Blood gas sampling in the intensive care unit: A prospective before-and-after interventional study on the effect of...

Blood gas analysis is the most commonly ordered test in the intensive care unit. Each investigation, however, comes with risks and costs to the patient and healthcare system. Evidence suggests that many tests are performed with no appropriate clinical indication.

The primary aim of our prospective interventional study was to investigate the proportion of blood gases undertaken with a valid clinical indication before and after an educational intervention. A secondary aim was to examine sleep interruption secondary to blood gas sampling.

A prospective, before-and-after interventional study was conducted across two metropolitan intensive care units in Melbourne, Australia. Adults aged ≥18 years who were admitted to intensive care were eligible for inclusion. Two observation periods were conducted across a 2-week period in May and September 2022 (Periods 1 and 2), where clinicians were encouraged to record the purpose of blood gas sampling and other relevant data via an electronic questionnaire. These data were reviewed with corresponding electronic medical records. In between these periods, an interventional educational program to inform the clinical rationale for blood gas testing was delivered during July and August 2022, including introduction of a clinical guideline.

There were 68 patients with 688 tests included in Period 1 compared to 69 patients with 756 tests in Period 2. There was no significant difference between the median number of blood gas analyses performed per patient before and after the educational intervention (6.0 tests per patient vs 5.0 tests per patient, p = 0.609). However, there was a significant increase in the percentage of tests with a valid clinical indication (49.0% vs 59.7%, p = 0.0025). The most common indications selected were routine measurement, monitoring a clinical value, change in ventilator settings/oxygen therapy, and clinical deterioration. In addition, there were a large number of patients who were awakened upon drawing of a blood sample for analysis (26.1% for Period 1 and 37.6% for Period 2, p = 0.06).

The implementation of an educational program resulted in a significant increase in the proportion of blood gases performed with an appropriate clinical indication. There was, however, no reduction in the overall number of blood gases performed.