Preeclampsia, typically characterized by elevated blood pressure and proteinuria after 20 weeks of pregnancy, is one of the leading causes of maternal–fetal morbidity and mortality.1–10 In industrialized countries, the incidence of preeclampsia is approximately 3–5 per 100 births.4,11 Most industrialized countries have experienced a decline in the incidence of preeclampsia over the past decade, although isolated studies report a temporal increase in frequency.12 Preeclampsia is a serious obstetric condition; in the United States, complications of preeclampsia account for up to 30% of maternal deaths during delivery hospitalization.6,10
Preeclampsia has been increasingly recognized as two different conditions: early-onset preeclampsia occurring at less than 34 weeks of gestation, and late-onset disease occurring at 34 or more weeks of gestation.13–15 Early-onset and late-onset disease have different implications for the fetus and neonate, with an approximately 10-fold higher risk of perinatal death observed among mothers with early-onset disease, and a twofold increased risk evident among mothers with late-onset disease (compared with mothers without preeclampsia).16 However, the consequences of early-onset compared with late-onset preeclampsia on maternal health have not been adequately quantified through population-based studies. We therefore carried out a study to examine the effects of early-onset and late-onset preeclampsia on severe maternal morbidity using population-based data. We also examined temporal trends in early-onset and late-onset preeclampsia, and the trend in severe maternal morbidity associated with these conditions.
MATERIALS AND METHODS
Our data source included all deliveries in Washington State during the period from 2000 to 2008 (n=766,359). Deliveries after a multifetal pregnancy, deliveries with missing information on plurality and gestational age, and deliveries to nonresidents of Washington State were excluded, leaving a study population of 670,120 singleton deliveries (Fig. 1). Information on pregnancy complications, maternal characteristics, and gestational age at delivery was obtained from the Birth Events Record Database, which included birth records of all liveborn neonates and fetal deaths. Information about the diagnosis of preeclampsia was obtained from the Comprehensive Discharge Abstract Database, which included all hospitalizations in Washington State. Women with a diagnosis of preeclampsia (including eclampsia and preeclampsia superimposed on chronic hypertension) during any pregnancy hospitalization were identified from the Comprehensive Discharge Abstract Database using International Classification of Diseases, 9th Revision (ICD-9) diagnostic codes 642.4, 642.5, 642.6, and 642.7. The number of weeks between the hospitalization when the preeclampsia diagnosis was made and the birth hospitalization was calculated based on the Comprehensive Discharge Abstract Database and Birth Events Record Database record linkage. Early-onset preeclampsia was diagnosed if preeclampsia occurred at less than 34 weeks of gestation; late-onset disease included preeclampsia diagnosed at 34 or more weeks of gestation, irrespective of the gestational week at delivery.
Information on maternal death was obtained from hospital discharge data (delivery hospitalization). Maternal morbidity was identified using ICD-9 codes for diagnoses and procedures during the delivery hospitalization (Appendix 1). Severe maternal morbidity was defined as any potentially life-threatening condition, and included several partially overlapping categories:
- Respiratory morbidity (obstetric pulmonary embolism, amniotic fluid embolism, pulmonary collapse, acute pulmonary edema, adult respiratory distress syndrome, acute cor pulmonale, or any pulmonary embolism)
- Cardiovascular morbidity (peripartum cardiomyopathy, cerebrovascular disorders, heart failure, atrial or ventricular fibrillation and flutter, acute myocardial infarction, cardiac arrest, hemopericardium, constrictive pericarditis, cardiac tamponade, malignant essential hypertension with or without kidney involvement, malignant hypertension, arterial embolism or thrombosis of the abdominal or thoracic aorta, aortic aneurism and dissection, or aneurism of pulmonary artery)
- Thromboembolism (including cerebral venous thrombosis, deep phlebothrombosis, arterial embolism and thrombosis of the abdominal aorta, embolism or thrombosis of thoracic aorta)
- Central nervous system morbidity (subarachnoid, intracerebral, subdural, or intracranial hemorrhage, occlusion or stenosis of precerebral or cerebral arteries, cerebral seizure or apoplexy, encephalopathy, hemiplegia, hemiparesis, or other paralysis)
- Blood transfusion (of any blood product)
- Acute renal failure (acute and subacute renal failure or other unspecified renal failure)
- Acute liver failure (acute and subacute liver failure, liver disorders in pregnancy including acute yellow atrophy, icterus gravis, or necrosis)
- Obstetric trauma (rupture of uterus before onset of or during labor, inversion of uterus, laceration of cervix, high vaginal laceration, other injury to pelvic organs, damage to pelvic joints and ligaments, third-degree or fourth-degree perineal tear)
- Other (complications of obstetric procedures, disseminated intravascular coagulation, obstetric shock, infection or sepsis, or thromboembolism).
