Keywords HELLP syndrome - arterial hypertension - hypertensive crisis - cardiovascular risk
- hypertension in pregnancy
Schlüsselwörter HELLP-Syndrom - arterielle Hypertonie - hypertensive Krise - kardiovaskuläres Risiko
- Hypertonie in der Schwangerschaft
Background
HELLP (hemolysis, elevated liver enzymes, and low platelets) syndrome is an important
clinical challenge due to the limited therapeutic options and high mortality rates
of both mother and child [1 ]
[2 ]
[3 ]
[4 ]. HELLP syndrome occurs in 0.2–0.8% of all pregnancies and 70–80% are associated
with preeclampsia [5 ]. With regards to its etiopathogenesis, there are similarities between hypertensive
pregnancy disorders such as preeclampsia and gestational hypertension and HELLP syndrome.
Some authors classify HELLP syndrome as a complication of preeclampsia. But although
hypertension is mandatory for the definition of preeclampsia, HELLP
syndrome can be normotensive [6 ]. Poor placentation is believed to be responsible for the development of HELLP syndrome
as it can lead to systemic inflammatory disorder initiated by the complement system
[7 ]. However, the pathogenesis of HELLP syndrome is still not fully understood, and
the only therapy is delivery of the child. Delivery is regularly initiated from 34
weeks of gestational age; between 24 and 34 weeks, the timing of delivery is based
on an evaluation of possible neonatal versus maternal morbidity and mortality.
The maternal outcome is often unpredictable, particularly as relapses are common and
the course may be progressive [8 ]
[9 ]. HELLP syndrome mostly occurs between weeks 27 and 37 of gestation. However, up
to one third of cases develop postpartum within the first 48 hours after delivery
[9 ]. The clinical symptoms vary and include right-sided epigastric pain, nausea, and
hypertension, and neurological symptoms such as visual disturbances and headaches.
Possible maternal complications are liver hematoma, rarely leading to hepatic capsular
rupture, acute kidney failure, eclampsia, disseminated intravascular coagulation (DIC),
intracerebral hemorrhage, and hypertensive crises [6 ]. Complications affecting
the course of pregnancy are placental abruption and intrauterine fetal death [2 ]
[3 ].
Despite its fulminant course and potential for complications, there still is no parameter
which can predict the outcome of women with HELLP syndrome. Existing classification
systems such as the Tennessee/Mississippi classification are not used in routine clinical
care, as there is no reliable correlation between laboratory results and the outcome
of women with HELLP syndrome [9 ]
[10 ]
[11 ]. Therefore, timing the delivery correctly, especially in women with early-onset
HELLP syndrome, remains challenging. Decisions are not only based on clinical evidence
but on subjective considerations of maternal risk and the child’s risk of preterm
birth. Thus, clinical markers and an algorithm for decision-making are needed to estimate
the severity of the individual course of
disease and decide whether prolongation of the pregnancy is justified or delivery
should be initiated.
A previous study focusing on preeclampsia pointed out that the likelihood of an adverse
maternal outcome is elevated when maternal blood pressure is high [12 ]. Based on this study, we hypothesized that patients with HELLP syndrome and hypertension
have a higher risk of adverse outcomes compared to normotensive patients. This study
aims to evaluate the severity of hypertension as a potential marker of maternal outcome
in women with HELLP syndrome. Due to the close relationship between HELLP syndrome
and preeclampsia, we sought to identify the effect of maternal blood pressure as well
as additional associated parameters on the outcome of HELLP patients to optimize individual
risk stratification.
Material and Methods
Patients included in the study and data collection
In this retrospective cohort study carried out at the University Hospital Münster,
a tertiary care obstetric center, we reviewed all births between 1 January 2010 and
31 December 2020. This study was designed in accordance with the Declaration of Helsinki
and approved by the Ethics Committee of the local medical council and the University
of Münster (trial no. 2023–074-f-S).
