Is eclampsia preventable? A case control review of consecutive cases from an urban underserved region.

Objectives: The purpose of this study was to determine whether eclampsia is potentially preventable.

Methods: This was a retrospective case control study comparing 25 cases of eclampsia, 33 cases of preeclampsia and 50 pregnant controls. Eclampsia cases were classified as follows: group A was potentially preventable by patient education if there were preeclamptic symptoms before presenting to the hospital; group B was potentially preventable by health care provider if patient presented to health care with evidence of preeclampsia and was not appropriately treated; group C was not preventable if eclampsia developed without symptoms or preeclampsia. Statistical analysis was done as applicable.

Results: The mean gestational age at seizure was 34 weeks; 50% of seizures occurred outside a hospital and 44% had multiple seizures. Twenty were antepartum, three intrapartum, and two postpartum. Fourteen (56%) cases were classified as group A, 4 (16%) as group B and 7 (28%) as group C. Compared to controls, African-Americans, nulliparity, premature delivery, and increased neonatal morbidity were significantly associated with eclampsia. Compared with preeclamptics, eclamptic subjects had more symptomatology and more adverse maternal complications.

Conclusions: In this underserved population, a significant percentage of eclampsia cases are potentially preventable by patient education or health care response. In a smaller percentage, eclampsia may develop without any symptomatology or preeclampsia.

Key Words: eclampsia, preeclampsia, pregnancy outcome, prevention

**********

Eclampsia is a complication of pregnancy seen in approximately 1 in 2,000 to 1 in 3,000 deliveries. (1) Eclampsia is the onset of grand mal seizures in pregnant women with preeclampsia. It is generally agreed that eclampsia is usually preceded by a prodrome, which can include headaches, visual disturbance, nausea, vomiting, or epigastric pain. (2) Thus, with early detection of this complex of symptoms, eclampsia may be preventable. Preventive measures should be optimal if: 1) preeclampsia is treated appropriately; 2) the patient is started on anticonvulsant therapy and 3) patients with the above mentioned symptoms are delivered promptly. For this reason, it has been suggested that the incidence of eclampsia may actually reflect the adequacy (or inadequacy) of prenatal care. It is postulated that adequate prenatal care would be associated with appropriate management of severe preeclampsia: educated patients would report prodromal symptoms to health care providers who would treat promptly, resulting in a decrease in eclampsia. Other authors disagree with the notion of eclampsia as a continuum of preeclampsia. (3)

Further analysis of the relationship between the incidence of eclampsia and adequacy of prenatal care may be useful, especially in the setting of an inner city comprised of lower socioeconomic-status populations, such as south-central Los Angeles.

Thus, the objectives of this study were 1) to determine whether eclampsia is preventable, and 2) to evaluate risk factors, symptoms, and perinatal outcomes associated with eclampsia in comparison to healthy controls and preeclampsia cases. Our hypotheses were 1) that there is a difference between preeclampsia and eclampsia in symptomatology and outcomes, and 2) that with the exception of a certain subset of cases, eclampsia is preventable.

Materials and Methods

This was a retrospective case control study of cases encountered at King-Drew Medical serving mainly low-income African-American and Hispanic people. The study was conducted by medical chart review after approval was obtained from the Institutional Review Board.

All cases of eclampsia during the study period were identified from the hospital database.

Each case was matched to 2 healthy controls. The controls were selected by a randomized match to the year of delivery of each case of eclampsia. Two cases of severe preeclampsia closest in time to each case of eclampsia were also selected and used as controls for each case of eclampsia.

To determine whether eclampsia was preventable, eclamptic subjects were classified in three groups. Group A included cases whose chart review demonstrated that they had symptoms and signs of severe preeclampsia (headaches, visual disturbances and epigastric pain) and did not seek immediate medical attention. They were then classified as preventable by patient education and compliance. Group B included subjects who had presented within 7 days to a health care provider with symptoms or findings consistent with preeclampsia, but were sent home or not appropriately treated. This group was classified as potentially preventable by health care provider. Group C referred to cases without prodrome/preeclampsia who had seizures first and then developed signs and symptoms of preeclampsia.

