9.2 Acute Respiratory Failure and ARDS

Two Kinds of Failure, Five Mechanisms of Hypoxemia

Test-plan item II.A.2 asks you to reason about oxygenation and ventilation failure in a patient whose heart is also the problem. Start with the two-category framework:

• Type I (hypoxemic) respiratory failure: PaO2 under 60 mmHg on room air with a normal or low PaCO2. This is a lung/oxygen-transfer problem — cardiogenic pulmonary edema, ARDS, pneumonia, PE, atelectasis after sternotomy.
• Type II (hypercapnic) respiratory failure: PaCO2 above 45-50 mmHg with pH under 7.35. This is a pump/drive/load problem — opioid and sedative excess after cardiac procedures, obesity hypoventilation, neuromuscular weakness, COPD with an acute exacerbation, or exhaustion after prolonged work of breathing.

Both may be present. A heart failure patient with flash pulmonary edema starts hypoxemic and hypocapnic, then becomes hypercapnic as respiratory muscles fatigue — a rising PaCO2 in a tachypneic patient who was previously blowing off CO2 is a pre-intubation finding, not reassurance.

The Five Mechanisms of Hypoxemia

Shunt does not correct with supplemental oxygen because blood bypassing ventilated alveoli never contacts oxygen at any FiO2. That single fact answers a great many CMC items. Clinically: a patient on 100% via non-rebreather with a PaO2 of 55 mmHg has a shunt, and increasing the FiO2 further will not help — positive pressure (CPAP, BiPAP or intubation with PEEP) to recruit collapsed or flooded alveoli is what fixes it.

A uniquely cardiac version: platypnea-orthodeoxia, hypoxemia that worsens when the patient sits upright, caused by right-to-left shunting through a patent foramen ovale when right atrial pressure rises (RV infarction, PE, pulmonary hypertension, or after pneumonectomy). It is refractory to oxygen and is diagnosed with a bubble study, not a chest CT.

ARDS: The Definition and the Numbers

Acute respiratory distress syndrome (ARDS) is defined by the Berlin criteria, extended by the 2023 global definition. All four elements are required:

1. Timing: within 1 week of a known clinical insult or of new or worsening respiratory symptoms.
2. Imaging: bilateral opacities on chest radiograph or CT not fully explained by effusions, lobar/lung collapse, or nodules. The global definition also accepts bilateral B-lines and/or consolidation on lung ultrasound.
3. Origin of edema: respiratory failure not fully explained by cardiac failure or fluid overload. If no ARDS risk factor is present, an objective assessment such as echocardiography is required to exclude hydrostatic edema. Note the wording: "not fully explained by" — heart failure and ARDS routinely coexist, and the presence of a low ejection fraction does not exclude ARDS.
4. Oxygenation: the PaO2/FiO2 (P/F) ratio measured with PEEP or CPAP of 5 cmH2O or more.

The 2023 global definition adds a non-intubated category for patients on high-flow nasal oxygen at 30 L/min or more, and permits SpO2/FiO2 of 315 or less (when SpO2 is 97% or less) as a substitute for the P/F ratio when arterial blood gases are unavailable.

Cardiogenic Pulmonary Edema versus ARDS

The CMC-level judgment is that these are not mutually exclusive: the patient with an EF of 30% and aspiration pneumonitis has both, and the treatment must satisfy both — enough diuresis to relieve hydrostatic congestion without so much that cardiac output falls.

Lung-Protective Ventilation and Its Cardiac Cost

The Settings

• Tidal volume 4-6 mL/kg of predicted body weight (PBW), never actual body weight. PBW for men = 50 + 2.3 x (height in inches - 60); for women = 45.5 + 2.3 x (height in inches - 60). This matters enormously: a 5-foot-4-inch woman weighing 92 kg has a PBW of about 54.7 kg, so a 6 mL/kg tidal volume is roughly 330 mL, not the 550 mL an actual-weight calculation would produce. Ventilating obese ARDS patients on actual weight is one of the most common real-world and test-item errors.
• Plateau pressure 30 cmH2O or less, measured with a 0.3-0.5 second inspiratory hold.
• Driving pressure (plateau minus PEEP) 15 cmH2O or less. Driving pressure normalizes tidal volume to the size of the aerated lung and is the single ventilator variable most tightly associated with survival.
• Respiratory rate up to 35/min to partially compensate for the small tidal volume.
• PEEP set from an ARDSNet lower- or higher-PEEP FiO2/PEEP table, with higher-PEEP strategies favored in moderate to severe disease; recruitment maneuvers are used selectively because sustained high-pressure recruitment increased mortality in the ART trial.
• Oxygenation targets: SpO2 88-95%, PaO2 55-80 mmHg. Hyperoxia is not benign in the cardiac patient — it causes coronary and systemic vasoconstriction.

