EBQ:Titrated Oxygen Therapy for COPD Exacerbation

incomplete Journal Club Article
Austin MA et. al.. "Effect of high flow oxygen on mortality in chronic obstructive pulmonary disease patients in prehospital setting: randomised controlled trial". BMJ. 2010. 341:c5462.
PubMed Full text PDF

Clinical Question

In the prehospital setting, does titrating oxygen saturation of patients with presumed AECOPD to SpO2 88-92% compared to high flow oxygen for all decrease prehospital and in-hospital mortality?


Compared to high flow oxygen, titrating supplemental oxygen to a target SpO2 of 88-92% decreased mortality overall, and decreased hypercapnia and acidosis in patients treated according to protocol.

Major Points

  • Patients presenting with dyspnea often receive supplemental oxygen empirically to correct hypoxemia and for symptomatic relief [1]. However, in patients with COPD hyperoxia has been shown to increase the risk of hypercapnia and acidosis, possibly through decreasing respiratory drive and worsening VQ-mismatch by disruption of hypoxic pulmonary vasoconstriction[2]. Furthermore, observational studies have shown an association between high flow oxygen administration and length of stay, ICU admission and mortality[3][1].
  • In this study, paramedics were randomized to administer either high flow oxygen or oxygen titrated to 88-92% SpO2 to patients with suspected AECOPD along with standard treatments such as nebulized bronchodilators.

Study Design

  • Cluster randomized controlled single centre parallel group trial.
  • N=405
    • 214 had confirmed COPD by PFTs in the past 5 years
  • Enrolment: June 2006 to July 2007
  • Setting: Ambulance service in Hobart, Tasmania, Australia


Patient Demographics

  • All patients (n=405)
    • Male: 49%
    • Mean Age: 69
  • Confirmed diagnosis of COPD (n=214)
    • Male: 48%
    • Mean Age: 68
    • FEV1 from most recent PFT: 42.6% predicted
    • Avg Pack years: 45.5

Inclusion Criteria

  • Age ≥ 35
  • Paramedic diagnosis of AECOPD based upon one or more of:
    • appropriateness of symptoms
    • patient-reported history of COPD
    • >10 pack-year smoking history

Exclusion Criteria


  • Paramedics were randomized to administer one of two treatments:
    • Active arm: Oxygen via nasal prongs to target an SpO2 or 88-92% + bronchodilators administered via nebulizer face mask over NPs driven by compressed air.
    • Control arm: High flow oxygen via non-rebreather at 8-10 L/min + bronchodilators administered via nebulizer with oxygen 6-8L/min


Primary Outcome

Prehospital and in-hospital mortality

  • 21/226 (9%) in high flow arm vs. 7/179 (4%) in the titrated oxygen arm. RR 0.42 (95% CI: 0.20 to 0.89) p=0.02

Secondary Outcomes

Incidence of both invasive and non-invasive ventilation

  • 19/213 (9%) in high flow arm vs. 13/166 (8%) in the titrated oxygen arm. RR 0.88 (95% CI: 0.45 to 1.72) p=0.7.
    • NIV 9/226 vs 8/179

Subgroup analysis

Confirmed COPD. (Analysis of PFT within last 5 years + medical records by respirologist blinded to treatment allocation)

  • Mortality
    • 11/117 (9%) in high flow arm vs. 2/97 (2%) in the titrated oxygen arm. RR 0.22 (95%CI: 0.05 to 0.91) p=0.04
  • Mean blood gasses (note the low # of patients who had blood gasses taken)
    • pH: 7.29 (n=19) in high flow arm vs. 7.35 (n=19) in the titrated oxygen arm. Mean difference 0.06. p=0.11
    • paCO2 (mmHg) : 77.8 (n=20) in high flow arm vs. 54.7 (n=20). Mean difference −23.1. p=0.06.

Criticisms & Further Discussion

  • Treatment protocol violation occurred in 56% of patients in the titrated oxygen arm based on ambulance records. All violations involved administration of high flow oxygen at some point pre-hospital.
  • Very few patients had a blood gas taken within 30 minutes of arrival at hospital. Those who did may represent a subgroup of sicker patients.
  • Though hospital staff were instructed to treat according to patient assignment, data regarding whether patients received high flow oxygen in-hospital is largely missing.
  • COPD subgroup defined retrospectively. In addition, this group excluded any patients who lacked recent PFTs, likely excluding patients with bona fide AECOPD.
  • All patients were admitted to hospital which may not represent the typical disposition of all patients brought to EDs by ambulance for suspected AECOPD.
  • Relatively little use of non-invasive ventilation, though this has been shown to improve hypercapnia and other outcomes in AECOPD.
  • Nebulized treatments administered differently in each group.
  • Single centre study, no blinding post-treatment initiation.

External Links

See Also


  • Australian College of Ambulance Professionals (ACAP)
  • FlaemNuova (equipment)