Postpartum hemorrhage is a leading cause of maternal morbidity and mortality.1–3 To decrease maternal harm from postpartum hemorrhage, the American College of Obstetricians and Gynecologists (ACOG)4 and the California Maternal Quality Care Collaborative (CMQCC)5 implemented obstetric safety bundles. Critical to these bundles is assessment of hemorrhage risk before delivery, which allows advanced planning and unit readiness (eg, type and crossmatch, hemorrhage cart with supplies). Yet, certain risk factors for postpartum hemorrhage, such as platelet count, are poorly studied.
Platelet count thresholds for risk of postpartum hemorrhage are variable. The American College of Obstetricians and Gynecologists classifies platelet count less than 70 k/microliter, and CMQCC less than 100 k/microliter, as a high risk factor for postpartum hemorrhage. Although such low platelet counts are associated with postpartum hemorrhage,6,7 they often indicate an underlying disorder, such as immune thrombocytopenia or preeclampsia with severe features. In women with uncomplicated pregnancies, the platelet count is rarely less than 70 k/microliter (0.1%), and infrequently less than 100 k/microliter (1%).8 Mild thrombocytopenia (platelet count 100–149 k/microliter) is present in up to 10% of all delivering women,8–11 but its association with postpartum hemorrhage is unclear.
We sought to assess whether mild thrombocytopenia is associated with postpartum hemorrhage. To address this question, we evaluated nulliparous women with term, singleton, vertex pregnancies undergoing labor.
We performed a retrospective cohort study of all nulliparous women with term, singleton, vertex pregnancies who delivered at Cedars-Sinai Medical Center between August 2016 and September 2017. This study was approved by Cedars-Sinai institutional review board. Women undergoing cesarean delivery without labor and those without a platelet count result before delivery owing to platelet clumping or precipitous delivery were excluded.
Our primary study group included women with mild thrombocytopenia, defined by platelet count 100–149 k/microliter. The primary comparator group included those with normal platelet count 150 k/microliter or greater. Women with severe thrombocytopenia (platelet count less than 100 k/microliter) were included for descriptive purposes but excluded from all analyses. Platelet count (k/microliter) data, as well as hemoglobin (g/dL), were abstracted from the first complete blood count (CBC) on admission for labor and before delivery.
The primary outcome was postpartum hemorrhage, based on data extracted from the electronic medical record, using International Classification of Diseases, Tenth Revision (ICD-10) codes12 and the hospital discharge problem list. The ICD-10 codes used for postpartum hemorrhage included O72.0, O72.1, O72.2, O72.3, and O75.89. Random chart audit was performed in 10% of cases to validate the ICD-10 diagnosis of postpartum hemorrhage. Secondary outcomes included use of Methergine (methylergonovine maleate) or Hemabate (carboprost tromethamine) as uterotonic agents at delivery, total estimated blood loss 1,000 mL or greater, and blood transfusion. Data on uterotonic agents and blood transfusion were abstracted from medication administration records, and estimated blood loss from delivery or operative reports.
The diagnosis of postpartum hemorrhage and determination of total estimated blood loss was made by the delivery provider. In the study period, providers used ACOG's reVITALize Obstetric Data Definitions for postpartum hemorrhage, published in 2014.13 Using this definition, postpartum hemorrhage was defined by total estimated blood loss 1,000 mL or greater or any blood loss accompanied by signs or symptoms of hypovolemia after birth (eg, hypotension, tachycardia). There was no standardized approach to quantify estimated blood loss. Oxytocin was routinely employed for active management of third stage of labor (30 units in 500 mL normal saline, intravenous infusion), and methylergonovine maleate and carboprost tromethamine were recommended as first-line agents to treat postpartum hemorrhage.
Demographic data, including maternal age, race–ethnicity, and body mass index (BMI, calculated as weight in kilograms divided by height in meters squared), were abstracted from the electronic medical record, as were data on morbidities such as pregestational or gestational diabetes and gestational hypertension or preeclampsia. We also abstracted intrapartum data on total length of labor (defined as time from admission to delivery), length of second stage, magnesium use, chorioamnionitis, third- or fourth-degree perineal lacerations, and mode of delivery.