Permissive Hypercapnia Is Not Free in a Cardiac Patient

Small tidal volumes raise PaCO2. Permissive hypercapnia accepts a PaCO2 of 50-70 mmHg or higher as long as the pH stays at or above 7.20-7.25. The general critical-care teaching is that this is well tolerated. In a cardiac patient it often is not, and CMC will test exactly this:

• Hypercapnia and acidemia are potent pulmonary vasoconstrictors — PVR rises, RV afterload rises, and acute cor pulmonale develops in roughly 20-25% of moderate-to-severe ARDS. A patient with pre-existing pulmonary hypertension, chronic RV dysfunction, or a recent RV infarct has far less reserve.
• Acidemia below pH 7.20 reduces myocardial contractility and blunts the response to catecholamines, so vasopressor requirements climb.
• Hypercapnia causes sympathetic activation with tachycardia and arrhythmia, and raises intracranial pressure — relevant after cardiac arrest.

What the nurse watches for: a rising CVP, a new or louder tricuspid regurgitation murmur, escalating norepinephrine requirement, a falling SvO2 or ScvO2, new hepatic congestion, and on echo a dilated RV with septal flattening. The response is to improve CO2 clearance (increase respiratory rate within limits, reduce dead space by shortening circuit tubing, check for auto-PEEP), correct acidemia, and consider an inhaled pulmonary vasodilator — not to accept a worsening RV because "permissive hypercapnia is standard."

Rescue Therapies

Prone positioning is the intervention with the strongest mortality evidence. PROSEVA enrolled patients with a P/F ratio under 150 on an FiO2 of 0.6 or more with a PEEP of 5 or more, proned for at least 16 consecutive hours per day, and cut 28-day mortality from 32.8% to 16.0%. Proning homogenizes transpulmonary pressure, recruits dorsal lung units, improves V/Q matching, and unloads the RV, which is why it often improves hemodynamics rather than worsening them.

Nursing logistics matter and are testable. Pre-turn: a team of 5-6 plus a dedicated airway lead; verify and secure the endotracheal tube and document depth; hold enteral feeding per protocol or use post-pyloric access; lubricate and tape the eyes closed; move ECG electrodes to the back; empty ostomy appliances and the urinary drainage bag; pre-oxygenate; ensure all lines and the chest tube have slack, and know which lines will be pulled first if something goes wrong. During the turn expect transient desaturation and arrhythmia. In position: swimmer's position with the head and arms repositioned every 2 hours, reverse Trendelenburg to reduce facial edema and aspiration risk. Complications: facial and periorbital edema, pressure injuries at the forehead, chin, chest, iliac crests and knees, brachial plexus injury, corneal abrasion, endotracheal tube obstruction or dislodgement, and line loss. Relative contraindications include unstable spine, open abdomen, and recent sternotomy — though selected post-sternotomy patients are proned with sternal support. If the patient arrests while prone, begin compressions in the prone position over the T7-T10 vertebral bodies while the team turns the patient supine.

Neuromuscular blockade. Cisatracurium (the ACURASYS/ROSE regimen: 15 mg IV bolus, then a fixed 37.5 mg/hour infusion — note this is mg/hour, not a weight-based rate; typical titrated ICU maintenance is 1-3 mcg/kg/min) for up to 48 hours in early moderate-to-severe ARDS improved outcomes in ACURASYS but not in ROSE, where the control arm used lighter sedation. Current practice is selective: severe dyssynchrony, a high driving pressure that cannot otherwise be reduced, or proning intolerance. Deep sedation and analgesia must be confirmed before and throughout paralysis; monitor train-of-four and provide eye care and pressure-injury prevention.

Inhaled pulmonary vasodilators. Inhaled nitric oxide 5-20 ppm or inhaled epoprostenol improve oxygenation and lower PA pressure without systemic hypotension. Neither improves survival in ARDS, so they are a bridge — but in ARDS with acute cor pulmonale in a cardiac patient they are frequently the intervention that stabilizes the RV. Monitor methemoglobin with nitric oxide and never discontinue abruptly (rebound pulmonary hypertension).

