Thromboembolism is a leading cause of maternal mortality. A World Health Organization systematic review of maternal death determined embolism to be responsible for 14.9% of maternal deaths in developed countries.1 Data from the United Kingdom's Confidential Enquiries into Maternal Death found thromboembolism to be the cause of 31.1% of deaths directly related to pregnancy between 2003 and 2005.2 Women who undergo a cesarean delivery are at high risk for venous thromboembolism (defined as deep vein thrombosis and pulmonary embolism), particularly if other risk factors such as obesity, cesarean delivery during labor, and preeclampsia are present.3–6
Thromboembolism prophylaxis may reduce maternal mortality from thromboembolic disease. A successful national strategy in the United Kingdom that included postcesarean pharmacologic prophylaxis reduced death from this cause by more than half. From 2003–2005 to 2006–2008, maternal deaths from thromboembolism decreased from 1.94 to 0.79 maternal deaths per 100,000 deliveries.2 A report from the Hospital Corporation of America identified postcesarean thromboprophylaxis as the sole means of systematically reducing maternal mortality in the United States.5 Decision analyses support the benefit of universal mechanical thromboembolism prophylaxis after a cesarean delivery,7 and economic analyses have found routine postcesarean prophylaxis to be cost-effective.8 The American College of Obstetricians and Gynecologists subsequently has recommended routine placement of pneumatic compression devices before cesarean delivery.9 For patients with additional risk factors, postcesarean heparin prophylaxis may offer benefit.10 The American College of Chest Physicians and national guideline recommendations from the United Kingdom, Sweden, and Canada recommend consideration of postcesarean heparin prophylaxis for high-risk patients.11–14
In the absence of a comprehensive national strategy, thromboembolism has remained a leading cause of maternal death in the United States. Although thromboembolism remained constant at 10.2% of maternal deaths for the time periods 1991–1997 and 1998–2005, the overall risk of death from this cause increased from 1.17 to 1.48 deaths per 100,000 women.15 It is unclear to what degree evidence-based interventions have been incorporated into clinical practice. There are often lengthy delays in clinical adoption of evidence-based medicine.16 The purpose of this study was to characterize contemporary practice patterns for postcesarean thromboembolism prophylaxis and determine whether opportunities to substantially decrease maternal mortality and morbidity in this clinical setting are being missed.
PATIENTS AND METHODS
The Perspective (Premier) is a voluntary, fee-supported database that captures hospitalization data from more than 600 acute care hospitals in the United States and was used for the analysis. The database is maintained by Premier Incorporated (Charlotte, North Carolina), and included are patient demographic information, disease and procedure codes, and hospital and health care provider characteristics. Additionally, the database contains all billed services such as medications, devices, laboratory tests, and radiologic imaging. Data undergo a rigorous quality control process including 95 separate quality assurance and data validation checks before being used for research.17 For each hospital included in the data set, 100% of discharge data are included. The hospitalization database has been used in numerous outcomes studies18–20 including evaluations of postsurgical thromboprophylaxis.21–26 In 2006, approximately 15% of all hospitalizations within the United States (almost 5.5 million hospital discharges) were captured in the hospitalization database.18 All data were deidentified and the analysis was approved by the Columbia University institutional review board.
We analyzed patients who underwent cesarean delivery between January 2003 and March 2010. Patients were identified using an enhanced methodology to capture delivery hospitalizations based on billing, procedure, and diagnosis-related group codes.27 This cohort included only patient hospitalizations in which the day of cesarean delivery relative to the hospitalization was known. January 2003 was used as a start date for this analysis because that is the earliest date within the data set that allows tracking of medication and device administration in relation to day of surgery.
The primary outcome of interest was the use of venous thromboembolism prophylaxis. Venous thromboembolism prophylaxis was classified as none, mechanical, pharmacologic, or combination pharmacologic and mechanical. Patients who received either graduated compression stockings or intermittent pneumatic compression were coded as receiving mechanical prophylaxis. Patients receiving unfractionated heparin, low-molecular-weight heparin (including enoxaparin sodium, tinzaparin sodium, or dalteparin sodium), or fondaparinux sodium were classified as having received pharmacologic prophylaxis. Women who received mechanical prophylaxis in combination with pharmacologic prophylaxis were included in the combination prophylaxis group. Patients were classified as having received mechanical prophylaxis if they received an appropriate device any hospital day up until postoperative day one. Patients were classified as having received pharmacologic prophylaxis if they received an appropriate drug starting postoperative day zero or one.
