Pulmonology · Pathway

COPD management

GOLD 2026 in one place. The ABE group and initial therapy, then the follow-up algorithm for the patient's current inhalers: dyspnea track or exacerbation track, ICS by eosinophils, roflumilast, azithromycin and biologics. A separate exacerbation view covers Rome severity, NIV, ICU, steroids and antibiotics.

GOLD 2026 report v1.3ABE · ≥1 exacerbation = ERome severityUS labelsReviewed Oct 2026

1Symptoms & exacerbations

Enter either. If both are entered, the CAAT decides the group (GOLD prefers a comprehensive symptom score).

2Labs & lung function

3Current maintenance therapy

4History & risk factors

Drug cautions (tick any present; they gate add-on doses)

How it works

ABE group and initial therapy (GOLD 2026 Figures 2.13, 3.8)

New in 2026: group E starts at one moderate or severe exacerbation in the previous year. Before 2026, E needed ≥2 moderate exacerbations or ≥1 hospital admission.

GroupCriteriaInitial therapy (treatment-naïve)
A0 exacerbations · mMRC 0–1 / CAAT <10A bronchodilator. A long-acting one is preferred unless breathlessness is very occasional.
B0 exacerbations · mMRC ≥2 / CAAT ≥10LABA + LAMA
E≥1 moderate or severe exacerbationLABA + LAMA. Consider LABA + LAMA + ICS if eosinophils ≥300 cells/µL (a practical recommendation).

CAAT™ is the renamed CAT™. GOLD says the two are equivalent and the scores are interchangeable. LABA + ICS is not encouraged in COPD: when ICS is indicated, LABA + LAMA + ICS is the preferred choice. With concomitant asthma, treat as asthma; ICS is mandatory. Prescribe rescue short-acting bronchodilators for everyone.

Grades by post-BD FEV₁ (with FEV₁/FVC <0.7): GOLD 1 ≥80% · GOLD 2 50–79% · GOLD 3 30–49% · GOLD 4 <30%.

Follow-up algorithm (GOLD 2026 Figure 3.9)

This algorithm applies to anyone already on maintenance therapy, whatever group they started in. If the response is adequate, continue. If not, check adherence, inhaler technique and comorbidities, then choose a track. When both exacerbations and dyspnea need treating, use the exacerbation track.

Dyspnea track: LABA or LAMA → LABA + LAMA. If dyspnea persists on LABA + LAMA: switch the device or molecules, start or escalate non-drug treatment (for example pulmonary rehabilitation), consider adding ensifentrine where available, and look for other causes of dyspnea.

Exacerbation track (≥1 moderate or severe exacerbation):

  • LABA or LAMA: eosinophils <300 → LABA + LAMA; ≥300 → LABA + LAMA + ICS.
  • LABA + LAMA: eosinophils ≥100 → LABA + LAMA + ICS. Below 100: azithromycin (if not currently smoking) or roflumilast (if FEV₁ <50%, chronic bronchitis and a prior hospitalized exacerbation).
  • LABA + LAMA + ICS: azithromycin or roflumilast as above. A biologic needs two moderate or one severe exacerbation and eosinophils ≥300 (the gate this page uses). Choose dupilumab if chronic bronchitis is present; mepolizumab works with or without it. Label trial criteria differ by drug. Dupilumab (BOREAS/NOTUS): eosinophils ≥300 at screening, chronic productive cough ≥3 months, MRC ≥2. Mepolizumab: MATINEE required ≥300 at screening; METREX had no minimum, and its efficacy population had ≥150 at screening or ≥300 in the past year.
  • LABA + ICS, no asthma (Figure 3.12): current exacerbations → LABA + LAMA + ICS if eosinophils ≥100, or LABA + LAMA if <100. No current exacerbations: if earlier exacerbations responded to LABA + ICS, continue (low symptom load) or consider escalating to LABA + LAMA + ICS (high symptom load); with no relevant exacerbation history, consider changing to LABA + LAMA.

ICS withdrawal: keep the ICS unless it was started inappropriately, there was no response, or there are significant side effects or severe/recurrent pneumonia. With eosinophils ≥300, de-escalation is more likely to be followed by exacerbations.

How this page reads "severe": GOLD defines a severe exacerbation in this context as one needing hospital admission. The US biologic labels also count ED or urgent-care observation over 24 h.

ICS: factors to consider (GOLD 2026 Figure 3.10)
StrengthFactor
Strongly favorsHospitalization for an exacerbation# · ≥2 moderate exacerbations/year# · eosinophils ≥300 · history of, or concomitant, asthma
Favors1 moderate exacerbation/year# · eosinophils 100 to <300
AgainstRepeated pneumonia events · eosinophils <100 · history of mycobacterial infection

#Despite appropriate long-acting bronchodilator maintenance therapy. Eosinophils are a continuum and the cut-points are approximate. ICS has little or no effect below 100 cells/µL. Pneumonia risk on ICS is higher in current smokers, age ≥55, prior exacerbations or pneumonia, BMI <25 kg/m², a poor mMRC grade and severe airflow obstruction (Appendix 3).

Boundaries on this page: eosinophils are classified on the value you enter. 99 is <100, 100 is "100 to <300", 299 is "100 to <300" and 300 is ≥300. The number is shown with enough decimals to match its band.