To identify all patients with HELLP syndrome in the hospital’s database, we searched
the documentation system using the ICD-10 codes O10–O16. These codes include chronic
hypertension with complications in pregnancy, hypertensive pregnancy disorders, preeclampsia,
eclampsia and HELLP syndrome. We then proceeded as shown in [Fig. 1 ] and excluded all duplicates, all patients who did not meet the criteria for a diagnosis
of HELLP syndrome, and all patients for whom data was incomplete.
Fig. 1
Flow chart of patients screened and included in the study. Abbreviations: PE = preeclampsia,
IUFD = intrauterine fetal death.
A total of 59 women who met the following diagnostic criteria for HELLP syndrome were
included in the study. The criteria for HELLP syndrome listed below are based on the
definition of HELLP syndrome of the American College of Obstetricians and Gynecologists
(ACOG) and the Tennessee Classification [9 ]
[13 ]:
Low Platelet count : platelet count < 100000 cells/µl
Hemolysis : lactate dehydrogenase (LDH) ≥ twice the upper limit for normal values and/or haptoglobin
below normal values
Elevated liver enzymes : alanine aminotransferase (GPT) and/or aspartate aminotransferase (GOT) elevated
to more than twice the upper limit for normal values.
We excluded three cases of intrauterine fetal death (IUFD) and two terminations of
pregnancy (due to maternal conditions) because of incomplete data. These patients
were either transferred to our hospital after a diagnosis of IUFD and/or were primarily
treated in another department and complete data collection for this study was therefore
not possible.
We collected the following data from the clinical records of all patients with HELLP
syndrome: demographic data, medical history, blood pressure recordings, laboratory
results, clinical symptoms, and obstetric outcomes. The clinical symptoms we focused
on were epigastric pain, headaches, hyperreflexia, visual disturbances, and other
neurological symptoms (e.g., vomiting, nausea).
Hypertension groups
According to previous studies on maternal outcomes with preeclampsia and based on
the current international guidelines on arterial hypertension, we divided the patients
into three groups according to their maximum prepartum blood pressure [12 ]
[13 ]
[14 ]:
Group 1 included all patients with mild to moderate or no hypertension (blood pressure
below 160/110 mmHg).
Group 2 consisted of patients with severe hypertension (systolic blood pressure between
160 and 179 mmHg and/or diastolic pressure between 110 and 119 mmHg)
Group 3 included patients with hypertensive crisis (systolic blood pressure ≥ 180 mmHg
and/or diastolic blood pressure ≥ 120 mmHg).
The Results section presents patient characteristics such as mean age, BMI, pre-existing
medical conditions, laboratory results, and HELLP-specific symptoms and compares the
characteristics of the three hypertension groups. Other presented data include birth
and perinatal outcome, including gestational age at delivery, mode of delivery, birth
weight, and APGAR score. Maternal outcomes are presented, and blood pressure and antihypertensive
medication at discharge from hospital are shown for the different groups, as well
as postpartum anemia and eclampsia. Lastly, an individual risk stratification based
on maternal outcome and severity of hypertension is discussed.
Endpoint for adverse maternal outcome
A combined endpoint was used to analyze adverse maternal outcomes. The endpoint included
the following criteria: blood pressure above 140/90 mmHg at discharge, necessity of
anti-hypertensive medication at discharge, postpartum hemoglobin levels below 8 g/dl,
or onset of eclampsia as a complication of HELLP syndrome. To confirm our hypothesis,
we tested the probability of adverse maternal outcomes in the hypertension groups.
Investigations also included analyzing the correlation between systolic and diastolic
blood pressure and each outcome parameter individually as well as for the combined
endpoint.
Statistical analysis
All statistical calculations were performed using SPSS Statistics, version 28 (IBM,
Armonk, NY, USA). Mean values and standard deviation were used for descriptive analysis
of continuous variables. Categorical data were expressed as frequencies/percentages.
