International Journal of Gynecology & Obstetrics 72 Ž2001.

17᎐23

Article

Pregnancy complicated by pre-eclampsia᎐eclampsia

with HELLP syndrome

P. Vigil-De GraciaU
Gynecology and Obstetric Unit, Complejo Hospitalario Metropolitano de la Caja de Seguro Social, Panama, Panama

Received 23 December 1999; received in revised form 18 April 2000; accepted 27 April 2000

Abstract

Objecti¨ e: To determine the incidence of, and assess the relationship between liver enzymes and platelet counts
with the severity of HELLP Žhemolysis elevated liver enzymes and low platelet count. syndrome, and describe
incidences of serious maternal complications. Materials and methods: Retrospective descriptive study of patients with
pre-eclampsia᎐eclampsia complicated by HELLP syndrome that occurred over a 3-year period in Panama. The
primary outcome included: platelet count; serum aspartate aminotransferase; serum alanine aminotransferase;
symptoms and complications among class 1, 2, and 3 HELLP. Results: There were 558 pregnancies complicated by
severe pre-eclampsia and 26 by eclampsia. The incidence of HELLP syndrome among women with severe
pre-eclampsia in our population was 12% and among women with eclampsia was 34.6%, Ž P- 0.0008.; with a total
incidence of 16%. Epigastric pain, visual symptoms and hematuria increased with the severity of HELLP syndrome.
Hematuria was the fourth symptom, but was significantly Ž Ps 0.002. associated with class 1 HELLP. There were
significant differences in the platelet count, and liver enzymes among the classes of HELLP syndrome. Abruptio
placentae, acute renal failure and disseminated intravascular coagulation were the most frequent maternal complica-
tions. There were two maternal deaths. Conclusions: This study supports the theory that HELLP syndrome is
associated with increased maternal morbidity and mortality. Our data suggest that certain subgroups of patients with
class 1 HELLP syndrome Ž‘classic or true HELLP’. are at increased risk for serious maternal complications,
including those with: platelet counts below 50 000r␮l; lactic dehydrogenase G 2000 IUrl; aspartate aminotranferase
G 500 IUrl; alanine aminotransferase G 300 IUrl; and hematuria. 䊚 2001 International Federation of Gynecology
and Obstetrics. All rights reserved.

Keywords: Pre-eclampsia᎐eclampsia; HELLP syndrome; Maternal mortality; Thrombocytopenia

U
Apartado Postal: 87 32 24, Zona 7, Panama,´ Panama.´ Tel.: q507-230-3821.
E-mail address: pvigil-d@cwp.net.pa ŽP. Vigil-De Gracia..

0020-7292r01r$20.00 䊚 2001 International Federation of Gynecology and Obstetrics. All rights reserved.
PII: S 0 0 2 0 - 7 2 9 2 Ž 0 0 . 0 0 2 8 1 - 2
18 P. Vigil-De Gracia r International Journal of Gynecology and Obstetrics 72 (2001) 17᎐23
1. Introduction
Pre-eclampsia is a major cause of maternal and
perinatal mortality and morbidity worldwide, par-
ticularly in developing countries. High blood pres-
sure complicates almost 10% of all pregnancies
w1x.
As described by Weinstein in 1982 w2x, ‘HELLP’
is the acronym for the syndrome of hepatic dys-
function, thrombocytopenia, and hemolytic
anemia. Between 4 and 18.9% of patients with
pre-eclampsia᎐eclampsia develop HELLP syn-
drome w3,4x. Patients whose pregnancies are com-
plicated by HELLP syndrome are at a higher risk
for renal failure, consumptive coagulopathy,
abruptio placentae, pulmonary and cerebral
edema, subcapsular liver hematoma, and hypo-
volemic shock w5x.
The definitive therapy for severe pre-
eclampsia᎐eclampsia, with or without HELLP
syndrome, is removal of all gestational products
from the uterus. The use of antepartum cortico-
steroids, rescue surfactant, neonatal intensive care
unit technology, and the maternal transport of
premature and immature pregnancies to tertiary
care facilities have collectively increased the sur-
vivability of these infants.
The clinical and laboratory courses of all women
diagnosed with HELLP syndrome were analyzed
to: Ž1. determine the incidence of HELLP syn-
drome; Ž2. to asses relationship between liver
enzymes and platelet count with the severity of
HELLP syndrome; Ž3. describe incidences of seri-
ous maternal morbidities associated with HELLP
syndrome; Ž4. detect potential clinical factors that
may affect maternal and neonatal morbidity and
its association with the triple-classification system
w6x; and Ž5. assess neonatal results associated with
HELLP syndrome.
2. Materials and methods
The ‘Complejo Hospitalario Metropolitano de
la Caja de Seguro Social’ in Panama City serves
as a hospital of tertiary care facilities for the
Republic of Panama, where there are approxi-
mately 5000 annual deliveries. During the 3-year
period from 1 July 1996, through 30 June 1999, 86
of the pregnancies managed at this hospital were
complicated by HELLP syndrome. Diagnosis of
HELLP syndrome was based on the clinical diag-
nosis of pre-eclampsia and the following labora-
tory abnormalities: Ž1. hemolysis, characteristic
peripheral blood smear, serum lactic dehydroge-
nase ŽLDH. G 600 Url, total bilirubin G 1.2
mgrdl, decreased hemoglobin and hematocrit; Ž2.
elevated liver enzymes, defined as aspartate
aminotransferase ŽAST. G 70 Url, alanine
aminotransferase ŽALT. G 50 Url and LDHG
600 Url; and Ž3. low platelet count defined as:
class 1 HELLP, which has a maternal platelet
nadir F 50 000r␮l, class 2 HELLP has a platelet
nadir between ) 50 000 and 100 000r␮l, and class
3 HELLP has ) 100 000 to F 150 000r␮l.
Routine laboratory evaluations included the se-
rial measurement of liver function tests, complete
blood cell count, coagulation profile, and renal
function tests.
Acute renal failure was diagnosed in the pres-
ence of oliguria᎐anuria in association with a se-
vere reduction in renal function Želevated serum
creatinine and serum urea, diminished creatinine
clearance ..
Disseminated intravascular coagulation ŽDIC.
was defined as the presence of low platelets Ž-
100 000r␮l., low fibrinogen Ž- 250 mgrdl., pro-
longed prothrombin and partial thromboplastin
times, and estimates of fibrinogen degradation
products or D-dimmer. Gestational age was de-
termined by the last menstrual period, uterine
size at the first prenatal visit or early sonography,
if obtained. Intrauterine growth restriction was
defined as a birth weight below the 10th per-
centile for that gestational age.
To prevent and control seizures, women with
HELLP syndrome routinely received magnesium
sulfate as a 4-g intravenous loading dose followed
by a 1-g dose intravenous per h. Bolus doses of
hydralazine or nifedipine were administered to
control severe hypertension. Blood and blood
products were used to correct coagulation abnor-
malities and anemia as needed.
The data have been presented as the mean "
S.D. Statistical comparisons were performed with
␹2 and Fisher exact tests, where appropriate for
 P. Vigil-De Gracia r International Journal of Gynecology and Obstetrics 72 (2001) 17᎐23 19

Table 1
Clinical characteristics of 86 women with HELLP syndromea

HELLP Žall. 1 HELLP 2 HELLP 3 HELLP

Ž N s 86. Ž N s 21. Ž N s 42. Ž N s 23.

