To explore the effects of fluid therapy with the synthetic colloids hydroxyethyl starch (HES) and gelatin (GEL) on the incidence of acute kidney injury (AKI) and need for renal replacement therapy (RRT) in patients undergoing cardiac surgery.

Secondary analysis of a prospective observational study in cardiac surgical patients.

University hospital.

The study included 584 elective patients (excluding patients on preoperative dialysis).

Anamnestic and surgical core data, hemodynamics, and hemodynamic treatments were recorded intraoperatively and postoperatively. Postoperative kidney dysfunction was graded according to the Acute Kidney Injury Network criteria from perioperative changes in plasma creatinine and urine flow. Statistical analyses were performed descriptively, by logistic and probit regression, omitting inotropic and vasoactive medications as established renal risk factors. The incidence of AKI and new renal replacement therapy was 28.6% and 7.5%, respectively. Patients with AKI were older, had a higher additive Euroscore, lower preoperative glomerular filtration rates and hemoglobin level, and presented with a longer duration of cardiopulmonary bypass and surgery and higher postoperative drainage loss. HES (1 [0-2] units of 500 mL) and GEL (3 [2-5] units of 500 mL) were used in 317 and 563 patients, respectively. Crystalloids were used in all patients (4,560 [4,080-5,042] mL). Patients presenting with AKI or new RRT were treated with significantly higher amounts of GEL. The use of HES and crystalloids did not differ between these groups. Probit regression showed significant dose-response relationships between the amount of infused gelatin and the probability of AKI and new RRT. Probit regression showed significant (p = 0.0001 and 0.0003, respectively) dose-response relationships between the total units of gelatin polysuccinate infused and the probability of AKI and new RRT (Fig 1). Logistic regression revealed a statistically significant odds ratio (OR) of 1.9741 (95% CI: 1.3104-2.9740; p = 0.0011) for an association between the number of gelatin units infused and AKI (grade 1-3) but no direct association between the number of gelatin units administered and new RRT. No association between a decrease in kidney function and the application of HES was observed.

Taking into account the limitations of the small sample size and a low event rate, the nonconsideration of established renal risk factors such as inotropes and vasopressors, and potentially unmeasured confounders, these findings suggested that gelatin solutions may have deleterious effects on renal function in cardiac surgical patients. The adverse clinical effects of HES on kidney function observed in other studies may have been blunted by the restrictive use of this synthetic colloid.