Inferential analyses of differences between the three groups were performed using
one-way ANOVA for normally distributed continuous variables, Kruskal–Wallis test for
non-normally distributed ordinal or continuous variables, and Fisher’s exact test
for categorical variables. In addition, the datasets for the hypertension groups were
preselected to carry out a pairwise Fisher’s exact test to compare significant variables
between the groups. Using closed testing procedures, all non-significant variables
were not tested pairwise [15 ]. Univariable binary logistic regression was done to evaluate the statistical relationship
between blood pressure and
adverse maternal outcome. The Box–Tidwell test was performed to verify linearity as
an assumption for logistic regression. A p value ≤ 0.05 was considered statistically
significant for all statistical tests. The results should be considered as exploratory
as no adjustment for multiple testing was done.
The program R (The R Foundation for Statistical Computing, Vienna, Austria, https://www.R-project.org/ ) was used to plot the odds ratios in forest plots.
Results
Patient characteristics
The prevalence of HELLP syndrome in all women who gave birth in University Hospital
Münster between 2010 and 2020 was 0.45%.
The mean maternal age of the study population of 59 patients with HELLP syndrome was
31.27 ± 4.54 years. The mean BMI was 26.80 ± 5.42 kg/m2 ; 22% of patients had a preconception BMI of more than 30 kg/m2 . Most women were primiparous (66.1%) and had no pre-existing medical conditions (66.1%).
Five of seven women (11.9%) with a history of pre-existing hypertension were on long-term
antihypertensive medication prepartum. 10.2% of patients had a coagulation disorder
and three women had suffered from thrombosis in the past. While none of the patients
had pre-existing diabetes, three developed gestational diabetes in pregnancy. One
out of five pregnancies in our cohort was conceived by assisted reproductive technology.
There were nine twin pregnancies (15.3%) and 14 cases with fetal growth restriction
(23.7%). Fifteen patients (25.4%) developed HELLP syndrome peri- or postpartum.
Mean systolic blood pressure was 178.53 mmHg ± 20.42 and mean diastolic blood pressure
was 107.69 mmHg ± 14.13. The most common symptom was right upper quadrant abdominal
pain or epigastric pain (72.9%), and 59.3% of all women suffered from headaches. Visual
impairment (15.3%), hyperreflexia (18.6%), and other neurological symptoms (15.3%)
were less common. In total, 7 cases (11.9%) had none of the above-listed symptoms.
With regards to HELLP-specific laboratory results, all patients showed distinct deviations
from the norm, as this was part of the inclusion criteria. The mean value of the lowest
platelet level measured during the hospital stay was 55.5 × 103 /µl ± 20.43. The mean value of the highest measured GPT levels was 484.78 U/l ± 887.39
and the mean value of the highest LDH levels was 1266.03 U/l ± 1596.80. In addition,
23.7% of women had elevated creatinine levels above the upper limit of normal.
IUFD/termination of pregnancy due to HELLP syndrome
Due to incomplete data, patients with IUFD or termination of pregnancy were not included
in the statistical analysis. The mean age of these patients (n = 5) was 32.4 ± 5.13
years, and their mean BMI was 24.5 ± 3.87 kg/m2 . All patients were primiparous. The mean gestational age was 21 + 4 weeks. One patient
had a pre-existing chronic kidney disease and the fetus of one patient had a congenital
malformation. In the hypertension groups, one patient had mild to moderate hypertension,
one had severe hypertension and three patients experienced hypertensive crises.
Patient characteristics in the hypertension groups
Patients were divided into three groups, according to the highest blood pressure measured
before delivery: eight patients (15.25%) had mild to moderate hypertension and one
patient was normotensive (Group 1). Nineteen patients (32.20%) had severe hypertension
(Group 2), and 31 patients (52.54%) experienced at least one hypertensive crisis during
pregnancy (Group 3). [Fig. 2 ] shows the systolic and diastolic blood pressure of every patient in each hypertension
group.