Maternal age Žyear. 29.18" 6.19 28.14" 6.82 29.64" 6.0 29.13" 6.21
Nulliparity Ž%. 43 43 43 43
Blood pressure
Systolic G 160 mmHg Ž%. 60.5 62 67 48
Diastolic G 110 mmHg Ž%. 46.5 52 48 39
Diastolic F 90 mmHg Ž%. 15 14 17 13
No prenatal care Ž%. 4.6 5 4
Proteinuria 3᎐4 q Ž%. 56 71 62 30 e
No proteinuria Ž%. 10.5 5 9.5 17
Platelet count = 10 3r␮l b 78.68" 36.7 28 " 7.69 79 " 14.6 124 " 13.8
LHD ŽUrl. c 1202 " 779 1745 " 1202 1081 " 525 908 " 289
AST ŽUrl. 225 " 281 482 " 380 d 171 " 216 c 98 " 67
ALT ŽUrl. 169 " 167 328 " 219 b 136 " 128 c 90 " 52
a
LDHs lactic dehydrogenase, ASTs aspartate aminotransferase, ALTs alanine aminotransferase.
b
P- 0.001: platelets count Žamong HELLP group., ALT Ž1 HELLP vs. 2 HELLP..
c
P- 0.05: LDH Žamong HELLP group., AST Ž2 HELLP vs. 3 HELLP., ALT Ž2 HELLP vs. 3 HELLP..
d
P - 0.01: AST Ž1 HELLP vs. 2 HELLP..
e
P- 0.004.

categoric variables, and one-way analysis of vari- Žrange 17 000᎐150 000.. The mean platelet count
ance for continuous variables with a P-value of of class 1 HELLP was 28 000r␮l, for 2 HELLP,
- 0.05 considered significant. 79 000r␮ l, and for class 3 HELLP was
124 000r␮l. There were significant differences in
platelet counts LDH, AST and ALT between the
3. Results class 1, 2 and 3 HELLP syndromes. In class 1
HELLP syndrome, the maximal LDH levels
ranged from 654᎐5600 IUrl with a mean of 1745
During the study period there were 558 preg-
IUrl; for class 2 the mean was 1081 IUrl Žrange
nancies complicated by severe pre-eclampsia and
579᎐2961.; and for class 3 the mean level was 908
26 by eclampsia. Eighty-six pregnancies were di-
agnosed with HELLP syndrome, of whom 77 had
severe pre-eclampsia and 9 had eclampsia. The
incidence of HELLP syndrome among women
with severe pre-eclampsia in our population was
12% and among women with eclampsia was 34.6%
Ž P- 0.0008.; with a total incidence of 16%.
Six women Ž6.9%. had pre-existing chronic hy-
pertension, five Ž5.8%. had multiple gestation and
seven Ž8.1%. had pre-existing uterine myomas.
The maternal age, parity, blood pressure and
prenatal care were similar between classes 1, 2, 3
of HELLP syndrome ŽTable 1.. The mean mater-
nal age was 29.18" 6.19 Žrange 16᎐41 years..
Table 1 summarizes the laboratory findings in Fig. 1. Comparison of mean peak concentrations of lactic
these 86 women with HELLP syndrome by classes. dehydrogenase, aspartate aminotransferase and alanine
The mean platelet count was 78 686 " 36 744 aminotransferase.
20 P. Vigil-De Gracia r International Journal of Gynecology and Obstetrics 72 (2001) 17᎐23

IUrl with ranged from 535᎐1590 IUrl. Aspartate
aminotranferase and ALT changed significantly
between classes of HELLP syndrome, according
to their severity. Serum concentrations of AST
and ALT generally paralleled LDH during the
course of HELLP syndrome, but at comparatively
more reduced serum concentrations. The mean
peak serum levels of LDH, AST and ALT were
significantly higher in the class 1 HELLP syn-
drome group; also serum concentrations of LDH,
AST, ALT diminished between HELLP 1 and
HELLP 3, but continued to show significant dif-
ferences ŽFig. 1..
Table 2 summarizes the obstetric outcome,
main complications and post-partum stay in the
hospital. The number of patients of gestational
age F 32 weeks or ) 32 weeks were not signifi-
cant between HELLP syndrome groups. There
was no association between the severity of HELLP
syndrome and cesarean delivery. Indications for
cesarean section included fetal distress, malpre-
sentations, previous cesarean section, failed
prostaglandin᎐oxytocin induction of labor, unfa-
vorable cervix and deteriorating maternal condi-
tion. Fifty-three patients had epidural anesthesia,
the range of platelet count in this group of
patients was 19 000᎐150 000r␮l.
The 86 pregnancies resulted in 92 births Žfour
sets of twins and one set of triplets.. Birth weight
as a reflection of gestational age trended down-
wards with progressively severe HELLP syn-
drome, but these differences did not vary signifi-
cantly among the groups in this study; this simi-
larly occurred with intrauterine growth restriction
ŽTable 2.. There were seven stillbirths and eight
neonatal deaths for a perinatal mortality rate of
163 per 1000 Ž16%..
Abruptio placentae Ž12%. and acute renal fail-
ure Ž12%. were the most frequent maternal com-
plications, followed by disseminated intravascular
coagulation Ž5%.. Patients with class 3 HELLP
syndrome did not have severe maternal complica-
tions. Overall, 29% of all patients required blood
or blood products to correct hypovolemia, anemia
or coagulopathy. Transfusion of some type of
blood product was considered to be clinically indi-
cated for 47% of patients with class 1 HELLP
syndrome, for 29% with class 2 HELLP syndrome
and for only 13% of patients with class 3 HELLP
syndrome ᎏ a clinically significant trend Ž P-
0.05..
There were two maternal deaths in this group.
One patient had eclampsia, cerebral edema and
cardiac arrest. The remaining maternal death had