The study was powered to detect a 60% difference in the risk of postpartum hemorrhage between groups, assuming a 10% incidence of mild thrombocytopenia and 9% rate of postpartum hemorrhage in the study population, with α=0.05 and β=0.80. For statistical analyses we used χ2 or Fisher exact test for categorical variables, Student's t test for continuous variables, and rank sum test for nonparametric test of medians. Multivariable logistic regression was performed to adjust for potential confounders of the association between platelet count and postpartum hemorrhage. Analyses were performed using Stata 15.1, with significance determined by α <0.05 for the primary outcome and α <0.01 for secondary outcomes.
A total of 3,166 nulliparous women with term, singleton, vertex pregnancies were delivered at our institution between August 2016 and September 2017. Of these, 2,858 met study criteria and were included, as shown in the study flow diagram (Fig. 1). We identified 2,579 (90.2%) women with normal platelet count (150 k/microliter or greater), 266 (9.3%) with mild thrombocytopenia (platelet count 100–149 k/microliter), and 13 (0.5%) with severe thrombocytopenia (platelet count less than 100 k/microliter).
In the group with normal platelet count the median value was 216 k/microliter (interquartile range 186–251 k/microliter), and in the mild thrombocytopenia group the median value was 135 k/microliter (interquartile range 124–143 k/microliter). Compared with women with normal platelet, those with mild thrombocytopenia had lower BMI and were more often white and less often Asian (Table 1). Other antepartum factors (gestational and pregestational diabetes, gestational hypertension and preeclampsia) and intrapartum factors (magnesium sulfate use, chorioamnionitis, length of labor, length of second stage, third- or fourth-degree perineal lacerations, cesarean delivery) were not different between groups.
The primary outcome of postpartum hemorrhage was significantly more common in women with mild thrombocytopenia compared with those with normal platelet count (16.9% vs 8.5%, P<.001) (Table 2). Secondary outcomes, including use of methylergonovine maleate or carboprost tromethamine or both, and estimated blood loss 1,000 mL or greater, were also more common in women with mild thrombocytopenia compared with those with normal platelet count (Table 2). However, blood transfusion was no different between groups.
When restricting the analysis to women with vaginal deliveries only, postpartum hemorrhage remained more common in those with mild thrombocytopenia (16.0% vs 7.9%, P<.001; odds ratio [OR] 2.1, 95% CI 1.5–3.2, P<.001). Use of methylergonovine maleate or carboprost tromethamine or both was also more common in the mild thrombocytopenia group (Table 3). However, the association between mild thrombocytopenia and estimated blood loss 1,000 mL or greater was no longer seen. When restricting the analysis to women with vaginal delivery and without gestational hypertension or preeclampsia, the findings were similar (Appendix 1, available online at http://links.lww.com/AOG/B836). Mild thrombocytopenia remained associated with postpartum hemorrhage and use of methylergonovine maleate or carboprost tromethamine or both, but not estimated blood loss 1,000 mL or greater.
The association between other antepartum and intrapartum risk factors and postpartum hemorrhage is shown in Table 4. Body mass index, race–ethnicity, gestational hypertension or preeclampsia, length of labor, chorioamnionitis, and cesarean delivery were associated with postpartum hemorrhage in univariable analysis and were thus assessed as potential confounders. After multivariable logistic regression, the association between mild thrombocytopenia and postpartum hemorrhage persisted after stepwise adjustment for antepartum factors (adjusted OR 2.2, 95% CI 1.5–3.4, P<.001) and then intrapartum factors (adjusted OR 2.2, 95% CI 1.5–3.2, P<.001) (Table 5).
Among nulliparous women with term, singleton, vertex pregnancies undergoing labor, mild thrombocytopenia (platelet count 100–149 k/microliter) was independently associated with postpartum hemorrhage. Specifically, those with mild thrombocytopenia had a twofold greater likelihood of postpartum hemorrhage compared with those with normal platelet count. Such women were also more likely to receive uterotonics at delivery or have an estimated blood loss greater than one liter.