The charts for all cases were reviewed for variables which included: age, parity, race, onset of prenatal care, medical problems, social history, drug, alcohol or tobacco use, blood pressure, laboratory data, delivery records, birth weight, neonatal complications, and seizure characteristics. Severe preeclampsia was defined by blood pressure greater than 160/110, proteinuria greater than 5 grams in 24 hours. HELLP syndrome was the presence of thrombocytopenia (platelet count less than 150,000 per cubic millimeter) elevated liver enzymes and hemolytic anemia. Renal failure was evidenced by requirement of dialysis. Gestational age was calculated from last menstrual period or from an early ultrasound. All neonates were evaluated and managed by a certified neonatologist. Neonatal morbidity included hyaline membrane disease, bronchopulmonary dysplasia, severe intraventricular hemorrhage (Grade 3 or 4), severe retinopathy (Stage 3 or 4), meconium aspiration syndrome, pulmonary hemorrhage and perinatal mortality. Hyaline membrane disease was defined by the need for supplemental oxygen, ventilatory support, and characteristic radiographic findings. Bronchopulmonary dysplasia was defined in premature infants dependent on supplemental oxygen at 36 weeks corrected gestational age.

All parameters were compared among eclampsia, preeclampsia and control cases. Results were analyzed using two-tailed Student t test, Fisher exact test, Pearson [chi square] test, and one-way analysis of variance and logistic regression analysis, as appropriate. P value < 0.05 was considered statistically significant. Odds ratios and 95% confidence intervals (CI) were calculated. Statistical analysis was performed with the Statistical Package for the Social Sciences for Windows, version 10 (SPSS, Inc., Chicago, IL).

Results

During the period of 1983 to 2000, there were 51 cases of eclampsia and a total of 100,848 deliveries, with an incidence of 1 in 2000 deliveries for eclampsia. Of the cases of eclampsia, 33 charts were found and 25 charts had complete data. Of the 50 cases of severe preeclampsia selected from the database, 33 complete records were found. Thus the analysis was performed on 25 cases of eclampsia, 33 cases of severe preeclampsia and 50 healthy controls.

The mean gestational age at which seizure occurred was 34 ([+ or -]0.89) weeks; 20 (80%) were antepartum, 3 (12%) intrapartum, and postpartum eclampsia occurred in 2 (8%) cases. Thirteen (52%) had one seizure, 9 (36%) had two seizures, and 3 (12%) had three seizures. Seventeen (68%) of patients had their first convulsions at home or in the ambulance. Secondary seizures were encountered more often in the hospital setting (76%) than at home (24%).

Fourteen patients (56%) were classified as group A, in whom eclampsia was potentially preventable by patient education. Four (16%) belonged to group B, with eclampsia potentially preventable by the health care system. Seven (28%) were classified as group C, in whom eclampsia was thought to be nonpreventable. The mean seizure-to-delivery interval was 7.84 ([+ or -]2.27) hours.

Table 1 shows that subjects with eclampsia were significantly younger than healthy controls (22.4 versus 27 years, P = 0.003). Eclamptics were 4 times more likely to be African-American than Hispanic. Nulliparity, tobacco use and hypertension were also significantly associated with eclampsia. There was no correlation with lack of prenatal care, body mass index, or weight gain in pregnancy. Preeclamptics compared with eclamptics were significantly older (28.66 versus 22.4 yr, P = 0.002) and also had a larger body mass index (30.2 versus 37.7, P = 0.005).

As expected, there were no symptoms in the healthy control group. Only one healthy subject had leg edema. Table 2 shows that 72% of subjects with eclampsia had one or more of the prodrome symptoms of headaches, visual disturbance, epigastric tenderness or nausea and vomiting, versus 42% in subjects with preeclampsia (P = 0.034). For specific symptoms, headaches and nausea/vomiting were significantly associated with eclampsia. It was interesting to note that overall edema was significantly higher in preeclamptics.

