Article Text

PDF

The HELLP syndrome in the antiphospholipid syndrome: retrospective study of 16 cases in 15 women
  1. D Le Thi Thuong1,
  2. N Tieulié1,
  3. N Costedoat1,
  4. M-R Andreu1,
  5. B Wechsler1,
  6. D Vauthier-Brouzes2,
  7. O Aumaître3,
  8. J-C Piette1
  1. 1Departments of Internal Medicine, Groupe Hospitalier Pitié-Salpêtrière, 47-83 Boulevard de l’Hôpital, 75651 Paris cedex 13, France
  2. 2Departments of Obstetrics, Groupe Hospitalier Pitié-Salpêtrière, 47-83 boulevard de l’Hôpital, 75651 Paris cedex 13, France
  3. 3Department of Internal Medicine, Hôpital Gabriel Montpied, BP 69, 63003 Clermond-Ferrand cedex 1, France
  1. Correspondence to:
    Dr D Le Thi Thuong
    Departments of Internal Medicine, Groupe Hospitalier Pitié-Salpêtrière, 47-83 Boulevard de l’Hôpital, 75651 Paris cedex 13, France; du.boutinpsl.ap-hop-paris.fr

Abstract

Objective: To study the characteristics of the haemolysis, elevated liver enzymes, low platelets (HELLP) syndrome in the antiphospholipid syndrome (APS) and its influence on the subsequent pregnancies.

Methods: This was a retrospective analysis of 16 episodes of HELLP complicating APS in 15 women.

Results: HELLP was complete in 10 cases and partial in six. It occurred during the second trimester in seven cases (the earliest at 18 weeks’ gestation), the third trimester in seven cases, and the day following delivery in two cases. Pre-eclampsia was present in six cases and eclampsia in five. Outcome of pregnancies was: live birth (n = 8), stillbirth (n = 2) and fetal death (n = 6). APS was primary in nine women and secondary to systemic lupus erythematosus (SLE) in six. HELLP revealed primary APS in six cases. Seven women were not treated. Low dose aspirin was empirically prescribed in one woman whose APS had been undiagnosed despite a history of two fetal deaths. In the other women, therapy consisted of aspirin (n = 8), low molecular weight heparin with a dose varying between 3000 and 12 000 U daily (n = 5), and high dose immunoglobulin every 4 weeks (n = 2), hydroxychloroquine (n = 4), and prednisone (n = 6). Six women had seven subsequent pregnancies, 3–6 years after the complicated pregnancy. HELLP recurred at 33 weeks’ gestation in one woman with SLE treated with prednisone, hydroxychloroquine, aspirin, and enoxaparin, and pregnancy ended in live birth. One woman became pregnant after in vitro fertilisation and embryo transfer, but pregnancy ended in fetal death despite prednisone, hydroxychloroquine, and enoxaparin. Four women had five uneventful pregnancies with 100 mg daily aspirin and heparin.

Conclusions: APS may be revealed by HELLP. In APS, HELLP is associated with pre-eclampsia/eclampsia in most cases and seems to occur earlier than in the general population. Heparin plus aspirin may prevent obstetric complications in the subsequent pregnancies.

Statistics from Altmetric.com

Incidence of haemolysis, elevated liver enzymes, low platelets (HELLP) syndrome in pregnancy has been estimated at 0.01–0.2% in the general population, and at 10–12% in pregnancies complicated with pre-eclampsia/eclampsia. During a 15 year period in a tertiary care centre serving a population area with 40 000 deliveries per year, 442 HELLP syndromes were observed.1 During the same period, 2331 pregnancies were complicated by pre-eclampsia or eclampsia. Reports of HELLP syndrome complicating APS are scarce. We report a series of 16 episodes of HELLP syndrome complicating primary or secondary APS in 15 women, and describe its characteristics and its influence on the subsequent pregnancies.

PATIENTS AND DEFINITIONS

We retrospectively reviewed the data of 16 pregnancies in 15 women with APS complicated with HELLP syndrome. Diagnosis of APS was based on the international criteria.2 Lupus anticoagulant (LA) was usually detected by activated partial thromboplastin time (PTT), diluted thromboplastin time, or kaolin clotting time. Abnormal coagulation times were confirmed by mixing patient and control plasma on a 1:1 ratio in order to exclude clotting factor deficiencies. IgG and IgM anticardiolipin (aCL) antibodies were measured by ELISA. IgG and IgM rates were reported as negative if <15 U/ml, as low positive at 16–25 U/ml, as medium positive at 26–80 U/ml and as high positive if >80 U/ml.

