Objective
Although major international guidelines recommend venous thromboembolism (VTE) prophylaxis during vaginal delivery hospitalization for women with additional risk factors, US guidelines recommend prophylaxis for a very small number of women who are at particularly high risk for an event. The purpose of this study was to characterize practice patterns of VTE prophylaxis in the United States during vaginal delivery hospitalizations and to determine VTE incidence in this population.
Study Design
A population-level database was used to analyze VTE incidence and use of VTE prophylaxis during vaginal delivery hospitalizations in the United States between 2006 and 2012 (n = 2,673,986). We evaluated whether patients received either pharmacologic or mechanical prophylaxis. Hospital-level factors and patient characteristics were included in multivariable regression analysis that evaluated prophylaxis administration.
Results
We identified 2,673,986 women who underwent vaginal delivery. Incidence of VTE increased during the study period from 15.6-29.8 events per 100,000 delivery hospitalizations. Within the cohort, 2.6% of patients (n = 68,835) received VTE prophylaxis. Pharmacologic prophylaxis was rare; <1% of women received unfractionated or low-molecular-weight heparin. Although patients with thrombophilia or a previous VTE event were likely to receive prophylaxis (60.8% and 72.8%, respectively), patients with risk factors for VTE such as obesity, smoking, and heart disease were unlikely to receive prophylaxis (rates of 5.9%, 3.3%, and 6.2%, respectively).
Conclusion
Our findings demonstrate that the administration of VTE prophylaxis outside a small group of women at extremely high risk for VTE is rare during vaginal delivery hospitalization. Given that VTE incidence is rising in this population, further research to determine whether broadening prophylaxis for VTE may reduce severe maternal morbidity and death is indicated.
Venous thromboembolism (VTE) is a leading cause of maternal death. A systematic review of maternal deaths that was performed by the World Health Organization implicated embolism in 14.9% of maternal deaths in developed countries ; the Centers for Disease Control and Prevention estimates that thrombotic pulmonary embolism accounted for 9.4% of pregnancy-related deaths from 2006-2009. In the United States, strategies to reduce VTE have focused primarily on perioperative cesarean prophylaxis and prenatal risk assessment of women who are at particularly high risk for events. Despite these efforts that included increasing use of mechanical prophylaxis during cesarean delivery, obstetric thromboembolism has increased 72% during delivery hospitalizations from 1998-2009 according to data from the Nationwide Inpatient Sample.
Recommendations for thromboprophylaxis during vaginal delivery in the United States have focused on women at the highest risk for VTE: women with acquired or hereditary thrombophilia and/or previous thromboembolic events. VTE is twice as common after cesarean delivery compared with vaginal deliveries ; because more women deliver vaginally, many events occur among women who do not undergo cesarean delivery. The prevalence of risk factors for VTE is rising, with obesity, advanced maternal age, and major medical comorbidities becoming increasingly common. In the United Kingdom, national guidelines recommend postpartum pharmacologic prophylaxis for women with previous VTE events or thrombophilias. Additionally, these guidelines recommend prophylaxis for other common risk factors that include obesity, maternal age ≥35 years, smoking, preeclampsia, postpartum hemorrhage, and prolonged labor ( Table 1 ). In the setting of a comprehensive strategy to reduce VTE, death from this cause in the United Kingdom decreased by more than one-half, from 1.94 maternal deaths per 100,000 deliveries from 2003-2005 to 0.79 maternal deaths per 100,000 from 2006-2008.
| Variable | Recommendation |
|---|---|
| Major risk factors | At least 7 days of postnatal prophylactic low-molecular-weight heparin is recommended if any 1 risk factor is present |
| Any previous venous thromboembolism a | |
| Anyone requiring antenatal low-molecular-weight heparin a | |
| Cesarean delivery in labor | |
| Asymptomatic thrombophilia (inherited or acquired) | |
| Obesity (body mass index, >40 kg/m 2 ) b | |
| Prolonged hospital admission | |
| Medical comorbidities (eg, heart or lung disease, systemic lupus erythematosus, cancer, inflammatory conditions, sickle cell disease, intravenous drug user) | |
| Minor risk factors | At least 7 days of postnatal prophylactic low-molecular-weight heparin is recommended if ≥2 risk factors are present |
| Age >35 y | |
| Obesity (body mass index, >30 kg/m 2 ) | |
| Parity ≥ 3 | |
| Smoker | |
| Elective cesarean delivery | |
| Any surgical procedure in the puerperium | |
| Gross varicose veins c | |
| Current systemic infection | |
| Immobility (eg, paraplegia, symphysis pubis dysfunction with reduced mobility, long distance travel d ) | |
| Preeclampsia | |
| Midcavity rotational operative delivery | |
| Prolonged labor (>24 hr) | |
| Postpartum hemorrhage >1 L or blood transfusion |
a At least 6 weeks postnatal prophylaxis required
b Based on earliest documented weight during prenatal care
c Symptomatic, above the knee or associated with phlebitis/edema, skin changes
The objectives of this study were to (1) characterize contemporary practice patterns for thromboembolism prophylaxis during vaginal-delivery hospitalizations in the United States, (2) characterize the incidence of thromboembolism, and (3) to determine whether potential opportunities to reduce risk in this clinical setting are being missed.