The intrapartum platelet count threshold, below which the risk of postpartum hemorrhage increases, is unclear. In their toolkits for obstetric hemorrhage, ACOG and CMQCC classify platelet count less than 70 k/microliter and less than 100 k/microliter, respectively, as high risk for postpartum hemorrhage.4,5 For women meeting these criteria, patient and unit readiness may include placement of a large-bore intravenous line, type and crossmatch, preparation of a hemorrhage cart, and active management of third stage of labor.4,5,14 However, in our cohort of 2,858 nulliparous women with term, singleton vertex pregnancies, only 13 (0.5%) had platelet count less than 100 k/microliter and only four (0.1%) had platelet count less than 70 k/microliter. Meanwhile, 9.3% of women had mild thrombocytopenia, which carries twofold greater odds of postpartum hemorrhage.
Our findings are consistent with a study by Carlson et al.7 Among 54,597 women delivering by cesarean or vaginal birth after cesarean, Carlson et al found that mild thrombocytopenia was associated with both laboratory and clinical evidence of hemorrhage. Although our data, and the data from Carlson et al, show that mild thrombocytopenia is associated with postpartum hemorrhage, there is limited guidance on platelet count and hemorrhage risk in pregnant women. Practice guidelines often focus on platelet count in women with medical conditions such as immune thrombocytopenia or preeclampsia, or those undergoing surgery or neuraxial anesthesia.15–20 In these scenarios, platelet counts less than 70–100 k/microliter raise concern for increased bleeding. However, the mechanisms linking platelet count to postpartum hemorrhage are unclear, and mild reductions in platelet count, as seen in our study, may impart an increased risk for postpartum hemorrhage.
The strengths of our study include its emphasis on mild thrombocytopenia and risk of postpartum hemorrhage, with exclusion of those with severe thrombocytopenia. By excluding those with severe thrombocytopenia, we limited the effect of serious medical conditions such as immune thrombocytopenia, systemic lupus erythematosus, and hemolysis, elevated liver enzymes, low platelet count syndrome. We used a relatively homogeneous population of nulliparous women with term, singleton vertex pregnancies, which minimized the effect of potential confounders. Moreover, our results were robust despite multivariable regression and sensitivity analysis, with the exclusion of cesarean deliveries and gestational hypertension or preeclampsia.
Our study is not without limitations. Patients were identified by ICD-10 codes, which may not capture all cases of postpartum hemorrhage. Butwick et al21 found that ICD-9 codes had 97% specificity for postpartum hemorrhage, but only 28% sensitivity. Although identification of postpartum hemorrhage cases may be better with ICD-10 coding, we performed chart audit to validate postpartum hemorrhage diagnoses and assessed use of uterotonic agents and total blood loss greater than 1,000 mL to support the results of our primary outcome (postpartum hemorrhage). This was a single center retrospective study, which may limit overall generalizability. However, mild thrombocytopenia was associated with BMI and race–ethnicity, which may aid in applicability to other populations. We did not find a difference in blood transfusion between groups, which may be a result of the small number of patients requiring transfusion in our study. The mild thrombocytopenia group received uterotonic agents more frequently, which may have decreased need for blood transfusion.
We conclude that, among nulliparous women with term, singleton, vertex pregnancies undergoing labor, those with mild thrombocytopenia (platelet count 100–149 k/microliter) have a twofold greater likelihood of postpartum hemorrhage compared with those with normal platelet count. A platelet count threshold of 150 k/microliter at the time of delivery may identify more women at increased risk for postpartum hemorrhage, but additional studies are needed to evaluate whether mild thrombocytopenia should be included in risk stratification toolkits for postpartum hemorrhage.
1. Bateman BT, Berman MF, Riley LE, Leffert LR. The epidemiology of postpartum hemorrhage in a large, nationwide sample of deliveries. Anesth Analg 2010;110:1368–73.
2. Mehrabadi A, Hutcheon JA, Lee L, Liston RM, Joseph KS. Trends in postpartum hemorrhage from 2000 to 2009: a population-based study. BMC Pregnancy Childbirth 2012;12:108.
3. Kramer MS, Berg C, Abenhaim H, Dahhou M, Rouleau J, Mehrabadi A, et al. Incidence, risk factors, and temporal trends in severe postpartum hemorrhage. Am J Obstet Gynecol 2013;209:449.e1–7.
