOR 19.2, 95% CI 16.7–22.0), hysterectomy (aOR 15.7, 95% CI 4.4–56.7), and obesity (aOR 3.5, 95% CI 3.0–4.0). Existing literature supports that these factors are strongly associated with postpartum VTE, and they are included in most thromboprophylaxis risk algorithms.6,8,10,24,25 Similarly, other characteristics associated with postpartum thromboprophylaxis administration in this analysis, including co-morbid conditions, hypertensive disorders of pregnancy, and infection, have also been found to be associated with postpartum VTE and are included in many risk algorithms.6,8,9,25
Aspirin use was associated with postpartum thromboprophylaxis administration in this analysis, although unlike the other characteristics, aspirin is not a known risk factor for VTE. As low-dose aspirin prophylaxis is recommended by the U.S. Preventative Services Task Force for individuals at risk for preeclampsia during pregnancy, the identified association between aspirin and pharmacologic prophylaxis likely identifies the presence of risk factors for VTE in those individuals on aspirin.26
Research Implications
Evidence on the relationship between pharmacologic thromboprophylaxis and postpartum complications has been mixed. In a single center case-control study of individuals meeting criteria for post-cesarean pharmacologic thromboprophylaxis, individuals receiving thromboprophylaxis (n=653) compared with those not receiving prophylaxis (n=1,042) experienced higher rates of wound separation (6.8% vs 3.6%, p=0.003) and rehospitalization (2.1% vs 0.8%, p=0.017) but there was no difference in rates of wound hematoma (1.7% vs 1.1%, p=0.28).27 In contrast, in a single center retrospective cohort study, Lu et al found rates of wound hematoma increased from 0.4% (n=50/11,799) to 0.7% (n=90/12,430) (aOR 2.34, 95% CI 1.54–3.57) after implementation of a heparin-based postpartum thromboprophylaxis protocol.19 Other studies that have examined the association between postpartum pharmacologic prophylaxis and complications have been limited in their ability to discern associations due to their small sample sizes and inclusion of those who were receiving therapeutic anticoagulation.28–31 Postpartum pharmacologic thromboprophylaxis was not associated with surgical site infections or bleeding complications in this analysis. The absolute rates for complications across both the prophylaxis and no prophylaxis groups were not greater than baseline population rates.27,32,33
The study sample size was sufficient to detect clinically meaningful differences in administration of prophylaxis and multiple adverse outcomes, but there was insufficient statistical power for the rare outcome of VTE. In the absence of prospective trials addressing efficacy of postpartum pharmacologic thromboprophylaxis for prevention of VTE, existing guidelines are informed by expert opinion, observational obstetric data, and extrapolation from non-obstetric populations.34,35 The association between postpartum pharmacologic thromboprophylaxis and VTE in this analysis likely reflects the fact that those who received thromboprophylaxis were at much higher risk for VTE at baseline. It is unknown whether the administration of pharmacologic prophylaxis reduced the risk in this higher risk group, or if it was simply ineffective in this population.
The optimal approach for postpartum pharmacologic prophylaxis remains unknown. The current findings add to the ongoing discourse by identifying low absolute rates of complications with prophylaxis. Further prospective research is needed to better elucidate both the efficacy and safety of pharmacologic prophylaxis to inform postpartum prophylaxis guidelines.
Strengths & Limitations
This study has several strengths. The selected cohort was large and diverse, which increases generalizability of the findings. Data were collected through medical record abstraction by trained research staff improving ascertainment of pharmacologic prophylaxis and complications as compared to administrative data. The findings provide an updated estimate of national rates of postpartum thromboprophylaxis and factors most strongly associated with administration.
This study also has limitations. Data were primarily from academic-affiliated institutions and therefore findings may not be generalizable to all hospital types. We were unable to address pharmacologic thromboprophylaxis dosing, which may be influential when considering complications and efficacy to prevent VTE. The low event rates for secondary outcomes may have resulted in insufficient statistical power and increased the risk for type II error. Low event rates also limited multivariable modeling. Therefore, results should be interpreted with caution. Some wound complications were only assessed if they led to readmission in the parent study. Thus, less severe wound complications that were managed conservatively in the outpatient setting could not be evaluated. While we excluded those with known SARS-CoV-2 infection from analysis, individuals with an asymptomatic infection that were never tested for SARS-CoV-2 may have been included.
Conclusions
In summary, we were able to characterize current practice for postpartum pharmacologic thromboprophylaxis across a diverse population. Rates of postpartum pharmacologic thromboprophylaxis were higher than those previously reported using administrative data, but rates varied widely by mode of delivery and institution. Use of thromboprophylaxis was not associated with increased risk of adverse outcomes but results should be interpreted with caution as the event rate for these secondary outcomes was low. Further prospective study is necessary to inform efficacy and safety of postpartum pharmacologic thromboprophylaxis.