Table 2
Pregnancy outcome of 86 women with HELLP syndrome

HELLP Žall. 1 HELLP 2 HELLP 3 HELLP

Ž N s 86. Ž N s 21. Ž N s 42. Ž N s 23.

Gestational age Žweek.

- 28 Ž%. 5.8 5 7 4
28.0᎐32.0 Ž%. 25.6 43 19 22
32.1᎐36.0 Ž%. 28 14 36 26
) 36.1 Ž%. 40.6 38 38 47
Cesarean delivery Ž%. 71 71 71 70
Birth weight Žgm. 1938 " 777 1768 " 771 1974 " 772 2018 " 805
Intrauterine grown restr a Ž%. 8 17 7 4
Abruptio placentae Ž%. 12 5 21 0
Acute renal failure Ž%. 12 9.5 19 0
P-P-hospitalization days Ž%. a
4᎐6 63 57 55 83 b
7᎐10 29 33 36 13
G 11 8 10 9 4
a
restr s restriction; P-P s post-partum.
b
P- 0.02: 3 HELLP vs. 1 HELLP and 2 HELLP.
 P. Vigil-De Gracia r International Journal of Gynecology and Obstetrics 72 (2001) 17᎐23
severe pre-eclampsia, pulmonary edema, dissemi-
nated intravascular coagulopathy, acute renal fail-
ure and death by multiple organ failure.
The post-partum stay is depicted in Table 2.
The post-partum stay diminished significantly with
the class 3 HELLP syndrome patient group Ž Ps
0.02..
Table 3 describes the presenting symptoms of
women with HELLP syndrome for each group.
Right-upper quadrant or epigastric pain, visual
symptoms and hematuria increased in likelihood
as the severity of HELLP syndrome worsened.
Hematuria Ž‘urine with apple-drink color’. was
the fourth symptom in all HELLP syndrome
groups, but significantly Ž Ps 0.002. associated
with class 1 HELLP syndrome.
4. Comment
The debate concerning HELLP syndrome in
pre-eclampsia᎐eclampsia has been the subject of
several reports w4᎐7x. In our hospital, we used the
triple-classification of HELLP syndrome w6x to
standardize the evaluation of these patients and
to permit comparative assessment of a number of
parameters pertinent to clinical management
concerns. The incidence among patients with se-
vere pre-eclampsia᎐eclampsia in this study was
16%. These high incidences emphasize our role
as a national referral institution. A high and
significant association between HELLP syndrome
and eclampsia was observed Ž34.6%.. This strong
association was reported previously by Miles et al.
w8x. These findings are in contrast to those of
Table 3
Presenting symptoms of 86 women with HELLP syndrome
HELLP Žall.
Ž N s 86.
Headache Ž%. 73
Right-upper quadrant
or epigastric pain Ž%. 45
Visual symptoms Ž%. 29
Hematuria Ž%. 22
Tinnitus Ž%. 12
a
Ps 0.002.
21
Sibai et al. w4x who reported a weak association
between HELLP syndrome and eclampsia.
Probably its presence may be a predisposing fac-
tor in development of eclampsia, but its severity
was not associated with eclampsia. Nevertheless,
in our study, the number of eclampsia cases in
each group was small.
Interestingly, the mean maternal age Ž29.18"
6.19. and nulliparity Ž43%. was very different with
respect to other studies w2,4,6,9x. In our series,
patients with HELLP syndrome were 5᎐6 years
or more older than those reported by Weinstein
w2x, Sibai et al. w4x and Martin et al. w6x. The
incidence of the syndrome was higher among
multiparous patients Ž57%.; these findings are in
contrast to previous findings w2,6,9x. The differ-
ence between our results and those of others in
the E.U.A could be attributed in part, to differ-
ences in the populations; and multiparity was
associated with a trend toward older mothers.
The clinical presentations of patients with
