Incidence and risk factors of hyperoxemia among potentially critically ill patients treated in hospital wards: a retrospective study.

BACKGROUND/AIM: The literature shows a link between hyperoxemia and poor outcomes, whereas this association remains unclear in hospital wards. This study aims to determine the incidence of hyperoxemia in hospital wards and its risk factors.

MATERIALS AND

METHODS

Patients aged ≥ 18 years who underwent an evaluation by an intensivist between 1 January 2020 and 31 December 2020, while receiving treatment in hospital wards, were included in the study following ethics committee approval. Patients with hypoxemia (partial pressure of oxygen [PaO2] < 60 mmHg), a condition related to Coronavirus disease 2019, a hospital stay < 1 day, or missing data were excluded. Patients were divided into two groups: normoxemia (60 mmHg ≤ PaO2 < 120 mmHg) and hyperoxemia.

RESULTS

The incidence of hyperoxemia was 42.2%. Patients with hyperoxemia had longer hospital stays and higher intensive care unit admission rates than those with normoxemia. Partial pressure of carbon dioxide < 30 mmHg (OR, 1.61; 95% CI, 1.16-2.25; p = 0.005), hemoglobin ≤ 10.3 g/dL (OR, 1.33; 95% CI, 1.01-1.75; p = 0.044), positive pressure ventilation (OR, 1.73; 95% CI, 1.09-2.74; p = 0.021), fraction of inspired oxygen ≥ 50% (OR, 1.71; 95% CI, 1.10-2.65; p = 0.018), type IV respiratory failure (OR, 1.62; 95% CI, 1.05-2.51; p = 0.030), and receiving treatment on surgical units (OR, 1.47; 95% CI, 1.02-2.12; p = 0.038) were independently associated with hyperoxemia. Charlson comorbidity index > 6 (OR, 0.64; 95% CI, 0.49-0.85; p = 0.002), and chronic obstructive pulmonary disease exacerbation (OR, 0.51; 95% CI, 0.29-0.89; p = 0.017) were independently associated with normoxemia.

CONCLUSION

Monitoring potentially critically ill patients receiving oxygen therapy in wards is essential to mitigate hyperoxemia and optimize the use of healthcare resources. Further research could focus on developing strategies to accomplish this objective.