REVIEW ARTICLE

Using case reports to determine when liver bleeding occurs during

disease
progression in HELLP syndrome
Marie Darby
a,
, James N. Martin Jr
a
, Sarah Q. Mitchell
b
, Michelle Y. Owens
a
, Kedra Wallace
b
a
Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Mississippi Medical Center, Jackson, USA
b
Department of Obstetrics and Gynecology, University of Mississippi Medical Center, Jackson, USA
a b s t r a c t a r t i c l e i n f o
Article history:
Received 7 December 2012
Received in revised form 26 March 2013
Accepted 21 June 2013
Keywords:
Dexamethasone
Hemolysis, elevated liver enzymes, low platelets
Liver hematoma
Liver rupture
Background: Hepatic hemorrhage occurs in less than 5% of patients with hemolysis, elevated liver enzymes,
low platelets (HELLP) syndrome but it is a profound cause of maternal/perinatal morbidity and mortality.
Objectives: To determine when liver bleeding occurs during the development of HELLP syndrome. Search
strategy: The English literature was searched for all reports of HELLP syndrome associated with liver bleeding.
Selection criteria: Eighty-seven case summaries of liver bleeding in the setting of HELLP syndrome were in-
cluded. The standard denition of HELLP syndrome was used with expansion into the Mississippi classication
system, supplemented by patients with partial HELLP syndrome. Data collection and analysis: Demographic and
clinical data were collected and recorded in an Excel database. Main results: Liver bleeding was detected in 18
(20.7%) patients with class 1 HELLP syndrome, 24 (27.6%) with class 2 HELLP syndrome, and 12 (13.8%) with
class 3 or partial HELLP syndrome. In 33 (37.9%) patients, the exact class of HELLP syndrome at the time liver
bleeding was detected could not be determined from the published descriptions. Conclusions: Liver bleeding
can occur early during HELLP syndrome development, not only in patients with advanced, class 1 illness.
2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Pregnancy-associated hepatic hemorrhage and rupture, in the ab-
sence of direct maternal trauma, is encountered almost exclusively
(98%) in patients with pre-eclampsia [1]. Among the 141 patients
with liver bleeding as hepatic hemorrhage and/or rupture described
in English literature reports between 1960 and 1997, more than 75%
had evidence of hemolysis, elevated liver enzymes, low platelets
(HELLP) syndrome. Although rare, pregnancy-associated liver bleeding
seems to occur most often in patients with atypical pre-eclampsia/
eclampsia and HELLP syndrome where there is evidence of hemolysis,
elevated liver enzymes, and thrombocytopenia [2,3]. As a complication
of HELLP syndrome, liver bleeding occurs in less than 5% of cases but,
when it does occur, it is a profound cause of maternal and perinatal
morbidity and mortality [1,4].
It is unclear when signicant hepatic micro- and/or macro-injury
leading to liver bleeding is initiated during the natural history of
HELLP syndrome development. Over the past 20 years, a spectrum
of complete or incomplete/partial HELLP syndrome, as well as non-
HELLP severe pre-eclampsia, has been described. Evaluation of these
patients has revealed that those with the most advanced stage of
HELLP syndrome (class 1; Table 1) have the highest prevalence of
major maternal morbidity, including hepatic injury and bleeding
[5,6]. However, it has been our experience that signicant liver injury
with liver bleeding can precede the development of complete HELLP
syndrome (class 1, class 2) and is not just a feature of advanced, class
1 HELLP.
The aim of the present review was to determine when liver bleed-
ing occurs as HELLP syndrome develops.
2. Materials and methods
The English literature was searched for HELLP AND liver hemato-
ma, HELLP AND liver rupture, pregnancy AND liver hematoma,
and pregnancy AND liver rupture in order to identify all recently
published reports of patients with HELLP syndrome-associated liver
bleeding as hepatic hemorrhage and/or rupture. These publications
were then reviewed to determine whether, using the triple-class classi-
cation system for HELLP syndrome, the timing of liver bleeding could
be identied relative to the stage of HELLP syndrome development.