Table 3 shows delivery and neonatal characteristics. Mean gestational age at delivery was significantly lower with eclampsia than in the control population (34.16 versus 38.92 weeks, P = 0.000). Eighty percent of subjects with eclampsia had a cesarean delivery compared with 6.3% of the controls (P = 0.000). Birth weight was significantly lower among neonates of eclamptic cases compared with controls, (2,211.7 versus 3,384.6 g, P = 0.000). Umbilical artery pH and Apgar scores at 1 minute and 5 minutes were considerably higher in the neonates of the control subjects. The incidence of congenital anomalies (36% versus 2%), days spent in the neonatal intensive care unit (NICU) (14 versus 1 day) and overall neonatal morbidity (52% versus 6%) were much higher in the neonates of eclamptic as compared to control pregnancies. For the preeclampsia group, the mean gestational age at delivery was 35.21 ([+ or -]0.9) weeks, 60% had a cesarean delivery, the mean birth weight was 2,345 ([+ or -]185) grams, the mean Apgar score at 1 minute was 7.69 ([+ or -]0.28) and at 5 minutes was 8.3 ([+ or -]0.32), with mean NICU days of 17.3 ([+ or -]6.3) days and 15 (57%) with neonatal morbidity. On comparison, there was no statistical difference between any of these values and those of subjects with eclampsia.

Table 4 compares the postpartum sequelae between eclampsic and preeclampsic subjects. The mean postpartum systolic blood pressure in eclampsics (176) was significantly higher than the mean systolic blood pressure in preeclamptics (162) (P = 0.16). Even though the liver transaminases were elevated in both groups, they were significantly higher in eclamptic subjects. The eclamptics were significantly more anemic (26.6 versus 33.3, P = 0.001) and had more thrombocytopenia (173 versus 127, P = 0.05). There was no individual significant difference between the two groups in regard to neurologic disorder, disseminated intravascular coagulation (DIC), renal failure, pulmonary edema or blindness; however 16 (64%) of eclamptics had at least one or more of these complications compared with 8 (24%) of preeclamptics (P = 0.008).

Discussion

Previous studies have shown a higher incidence of eclampsia in African-American women compared with whites (4), but a MEDLINE search did not reveal any comparisons with Hispanic women. In this study, African-American women were four times more likely than healthy controls to present with eclampsia than Hispanic women. Statistically significant associations with eclampsia when compared with controls were: younger maternal age, nulliparity, tobacco use and hypertension, which have been reported previously. (1,5,6) There was no significant racial difference when preeclamptics were compared with eclamptics but preeclamptics were still older, had a higher body mass index, and were more likely to have edema.

In 56% of the cases studied, eclampsia was potentially preventable by patient education and compliance. These were patients with prodromal symptomatology who stayed home instead of seeking immediate medical attention. In 16% of cases the health care provider could have prevented the progression to eclampsia. These were patients who presented with symptoms and signs of preeclampsia to a health care provider and were sent home. Only 28% were felt to be nonpreventable since the eclampsia occurred in hospital with seizure being the first sign of preeclampsia. Given that about 72% of cases were potentially preventable by patient education/compliance or appropriate provider care, it is not surprising that there was no difference in prenatal care between controls, preeclamptics, and eclamptics. With adequate prenatal care, patients seek medical attention appropriately, with initiation of proper management to prevent development of eclampsia. This phase would include prompt diagnosis, initiation of seizure prophylaxis, and immediate delivery. (4) Thus effective and adequate prenatal care involves patient education, compliance and response by health care professionals.