Methods
The Perspective (Premier, Charlotte, NC) was used for the analysis. This voluntary, fee-supported database captures hospitalization data from >600 acute care hospitals in the United States. Patient demographic information, disease and procedure codes, and hospital and provider characteristics are included. The database also contains all billed services such as medications, devices, laboratory tests, and radiologic imaging. Data undergo a quality control process that includes 95 separate quality assurance and data validation checks that confirm accuracy before being used for research. For each individual hospital that is included in the dataset, 100% of discharge data is included. Perspective has been used in numerous outcomes studies that include evaluations of postsurgical thromboprophylaxis. In 2006, approximately 15% of all hospitalizations within the United States (almost 5.5 million hospital discharges) were captured in Perspective. All data were deidentified, and the analysis was approved by the Columbia University institutional review board.
We analyzed the cases of women who underwent vaginal delivery from 2006-2012. Patients were identified with the use of an enhanced method to capture delivery hospitalization based on International Classification of Diseases–9th Revision (ICD-9) billing codes V27 and 650 and diagnosis-related group codes 370-375. Patients were excluded if they underwent cesarean delivery with a previously described method. The primary outcome of interest was the use of any VTE prophylaxis during the delivery hospitalization. VTE prophylaxis was classified as mechanical, pharmacologic, or combination pharmacologic/mechanical. Cases that received either graduated compression stockings or intermittent pneumatic compression were coded as receiving mechanical prophylaxis. Patients who received unfractionated heparin, low-molecular-weight heparin (including enoxaparin sodium, tinzaparin sodium, or dalteparin sodium), or fondaparinux sodium were classified as having received pharmacologic prophylaxis. Patients were classified as having received mechanical prophylaxis if they received an appropriate device during any hospital day. Women were classified as having received pharmacologic prophylaxis if they received an appropriate drug during any hospital day.
Epidemiologic literature was reviewed to identify relevant medical, surgical, and obstetric risk factors that were associated with obstetric thromboembolism. Through an iterative process, clinical risk factors that were demonstrated to be associated with increased postpartum thromboembolism risk in large observational and population-based cohorts were chosen for inclusion in the analysis. Hospital characteristics included location (urban vs rural), teaching status (teaching vs nonteaching), annualized vaginal delivery volume (by quartile), geographic region (Midwest, Northeast, South, West), and hospital size based on the number of beds (<400, 400-600, or >600 beds). Patient demographics included age, race, year of hospitalization, and marital status. Comorbidity was estimated with the Elixhauser index that combines comorbid conditions based on ICD coding into an overall measure of medical comorbidity that is used in large administrative data. Comorbidity was also estimated with an index that was designed specifically for use in obstetric patient populations. This index uses demographic and medical risk factors (such as congenital heart disease, advanced maternal age, diabetes mellitus) that are predictive of maternal death and end-organ injury (acute renal failure, pulmonary edema, stroke) to provide a simple measure for summarizing the burden of maternal illness in an obstetric population.
The association between thromboembolism prophylaxis and clinical and demographic variables was compared with the use of χ 2 tests. Rates of deep vein thrombosis, pulmonary embolism, and VTE are reported per 100,000 delivery hospitalizations. VTE was defined as the sum of deep vein thrombosis and pulmonary embolism cases. Cases of VTE were identified by ICD-9 codes for pulmonary embolism (415.1, 415.11, 415.12, 415.19, v12.51, 673.20, 673.21, 673.22, 673.24, 673.80, 673.81, 673.82, 673.84) and deep vein thrombosis (451.1, 451.11, 451.19, 451.2, 451.81, 451.9, 453.4, 453.40, 453.41, 453.42, 453.9, 453.8, 671.40, 67.42, 671.44, 997.2, 999.2). To account for the influence of clinical and demographic factors on the use of prophylaxis, we developed mixed effects log-binomial regression models to examine the use of prophylaxis. These models included all the clinical and demographic characteristics and a hospital-specific random intercept-term to account for hospital-level clustering. Results are reported as risk ratios with 95% confidence intervals (95% CI). A probability value of < .05 was considered statistically significant. All analyses were performed with SAS software (version 9.3; SAS Institute Inc, Cary, NC).
