Platelet Limbo

Two cases of pregnancy-thrombocytaenia last week brought out the old discussion about the use of regional anaesthesia when the platelet count is low. The first case was HELLP syndrome: unwell, RUQ pain, and platelet count of 25,000. She had no petechiae on her mucosa, or under the sphygomanometer cuff, and the venepuncture sites had healed well. Her airway looked alright for intubation. After dexamethasone and several hours of magnesium infusion overnight Caesarean delivery of the 34 week baby was successfully undertaken with general anesthesia (with isoflurane of course!) without a platelet transfusion. Clots formed in the operative field and the estimated loss was less than average. She was extubated and tranferred to HDU and did well. The strategy here was to treat the patient rather than a number: there was no clinical sign of thrombocytopenia and there was no bleeding- thus no platelets were given and the ones that were going around seemed to do the job enough.

The other case was one of moderately severe preeclampsia for cesarean delivery; her platelet count was 79,000. Again there was no overt sign of low platelets and spinal anesthesia was successsfully used without a problem. She presented a good scenario for discussion because her oedema and adiposity would have made for a worrying attempt intubation, and this is often the case because obese, older and oedematous patients are the ones with more severe forms of preeclampsia and the more difficult looking airways. It is a nonsense that a neuraxial block is safe at 50,000 but not at 49000. In these cases induction of general anaesthesia can result in a failed airway and awake intubation can provoke a marked hypertensive response which, ironically in a preeclamptic, could precipitate an intracranial hematoma. So how low can you go with a platelet count and a spinal? Probably lower than we would normally like. Look at the patient and consider the alternatives and the circumstances. Even if a spinal hematoma does develop, the disaster is if it unrecognised and untreated, not if it occurs. It is better to be calling for a neurosurgeon rather than a forensic pathologist.