The composite outcome “severe maternal morbidity” included any of the above-stated conditions. In addition, we examined other indicators of maternal morbidity such as postpartum hemorrhage (defined by a diagnosis of postpartum hemorrhage, obstetric tamponade of uterus and vagina, eg, balloon tamponade, blood transfusion, or a combination of these factors), antepartum hemorrhage (including placental abruption, placenta previa, and hemorrhage from unknown causes), and maternal hospital length of stay. Antepartum and postpartum hemorrhage were not included in the severe maternal morbidity composite outcome, as we could not clearly distinguish between mild and severe forms of these two conditions. Instead, blood transfusion, which clearly identified cases of severe hemorrhage, was used to capture such cases within the composite outcome.
Univariable and multivariable analyses were performed using ongoing pregnancies as the denominator. Although all pregnant women at 20 weeks of gestation were included in the analyses of early-onset preeclampsia, only those with ongoing pregnancies at 34 weeks were included in the analyses of late-onset preeclampsia. Temporal trends were assessed using the Cochran-Armitage test for linear trend in proportions. Logistic regression was used to obtain adjusted odds ratios (aOR) and 95% confidence intervals (95% CI) adjusted for maternal age (less than 20 years and 35 or more years compared with 20–34 years), parity (number of previous live births, none compared with one or more), marital status (single, widowed, or separated compared with married or common-law relationship), education (less than high school compared with high school education or greater), race (non-Hispanic white compared with Hispanic, African American, Native American, and other), smoking during pregnancy (yes or no), diabetes mellitus (yes or no), male sex of the fetus, and congenital anomalies (yes or no). Each specific severe maternal morbidity was modeled separately using logistic regression and the models were adjusted for all the above-mentioned maternal characteristics and clinical risk factors.
All analyses were performed on publicly accessible deidentified data. An exemption from ethics approval was granted by the Department of Social and Health Services, State of Washington. Analyses were carried out using SAS 9.3. A two-tailed P value of less than .05 was considered significant.
All 670,120 women residents of Washington State who had singleton hospital deliveries at 20 weeks of gestation or later between 2000 and 2008 and met study criteria were included in the study. The preeclampsia rate was 3.0 per 100 singleton pregnancies (95% CI 3.0–3.1); the rate of early-onset preeclampsia was 0.3 per 100 singleton pregnancies (95% CI 0.3–0.4), and the rate of late-onset disease was 2.7 per 100 singleton pregnancies (95% CI 2.7–2.8). We observed a 6.9% increase in the incidence of preeclampsia in Washington state, from 2.9 to 3.1 per 100 singleton deliveries between 2000 and 2008 (trend P<.001). A significant 33% increase was observed in early-onset disease (from 0.3 in 2000 to 0.4/100 singleton deliveries in 2008; trend P<.001). Late-onset disease also increased significantly by 3.7% (from 2.7 to 2.8/100 singleton deliveries, trend P<.001; Fig. 2). After adjustment for maternal risk factors (including maternal age, parity, marital status, education, race, smoking during pregnancy, diabetes mellitus, sex of the fetus, and congenital anomalies), the increase in rate of early-onset disease was 36.0% (95% CI 18.4–46.4, ie, 4.5%/year, 95% CI 2.3–5.8). There was no significant increase in late-onset disease after adjustment for changes in maternal characteristics.
Women with early-onset preeclampsia were more likely to be older, African American, unmarried, and nulliparous, and they were more likely to have diabetes mellitus and a fetus with a congenital anomaly compared with women without early-onset disease (Table 1). Women with late-onset disease were more likely to be younger, unmarried, nulliparous, and have diabetes mellitus. However, they were less likely to be Hispanic, less likely to smoke during pregnancy, and less likely to have a female fetus.
