Critical Care Medicine: Common ICU Conditions — Acute Respiratory Failure

Acute Respiratory Failure: A Structured Assessment

Acute respiratory failure means the lungs can no longer do one of their two jobs: oxygenating blood (hypoxemic failure) or eliminating carbon dioxide (hypercapnic failure). Many patients have both at once. History and physical examination are interpreted together with the chest radiograph, chest CT when needed, arterial blood gas (ABG) analysis, and pulse oximetry — and the chest radiograph pattern is one of the fastest ways to narrow the differential (Table 1). MKSAP organizes the initial approach into three sequential questions (Figure 25): secure the airway (Is it patent? Can the patient protect it?), assess oxygenation (Is saturation falling? Is the alveolar–arterial [A–a] gradient widened?), and assess ventilation (Is respiratory drive adequate? Is minute ventilation adequate? Does the ABG show hypercapnia?).

Acute Upper Airway Management

When immediate intubation is not feasible, oropharyngeal or nasopharyngeal airways, supraglottic devices, and laryngeal mask airways can sustain oxygenation and ventilation until the airway is secured definitively. Respiratory failure may also result from obstruction or compression of the upper airway. Partial obstruction presents with tachypnea, stridor, increased respiratory effort, and an upright (tripod) posture with accessory muscle use; absent air movement, inability to speak, or cyanosis indicates complete obstruction — a medical emergency. The character of stridor localizes the lesion: inspiratory stridor points to obstruction at or above the vocal cords, while expiratory stridor and wheezing suggest an intrathoracic process. Patients with respiratory distress or a high risk of deterioration should be intubated early. A normal pulse oximeter reading does not exclude a threatened airway: in partial upper airway obstruction with stridor, oxygen saturation is usually still normal, so the oximeter cannot be used for reassurance — stridor plus distress means the airway is narrowing now, and early controlled intubation is safer than waiting for desaturation, a late and ominous sign.

Hypoxemic Respiratory Failure: Shunt Physiology

Hypoxemic respiratory failure is inadequate oxygenation of hemoglobin. The most common mechanism in the ICU is shunt — an extreme ventilation/perfusion (V/Q) mismatch in which perfused lung units receive no ventilation at all (V/Q = 0). Shunt has two principal causes: alveolar collapse (atelectasis) and alveolar filling with blood, cells, protein, or water. The signature of shunt is a widened A–a oxygen gradient that does not correct with supplemental oxygen. The A–a gradient is derived by subtracting the measured arterial PO₂ from the alveolar PO₂ calculated with the alveolar gas equation:

Alveolar PO₂ = (FiO₂ × [P_atm − 47]) − (1.25 × PaCO₂) — where FiO₂ is the inspired oxygen fraction, P_atm is atmospheric pressure (760 mm Hg at sea level), and 47 mm Hg is the partial pressure of water at 37 °C.

Clinically, acute hypoxemic failure presents with recent onset of increased work of breathing, tachypnea, and anxiety; cyanosis of lips or fingers appears when hemoglobin oxygenation falls severely, typically below 80%. Auscultatory findings help point to the cause — crackles, wheezing, egophony, or rhonchi — while clubbing suggests an underlying chronic disease. Treatment rests on supplemental oxygen, mechanical ventilation with PEEP to reopen alveoli, and, in severe refractory cases, extracorporeal membrane oxygenation (ECMO); disease-specific management follows.

Specific Causes of Hypoxemic Respiratory Failure

Cardiogenic Pulmonary Edema

Acute cardiogenic pulmonary edema is a frequent cause of hypoxemic failure, and whenever no clear ARDS trigger exists, cardiac causes must be actively excluded — cardiomyopathy, mitral or aortic valve disease, myocardial ischemia, and arrhythmias (especially atrial fibrillation with rapid ventricular response). The evaluation includes assessment of volume overload (jugular venous distention, S₃, peripheral edema), electrocardiogram, B-type natriuretic peptide, serial troponins, and echocardiography. Treatment is prompt diuresis plus afterload reduction with vasodilators or inodilators.

Pneumonia

Pneumonia is a common cause of ICU respiratory failure and the most common cause of ARDS developing outside the hospital. In the right clinical setting the chest radiograph is the diagnostic standard — but it is imperfect.

