Effects of changes in inspired oxygen fraction on urinary oxygen tension measurements

Continuous measurement of urinary PO₂ (PuO₂) is being applied to indirectly monitor renal medullary PO₂. However, when applied to critically ill patients with shock, its measurement may be affected by changes in FiO₂ and PaO₂ and potential associated O₂ diffusion between urine and ureteric or bladder tissue. We aimed to investigate PuO₂ measurements in septic shock patients with a fiberoptic luminescence optode inserted into the urinary catheter lumen in relation to episodes of FiO₂ change. We also evaluated medullary and urinary oxygen tension values in Merino ewes at two different FiO₂ levels.

In 10 human patients, there were 32 FiO₂ decreases and 31 increases in FiO₂. Median pre-decrease FiO₂ was 0.36 [0.30, 0.39] and median post-decrease FiO₂ was 0.30 [0.23, 0.30], p = 0.006. PaO₂ levels decreased from 83 mmHg [77, 94] to 72 [62, 80] mmHg, p = 0.009. However, PuO₂ was 23.2 mmHg [20.5, 29.0] before and 24.2 mmHg [20.6, 26.3] after the intervention (p = 0.56). The median pre-increase FiO₂ was 0.30 [0.21, 0.30] and median post-increase FiO₂ was 0.35 [0.30, 0.40], p = 0.008. PaO₂ levels increased from 64 mmHg [58, 72 mmHg] to 71 mmHg [70, 100], p = 0.04. However, PuO₂ was 25.0 mmHg [IQR: 20.7, 26.8] before and 24.3 mmHg [IQR: 20.7, 26.3] after the intervention (p = 0.65). A mixed linear regression model showed a weak correlation between the variation in PaO₂ and the variation in PuO₂ values. In 9 Merino ewes, when comparing oxygen tension levels between FiO₂ of 0.21 and 0.40, medullary values did not differ (25.1 ± 13.4 mmHg vs. 27.9 ± 15.4 mmHg, respectively, p = 0.6766) and this was similar to urinary oxygen values (27.1 ± 6.17 mmHg vs. 29.7 ± 4.41 mmHg, respectively, p = 0.3192).

Changes in FiO₂ and PaO₂ within the context of usual care did not affect PuO₂. Our findings were supported by experimental data and suggest that PuO₂ can be used as biomarker of medullary oxygenation irrespective of FiO₂.