Diagnosis of HELLP was based on the presence of haemolysis (anaemia with characteristic peripheral blood smear), lactate dehydrogenase (LDH) over the upper normal value or total bilirubin >12 mg/l, elevated alanine aminotransferase (AAT) >twofold the upper normal value, and nadir platelet count below 125 000/mm3. Complete HELLP was defined by presence of all criteria, and partial HELLP by the presence of two criteria.3 Diagnosis of systemic lupus erythematosus (SLE) was based on the revised ACR criteria.4 The other definitions were: (a) embryonic loss (spontaneous termination of pregnancy prior to 10 weeks’ gestation)); (b) fetal death (according to Branch’s definition,5 death of a fetus demonstrated to be alive at or beyond 10 weeks’ gestation); (c) premature birth (termination of pregnancy with a live birth below 37 weeks’ gestation); (d) full term birth (termination of pregnancy with a live birth between 38 and 40 weeks); (e) intrauterine growth retardation (IUGR) (birth weight below the 10th centile for the stated gestation); (f) hypertension (diastolic pressure >90 mm Hg); (g) pre-eclampsia (hypertension complicated with proteinuria ⩾0.5 g/24 hours or oedema or both); and (h) eclampsia (pre-eclampsia complicated with seizures or fitting.

RESULTS

At the first episode of HELLP, the mean age was 30 years (range 22–37). Seven women were nulliparous. Pregnancy was the second in four women, the third in two women, the fourth for one woman, and the tenth for one other woman. Eight women had been previously pregnant with a total of 15 pregnancies. Two women had had a total of four normal pregnancies before SLE/APS onset. Hence, 11 pregnancies were complicated with fetal death (n = 9), embryonic loss (n = 1), and live birth at full term of a growth retarded baby (n = 1).

APS was primary in nine women and secondary to SLE in six. Primary APS was known in only one of the women, who had a history of repeated fetal loss and stroke; 2/8 women with unknown primary APS had been undiagnosed, although there was a suggestive history. One woman had a history of two successive fetal deaths, and another woman had a false positive syphilis test. Hence, HELLP revealed primary APS in six cases.

Of the six women with SLE, APS was known in four, who had a history of thrombophlebitis with fetal death (n = 2), repeated fetal deaths (n = 1), and ischaemic cerebral stroke with embryonic loss (n = 1). Two women with SLE had prior asymptomatic aCL until onset of HELLP complicated with fetal death (n = 2).

Seven women whose pregnancy was complicated by HELLP had not been previously treated. Low dose aspirin was empirically prescribed in one woman whose APS had been undiagnosed despite a history of two fetal deaths. Three women were treated with aspirin plus low molecular weight heparin (LMWH) with a dose varying between 3000 and 8000 U daily. Two women with prior complicated pregnancy under aspirin and LMWH were treated with LMWH 12 000 U daily, aspirin 100 mg/day, and high dose immunoglobulin 80 g twice daily every 4 weeks. One woman with primary ASP complicated with autoimmune thrombocytopenia was treated with hydroxychloroquine and 5 mg daily prednisone. All women with SLE, except one woman who was not treated, received prednisone with a dose ranging between 5 and 30 mg/day and aspirin 100 mg/day. Hydroxychloroquine was maintained in three women and withdrawn in two others.

HELLP was complete in 10 cases and partial in six. It occurred during the second trimester in seven cases (the earliest at 18 weeks’ gestation), the third trimester in seven, and in two cases, it appeared the day following delivery (fig 1). The main characteristics of HELLP are summarised in table 1.

Table 1

 Main characteristics of HELLP complicating the APS and comparison with series of HELLP in referral obstetric centres

Figure 1

 Date of HELLP onset in APS.