Atelectasis

Atelectasis is a common postoperative complication: pain and impaired respiratory mechanics after thoracic or abdominal surgery lead to shallow breathing and suppressed coughing. Patients may be asymptomatic or show diminished breath sounds at the bases, consolidation on imaging, egophony, rhonchi, and labored breathing. Management is pain control, early mobility, and secretion management; inspiratory muscle training, deep-breathing exercises, mobility programs, and percussion/vibration therapy all help prevent it. Routine bronchoscopy for mucus clearance is no better than chest physiotherapy and is reserved for atelectasis persisting after 24 hours of physiotherapy; when secretions are minimal, positive expiratory pressure devices can recruit collapsed alveoli. Mucolytics such as N-acetylcysteine have no proven role.

Diffuse Parenchymal Lung Disease

Acute exacerbations of diffuse parenchymal lung disease — particularly idiopathic pulmonary fibrosis (IPF) — may follow an inciting event (infection, procedure, drug reaction) or arise without identifiable cause. Diagnostic criteria include a previous or concurrent diagnosis of diffuse parenchymal lung disease, acute respiratory worsening within 1 month, and new ground-glass opacities on imaging.

Pulmonary Embolism

Pulmonary embolism (PE) causes respiratory failure primarily through V/Q mismatch. Acute embolism raises pulmonary vascular resistance both by mechanical obstruction and by vasospasm of adjacent vessels mediated by inflammatory mediators; blood flow is redirected to normal lung that becomes overperfused, and hypoxemia results whenever alveolar oxygen in that lung cannot fully saturate the excess flow. Chest imaging may show peripheral, wedge-shaped, ground-glass opacities (Hampton hump) representing infarcted lung with surrounding atelectasis, and flow through atelectatic lung adds true shunt. Large emboli strain the right ventricle (RV), impairing cardiac output and lowering mixed venous oxygen saturation — a second hit that deepens hypoxemia.

Most hypoxemic PE patients are managed with supplemental oxygen alone, but some need mechanical ventilation — and the physiologic effects of positive-pressure ventilation matter here. Increased intrathoracic pressure reduces venous return and RV preload, while increased alveolar pressure compresses capillaries and raises RV afterload; together these can worsen RV dysfunction and precipitate hemodynamic collapse. There is no absolute contraindication to ventilation in PE, but instability must be anticipated and managed with vasopressors (norepinephrine), oxygen, judicious PEEP, and sometimes inhaled pulmonary vasodilators (nitric oxide or epoprostenol). In hemodynamic collapse, thrombolysis decreases mortality and improves clinical and echocardiographic parameters, weighed against an intracranial hemorrhage risk of up to 2%; when thrombolysis is contraindicated or fails, surgical or catheter-based embolectomy is an option where available.

Acute Inhalational Injuries

Smoke Inhalation

Pulmonary complications are a leading cause of morbidity and mortality in burn patients and in anyone with significant smoke exposure. Direct thermal injury from smoke is usually confined to the upper airways — the exception is steam, which carries heat efficiently enough to burn the entire tracheobronchial tree. After smoke inhalation, about one third of patients develop airway edema or mucosal sloughing from epithelial necrosis, and chest physiotherapy with serial bronchoscopy is often needed to keep the airway clear. Chemical components of smoke injure the distal airways and parenchyma, producing bronchoconstriction, pulmonary edema, V/Q mismatch, and bronchial cast formation; treatment is supportive. Secondary infection is common and a major driver of death — pneumonia is the most frequent complication, particularly from Staphylococcus and Pseudomonas species. Carbon monoxide and hydrogen cyanide toxicity are common after smoke inhalation and should be suspected in every exposed patient.

Chemical Vapor Inhalation

Chemical vapor injuries are uncommon and arise mostly from warfare, industry, farming, and household cleaning or pesticide products. Solubility predicts the site and timing of injury: water-soluble agents strike the mucosal surfaces of the upper airway with rapid symptom onset, whereas water-insoluble agents reach the distal airways and parenchyma with symptoms often delayed. Edema, bronchospasm, asphyxiation, and direct systemic toxicity are common (Table 2); treatment is supportive. After cholinesterase inhibitor exposure (typically organophosphate pesticides), high-dose atropine is given until bronchorrhea and bronchospasm are controlled.

E-Cigarette or Vaping Product Use–Associated Lung Injury (EVALI)

EVALI is a third form of inhalational injury. Vitamin E acetate, used to thicken illicit vaping products, has been linked to the toxicity. Suspicion should be high in patients who use e-cigarettes or vaping devices and present with hypoxemia plus respiratory symptoms (cough, dyspnea) or gastrointestinal symptoms (abdominal pain, vomiting, diarrhea); constitutional symptoms such as fever and chills may accompany. Reported cases fell sharply after the epidemic peaked in September 2019.