Epidemiologic literature3,4,6,14,28–40 was reviewed to identify relevant medical, surgical, and obstetric risk factors associated with postcesarean thromboembolism. Through an iterative process, clinical risk factors 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 compared with rural), teaching status (teaching compared with nonteaching), delivery volume (by tertile), geographic region (Midwest, Northeast, South, West), and hospital size based on bed number (fewer than 400 beds, 400–600 beds, greater than 600 beds). Patient demographics included age, race, year of hospitalization, and marital status. Comorbidity was estimated using the Elixhauser comorbidity index, which combines comorbid conditions based on International Classification of Diseases coding into an overall measure of medical comorbidity that is used in large administrative data sets.41
The association between thromboembolism prophylaxis and clinical and demographic variables were compared using the χ2 test. To account for the influence of clinical and demographic factors on use of prophylaxis, we developed mixed effects regression models to examine use of prophylaxis. In addition to clinical and demographic characteristics, the models included hospital ID as a random intercept term to account for hospital-level clustering. We first modeled receipt of any prophylaxis (mechanical, pharmacologic, or combination) in the entire cohort. A second model was developed to examine use of pharmacologic prophylaxis (either pharmacologic prophylaxis alone or combination prophylaxis) in the subset of women who received some form of prophylaxis. Results are reported as risk ratio with 95% confidence interval (CI).
A series of sensitivity analyses were performed in which prophylaxis was considered to have been given if any dose of pharmacologic prophylaxis was administered within 48 or 72 hours of surgery. All analyses were performed with SAS 9.3.
A total of 1,263,205 women hospitalized for cesarean delivery were identified and included in the analysis. A total of 955,787 women (75.7%) received no thromboembolism prophylaxis, whereas 278,669 (22.1%) received mechanical prophylaxis alone, 16,639 (1.3%) received pharmacologic prophylaxis, and 12,110 (1.0%) received combination prophylaxis. Figure 1 displays the use of prophylaxis based on the year of hospital admission. Use of prophylaxis increased over time from 8.4% in 2003 to 41.6% in 2010 (P<.001). This increase was primarily the result of increased utilization of mechanical as opposed to pharmacologic and combination prophylaxis. Pharmacologic and combination prophylaxis increased 0.7% and 1.1%, respectively, between 2003 and 2010.
Table 1 displays the univariable analysis of demographic factors associated with prophylaxis. Prophylaxis varied significantly by geographic region: 32.9% of women in the Northeast received prophylaxis compared with 16.3% of women in the West, 20.2% in the Midwest, and 26.2% in the South (P<.001). Older age, black race, rural hospital location, and intermediate hospital size were all factors associated with increased use of prophylaxis (P<.001 for all). Although univariable analysis of clinical risk factors for thromboembolism demonstrated higher rates of prophylaxis (analysis not shown) for many risk factors, the increased use of prophylaxis was marginal.
The first results column of Table 2 displays a multivariable model of factors associated with use of any prophylaxis. All risk factors included in the multivariable model are listed in Table 2. Year of delivery was significantly associated with probability of receiving prophylaxis: women delivering in 2010 were more likely to receive prophylaxis than women delivering in 2003 (rate ratio [RR] 4.9, 95% CI 4.8–5.1). Geographical region remained significant in the multivariable analysis: rates of prophylaxis were lower in the West compared with the Northeast (RR 0.4, 95% CI 0.2–0.6). Medical risk factors associated with increased use of thromboprophylaxis included: history of thromboembolism (RR 2.3, 95% CI 2.2–2.4), concurrent surgical procedure (RR 2.1, 95% CI 1.9–2.3), and hypercoagulability (RR 2.14, 95% CI 2.06–2.22). Medical risk factors with more modest increased thromboembolism risk such as heart disease, systemic lupus erythematosus, and renal disease were associated with lesser increases in use of prophylaxis. Common obstetric risk factors for thromboembolism such as multiparity (RR 0.98, 95% CI 0.93–1.04), preeclampsia (RR 1.07, 95% CI 1.06–1.08), and cesarean delivery after labor (RR 0.94, 95% CI 0.93–0.95) were not associated with major differences in probability of prophylaxis. Age and many other medical risk factors were marginally associated with different rates of prophylaxis.
