Does preoperative hyponatremia potentiate the effects of left ventricular dysfunction on mortality after cardiac surgery?

Objective: Left ventricular dysfunction and preoperative hyponatremia are associated with adverse outcomes after cardiac surgery. However, the interactions between them are unknown. Thus, we evaluated the interaction of low left ventricular ejection fraction (<40%) and preoperative hyponatremia (Na <135 mEq/L) with morbidity and mortality after cardiac surgery. Methods: The interaction of hyponatremia and ejection fraction with hospital complications, length of stay, and mortality was analyzed using logistic and Cox regression analysis in 2247 patients who underwent cardiac surgery between 2005 and 2008 at The Ohio State University Wexner Medical Center. Results: Of the patients, 68.5% had normal ejection fraction. Hyponatremia was present in 18% of patients with normal ejection fraction and 35% of patients with low ejection fraction. Hyponatremic patients had higher rates of New York Heart Association class III and IV, more comorbidities, and higher Society of Thoracic Surgeons score and European System for Cardiac Operative Risk Evaluation irrespectively of their ejection fraction. The correlation between preoperative sodium and ejection fraction was weak (r(2) = 0.04). Hyponatremia increased the rate of postoperative complications and hospital stay, and decreased 1- and 3-year survivals in patients with both normal and low ejection fraction. Hyponatremia was independently associated with longer hospital stay for normal ejection fraction (multiplier, 1.18; confidence interval, 1.09-1.27; P < .001) and low ejection fraction (multiplier, 1.10; confidence interval, 1.0-1.21; P = .05), increased need for dialysis for normal ejection fraction (odds ratio, 2.16; confidence interval, 1.08-4.32; P = .03), and increased risk of mortality for normal ejection fraction (hazard ratio, 1.56; confidence interval, 1.20-2.05; P = .001), but not for patients with low ejection fraction (hazard ratio, 1.21; confidence interval, 0.89-1.65; P = .21). Conclusions: Hyponatremia is more common in patients with low ejection fraction. Although preoperative hyponatremia is independently associated with adverse outcomes in patients with normal ejection fraction, an association with adverse outcomes in patients with low ejection fraction was not demonstrated.