Fig. 2
Scatterplot of systolic and diastolic blood pressure levels for each group according
to the severity of hypertension. Each marker represents one patient. Abbreviation:
BP = blood pressure.
The demographic data of the different blood pressure groups are displayed in [Table 1 ]. Most parameters did not differ significantly between groups. When analyzing clinical
symptoms, women with a hypertensive crisis were more likely to develop headaches (p = 0.039).
Table 1
Patient characteristics in the three blood pressure groups.
Mild to moderate/no hypertension (Group 1), n = 9
Severe hypertension (Group 2), n = 19
Hypertensive crisis (Group 3), n = 31
p value
Abbreviations: GPT = alanine aminotransferase; LDH = lactate dehydrogenase; n.s. =
not significant; SD = standard deviation
Baseline characteristics
Mean age ± SD
31.22 ± 1.64
31.32 ± 4.92
31.26 ± 4.95
n.s. (p = 0.998)
Mean BMI ± SD
24.80 ± 2.43
27.32 ± 4.60
26.83 ± 6.42
n.s. (p = 0.425)
Pre-existing hypertension
1 (11.1%)
3 (15.8%)
3 (9.7%)
n.s. (p = 0.858)
Antihypertensive medication prepartum
0
1 (5.26%)
4 (12.9%)
n.s. (p = 0.550)
Multiple pregnancies
2 (22.2%)
5 (26.3%)
2 (6.5%)
n.s. (p = 0.110)
Assisted reproduction
0
3 (15.8 %)
9 (29.0 %)
n.s. (p = 0.160)
Nulliparous
8 (88.9%)
14 (73.7%)
17 (54.8%)
n.s. (p = 0.099)
Symptoms
No symptoms
1 (11.1%)
4 (21.1%)
2 (6.5%)
n.s. (p = 0.288)
Right upper quadrant abdominal/epigastric pain
6 (66.7%)
12 (63.2%)
25 (80.6%)
n.s. (p = 0.345)
Headache
3 (33.3%)
9 (47.4%)
23 (74.2%)
p = 0.037
p = 0.687 (1 vs. 2)
p = 0.044 (1 vs. 3)
p = 0.073 (2 vs. 3)
Laboratory findings
Potassium (mmol/l) ± SD
4.76 ± 0.45
4.86 ± 0.46
4.92 ± 0.40
n.s. (p = 0.491)
Creatinine (mg/dl) ± SD
1.67 ± 1.43
1.02 ± 0.42
1.29 ± 1.36
n.s. (p = 0.604)
Uric acid (mg/dl) ± SD
7.70 ± 2.23
7.50 ± 1.79
7.60 ± 1.46
n.s. (p = 0.953)
GPT (U/l) ± SD
781.11 ± 1683.56
371.16 ± 476.35
468.40 ± 771.79
n.s. (p = 0.989)
LDH (U/l) ± SD
1302.44 ± 1733.34
951.37 ± 890.62
1448.32 ± 1885.47
n.s. (p = 0.243)
Hemoglobin prepartum (g/dl) ± SD
10.64 ± 1.76
10.79 ± 0.92
11.18 ± 1.69
n.s. (p = 0.282)
Hemoglobin postpartum (g/dl) ± SD
8.26 ± 1.47
9.07 ± 1.67
8.92 ± 1.63
n.s. (p = 0.338)
Platelet count (103 /µl) ± SD
55.30 ± 16.29
53.84 ± 19.90
56.74 ± 22.26
n.s. (p = 0.681)
Birth and perinatal outcome
The mean gestational age at delivery was 33 weeks + 6 days. In total, 72.88% of births
were preterm deliveries and 49.15% of all births were delivered early due to HELLP
syndrome. In 76.27% of cases, delivery was by cesarean section. The mode of delivery
in the different hypertension groups is shown in [Table 2 ]. The mean time between the diagnosis of HELLP syndrome and delivery was 2.78 ± 5.75
days. As regards the perinatal outcome, 44 newborns (74.58%) were admitted to the
neonatal care unit. The mean birth weight was 2020.03 ± 961.52 g and the mean birth
weight percentile was 28.72 ± 21.43. Mean umbilical cord arterial pH was 7.25 ± 0.11,
but 27.11% of all newborns had pH levels of less than 7.2. Fourteen newborns (23.73%)
had an APGAR score of less than 8 after 5 or 10 minutes. When the groups were compared,
no significant differences were found between groups (see [Table 2 ] for details and p values).