The second results column of Table 2 demonstrates use of pharmacologic and combined prophylaxis compared with mechanical prophylaxis alone. High-risk conditions such as cancer, prior thromboembolism, and hypercoagulable state as well as medical conditions associated with thromboembolism risk were associated with increased rates for using pharmacologic and combined prophylaxis compared with mechanical prophylaxis. These findings were largely unchanged in sensitivity analyses in which the prophylaxis within 48 or 72 hours of surgery was examined.
Although our findings demonstrate increasing use of postcesarean mechanical prophylaxis, at the end of the study period, there was still significant underuse of this recommended means of reducing maternal morbidity and mortality. Although research in obstetric populations is limited, medical and surgical evidence supports mechanical prophylaxis as an effective means to reduce thromboembolism.42,43 Compression stockings and intermittent pneumatic compression devices are rarely contraindicated in obstetric patients and universal postcesarean mechanical prophylaxis is recommended by the American College of Obstetricians and Gynecologists.9
Our findings support the need for clear guidelines and protocols for thromboembolism prophylaxis. Not only did fewer than half of patients receive thromboembolism prophylaxis in 2010, but significant variation based on hospital factors occurred as well, suggesting uneven care quality. Suboptimal prophylaxis rates may be attributable in part to lack of clear indications for prophylaxis. Medical and obstetric risk factors were generally associated with only marginally higher risks of thromboprophylaxis and because multiple variables were tested for some significant associations may be the result of chance. Outside of a personal history of thromboembolism that carries high risk for postpartum recurrence, indications for thromboprophylaxis are not well defined and many risk factors are relatively common. Given the relative complexity of epidemiologic risk factors, a clinical decision regarding prophylaxis may be cumbersome, time-consuming, and complicated for many health care providers. The causes of increased prophylaxis at intermediate-volume and rural hospitals is unclear; however, it possible that hospital size and location may be related to the ability to quickly adopt clinical protocols. Risk assessment tools that simplify decision-making may aid hospitals in providing uniform, high-quality care.
Data from this study showed pharmacologic prophylaxis was used in only a small percentage of patients, much less frequently than would be indicated by British and Scandinavian guidelines.11,12 Although mechanical prophylaxis alone has been demonstrated to reduce deep vein thrombosis, reliance solely on this method for postoperative venous thromboembolism prophylaxis has well-documented shortcomings. First, studies of postoperative patients have demonstrated suboptimal compliance44 and studies evaluating educational programs directed at hospital personnel intended to improve compliance have not shown benefit.45 Second, because there are many devices on the market, and because even with high-risk surgeries, venous thromboembolism events are relatively rare, adequately powered studies validating individual device performance are lacking.46–48 Third, because many risk factors for increased thromboembolism risk such as preeclampsia, cesarean delivery during labor, and obesity are relatively common, many patients may still have a significant residual risk for events even if they receive mechanical prophylaxis. Low use of pharmacologic prophylaxis may be secondary to the fact that although research has found that for high-risk procedures, combination thromboprophylaxis with heparin offers greater risk reduction than mechanical prophylaxis alone,49 randomized trials focusing on effectiveness of pharmacologic prophylaxis after cesarean delivery are underpowered to assess outcomes and are inconclusive.40 The relative benefits of pharmacologic prophylaxis may be greater than suggested by prior decision analysis work7 because of the lower incidence of heparin-induced thrombocytopenia associated with low-molecular-weight heparins.50
Although our study benefits from the inclusion of a large cohort of patients in diverse geographic and hospital settings, we recognize important limitations. First, the primary purpose of claims data is billing and we cannot exclude the possibility of prophylaxis misclassification in some patients. However, this number is likely small because the hospitalization database has been validated in several studies that examine drug and device use.49,51,52 Second, although it is possible to estimate the number of patients who received prophylaxis, we are not able to examine the quality of prophylaxis. As already noted, compliance with mechanical prophylaxis among patients is poor. Third, because administrative data do not allow for direct determination of patient attributes such as body mass index, it is likely that the prevalence of some clinical risk factors is underestimated. Additionally, reported use of pharmacologic prophylaxis does not ensure that the proper dose of the drug was administered throughout the hospitalization. Finally, like with any observational study, we are unable to capture individual patient and physician preferences that likely influence prophylaxis.