VV-ECMO versus VA-ECMO. This distinction is heavily tested on CMC. Venovenous ECMO drains venous blood and returns it to the venous system (femoral drainage with internal jugular return, or a dual-lumen cannula); it provides gas exchange only and no hemodynamic support — the patient's own heart must generate the cardiac output, and arterial saturation is expected to be well below 100% with recirculation as a key troubleshooting concept. Venoarterial ECMO drains venous blood and returns it to the arterial system, providing full cardiopulmonary support for cardiogenic shock and refractory arrest, at the price of increased LV afterload (requiring venting strategies) and differential hypoxemia (Harlequin or north-south syndrome) when native lung function is poor. Choosing VA-ECMO for isolated respiratory failure adds arterial cannulation risk and LV distension for no benefit. EOLIA-derived triggers for VV-ECMO: P/F under 50 for more than 3 hours, P/F under 80 for more than 6 hours, or pH under 7.25 with PaCO2 60 mmHg or greater for more than 6 hours despite optimized ventilation.

Fluids: The Central Tension

The FACTT trial showed that a conservative fluid strategy in ARDS improved oxygenation and increased ventilator-free days (14.6 versus 12.1) without increasing shock or dialysis, though it did not change mortality. That is straightforward in a septic patient who has been resuscitated. It is not straightforward in a cardiac patient, where too little preload drops cardiac output and coronary perfusion, and too much floods the lung and dilates the RV.

The workable rule is resuscitate first, then de-resuscitate, guided by dynamic rather than static measures:

• Use stroke volume variation or pulse pressure variation (a value under about 13% suggests the patient is not fluid-responsive), passive leg raise, or echocardiographic assessment rather than a CVP number alone.
• Follow perfusion endpoints — lactate clearance, urine output, capillary refill, ScvO2 or SvO2, and mental status — not the chest radiograph alone.
• In RV-predominant physiology, diuresis often improves cardiac output by reducing RV distension and restoring septal geometry. In LV-predominant low output, over-diuresis drops output. Knowing which ventricle is limiting is the assessment that decides the plan.

Nursing Priorities in Respiratory Failure and ARDS

• Ventilator-associated event prevention. Head of bed 30-45 degrees, subglottic secretion drainage endotracheal tubes, cuff pressure 20-30 cmH2O, oral care with toothbrushing (chlorhexidine is no longer universally recommended), avoidance of unnecessary circuit changes, and daily coordinated spontaneous awakening and spontaneous breathing trials. Know the surveillance definition: a ventilator-associated condition is a sustained increase of 3 cmH2O or more in daily minimum PEEP, or 20 percentage points or more in daily minimum FiO2, for 2 or more days after 2 or more days of stable or decreasing settings.
• Sedation and analgesia. Analgesia first, then the lightest effective sedation targeted to a Richmond Agitation-Sedation Scale of 0 to -2, except during neuromuscular blockade or the first hours of proning. Deep sedation lengthens ventilation and increases delirium. Screen with CAM-ICU or ICDSC every shift.
• The ABCDEF bundle in full: Assess/treat pain, Both SAT and SBT, Choice of sedation, Delirium monitoring, Early mobility, Family engagement. Early mobility is feasible on the ventilator and is not withheld solely because a femoral line or ECMO cannula is present in units with the appropriate protocol.
• Recognize dyssynchrony: double-triggering, flow starvation with a scooped-out pressure waveform, and ineffective triggering from auto-PEEP. Report waveform patterns, not just "the patient is fighting the vent."
• Suction only for indication (visible secretions, sawtooth flow waveform, rising peak pressure, desaturation), using closed inline suction to preserve PEEP, and pre-oxygenate to avoid derecruitment and hypoxemia-driven pulmonary vasoconstriction.
• Prophylaxis and skin: venous thromboembolism prophylaxis, stress ulcer prophylaxis when indicated, glycemic control, and a proactive pressure-injury plan (ARDS patients are proned, edematous, paralyzed and hemodynamically fragile — a perfect storm).
• Family communication and goals of care. Severe ARDS with refractory shock in a patient with advanced cardiac disease requires early, honest prognostic conversation; the CMC nurse initiates and participates in it.