The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR

Registration number
Ethics application status
Date submitted
Date registered
Date last updated
Type of registration
Prospectively registered

Titles & IDs
Public title
Multicentre prospective study of pulse oximetry to evaluate arterial oxygen saturation in the clinical setting.
Scientific title
In patients requiring an arterial blood gas, a comparison between arterial oxygen saturation and pulse oximetry saturation measurements.
Secondary ID [1] 285529 0
Universal Trial Number (UTN)
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Any condition requiring measurement of arterial blood gas 293349 0
Condition category
Condition code
Respiratory 293619 293619 0 0
Other respiratory disorders / diseases

Study type
Patient registry
Target follow-up duration
Target follow-up type
Description of intervention(s) / exposure
Participants who require an arterial blood gas (ABG) will be identified. At the time of the ABG an oximetry value (SPO2) will be measured, taken from an earlobe or finger probe. The SpO2 value will be the first value measured following visualisation of blood entering the collection vial for the ABG analysis. Only one paired ABG/SpO2 measurement will be recorded for each participant.
Intervention code [1] 290474 0
Not applicable
Comparator / control treatment
The SpO2 value recorded by the pulse oximeter will be paired with the ABG measured oxygen saturation (SaO2) value for comparison.
Control group

Primary outcome [1] 293424 0
SpO2, measured by pulse oximetry at the same time an ABG is performed.
Timepoint [1] 293424 0
The SpO2 value will be the first value measured following visualisation of blood entering the collection vial for the ABG analysis.
Primary outcome [2] 293425 0
SaO2, measured by ABG (while the pulse oximeter is in place).
ABG is assessed by means of a blood sample taken from an artery which is then sent to the laboratory for analysis.
Timepoint [2] 293425 0
Taken at time of the clinically indicated ABG.
Secondary outcome [1] 311141 0
Timepoint [1] 311141 0

Key inclusion criteria
Patients requiring an ABG measurement as part of their clinical care at: Wellington Regional and Christchurch Hospitals in New Zealand; Westmead Prince Charles, Royal North Shore and Concord Hospitals in New South Wales, Australia; Austin Hospital, Victoria, Australia; or Royal Hobart Hospital, Tasmania, Australia.
Minimum age
16 Years
Maximum age
No limit
Both males and females
Can healthy volunteers participate?
Key exclusion criteria
1. Diagnosis of sickle cell anaemia, methaemoglobinemia, or carbon monoxide (CO) poisoning
2. Patients previously recruited to the study with paired SpO2 and SaO2 values successfully recorded
3. Any other condition which, at the investigator’s discretion, is believed may present a safety risk or impact upon the feasibility of the study or the interpretation of the study results.

Study design
Natural history
Defined population
Statistical methods / analysis
Summary of Objectives:

1. To describe the agreement between SpO2 and SaO2 measurement in terms of bias and limits of agreement.

2. To estimate the influence on bias of:
a. The mean oxygen saturation (average of SpO2 and SaO2)
b. Location of measurement (ED, HDU, ward, or outpatient department)
c. Position of the probe (finger or ear)
d. Doctor’s diagnosis of recognised condition associated with chronic respiratory failure*
e. Current tobacco smoking status (current versus ex or non-smoker)
f. Skin pigmentation (based on modified Fitzpatrick scale with patient skin colour classified as either: Light (Type I to Type II), Medium (Type III to Type IV) or Dark (Type V to Type VI))
g. A doctor’s diagnosis of Diabetes Mellitus.

3. To describe the diagnostic performance of SpO2 to detect clinically important boundaries of SaO2 (90 %) and PaO2 (60 mmHg).

4. To estimate the variance component of the difference between SpO2 and SaO2 difference due to different oximetry devices.

* Chronic obstructive pulmonary disease, obesity hypoventilation syndrome, bronchiectasis, cystic fibrosis, neuromuscular disease and chest wall deformities such as severe kyphoscoliosis

Objective 1: Bland-Altman plots and estimation of bias and limits of agreement.
Objective 2: Analysis of Covariance. Should important predictors of bias be identified Bland-Altman methods will be used to determine whether there is also an effect on limits of agreement.
Objective 3: Estimation of variance components and associated intra-class correlation coefficients for the effect of oximeters as well as Best Linear Unbiased Predictors of the effect of individual oximeters; all by mixed linear models and estimation by Restricted Maximum Likelihood.
Objective 4: Receiver Operating Characteristic curve estimation by logistic regression and associated prediction equations.

