Dexamethasone compared with betamethasone for glucocorticoid treatment of postpartum HELLP syndrome.

OBJECTIVES: To compare the efficacy of dexamethasone and betamethasone to ameliorate the course of postpartum hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome. METHODS: A prospective, mixed randomized/non-randomized clinical investigation of patients with postpartum HELLP syndrome. Treatment with either dexamethasone or betamethasone was continued until there was evidence of disease recovery. RESULTS: Baseline characteristics of both the dexamethasone (n=18) and betamethasone (n=18) groups were similar. Although the time to discharge from the obstetrical recovery room was not statistically significant between groups, reduction in mean arterial blood pressure was more pronounced in the dexamethasone group as compared with the betamethasone group (-15.3+/-1.4 mmHg vs. -7.5+/-1.4 mmHg, respectively, P<0.01). Patients in the dexamethasone group required less antihypertensive treatment than the betamethasone group (6% vs. 50%, P=0.01) and also had a decreased need for readmission to the obstetrical recovery room (0% vs. 22%, P=0.03). CONCLUSION: This investigation supports the use of dexamethasone as the superior glucocorticoid to use for patients with postpartum HELLP syndrome.

Dye-dilution techniques using aminohippurate sodium: do they accurately reflect amniotic fluid volume?

OBJECTIVE: To determine whether the dye-dilution technique using aminohippurate sodium accurately measures amniotic fluid volume. METHODS: Singleton pregnancies with intact membranes undergoing a Cesarean delivery had their amniotic fluid volume assessed by the dye-dilution technique and direct measurement. RESULTS: Fifteen women were prospectively assessed. Six patients had their amniocentesis on the delivery table and nine patients at 4-24 h prior to the Cesarean delivery. The six women undergoing an amniocentesis just before delivery had good concordance between the dye-determined and direct measurement of amniotic fluid volume (r = 0.99, p = < 0.001). Among the nine women with varying times from amniocentesis to direct measurement, the correlation was not significant (r = 0.36, p = 0.08). The percentage difference between the dye-determined and directly measured amniotic fluid volume was significantly smaller in the women undergoing amniocentesis just prior to delivery (7%) than in the women with varying times from amniocentesis to delivery (37%, P < 0.001). CONCLUSION: Dye-determined amniotic fluid volume accurately reflects actual amniotic fluid volume but the dye-determined concentrations, in vivo, may undergo rapid changes.

Interleukin-10 administration and bacterial endotoxin-induced preterm birth in a rat model.

OBJECTIVE: To determine whether intra-uterine infusion of interleukin-10 prevents preterm delivery in rats treated with endotoxin. METHODS: Pregnant rats underwent implantation of uterine catheters and were randomly assigned to receive intrauterine infusion of either normal saline, 50 microg lipopolysaccharide endotoxin, or 50 microg lipopolysaccharide with 500 ng interleukin-10 administered either concurrently or 24 hours later. The interval from infusion to delivery for each group was recorded, along with the number of live born pups and their birth weight. We calculated that to obtain a power of 80%, assuming a 24-hour difference in the treatment to delivery times between the test and control subjects, at least six animals would be needed in each group. RESULTS: In females receiving lipopolysaccharide (50 microg) alone, the interval to delivery (P <.05), live birth rate (P <.05), and pup weight (P <.001) were reduced compared with the saline-infused controls. In contrast, females receiving interleukin-10 at the time of the endotoxin challenge or 24 hours after delivered at term with no difference in litter size or live birth weight compared with the controls. CONCLUSION: Animals treated with both lipopolysaccharide and interleukin-10, administered concurrently or 24 hours after the endotoxin challenge, delivered normal weight pups at term with a similar litter size as the saline-infused controls. Interleukin-10 appears to be effective in preventing endotoxin-induced preterm birth and fetal wastage in pregnant rats.

The myth of transient hypertension: descriptor or disease process?

The objective of this study is to describe the incidence of transient hypertension and to evaluate if transient hypertension is associated with increased maternal or fetal morbidity as compared to other hypertensive disorders of pregnancy and normotensive controls. Data were collected from all deliveries at the University of Mississippi Medical Center from July 1, 1996 through January 1, 1997. Patients were grouped according to ACOG criteria for pregnancy induced hypertension. Specific maternal and fetal morbidities were compared among the groups and controls. There were 1489 deliveries during the study period. Nearly 30% of patients met the criteria for transient hypertension. There were no significant differences between patients with transient hypertension and controls in regard to maternal and fetal outcomes. Transient hypertension occurs more often than expected, however, it appears to be of limited clinical significance.

