Acute COPD Exacerbation in the ICU

[1]

Definition

An acute exacerbation of COPD (AECOPD) is a sustained worsening of dyspnoea, cough, or sputum beyond the day-to-day variation, that is acute in onset and warrants a change in regular medication. The commonest triggers are respiratory infections (around 70 per cent — viral such as rhinovirus or influenza, or bacterial such as Haemophilus influenzae, Streptococcus pneumoniae, Moraxella catarrhalis), with environmental exposures, pneumothorax, pulmonary embolism, cardiac failure, and poor adherence to inhalers as other recognised precipitants.[9][10]

In the ICU, the entity of interest is the severe AECOPD with hypercapnic (type 2) respiratory failure and respiratory acidosis — the patient whose ABG shows a low pH with a high PaCO2. This is the patient for whom bilevel NIV was designed and proven.[1]

Anthonisen criteria — is the exacerbation infective?

The Anthonisen criteria stratify the likelihood of a bacterial trigger and so guide the decision to give antibiotics. A bacterial exacerbation is likely with two or (especially) three of the following cardinal features:[6]

1. Increased sputum volume
2. Increased sputum purulence
3. Increased dyspnoea [1]

Antibiotics reduce treatment failure and mortality when all three (type 1) or two of three (type 2) are present; benefit is marginal with only one feature (type 3).[6]

Pathophysiology — why the COPD patient retains CO2

The respiratory failure of an AECOPD is type 2 (hypercapnic) with a superimposed metabolic acidosis from hypoxaemic lactic acidosis. Four mechanisms operate:[9]

1. Increased load, reduced capacity. Bronchospasm, mucosal oedema, and secretions raise airway resistance; dynamic hyperinflation (gas trapping) adds an elastic threshold load — the patient must generate a pressure equal to the intrinsic PEEP (auto-PEEP) before any air flows. Respiratory muscle fatigue follows, alveolar ventilation falls, and PaCO2 climbs. [1]

2. V/Q mismatch. Areas of low ventilation relative to perfusion increase the physiological dead space and venous admixture, raising PaCO2 and lowering PaO2. [1]

3. The Haldane effect. Deoxygenated haemoglobin carries CO2 more readily (as carbamino compounds and bicarbonate). Oxygenating haemoglobin releases this CO2. In a patient already ventilation-limited, the released CO2 cannot be blown off and PaCO2 rises acutely — the molecular basis of oxygen-induced hypercapnia. [1]

4. Worsened V/Q matching from relieved hypoxic pulmonary vasoconstriction. High FiO2 abolishes hypoxic pulmonary vasoconstriction, redistributing blood to poorly ventilated units and increasing dead space. [1]

Clinical presentation

The patient is usually a known COPD sufferer, though AECOPD may be the first presentation. Typical features: [1]

• Breathlessness that has worsened over hours to days, with accessory muscle use, pursed-lip breathing, and inability to speak in full sentences
• Tachypnoea initially, then bradypnoea as the patient tires — a pre-arrest sign
• Wheeze and a prolonged expiratory phase; a silent chest is exhaustion and impending arrest
• Productive cough, sputum volume or purulence increased if infective
• Confusion or drowsiness from CO2 narcosis (a low GCS is a relative contraindication to NIV and a trigger to intubate)
• Signs of haemodynamic compromise (tachycardia, hypotension) in the advanced case [1]

A sudden unilateral deterioration with tracheal deviation suggests a tension pneumothorax — treat it before treating the exacerbation. [1]

Severity — the ABG-driven staging

The arterial blood gas is the single most important investigation. The pH defines severity and drives the level of respiratory support.[1][7]

Investigations

1. Arterial blood gas — the pivotal test. Record pH, PaCO2, PaO2, bicarbonate, base excess. A compensated respiratory acidosis (high bicarbonate) confirms chronic CO2 retention. Recheck after one hour of NIV.
2. Chest X-ray — to find a pneumothorax, consolidation, pulmonary oedema, or a large pleural effusion (i.e. to find the trigger and the mimics).
3. ECG — atrial fibrillation and right heart strain (cor pulmonale) are common; exclude myocardial ischaemia.
4. Full blood count, electrolytes, CRP — polycythaemia from chronic hypoxia, the hypokalaemia of beta-agonist therapy, and an infective marker.
5. Sputum and blood cultures — if the exacerbation is infective, before antibiotics.
6. Troponin and BNP — if a cardiac contribution is suspected.
7. Venous blood gas — a useful screen; a normal venous pH with an acceptable PaCO2 estimate makes a significant acidosis unlikely, but an arterial gas is definitive.
8. Theophylline level — only if the patient is taking a methylxanthine. [1]