Acute Respiratory Distress Syndrome

ARDS is a syndrome of acute hypoxemia and diffuse lung injury triggered by an inciting insult. The most common causes are sepsis and pneumonia, but the full menu of direct and indirect lung injuries is broad (Table 3) — and aspiration of gastric contents is among the classic direct insults.

Pathophysiologically, disruption of surfactant together with vascular endothelial and alveolar epithelial injury allows protein-rich fluid to flood the interstitium and alveoli, causing alveolar collapse, V/Q mismatch, and reduced lung compliance; the histologic correlate is diffuse alveolar damage with hyaline membranes (Figure 26). Harrison's describes the time course in phases: an exudative phase (first week) with neutrophil-rich inflammation, dependent alveolar edema, and intrapulmonary shunting — plus microvascular occlusion that increases dead space, which is why hypercapnia can accompany severe ARDS; a proliferative phase (days 7–21) when many patients improve as type II pneumocytes regenerate the epithelium and restore surfactant; and, in a minority, a fibrotic phase with alveolar-duct and interstitial fibrosis, emphysema-like bullae, progressive vascular occlusion, and a heavy burden of morbidity.

ARDS has no single diagnostic test; diagnosis is clinical, and mortality rises with severity — in the LUNG SAFE study, hospital mortality was roughly 35% for mild, 40% for moderate, and 46% for severe ARDS. Crucially, death usually results from the underlying disease, secondary infection, or multiorgan dysfunction rather than from refractory respiratory failure itself.

Diagnosis: The Berlin Definition — and Its Successor

Ventilatory Management

Most patients with ARDS require invasive mechanical ventilation, and PEEP plus low tidal volume ventilation are the cornerstones of care because they improve survival. The injury in ARDS is heterogeneous — stiff, injured lung lies next to relatively normal lung — so the compliant regions are vulnerable to overdistention by each tidal volume (volutrauma), while repeated opening and closing of dependent alveoli causes atelectrauma. The landmark ARDS Network trial of low tidal volume ventilation (6 mL/kg predicted body weight, plateau pressure ≤30 cm H₂O) versus conventional ventilation (12 mL/kg) reduced absolute mortality by roughly 9 to 11 points (31% vs 40%), and guidelines recommend tidal volumes of 4 to 8 mL/kg of predicted body weight with plateau pressure below 30 cm H₂O.

PEEP counters atelectrauma, improves the homogeneity of the lung parenchyma, and maintains alveolar recruitment to improve V/Q matching. No method of setting PEEP has proven definitively superior: the favored approach is a PEEP–FiO₂ table that matches PEEP to the FiO₂ needed to hold saturations at 88% to 95%, with higher PEEP recommended in moderate to severe ARDS. Notably, a trial pairing higher PEEP with routine recruitment maneuvers against a lower-PEEP strategy in moderate to severe ARDS was stopped amid increased mortality in the higher-PEEP arm — recruitment is not benign. Prone positioning relieves compression of dorsal lung by the heart and mediastinum and improves V/Q matching; a large randomized trial showed a mortality benefit when patients with a PO₂/FiO₂ below 150 were proned early (<48 hours) alongside low tidal volume ventilation. Two large ECMO trials in severe ARDS showed feasibility and trends toward lower mortality, making ECMO a therapeutic option for severe refractory failure — and patients with severe ARDS should be evaluated for it early. Other proposed refinements (recruitment maneuvers, inverse-ratio ventilation, airway pressure release ventilation, esophageal- or driving-pressure–guided PEEP titration, high-frequency oscillatory ventilation) have not shown mortality benefit.

Nonventilatory Management

ARDS mortality has fallen, but morbidity among survivors is substantial, and ICU interventions shape long-term outcomes. Sedation and analgesia strategies aim to limit the complications of sedatives and opioids; current practice favors lightening sedation as soon as safely possible so patients can interact — a multicenter trial found no difference in 90-day mortality between a no-sedation strategy and light sedation with daily interruption, underscoring the value of daily awakening and minimizing mind-altering drugs.

Volume overload is a recognized source of morbidity. In the Fluids and Catheters Treatment Trial, a conservative fluid strategy did not change mortality compared with liberal fluids but improved oxygenation and shortened both ventilator time and ICU stay; hemodynamically stable patients without end-organ hypoperfusion should receive protocol-directed diuresis with deliberate minimization of fluid administration.

A range of other therapies — nutritional modifications, glucocorticoids, vitamin C, macrolide antibiotics, inhaled nitric oxide, prostacyclin analogues, and stem cells or granulocyte–macrophage colony-stimulating factor — have conflicting or limited evidence and are not recommended for ARDS itself; the nuance is that low-dose hydrocortisone is supported for two conditions that frequently accompany ARDS, refractory septic shock and severe community-acquired pneumonia.