Change in platelet count predicts eventual maternal outcome with syndrome of hemolysis, elevated liver enzymes and low platelet count.

OBJECTIVE: To profile the rates of change for platelets and lactate dehydrogenase (LDH) in patients with the syndrome of hemolysis, elevated liver enzymes and low platelet count (HELLP) to reflect and possibly predict disease severity. METHODS: Serial determinations of the platelet count and serum LDH were analyzed retrospectively for 545 patients with HELLP syndrome and 94 severely pre-eclamptic patients. A repeated measures (mixed model) analysis of variance (ANOVA) was utilized to estimate the rates of change for LDH and platelets in each group. RESULTS: Patients who developed class 1 or 2 HELLP syndrome exhibited a rapid deterioration in platelet count of 45-50,000/microl per day compared to 30,000/microl per day for class 3 and non-HELLP severe pre-eclampsia. For LDH, values increased at a rate of approximately 1400 IU/l per day, 600 IU/l per day, 300 IU/l per day and 200 IU/l per day for patients with classes 1, 2 and 3 and for non-HELLP severe pre-eclampsia, respectively. CONCLUSION: The rate of change of platelets and LDH appeared to correlate well with eventual syndrome severity and this can be used to enhance patient assessment beyond the value of a single test for either laboratory parameter.

Pregnancy outcome in women with preterm labor symptoms without cervical change.

OBJECTIVE: This study was undertaken to determine pregnancy outcome in women who have preterm labor symptoms without cervical change according to fetal fibronectin status. STUDY DESIGN: Patients who were examined at the obstetric emergency department with symptoms of preterm labor but without cervical change underwent fetal fibronectin collection. Pregnancy outcome and fetal fibronectin results were analyzed after delivery. RESULTS: Of the 235 patients sampled, 20% (n = 48) had positive fetal fibronectin results. The mean +/- SD gestational age at delivery was lower in women with positive fetal fibronectin results (34.2 +/- 4.1 vs 37.7 +/- 2.3 weeks; P <.001); these women were more likely to deliver preterm as a result of preterm labor than women with other obstetric indications (46% vs 19%; P <.001). Infants born to these women demonstrated lower birth weight (2317 +/- 895 g vs 2877 +/- 557 g; P =.003), were more likely to be admitted to the neonatal intensive care unit (42% vs 14%; P <.001), and were more likely to die in the neonatal period (11% vs 0%; P <.001). CONCLUSION: Patients with symptoms of preterm labor but without cervical change who have negative fetal fibronectin results are less likely to deliver preterm. Therefore in women with symptoms but without cervical change fetal fibronectin should be considered for risk assessment.

Single umbilical artery is associated with an increased incidence of structural and chromosomal anomalies and growth restriction.

The objective to characterize neonatal outcome associated with ultrasonographic identification of a single umbilical artery. Pregnancies diagnosed with single umbilical artery antenatally were identified. All prenatal/antenatal and pediatric records were reviewed for maternal demographics, associated anomalies, karyotypic analysis, pregnancy complications, and neonatal outcome. Twenty-seven pregnancies complicated by fetal single umbilical artery were identified. Of the 27 pregnancies, 5 (18.5%) underwent pregnancy termination and 1 (3.7%) experienced fetal demise. Of the 21 liveborn infants, 4 (19%) died within the first year of life. Sixty-seven percent of fetuses had an associated structural anomaly. Sixteen of the 27 pregnancies underwent amniocentesis and 7 of these were chromosomally abnormal. All of the karyotypically abnormal fetuses had a structural defect in addition to the single umbilical artery. Of the six fetuses without any associated structural or chromosomal anomalies, three (50%) demonstrated growth restriction. Single umbilical artery is relatively rare finding. When a single umbilical artery is identified, a vigilant search for associated anomalies should be undertaken. Pregnancies identified as having fetuses with associated structural anomalies should be offered amniocentesis. Pregnancies with isolated single umbilical artery should be carefully monitored for evidence of fetal growth restriction.

Randomized trial of intermittent or continuous amnioinfusion for variable decelerations.

