Peripartum management of a patient with Glanzmann’s thrombasthenia using Thrombelastograph®

S. Monte* and G. Lyons
Department of Obstetric Anaesthesia, St James’ University Hospital, Beckett Street, Leeds LS9 7TF, UK*Corresponding author
Accepted for publication: January 7, 2002

We describe the perioperative management of a 31-yr-old primipara with Glanzmann’s thrombasthenia, who required urgent Caesarean delivery at 33 weeks’ gestation. Peripartum haemorrhage was inevitable and was pre-empted by transfusion of multiple blood products, the effects of which were monitored by Thrombelastograph®. The blood products given are discussed.

Glanzmann’s thrombasthenia is a rare, autosomal recessive platelet disorder manifested by a lack of the glycoprotein IIb–IIIa complex in the platelet membrane.1 Platelet aggregation is severely impaired by the absence of this integral membrane complex, which holds the receptor for fibrinogen.2 Patients may have normal platelet counts but can exhibit very abnormal platelet function. The disorder is characterized by potentially major mucocutaneous bleeding, and is usually diagnosed in childhood following such an episode. Information regarding the long-term outcome of these patients is limited and, as a consequence, their management is not evidence based. It is considered hazardous for such patients to conceive, with a high risk of severe peripartum haemorrhage.

We describe the peripartum management of a 31-yr-old primipara with Glanzmann’s thrombasthenia. We make special reference to the use of the Thrombelastograph® analyser, which enabled us to assess the efficacy of various therapies.

A 31-yr-old primpara with a body mass index of 25, who was known to have Glanzmann’s thrombasthenia, was referred to the obstetric anaesthetic department in the third trimester of pregnancy.

As an infant she had presented with multiple nose bleeds and frequent bruising. Investigations led to a diagnosis of Glanzmann’s thrombasthenia. Her condition necessitated multiple whole-blood and platelet transfusions throughout childhood and adolescence. In early 2000 she had had an ectopic pregnancy, which was diagnosed only when it had ruptured. Following laparotomy, she required multiple transfusions of whole blood, platelets and factor VII, and had a prolonged stay in hospital.

She was referred to the antenatal clinic by her family doctor when 7 weeks pregnant. Her antenatal care was uneventful until 28 weeks’ gestation. At this stage, after ultrasound scanning revealed dilatation of the fetus’s stomach and evidence of poor blood flow in the fetus’s thoracic aorta, chorionic villus sampling was arranged to exclude an associated chromosomal abnormality. Platelet transfusion was advised for this procedure, but was complicated by the fact that the patient had anti-c antigens (albeit at low titre), and human leucocyte antigen (HLA), requiring HLA-matched platelets. Fetal chromosomes proved normal but, because of the scan abnormalities, the patient continued with twice-weekly Doppler assessments and cardiotocographs.

At 31 weeks’ gestation the patient developed unremitting epistaxis. Her nose was packed, she was prescribed tranexamic acid, and the bleeding subsided.
The anaesthetic team was made aware of the patient at this stage and, when assessed, her striking history of haemorrhage was noted. The presence of antiplatelet antibodies meant that platelet transfusions might not be effective. Native blood Thrombelastograph® analysis was performed at this stage, which showed normal clot initiation but poor clot strength (Fig. 1). With this information in mind, we decided that regional anaesthesia was contraindicated and that peripartum haemorrhage was inevitable.

At 33 weeks’ gestation, Doppler assessment revealed no end-diastolic flow in the fetus’s thoracic aorta and a markedly reduced liquor volume. It was therefore decided that Caesarean delivery was indicated with a degree of urgency, but as a planned procedure. A multidisciplinary team of obstetricians, haematologists and anaesthetists was involved in the care of the patient, and a regimen for perioperative platelet cover was planned. Six units of whole blood were cross-matched. The patient was given a gamma globulin infusion (Flebogamma) on the night before surgery, and lansoprazole 30 mg orally on the morning of surgery. Ten units of platelets were given 30 min before surgery, together with recombinant factor VII (rFVII) 90 µg kg–1. Intravenous ranitidine 50 mg was given preoperatively, together with oral 0.3 M sodium citrate 30 ml. Good venous access was achieved by the placement of two 16G cannulae in forearm veins. A rapid sequence induction of anaesthesia was performed after preoxygenation, using thiopental 6 mg kg–1 and succinylcholine 1.5 mg kg–1 i.v., followed by oral intubation of the trachea. Atracurium 0.5 mg kg–1, morphine 0.2 mg kg–1, and cefuroxime 750 mg as routine antibiotic prophylaxis were given intraoperatively. A bolus followed by an infusion of oxytocin (Syntocinon®) was administered after delivery of the baby. Anaesthesia was maintained intraoperatively with oxygen, nitrous oxide and isoflurane end-tidal concentration 1.1%, with an FIO2 of 0.4, via a circle system. A right-sided triple-lumen central line was sited via the external jugular vein after induction of anaesthesia.

The Caesarean section progressed uneventfully, with an estimated blood loss of 700–800 ml. There was cardiovascular stability throughout. Meticulous haemostasis was achieved before wound closure. In total, 12 units of platelets were given intraoperatively, in addition to the 10 units given preoperatively. Residual neuromuscular blockade was antagonized with neostigmine 50 µg kg–1 and glycopyrronium 10 µg kg–1 and the patient’s trachea was extubated uneventfully.

Serial Thrombelastograph® analysis was performed before and after administration of each blood product in our obstetric unit laboratory, which allowed us to make a qualitative perioperative assessment of the effect of platelets and rFVII. The patient was transferred to the intensive care unit for postoperative care because haemorrhage was anticipated. Advice from haematology was that the patient should receive rFVII 2 hourly for the first 12 h, then 3 hourly for 12 h, then 4 hourly on day 1, in a dose of 90 µg kg–1. She also received 10 units of platelets 12 hourly for the first 24 h.