Acute Hypercapnic (Ventilatory) Respiratory Failure

Hypercapnic respiratory failure means CO₂ produced by cellular metabolism is accumulating in the blood because alveolar ventilation is inadequate. The governing relationship is PaCO₂ ∝ CO₂ production ÷ alveolar ventilation — and alveolar ventilation is minute ventilation minus dead-space ventilation. This explains an apparent paradox: a patient can breathe rapidly with a high minute ventilation and still be hypoventilating if the dead-space fraction (V_D/V_T) is large. Hypercapnia therefore reflects one of three defects (Table 5): decreased respiratory drive, decreased tidal volume (V_T), or increased dead space relative to tidal volume.

Clinical features are variable and nonspecific: somnolence and myoclonic jerks suggest CO₂ narcosis; increased work of breathing may precede overt failure; and underlying chest wall deformity, neurologic weakness, or polycythemia should prompt evaluation for chronic hypercapnia. Hypercapnic failure frequently coexists with hypoxemia — and treating the hypoxemia can worsen the hypercapnia through increased V/Q mismatch, reduced respiratory drive, and the Haldane effect. This is particularly dangerous in neuromuscular disease, where supplemental oxygen can precipitate rapid CO₂ retention, obtundation, and respiratory arrest; every patient suspected of acute hypercapnic failure needs an ABG even if oxygen corrects the saturation. The pH establishes acuity and severity (Table 6): chronic hypercapnia is buffered by renal bicarbonate retention, so pH changes are modest. Interpretation is context-dependent — an elevated PaCO₂ may be a COPD patient's baseline, while in asthma the same value signals imminent respiratory failure requiring emergent intubation.

Decreased Respiratory Drive and Drug Overdose

Decreased drive diminishes alveolar CO₂ clearance, and patients typically present somnolent and unable to protect the airway. NPPV can be considered when the airway is protectable — it improves minute ventilation and gas exchange — but intubation and mechanical ventilation are indicated if airway protection is compromised, if respiratory acidosis is significant (pH <7.25), or if the patient is hemodynamically unstable. Sedating drugs (illicit or prescribed), anesthetics, and severe alcohol intoxication are the common depressant causes; contact a poison control center to guide agent-specific management and antidotal therapy.

Obesity Hypoventilation Syndrome

Obesity hypoventilation syndrome (OHS) is defined by the triad of obesity, sleep-disordered breathing, and persistent daytime hypercapnia (PaCO₂ >45 mm Hg). The hypercapnia results from low tidal volumes combined with an inappropriately blunted central respiratory response; Harrison's adds that an elevated serum bicarbonate (≥27 mmol/L) with normal renal function is a sensitive screening clue worth confirming with an ABG, and that roughly 90% of OHS patients have coexisting obstructive sleep apnea — often severe. When OHS presents as acute hypercapnic failure in hospital, it should be treated as a diagnosis of exclusion. The 2019 American Thoracic Society guideline recommends starting suspected inpatients on NIV — bilevel positive airway pressure or volume-targeted pressure support — and continuing it until outpatient obstructive sleep apnea evaluation can be completed. ABGs should be rechecked within 2 hours of starting NPPV to confirm improvement; patients who fail adequate NPPV require intubation. Respiratory stimulants (acetazolamide, theophylline, progestins) have only theoretical benefit and limited supporting data. Long-term, substantial weight loss (on the order of 20% to 25% of body weight) can normalize the PaCO₂, though bariatric surgery is often needed to achieve it.

Neuromuscular Weakness

The diaphragm does more than two thirds of the work of inspiration and is innervated by the phrenic nerves from the C3–C5 nerve roots; diaphragmatic weakness is common in the ICU, typically from postsurgical phrenic nerve injury or ICU-acquired weakness. It produces orthopnea, shallow breathing, and paradoxical chest–abdomen movement; weak cough, trouble managing secretions, or a voice change points to bulbar weakness (intercostal and abdominal muscles, supplied by lower cervical and upper thoracic roots, power the cough). Assessment of cranial nerves and respiratory muscle strength is essential, and effort-dependent pulmonary function measures are the practical tools (Table 7) — bearing in mind that bulbar, facial, or postural weakness limits accuracy and that severe obstructive disease can make pressures falsely low.

