Preeclampsia is responsible for major maternal and perinatal morbidity and mortality. The disorder affects approximately 5%–9% of pregnancies, and if untreated, it can lead to eclampsia, a serious obstetric complication that accounts for maternal deaths in the United States and abroad.1–3 There has been a marked reduction in the incidence of eclampsia in recent years, which may be attributed in part to improved prenatal care, including early detection of signs and symptoms of preeclampsia and prophylactic use of magnesium sulfate in the peripartum period.4 However, there has been a shift toward an increased frequency of eclampsia in the postpartum period with most cases occurring in the late postpartum period.5 This shift could be the result of the persistence and progression of preeclampsia that started during the intrapartum period or late manifesting disease after delivery but unrecognized before discharge.
Because preeclampsia abates with placental delivery, obstetric units tend to discontinue seizure prophylaxis within 48 hours postpartum. However, up to 26% of eclamptic seizures occur beyond 48 hours5 and as late as 6 weeks after delivery.6 In a rare case of an advanced extrauterine, intra-abdominal pregnancy, in which removal of the placenta was not possible, preeclampsia was reported to persist for 99 days instead of resolving after delivery of the fetus.7 These facts are consistent with a current theory on the pathogenesis of preeclampsia that suggests that the disease is caused by placental-related widespread endothelial cell dysfunction8 as a result of increased circulating levels of antiangiogenic factors such as soluble truncated FMS-like tyrosine kinase-1 receptor isoform9 and soluble endoglin.10 Therefore, the onset of symptoms and signs of the disease may be dictated by the degree and time course of resolution of the endothelial damage. Indeed, Blaauw and collaborators11 demonstrated evidence of persistent endothelial dysfunction in a group of women with early onset preeclampsia up to 11 months after delivery.
There is a paucity of data in the literature regarding the natural history, progression, management, and treatment outcome of delayed onset preeclampsia in patients more than 2 days postpartum who present after hospital discharge. In addition, the extent to which development or progression of preeclampsia is influenced by either the patient's demographic or antecedent obstetric characteristics is also unclear. In this study, our objective was to estimate the demographics, clinical course, and complications of patients who were readmitted with a diagnosis of postpartum preeclampsia after discharge from the hospital and to compare those in this cohort who developed eclamptic seizures with those who did not. Knowledge of these variables may be helpful in formulating strategies aimed at prevention, early diagnosis, and timely treatment of delayed-onset preeclampsia.
MATERIALS AND METHODS
This study was approved by the institutional review board of the Wayne State University (HIC No. 037909M1E). We conducted a chart review of patients who were discharged and later readmitted with the diagnosis of postpartum preeclampsia more than 2 days to 6 weeks or less after delivery between January 2003 and August 2009 at the Detroit Medical Center. Based on the International Classification of Diseases, 9th Revision, code, a total of 253 patients were coded to have been readmitted with postpartum preeclampsia from their diagnosis at discharge; 101 patients were excluded from the final analysis because they did not fulfill criteria for the diagnosis of postpartum preeclampsia. The medical records of 152 patients with a readmission diagnosis of preeclampsia and or eclampsia were reviewed for demographic information, obstetric and medical history, intrapartum and postpartum management, presenting symptoms and signs, management in the emergency department, and inpatient hospital course and complications. A separate article analyzes the characteristics of cases developing delayed onset postpartum preeclampsia with and those without antecedent hypertensive diatheses in the antepartum and early peripartum period before initial discharge in the index pregnancy.12
All statistical analyses were performed using Statistical Package for Social Sciences. All tests were conducted using P≤.05 for statistical significance. Normally distributed numerical variables were summarized with means±standard deviations, whereas other variables were summarized with medians and ranges. Frequencies and percentages with 95% confidence intervals (CIs) were used to summarize categorical variables. For numerical variables, univariable comparisons were conducted with either Student's t test (for normally distributed data) or Wilcoxon's rank test (for nonnormal data). Chi-square and Fisher's exact tests were used for categorical data. To model gravidity, the data were dichotomized into gravidity less than three and three or more. Categorization was done because the residuals for a linear model with gravidity did not meet the assumptions of normality. Adjusted odds ratios (ORs) and 95% CIs were computed by logistic regression for the risk of the eclampsia. Potential confounders were included in the full model if they were risk factors for eclampsia based on a P<.05.