Women with preeclampsia had a significantly higher rate of maternal death compared with women without preeclampsia (14.8 compared with 4.1 deaths/100,000 singleton deliveries; rate ratio 3.7, 95% CI 1.1–12.1). There was one maternal death among the 2,374 women with early-onset preeclampsia, yielding a maternal mortality rate of 42.1 per 100,000 singleton deliveries (95% CI 5.9–299.0). Among women with late-onset disease there were two deaths among 17,890 women, and the maternal mortality rate was 11.2 per 100,000 singleton deliveries (95% CI 2.8–45.2). Women with early-onset disease had a significantly higher rate of maternal mortality compared with women without preeclampsia (42.1 compared with 4.2/100,000 deliveries; OR 9.7, 95% CI 1.3–71.3); women with late-onset preeclampsia had a nonsignificantly increased maternal mortality rate (11.2 compared with 3.1/100,000 deliveries; OR 3.6, 95% CI 0.8–15.3) compared with women who did not have preeclampsia (Table 2). After adjustment for maternal age, race, education, marital status, nulliparity, diabetes mellitus, male fetus and congenital anomalies, the aOR for maternal death given late-onset disease was 5.1 (95% CI 1.1–22.9).
The rate of severe maternal morbidity was highest among women with early-onset preeclampsia, with particularly high rates observed for respiratory morbidity (aOR 21.0, 95% CI 15.6–28.3), cardiovascular morbidity (aOR 21.6, 95% CI 14.0–33.4), and acute renal failure (aOR 32.1, 95% CI 15.8–64.9) compared with women without early-onset disease. Rates of maternal morbidity were relatively lower among women with late-onset preeclampsia compared with women with early-onset preeclampsia. However, women with late-onset preeclampsia had significantly elevated rates of cardiomyopathy (aOR 19.7, 95% CI 8.9–43.6) and acute renal failure (aOR 18.5, 95% CI 11.1–30.7) compared with women without preeclampsia. Obstetric trauma occurred significantly less frequently among women with early-onset and late-onset disease (aOR 0.2, 95% CI 0.2–0.3, and aOR 0.7, 95% CI 0.7–0.8, respectively) compared with women without early-onset and late-onset disease. The rate of chorioamnionitis was also lower among women in the early-onset group (aOR 0.7, 95% CI 0.5–1.0; P=.03).
The rate of composite severe maternal morbidity was 14.2 per 100 singleton deliveries in the early-onset preeclampsia group (aOR 1.2, 95% CI 1.0–1.3; P=.01), and 12.3 per 100 singleton deliveries in the late-onset group (aOR 1.0, 95% CI 0.9–1.0; P=.63), compared with approximately 10 per 100 singleton births among women without preeclampsia. However, when obstetric trauma was excluded from the composite severe maternal morbidity outcome, rates were relatively higher among women with early-onset and late-onset preeclampsia compared with women without early-onset and late-onset disease (12.2 compared with 2.9/100 singleton births, and 5.5 compared with 1.7/100 singleton births, respectively; aOR 3.7, 95% CI 3.2–4.3, and aOR 1.7, 95% CI 1.6–1.9, respectively). Cesarean delivery rates were substantially higher among women with early-onset and late-onset preeclampsia (76.6% and 41.6% among women with early-onset and late-onset disease, respectively, compared with 25.2% among women without preeclampsia).
The rates of postpartum hemorrhage were 8.4 per 100 singleton births among women with both early-onset and late-onset preeclampsia, compared with a postpartum hemorrhage rate of approximately 4.1 per 100 singleton deliveries among women without preeclampsia (aOR 1.9, 95% CI 1.6–2.2, and aOR 2.1, 95% CI 2.0–2.2, respectively). Antepartum hemorrhage was more common among women with early-onset preeclampsia than late-onset preeclampsia (8.9 compared with 2.4/100 singleton deliveries) and significantly elevated when compared with women without early and late-onset preeclampsia (aOR 5.4, 95% CI 4.6–6.3, and aOR 1.8, 95% CI 1.6–2.0, respectively). The median length of stay for delivery hospitalization was 5 days and 3 days in the early-onset and late-onset preeclampsia groups, respectively, in contrast with a 2-day stay among women without preeclampsia (the quartile ranges were 4–8 days, 2–4 days, and 1–3 days, respectively; P<.001 for both).