HELLP syndrome varied, related in part to the
severity. Headache, epigastric pain, visual symp-
toms and hematuria appeared to increase in fre-
quency as the severity of HELLP syndrome wors-
ened. Interestingly, patients who had hematuria
in general were more ill, had a lower platelet
count, had higher laboratory values reflective of
hepatic dysfunction and hemolysis, and also had
the highest maternal morbidity and mortality.
It is important to appreciate that severe hyper-
tension ŽG 160r110 mmHg. is not a constant in
HELLP syndrome; 15% of the 86 patients in our
study had a diastolic blood pressure of F 90
mmHg. This finding is in agreement with those
1 HELLP 2 HELLP 3 HELLP
Ž N s 21. Ž N s 42. Ž N s 23.
81 71 70
62 40 39
33 31 22
48 a 22 0
14 9.5 13
22 P. Vigil-De Gracia r International Journal of Gynecology and Obstetrics 72 (2001) 17᎐23
reported by others w2,4,10᎐13x. Furthermore, the
level of proteinuria was variable: 48 women Ž56%.
had proteinuria 3᎐4 q ; 29 had 1᎐2 q ; and 10.5%
had no proteinuria. Thus, it is important that all
health professionals caring for pregnancies know
that in some women with HELLP syndrome, hy-
pertension and proteinuria may be absent or
slight.
The categorization of HELLP syndrome
patients with a platelet nadir of less than
150 000r␮l into class 1, class 2 and class 3 groups
facilitated clinical management and was highly
significant Ž P- 0.001.. Thus, the triple-classifica-
tion of HELLP syndrome is based on the platelet
count w6x; the use of LDH, AST, ALT by HELLP
syndrome patients group had a significant differ-
ence ŽTable 1.. Serum AST and ALT changed
significantly between classes of HELLP syn-
drome, according to the severity. The current
study has revealed that levels of LDH, AST, and
ALT serve, and might have utility in, the Missis-
sippi triple-classification system together with
platelet count. Consequently, with respect to the
classification of HELLP syndrome, it is my opin-
ion that LDH values Žhemolysis ᎏ H., the degree
of abnormality of liver enzymes ŽAST, ALT-EL.
should be included, and obviously, the platelet
count ŽLP..
Fifty-three patients in the current study re-
ceived epidural anesthetics and there was no
complication associated with this procedure. The
incidence of cesarean section did not increase
with the severity of HELLP syndrome. This find-
ing is similar to that reported by Magann et al.
w14x. There was no significant difference among
the three birth weight means, but the birth weight
trended downward with progressively severe
HELLP syndrome. Birth weight is a reflection of
gestational age; however, because the birth weight
and gestational age did not vary significantly
among groups, the downward trend appeared to
arise from the number of intrauterine growth
restrictions and the severity of HELLP syndrome.
This study supports our previous observation
that HELLP syndrome is associated with in-
creased maternal morbidity and mortality w5x.
Laboratory evidence of disseminated intravascu-
lar coagulation was found in only four Ž5%. of 86
patients with HELLP syndrome. These findings
are in contrast to those of Sibai et al. w4,9,10x who
reported a high incidence of disseminated in-
travascular coagulation Ž15᎐38%. in patients with
HELLP syndrome. There were two maternal
deaths in this group of HELLP syndrome Ž2.3%.;