In the first 24 h after surgery, the patient remained cardiovascularly stable, but had a continuous ooze from her abdominal drain site, and in total lost 1900 ml of blood. Her incision site remained dry and ultrasound scanning revealed no evidence of intra-abdominal free fluid. Blood loss subsided over the ensuing 36 h. In total, the patient received 12 units of whole blood, 120 units of platelets, 38 units of cryoprecipitate and rFVII 101 mg in the first 36 h after surgery. She was discharged from the intensive care unit on the second postoperative day and had no further problems. She is now at home with a healthy baby girl.

The incidence of morbidity and mortality associated with Glanzmann’s thrombasthenia is largely unknown. This reflects the low prevalence of the condition and possible under-reporting. There is anecdotal evidence of a fatal cerebral haemorrhage at the time of delivery,2 but this case has not been published separately. There are no other documented deaths from obstetric- or surgically induced haemorrhage. Various peripartum treatments are described to limit associated obstetric haemorrhage; however, there is no consensus as to best management.

Repeated platelet transfusions predispose to the development of antiplatelet antibodies,4 resulting in a variable response to subsequent platelet transfusions. In this case, a good response to platelet transfusion was demonstrated but in some cases, platelet transfusion may not help at all. A gamma globulin infusion was given to our patient on the night before surgery in an attempt to dampen this antiplatelet response. The rationale for prescribing rFVII was based on the theory that factor VII acts on platelets in the absence of tissue factor, to activate factors IX and X, thus enhancing thrombin generation. The increased generation of thrombin may then provide a strong signal for the recruitment of other platelets.5 Treatment with rFVII is well tolerated generally, although hugely expensive. The cost incurred for rFVII alone in our patient was £96 000.

Other treatments described include antibody removal by plasmapheresis, and transfusion of single-donor and HLA-matched platelets. Plasmapheresis has previously been used in a patient with multiple antibodies. This technique significantly decreases the antibody titre and enables the effective transfusion of compatible platelets.2 That patient did not receive rFVII. Clearly, good surgical haemostasis and oxytocin infusions also form an integral part of care.

Glanzmann’s thrombasthenia is an uncommon condition, and as there is no accepted or universally available monitor of platelet function, its clinical course is difficult to monitor. Techniques include measurement of bleeding time and platelet aggregometry. Critics of the bleeding time have questioned its validity and reproducibility.6 Platelet aggregometry is not widely available.

Thrombelastograph® analysis demonstrates various aspects of coagulation. One variable, the maximum amplitude, shows consistent reproducibility as a test of platelet function.7 8 The whole test has a response time of 20–30 min, making it suitable for serial sampling. We used the Thrombelastograph® to monitor this patient’s coagulation. The Thrombelastograph® picture of Glanzmann’s thrombasthenia has not been reported previously. It was clear from the initial Thrombelastograph® tracing that the clot strength was poor when compared with our normal pregnant reference intervals at similar gestation. It was apparent using this technique and by performing serial Thrombelastograph® tests, that there was a transient improvement in clot strength after platelet transfusion, which appeared to last for about 2 h. However, the effect on the Thrombelastograph® data after infusion of rFVII was less marked. This is not surprising, as the purpose of factor VII is to act locally in response to tissue trauma.

This patient’s peripartum care was complicated, involving obstetricians, anaesthetists, haematologists, ENT surgeons and a clinical immunologist. Transfusion of all the blood products involved considerable cooperation by the blood bank staff, and incurred huge expense. Such a complex problem is probably best managed in a tertiary referral centre.
Faced with this problem again, we would adopt a similar approach. Use of the Thrombelastograph® provided reassurance of the efficacy of platelet transfusions in the presence of antibodies. Because of the multitherapeutic approach taken, it was not possible to specifically identify a key role for any individual treatment. The good outcome may have been the result of the combination of therapies. The morbidity of this condition in association with pregnancy and delivery is unclear, but the involvement of the haematology services is invaluable.

1 Sherer DM, Lerner R. Glanzmann’s thrombasthenia in pregnancy: a case and review of the literature. Am J Perinatol 1999; 16: 297–301[ISI][Medline]
2 Ito K, Yoshida H, Matsumoto H. Antibody removal therapy used successfully at delivery of a pregnant patient with Glanzmann’s thrombasthenia and multiple antiplatelet antibodies. Vox Sang 1991; 61: 40–6[ISI][Medline]
3 Kashyap R, Kriplani A. Pregnancy in a patient with Glanzmann’s thrombasthenia with antiplatelet antibodies. J Obstet Gynaecol Res 1997; 23: 247–50[Medline]
4 Laursen B, Jacobsen S. Transfusion problems in patients with Glanzmann’s thrombasthenia. Eur J Haematol 1992; 49: 143–4[ISI][Medline]
5 Monroe D, Hoffman M. The factor VII-platelet interplay: effectiveness of recombinant factor VIIa in the treatment of bleeding in severe thrombocytopathia. Semin Thromb Haemost 2000; 26: 373–7[ISI]
6 Orlikowski CEP, Rocke DA. Coagulation monitoring in the obstetric patient. Int Anaesth Clin 1994; 32: 173–91
7 Tuman KJ, McCarthy RV. A comparison of thromboelastography and platelet aggregometry. Anaesthesiology 1991; 75: A433
8 Robinson L, Gorton H. Thromboelastography: validation of the Thrombelastograph3000® and Thrombelastograph5000®. Br J Anaesth 2001; 86: 309