OBJECTIVE: To determine whether continuous or intermittent bolus amnioinfusion is more effective in relieving variable decelerations. METHODS: Patients with repetitive variable decelerations were randomized to an intermittent bolus or continuous amnioinfusion. The intermittent bolus infusion group received boluses of 500 mL of normal saline, each over 30 minutes, with boluses repeated if variable decelerations recurred. The continuous infusion group received a bolus infusion of 500 mL of normal saline over 30 minutes and then 3 mL per minute until delivery occurred. The ability of the amnioinfusion to abolish variable decelerations was analyzed, as were maternal demographic and pregnancy outcome variables. Power analysis indicated that 64 patients would be required. RESULTS: Thirty-five patients were randomized to intermittent infusion and 30 to continuous infusion. There were no differences between groups in terms of maternal demographics, gestational age, delivery mode, neonatal outcome, median time to resolution of variable decelerations, or the number of times variable decelerations recurred. The median volume infused in the intermittent infusion group (500 mL) was significantly less than that in the continuous infusion group (905 mL, P =.003). CONCLUSION: Intermittent bolus amnioinfusion is as effective as continuous infusion in relieving variable decelerations in labor. Further investigation is necessary to determine whether either of these techniques is associated with increased occurrence of rare complications such as cord prolapse or uterine rupture.

Leukocytosis is proportional to HELLP syndrome severity: evidence for an inflammatory form of preeclampsia.

BACKGROUND: We investigated the possibility that HELLP syndrome is in part a systemic inflammatory response. METHODS: We evaluated total white blood cell (WBC) counts of all patients with severe preeclampsia with and without HELLP syndrome admitted to our hospital between 1995 and 1997. Patients were grouped by diagnosis and timing of platelet nadir. Analysis of variance and regression analysis were used for data analysis. RESULTS: Of 177 patients, 91 had HELLP syndrome, and 86 had severe preeclampsia alone. The WBC counts were significantly higher in patients with HELLP syndrome (12.5 +/- .442 x 10(9)/L) than in patients with severe preeclampsia (10.3 +/- .288 x 10(9)/L). Regression analysis showed that platelet counts varied inversely with WBC counts. Also, patients with class I HELLP syndrome had significantly higher WBC counts than patients with other classes of HELLP syndrome. CONCLUSION: The finding of an association between increasing leukocytosis and worsening thrombocytopenia early in the course of HELLP syndrome supports the hypothesis that it may represent an inflammatory process.

A prospective, randomized, controlled trial of high and low maintenance doses of magnesium sulfate for acute tocolysis.

OBJECTIVE: This study was undertaken to compare a high-dose protocol for magnesium sulfate tocolytic therapy with a low-dose regimen with respect to time needed to achieve tocolysis. STUDY DESIGN: Patients between 24 and 34 weeks' gestation with preterm labor were included. Patients with ruptured membranes or nonreassuring fetal assessments were excluded. Gravid women received a 4-g loading dose of magnesium sulfate and were prospectively randomly assigned to receive a maintenance dose of 2 or 5 g/h. RESULTS: The median times to tocolysis were 120 minutes (semi-interquartile range, 30 minutes) in the low-dose group and 90 minutes (semi-interquartile range, 28 minutes) in the high-dose group (P <.001). CONCLUSION: Patients treated with a higher maintenance dose of magnesium sulfate had a higher frequency of side effects; however, tocolysis was achieved more rapidly and they required shorter admissions to the labor and delivery unit without increased maternal or neonatal morbidity.

Intrauterine endotoxin infusion in rat pregnancy induces preterm delivery and increases placental prostaglandin F2alpha metabolite levels.

OBJECTIVE: This study was designed to examine the effects of intrauterine endotoxin (lipopolysaccharide) on rat pregnancy. STUDY DESIGN: Pregnant Sprague-Dawley rats (N = 26) were implanted with uterine catheters on day 15 or 16 of a 22-day gestation. Animals were randomly assigned to receive either lipopolysaccharide (25 or 50 microg) or sodium chloride solution (1 mL) on day 17 and then were either sacrificed on day 19 or observed until delivery. Placentas were harvested at the time of death, homogenates were prepared, and prostaglandin F(2)(alpha) metabolite levels were determined by means of radioimmunoassay. Data were analyzed by analysis of variance, Student-Newman-Keuls, and Mann-Whitney tests. RESULTS: Lipopolysaccharide-treated groups (25 and 50 microg) displayed a shorter interval to delivery (mean +/- SE, 82 +/- 13 and 63 +/- 8 hours, respectively) than control animals (117 +/- 3 hours). Pups of lipopolysaccharide-treated (25 and 50 microg) female animals had lower live birth weights (4.92 +/- 0.01 and 5.12 +/- 0. 24 g, respectively) compared with control animals (6.04 +/- 0.07 g). Placental homogenates from lipopolysaccharide-treated female animals contained higher levels of prostaglandin F(2)(alpha) metabolite (1567 +/- 64 and 1475 +/- 59 pg/mL) than those from sodium chloride solution-infused control animals (804 +/- 68 pg/mL). CONCLUSION: Bacterial products induce the preterm delivery of low-birth-weight pups in rats, possibly by increasing local prostaglandin biosynthesis.