There was a 23% temporal increase in severe maternal morbidity (excluding trauma) associated with late-onset disease, from 4.7 in 2000 to 6.1 per 100 singleton deliveries in 2008 (Fig. 3; P value for linear trend was .05). No significant temporal trend in severe maternal morbidity was observed among women with early-onset disease. Among women with early-onset disease, a significant decline was observed in maternal infection (specifically puerperal infection) and obstetric trauma. Among women with late-onset disease, a significant increase was observed in maternal infection (predominantly chorioamnionitis) and obstetric shock. Among women without preeclampsia, the rate of severe maternal morbidity increased by 33% (from 2.4 in 2000 to 3.2/100 singleton deliveries in 2008, P for trend less than .001). The rate of increase in severe maternal morbidity was not significantly different between women with late-onset preeclampsia and those without preeclampsia.
Our study showed that early-onset preeclampsia significantly increased the risk of severe maternal morbidity, particularly the risk of respiratory and cardiovascular morbidity and renal failure. The maternal mortality rate among women with early-onset preeclampsia was also significantly elevated, although this estimate was based on a single death. Rates of severe maternal morbidity were lower among women with late-onset disease compared with early-onset preeclampsia. However, specific maternal morbidity rates were significantly elevated among women with late-onset disease compared with women without late-onset disease, especially rates of cardiomyopathy and acute renal failure. A lower frequency of obstetric trauma was observed among women with early-onset and late-onset preeclampsia, and this was expected because women with preeclampsia had substantially higher rates of caesarean delivery and consequently lower rates of high vaginal lacerations and severe perineal trauma. The rate of preeclampsia in Washington State increased between 2000 and 2008, with the temporal increase more marked with regard to early-onset disease.
Although many studies have quantified the effects of preeclampsia on neonatal outcomes, maternal mortality and morbidity rates associated with early-onset and late-onset preeclampsia have not been adequately studied. The rates of severe maternal morbidity associated with preeclampsia in our study were similar to or slightly lower than those reported from hospital-based studies.17–19 The lower rates observed in our study are expected, as hospital-based studies typically include tertiary hospitals with a disproportionate representation of women with severe preeclampsia. In addition, our study population only included singleton pregnancies that are typically less complicated than multifetal pregnancies. Our results also show higher risks of severe maternal morbidity associated with early-onset disease. Preeclampsia occurring at early gestation presents clinicians with the challenge of needing to balance the risk of perinatal death and severe neonatal morbidity after delivery at very early gestation with the risk of worsening the maternal condition associated with expectant management.20–22 A few small randomized trials have compared expectant management and prompt early delivery among women with early onset preeclampsia (24–34 weeks of gestation),18,23–25 and meta-analyses show that expectant management is associated with a lower incidence of some neonatal morbidity (including intraventricular hemorrhage and hyaline membrane disease), without significant differences in maternal outcomes.26 A recent trial also failed to demonstrate any differences in neonatal and maternal outcomes after expectant management compared with early delivery.18 However, these trials were not powered to detect differences in severe maternal morbidity. It is possible that expectant management at early gestation may have contributed to our finding of a temporal increase in adverse maternal outcomes among women with early-onset preeclampsia.