The following information was collected and recorded in an Excel
(Microsoft, Redmond, WA, USA) database: demographic data; gesta-
tional age at delivery; mode of delivery; laboratory values relevant to
HELLP syndrome (platelet count, hematocrit, lactate dehydrogenase,
aspartate aminotransferase if available); most advanced HELLP class
recorded; presence/timing of symptoms; presence/absence of liver rup-
ture; diagnostic modalities used to detect liver bleeding; perinatal out-
comes; and maternal deaths. Interventions noted included transfusion
of blood components, use of intravenous (IV) dexamethasone or other
International Journal of Gynecology and Obstetrics 123 (2013) 79
Corresponding author at: Division of Maternal-Fetal Medicine, Department of Obstetrics
and Gynecology, 2500 North State Street, Jackson, MS 39216, USA. Tel.: +1 6019845358;
fax: +1 6018157094.
E-mail address: mmdarby@umc.edu (M. Darby).
0020-7292/$ see front matter 2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijgo.2013.06.007
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics
j our nal homepage: www. el sevi er . com/ l ocat e/ i j go
potent glucocorticoid, factor VIIa administration, abdominal packing,
and liver transplant.
The standard denition used for HELLP syndrome (complete
HELLP syndrome) was that recommended by Martin and Sibai [2,3],
with expansion into the Mississippi classication system(Table 1). Pa-
tients with a diagnosis of partial HELLP syndrome were required to
have evidence of pre-eclampsia with 2 of the 3 laboratory criteria
present for complete HELLP syndrome. Owing to the variety of labora-
tory systems used to measure lactate dehydrogenase and aspartate
aminotransferase in the published studies, the present analysis used
platelet count as the principal marker to reect the stage and progres-
sion of HELLP syndrome severity. Associations between lowest HELLP
class and outcomes/interventions were analyzed via
2
test. All statis-
tical analysis was performed using STATA version 12.1 (StataCorp,
College Station, TX, USA). Missing data were not included in the analysis.
3. Results
The initial literature search yielded 73 potentially useful English-
language reports published between 1976 and 2011 inclusive. Among
these were 49 reports containing 87 case summaries detailing liver
bleeding in patients with HELLP syndrome (Supplementary Material
S1). Among these 87 patients, liver bleeding as liver hematoma or rup-
ture was detected in 18 (20.7%) women with class 1 HELLP syndrome,
24 (27.6%) with class 2 HELLP syndrome, and 12 (13.8%) with class 3
or partial HELLP syndrome (Fig. 1). In 33 (37.9%) patients, the exact
class of HELLP syndrome at the time liver bleeding was detected could
not be determinedfromthe publisheddescriptions. Following detection
of liver bleeding as hepatic hemorrhage, 46 (54.1%) patients in the
series proceeded to frank hepatic rupture. The presence or absence of
liver rupture was not associated with the timing of diagnosis relative
to HELLP class, lowest HELLP class subsequently recorded, need for
interventions (blood transfusion, dexamethasone, factor VIIa adminis-
tration, abdominal packing, transplant), maternal death, or adverse
perinatal outcome.
Demographics for the study group included a mean maternal age of
30.2 4.4 years (range, 2143 years); 32 (36.8%) were nulliparous.
For 75 (86.2%) patients, race was indeterminate from published re-
ports. Mean gestational age at delivery was 33 5.3 weeks (range,
1840 weeks). Most patients with liver bleeding were delivered by
cesarean (53 [60.9%]). Fetal wastage was high, with 21 (24.1%) cases
involving fetal demise and 3 (3.4%) involving neonatal death. Live
birth occurred in 41 (47.1%) cases, while perinatal outcome was not
reported in 22 (25.3%). The diagnosis of liver bleeding in the present
series of 87 patients was made most often by some form of imaging
(50 [57.5%]) (Fig. 2).