Exacerbations: severity, ventilation, drugs (GOLD 2026 Chapter 4)

Rome severity (Figure 4.2). Mild is the default: VAS <5, RR <24, HR <95, SaO₂ ≥92% with a change ≤3%, CRP <10 mg/L. Moderate means at least three of five: VAS ≥5, RR ≥24, HR ≥95, SaO₂ <92% and/or a fall >3%, CRP ≥10. Severe: "Dyspnea, respiratory rate, heart rate, SaO₂ and CRP same as moderate", and the ABG shows PaO₂ ≤60 mmHg and/or hypercapnia with acidosis (PaCO₂ >45 mmHg and pH <7.35). So this page grades severe only when ≥3 of 5 moderate criteria are met and the gas is abnormal. An abnormal gas without the clinical criteria is flagged as respiratory failure (NIV, place of care), but it is not called Rome-severe, and no steroid dose is shown until the clinical items are entered. Gases entered in kPa are converted (×7.50062) and rounded to 0.1 mmHg before classification, so 8.0 kPa counts as PaO₂ 60 and 6.0 kPa as PaCO₂ 45. Missing items are never counted as normal: if CRP could change the grade, the page says so.

NIV (Figure 4.8), any of: respiratory acidosis (PaCO₂ ≥45 mmHg [6.0 kPa] and pH ≤7.35) · severe dyspnea with signs of respiratory muscle fatigue · persistent hypoxemia despite oxygen. The NIV gas threshold (≥45, ≤7.35) is slightly wider than the Rome "severe" gas (>45, <7.35). For hypoxemic respiratory failure, HFNT is the first mode; for hypercapnic failure, or no response to HFNT, use NIV unless contraindicated (Evidence A).

Antibiotics: each symptom is a required Yes/No, so "not indicated" appears only when every item is answered. Indicated with increased purulence plus at least one more of increased dyspnea, fever or increased sputum volume; or a positive sputum culture during a prior exacerbation; or mechanical ventilation, invasive or noninvasive (Evidence A). Course: 5 days (Evidence B). GOLD gives 5–7 days as the recommended length and ≤5 days for outpatients. The choice follows local resistance. GOLD names drug classes but no doses, so this page gives none.

Corticosteroids: 40 mg prednisone-equivalent daily for 5 days (Evidence A; REDUCE). Oral is as effective as IV. Oxygen: GOLD 2026 says controlled oxygen for hypoxemia but gives no number. The 88–92% target comes from BTS 2017 (pending blood gases) and the Austin 2010 RCT.

What this page leaves out
  • Doses of inhaled maintenance drugs and nebulized SABA/SAMA vary by product; follow the product label.
  • Antibiotic doses: GOLD names classes only.
  • Steroid equivalents: use the steroid converter.
  • Details of the 2025–26 US COVID-19 schedule: GOLD defers to current WHO/CDC advice, so this page does too.
  • Long-term oxygen, lung-volume reduction, palliative care and the GOLD "disease activity" concept: see the GOLD report.

Sources

  1. Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for Prevention, Diagnosis and Management of COPD: 2026 Report (v1.3, 8 Dec 2025), goldcopd.org — Figures 2.13 (ABE tool), 3.3, 3.6 (vaccines), 3.8 (initial therapy), 3.9 (follow-up therapy), 3.10 (ICS factors), 3.11 (biologics), 3.12 (LABA+ICS), 3.13 (inhalation), 3.16, 4.2 (Rome severity), 4.4 (place of care), 4.5, 4.6, 4.7, 4.8 (NIV), 4.9 (invasive ventilation), 4.10 (discharge), A3.3 and A4.1 (evidence levels); Chapter 3 follow-up text (pp. 65–66) and Chapter 4 pharmacological treatment (pp. 103–105). The current edition as of Oct 2026.
  2. GOLD 2026 Key Changes Summary (10 Nov 2025): group E now includes one moderate exacerbation; one moderate exacerbation is the threshold to consider escalation.
  3. Celli BR, Fabbri LM, Aaron SD, et al. An updated definition and severity classification of COPD exacerbations: the Rome proposal. Am J Respir Crit Care Med 2021;204:1251–8. doi:10.1164/rccm.202108-1819PP
  4. US prescribing information (DailyMed/openFDA): Dupixent (dupilumab), Apr 2026, §1.6, §2.8, §14; Nucala (mepolizumab), Aug 2025, §1.3, §2.1, §14.3; Ohtuvayre (ensifentrine), Apr 2026, §1, §2; Daliresp (roflumilast) §1, §2, §4; Zithromax (azithromycin) §5.4 (QT prolongation).
  5. O'Driscoll BR, Howard LS, Earis J, Mak V. BTS guideline for oxygen use in adults in healthcare and emergency settings. Thorax 2017;72(Suppl 1):ii1–ii90 (target 88–92% in known COPD, pending blood gases; reaffirmed by BTS, Dec 2019).
  6. Austin MA, Wills KE, Blizzard L, et al. Effect of high flow oxygen on mortality in COPD patients in prehospital setting: randomised controlled trial. BMJ 2010;341:c5462. doi:10.1136/bmj.c5462
  7. Leuppi JD, Schuetz P, Bingisser R, et al. Short-term vs conventional glucocorticoid therapy in acute exacerbations of COPD: the REDUCE randomized clinical trial. JAMA 2013;309:2223–31. doi:10.1001/jama.2013.5023
  8. Advisory Committee on Immunization Practices (via Immunize.org, accessed Oct 2026): RSV vaccination for all adults ≥75 years and for adults 50–74 years with risk conditions, including chronic lung disease such as COPD.
  9. Rochwerg B, Brochard L, Elliott MW, et al. Official ERS/ATS clinical practice guidelines: noninvasive ventilation for acute respiratory failure. Eur Respir J 2017;50:1602426. doi:10.1183/13993003.02426-2016

Educational reference only — not a substitute for clinical judgment, institutional protocols, or specialist advice. Calculations run in your browser; nothing you enter is sent or stored.