Secular changes in preeclampsia frequency may be expected as a consequence of temporal changes in maternal characteristics and obstetric practice. Thus recent increases in older maternal age and prepregnancy weight could have contributed to temporal increases in preeclampsia rates, whereas increases in early delivery through labor induction, cesarean delivery, or both could have had the opposite effect. Our finding of a temporal increase in the incidence of preeclampsia (predominantly early-onset disease) contrasts with temporal patterns observed in most European countries that have reported a temporal decline in preeclampsia during the past ten years.15 However, similar increases in preeclampsia rates have been observed in Massachusetts, where preeclampsia rates increased by 23% from approximately 3.0 to 3.7 per 100 births between 1998 and 2007. One potential explanation for the discrepant temporal trends between the United States and elsewhere relates to differences in preeclampsia coding.15 Countries with a temporal decline in preeclampsia were those that adopted ICD-10 coding standards (eg, Sweden and Australia), whereas data from studies showing a temporal rise in preeclampsia were based on the ICD-9 coding system (eg, Massachusetts and our study). The temporal increase in early-onset disease is also consistent with the temporal increase in chronic hypertension among pregnant women, which is more strongly associated with early-onset as opposed to late-onset disease.16
Our study has a few limitations. Gestational age at the onset of preeclampsia was estimated based on the time between hospital admission for preeclampsia and hospital admission for delivery. We were not able to capture women with preeclampsia who were not hospitalized and those who did not deliver in the hospital. However, such missed cases of preeclampsia were likely few and mild, that is, those not resulting in serious complications requiring hospitalization. As with any hospitalization database, the accuracy of diagnoses was contingent on documentation and abstraction from medical records. Other limitations of such databases include a lack of clinical detail. Diagnoses related to each hospitalization included the main diagnosis and other diagnoses made throughout the hospital stay, although the exact timing of the secondary diagnoses was not available. Nevertheless, linkage between hospitalization data and birth and neonatal death certificates increased data accuracy, and the accuracy of major obstetric diagnoses and procedures was relatively high in Washington State's linked data file.27 Other potential weaknesses of the study include limited information about risk factors, such as body mass index and medical interventions. The analyses of temporal trends in disease-specific maternal morbidity were exploratory, and some statistically significant results may have arisen by chance owing to multiple comparisons. Temporal trends in very rare maternal morbidity could not be analyzed.
The strengths of our study include a population-based approach including all deliveries in Washington State, with the onset of preeclampsia defined by gestational age at hospitalization rather than gestational age at delivery. The large study size provided sufficient statistical power to examine most specific severe maternal morbidity.
In conclusion, we have documented a temporal increase in preeclampsia in Washington State between 2000 and 2008 that occurred predominantly because of an increase in early-onset disease. The study also quantified the effect of early-onset and late-onset preeclampsia on severe maternal morbidity. Preeclampsia substantially increased rates of cardiovascular, respiratory, central nervous system, renal, hepatic, and other maternal morbidity, and women with early-onset disease had significantly higher rates of specific maternal morbidity compared with women without early-onset disease. Our finding of a temporal increase in the incidence of severe maternal morbidity after late-onset preeclampsia warrants further investigation.
1. Steegers EAP, von Dadelszen P, Duvekot JJ, Pijnenborg R. Pre-eclampsia. Lancet 2010;376:631–44.
2. Hutcheon JA, Lisonkova S, Joseph KS. Epidemiology of pre-eclampsia and the other hypertensive disorders of pregnancy. Best Pract Res Clin Obstet Gynaecol 2011;25:391–403.
3. Sibai B, Dekker G, Kupferminc M. Pre-eclampsia. Lancet 2005;365:785–99.
4. Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Van Look PF. WHO analysis of causes of maternal death: a systematic review. Lancet 2006;367:1066–74.
5. Centre for Maternal and Child Enquiries (CMACE). Saving mothers' lives: reviewing maternal deaths to make motherhood safer: 2006–08. The eighth report on confidential enquiries into maternal deaths in the United Kingdom. BJOG 2011;118(suppl 1):1–203.
6. Campbell KH, Savitz D, Werner EF, Pettker CM, Goffman D, Chazotte C, et al.. Maternal morbidity and risk of death at delivery hospitalization. Obstet Gynecol 2013;122:627–33.
7. Knight M. Eclampsia in the United Kingdom 2005. BJOG 2007;114:1072–8.
8. Liu S, Joseph KS, Bartholomew S, Fahey J, Lee L, Allen AC, et al.. Temporal trends and regional variations in severe maternal morbidity in Canada, 2003 to 2007. J Obstet Gynaecol Can 2010;32:847–55.
9. Berg CJ, Callaghan WM, Syverson C, Henderson Z. Pregnancy-related mortality in the United States, 1998 to 2005. Obstet Gynecol 2010;116:1302–9.