The notion of eclampsia as a continuum of preeclampsia remains quite controversial. Katz et al, (3) in a retrospective analysis of 51 pregnancies complicated by eclampsia showed that 60% of seizures occurred without warning signs of preeclampsia. In addition, only 5 patients had no prenatal care. Eclampsia was then suggested as a subset of preeclampsia, rather than as a consequence. In such cases, eclampsia may well be impossible to prevent. A review by Sibai (2) grouped cases of eclampsia into categories of physician error, patient failure, and failure of magnesium sulfate therapy. Upon their analysis, about 30% of cases were also found to be unpreventable and about 60% of the pregnancies had no prenatal care. In this study about 30% was also nonpreventable, which is in line with Sibai's study (2) and contrasts with the study by Katz et al. (3) This may suggest that there are 2 subsets of eclampsia, those that occur as a continuum from preeclampsia, and those that occur as the first sign of preeclampsia. Thus effective prenatal care may be expected to prevent the first subset and not the second. Therefore underserved communities with inadequate or no prenatal care at all will be expected to have a larger percentage of potentially preventable eclampsia, and communities with very effective prenatal care and education may be expected to have a larger percentage of nonpreventable eclampsia. This may explain the difference between Sibai's (2) and the current study's findings versus those of Katz et al (3).

Delivery and neonatal outcome of eclamptic pregnancies when compared with controls were primarily influenced by the increased rate of preterm delivery. This explains the decreased birth weight, poor overall neonatal status at birth, the significant neonatal morbidity and increased NICU stay. This study also assessed maternal morbidity associated with eclampsia. The findings confirm increased morbidity with eclampsia, which is supported by previous reports. (6-8) Furthermore, this report shows that eclampsia is associated with a worse maternal outcome than preeclampsia (64% versus 24%).

The main limitation of this study is the retrospective design with missing records. There is a possibility of selection bias in recall, since patients with uneventful deliveries are less likely to recall or report symptoms. Also, this was a small number of cases, but since eclampsia is uncommon, getting large numbers from a single institution is difficult.

Conclusion

In this underserved population, eclampsia is associated with significant maternal and neonatal morbidity. A significant percentage of eclampsia cases are potentially preventable by patient education or health care response. In a smaller percentage, eclampsia may develop without any symptomatology or preeclampsia.

Table 1. Demographics, social, and prenatal factors: eclampsia compared
with controls and eclampsia compared with preeclampsia (a)

Variable         Eclampsia       Controls      P value        95% CI

Maternal age     22.4 (1.4) (b)  27.06 (0.79)  0.003        24.03-27.01
African-
  American       10 (40%)         5 (10%)      0.005 [4]     1.76-20.36
Hispanic         15 (60%)        45 (90%)      0.005 [0.67]  0.48-0.93
Nulliparity      18 (72%)        13 (26%)      0.000 [7.30]  2.4-21.5
Body mass index  30.2 (1.84)     31.85 (0.95)  NS
Weight gain/
  week            0.63 (0.1)      0.64 (0.01)  NS
Alcohol abuse     2 (8%)          1 (2%)       NS
Tobacco use       5 (20%)         0            0.003         0.65-0.97
Cocaine use       2 (8%)          0            NS
Hypertension      3 (12%)         0            0.034         0.76-1
No prenatal
  care            6 (24%)         5 (10%)      NS
Gestational age
  at PNC         17.29 (2.42)    15.91 (1.2)   NS

Variable         Eclampsia      Preeclampsia   P value        95% CI

Maternal age     22.4 (1.4)     28.66 (1.04)   0.002        23.8-28
African-
  American       10 (40%)        7 (21%)       NS
Hispanic         15 (60%)       24 (72%)       NS
Nulliparity      18 (72%)       16 (50%)       NS
Body mass index  30.2 (1.84)    37.8 (1.72)    0.005        32-37.5
Weight gain/
  week            0.63 (0.001)   0.65 (0.001)  NS
Alcohol abuse     2 (8%)         2 (6%)        NS
Tobacco use       5 (20%)        1 (3%)        NS
Cocaine use       2 (8%)         1 (3%)        NS
Hypertension      3 (12%)        7 (21%)       NS
No prenatal
  care            6 (24%)        6 (18%)       NS
Gestational age
  at PNC         17.29 (2.42)   16.27 (1.43)   NS

(a) OR, odds ratio; CI, confidence interval; PNC, prenatal care; NS, not
significant.
(b) Standard error of the mean in parentheses; OR in brackets.

