Objective
The pathogenesis of the HELLP (hemolysis, enzyme liver, low platelets) syndrome is unknown. Recently soluble endoglin (sEng) was identified as a cause of the appearance of schistocytes and liver pathology in an animal model of preeclampsia (PE).
Study Design
We explored the value of sEng in 82 women who delivered in a context of normal pregnancy (NP, n = 10), PE (n = 49), or HELLP (n = 23).
Results
sEng was elevated in pathological pregnancies (66.7 ± 62 and 75.7 ± 48 pg/mL in PE and HELLP, respectively, vs 5.29 ± 1.25 in NP, P < .001 for both comparisons) and was correlated with an increase in transaminases (r 2 = 0.17; P = .05), but it was not statistically different between PE and HELLP.
Conclusion
Although recent literature findings demonstrated a role of sEng in the pathophysiology of HELLP syndrome in animal models, we found that, at the time of delivery, sEng was not specifically elevated in preeclamptic patients with HELLP.
Pathophysiology of the maternal syndrome of preeclampsia is not entirely understood. However, Maynard et al in 2003 published a seminal paper demonstrating the role of the soluble form of the vascular endothelial growth factor type 1 receptor (sFlt-1) in pregnancy-induced hypertension and albuminuria. As a result, the following schema has become widely accepted: because of poor pseudovasculogenesis, a placenta will produce excessive quantities of numerous soluble mediators.
The concentrations of these mediators progressively increase in maternal serum from the end of the first trimester through the end of pregnancy, altering endothelial function, thus leading to systemic hypertension, glomerular endotheliosis, and liver and brain disorders. Although sFlt-1 is important, other antiangiogenic factors are probably at work here.
Recently the soluble form of endoglin (sEng), a receptor for transforming growth factor beta on endothelial cells, has been incriminated on the basis of experimental findings in the pathophysiology of a severe variant of preeclampsia, the hemolysis, enzyme liver, low platelets (HELLP) syndrome. Briefly, it was shown in this paper that: (1) coinfection of gravid rats with 2 recombinant viruses encoding sEng and sFlt-1 could reproduce elements of the human HELLP syndrome in animals; and (2) sEng levels were significantly higher in blood from women delivering a baby in a context of the HELLP syndrome than in preeclampsia, even in its severe forms.
Later it was also reported in a large population that sEng was indeed increased in maternal blood from women with preeclampsia, relative to women with normal outcome, from the 20th week of gestation; however, this paper did not address the specific importance of sEng in respect of HELLP syndrome. Regardless, the relative concentration of several factors could, in conjunction with individual susceptibilities (ie, maternal endothelial thresholds), arouse different clinical syndromes.
As far as HELLP syndrome is concerned, an excessive concentration of sEng and sFlt-1 is thought to play a causal role. Because this has not yet been confirmed, neither prospectively nor retrospectively, we explored the value of sEng in women with pathological pregnancies and tried to establish correlations with clinical and biological data. Our hypothesis was that sEng would be significantly higher in patients with HELLP syndrome vs patients with the classical form of preeclampsia.
Materials and Methods
Population study
The aim of this retrospective study was to determine whether the maternal serum concentration of sEng would allow us to distinguish between patients with preeclampsia with and without HELLP syndrome. Every patient having delivered during the study period (from May 2003 to December 2007) in the Department of Obstetrics of the Hopital de Poissy in France with a final diagnosis of preeclampsia with or without HELLP syndrome (see definitions in the following text), were enrolled, provided they had EDTA-plasma samples available for analysis and drawn during the day of the diagnosis of preeclampsia or HELLP syndrome.
We included an additional group of 10 pregnant patients, with no previous history of hypertension or diabetes mellitus, matched (±10 days) for the age of gestation and having delivered with a final diagnosis of normal pregnancy. All patients had given their informed, open-ended consent for the storage of frozen plasma and the future analysis of biomarkers, and the study protocol was approved by our local institutional review board at Poissy.
The following routine clinical and biological data were recorded: age, body mass index, systolic and diastolic blood pressure, platelet count, asparate and alanine aminotransferase (peak values), lactate dehydrogenase (peak value), proteinuria (peak value), total length of gestation, and baby weight at delivery.
Definition criteria
Preeclampsia was defined as hypertension (ie, systolic blood pressure of at least 140 mm Hg, or diastolic blood pressure of at least 90 mm Hg, on 2 occasions) and proteinuria (>30 mg/dL on 2 consecutive dipsticks and confirmed biochemically) after 20 weeks of gestation. In addition, the diagnosis of preeclampsia was further confirmed by the reversal of the symptoms by 12 weeks postpartum.
HELLP syndrome was defined according to the Mississippi classification for HELLP: all patients had lactate dehydrogenase (LDH) greater than 600 IU/L; a low platelet count (class I: ≤50 g/L ; class II: ≤100 g/L; class III: ≤150 g/L); and a high aspartate aminotransferase (ASAT) or alanine aminotransferase (ALAT) level (class I and II: ≥70 IU/L ; class III: ≥40 IU/L). A secondary analysis was also performed using the Tennessee classification for the HELLP syndrome: ASAT 70 IU/L or greater, LDH 600 IU/L or greater, and platelet count 100 g/L or less.
Measurement of antiangiogenic factors
The total or free soluble endoglin and Flt-1 were measured in thawed plasma by A.H. and J.F. in September 2008, as previously described in a duplicate and blinded fashion by enzyme-linked immunoassay (quantikine human sVEGF R1/Flt-1 and endoglin/CD105; R&D Systems, Abingdon, England) in a 1:10 dilution, according to the manufacturer’s recommendations. These commercial kits measure total (bound and unbound) protein concentrations. We specifically found the coefficient of variation for the measurement of sFlt1 and sEng to be 5.7% and 6.2%, respectively.
Statistical analysis
The human data reported in the publication by Venkatesha et al indicated a mean sEng concentration of 40 ng/mL and 100 ng/mL in mild preeclampsia and HELLP, respectively, with corresponding SDs of 10 ng/mL for both groups. Therefore, it was calculated that a sample size of at least 13 patients per group would allow for the detection of a 25% increase in sEng levels, with a statistical power of 0.8 and a risk error of 0.05.
Statistical analysis was performed using the Statview version 5 software. Clinical and biological variables were compared using the Student t test. Linear regression analysis was performed to detect a correlation between the variables and the concentration of antiangiogenic factors in maternal plasma. Data are presented as mean ± SD. P < .05 was considered statistically significant.
Results
During the study period, 152 women delivered in our center with a final diagnosis of preeclampsia with or without HELLP syndrome. Among them, clinical data and frozen plasmatic samples were available in 72 patients (47.4%). Therefore, the population study consisted of 82 patients who received a final diagnosis of normal pregnancy (n = 10), preeclampsia (n = 49), or HELLP syndrome (n = 23). Patients with HELLP were classified as Mississippi class I (n = 3), II (n = 9), or III (n = 11).
The main characteristics of this population are shown in the Table . Apart from the biological features used for classification into HELLP (hemolysis, elevation of liver enzymes, and thrombocytopenia), women with HELLP syndrome were characterized by a more severe hypertension, a lower body mass index, and a lower frequency of smoking when compared with classical preeclampsia. The gestation length and the weight of the newborn were also significantly lower in the HELLP group. Proteinuria was comparable.
