Relationship between external jugular and central venous oxygen saturation in anesthetized dogs under varying hemodynamic conditions.

OBJECTIVE

To assess the correlation, agreement, and trending ability between external jugular venous oxygen saturation (SejvO2) and central venous oxygen saturation (ScvO2) across varying hemodynamic states in anesthetized dogs.

METHODS

This prospective, nonrandomized, repeated-measures study included 6 healthy dogs. Each dog was anesthetized, paralyzed, and mechanically ventilated. Six sequential hemodynamic states were induced: baseline, controlled hemorrhage, isotonic crystalloid resuscitation, autotransfusion, continuous rate infusion of dobutamine, and administration of high-concentration isoflurane. Simultaneous blood samples were collected from a central venous catheter and an external jugular venous catheter. Paired SejvO2 and ScvO2 values were analyzed for correlation, agreement, and trending ability.

RESULTS

36 paired samples were obtained. External jugular venous oxygen saturation and ScvO2 were strongly correlated (r = 0.71; 95% CI, 0.50 to 0.84). Bland-Altman analysis showed a mean bias of -4.47% (95% CI, -7.34% to -1.60%) with limits of agreement from -14.25% (95% CI, -17.70% to -10.81%) to 5.31% (95% CI, 1.87% to 8.76%), exceeding the predefined ± 8% threshold. The concordance rate was 66.7% (95% CI, 47.2% to 82.7%) and improved to 87.5% (95% CI, 61.7% to 98.4%) when applying an exclusion zone of ± 5%, indicating moderate trending ability.

CONCLUSIONS

SejvO2 correlated strongly with ScvO2 but lacked sufficient agreement for direct substitution. However, it demonstrated moderate ability to track directional changes under different hemodynamic conditions.

CLINICAL RELEVANCE: SejvO2 may serve as a less invasive option for serial monitoring of oxygen balance when direct ScvO2 measurement is not feasible, supporting its potential role in critical care monitoring.