Lack of utility of standard labor curves in the prediction of progression during labor induction.

OBJECTIVE: This study was undertaken to determine whether patients undergoing labor induction can be reliably evaluated by means of standard labor assessment curves. STUDY DESIGN: In this retrospective chart review of 123 patients who underwent cervical ripening and induction of labor, Friedman's standard labor curves were used for comparison. Statistical analysis was performed with the Student t test. RESULTS: Nulliparous and parous patients undergoing cervical ripening spent more time in active-phase labor than standard expectations of labor progression would indicate (12.7 +/- 7.8 vs 5. 9 +/- 3.4 hours for nulliparous women, P <.001; 7.9 +/- 6.4 vs 2.5 +/- 1.5 hours for parous women, P <.001). Nulliparous and parous patients who were delivered vaginally spent more time in active labor than did their respective standard historical control populations (10.3 +/- 8.0 vs 5.9 +/- 3.4 hours for nulliparous women, P <.001; 7.0 +/- 6.0 vs 2.5 +/- 1.5 hours for parous women, P <. 001). CONCLUSION: Standard methods for the evaluation of labor adequacy and prediction of the likelihood of vaginal delivery may not apply to patients undergoing cervical ripening.

Cardiopulmonary morbidity as a complication of severe preeclampsia HELLP syndrome.

OBJECTIVE: To profile the types and frequencies of cardiopulmonary morbidity encountered in patients with severe preeclampsia with or without hemolysis, elevated liver enzymes, and low platelets (HELLP syndrome). STUDY DESIGN: We initiated a retrospective study of 979 patients with severe preeclampsia with and without HELLP syndrome. Types of cardiopulmonary morbidity were analyzed among the three classes of HELLP syndrome and severe preeclampsia without HELLP syndrome. RESULTS: Cardiopulmonary morbidity occurred in 7.6% of study patients. As a group, patients with cardiopulmonary complications were more likely to have cesareans (11% vs 6%, p = 0.019) earlier in gestation (1366 +/- 700 gm birth weight versus 1734 +/- 892 gm birth weight, p = 0.021), with higher peak postpartum blood pressures (< 0.001) and with more abnormal laboratory values indicative of multisystem disease, compared with patients without this complication. Patients with cardiopulmonary complications required almost twice as long to achieve diuresis as comparison patients (22 +/- 23 hours versus 12 +/- 11 hours, p < 0.001). CONCLUSION: The probability of cardiopulmonary complications increases significantly when patients develop class 1 HELLP syndrome. Of all cardiopulmonary complications, acute lung injury/acute respiratory distress syndrome is most specific to class 1 HELLP syndrome. Transient renal dysfunction is closely related to cardiopulmonary morbidity.

Second-trimester uterine evacuation: A comparison of intra-amniotic (15S)-15-methyl-prostaglandin F2alpha and intravaginal misoprostol.

OBJECTIVE: Our purpose was to compare the efficacy, safety, and adverse effects of intra-amniotically administered (15S)-15-methyl-prostaglandin F(2alpha) and intravaginally administered misoprostol for second-trimester uterine evacuation. STUDY DESIGN: Fifty-one patients were randomly assigned to receive either a single 2.5-mg intra-amniotic injection of (15S)-15-methyl-prostaglandin F(2)(alpha) (n = 26) or two 200-microg intravaginal doses of misoprostol (n = 25) at 12-hour intervals. The primary outcome measured was evacuation of the uterus within 24 hours. RESULTS: The mean time from initiation of termination to uterine evacuation was less in the prostaglandin group than in the misoprostol group (17.5 +/- 8.6 hours vs 22.3 +/- 12.5 hours), but this was not statistically significant (P >.05). The rate of successful fetal evacuation at 24 hours was significantly higher in the prostaglandin group than in the misoprostol group (88% vs 60%, P =.02). The complete-abortion rate and the incidence of adverse effects were similar in both groups. CONCLUSION: The use of an intra-amniotic injection of (15S)-15-methyl-prostaglandin F(2alpha) for midtrimester pregnancy termination is safe and is associated with a greater number of successful uterine evacuations within 24 hours, without an increase in adverse effects, than intravaginal administration of misoprostol.