Supplementary Material
AJOG at a Glance:
- Why was this study conducted?
- This study aimed to (1) describe the rate of postpartum pharmacologic thromboprophylaxis and factors associated with use, and (2) to assess associated adverse outcomes.
- What are the key findings?
- In this secondary analysis of a multicenter cohort, postpartum pharmacologic prophylaxis was received by approximately one in 10 individuals. Cesarean delivery, hysterectomy, and obesity were predominant factors associated with prophylaxis administration.
- Postpartum thromboprophylaxis was not associated with adverse outcomes but event rates were low across the cohort limiting interpretation.
- What does this study add to what is already known?
- We report on contemporary rates of postpartum thromboprophylaxis.
- Previous studies have conflicting findings on the association between thromboprophylaxis and adverse outcomes. The current findings add to available literature on low absolute rates of complications related to prophylaxis use.
Acknowledgments:
The authors thank Kelly Clark, BSN, RN and Amber Sowles, RN, BSN, for protocol development and coordination between clinical research centers; and Rebecca G. Clifton, PhD for protocol development and oversight.
The members of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network that participated in this study are as follows:
Clinical Centers: Main centers are listed below. Personnel at subsites have the subsite listed in brackets.
University of Pittsburgh, Pittsburgh, PA – H. Simhan, M. Bickus, F. Facco
University of Alabama at Birmingham, Birmingham, AL – A. Tita, J. Grant, A. Leath, S. Longo [Ochsner Health], M. Hendricks [Ochsner Health], K. Arias [Ochsner Health]
The Ohio State University, Columbus, OH – M. Costantine, A. Bartholomew, M. Landon, K. Rood, P. Schneider, H. Frey, D. McKenna [Miami Valley Hospital], S. Wiegand [Miami Valley Hospital], E. K. Snow [Miami Valley Hospital], K. Fennig [Miami Valley Hospital], M. Habli [Good Samaritan Hospital], D. Lambers [Good Samaritan Hospital], M. McClellan [Good Samaritan Hospital]
University of Utah Health Sciences Center, Salt Lake City, UT – T. Metz, A. Sowles, A. Nelsen [Utah Valley], M. Varner, M.S. Esplin [Intermountain Medical Center]
Brown University, Providence, RI – D. Rouse, D. Allard, C. Pettker [Yale], J. Leventhal [Yale], J. Rousseau, J. Milano, L. Early
Columbia University, New York, NY – C. Gyamfi-Bannerman, S. Bousleiman, R. Wapner, D. Sutton, H. Manchon, M. Hoffman [Christiana Care], C. Kitto [Christiana Care], K. Palomares [St. Peter’s U. Hosp], I. Beche [St. Peter’s U. Hosp.], D. Skupski [NY Presbyterian Queens], R. Chan-Akeley [NY Presbyterian Queens]
University of Texas Medical Branch, Galveston, TX – G. Saade, A. Salazar, L. Pacheco, S. Clark, H. Harirah, S. Jain, G. Olson, A. Saad, M. McDonnold [St. David’s Women’s Center of Texas], C. Brown [St. David’s Women’s Center of Texas], L. Allen, G. Carrington, J. Cornwell, J. DeVolder, L. Thibodeaux, E. Welch
MetroHealth Medical Center-Case Western Reserve University, Cleveland, OH – J. Bailit, W. Dalton, A. Tyhulski, A. Mayle [University Hospitals]
University of Texas Health Science Center at Houston-Children’s Memorial Hermann Hospital, Houston, TX – S. Chauhan, H. Mendez-Figueroa, F. Ortiz
University of North Carolina at Chapel Hill, Chapel Hill, NC – J. Thorp, T. Manuck, K. Clark, B. Hughes [Duke], S. Timlin, L. Fried, H. Byers, C. Beamon, MD [WakeMed Health & Hospitals], J. Ferrara [Duke], A. Williams, K. Eichelberger [Greenville], A. Moore [Greenville]
Northwestern University, Chicago, IL – W. Grobman, G. Mallett, M. Ramos-Brinson, B. Plunkett [NorthShore University Evanston Hospital], K. Kearns [NorthShore University Evanston Hospital], A. Palatnik [Froedhert Hospital/ Medical College of Wisconsin], S. Northey [Froedert Hospital/Medical College of Wisconsin]
University of Pennsylvania, Philadelphia, PA – S. Parry, H. Sehdev, M. McCabe, C. Fazio, A. Filipczak, J. Craig, L. Muzzarelli
Data Coordinating Center:
The George Washington University Biostatistics Center, Washington, DC – R. Clifton, G. Sandoval, E. Thom, C. Nwachuku, V. L. Flowers-Fanomezantsoa
NIH:
Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, MD – M. Longo, M. Miodovnik, S. Archer
MFMU Network Steering Committee Chair: George A. Macones
Funding:
This work is funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (UG1 HD087230, UG1 HD027869, UG1 HD027915, UG1 HD034208, UG1 HD040500, UG1 HD040485, UG1 HD053097, UG1 HD040544, UG1 HD040545, UG1 HD040560, UG1 HD040512, UG1 HD087192, U24 HD036801) and the National Center for Advancing Translational Sciences (UL1TR001873). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Conflicts of Interest: The authors report no conflict of interest.
