Respiratory Failure and Mechanical Ventilation
Respiratory failure and mechanical ventilation rewards a mechanistic approach: classify the failure, choose the support, then measure whether it worked. The Board tests type 1 versus type 2 failure, ARDS severity, lung-protective settings, weaning physiology and the complications of positive-pressure breathing.
Type 1 versus type 2 failure
Type 1 is hypoxaemic (PaO2 below 60 mmHg on room air) with normal or low PaCO2: pneumonia, pulmonary oedema, pulmonary embolism, ARDS. Type 2 is hypercapnic (PaCO2 above 50 mmHg) from alveolar hypoventilation: COPD, neuromuscular weakness, opioid toxicity, obesity hypoventilation. Shunt - perfused but unventilated units - barely responds to inspired oxygen, whereas V/Q mismatch responds well. That distinction explains why lobar pneumonia needs PEEP and recruitment, while severe asthma needs bronchodilators, expiratory time and patience.
ARDS and lung protection
Berlin criteria: onset within one week of a known insult, bilateral opacities not fully explained by effusion, collapse or nodule, respiratory failure not fully explained by cardiac failure, and P/F ratio at or below 300 mmHg on PEEP of at least 5 cmH2O. Severity is mild 200-300, moderate 100-200, severe below 100. Set tidal volume 6 mL/kg predicted body weight, plateau pressure at or below 30 cmH2O, driving pressure (plateau minus PEEP) at or below 15. Permissive hypercapnia to about pH 7.20 is acceptable. Prone positioning for at least 16 hours daily reduces mortality in moderate-severe disease.
Worked example
A 175 cm man with pneumonia weighs 95 kg. Predicted body weight is 50 + 0.91 x (175 - 152.4) = about 70 kg, so tidal volume is 6 x 70 = 420 mL, not 6 x 95 = 570 mL. Plateau pressure reads 32 cmH2O on PEEP 12, a driving pressure of 20. You drop to 5 mL/kg, accept PaCO2 55 mmHg with pH 7.26, and leave PEEP alone: you are treating the lung, not the blood gas.
Traps
Peak versus plateau pressure - peak reflects airway resistance and falls with bronchodilators, plateau reflects alveolar pressure. Using actual instead of predicted body weight. Recruitment manoeuvres in hypovolaemia, which collapse venous return. Missing auto-PEEP: look for expiratory flow that never returns to zero. Reading weaning failure from saturation alone - a rapid shallow breathing index above 105 predicts failure better.