Results
A total of 4,076,078 women were identified as having a delivery hospitalization; 1,402,902 of whom were excluded for delivering by cesarean. A total of 2,673,986 women who had been hospitalized for vaginal delivery were identified and included in the analysis. The incidence of both pulmonary embolism and deep vein thrombosis increased over the course of the study period, and VTE incidence nearly doubled between 2006 and 2012 ( Figure 1 ); 414 women were diagnosed with deep vein thrombosis, and 236 women were diagnosed with pulmonary embolism. Use of either mechanical or pharmacologic prophylaxis was uncommon. Of the 68,835 women (2.6%) who received prophylaxis, most (67.5%; n = 46,474 women) received mechanical prophylaxis. Less than 1.0% of patients received either unfractionated or low-molecular-weight heparin (0.39% and 0.45%, respectively). Table 2 displays the patient demographics and hospital characteristics of the cohort. Prophylaxis increased with comorbidity and varied significantly based on geography, hospital delivery volume, and race ( P < .001). The use of prophylaxis increased over the course of the study period from 1.6% in 2006 to 2.8% in 2012 ( P < .001). This increase was largely due to mechanical prophylaxis increasing from 0.8% in 2006 to 2.1% in 2012.
| Characteristic | No prophylaxis | Any prophylaxis | P value | ||
|---|---|---|---|---|---|
| n | % | n | % | ||
| All patients | 2,605,151 | 97.4 | 68,835 | 2.6 | |
| Age, y | < .001 | ||||
| <20 | 282,392 | 97.8 | 6334 | 2.2 | |
| 20-24 | 666,383 | 97.7 | 15,400 | 2.3 | |
| 25-29 | 753,547 | 97.7 | 18,092 | 2.3 | |
| 30-34 | 593,825 | 97.5 | 15,499 | 2.5 | |
| >34 | 312,629 | 96.9 | 9885 | 3.1 | |
| Race | < .001 | ||||
| White | 1,354,569 | 97.7 | 32,356 | 2.3 | |
| Black | 346,027 | 96.6 | 12,245 | 3.4 | |
| Hispanic | 234,170 | 97.6 | 5780 | 2.4 | |
| Other/Unknown | 674,010 | 97.9 | 14,829 | 2.2 | |
| Year of delivery | < .001 | ||||
| 2006 | 366,317 | 98.4 | 5950 | 1.6 | |
| 2007 | 374,851 | 98.3 | 6662 | 1.8 | |
| 2008 | 352,438 | 97.8 | 7825 | 2.2 | |
| 2009 | 354,460 | 97.3 | 9884 | 2.7 | |
| 2010 | 367,470 | 96.9 | 11,675 | 3.1 | |
| 2011 | 402,359 | 97.1 | 11,911 | 2.9 | |
| 2012 | 390,881 | 97.2 | 11,303 | 2.8 | |
| Marital status | < .001 | ||||
| Married | 1,262,380 | 97.8 | 28,897 | 2.2 | |
| Single | 1,035,563 | 97.1 | 31,201 | 2.9 | |
| Unknown | 310,833 | 98.4 | 5112 | 1.6 | |
| Insurance status | < .001 | ||||
| Medicare | 14,737 | 96.0 | 616 | 4.0 | |
| Medicaid | 1,132,089 | 97.6 | 28,381 | 2.5 | |
| Commercial | 1,304,306 | 97.9 | 27,826 | 2.1 | |
| Uninsured | 70,279 | 97.6 | 1707 | 2.4 | |
| Unknown | 87,365 | 92.9 | 6680 | 7.1 | |
| Elixhauser index | < .001 | ||||
| None | 1,907,431 | 98.0 | 39,791 | 2.0 | |
| 1 | 489,590 | 97.0 | 15,169 | 3.0 | |
| 2 | 131,391 | 95.9 | 5642 | 4.1 | |