Presentation Information:
The 43rd Annual Pregnancy Meeting, Society for Maternal-Fetal Medicine, San Francisco, California, February 6–11, 2023
A list of the full members of the NICHD MFMU Network is available in the acknowledgements.
Tweetable statement: Postpartum pharmacologic thromboprophylaxis was administered in 12% of this U.S. cohort with administration associated with low absolute rates of complications.
References
- 1.Abe K, Kuklina EV, Hooper WC, Callaghan WM. Venous thromboembolism as a cause of severe maternal morbidity and mortality in the United States. Semin Perinatol 2019;43:200–204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Creanga AA, Syverson C, Seed K, Callaghan WM. Pregnancy-Related Mortality in the United States, 2011–2013. Obstet Gynecol 2017;130:366–373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Clark SL, Belfort MA. The Case for a National Maternal Mortality Review Committee. Obstet Gynecol 2017;130:198–202. [DOI] [PubMed] [Google Scholar]
- 4.Thromboembolism in Pregnancy. ACOG Practice Bulletin No. 196. American College of Obstetricians and Gynecologists. Obstet Gynecol 2018;132:e1–e17. [DOI] [PubMed] [Google Scholar]
- 5.Farmakis LT, Barco S, Hobohm L, Braekkan SK, Connors JM, Giannakoulas G, et al. Maternal mortality related to pulmonary embolism in the United States, 2003–2020. Am J Obstet Gynecol MFM 2022;5:100754. [DOI] [PubMed] [Google Scholar]
- 6.Tepper NK, Boulet SL, Whiteman MK, Monsour M, Marchbanks PA, Hooper WC, et al. Postpartum venous thromboembolism: incidence and risk factors. Obstet Gynecol 2014;123:987–96. [DOI] [PubMed] [Google Scholar]
- 7.D’Alton ME, Friedman AM, Smiley RM, Montgomery DM, Paidas MJ, D’Oria R, et al. National Partnership for Maternal Safety: Consensus Bundle on Venous Thromboembolism. Obstet Gynecol 2016;128:688–98. [DOI] [PubMed] [Google Scholar]
- 8.Royal College of Obstetricians and Gynaecologists. Thrombosis and embolism during pregnancy and the puerperium, reducing the risk. Green-top Guideline No. 37a. RCOG; 2015. [Google Scholar]
- 9.Bates SM, Greer IA, Middeldorp S, Veenstra DL, Prabulos AM, Vandvik PO. VTE, thrombophilia, antithrombotic therapy, and pregnancy: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e691S–e736S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pacheco LD, Saade G, Metz TD. Society for Maternal-Fetal Medicine Consult Series #51: Thromboembolism prophylaxis for cesarean delivery. Am J Obstet Gynecol 2020;223:B11–b17. [DOI] [PubMed] [Google Scholar]
- 11.Friedman AM, D’Alton ME. Expert review: prevention of obstetrical venous thromboembolism. Am J Obstet Gynecol 2021;225:228–236. [DOI] [PubMed] [Google Scholar]
- 12.Palmerola KL, D’Alton ME, Brock CO, Friedman AM. A comparison of recommendations for pharmacologic thromboembolism prophylaxis after caesarean delivery from three major guidelines. BJOG 2016;123:2157–2162. [DOI] [PubMed] [Google Scholar]
- 13.Friedman AM, Ananth CV, Lu Y-S, D’Alton ME, Wright JD. Underuse of Postcesarean Thromboembolism Prophylaxis. Obstet Gynecol 2013;122:1197–1204. [DOI] [PubMed] [Google Scholar]
- 14.Friedman AM, Ananth CV, Prendergast E, Chauhan SP, D’Alton ME, Wright JD. Thromboembolism incidence and prophylaxis during vaginal delivery hospitalizations. Am J Obstet Gynecol 2015;212:221.e1–12. [DOI] [PubMed] [Google Scholar]
- 15.Robison E, Heyborne K, Allshouse AA, Valdez C, Metz TD. Implementation of a Risk-Based Heparin Protocol for Postpartum Venous Thromboembolism Prevention. Obstet Gynecol 2017;130:262–269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gassmann N, Viviano M, Righini M, Fontana P, Martinez de Tejada B, Blondon M. Estimating the risk thresholds used by guidelines to recommend postpartum thromboprophylaxis. J Thromb Haemost 2021;19:452–459. [DOI] [PubMed] [Google Scholar]