4. Discussion
The present report attempted to associate presence of liver bleed-
ing with stage of HELLP syndrome using the Mississippi triple-class
classication system. Based on the data available from the English-
language literature regarding liver bleeding and HELLP syndrome
since 1976, it seems that liver bleeding as liver hemorrhage and/or
rupture can occur at any time in the spectrum of HELLP syndrome
development (i.e. class 1, class 2, or class 3/incomplete/partial HELLP
stages). The assumption that liver bleeding is present only in the most
advanced cases of HELLP syndrome (i.e. class 1) seems to be erroneous.
Instead, it appears that liver bleeding as hepatic hematoma and/or
rupture is encountered both in women with end-stage class 1 HELLP
syndrome (as expected) and, perhaps more importantly, in some pa-
tients relatively early in the development of diseaseeven before
widely accepted criteria for complete HELLP syndrome (class 1 or
class 2) are present. It is possible that early evidence of liver bleeding
could have been detected in more of the patients in the present series
had they presented for care earlier.
The present study was limited by its retrospective structure, the
relatively small number of useful case reports, and its reliance on
sometimes incomplete data contained within the available publica-
tions. It provides an argument for appealing to authors of reports on
patients with HELLP syndrome to render more detailed accounts
in the future and to use the Mississippi classication system when
reporting ndings. However, these shortcomings are to be expected
because liver rupture/hematoma in patients with HELLP syndrome
is relatively rare, occurring in less than 5% of cases and only 1 per
250 000 deliveries [7].
Having determined retrospectively that liver bleeding can be
detected early in the disease course of HELLP syndrome, it may be a
daunting exercise to explore the efcacy of an intervention to inter-
rupt the further progression of liver bleeding after it is identied. As
is the case in most investigations, more questions are raised than an-
swered by the present study ndings. The observation that the disease
course of a patient in a previous protocol study of HELLP syndrome
Table 1
Mississippi classication system for HELLP syndrome [3,7,9,10].
Platelets per L LDH, IU/L AST or ALT, IU/L
Class 1 HELLP 50 000 600 70
Class 2 HELLP N50 000 to 100 000 600 70
Class 3 HELLP N100 000 to 150 000 600 40
Abbreviations: ALT, alanine transaminase; AST, aspartate aminotransferase; HELLP,
hemolysis, elevated liver enzymes, low platelets; LDH, lactate dehydrogenase.
Fig. 1. Stage of hemolysis, elevated liver enzymes, low platelets syndrome relative
to time that liver bleeding is detected as a reection of liver hematoma formation or
rupture among 87 patients.
Ultrasound
20%
CT
30%
Sequential
Ultrasound and
CT
8%
Exploratory
Laparotomy
32%
Autopsy
2%
Not reported
8%
Fig. 2. Method of diagnosis for liver bleeding.
8 M. Darby et al. / International Journal of Gynecology and Obstetrics 123 (2013) 79
management using IV dexamethasone [7] was arrestedwith no pro-
gression of liver bleeding beyond class 3 HELLP syndromeraises the
question of whether this intervention might reduce the progression
of liver bleeding as hepatic hemorrhage/rupture below the 55% level
observed in the present patient aggregate. It would take a very large
prospective multicenter trial to investigate adequately the effective-
ness of such an intervention. The use of IV dexamethasone in combi-
nation with IV magnesium sulfate and strict blood pressure control
embedded within the Mississippi protocol has been shown to signif-
icantly reduce composite major maternal disease as HELLP syndrome
(poor laboratory parameters, disease progression to class 1 HELLP
syndrome, degree of hypertension, need for antihypertensive therapy,
use of transfusion, prolonged recovery) in more than 400 studied pa-
tients over the past 14 years [7,8].