Expression of the placental cytokines tumor necrosis factor alpha, interleukin 1beta, and interleukin 10 is increased in preeclampsia.

OBJECTIVE: We sought to determine whether placental cytokine expression is altered in patients with preeclampsia. STUDY DESIGN: Whole placental tissue was collected at cesarean delivery, and total ribonucleic acid was extracted. Reverse transcriptase-polymerase chain reaction was performed to determine cytokine expression. Product bands were quantitated by scanning densitometry, and results were expressed as a ratio of cytokine/housekeeping gene (cytokine expression index). Statistical analysis was performed by the Student t test and the Mann-Whitney U test. RESULTS: Placentas from 6 patients with preeclampsia and 4 normotensive patients were analyzed. Placental expression of interleukin 1beta and interleukin 10 was greater in preeclamptic women than in normotensive subjects (median interleukin 1beta cytokine expression index, 0.675; range, 0.394-0. 953; vs 0.106; range, 0.084-0.166; P =.011; median interleukin 10 cytokine expression index, 1.042; range, 0.672-1.192; vs 0.126; range, 0.062-0.398; P <.011). Tumor necrosis factor alpha messenger ribonucleic acid was detected in placentas of preeclamptic subjects but not in normotensive control subjects. CONCLUSION: Placentas from preeclamptic patients demonstrated increased expression of interleukin 1beta, interleukin 10, and tumor necrosis factor alpha. This may be in association with placental hypoxia and may contribute to the global endothelial dysfunction observed in preeclampsia.

Maternal mortality associated with HELLP (hemolysis, elevated liver enzymes, and low platelets) syndrome.

OBJECTIVE: The aim of this study was to determine factors contributing to deaths among women with HELLP (hemolysis, elevated liver enzymes, and low platelets) syndrome. STUDY DESIGN: Information from multiple sources was scrutinized to distinguish and profile maternal deaths associated with HELLP syndrome. RESULTS: Information was available regarding 54 maternal deaths. According to HELLP syndrome classification 60.0% had class 1 disease, 35.6% had class 2 disease, and 4.4% had class 3 disease. Events associated with maternal deaths included cerebral hemorrhage (45%), cardiopulmonary arrest (40%), disseminated intravascular coagulopathy (39%), adult respiratory distress syndrome (28%), renal failure (28%), sepsis (23%), hepatic hemorrhage (20%), and hypoxic ischemic encephalopathy (16%). Delay in diagnosis of HELLP syndrome was implicated in 22 of 43 patients' deaths (51.1%). CONCLUSIONS: It appears that (1) most maternal deaths occurred among women with class 1 HELLP syndrome, (2) delay in diagnosis was associated with mortal consequences, and (3) hemorrhage in the hepatic or central nervous system or vascular insult to the cardiopulmonary or renal system were associated with increased mortality risk.

Modern obstetric management and outcome of infants with gastroschisis.

OBJECTIVE: To determine whether outcomes of infants with gastroschisis differed by mode or site of delivery, diagnostic method, or when maternal-fetal medicine consultation was given. METHODS: Charts of 32 infants born at the University of Mississippi Medical Center or admitted to the neonatal intensive care unit between September 1992 and June 1998 were reviewed for maternal demographic characteristics and neonatal outcomes. Statistical analysis was done using Student t test, analysis of variance, chi2, and Kruskal-Wallis test with P<.05 considered statistically significant. RESULTS: There were no statistically significant differences in neonatal outcomes by method or site of delivery, diagnostic method, or maternal-fetal medicine consultation before delivery. Infants delivered vaginally had higher Apgar scores at 1 and 5 minutes (9 versus 7 and 9 versus 8, respectively, P<.05). Vaginally delivered infants required more days of antibiotic therapy than those delivered abdominally (10 versus 3 days, P<.05) but had a shorter interval to enteral feedings (15 versus 30 days, P<.05). CONCLUSION: Outcomes of infants with isolated gastroschisis were not significantly affected by method or site of delivery, diagnostic method, or maternal-fetal surveillance. Although the findings of this investigation were largely negative and the statistical power limited due to the rarity of this fetal disruption, small series of cases of gastroschisis need to be analyzed to resolve current controversies surrounding optimal treatment of gastroschisis.