During the study period, 48,498 deliveries were recorded, and preeclampsia occurred in 3,072 patients (6.3%, 95% CI 6.1–6.6). One hundred fifty-two patients (5.0%, 95% CI 4.2–5.8) were readmitted to the hospital within 6 weeks of their initial discharge with the diagnosis of postpartum preeclampsia or eclampsia. The mean±standard deviation and median (range) maternal age and gestational age at delivery were 27.6±6.8 and 28 (15–44) years and 38.2±2.5 and 39 (28–42) weeks, respectively. Twenty-three percent (23.3%, 95% CI 17.3–30.7) of the patients were primigravid, 32.9% (95% CI 26.3–41.2) were nulliparous, 78.9% (95% CI 71.8–84.7) were on Medicaid, and 96.7% (95% CI 92.5–98.6) were African American. Information about the gravidity and parity was missing in two patients. The mode of delivery was vaginal in 57.9% (95% CI 49.9–65.5), whereas 42.1% (95% CI 34.6–50.1) underwent cesarean deliveries. Four (2.6%, 95% CI 0.9–7.0) patients had twin gestations.
Of the 152 patients readmitted with the diagnosis of delayed postpartum preeclampsia or eclampsia, 96 (63.2%, 95% CI 55.3–70.4) had no antecedent diagnosis of hypertensive disease in the current pregnancy, whereas seven (4.6%, 95% CI 2.3–9.2) had gestational hypertension, 14 (9.2%, 95% CI 5.6–14.9) had chronic hypertension, 28 (18.4%, 95% CI 13.1–25.3) had preeclampsia, and seven (4.6%, 95% CI 2.3–9.2) were diagnosed with preeclampsia superimposed on chronic hypertension during the peripartum period. Twenty-seven (17.8%, 95% CI 12.5–24.6) of the patients readmitted with the diagnosis of delayed postpartum preeclampsia were treated with magnesium sulfate during the peripartum period. These patients were equally distributed across the eclampsia and noneclampsia groups (Table 1). The mean (range) number of days postpartum and between discharge and readmission were 7.6 days (3–23) and 4.7 days (0–19), respectively.
The highest systolic blood pressure on readmission was (mean [range]) 176 (116–240) mmHg and the highest diastolic blood pressure was 101 (72–131) mmHg. Symptoms and signs at presentation were not mutually exclusive. Headache (n=105, 69.1%, 95% CI 61.3–75.9) was the most common presenting symptom. Other symptoms included shortness of breath (n=41, 30.0%, 95% CI 20.6–34.5), blurry vision (n=32, 21.1%, 95% CI 15.3–28.2), nausea (n=19, 12.5%, 95% CI 8.2–18.7), vomiting (n=17, 11.2%, 95% CI 7.1–17.2), edema (n=16, 10.5%, 95% CI 6.6–16.4), seizure (n=6, 4.0%, 95% CI 1.8–8.4), other neurological deficit (n=8, 5.3%, 95% CI 2.7–10.0), and epigastric pain (n=8, 5.3%, 95% CI 2.7–10.0). Six patients were admitted with a history of eclamptic seizures at home, five seized while in the emergency department, and another 11 seized during the readmission period. The overall rate of eclampsia for patients readmitted with the diagnosis of postpartum preeclampsia was 14.5% (n=22, 95% CI 9.8–20.9). Over 90% (90.9%, 95% CI 72.2–97.5) of patients with eclampsia presented within 7 days after discharge from hospital. Eclampsia was more common (n=17, 77.3%, 95% CI 56.6–89.9) in patients with no antecedent hypertensive disease in the index pregnancy. None of the patients with preeclampsia superimposed on chronic hypertension had postpartum eclampsia. Three (13.6%, 95% CI 3.6–36.0) patients with peripartum preeclampsia, one (4.6%, 95% CI 0.2–24.9) with chronic hypertension, and one (4.6%, 95% CI 0.2–24.9) with gestational hypertension also developed postpartum eclampsia on readmission.