10. MacKay AP, Berg CJ, Atrash HK. Pregnancy-related mortality from preeclampsia and eclampsia. Obstet Gynecol 2001;97:533–8.
11. Abalos E, Cuesta C, Grosso AL, Chou D, Say L. Global and regional estimates of preeclampsia and eclampsia: a systematic review. Eur J Obstet Gynecol Reprod Biol 2013;170:1–7.
12. Roberts JM, Hubel CA. The Two Stage model of preeclampsia: variations on the theme. Placenta 2009;30(suppl A):S32–7.
13. Ness RB, Roberts JM. Heterogeneous causes constituting the single syndrome of preeclampsia: a hypothesis and its implications. Am J Obstet Gynecol 1996;175:1365–70.
14. von Dadelszen P, Magee LA, Roberts JM. Subclassification of preeclampsia. Hypertens Pregnancy 2003;22:143–8.
15. Roberts CL, Ford JB, Algert CS, Antonsen S, Chalmers J, Cnattingius S, et al.. Population-based trends in pregnancy hypertension and pre-eclampsia: an international comparative study. BMJ Open 2011;1:e000101.
16. Lisonkova S, Joseph KS. Incidence of pre-eclampsia: risk factors and outcomes associated with early- versus late-onset disease. Am J Obstet Gynecol 2013;209:544.e1–12.
17. vonDadelszen P, Payne B, Li J, Ansermino JM, Broughtoun-Pipkin F, Côté AM, et al.. Prediction of adverse maternal outcomes in pre-eclampsia: development and validation of the fullPIERS model. Lancet 2010;377:219–27.
18. Vigil-De Gracia P, Reyes Tejada O, Calle Miñaca AC, Tellez G, Chon VY, Herrarte E, et al.. Expectant management of severe preeclampsia remote from term: the MEXPRE Latin Study, a randomized, multicenter clinical trial. Am J Obstet Gynecol 2013;209:425.e1–8.
19. Bombrys AE, Barton JR, Nowacki EA, Habli M, Pinder L, How H, et al.. Expectant management of severe preeclampsia at less than 27 weeks' gestation: maternal and perinatal outcomes according to gestational age by weeks at onset of expectant management. Am J Obstet Gynecol 2008;199:247.e1–6.
20. Sibai BM, Barton JR. Expectant management of severe preeclampsia remote from term: patient selection, treatment, and delivery indications. Am J Obstet Gynecol 2007;196:514.e1–9.
21. Publications Committee, Society for Maternal-Fetal Medicine, Sibai BM. Evaluation and management of severe preeclampsia before 34 weeks' gestation. Am J Obstet Gynecol 2011;205:191–8.
22. Magee LA, Yong PJ, Espinosa V, Côté AM, Chen I, von Dadelszen P. Expectant management of severe preeclampsia remote from term: a structured systematic review. Hypertens Pregnancy 2009;28:312–47.
23. Mesbah EMM. Severe preterm preeclampsia: aggressive or expectant management? Med J Cairo Univ 2003;71:175–82.
24. Odendaal HJ, Pattinson RC, Bam R, Grove D, Kotze TJ. Aggressive or expectant management for patients with severe preeclampsia between 28–34 weeks' gestation: a randomized controlled trial. Obstet Gynecol 1990;76:1070–5.
25. Sibai BM, Mercer BM, Schiff E, Friedman SA. Aggressive versus expectant management of severe preeclampsia at 28 to 32 weeks' gestation: a randomized controlled trial. Am J Obstet Gynecol 1994;171:818–22.
26. Churchill D, Duley L, Thornton JG, Jones L. Interventionist versus expectant care for severe pre-eclampsia between 24 and 34 weeks' gestation. The Cochrane Database of Systematic Reviews 2013, Issue 7. Art. No.: CD003106. DOI: 10.1002/14651858.CD003106.pub2.
27. Lydon-Rochelle MT, Holt VL, Nelson JC, Cárdenas V, Gardella C, Easterling TR, et al.. Accuracy of reporting maternal in-hospital diagnoses and intrapartum procedures in Washington State linked records. Paediatr Perinat Epidemiol 2005;19:460–71.
Diagnostic Codes for Maternal Morbidity