The ndings fromthe present study illustrate the value of using the
triple-class classication system for patients with HELLP syndrome in
terms of presentation and interpretation of ndings. HELLP syndrome
is a spectral condition that begins with few apparent laboratory or
clinical abnormalities; thereafter, it can progress variably through
stages to the most advanced level of class 1 disease. More than 2 de-
cades ago, it was rst shown that class 1 HELLP syndrome is associated
with the highest maternal and perinatal morbidity and mortality com-
pared with the other classes; this has been conrmed more recently
with data from 2000 to 2011 [3,7,9,10]. Study of all published cases
of liver bleeding with HELLP syndrome in relation to initial presenta-
tion timing and the stage of developing HELLP syndrome facilitated
the documentation of liver bleeding as an event that can become evi-
dent anywhere along the spectrum of HELLP syndrome development
and progression. Thus, composite maternal morbidity is highest in
class 1 HELLP, but hepatic morbidity is a potential risk at all stages of
HELLP syndrome.
Evidence of signicant liver injury with liver bleeding can even
precede the full clinical laboratory expression of complete HELLP
syndrome. In patients who develop liver bleeding with HELLP syn-
drome, it is usually evident before class 1 HELLP syndrome is reached.
There is a need for further investigation of therapies to interrupt the
progression of HELLP syndrome and to prevent the consequences of
maternal morbidity and mortality associated with progression to
class 1 disease, including liver bleeding.
Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.ijgo.2013.06.007.
Conict of interest
The authors have no conicts of interest.
References
[1] Rinehart BK, Terrone DA, Magann EF, Martin RW, May WL, Martin Jr JN. Preeclampsia-
associated hepatic hemorrhage and rupture: mode of management related to maternal
and perinatal outcome. Obstet Gynecol Surv 1999;54(3):196202.
[2] Sibai BM, Ramadan MK, Usta I, Salama M, Mercer BM, Friedman SA. Maternal
morbidity and mortality in 442 pregnancies with hemolysis, elevated liver en-
zymes, and low platelets (HELLP syndrome). Am J Obstet Gynecol 1993;169(4):
10006.
[3] Martin Jr JN, Rinehart BK, May WL, Magann EF, Terrone DA, Blake PG. The spec-
trumof severe preeclampsia: comparative analysis by HELLP (hemolysis, elevated
liver enzyme levels, and low platelet count) syndrome classication. Am J Obstet
Gynecol 1999;180(6 Pt 1):137384.
[4] Zarrinpar A, Farmer DG, Ghobrial RM, Lipshutz GS, Gu Y, Hiatt JR, et al. Liver trans-
plantation for HELLP syndrome. Am Surg 2007;73(10):10136.
[5] Martin Jr JN, Thigpen BD, Moore RC, Rose CH, Cushman J, May W. Stroke and se-
vere preeclampsia and eclampsia: a paradigm shift focusing on systolic blood
pressure. Obstet Gynecol 2005;105(2):24654.
[6] MartinJr JN, Files JC, Blake PG, Perry Jr KG, MorrisonJC, NormanPH. Postpartumplas-
ma exchange for atypical preeclampsia-eclampsia as HELLP (hemolysis, elevated
liver enzymes, and low platelets) syndrome. Am J Obstet Gynecol 1995;172(4 Pt 1):
110727.
[7] Martin Jr JN, Owens MY, Keiser SD, Parrish MR, Tam Tam KB, Brewer JM, et al.
Standardized Mississippi Protocol treatment of 190 patients with HELLP syn-
drome: slowing disease progression and preventing new major maternal morbid-
ity. Hypertens Pregnancy 2012;31(1):7990.
[8] Martin Jr JN, Thigpen BD, Rose CH, Cushman J, Moore A, May WL. Maternal benet
of high-dose intravenous corticosteroid therapy for HELLP syndrome. Am J Obstet
Gynecol 2003;189(3):8304.
[9] Martin Jr JN. Milestones in the quest for best management of patients with HELLP
syndrome (microangiopathic hemolytic anemia, hepatic dysfunction, thrombocy-
topenia). Int J Gynecol Obstet 2013;121(3):2027.
[10] Martin Jr JN, Brewer JM, Wallace K, Sunesara I, Canizaro A, Blake PG, et al. HELLP
syndrome and composite major maternal morbidity: importance of Mississippi
classication system. J Matern Fetal Neonatal Med 2013 in press.
9 M. Darby et al. / International Journal of Gynecology and Obstetrics 123 (2013) 79