Management — the bundle and the escalation

The management of the severe AECOPD runs in parallel: the medical bundle (oxygen, bronchodilators, steroids, antibiotics) is applied to every patient, while the respiratory support is escalated according to the ABG pH.[9][10]

The medical bundle

[1]

The escalation pathway — NIV to intubation

The respiratory support is ABG-driven. The following FlowSteps trace the complete pathway from the acidotic patient on the ward to the intubated patient in the ICU.[1][7]

[1]

Bilevel NIV — the settings and the technique

NIV in AECOPD is bilevel (BiPAP), not CPAP. CPAP alone does nothing for the work of breathing against auto-PEEP and is not first-line. The IPAP (inspiratory positive airway pressure) unloads the fatigued muscles; the EPAP (expiratory positive airway pressure) splints the airways open and counteracts intrinsic PEEP.[2][3][7]

Technique points:

• Use a well-fitting oronasal (full-face) mask first; switch to a nasal or total-face mask if leak or claustrophobia is a problem.
• Let the patient hold the mask for the first few minutes to build tolerance before strapping it on — reduces claustrophobia and refusal.
• Start low and titrate up — an aggressive initial pressure causes leak, mask intolerance, and aerophagia.
• Set a backup respiratory rate modestly below the patient's own (e.g. 12-14) so it does not over-ventilate.
• Reassess at one hour: clinical (RR, accessory muscle use, GCS) and biochemical (ABG pH and PaCO2). [1]

Contraindications to NIV

Absolute (or near-absolute) — favour intubation: respiratory arrest, inability to protect the airway (low GCS, copious secretions), severe agitation or confusion precluding cooperation, haemodynamic instability, facial trauma/surgery, upper airway obstruction. [1]

Relative — proceed with caution in a monitored area: pH <7.25, pneumonia on CXR, and the patient who fails to improve at one hour.[7]

NIV versus invasive ventilation — when to choose which

Invasive ventilation settings for the intubated COPD patient

The goal in the intubated COPD patient is to deliver an adequate tidal volume while leaving enough time for full exhalation, so that gas is not trapped and auto-PEEP does not accumulate. Failure to do this produces dynamic hyperinflation, hypotension (impaired venous return), barotrauma, and the auto-PEEP arrest.[9]

[1] [1]

Pharmacotherapy in detail

Bronchodilators

Nebulised salbutamol 5 mg with ipratropium 500 micrograms, four- to six-hourly, is the standard; severe exacerbations may need it continuously. In the patient not yet on NIV, drive the nebuliser with air, not oxygen, to avoid worsening hypercapnia. Once on NIV the oxygen is controlled by the ventilator. There is no role for intravenous beta-agonists.[9]

Corticosteroids

Systemic steroids shorten recovery, reduce early relapse, and improve PaO2 and dyspnoea. The REDUCE trial (JAMA 2013) showed a 5-day course of prednisolone 40 mg daily was non-inferior to a 14-day course — so 5 days is now standard, stopping without taper.[5] The earlier Niewoehner VA Cooperative trial established benefit over placebo but warned of hyperglycaemia and neuromuscular complications with prolonged courses.[4] Give hydrocortisone 100 mg IV if oral absorption is uncertain.

Antibiotics

Give antibiotics for an Anthonisen type 1 or 2 exacerbation (the classic evidence is Anthonisen 1987, showing benefit in type 1).[6] First-line oral: amoxicillin-clavulanate 500/125 mg TDS or doxycycline 100 mg BD for 5-7 days. In the severe/ICU patient, broaden to cover Pseudomonas if risk factors (recent hospitalisation, antibiotics, severe airflow limitation, bronchiectasis) are present. A procalcitonin-guided strategy safely reduces antibiotic exposure.