In conclusion, as of 2010, fewer than half of women who underwent cesarean delivery received venous thromboembolism prophylaxis with significant variation noted despite an increased adoption of prophylaxis overall since 2003. Although patients at high risk for venous thromboembolism were more likely to receive mechanical prophylaxis, pharmacologic prophylaxis was rarely used. Thromboembolism prophylaxis is underused and represents a major opportunity to reduce maternal morbidity and mortality. Risk assessment tools and thromboprophylaxis guidelines among other validated measures are needed to assure high-quality, uniform care. Comparative effectiveness research is needed to further characterize the safety, efficacy, and cost of different prophylactic devices, medications, and regimens.
1. Khan KS, Wojdyla D, Say L, Gulmezoglu AM, Van Look PF. WHO analysis of causes of maternal death: a systematic review. Lancet 2006;367:1066–74.
2. Cantwell R, Clutton-Brock T, Cooper G, Dawson A, Drife J, Garrod D, et al.. Saving Mothers' Lives: Reviewing maternal deaths to make motherhood safer: 2006–2008. The Eighth Report of the Confidential Enquiries into Maternal Deaths in the United Kingdom. BJOG 2011;118(suppl 1):1–203.
3. Jacobsen AF, Skjeldestad FE, Sandset PM. Incidence and risk patterns of venous thromboembolism in pregnancy and puerperium—a register-based case-control study. Am J Obstet Gynecol 2008;198:233.e1–7.
4. 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.
5. Clark SL, Belfort MA, Dildy GA, Herbst MA, Meyers JA, Hankins GD. Maternal death in the 21st century: causes, prevention, and relationship to cesarean delivery. Am J Obstet Gynecol 2008;199:36.e1–5.
6. Sultan AA, Tata LJ, West J, Fiaschi L, Fleming KM, Nelson-Piercy C, et al.. Risk factors for first venous thromboembolism around pregnancy: a population-based cohort study from the United Kingdom. Blood 2013;121:3953–61.
7. Quinones JN, James DN, Stamilio DM, Cleary KL, Macones GA. Thromboprophylaxis after cesarean delivery: a decision analysis. Obstet Gynecol 2005;106:733–40.
8. Casele H, Grobman WA. Cost-effectiveness of thromboprophylaxis with intermittent pneumatic compression at cesarean delivery. Obstet Gynecol 2006;108:535–40.
9. James A; Committee on Practice Bulletins–Obstetrics. Practice bulletin no. 123: thromboembolism in pregnancy. Obstet Gynecol 2011;118:718–29.
10. Blondon M, Perrier A, Nendaz M, Righini M, Boehlen F, Boulvain M. Thromboprophylaxis with low-molecular-weight heparin after cesarean delivery. Thromb Haemost 2010;103:129–37.
11. Lindqvist PG, Hellgren M. Obstetric thromboprophylaxis: the Swedish guidelines. Adv Hematol 2011;2011:157483.
13. Kent N, Leduc L, Crane J, Farine D, Hodges S, Reid GJ, et al.. Prevention and treatment of venous thromboembolism (VTE) in obstetrics. J SOGC 2000;22:736–49.
14. Bates SM, Greer IA, Pabinger I, Sofaer S, Hirsh J; American College of Chest Physicians. Venous thromboembolism, thrombophilia, antithrombotic therapy, and pregnancy: American College of Chest physicians evidence-based clinical practice guidelines (8th Edition). Chest 2008;133(suppl):844S–86S.
15. Callaghan WM. Overview of maternal mortality in the United States. Semin Perinatol 2012;36:2–6.
16. Committee on Quality of Health Care in America. Crossing the quality chasm: a new health system for the 21st century. Washington (DC): Institute of Medicine; 2001.