- 17.Kotaska A Postpartum Heparin Thromboprophylaxis: More Harm Than Good. Obstet Gynecol 2021;138:527–529. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sibai BM, Rouse DJ. Pharmacologic Thromboprophylaxis in Obstetrics: Broader Use Demands Better Data. Obstet Gynecol 2016;128:681–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lu MY, Blanchard CT, Ausbeck EB, Oglesby KR, Page MR, Lazenby AJ, et al. Evaluation of a Risk-Stratified, Heparin-Based, Obstetric Thromboprophylaxis Protocol. Obstet Gynecol 2021;138:530–538. [DOI] [PubMed] [Google Scholar]
- 20.Metz TD, Clifton RG, Hughes BL, Sandoval GJ, Grobman WA, Saade GR, et al. Association of SARS-CoV-2 Infection With Serious Maternal Morbidity and Mortality From Obstetric Complications. JAMA 2022;327:748–759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Metz TD, Clifton RG, Hughes BL, Sandoval GJ, Grobman WA, Saade GR, et al. Disease Severity and Perinatal Outcomes of Pregnant Patients With Coronavirus Disease 2019 (COVID-19). Obstet Gynecol 2021;137:571–580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007;370:1453–7. [DOI] [PubMed] [Google Scholar]
- 23.Bruno AM, Allshouse AA, Einerson BD, Campbell HM, Branch DW, Silver RM, et al. Trends in postpartum venous thromboembolism and chemical prophylaxis among insured US patients. Am J Obstet Gynecol MFM 2022;4:100620. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Mauney L, Barth WH Jr., Clapp MA. Association between peripartum hysterectomy and venous thromboembolism. Am J Obstet Gynecol 2022;226:119.e1–119.e11. [DOI] [PubMed] [Google Scholar]
- 25.James AH, Jamison MG, Brancazio LR, Myers ER. Venous thromboembolism during pregnancy and the postpartum period: incidence, risk factors, and mortality. Am J Obstet Gynecol 2006;194:1311–5. [DOI] [PubMed] [Google Scholar]
- 26.LeFevre ML. Low-dose aspirin use for the prevention of morbidity and mortality from preeclampsia: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med 2014;161:819–26. [DOI] [PubMed] [Google Scholar]
- 27.Ferres MA, Olivarez SA, Trinh V, Davidson C, Sangi-Haghpeykar H, Aagaard-Tillery KM. Rate of wound complications with enoxaparin use among women at high risk for postpartum thrombosis. Obstet Gynecol 2011;117:119–124. [DOI] [PubMed] [Google Scholar]
- 28.Kominiarek MA, Angelopoulos SM, Shapiro NL, Studee L, Nutescu EA, Hibbard JU. Low-molecular-weight heparin in pregnancy: peripartum bleeding complications. J Perinatol 2007;27:329–34. [DOI] [PubMed] [Google Scholar]
- 29.Moriuchi K, Chigusa Y, Kondoh E, Murakami R, Ueda Y, Mogami H, et al. Enoxaparin administration within 24 hours of caesarean section: a 6-year single-centre experience and patient outcomes. J Obstet Gynaecol 2019;39:451–454. [DOI] [PubMed] [Google Scholar]
- 30.Cox S, Eslick R, McLintock C. Effectiveness and safety of thromboprophylaxis with enoxaparin for prevention of pregnancy-associated venous thromboembolism. J Thromb Haemost 2019;17:1160–1170. [DOI] [PubMed] [Google Scholar]
- 31.Limmer JS, Grotegut CA, Thames E, Dotters-Katz SK, Brancazio LR, James AH. Postpartum wound and bleeding complications in women who received peripartum anticoagulation. Thromb Res 2013;132:e19–23. [DOI] [PubMed] [Google Scholar]
- 32.Callaghan WM, Kuklina EV, Berg CJ. Trends in postpartum hemorrhage: United States, 1994–2006. Am J Obstet Gynecol 2010;202:353.e1–353.e6. [DOI] [PubMed] [Google Scholar]
- 33.Temming LA, Raghuraman N, Carter EB, Stout MJ, Rampersad RM, Macones GA, et al. Impact of evidence-based interventions on wound complications after cesarean delivery. Am J Obstet Gynecol 2017;217:449.e1–449.e9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Bates SM, Middeldorp S, Rodger M, James AH, Greer I. Guidance for the treatment and prevention of obstetric-associated venous thromboembolism. J Thromb Thrombolysis 2016;41:92–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Middleton P, Shepherd E, Gomersall JC. Venous thromboembolism prophylaxis for women at risk during pregnancy and the early postnatal period. Cochrane Database Syst Rev 2021;3:Cd001689. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.