| >2 | 80,364 | 94.6 | 4608 | 5.4 | |
| Hospital location | < .001 | ||||
| Rural | 277,621 | 98.0 | 5745 | 2.0 | |
| Urban | 2,331,155 | 97.5 | 59,465 | 2.5 | |
| Hospital teaching status | < .001 | ||||
| Nonteaching | 1,654,770 | 97.7 | 38,812 | 2.3 | |
| Teaching | 954,006 | 97.3 | 26,398 | 2.7 | |
| Hospital size: beds, n | < .001 | ||||
| <400 | 1,468,487 | 97.8 | 32,571 | 2.2 | |
| 400-600 | 720,589 | 97.2 | 20,465 | 2.8 | |
| >600 | 419,700 | 97.2 | 12,174 | 2.8 | |
| Hospital region | < .001 | ||||
| Midwest | 506,126 | 98.2 | 9132 | 1.8 | |
| Northeast | 410,908 | 98.4 | 6763 | 1.6 | |
| South | 1,084,933 | 97.2 | 31,028 | 2.8 | |
| West | 606,809 | 97.1 | 18,287 | 2.9 | |
| Hospital delivery volume quartile | < .001 | ||||
| Lowest volume | 646,488 | 97.6 | 16,170 | 2.4 | |
| 2nd | 652,837 | 97.9 | 14,122 | 2.1 | |
| 3rd | 662,416 | 98.3 | 11,308 | 1.7 | |
| Highest volume | 647,035 | 96.5 | 23,610 | 3.5 | |
| Length of hospitalization, d | < .001 | ||||
| <7 | 2,596,670 | 97.8 | 59,770 | 2.2 | |
| ≥7 | 12,106 | 69.0 | 5440 | 31.0 | |
| Obstetric comorbidity index | < .001 | ||||
| 0 | 1,989,392 | 98.0 | 40,325 | 2.0 | |
| 1 | 401,979 | 97.1 | 11,900 | 2.9 | |
| 2 | 156,034 | 95.8 | 6814 | 4.2 | |
| 3 | 33,401 | 93.7 | 2229 | 6.3 | |
| 4 | 9,489 | 91.2 | 911 | 8.8 | |
| 5 | 13,821 | 87.0 | 2069 | 13.0 | |
| >5 | 4,660 | 82.9 | 962 | 17.1 | |
Analysis of medical and obstetric risk factors for VTE ( Table 3 ) demonstrated that patients with thrombophilia and previous thromboembolism were likely to receive prophylaxis; 60.8% and 72.8%, respectively, of patients with these diagnoses received prophylaxis. Prophylaxis for other risk factors for VTE such as obesity, smoking, and preeclampsia was rare; women who were diagnosed with these conditions received prophylaxis in 5.9%, 3.0%, and 5.9% of cases, respectively. In multivariate regression analysis, the risk ratios for prophylaxis for a history of thromboembolism and hypercoagulability were 10.14 (95% CI, 9.74–10.56) and 9.32 (95% CI, 8.96–9.71), respectively ( Table 4 ). Other medical risk factors for thromboembolism were associated generally with either a marginally increased probability of prophylaxis or no increase at all. For example, the risk ratios for obesity, smoking, and preeclampsia were 1.29 (95% CI, 1.25–1.34), 1.03 (95% CI, 1.00–1.07), and 1.23 (95% CI, 1.19–1.27), respectively. The obstetric comorbidity index demonstrated that a relatively small number of patients in this cohort were at particularly high risk for major maternal morbidity and/or death. Figure 2 shows the number of patients by obstetric comorbidity index score on a logarithmic scale and the rate of prophylaxis by comorbidity score. The logarithmic scale was used to better represent the relatively small number of patients with high obstetric comorbidity index scores.