Table 2. Eclampsia symptomatology: eclampsia compared with preeclampsia
(a)

Variable               Eclampsia  Preeclampsia  P value (OR)  95% CI

Headaches              18 (72%)    9 (27.3%)    0.001 (2.64)  1.44-4.85
Visual disturbance      9 (36%)    5 (15.2%)    NS
Epigastric tenderness   9 (36%)    8 (24.2%)    NS
Nausea/vomiting         9 (36%)    3 (9.1%)     0.02 (4)      1.19-13.1
Any symptom            18 (72%)   14 (42.4%)    0.034 (1.7)   1.06-2.7
Face edema              7 (28%)    9 (28.1%)    NS
Leg edema              14 (56%)   21 (65.6%)    NS
Hand edema              5 (20%)    6 (18.8%)    NS
Any edema              14 (56%)   23 (71.9%)    0.000 (2.8)   3.9-201

"OR, odds ratio; NS, not significant; CI, confidence interval.

Table 3. Delivery and neonatal outcome: eclampsia compared with controls
(a)

Variable                     Eclampsia           Controls

Gestational age at delivery    34.16 (1.14) (a)    38.92 (0.23)
Cesarean delivery              20 (80%)             2 (6.3%)
Birth weight                 2211.68 (177.4)     3384.56 (67.17)
Apgar at 1 minute               6.72 (0.46)         8.98 (0.009)
Apgar at 5 minutes              7.8 (0.41)          9.02 (0.009)
Umbilical artery pH             7.27 (0.0001)       7.43 (0.009)
NICU days                      13.84 (4.20)         1.16 (0.51)
Neonatal morbidity             13 (52%)             3 (6%)
Congenital anomalies            9 (36%)             1 (2%)

Variable                     P value (OR)  95% CI

Gestational age at delivery  0.000           36.4-38.3
Cesarean delivery            0.000 [12.8]     3.2-49.7
Birth weight                 0.000         2799.15-3188.05
Apgar at 1 minute            0.000            7.84-8.62
Apgar at 5 minutes           0.000            8.31-8.91
Umbilical artery pH          0.000            7.25-7.37
NICU days                    0.000            2.23-8.54
Neonatal morbidity           0.000 [8.7]      2.7-27.6
Congenital anomalies         0.000 [18]       2.4-134.3

"OR, odds ratio; CI, confidence interval; NICU, neonatal intensive care
unit.
Standard error of the mean in parentheses; OR in brackets.
Neonatal morbidity = respiratory distress syndrome, bronchopulmonary
dysplasia, intraventricular hemorrhage grade 3 or 4, retinopathy 3 or 4,
meconium aspiration syndrome, pulmonary hemorrhage, and perinatal
mortality.

Table 4. Postpartum eclampsia complications: eclampsia compared with
preeclampsia (a)

Postpartum variable     Eclampsia       Preeclampsia   P value (OR)

