Post-laparotomy desaturation with basal collapse and overnight transfusion — forks, ventilation, reversal, analgesia and NIV timing

Examiner probes

1. She desaturates on day 2 with basal crackles after an emergency laparotomy. Name the five failure forks and score her ARISCAT-style risk — with the derivation numbers and the external humility?^[2]
2. What should her intraoperative ventilation have been — and what do PROVHILO, PROBESE, iPROVE and the 2026 DESIGNATION close share?^{[5][6][7][10]}
3. Driving pressure keeps appearing. Give the meta-analysis numbers and the PEEP paradox in one sentence.^[8]
4. Rocuronium was used. Sugammadex or neostigmine — and what happens if you let the block wear off?^[17]
5. Her epidural is running. What lung dividend does MASTER allow you to claim — and what must you not claim?^[23]
6. She is now in established hypoxaemic failure. Prophylactic CPAP, HFNC, or NIV — with PRISM, OPERA and Jaber numbers — and where do spirometry and the overnight FFP fit?^[29][26][34]

Model responses

1. Five forks, seven factors, humble scores. Atelectasis/splinting, aspiration, transfusion injury, ARDS, simple hypoxaemia — the ARISCAT composite (infection, failure, bronchospasm, atelectasis, effusion, pneumothorax, aspiration pneumonitis).^[2] Seven factors: low saturation, recent chest infection, age, anaemia, upper abdominal/thoracic surgery, ≥2 h duration, emergency surgery — 2464 patients, 5 percent events, death 19.5 versus 0.5 percent, AUC 0.90/0.88.^[2] Humility: Catalonia validation AUC 0.567, ICU comparison all scores below 0.70 — stratify, never dictate.

2. Low stretch, low PEEP, no routine recruitment. PROVHILO: 900 patients at 8 mL/kg — high PEEP 12 with recruitment versus low PEEP ≤2 — PPC 40 versus 39 percent (RR 1.01) with more hypotension on high PEEP.^[5] PROBESE (2013 obese patients, PPC 21.3 versus 23.6 percent), iPROVE (967 analysed, 46 to 51 percent across four arms) and the 2026 DESIGNATION close (1435 patients, 19.8 versus 17.4 percent) repeat the null in obese, individualised and driving-pressure-guided forms — routine recruitment protects nothing at hypotension cost.^[6][7][10]

3. OR 1.16 per driving-pressure unit, the only mediator — and PEEP that raises driving pressure raises PPC odds 3.11-fold. 17 trials, 2250 patients; tidal volume null.^[8]

4. Sugammadex — and never by waiting. SNaPP: 3498 patients, PPC-or-death 19.0 versus 21.5 percent (RR 0.88) via atelectasis, pneumonia and aspiration unchanged, no ARDS.^[17] Spontaneous recovery is an independent PPC risk factor (reversal protected, OR 0.62) — waiting is the risk factor, not the plan.^[21]

5. Claim the lung, not the life. MASTER: 888 analysed — any-morbidity-or-death 57.1 versus 60.7 percent equal, death equal — only respiratory failure of eight endpoints improved (23 versus 30 percent), with better pain and no major catheter harm.^[23] Epidurals are lung therapy through splinting relief, not mortality therapy.

6. NIV now; prophylaxis never; bundle not talisman; split the transfusion. Established failure gets NIV: Jaber cut reintubation 33.1 versus 45.5 percent with fewer infections.^[29] Prophylactic CPAP changed nothing (PRISM 8.1 versus 8.2 percent) and preventive HFNC changed nothing (OPERA) — pressure treats, it does not prevent.^[26] Spirometry alone failed twice — prescribe the bundle. The FFP fork: donor strategy OR 0.61, male-only FFP OR 0.219 against distress — split TRALI permeability from overload before treating.^[34]