| Risk factor | No prophylaxis | Any prophylaxis | P value | ||
|---|---|---|---|---|---|
| n | % | n | % | ||
| All patients | 2,605,151 | 97.4 | 68,835 | 2.6 | |
| Previous venous thromboembolism | < .001 | ||||
| No | 2,607,347 | 97.7 | 61,378 | 2.3 | |
| Yes | 1429 | 27.2 | 3832 | 72.8 | |
| Obesity | < .001 | ||||
| No | 2,548,640 | 97.7 | 61,449 | 2.4 | |
| Yes | 60,136 | 94.1 | 3761 | 5.9 | |
| Smoking | < .001 | ||||
| No | 2,472,954 | 97.6 | 60,967 | 2.4 | |
| Yes | 135,822 | 97.0 | 4243 | 3.0 | |
| Immobility | < .001 | ||||
| No | 2,608,718 | 97.6 | 65,177 | 2.4 | |
| Yes | 58 | 63.7 | 33 | 36.3 | |
| Varicose veins | < .001 | ||||
| No | 2,605,123 | 97.6 | 64,887 | 2.4 | |
| Yes | 3653 | 91.9 | 323 | 8.1 | |
| Multiparity | < .001 | ||||
| No | 2,592,719 | 97.6 | 64,358 | 2.4 | |
| Yes | 16,057 | 95.0 | 852 | 5.0 | |
| Hyperemesis | < .001 | ||||
| No | 2,608,315 | 97.6 | 65,175 | 2.4 | |
| Yes | 461 | 92.9 | 35 | 7.0 | |
| Multiple gestation | < .001 | ||||
| No | 2,596,226 | 97.6 | 64,098 | 2.4 | |
| Yes | 12,550 | 91.9 | 1112 | 8.1 | |
| Assisted reproductive technology | < .001 | ||||
| No | 2,606,959 | 97.6 | 65,123 | 2.4 | |
| Yes | 1817 | 95.4 | 87 | 4.6 | |
| Preeclampsia | < .001 | ||||
| No | 2,466,523 | 97.8 | 56,261 | 2.2 | |
| Yes | 142,253 | 94.1 | 8949 | 5.9 | |
| Placental abruption | < .001 | ||||
| No | 2,593,104 | 97.6 | 63,943 | 2.4 | |
| Yes | 15,671 | 92.5 | 1267 | 7.5 | |
| Endometritis | < .001 | ||||
| No | 2,604,788 | 97.6 | 64,834 | 2.4 | |
| Yes | 3988 | 91.4 | 376 | 8.6 | |
| Pyelonephritis | < .001 | ||||
| No | 2,607,415 | 97.6 | 65,097 | 2.4 | |
| Yes | 1361 | 92.3 | 113 | 7.7 | |
| Systemic inflammatory response syndrome | < .001 | ||||
| No | 2,608,763 | 97.6 | 65,206 | 2.4 | |
| Yes | 13 | 76.5 | 4 | 23.5 | |
| Sepsis | < .001 | ||||
| No | 2,608,588 | 97.6 | 65,053 | 2.4 | |
| Yes | 188 | 54.5 | 157 | 45.5 | |
| Pneumonia | < .001 | ||||
| No | 2,608,297 | 97.6 | 64,972 | 2.4 | |
| Yes | 479 | 66.8 | 238 | 33.2 | |
| Influenza | < .001 | ||||
| No | 2,608,433 | 97.6 | 65,181 | 2.4 | |
| Yes | 343 | 92.2 | 29 | 7.8 | |
| Adult respiratory distress syndrome | < .001 | ||||
| No | 2,608,686 | 97.6 | 65,115 | 2.4 | |
| Yes | 90 | 48.7 | 95 | 51.4 | |
| Postpartum hemorrhage | < .001 | ||||
| 2,535,267 | 97.6 | 61,374 | 2.4 | ||
| 73,509 | 95.0 | 3836 | 5.0 | ||
| Transfusion | < .001 | ||||
| 2,597,084 | 97.6 | 63,377 | 2.4 | ||
| 11,692 | 86.5 | 1833 | 13.5 | ||
| Heart disease | < .001 | ||||
| No | 2,593,200 | 97.6 | 64,179 | 2.4 | |
| Yes | 15,576 | 93.8 | 1031 | 6.2 | |
| Sickle cell | < .001 | ||||
| No | 2,608,214 | 97.6 | 65,163 | 2.4 | |
| Yes | 562 | 92.3 | 47 | 7.7 | |
| Systemic lupus erythematosus | < .001 | ||||
| No | 2,606,838 | 97.6 | 64,876 | 2.4 | |
| Yes | 1938 | 85.3 | 334 | 14.7 | |
| Renal disease | < .001 | ||||
| No | 2,608,131 | 97.6 | 64,982 | 2.4 | |
| Yes | 645 | 73.9 | 228 | 26.1 | |
| Hypercoagulability | < .001 | ||||
| No | 2,606,331 | 97.7 | 61,423 | 2.3 | |
| Yes | 2445 | 39.2 | 3787 | 60.8 | |
| Surgical | < .001 | ||||
| No | 2,608,428 | 97.6 | 64,808 | 2.4 | |
| Yes | 348 | 46.4 | 402 | 53.6 | |
| Cancer | < .001 | ||||
| No | 2,608,509 | 97.6 | 65,188 | 2.4 | |
| Yes | 267 | 92.4 | 22 | 7.6 | |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