Neonatal sepsis and death caused by resistant Escherichia coli: possible consequences of extended maternal ampicillin administration.

OBJECTIVE: Our goal was to evaluate the relationship between neonatal death caused by sepsis associated with ampicillin-resistant organisms and length of antibiotic exposure. STUDY DESIGN: All neonatal deaths from culture-positive sepsis over a 3-year period were examined. Infants who were delivered at either the University of Mississippi Medical Center or at Saint Barnabas Medical Center at >/=24 weeks' gestation and died within 7 days of life were included. Information on the organism causing sepsis and its sensitivities was collected, and the number of doses of ampicillin administered to the mother before delivery was determined. RESULTS: Of the 78 neonatal deaths, 35 met the inclusion criteria. There were 8 cases of sepsis from ampicillin-resistant Escherichia coli and 27 cases caused by other organisms. There was a statistically significant difference between the mean number of doses of ampicillin received by the ampicillin-resistant Escherichia coli group (17.6 +/- 5.5) compared with the other organisms group (4.9 +/- 3.6) (P <.001). CONCLUSION: A relationship exists between neonatal death caused by ampicillin-resistant Escherichia coli and prolonged antepartum exposure to ampicillin.

Multiple courses of betamethasone to enhance fetal lung maturation do not suppress neonatal adrenal response.

OBJECTIVE: Our purpose was to evaluate the neonatal adrenal gland by provocative testing in neonates of mothers who had received multiple courses of betamethasone to enhance fetal lung maturity. STUDY DESIGN: Infants of mothers who had received >/=3 courses of betamethasone for fetal lung maturation were enrolled in the study. Twenty-four hours after delivery a baseline serum cortisol concentration was obtained. A synthetic adrenocorticotropic hormone (Cortrosyn) was administered (0.25 mg/1.73 m2). Two hours later a second serum cortisol concentration was obtained. An increase in serum cortisol in response to Cortrosyn was considered a positive test result. Nominal data were compared by means of the Student t test. RESULTS: There were 9 infants enrolled in the study. The mean number of betamethasone treatment cycles was 4.8 +/- 1.09. The mean baseline cortisol level was 2.23 +/- 0.52 microgram/dL, and the mean post-adrenocorticotropic hormone cortisol level was 9.86 +/- 1.70 microgram/dL. All neonates had a positive adrenocorticotropic hormone test result. Stepwise linear regression showed no association between the number of courses of betamethasone treatment cycles and the post-adrenocorticotropic hormone cortisol concentration. CONCLUSION: Multiple weekly treatment cycles of betamethasone for fetal lung maturity administered between 24 and 34 weeks' gestation do not appear to cause adrenal suppression.

The spectrum of severe preeclampsia: comparative analysis by HELLP (hemolysis, elevated liver enzyme levels, and low platelet count) syndrome classification.

OBJECTIVE: This study was undertaken to explore the spectrum of maternal disease with a triple classification system of HELLP (hemolysis, elevated liver enzyme levels, and low platelet count) syndrome and compare these classes with severe preeclampsia without HELLP syndrome. STUDY DESIGN: In this retrospective analytic study the pregnancies of 777 patients with class 1, 2, or 3 HELLP syndrome were compared and contrasted with those of 193 women with severe preeclampsia but without HELLP syndrome. RESULTS: Eclampsia, epigastric pain, nausea and vomiting, significant proteinuria, major maternal morbidity, and stillbirth increased as HELLP syndrome worsened from class 3 to class 1. In contrast, headache and diastolic hypertension were more common among the significantly heavier patients with severe preeclampsia without HELLP syndrome. Approximately half of pregnancies complicated by class 1 HELLP syndrome exhibited significant maternal morbidity, compared with only 11% of those complicated by severe preeclampsia without HELLP syndrome. Although a significant trend was apparent in increasing levels of lactate dehydrogenase, aspartate aminotransferase, and uric acid as HELLP syndrome worsened, there was considerable variation within groups. CONCLUSION: Laboratory and clinical indices of disease severity in patients with severe preeclampsia or eclampsia generally were highest with class 1 HELLP syndrome and were lowest when HELLP syndrome was absent. Class 3 HELLP syndrome is considered a clinically significant transitional group.