Table 2
Pregnancy and birth-related characteristics for the three blood pressure groups.
Mild to moderate/no hypertension (Group 1), n = 9
Severe hypertension (Group 2), n = 19
Hypertensive crisis (Group 3), n = 31
p value
Abbreviations: n.s. = not significant; SD = standard deviation
Pregnancy and birth characteristics
Gestational age at delivery (weeks) ± SD (days)
36 ± 3.11
33 ± 1.82
33 ± 2.36
n.s. (p = 0.119)
Number of preterm deliveries (%)
5 (55.6%)
14 (73.7%)
24 (77.4%)
n.s. (p = 0.481)
Birth weight ± SD (grams)
2705.00 ± 553.46
1848.79 ± 1066.46
2200.00 ± 650.538
n.s. (p = 0.063)
Umbilical cord arterial pH-value ± SD
7.25 ± 0.12
7.24 ± 0.97
7.25 ± 0.111
n.s. (p = 0.832)
APGAR (5 min) ± SD
8.78 ± 2.05
8.00 ± 1.83
8.23 ± 1.5
n.s. (p = 0.185)
APGAR (10 min) ± SD
9.56 ± 1.01
9.11 ± 0.94
9.13 ± 0.89
n.s. (p = 0.249)
Time between diagnosis and delivery ± SD (days)
1.22 ± 1.986
2.53 ± 3.24
3.39 ± 7.451
n.s. (p = 0.605)
Delivery indicated because of HELLP syndrome (%)
6 (66.7%)
7 (36.8%)
16 (51.6%)
n.s. (p = 0.308)
Mode of delivery
Cesarean section (%)
5 (55.6%)
15 (79.0%)
25 (80.7%)
n.s. (p = 0.390)
Spontaneous vaginal delivery (%)
2 (22.2%)
1 (5.3%)
5 (16.1%)
Operative vaginal delivery (%)
2 (22.2%)
3 (15.8%)
1 (3.2%)
Adverse maternal outcome
78% of patients had an adverse maternal outcome as defined by the criteria given above.
Five women developed eclampsia (8.5%; two patients from Group 1 and three patients
from Group 3). 81.4% of patients received intravenous magnesium as prophylaxis against
eclampsia. If magnesium was indicated, it was administered for at least 24 hours postpartum.
Severe postpartum anemia (hemoglobin < 8 g/dl) was found in 28.8% of women. [Table 3 ] shows the number of patients in each hypertension group who received intravenous
magnesium as well as the number of patients with severe anemia. Online Supplement
Fig. S1 shows how many poor outcome parameters were reached in each hypertension group. Six
patients only met one criterion for adverse maternal outcome when they presented with
severe postpartum anemia.
Table 3
Outcome parameters in the three blood pressure groups.