Other complications diagnosed after readmission included pulmonary edema (n=17, 11.2%, 95% CI 7.1–17.2), cardiomyopathy (n=4, 2.6%, 95% CI 0.9–7.0), hemolysis, elevated liver enzymes, low platelet count syndrome (n=3, 2.0%, 95% CI 0.5–6.1), pneumonia (n=1, 0.7%, 95% CI 0.03–4.2), and maternal death secondary to hemorrhagic stroke (n=1, 0.7%, 95% CI 0.03–4.2). Blood pressure medications were used in the emergency department in 103 (67.8%, 95% CI 60.0–74.7) patients and 74 (48.7%, 95% CI 40.9–56.6) who were discharged on oral medications were equally distributed across the eclamptic and noneclamptic groups (Table 2). Calcium channel blockers were the most common antihypertensive medication prescribed and were used by 47 (30.9%, 95% CI 24.1–38.7) patients. One hundred twenty-nine (84.9%, 95% CI 78.3–89.7) patients were treated with magnesium sulfate during their admission. Compared with women who did not develop postpartum eclampsia during their second admission, those who did were significantly younger (P=.001) with a relatively lower gravidity (P=.03) and a higher serum hemoglobin (P=.03). When these variables were entered into a logistic regression model, young age was associated with greater odds of developing eclampsia (adjusted OR 1.13, 95% CI 1.02–1.26, P=.03), whereas a lower readmission hemoglobin was associated with a lower odds of progression to eclampsia (adjusted OR 0.75, 95% CI 0.57–0.98, P=.04). Gravidity was not associated with risk of eclampsia (gravidity less than three compared with three or more; adjusted OR 1.29, 95% CI 0.37–4.45, P=.69). Although patients who developed eclampsia appeared less likely to have peripartum preeclampsia, chronic hypertension, and diabetes, these differences were not statistically significant (Table 3). There were no significant differences between patients with postpartum preeclampsia with and without seizures with respect to other demographic variables, medical disorders associated with pregnancy, hospital course, or laboratory findings (Tables 1 and 3). Of 543 patients diagnosed with preeclampsia superimposed on chronic hypertension during the study period, only 28 (5.2%, 95% CI 3.6–7.4) were readmitted to the hospital for postpartum preeclampsia and none developed eclampsia.
In this large series of 152 patients with delayed postpartum preeclampsia who were readmitted after initial discharge from the hospital, we identified several variables that could help stratify patients who are more likely to develop the disease. In our study, over 96% of the patients were African American, 78% were on Medicaid, and 63.2% had no antecedent diagnosis of hypertensive disease in the index pregnancy. Similar to the 78% reported by Chames and colleagues,5 those with no antecedent diagnosis of hypertensive disease seem to be particularly at risk, because the highest rate of eclampsia (77.3%) occurred in this group of patients. This could be the result of the fact that patients and their physicians might not be aware that there is still a measurable risk of preeclampsia even after postpartum discharge despite an uneventful pregnancy and delivery. In contrast, of the 543 patients diagnosed with preeclampsia superimposed on chronic hypertension during the study period, only 28 (5.2%) were readmitted to the hospital for postpartum preeclampsia and none developed eclampsia. However, it is possible that some of these patients were discharged from the emergency department after merely increasing the dose of their antihypertensive(s) if they were asymptomatic and therefore were not recorded.
Only maternal age and preadmission hemoglobin distinguished between those who did or did not develop postpartum eclampsia. In contrast, other maternal demographic variables, history of medical disorders, symptoms and signs on presentation, hospital course, and laboratory data were not associated with the development of postpartum eclampsia. Similarly, intrapartum use of magnesium sulfate was not associated with the development of delayed postpartum eclampsia.
Although more common in the ante- and intrapartum periods, 44% of eclampsia occurred postpartum in one study,13 and nearly one-third of patients developed eclamptic seizures more than 48 hours postpartum in another.5 Thus, it is essential that patients at risk for this condition be identified so that prompt diagnosis of preeclampsia can be made and measures to prevent progression instituted.
Headache was the most common presenting symptom in our study population and was present in over two-thirds (69.1%) of the patients. Furthermore, we found that all of our patients (100%) admitted with postpartum eclampsia presented with prodromal symptoms, a rate slightly higher than the 91% reported in a previous study.14 This stresses the importance of evaluating postpartum patients carefully for the signs and symptoms enumerated previously, especially when they are associated with elevated blood pressures. Given that many patients will be seen in the emergency department, as most of our patients were, it is important for emergency department physicians to have a high index of suspicion for postpartum preeclampsia including the early involvement of obstetric staff in the care of such patients.
At the present time, there are no data to suggest that the pathology of preeclampsia in the postpartum period is different from those that occur in the antepartum and the peripartum periods. In addition, laboratory values were no different between those who developed eclampsia and those who did not; therefore, the findings of normal laboratory values should not preclude the use magnesium sulfate prophylaxis for seizure prophylaxis. One way to avoid complications of eclampsia is to continue to educate all patients and our emergency department colleagues not to ignore the symptoms, signs, and the possibility of preeclampsia or its progression. This should take place before patients are discharged from the hospital and should include verbal instructions and a printed instruction sheet.14 Patients should be informed that they may be at risk for preeclampsia or eclampsia up to 6 weeks after delivery.
