An audit of oxygen therapy on the medical ward in 2 different hospitals in Central Saudi Arabia

OBJECTIVE: To study the use of oxygen therapy on the medical wards in 2 hospitals in Riyadh, Kingdom of Saudi Arabia. One was academic, King Khalid University Hospital and the other a community hospital, Riyadh Medical Complex.

METHODS: This study was carried out over a one year period, 6 April 2000 through to 6 April 2001. Oxygen saturation was measured randomly by pulse oximetry in patients receiving oxygen therapy. Charts of the patients were inspected for relevant information related to oxygen therapy, including indications, dose, monitoring and documentation of the order. If oxygen saturation was >97% oxygen flow was reduced to maintain oxygen saturation between 92% and 94%. The potential savings by such reduction were calculated.

RESULTS: A total of 108 patients were studied. The most frequent indications for oxygen therapy were hypoxemia and dyspnea. Arterial oxygen tension before starting oxygen was carried out for 78 patients (72.2%) and showed that the majority (60 patients, 76.9%) were hypoxemic arterial oxygen tension <= 65 mm Hg). The last documented arterial oxygen tension values after initiating oxygen were found in 79 patients (73.1%) and these were carried out at mean interval of 111 hours (range one-1200) before our assessment. Most patients (32 patients, 40.5%) had excessive values (>85 mmHg), 24 patients (30.4%) were hypoxemic (arterial oxygen tension <=65 mmHg) and only 23 patients (29.1%) had acceptable values (arterial oxygen tension > 65-85 mmHg). Our measurements also showed that arterial oxygen tension was excessive (>97%) in 59 patients (54.6%), adequate (>=92%-97%) in 44 patients (40.7%), and only a minority (5 patients, 4.6%) were hypoxemic (<92%). Oxygen dose could be reduced in 31 patients (28.7%) by a mean of 42% (range 18%-66%) and stopped in 38 patients (35%) while maintaining aterial oxygen tension between 92%-94%. Errors in oxygen prescription were more apparent in the non-academic setting (P<0.05%).

CONCLUSION: Oxygen prescription was sub-optimal in both the academic and non-academic setting. The study highlights the need to adopt and evaluate cost-effective measures such as oxygen titration protocols using pulse oximetry, and physician education programs.