Maximum systolic BP     176.56 (5.45)   161.96 (2.63)  0.016
Maximum diastolic BP    102.56 (3.29)    97.14 (2.14)  NS
Minimum platelet count  127.4 (18)      172.7 (13.5)   0.045
Maximum ALT             170.06 (58.53)   70.78 (16.1)  0.056
Maximum AST             196.28 (48.3)    92.6 (19.6)   0.028
Maximum LDH             546.9 (76.8)    359.2 (73.8)   NS
Minimum hematocrit       26.63 (1.07)    33.3 (1.2)    0.001
Maximum BUN              12.3 (0.9)      12.67 (1.3)   NS
Minimum serum albumin     2.3 (0.12)      2.95 (2.1)   NS
Maximum uric acid         8 (0.38)        6.7 (0.29)   0.008
Maximum bilirubin         1.8 (0.36)      1.2 (0.74)   NS
Maximum creatinine        0.98 (0.009)    0.89 (.002)  NS
Preeclampsia             10 (40%)
HELLP                    15 (60%)        16 (50%)      NS
Neurologic disorder       6 (24%)         1 (3.0%)     NS
Blindness/papilloedema    5 (20%)         2 (6.1%)     NS
DIC                       4 (16%)         2 (6%)       NS
Pulmonary edema           3 (12%)         5 (15%)      NS
Renal failure             4 (16%)         5 (15%)      NS
Any complication         16 (64%)         8 (24%)      0.008 [2.6]

"OR, odds ratio; BP, blood pressure; ALT, alanine aminotransferase; AST,
aspartate aminotransferase; LDH, lactate dehydrogenase; BUN, blood urea
nitrogen; DIC, disseminated intravascular coagulation; NS, not
significant.
Standard error of the mean in parentheses; OR in brackets.
Any complication = maternal death, pulmonary edema, hemodialysis, DIC,
or neurologic and visual disorders.

Accepted September 30, 2003.

Copyright [c] 2004 by The Southern Medical Association

0038-4348/04/9705-0440

A chip on the shoulder is a sure indication that there is more wood higher up.

--Aldous Huxley

References

1. Roberts JM. Pregnancy related hypertension, in Creasy RK, Resnik R (eds): Maternal-Fetal Medicine. Philadelphia, W.B. Saunders Co., 1999, ed 4, pp 833-872.

2. Sibai BM. Eclampsia: Part VI--Maternal-perinatal outcome in 254 consecutive cases. Am J Obstet Gynecol 1990;163:1049-1055.

3. Katz VL, Farmer R, Kuller JA. Preeclampsia into eclampsia: toward a new paradigm. Am J Obstet Gynecol 2000;182:1389-1396.

4. Sibai BM, Mercer BM, Schiff E, et al. Aggressive versus expectant management of severe preeclampsia at 28 to 32 weeks' gestation: a randomized controlled trial. Am J Obstet Gynecol 1994;171:818-822.

5. Mattar F, Sibai BM. Eclampsia: Part VIII--Risk factors for maternal morbidity. Am J Obstet Gynecol 2000;182:307-312.

6. Martin JN Jr, May WL, Rinehart BK, et al. Increasing maternal weight: a risk factor for preeclampsia/eclampsia but apparently not for HELLP syndrome. South Med J 2000;93:686-691.

7. Witlin AG, Saade GR, Mattar F, et al. Risk factors for abruptio placentae and eclampsia: analysis of 445 consecutively managed women with severe preeclampsia and eclampsia. Am J Obstet Gynecol 1999;180:1322-1329.

8. Kaplan PW, Neurologic issues in eclampsia. Rev Neurol (Paris) 1999;155:335-341.

RELATED ARTICLE: Key Points

* Eclampsia was associated with patients who were African-American, nulliparous, tobacco users and hypertensive.

* Eclampsia was associated with adverse maternal and neonatal outcome.

* A significant percentage of eclampsia cases are potentially preventable by patient education or health care response.

* In a smaller percentage, eclampsia may develop without any symptomatology or preeclampsia.

Dotun Ogunyemi, MD, Jean-Louis Benae, MD, and Chidi Ukatu, MD

From the Department of Obstetrics and Gynecology, Charles R. Drew University of Medicine & Science, University of California at Los Angeles, Los Angeles, CA.

We have no commercial or proprietary interest in any drug, device, or equipment mentioned in this article. There was no financial support or provision of supplies for this study.

Reprint requests to Dotun Ogunyemi, MD, Department of Obstetrics and Gynecology, Martin Luther King/Charles R. Drew Medical Center, 12021 S. Wilmington Avenue, Los Angeles, CA 90059-3019. Email: [e-mail]