17. Stulberg JJ, Delaney CP, Neuhauser DV, Aron DC, Fu P, Koroukian SM. Adherence to surgical care improvement project measures and the association with postoperative infections. JAMA 2010;303:2479–85.
18. Lindenauer PK, Pekow PS, Lahti MC, Lee Y, Benjamin EM, Rothberg MB. Association of corticosteroid dose and route of administration with risk of treatment failure in acute exacerbation of chronic obstructive pulmonary disease. JAMA 2010;303:2359–67.
19. Rothberg MB, Pekow PS, Lahti M, Brody O, Skiest DJ, Lindenauer PK. Antibiotic therapy and treatment failure in patients hospitalized for acute exacerbations of chronic obstructive pulmonary disease. JAMA 2010;303:2035–42.
20. Lindenauer PK, Pekow P, Wang K, Mamidi DK, Gutierrez B, Benjamin EM. Perioperative beta-blocker therapy and mortality after major noncardiac surgery. N Engl J Med 2005;353:349–61.
21. Fang MC, Maselli J, Lurie JD, Lindenauer PK, Pekow PS, Auerbach AD. Use and outcomes of venous thromboembolism prophylaxis after spinal fusion surgery. J Thromb Haemost 2011;9:1318–25.
22. Deitelzweig SB, Lin J, Hussein M, Battleman D. Are surgical patients at risk of venous thromboembolism currently meeting the Surgical Care Improvement Project performance measure for appropriate and timely prophylaxis? J Thromb Thrombolysis 2010;30:55–66.
23. Vekeman F, LaMori JC, Laliberté F, Nutescu E, Duh MS, Bookhart BK, et al.. In-hospital risk of venous thromboembolism and bleeding and associated costs for patients undergoing total hip or knee arthroplasty. J Med Econ 2012;15:644–53.
24. Ritch JM, Kim JH, Lewin SN, Burke WM, Sun X, Herzog TJ, et al.. Venous thromboembolism and use of prophylaxis among women undergoing laparoscopic hysterectomy. Obstet Gynecol 2011;117:1367–74.
25. Wright JD, Lewin SN, Shah M, Burke WM, Lee SM, Sun X, et al.. Quality of venous thromboembolism prophylaxis in patients undergoing oncologic surgery. Ann Surg 2011;253:1140–6.
26. Wright JD, Hershman DL, Shah M, Burke WM, Sun X, Neugut AI. Quality of perioperative venous thromboembolism prophylaxis in gynecologic surgery. Obstet Gynecol 2011;118:978–86.
27. Kuklina EV, Whiteman MK, Hillis SD, Jamieson DJ, Meikle SF, Posner SF. An enhanced method for identifying obstetric deliveries: implications for estimating maternal morbidity. Matern Child Health J 2008;12:469–77.
28. Chan WS. The ‘ART’ of thrombosis: a review of arterial and venous thrombosis in assisted reproductive technology. Curr Opin Obstet Gynecol 2009;21:207–18.
29. Danilenko-Dixon DR, Heit JA, Silverstein MD, Yawn BP, Petterson TM, Lohse CM, et al.. Risk factors for deep vein thrombosis and pulmonary embolism during pregnancy or postpartum: a population-based, case-control study. Am J Obstet Gynecol 2001;184:104–10.
30. Dargaud Y, Rugeri L, Vergnes MC, Arnuti B, Miranda P, Negrier C, et al.. A risk score for the management of pregnant women with increased risk of venous thromboembolism: a multicentre prospective study. Br J Haematol 2009;145:825–35.
31. Goldhaber SZ. Risk factors for venous thromboembolism. J Am Coll Cardiol 2010;56:1–7.
32. Jacobsen AF, Skjeldestad FE, Sandset PM. Ante- and postnatal risk factors of venous thrombosis: a hospital-based case-control study. J Thromb Haemost 2008;6:905–12.
33. James AH. Thromboembolism in pregnancy: recurrence risks, prevention and management. Curr Opin Obstet Gynecol 2008;20:550–6.
34. Larsen TB, Sorensen HT, Gislum M, Johnsen SP. Maternal smoking, obesity, and risk of venous thromboembolism during pregnancy and the puerperium: a population-based nested case-control study. Thromb Res 2007;120:505–9.