Mild to moderate/no hypertension (Group 1), n = 9
Severe hypertension (Group 2), n = 19
Hypertensive crisis (Group 3), n = 31
p value
Abbreviation: n.s. = not significant
Intravenous magnesium (%)
6 (66.7%)
15 (79.0%)
27 (87.1%)
n.s. (p = 0.351)
Number of patients with eclampsia (%)
2 (22.2%)
0
3 (9.7%)
n.s. (p = 0.094)
Number of patients with severe anemia (%)
3 (33.3%)
6 (31.6 %)
8 (25.8%)
n.s. (p = 0.791)
Diagnosis of HELLP syndrome postpartum (%)
2 (22.2%)
5 (26.3%)
8 (25.8%)
n.s. (p = 0.308)
Number of outcome criteria of the combined endpoint found in each group
1
4 (44.4%)
6 (31.6%)
14 (45.2%)
n.s. (p = 0.125)
2
0
6 (31.6%)
11 (35.5%)
3
0
1 (5.3%)
2 (6.3%)
4
1 (11.1%)
0
1 (3.2%)
Women with hypertensive crises were more likely to have an adverse outcome than women
in the other two blood pressure groups. The difference between Group 1 and Group 3
was statistically significant (p = 0.049), see [Fig. 3 ] for details.
Fig. 3
Percentage of patients who met the criteria for the combined endpoint “adverse maternal
outcome” in each hypertension group. The difference between Group 1 and Group 3 is
significant (p = 0.049).
Blood pressure and antihypertensive medication at discharge from hospital
At discharge, 24 women (40.7%) were hypertensive with blood pressure levels > 140/90 mmHg
and 29 women (49.2%) needed antihypertensive medication. When the blood pressure groups
were analyzed, patients with higher blood pressure levels during pregnancy were more
likely to have persistent hypertension at discharge than patients with lower blood
pressure levels during pregnancy (see [Fig. 4 ]
a ). Statistical analysis showed that Group 3 patients were more likely to need a combination
of antihypertensive agents at discharge from hospital than women in Group 1 or 2.
The difference between blood pressure groups with regards to the number of prescribed
antihypertensive agents at the time of discharge was significant (p = 0.013), as shown
in [Fig. 4 ]
b .
Fig. 4
Blood pressure levels (a ) and number of antihypertensive agents prescribed (b ) at hospital discharge for each blood pressure group. The difference in the number
of prescribed antihypertensive agents between the three groups was significant (p = 0.013).
Abbreviation: BP = blood pressure.
Risk stratification
Statistical evaluation showed that the risk of an adverse maternal outcome increased
with the severity of hypertension. Women in Group 3 were more likely to meet the criteria
for the defined endpoint “adverse maternal outcome”. Univariable logistic regression
for the endpoint “adverse maternal outcome” showed that an increase in systolic blood
pressure of 1 mmHg increased the chances of meeting the criteria for the endpoint
by 5.7%. An increase of 10 mmHg raised the risk of an adverse outcome by 74%. The
forest plot in [Fig. 5 ]
a shows an odds ratio of 1.74 (95% CI: 1.22–2.66) as well as the odds ratios for each
outcome parameter in relation to a stepwise increase in systolic blood pressure by
10 mmHg. An increase in diastolic blood pressure of 10 mmHg raised the risk of an
adverse maternal outcome by 55% (odds ratio = 1.55; 95% CI: 0.99–1.58). The odds ratios
for all outcome parameters and
for the combined endpoint in relation to diastolic blood pressure are displayed in
[Fig. 5 ]
b .
Fig. 5
Forest plot showing the increased probability of meeting the criteria for the combined
endpoint “adverse maternal outcome” in relation to an increase in blood pressure by
10 mmHg for a systolic and b diastolic blood pressure. The individual parameters of the combined endpoint are
shown in both diagrams. An increase in systolic blood pressure of 10 mmHg raised the
risk of an adverse outcome by 74% (OR = 1.74; 95% CI: 1.22–2.66), and an increase
of diastolic blood pressure by 10 mmHg increased the risk by 55% (OR = 1.55; 95% CI:
0.99–1.58).
