In the emergency department and intensive care unit, blood gas analysis is a crucial tool in the assessment of critically ill patients. Blood gas analysis is quick and repeatable at the bedside. The sampling can be done from both arterial and venous samples, with arterial samples generally considered to be more reliable and by that widely used as the standard method.

The purpose of this project is to compare venous and arterial blood gas parameters in undifferentiated, critically ill patients. We plan to evaluate the correlation between different parameters through a prospective observational study. In particular, carbon dioxide partial pressure (pCO2) correlation between venous and arterial samples is investigated by using and comparing different conversion models proposed in the literature.

250 patients deemed to be in need of arterial blood gas sampling based on their clinical condition will be included in a consecutive fashion at all hours.

The long-term goal is to clinically translate the findings into a limitation on the use of arterial sampling, which could potentially reduce pain and complication risks in the many patients who undergo arterial blood gas sampling every day.

== Practical procedures & flowchart ==

Patients included in this study are treated according to clinical routine. After inclusion, sampling of an arterial blood gas and a venous blood gas is performed simultaneously or as close in time as possible. Because the venous blood gas analysis sample can be taken directly from the patient's peripheral venous catheter, no additional venous puncture is usually required and the additional amount of blood that may be required is approx. 1 ml (the assay requires about 500 µl). In the very rare cases where it is not possible to establish a peripheral venous catheter within a reasonable time, the sample will be acquired from a new venous puncture together with other venous routine samples.

Sampling can be carried out by doctors or nurses. The arterial and venous blood gas analysis is performed with a blood gas analyzer (ABL90) in the emergency department and is performed immediately after the sampling. Sampling times and analysis times are documented.

Survey results are routinely stored in electronic form in patient health record and also locally on the device ABL90 with social security number as ID. The data is saved for statistical analysis.

== Documentation ==

The following parameters are documented during sampling:

• Chief complaint
• Indication for sampling
• The patient's current vital signs at the time of sampling
• Ongoing oxygen treatment (flow rate)
• Size of cannula, sampling location
• Sampling times and analysis times
• Number of puncture attempts

After sampling, the following is documented:

• Analysis results of venous and arterial blood gas sample
• Final diagnosis at discharge from the emergency department

The documentation will primarily be on worksheets, which will be marked with a patient label (name / social security number) and a serial number. The worksheet will be stored together with consent forms in a locked space of Akutkliniken US Linköping.

The data will be archived in the local emergency department or at the Regional Archives in Östergötland according to applicable laws and regulations.

The result of blood gas analysis is obtained from the digital lab system in the patient record (Lab-ROS) or directly exported from the internal memory of the analyzers at the emergency department. In both cases, the results will be used for digital compilation of study data. All electronic storage of study data will be pseudonymized by linking to patients serial number.

== Statistics ==

In previous studies, the correlation between arterial and venous samples has been evaluated using several different statistical methods:

• T-test, Pearson's and Spearman's R variable
• Descriptive data are presented as mean and +/- from SD

Our intention in this study is to make a parallel comparison of previously published methods, as well as to examine, based on collected data, whether a new formula with better precision can be derived.

== The potential significance of the project ==

At present, arterial blood gas sampling is used as the standard procedure for assessing a critically ill patient in the emergency department.

The project aims to clarify the feasibility of replacing routine arterial blood gas sampling in the emergency department with venous punctures without impairing test results and patient care. If feasible, such a change in sampling routines could potentially mean a great improvement for patients in terms of pain and complication risks. Also, workflow and lead times in the emergency department could be improved by the fact that venous samples may be drawn by all clinical staff, whereas arterial samples are normally reserved for doctors and nurses with dedicated training.

Inclusion Criteria:

• Patients able to understand and process oral and written information and give his/her consent (within 12h), or consent provided by a close relative.
• Aged > 18 years
• Assessed to be in need of arterial blood gas analysis

Exclusion Criteria:

• Unwilling to participate, or unable to understand the provided information
• Inability to give subsequently consent (> 12 hours) or unwillingness by close relative to provide consent.
• Pregnant
• Aged < 18 years

For enrolment, we will consider all patients who are admitted to the emergency department irrespective of mode of arrival (ambulance, walk-in etc). To be eligible, patients will need to be classified as critical ill and there has to be a clear clinical indication for an arterial blood gas sampling. Indications for arterial blood gas sampling, in turn, are primarily dictated by local SOPs (Standard Operating Procedures; "PM/ vårdprocessåtgärder") for appropriate sampling based on chief complaint, age, and vital signs upon arrival. The aim is to include these patients consecutively around-the-clock.