Outcome of pregnancy was: live birth of a eutrophic (n = 3) or a hypotrophic baby (n = 5), at full (n = 2) or premature term (n = 6), stillbirth (n = 3), and fetal death (n = 5). Table 2 summarises the pregnancy outcome according to the date of HELLP onset. Fetal death and stillbirth were observed only in HELLP occurring during the second trimester. Apart from two women (one who developed HELLP at 36 weeks’ gestation and another in the post-partum period), all deliveries were made by caesarean section. The main paediatric complications were observed in a 25 week old premature newborn, who developed a severe membrane hyaline disease evolving towards bronchopulmonary dysplasia. The child needed mechanical ventilation for 43 days, and this period was complicated by multiple infections, metabolic anomalies, enterocolitis, neutropenia, and thrombocytopenia. Craniosynostosis needed surgical repair 1 year later. At the present time, this child continues to have retarded stature.

Table 2

 Pregnancy outcome in our series according to the date of HELLP onset

Pre-eclampsia was present in six cases and eclampsia in five. Among these 11 cases, outcome of pregnancy was: fetal deaths (n = 4), stillbirth (n = 3), and delivery of hypotrophic (n = 3) or eutrophic (n = 1) liveborns. One woman had an abrupted placenta. No hepatic infarction was observed.

Maternal therapy included anti-hypertensive therapy, transfusion, and other supportive care, including mechanical ventilation for 4−14 days in three women. Dexamethasone or betamethasone were prescribed in order to accelerate fetal maturation. No patient needed plasma exchanges, and none of the women died. One woman remained hypertensive and needed permanent medication.

Six women had seven subsequent pregnancies, 3–6 years after the pregnancy complicated by HELLP (table 3). HELLP recurred at 33 weeks’ gestation in one woman (patient 1) with SLE treated with prednisone, hydroxychloroquine, aspirin, and enoxaparin. Caesarean section delivered a 2100 g live born child, and a rapid remission of laboratory abnormalities followed. One woman became pregnant after in vitro fertilisation and embryo transfer but pregnancy ended in fetal death despite prednisone, hydroxychloroquine, and enoxaparin (patient 2). Four women had five uneventful pregnancies with 100 mg daily aspirin and heparin (patients 3–6).

Table 3

 Characteristics of six women with APS complicated with HELLP who had seven subsequent pregnancies

At the present time, all women are doing well, with a mean follow up of 7 years. No vascular complication was observed. All the women with APS characterised only by obstetric complications are being treated with 100 mg daily aspirin, except one woman who has had no therapy. All women with APS with both history of vascular occlusion and obstetric complication are treated with oral anticoagulation. All six women with SLE are in remission with prednisone (n = 4, with a dose ranging between 2 and 12.5 mg daily) and/or hydroxychloroquine (n = 6).

DISCUSSION

HELLP syndrome was described in 1982 by Westein as the association of microangiopathic haemolytic anaemia, hepatitis, and thrombocytopenia in a particular group of pre-eclamptic women, but there are no consensus criteria for its diagnosis. In Martin’s study,6 HELLP was diagnosed on the basis of thrombocytopenia ⩽ 150 000/μL), hepatic dysfunction (increased AAT or alanine aminotransferase ⩾40 U/l), and haemolysis (increased LDH ⩾600 U/l, progressive anaemia). Martin also classified HELLP on the basis of the platelet count nadir: class 1 when platelet nadir is ⩽50 000/mm3, class 2 at 50 000–100 000/mm3, and class 3 at 100 000–150 000/mm3. Incidence of eclampsia, epigastric pain, major maternal morbidity, and stillbirth increases as HELLP worsens from class 3 to class 1. In Sibai’s study,1 HELLP was diagnosed when pre-eclampsia was associated with all the following laboratory abnormalities: characteristic peripheral blood smear, LDH>600 U/l (or total bilirubin >12 mg/l), AAT>70 U/l, and platelet count <100 000/mm3. Audibert3 defined complete HELLP by the presence of all three of the following criteria: haemolysis (characteristic peripheral blood smear and LDH ⩾600 U/l), AAT⩾70 U/l, and platelet count <100 000/μL. Partial HELLP was defined by the presence of one or two features of HELLP. Patients with complete HELLP had a higher risk of caesarean section, disseminated intravascular coagulation, and need for transfusion than those with partial HELLP or pre-eclampsia.