Evidence and landmark trials

[1]

Complications

• NIV-related: mask pressure sores, nasal bridge ulceration, aerophagia and gastric distension, conjunctivitis from leak, patient intolerance. Mostly avoided with good mask fit and titration.
• Invasive-ventilation-related: ventilator-acquired pneumonia, dynamic hyperinflation and the auto-PEEP arrest, barotrauma (pneumothorax), weaning difficulty, critical-illness myopathy (steroid- and immobility-associated).
• Pharmacotherapy: hyperglycaemia and steroid myopathy (corticosteroids), hypokalaemia and tachycardia (beta-agonists), Clostridioides difficile and resistance (antibiotics).
• Disease-related: cor pulmonale, atrial fibrillation, pneumothorax, pulmonary embolism. [1]

Early mobilisation and weaning

Once the patient is improving — the pH is normalising, the PaCO2 is falling, secretions are controlled, and the trigger is treated — begin weaning NIV: reduce the IPAP in 2 cmH2O steps, increase off periods, and move to a Hudson mask for meals. Do not stop NIV abruptly.[10]

For the intubated patient, use daily sedation breaks and a spontaneous breathing trial (SBT) once FiO2 is <40 per cent, PEEP is <8, and the patient is triggering. COPD patients are at high risk of prolonged ventilation — a protocolised weaning approach and early mobilisation reduce ventilator-days. Early mobilisation (sitting out of bed, marching on the spot, active limb exercises within 24-48 hours of stability) prevents deconditioning and steroid myopathy and is part of standard ICU rehabilitation. [1]

Prognosis

Hospital mortality for an acidotic AECOPD treated with NIV is around 8-10 per cent, rising to 20-25 per cent when invasive ventilation is required.[1][11] Readmission within 90 days is common (~20 per cent) and predicts poorer long-term survival. Poor prognostic features include advanced age, low BMI, comorbid cardiac disease, ongoing smoking, and severe baseline airflow obstruction.

Exam practice

[1] [1]

Clinical pearls

Red flags

References

1. [1]Plant PK, Owen JL, Elliott MW. Early use of non-invasive ventilation for acute exacerbations of chronic obstructive pulmonary disease on general respiratory wards: a multicentre randomised controlled trial Lancet, 2000.PMID 10859037
2. [2]Brochard L, Isabey D, Piquet J, et al. Reversal of acute exacerbations of chronic obstructive lung disease by inspiratory assistance with a face mask N Engl J Med, 1990.PMID 2122253
3. [3]Brochard L, Mancebo J, Wysocki M, et al. Noninvasive ventilation for acute exacerbations of chronic obstructive pulmonary disease N Engl J Med, 1995.PMID 7651472
4. [4]Niewoehner DE, Erbland ML, Deupree RH, et al. Effect of systemic glucocorticoids on exacerbations of chronic obstructive pulmonary disease. Department of Veterans Affairs Cooperative Study Group N Engl J Med, 1999.PMID 10379017
5. [5]Leuppi JD, Schuetz P, Bingisser R, et al. Short-term vs conventional glucocorticoid therapy in acute exacerbations of chronic obstructive pulmonary disease: the REDUCE randomized clinical trial JAMA, 2013.PMID 23695200
6. [6]Anthonisen NR, Manfreda J, Warren CP, et al. Antibiotic therapy in exacerbations of chronic obstructive pulmonary disease Ann Intern Med, 1987.PMID 3492164
7. [7]Osadnik CR, Tee VS, Carson-Chahhoud KV, et al. Non-invasive ventilation for the management of acute hypercapnic respiratory failure due to exacerbation of chronic obstructive pulmonary disease Cochrane Database Syst Rev, 2017.PMID 28702957
8. [8]Ram FS, Picot J, Lightowler J, Wedzicha JA. Non-invasive positive pressure ventilation for treatment of respiratory failure due to exacerbations of chronic obstructive pulmonary disease Cochrane Database Syst Rev, 2004.PMID 15266518
9. [9]Barnes PJ, Burney PG, Silverman EK, et al. Chronic obstructive pulmonary disease Nat Rev Dis Primers, 2015.PMID 27189863
10. [10]Rabe KF, Wedzicha JA. Controversies in treatment of chronic obstructive pulmonary disease Lancet, 2011.PMID 21907867
11. [11]Waeijen-Smit K, Crutsen M, Keene S, Miravitlles M. Global mortality and readmission rates following COPD exacerbation-related hospitalisation: a meta-analysis of 65 945 individual patients ERJ Open Res, 2024.PMID 38410700