In our study, 90% of those who developed eclampsia presented within 7 days of being discharged in our study. Two other patients that developed eclampsia did so on the 11th and 12th days after they were discharged from the hospital. Whether active postpartum surveillance at home for these eclamptic patients for at least 1 week after discharge might have prevented the development of eclampsia is an intriguing question.
Several limitations of our study must be acknowledged. First, this was a retrospective cohort study and some results did not include all of the participants because of missing data. Although we attempted a comprehensive review, we may not have identified all patients with delayed postpartum preeclampsia because mild forms of the disease may have resolved spontaneously without patients seeking hospital care. In addition, some of our patients may have sought care elsewhere rather than at our institution. Furthermore, asymptomatic patients with a history of chronic hypertension who presented with atypical symptoms may have been discharged by emergency department physicians with adjustment of their blood pressure medication(s) without work-up to determine whether they had postpartum preeclampsia superimposed on chronic hypertension. Lastly, the lack of association between antecedent hypertension, preeclampsia, renal disease, and diabetes and delayed postpartum eclampsia may have been the result of a type 2 error. However, the data for those with history of gestational hypertension and renal disease show little evidence of trends that might achieve statistical significance in a larger sample size. A larger data set may show that patients with diabetes and chronic hypertension are less likely to develop delayed postpartum eclampsia.
In summary, we found that younger women are at a higher risk for eclamptic seizures in the late postpartum period and that the overwhelming majority of seizures occur within 7 days of postpartum hospital discharge. Education about the possibility of delayed postpartum preeclampsia and eclampsia should occur after delivery whether or not a patient develops hypertensive disease before discharge from the hospital.
1. Diagnosis and management of preeclampsia and eclampsia. ACOG Practice Bulletin No. 33. American College of Obstetricians and Gynecologists. Obstet Gynecol 2002;99:159–67.
2. Hogberg U. The World Health Report 2005: ‘make every mother and child count’—including Africans. Scand J Public Health 2005;33:409–11.
3. Magee LA, Sibai B, Easterling T, Walkinshaw S, Abalos E, von Dadelszen P. How to manage hypertension in pregnancy effectively. Br J Clin Pharmacol 2011;72:394–401.
4. Lopez-Llera M. Main clinical types and subtypes of eclampsia. Am J Obstet Gynecol 1992;166:4–9.
5. Chames MC, Livingston JC, Ivester TS, Barton JR, Sibai BM. Late postpartum eclampsia: a preventable disease? Am J Obstet Gynecol 2002;186:1174–7.
6. Matthys LA, Coppage KH, Lambers DS, Barton JR, Sibai BM. Delayed postpartum preeclampsia: an experience of 151 cases. Am J Obstet Gynecol 2004;190:1464–6.
7. Piering WF, Garancis JG, Becker CG, Beres JA, Lemann J Jr. Preeclampsia related to a functioning extrauterine placenta: report of a case and 25-year follow-up. Am J Kidney Dis 1993;21:310–3.
8. Levine RJ, Maynard SE, Qian C, Lim KH, England LJ, Yu KF, et al.. Circulating angiogenic factors and the risk of preeclampsia. N Engl J Med 2004;350:672–83.
9. Venkatesha S, Toporsian M, Lam C, Hanai J, Mammoto T, Kim YM, et al.. Soluble endoglin contributes to the pathogenesis of preeclampsia. Nat Med 2006;12:642–9.
10. Roberts JM, Redman CW. Pre-eclampsia: more than pregnancy-induced hypertension. Lancet 1993;341:1447–51.
11. Blaauw J, Graaff R, van Pampus MG, van Doormaal JJ, Smit AJ, Rakhorst G, et al.. Abnormal endothelium-dependent microvascular reactivity in recently preeclamptic women. Obstet Gynecol 2005;105:626–32.
12. Filetti LC, Imudia AN, Al-Safi Z, Hobson DT, Awonuga AWO, Bahado-Singh RO. New onset delayed postpartum preeclampsia: different disorders? J Matern Fetal Neonatal Med 2011 [Epub ahead of print].
13. Douglas KA, Redman CW. Eclampsia in the United Kingdom. BMJ 1994;309:1395–400.
14. Barton JR, Witlin AG, Sibai BM. Management of mild preeclampsia. Clin Obstet Gynecol 1999;42:455–69.