The risk of HELLP is probably increased in APS. However, its incidence is difficult to estimate from the literature, especially after the recent changes in APS criteria.2 The number of HELLP complicating APS cases reported in the English literature is <30. A recent literature review found 10 cases,7 to which should be added their two cases, seven cases published in 2002–2003,19–22 six earlier cases,8,9,13,15,16 and a few cases cited in APS pregnancies series.23–25 Our series of 75 pregnancies in primary or secondary APS23 included seven women (also included in this study), who had eight episodes of HELLP. Backos24 reported two HELLP in a series of 150 APS pregnancies. In a placebo controlled study of intravenous immunoglobulin treatment for APS, Branch observed that 6/16 women developed severe pre-eclampsia and/or HELLP.25 However, most series of APS pregnancies did not report any case of HELLP,26–36 although the rate of pre-eclampsia varied between 0 and 51%.

The high rate of HELLP observed in our series (10.6%), contrasts with 0.66% in that of Backos;24 this can be explained by different recruitment criteria, as Backos included only primary APS without history of thrombophlebitis. On the other hand, the frequency of APS in HELLP is not well known. In our series, HELLP revealed APS in 8/15 cases. HELLP also revealed APS in half of the 24 well described cases in the literature (table 4). A systematic screening of thrombophilia in 93 women with a history of severe pre-eclampsia/HELLP found antiphospholipid antibodies in 66% of the cases.37 Although the cut off level was low (5 U/ml), the difference in prevalence compared with controls was significant (p<0.001). In France, a systematic search for antiphospholipid antibodies in 68 consecutive cases of complete HELLP among 15 000 deliveries, displayed nine positive cases. Seven of these were subsequently tested, and only two were diagnosed as primary APS. Hence, APS prevalence in HELPP is estimated at 3−6%.7

Table 4

 Literature review of 24 well described cases of HELLP syndrome complicating APS

In our series, HELLP occurred in 44% of the cases during the second trimester and 12.5% at 18–20 weeks. In a literature review of the 24 well described cases of HELLP complicating APS, HELLP occurred also before 27 weeks’ gestation in 16 cases (table 1). Early HELLP onset was described in association with APS as early as at 8 weeks’ gestation.20 In Sibai’s study,1 HELLP occurred in 2.5% of the cases at 17–20 weeks, in 71% of the cases at 27–36 weeks, in 18% at 37–42 weeks, and in 30% during the post-partum period.

HELLP appears more severe in APS than in the general population. In our series, more than two thirds of the cases were associated with pre-eclampsia/eclampsia, and two cases were complicated by microangiopathic coagulopathy. Six of the 24 cases of well described HELLP complicating APS were associated with pre-eclampsia (table 1). We did not observe any case of liver infarction, which Pauzner found to be almost always associated with APS.20 In one case of HELLP in APS, hepatic involvement was due to complete thrombosis of hepatic veins.13

HELLP may also be part of the catastrophic syndrome, similarly to Hochfeld’s case.9 Catastrophic APS is an acute and devastating disseminated coagulopathy defined by the clinical involvement of at least three different organ systems over a short period.38 One fatal case of HELLP with catastrophic syndrome was characterised by bone marrow necrosis.19

In our study, all HELLP resolved with delivery and/or supportive care. There was no case of maternal death, similarly to the series of Backos and Branch.24,25 In the general population, HELLP syndrome mortality is about 1%.1 In the literature, besides the catastrophic syndrome, some cases of HELLP complicating APS were refractory to corticosteroids and anticoagulation. Treatment of refractory HELLP is not standardised. Remission may be observed after plasma exchanges.22 Dexamethasone appeared to be more effective than betamethasone.39 Combination of plasma exchanges with fresh frozen plasma has been advocated.1 In our studies, six women had seven subsequent pregnancies treated with aspirin plus LMWH; one pregnancy was complicated by HELLP recurrence but resulted in live birth, another one with embryonic loss. The other pregnancies ended in live birth. There are no data concerning the risk of HELLP recurrence in APS. Of four cases of HELLP complicating APS, Pauzner reported one recurrence of liver infarction despite aspirin and 60 mg daily enoxaparin.20 In the general population, HELLP recurred in 5% of subsequent pregnancies in women with a history of HELLP at ⩽30 weeks’ gestation.40

In conclusion, APS may be revealed by the HELLP syndrome. Search for antiphospholipid antibodies is mandatory in such circumstances. In APS, HELLP is associated with pre-eclampsia/eclampsia in most cases, and seems to occur earlier than in the general population. Heparin plus aspirin may prevent obstetric complications in the subsequent pregnancies.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

REFERENCES