During the outbreak of coronavirus disease 2019 (COVID-19), many studies have investigated laboratory biomarkers in management and prognostication of COVID-19 patients, however to date, few have investigated arterial blood gas, acid–base and blood pressure patterns. The aim of the study is to assess the arterial blood gas and acid–base patterns, blood pressure findings and their association with the outcomes of COVID-19 patients admitted to an intensive care unit.

A single-centre retrospective, observational study in a dedicated COVID-19 intensive care unit in Cape Town, South Africa. Admission arterial blood gas, serum electrolytes, renal function and blood pressure readings performed on COVID-19 patients admitted between 26 March and 2 June 2020 were analysed and compared between survivors and non-survivors.

A total of 56 intensive care unit patients had admission arterial blood gas performed at the time of intensive care unit admission. An alkalaemia (pH > 7.45) was observed in 36 (64.3%) patients. A higher arterial pH (median 7.48 [interquartile range: 7.45–7.51] versus 7.46 [interquartile range: 7.40–7.48], P = 0.049) and partial pressure of oxygen in arterial blood (median 7.9 kPa [interquartile range: 7.3–9.6] versus 6.5 kPa [interquartile range: 5.2–7.3], P = <0.001) were significantly associated with survival. Survivors also tended to have a higher systolic blood pressure (median: 144 mmHg [interquartile range: 134–152] versus 139 mmHg [interquartile range: 125–142], P = 0.078) and higher arterial HCO₃ (median: 28.0 mmol/L [interquartile range: 25.7–28.8] versus 26.3 mmol/L [interquartile range: 24.3–27.9], P = 0.059).

The majority of the study population admitted to intensive care unit had an alkalaemia on arterial blood gas. A higher pH and lower partial pressure of oxygen in arterial blood on arterial blood gas analysis were significantly associated with survival.