Discussion
This study shows that hypertension does not only play an important role in preeclampsia
but also in HELLP syndrome and affects patient outcomes. We found that when hypertension
increased, women were more likely to have an adverse outcome, were more likely to
be hypertensive at discharge and to need antihypertensive medication. According to
our analysis, even a small increase in systolic blood pressure significantly increases
the chance of an adverse maternal outcome. An increase in diastolic blood pressure
did not show this correlation. We suggest that maternal blood pressure could serve
as a predictive parameter for adverse outcomes in patients with HELLP syndrome and
that this could influence the therapeutic strategy.
To date, the role of elevated blood pressure in women with HELLP syndrome has not
been fully understood. HELLP syndrome is considered a subtype or complication of preeclampsia,
even though patients with HELLP syndrome may be normotensive. In our study, almost
all patients were hypertensive, and the extent of hypertension seemed to play an important
part in maternal outcome. Whether systolic or diastolic blood pressure or a combination
of both has the greatest impact cannot be clearly answered based on the data from
our study.
Tita et al. showed that patients with chronic hypertension benefit from antihypertensive
therapy during pregnancy, even if hypertension is mild [16 ]. In the past, most of these patients did not receive antihypertensive medication
during pregnancy. It is not clear whether an improvement in maternal outcomes of patients
with gestational hypertension and preeclampsia could be achieved by adjusting the
indication for antihypertensive treatment. It is also not clear whether close monitoring
and strict control of blood pressure could improve maternal outcomes of patients with
HELLP syndrome.
In a previous study, we showed that patients with preeclampsia and very high blood
pressure during pregnancy developed HELLP syndrome significantly more often than patients
with lower blood pressure levels [12 ]. Therefore, strict blood pressure control might be a possible preventative option
to delay or even avoid the severe complications of preeclampsia and related disorders
such as HELLP syndrome. Patients’ blood pressure levels should be recorded frequently
and need to be considered when scheduling delivery.
So far, little attention has been paid to blood pressure as a possible predictive
parameter for maternal outcome in HELLP syndrome. Attention has focused on laboratory
parameters and clinical symptoms. Aziz et al. noted that laboratory results and clinical
symptoms were unable to predict maternal outcome but suggested placental factors and
underlying medical conditions could play a role in adverse maternal outcomes [17 ]. Due to the rather small study population of 74 women and the subjectivity of clinical
symptoms, the power of their study is limited. In contrast, Miranda et al. were able
to show a correlation between the severity of symptoms and the rate of maternal complications.
Postpartum onset of symptoms seemed to be particularly associated with a more adverse
outcome [18 ]. Lind Malte et al. evaluated a combination of sFlt-1,
CT-pro-ET-1, and blood pressure to predict the course of severe preeclampsia and HELLP
syndrome. The authors suggested that a combination of these parameters could be used
to identify patients at risk for progression [19 ]. Placental factors such as sFlt-1, PIGF, and sFlt-1/PIGF ratio are already used
and being discussed as predictors for maternal outcome in patients with preeclampsia
[20 ].
The high number of patients in our cohort with eclampsia (n = 5) emphasizes the need
for improved risk stratification. According to our hypothesis, stricter blood pressure
control could lead to more favorable outcomes for patients with HELLP syndrome, especially
if blood pressures are very high. An alternative explanation for the high number of
adverse outcomes could be the narrow definition of HELLP syndrome we chose for our
study, which may have led to a selection bias with a greater inclusion of more severely
affected women. Comparable studies include patients who only meet one or two of the
diagnostic criteria for HELLP syndrome, also known as partial HELLP syndrome (e.g.,
the study by Huang et al. [21 ]).