Guillain-Barré syndrome and myasthenic crisis are the most common causes of acute neurologic respiratory failure in the ICU. Guillain-Barré syndrome presents with ascending symmetric paralysis and areflexia evolving over 2 to 4 weeks; dysautonomia is common and can cause hemodynamic instability or arrhythmias. Myasthenic crisis is defined by muscle fatigability — diplopia, ptosis, dysarthria, limb weakness, and weak cough are typical. About 25% of Guillain-Barré patients develop respiratory failure and require intubation, since respiratory function may take days to weeks to recover; in myasthenic crisis, early NPPV can prevent intubation, provided the ability to protect the airway and to cough is assessed carefully. Acute spinal cord injury at or above C5 invariably requires mechanical ventilation; some recovery of independent breathing is possible (complete injury below C3, incomplete injury above C3), but atelectasis, aspiration pneumonia, and pulmonary embolism cause recurrent hypercapnic failure, making NPPV and mechanical cough-assist devices essential.

Restrictive Chest Wall Disease

Restriction from parenchymal lung disease causes hypoxemia without hypercapnia; hypercapnic failure is instead the signature of extrapulmonary chest wall restriction — pectus deformity, scoliosis, kyphosis — which compromises respiratory mechanics. Ascites and severe bowel distention do the same by pushing the diaphragm cephalad. These patients usually live with poor ventilatory reserve rather than overt failure, but acute insults such as infection or sedating medications can tip them into hypercapnic respiratory failure; NPPV or invasive ventilation supports them while the precipitant is treated. For thoracic cage deformity, NPPV is frequently sufficient.

Obstructive Lung Disease: Asthma and COPD

The physiologic hallmark of obstructive disease is hyperinflation. High airway resistance — from structural change, inflammation, cholinergic tone, and mucus plugging — combined with the loss of elastic recoil of emphysema increases total lung capacity and functional residual capacity. Because air cannot empty before the next breath begins, functional residual capacity climbs dynamically, and when intrathoracic pressure remains positive at end-exhalation because of this trapped volume, the result is auto-PEEP. Dynamic hyperinflation, auto-PEEP, and reduced recoil together drive disordered mechanics, increased work of breathing, and respiratory muscle fatigue; during mechanical ventilation, auto-PEEP can also reduce venous return and cause hemodynamic instability. The two levers that reduce auto-PEEP are prolonging expiratory time (lower respiratory rate) and bronchodilators.

COPD exacerbations present with increased work of breathing and, when hypercapnia is present, somnolence. Excess oxygen worsens hypercapnia through three mechanisms — increased V/Q mismatch, decreased respiratory drive, and the Haldane effect — so oxygen is titrated to an SpO₂ of 88% to 92%. For hypercapnic failure from COPD exacerbation, NPPV is the preferred initial means of lowering PaCO₂, with bilevel positive airway pressure the most used and most effective mode; patients who cannot protect their airway, who are hemodynamically unstable, or who fail NPPV require intubation.

Severe asthma shows its own warning signs: upright posture, inability to speak in full sentences, diaphoresis, nasal flaring, and inspiratory accessory muscle use signal impending respiratory failure. Blood gases typically show respiratory alkalosis from rapid shallow breathing — so a normal or elevated PaCO₂ in a tiring asthmatic is an ominous sign of imminent collapse. Oxygen is given to maintain saturations of 90% to 95%. For exacerbations refractory to conventional therapy, magnesium sulfate or heliox may be tried despite lower levels of evidence; other adjuncts lack demonstrated efficacy, and antibiotics are reserved for documented infection. Little evidence supports NPPV in asthma — if it is used, close monitoring is mandatory — and for most asthmatics with respiratory failure, intubation and mechanical ventilation are recommended. As in COPD, the ventilator strategy centers on maximizing expiratory time to reduce auto-PEEP, which often requires sedation or paralysis and may lower minute ventilation (with permissive rises in PaCO₂); frequent ABGs guide management.

Post-Intensive Care Syndrome

Survival is only the first outcome that matters. Post-intensive care syndrome describes the constellation of symptoms that follow an episode of critical illness, grouped by domain: physical impairment, mental health disorders, and cognitive impairment. Affected patients have increased health care use, increased morbidity and mortality, and impaired quality of life — and the syndrome reaches beyond the patient, with family members and caregivers experiencing anxiety, depression, and posttraumatic stress disorder. Current research and quality-improvement efforts focus on better recognition, prevention, diagnosis, and management of these sequelae, tying directly into the ARDS survivorship data showing that physical limitation and psychological burden persist even when lung function recovers. The next sections of this chapter move from respiratory failure to the circulation: shock and sepsis.