35. Lindqvist P, Dahlback B, Marsal K. Thrombotic risk during pregnancy: a population study. Obstet Gynecol 1999;94:595–9.
36. Lindqvist PG, Torsson J, Almqvist A, Bjorgell O. Postpartum thromboembolism: severe events might be preventable using a new risk score model. Vasc Health Risk Manag 2008;4:1081–7.
37. Lussana F, Coppens M, Cattaneo M, Middeldorp S. Pregnancy-related venous thromboembolism: risk and the effect of thromboprophylaxis. Thromb Res 2012;129:673–80.
38. Simpson EL, Lawrenson RA, Nightingale AL, Farmer RD. Venous thromboembolism in pregnancy and the puerperium: incidence and additional risk factors from a London perinatal database. BJOG 2001;108:56–60.
39. Sultan AA, West J, Tata LJ, Fleming KM, Nelson-Piercy C, Grainge MJ. Risk of first venous thromboembolism in and around pregnancy: a population-based cohort study. Br J Haematol 2012;156:366–73.
40. Tooher R, Gates S, Dowswell T, Davis LJ. Prophylaxis for venous thromboembolic disease in pregnancy and the early postnatal period. The Cochrane Database of Systematic Reviews 2010, Issue 5. Art. No.: CD001689. DOI: 10.1002/14651858.CD001689.pub2.
41. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care 1998;36:8–27.
42. Sachdeva A, Dalton M, Amaragiri SV, Lees T. Elastic compression stockings for prevention of deep vein thrombosis. The Cochrane Database of Systematic Reviews 2010, Issue 7. Art. No.: CD001484. DOI: 10.1002/14651858.CD001484.pub2.
43. Lippi G, Favaloro EJ, Cervellin G. Prevention of venous thromboembolism: focus on mechanical prophylaxis. Semin Thromb Hemost 2011;37:237–51.
44. Bockheim HM, McAllen KJ, Baker R, Barletta JF. Mechanical prophylaxis to prevent venous thromboembolism in surgical patients: a prospective trial evaluating compliance. J Crit Care 2009;24:192–6.
45. Macatangay C, Todd SR, Tyroch AH. Thromboembolic prophylaxis with intermittent pneumatic compression devices in trauma patients: a false sense of security? J Trauma 2008;15:12–5.
46. Feist WR, Andrade D, Nass L. Problems with measuring compression device performance in preventing deep vein thrombosis. Thromb Res 2011;128:207–9.
47. Morris RJ, Woodcock JP. Intermittent pneumatic compression or graduated compression stockings for deep vein thrombosis prophylaxis? A systematic review of direct clinical comparisons. Ann Surg 2010;251:393–6.
48. Zhao JM, He ML, Xiao ZM, Li TS, Wu H, Jiang H. Different types of intermittent pneumatic compression devices for preventing venous thromboembolism in patients after total hip replacement. The Cochrane Database of Systematic Reviews 2012, Issue 11. Art. No.: CD009543. DOI: 10.1002/14651858.CD009543.pub2.
49. Lindenauer PK, Pekow P, Gao S, Crawford AS, Gutierrez B, Benjamin EM. Quality of care for patients hospitalized for acute exacerbations of chronic obstructive pulmonary disease. Ann Intern Med 2006;144:894–903.
50. Junqueira DR, Perini E, Penholati RR, Carvalho MG. Unfractionated heparin versus low molecular weight heparin for avoiding heparin-induced thrombocytopenia in postoperative patients. The Cochrane Database of Systematic Reviews 2012, Issue 9. Art. No.: CD007557. DOI: 10.1002/14651858.CD007557.pub2.
51. Amin AN, Stemkowski S, Lin J, Yang G. Preventing venous thromboembolism in US hospitals: are surgical patients receiving appropriate prophylaxis? Thromb Haemost 2008;99:796–7.
52. Amin AN, Stemkowski S, Lin J, Yang G. Inpatient thromboprophylaxis use in U.S. hospitals: adherence to the seventh American College of Chest Physicians' recommendations for at-risk medical and surgical patients. J Hosp Med 2009;4:e15–21.