The planned sample size is based on three considerations. Firstly, for the analysis of variables that predict the size of the bias we seek to have between 20 and 40 participants for each degree of freedom in the ANCOVA. This requires between 200 and 400 participants. Secondly the estimates of paired SD for the SpO2 to SaO2 difference from the papers of Van de Leow and Wouters were 2.1%, and 2.2% respectively. If there were two equal sized groups of 42 participants, 21 in each group, there is 80% power, with a type I error rate of 5%, to detect this size difference. This suggests if we have at least 20 participants with a particular characteristic we will have greater than 80% power to detect a difference of 2% between groups for dichotomous variables. We think that a sample size of 400 makes it likely that each individual characteristic will have at least 20 participants. Finally for estimation of variance of components for the different pulse oximeters by Best Unbiased Linear Predictors we need between 20 and 25 participants per oximeter brand. We anticipate between 10 and 20 oximeter brands and this leads to a requirement for between 200 and 400 participants.

Recruitment status
Date of first participant enrolment
Date of last participant enrolment
Date of last data collection
Sample size
Accrual to date
Recruitment in Australia
Recruitment state(s)
Recruitment outside Australia
Country [1] 6429 0
New Zealand
State/province [1] 6429 0
Wellington & Christchurch

Funding & Sponsors
Funding source category [1] 290135 0
Name [1] 290135 0
Medical Research Institute of New Zealand
Address [1] 290135 0
Wellington Regional Hospital,
Riddiford Street,
Wellington 6021
Country [1] 290135 0
New Zealand
Primary sponsor type
Dr Janine Pilcher
Wellington Regional Hospital,
Riddiford Street,
Wellington 6021
New Zealand
Secondary sponsor category [1] 288877 0
Name [1] 288877 0
Address [1] 288877 0
Country [1] 288877 0
Other collaborator category [1] 278220 0
Name [1] 278220 0
Dr Leonie Eastlake
Address [1] 278220 0
Wellington Regional Hospital,
Riddiford Street,
Wellington 6021
Country [1] 278220 0
New Zealand

Ethics approval
Ethics application status
Ethics committee name [1] 291840 0
Health and Disability Ethics Committee (Northern B)
Ethics committee address [1] 291840 0
Ministry of Health
Level 6, Deloitte House
10 Brandon Street
PO Box 5013
Ethics committee country [1] 291840 0
New Zealand
Date submitted for ethics approval [1] 291840 0
Approval date [1] 291840 0
Ethics approval number [1] 291840 0

Brief summary
Pulse oximetry is commonly used in the clinical setting to titrate oxygen therapy to a target oxygen saturation range in order to avoid the risks of hypoxaemia and hyperoxaemia.
The purpose of this study is to investigate the level of agreement between pulse oximeter measured oxygen saturations and arterial blood gas measured oxygen saturations (the gold standard for oxygen saturation measurement) in Australian and New Zealand hospitals.
Trial website
Trial related presentations / publications
Public notes

Principal investigator
Name 52234 0
Dr Janine Pilcher
Address 52234 0
MRINZ, Wellington Regional Hospital, Riddiford Street, Newtown, Wellington 6021
Country 52234 0
New Zealand
Phone 52234 0
+64 4 8050241
Fax 52234 0
+64 4 3895707
Email 52234 0
Contact person for public queries
Name 52235 0
Dr Janine Pilcher
Address 52235 0
MRINZ, Wellington Regional Hospital, Riddiford Street, Newtown, Wellington 6021
Country 52235 0
New Zealand
Phone 52235 0
+64 4 8050241
Fax 52235 0
+64 4 3895707
Email 52235 0
Contact person for scientific queries
Name 52236 0
Dr Janine Pilcher
Address 52236 0
MRINZ, Wellington Regional Hospital, Riddiford Street, Newtown, Wellington 6021
Country 52236 0
New Zealand
Phone 52236 0
+64 4 8050241
Fax 52236 0
+64 4 3895707
Email 52236 0

No information has been provided regarding IPD availability
Summary results
Have study results been published in a peer-reviewed journal?
Other publications
Have study results been made publicly available in another format?
Results – basic reporting
Results – plain English summary