one with evidence of disseminated intravascular
coagulation, and both had class 1 HELLP. Inter-
estingly, there were no severe maternal complica-
tions in patients with class 3 HELLP syndrome.
Finally, approximately half of the patients in
the class 1 HELLP group had significant maternal
morbidity. Our data suggest that certain sub-
groups of patients with class 1 HELLP syndrome
Ž‘classic or true HELLP’. are at increased risk for
serious maternal complications, including those
with platelet counts below 50 000r␮l, LDH)
2000 IUrl, AST) 500 IUrl, ALT) 300 IUrl,
and hematuria.
References
w1x National High Blood Pressure Education Program
Working Group, Report on high blood pressure in preg-
nancy. Am J Obstet Gynecol 1990;163:1691᎐1712.
w2x Weinstein L. Syndrome of hemolysis, elevated liver en-
zymes, and low platelet count: a severe consequence of
hypertension. Am J Obstet Gynecol 1982;142:159᎐167.
w3x Barton JR, Sibai BM. Care of the pregnancy compli-
cated by HELLP syndrome. Obstet Gynecol Clin North
Am 1991;18:165᎐179.
w4x Sibai BM. Maternal morbidity and mortality in 442
pregnancy with hemolysis, elevated liver enzymes, and
low platelets ŽHELLP syndrome.. Am J Obstet Gynecol
1993;169:1000᎐1006.
w5x Vigil-De Gracia PE, Tenorio-Maranon ˜´ RF, Cejudo-Car-
´ ´
ranza E, Helguera-Martinez A, Garcıa-Caceres E. Dif-
ferences among pre-eclampsia, HELLP syndrome and
eclampsia. Maternal evaluation. Gin Obstet Mex 1996;
64:337᎐382.
w6x Martin Jr. JN, Blake PG, Perry Jr. KG, McCaul JF,
Hess LW, Martin RW. The natural history of HELLP
syndrome: patterns of disease progression and regres-
sion. Am J Obstet Gynecol 1991;164:1500᎐1513.
w7x Barton JR, Sibai BM. Pregnancy and liver disease.
HELLP and liver disease of pre-eclampsia. Clin Liver
Dis 1999;3:31᎐48.
w8x Miles Jr. JF, Martin Jr. JN, Blake PG et al. Post-partum
eclampsia: a recurring perinatal dilemma. Obstet Gy-
necol 1990;76:328᎐331.
w9x Audibert F, Friedman SA, Frangieh AY, Sibai BM.
Clinical utility of strict diagnostic criteria for the HELLP
 P. Vigil-De Gracia r International Journal of Gynecology and Obstetrics 72 (2001) 17᎐23
Žhemolysis, elevated liver enzymes, and low platelets.
syndrome. Am J Obstet Gynecol 1996;175:460᎐464.
w10x Sibai BM. Maternal-perinatal outcome associated with
the syndrome of hemolysis, elevated liver enzymes, and
low platelets in severe pre-eclampsia᎐eclampsia. Am J
Obstet Gynecol 1986;155:501᎐507.
w11x Aarnoudse JG, Houthoff HJ, Wieits J, Vallenga E,
Huisjes HJ. A syndrome of liver damage and intravascu-
lar coagulation in the last trimester of normotensive
pregnancy. A clinical and histopathological study. Br J
Obstet Gynaecol 1986;93:145᎐155.
w12x Mackenna J, Dover NL, Brame RG. Pre-eclampsia asso-
ciated with hemolysis, elevated liver enzymes, and low
23
platelets ᎏ an obstetric emergency? Obstet Gynecol
1983;62:751᎐754.
w13x Martin Jr. JN, Blake PG, Lowry SL, Perry Jr. KG, Files
JC, Morrison JC. Pregnancy complicated by pre-
eclampsia᎐eclampsia with the syndrome of hemolysis,
elevated liver enzymes, and low platelet count: how
rapid is post-partum recovery? Obstet Gynecol 1990;
76:737᎐741.
w14x Magann EF, Perry Jr. KG, Chauhan SP, Graves GR,
Blake PG, Martin Jr. JN. Neonatal salvage by weeks’
gestation in pregnancies complicated by HELLP syn-
drome. J Soc Gynecol Invest 1994;1:206᎐209.