Massive obstetric haemorrhage and transfusion — structured oral station (12 minutes)

Opening prompt — "Interpret this data for me."

Model response — go abnormality by abnormality, with a threshold attached to each: ^[2][3]

• "She has a massive obstetric haemorrhage with an established coagulopathy and the full lethal triad, plus severe hypocalcaemia."
• "Fibrinogen 1.4 g/L is critically low. In the Charbit cohort a fibrinogen of 2 g/L or below had a 100% positive predictive value for severe postpartum haemorrhage, and the risk rose 2.63-fold for each 1 g/L fall."^[2]
• "FIBTEM A5 of 6 mm confirms that at the bedside. About 12 mm corresponds to a Clauss fibrinogen of 2 g/L, and the published obstetric algorithm treated below 7 mm outright."^[3][4]
• "The EXTEM–FIBTEM pair is the key. EXTEM A5 31 mm with FIBTEM A5 6 mm tells me the deficit is fibrinogen, not platelets. Maximum lysis of 4% tells me she is not hyperfibrinolytic — the tranexamic acid has done its job."^[3][4]
• "Ionised calcium 0.71 mmol/L is severe hypocalcaemia. Temperature 35.1 degrees Celsius and pH 7.19 with a base deficit of 11 and lactate 6.4 mmol/L complete the triad. Urine output 10 mL in an hour means she is already hypoperfusing her kidneys."^[10][11]

Examiner is listening for: named thresholds rather than "abnormal", explicit use of the EXTEM–FIBTEM pair, and recognition of the calcium — which candidates skip most often. ^[3][10]

Probe 1 — "Prescribe. What do you give, in what order, and to what target?"

• "Fibrinogen first — cryoprecipitate or fibrinogen concentrate, targeting a fibrinogen at or above 2 g/L and a FIBTEM A5 above 12 mm, then re-test."^[3][5]
• "Calcium now, not after — 10 mL of 10% calcium gluconate IV, or 10 mL of 10% calcium chloride if she has central access, then re-measure and repeat as transfusion continues."^[10][1]
• "Fresh frozen plasma for the prothrombin time at 1.6 times reference, targeting under 1.5 times — accepting that plasma will worsen her calcium, which is why I check it again."^[1][10]
• "Platelets if the count falls below 50 times 10 to the ninth per litre with ongoing bleeding, or if EXTEM A5 stays low once FIBTEM is corrected. At 92 she does not need them yet."^[1][4]
• "Red cells to maintain oxygen carriage while she is bleeding, all through a warmer. Stop the crystalloid — she has had 3000 mL and it is diluting her."^[1][11]
• "Second dose of tranexamic acid 1 g IV, since bleeding has continued beyond 30 minutes from the first dose."^[15]
• "Active rewarming: warmed products through a rapid infuser, forced-air warming, theatre temperature up, targeting above 36 degrees Celsius."^[11][5]

Probe 2 — "Your consultant says 'just run four units of red cells and four of plasma and two pools of platelets, that's the shock pack'. Do you agree?"

The examiner wants to see you disagree respectfully and with data. ^[4][6]

• "I would give a ratio-based pack if I had no numbers. But I do have numbers, and they tell me she needs fibrinogen, not platelets."
• "The obstetric ROTEM cohort concluded exactly this: coagulopathy is not present in every woman with obstetric haemorrhage and cannot be predicted from blood loss alone, therefore formulaic treatment with blood products is not justified."^[4]
• "The national prospective data support that. In the Welsh cohort of 349 women with massive postpartum haemorrhage, only 22.9% received any coagulation product and only 17.1% had a documented clotting abnormality."^[13]
• "And there is a harm side: introducing fibrinogen concentrate in place of shock packs cut median blood components from 8 to 3 and removed transfusion-associated circulatory overload entirely — 4 of 42 in the shock-pack phase versus none of 51."^[5]
• Then the collegial close: "So I would give the fibrinogen now, hold the platelets until the count or EXTEM justifies them, and re-test in ten minutes. If you would prefer the pack, can we agree to give fibrinogen first and reassess together?"^[4][5]

Probe 3 — "Where does 1:1:1 come from, and what did that trial actually show?"