In addition to the importance of improving the perinatal outcomes of mother and child,
long-term follow-up is necessary, as women with a history of preeclampsia face higher
cardiovascular risks in later life [12 ]
[13 ]
[22 ]. Results from previous studies have shown that patients who previously suffered
from HELLP syndrome had a higher risk of obstetric complications and a higher risk
of recurrence of hypertensive disorders during pregnancy of up to 20.7% [23 ]
[24 ]
[25 ]. According to Sibai et al., the risk of hypertension is related to pre-existing
hypertension rather than to HELLP syndrome itself. There is also a higher prevalence
of cardiovascular impairment such as cardiovascular remodeling including hypertrophy
of the left ventricle [26 ]. Sciatti et al. showed that HELLP syndrome was associated with myocardial dysfunction
and impairment of end-systolic elastance due to a larger aorta following HELLP syndrome
[27 ]
[28 ]. They proposed initiating a form of cardiovascular risk management for women with
a history of HELLP syndrome. In our study, nearly half of the cohort needed antihypertensive
medication at discharge, which demonstrates the importance of cardiovascular follow-up
to prevent long-term cardiovascular morbidity in patients with preeclampsia and HELLP
syndrome [29 ]. According to Riemer et al., postpartum lifestyle interventions including aerobic
endurance exercise can lower cardiovascular morbidity by having a positive effect
on arterial stiffness [30 ].
One limitation of this study is the small population size. This could be a general
problem due to the low prevalence of HELLP syndrome, and the patient populations in
comparable studies by Huang et al. and Li et al. were no bigger [21 ]
[31 ]. In addition, maternal outcome was based on a retrospective analysis and a constructed
endpoint with chosen parameters; this cannot portray the full complexity of adverse
maternal outcomes. Long-term follow-up is needed to understand more about obstetric
complications and the risk of recurrence of HELLP syndrome in subsequent pregnancies
as well as the increased risk of cardiovascular disease in general. The indication
for antihypertensive treatment in pregnancy is currently changing and the trend is
to treat women with mild hypertension as well. The study of Tita et al. showed no
increased risk of
small-for-gestational-age fetuses when women with mild chronic hypertension were treated
with antihypertensive agents [16 ]. Since our study investigated patients with HELLP syndrome over a longer period
(10 years), there was a trend to control blood pressure more strictly in the last
years of the study. But German guidelines on the indications for treatment with antihypertensive
agents did not change during this period and neither did the in-house guidelines,
therefore we assume that the impact was low [8 ]
[13 ].
One of the strengths of this study is the clear definition of HELLP syndrome and the
focus only on patients with HELLP syndrome. Other studies often did not differentiate
between severe preeclampsia and HELLP syndrome. Huang et al. compared patients with
preeclampsia complicated by HELLP syndrome with patients with only preeclampsia. He
found significant differences in gestational age at onset and in laboratory results,
indicating that are independent risk factors for HELLP syndrome [21 ]. This suggests that HELLP syndrome and preeclampsia cohorts should be evaluated
separately. Another problem is that there is no universal consensus regarding the
cut-off values of laboratory results or the classification of HELLP syndrome, which
makes it difficult to compare different studies on HELLP syndrome. Compared to other
studies such as the study by Huang et al. [21 ], the inclusion criteria in our study were strict (no patients with partial HELLP
syndrome). This may have led to a selection bias towards more severely affected women
but it will also have increased the specificity of the diagnosis and lowered the risk
of including patients with a false diagnosis. So far, little attention has been paid
to the role of hypertension in HELLP syndrome. The correlation between preeclampsia
and HELLP syndrome with regards to pathogenesis suggests that hypertension might play
an important role in HELLP syndrome and its risk stratification.
In conclusion, more HELLP syndrome-specific research is necessary to fully understand
its pathophysiology, discover possible new parameters for the prediction of maternal
outcomes and develop better therapeutic approaches. Our hypothesis that the severity
of hypertension could function as a predictive marker for adverse maternal outcome
needs to be tested in different and larger study populations. A prospective study
is needed to investigate whether good blood pressure management can reduce the rate
of complications in women with HELLP syndrome.
Supplementary Material
Fig. S1 Number of outcome parameters of the combined endpoint for adverse maternal outcome
identified in the three hypertension groups.