4. American College of Obstetrics and Gynecologists. District II safe motherhood initiative: maternal safety bundle for obstetric hemorrhage. Available at: https://www.acog.org/-/media/project/acog/acogorg/files/forms/districts/smi-ob-hemorrhage-bundle-slides.pdf
. Retrieved April 27, 2020.
5. Lyndon A, Lagrew D, Shields L, Main E, Cape V. Improving health care response to obstetric hemorrhage. California Maternal Quality Care Collaborative Toolkit to Transform Maternity Care. Stanford, CA: California Maternal Quality Care Collaborative; 2015.
6. Simon L, Santi TM, Sacquin P, Hamza J. Pre-anaesthetic assessment of coagulation abnormalities in obstetric patients: usefulness, timing and clinical implications. Br J Anaesth 1997;78:678–83.
7. Carlson LM, Dotters-Katz SK, Smid MC, Manuck TA. How low is too low? Postpartum hemorrhage risk among women with thrombocytopenia. Am J Perinatol 2017;34:1135–41.
8. Reese JA, Peck JD, Deschamps DR, McIntosh JJ, Knudtson EJ, Terrell DR, et al. Platelet counts during pregnancy. N Engl J Med 2018;379:32–43.
9. Boehlen F, Hohlfeld P, Extermann P, de Moerloose P. Maternal antiplatelet antibodies in predicting risk of neonatal thrombocytopenia. Obstet Gynecol 1999;93:169–73.
10. Burrows RF, Kelton JG. Fetal thrombocytopenia and its relation to maternal thrombocytopenia. N Engl J Med 1993;329:1463–6.
11. Boehlen F, Hohlfeld P, Extermann P, Perneger TV, de Moerloose P. Platelet count at term pregnancy: a reappraisal of the threshold. Obstet Gynecol 2000;95:29–33.
12. ICD10 data. Available at: https://www.icd10data.com
. Retrieved December 12, 2019.
13. Menard MK, Main EK, Currigan SM. Executive summary of the reVITALize initiative: standardizing obstetric data definitions. Obstet Gynecol 2014;124:150–3.
14. Main EK, Goffman D, Scavone BM, Low LK, Bingham D, Fontaine PL, et al. National partnership for maternal safety: consensus bundle on obstetric hemorrhage. Obstet Gynecol 2015;126:155–62.
15. Kaufman RM, Djulbegovic B, Gernsheimer T, Kleinman S, Tinmouth AT, Capocelli KE, et al. Platelet transfusion: a clinical practice guideline from the AABB. Ann Intern Med 2015;162:205–13.
16. Practice guidelines for obstetric anesthesia: an updated report by the American society of anesthesiologists task force on obstetric anesthesia and the society for obstetric anesthesia and perinatology. Anesthesiology 2016;124:270–300.
17. Lee LO, Bateman BT, Kheterpal S, Klumpner TT, Housey M, Aziz MF, et al. Risk of epidural hematoma after neuraxial techniques in thrombocytopenic parturients: a report from the multicenter perioperative outcomes group. Anesthesiology 2017;126:1053–63.
18. Levy N, Goren O, Cattan A, Weiniger CF, Matot I. Neuraxial block for delivery among women with low platelet counts: a retrospective analysis. Int J Obstet Anesth 2018;35:4–9.
19. Estcourt LJ, Malouf R, Hopewell S, Doree C, Van Veen J. Use of platelet transfusions prior to lumbar punctures or epidural anaesthesia for the prevention of complications in people with thrombocytopenia. The Cochrane Database Systematic Review 2018, Issue 4. Art. No.: CD011980. doi: 10.1002/14651858.CD011980.pub3.
20. Thrombocytopenia in pregnancy. ACOG Practice Bulletin No. 207. American College of Obstetricians and Gynecologists. Obstet Gynecol 2019;133:e181–93.
21. Butwick AJ, Walsh EM, Kuzniewicz M, Li SX, Escobar GJ. Accuracy of International Classification of Diseases, Ninth Revision, codes for postpartum hemorrhage among women undergoing cesarean delivery. Transfusion 2018;58:998–1005.