• "PROPPR, in trauma. It randomised severely injured patients to plasma, platelets and red cells in a 1:1:1 versus 1:1:2 ratio. There was no significant mortality difference at 24 hours or at 30 days. More patients in the 1:1:1 arm achieved haemostasis, and fewer died of exsanguination by 3 hours."^[6]
• "Two caveats for obstetrics: it was a trauma population, and pregnant women start with a fibrinogen of roughly 4 to 6 g/L, so the coagulopathy has a different natural history. Ratios are a bridge until the point-of-care result arrives, not a destination."^[6][7]

Probe 4 — "Three randomised trials of fibrinogen concentrate in postpartum haemorrhage were negative. Why are you giving it?"

This is the discriminating probe of the station. ^[7][8][9]

• "Because none of those trials tested what I am doing. FIB-PPH gave 2 g at a mean baseline fibrinogen of 4.5 g/L and found transfusion in 20% versus 22%. FIDEL gave 3 g at a mean baseline of 4.1 g/L, with failure rates of 40.0% versus 42.4%. OBS2 used a FIBTEM A5 trigger of 15 mm or less and found no benefit."^[8][9][7]
• "All three treated women whose fibrinogen was normal. OBS2's own pre-specified subgroup analysis concluded that replacement is not required above a FIBTEM A5 of 12 mm or a Clauss fibrinogen above 2 g/L, and that the raised term fibrinogen is a physiological buffer rather than a haemostatic requirement."^[7]
• "My patient has a fibrinogen of 1.4 g/L and a FIBTEM A5 of 6 mm. She is below every threshold those trials never tested. The rule is: replace a measured low fibrinogen; do not top up a normal one."^[7][3]
• "There is also a timing signal. In the French population-based analysis, fibrinogen given more than 3 hours after red cell transfusion started carried more than double the risk of near-miss or death compared with giving it before or within 30 minutes."^[14]

Probe 5 — "The bleeding continues. When do you operate, and who decides?"

• "I name the decision point out loud now, not later: if she continues to bleed despite a correctly filled balloon, corrected fibrinogen, correct calcium and warmth, with an ongoing transfusion requirement and haemodynamic instability, we go to theatre."^[1][13]
• "In theatre the ladder is compression, brace suture, stepwise devascularisation, then hysterectomy — and I would make that decision with my consultant now rather than in twenty minutes, because delay is what kills."^[1]
• "I would also re-ask the surgical question directly: is there a bleeding point a clamp or a suture will fix? Genital tract trauma was the leading cause of massive postpartum haemorrhage in the Welsh national cohort at 30.6%, so I want the tract re-inspected under anaesthesia."^[13][1]

Probe 6 — "Anything you would do differently as a unit, rather than as an individual?"

• "Yes, and there is national evidence for it. OBS Cymru introduced quantitative blood-loss measurement, universal risk assessment, structured escalation and viscoelastic-guided early fibrinogen replacement across all 12 Welsh units and 61 094 births."^[12]
• "Massive postpartum haemorrhage fell by 1.10 per 1000 maternities per year, progression from moderate to massive fell from 7.0% to 5.0%, and red cells transfused fell by 7.4 units per 1000 maternities per year — with no increase in women whose lowest haemoglobin was below 80 g/L."^[12]
• "So the system change I would argue for is measurement plus point-of-care testing plus a structured escalation ladder, not a new drug."^[12][4]

Probe 7 — communication. "She is awake, frightened, and asks what is happening. Then, six weeks later, what will you ask her at follow-up?"

Two scored domains in one probe. ^[1][16]

• Now: move to her eye level, use her name, and brief plainly. "You are bleeding more than we would like after the birth. There is a whole team here. We are giving you blood and clotting products, keeping you warm, and we may need to take you to theatre to stop the bleeding. I will keep telling you what is happening."
• Acknowledge fear, keep her partner informed, allocate a named staff member to stay with them, and commit to a debrief.^[1]
• At six weeks, ask three questions deliberately: "Have you been able to breastfeed? Have your periods returned? How are you sleeping and how do you feel about coming back to hospital?"
• "The first two screen for Sheehan syndrome — postpartum pituitary necrosis after severe bleeding, which presents with failed lactation and amenorrhoea and then non-specific hypopituitarism, and which is typically diagnosed years late. The third screens for post-traumatic stress."^[16][1]