Postoperative
hyponatremia was present in 59%. Hyponatremic patients were older (mean ±
SD, 62 ± 13 vs 61 ± 14 years, p = 0.001), had lower left ventricle ejection fraction (mean ± SD, 44% ± 16% vs 48% ± 13%, p < 0.001), higher mean pulmonary artery pressures (mean ± SD, 30 ± 11 vs 27 ± 9 mmHg, p < 0.001), lower glomerular filtration rate (mean ± SD, 72 ± 29 vs 74 ± 27 mg/min/1.73 m(2), p = 0.01), higher EuroSCORE (median, 15% vs 6%, p < 0.001), higher New York Heart Association class IV (31% vs 26%, p = 0.002), prevalence of
COPD (23% vs 14%, p < 0.001), and
peripheral vascular disease (16% vs 12%, p < 0.001).
Hyponatremia increased overall (24% vs 18.2%, p < 0.001) and late mortality (18.6% vs 13.9%, p < 0.001) and
length of stay (LOS; 11 vs 7 days, p < 0.001). Mortality increased with the severity of the
hyponatremia. After adjusting for baseline and procedure variables, postoperative
hyponatremia was associated with increase in mortality (hazard ratio 1.22, 95% CI 1.06-1.4, p = 0.004), LOS (multiplier 1.34, 95% CI 1.22-1.49, p < 0.001), infectious (odds ratio [OR] 2.32, 95% CI 1.48-3.62, p < 0.001), pulmonary (OR 1.82, 95% CI 1.49-2.21, p < 0.001), and
renal failure complications (OR 2.46, 95% CI 1.58-3.81, p < 0.001) and need for dialysis (OR 3.66, 95% CI 1.72-7.79, p = 0.001).
CONCLUSIONS: