Last updated: Lesson of the Month - January 2019…
on 01 Jan 2019

December 2012

Matthew Graham-Brown

A 38 year old male patient was admitted from home. He had been found by a family member collapsed and unresponsive, and they immediately called 999. The ambulance crew noted that he was not moving his left side and called a 'FAST POSITIVE' alert to the local hospital. On arrival he was met by the advanced nurse practitioner of the stroke team, who agreed the patient was not moving his left side, and that an urgent CT head was required; after which the patient was to be sent to the acute stroke unit for further review by the medical team. GCS at the time was 11/15 (E-3, V-3, M-5). The CT head was completed and reported as entirely normal: 'no signs of ischaemia or intracerabral haemorrhage. No space occupying lesion identified. No evidence of small vessel disease - unremarkable scan'. The patient had routine bloods taken, including FBC, U&E, LFTs, CRP and clotting screen, and was sent up to the stroke unit as planned.

On arrival on the medical unit baseline observations were done:

  1. BP - 101/63
  2. Pulse - 110
  3. RR - 24
  4. Sats - 100% OA
  5. Temp - 39.1

Low grade temperature had been noted in the emergency department, but not this high. On review by the medical staff the patient was now more rousable, and moving all four limbs equally. He was saying words, but was not coherently. GCS was 13/15 (E-4, V-4, M-5). Clinically, the picture was no longer consistent with acute stroke, and on discussion with the family member who brought him to the hospital, it was discovered that the patient had been behaving strangely over the previous 3-4 days, although no specific symptoms could be elicited. The stroke consultant reviewing the patient felt considering the fluctuant GCS, changing neurological signs, prodromal history and pyrexia, it was essential to cover the patient for potential meningitis (although it was noted that on full examination there were no signs of rash, photophobia or meningism). Intravenous antibiotics were given as per the trust protocol as well as intravenous acyclovir. A plan was made and handed over in person to the on-call SHO to review bloods; and that it was safe to perform a lumbar puncture. The team also flagged the patient up to critical care outreach as a potentially sick patient who would require full escalation to ITU if he deteriorated.

The evening SHO dutifully chased up the blood results:

  1. FBC: Hb - 8.3, Plt - NA, WCC - 7.32, (with normal white cell differential)
  2. U&E: Na - 140, K - 3.9, Ur 14.3, Cr 128, eGFR - 48
  3. LFT: ALT - 20, ALP - 81,  Billi - 38
  4. CRP - 7
  5. INR - 1.0 with normal clotting screen

The SHO noted the normal clotting, and also the plan for lumbar puncture. He also noticed that for some reason the platelet count was not available, and on discussion with the laboratory he discovered this was due to 'platelet clumping'. He re-sent a repeat sample in a citrate tube, and asked the lab to process it immediately, and to contact the oncall team as soon as the platelet count was available. The SHO then attended the evening handover where he flagged the patient up to the night team (including medical registrar and SHO); asking them to review the repeat platelet count, to review the patient's clinical state, and to consider LP as per the day plan, if safe.

The laboratory called the night SHO as agreed with the platelet result:

Platelet count - 5 (!)

The night SHO immediately asked the laboratory to prepare a blood film and to process an urgent LDH on the biochemistry sample that the lab had already received. He went straight to the patient to review the clinical state, and the notes, and very quickly put together the following summary:

'38 year old patient admitted with fluctuating conscious level and altering neurological signs, pyrexia, low platelets and anaemia with normal clotting, normal inflammatory markers and deranged renal function'. (He also noted the patient had a slightly raised billirubin with otherwise normal LFTs). The SHO felt the patient had many features of thrombotic thombocytopaenic purpura (TTP), and that this should now be considered the most likely diagnosis. He rang the on-call medical registrar, and informed them of the platelet result and his desire to contact the on-call haematologist to review the blood film and confirm the diagnosis. The registrar agreed with the plan and the consultant haematologist was immediately contacted. He came into the hospital to review the blood film that had been prepared: 

The blood film showed red cell fragments (see arrows on blood film - aka 'Helmet cells') and confirmed a microangiopathic haemolytic anaemia (MAHA). LDH was also extremely high clinching the diagnosis of TTP. A cross match blood sample was sent to the lab and the haematology consultant himself brought FFP to be given overnight. He also contacted the oncall renal physician; and arranged for transfer to the renal unit for plasma exchange the following day, under joint care of renal physicians and haematologists

The patient had several sessions of plasma exchange and improved dramatically over the following week. ADAMTS13 was sent and confirmed the diagnosis. The patient was discharged 17 days after admission when platelet count was greater than 150, having been started on immunosuppression, aspirin and low molecular weight heparin, with haematology follow up


  1. The communication between medical staff in this case was exemplary. When plans were made, they were completed, followed up and acted upon appropriately. The Dr's reacted to unexpected results, reviewed diagnoses and treatment plans appropriately - and, crucially, responded to the developing clinical picture. A very difficult diagnosis was reached with 8 hours of the patient arriving in hospital, meaning the correct treatment was started very quickly
  2. The stroke team who initially reviewed the patient made a sensible plan based on their clinical findings, and the limited history that was available, but they also ensured other bases were covered by doing baseline blood tests, following up the CT scan and reviewing the patient properly after the scan
  3. The second diagnosis of meningitis was also perfectly reasonable given the information available at that time and the clincal findings. They also ensured relevant teams (critical care and ITU) were aware of the patient's current status, and the need for higher level care should he deteriorate
  4. The on-call SHO in the evening did extremely well to continue to ensure patient safety by not proceeding to LP without knowing the platelet count. It would have been very easy to assume that with normal clotting and 'clumped' platelets that it would be safe to proceed to LP. But he did exactly the right thing, and avoided potential disaster by not doing an LP on someone with very low platelets
  5. When the night SHO got the crucial information about the low platelets, he again responded appropriately by reassessing how this new piece of information fitted in with the clinical case. He also made the correct diagnosis and asked (appropriately) for an urgent blood film and LDH. He then involved his immediate senior so they were aware of the development, and the oncall haematologist who was required to review the film
  6. The haematologist not only confirmed the diagnosis, but admirably brought the FFP up to the ward himself to ensure there was no delay in getting it to the patient. He also liased with the renal team for transfer himself, ensuring the patient received the absolute best care


TTP is charactersied by a pentad:

  1. Thrombocytopaenia
  2. Micro-angiopathic haemolytic anaemia (MAHA)
  3. Pyrexia
  4. Renal Failure
  5. Fluctuating neurological signs

However, not all 5 have to be present for a diagnosis of TTP to be made. Indeed up to 35% of patients do NOT have neurological signs at presentation and pyrexia and renal failure are often not prominent features. Revised criteria suggest that TTP must be considered in the presence of thrombocytopaenia and MAHA alone

TTP is rare, with a reported incidence of 6 cases per million per year in the UK. It is essential to make the diagnosis as untreated mortality is ~90%. Patients with TTP have a defective von Willebrand factor (vWF) cleaving protease (ADAMTS13), which is an enzyme that normally degrades large vWF multimers (deficiency can be congenital or acquired). Deficiency of this enzyme leads to accumulation of vWF multimers and systemic platelet activation under conditions of high sheer stress (as in the microcirculation), with thrombosis. The rationale for plasma infusions and exchange is to replace the vWF cleaving protease, to reduce circulating antibodies, and to remove large vWF multimers from the circulation

For further information see the British Journal of Haematology guideline.

Back To Index