Although low-dose dopamine is widely used in intensive care units as a renal protector in patients at risk for renal impairment, no evidence exists that it decreases the incidence of renal failure or mortality [7-9]. In dogs and humans, dopamine is ineffective as a renal protective agent [10, 11]. In liver transplant recipients, the evidence is conflicting. Early work by Polson and colleagues [9] claimed improved renal function with dopamine, but a more recent study showed no benefit [12].

In view of the uncertainty about the efficacy of routine low-dose dopamine as a renal protective agent, we did a study to assess the effect of low-dose dopamine on renal function in euvolemic patients after major elective vascular surgery.

Methods

Top Methods Results Discussion Author & Article Info References

After receiving approval from the ethics committee of Princess Alexandra Hospital, informed consent was obtained for this study from 37 consecutive patients presenting to the vascular surgical unit (during a 7-month period) for elective infrarenal abdominal aortic aneurysm repair or aortobifemoral grafting. The vascular surgical service is 1 of 3 major public vascular units serving about 1 million persons in the Brisbane area.

On the day before surgery, plasma urea and creatinine levels were assayed. Creatinine clearance was measured using an accurately timed 2-hour collection [13]. The anesthetic technique was similar for all patients and included an opioid-local anesthetic epidural for postoperative analgesia.

The hospital pharmacy randomly assigned dopamine and an equal volume of saline to identical containers using a random numbers table. The containers were identified only by a number (numbers 1 to 40), and the contents were unknown to the staff and the patients of the intensive care unit. All patients were admitted to the intensive care unit postoperatively and were started on an infusion of either dopamine (3 µg/kg per minute) for 24 hours or placebo (saline), which was chosen in a double-blind manner by selecting the next number in the sequence 1 to 40. The management in the intensive care unit was otherwise usual for both groups; it included an infusion of Plasmalyte 56 (Baxter Health Care; Old Toongabbie, New South Wales, Australia), 3000 mL during 24 hours, and included bolus infusions of Plasmalyte 148 (Baxter Health Care) to maintain a urine flow of 1 to 1.5 mL/kg per hour. Mannitol and diuretic agents were not used. Postoperative fluid requirement and urine output during the first 24 hours in the intensive care unit were recorded. Plasma creatinine levels, serum urea levels, and creatinine clearance were measured at 24 hours and 5 days. Urine output and fluid requirements were indexed to body weight, whereas clearance values were indexed to the body surface area. All biochemical measurements were made using a SMAC II Autoanalyzer (Technicon Equipment, Sydney, Australia). Statistical analysis was done using SPS-PC software (SPS Australasia, Crows Nest, New South Wales, Australia), with the data presented as means and 95% confidence intervals (CIs).

Results

Top Methods Results Discussion Author & Article Info References

Eighteen patients received dopamine, and 19 patients received placebo. One patient from the placebo group and 3 from the dopamine group had myocardial infarctions. Two patients died during the study and both were in the dopamine group (1 patient, myocardial infarction; 1 patient, renal failure). These patients were included in the 24-hour analysis but were excluded from the 5-day analysis. The groups were similar with respect to age, preoperative morbidity, diuretic use, radionucleotide ejection fraction, urea creatinine levels, type of surgery, duration of aortic cross-clamping, and blood loss (Tables 1 and 2). The mean fluid requirements in the first 24 hours were slightly less in the dopamine group than in the placebo group (4.06 mL/kg per hour for dopamine compared with 5.29 mL/kg per hour for placebo, a difference of 1.23 [95% CI, –0.11 to 2.57]), whereas the mean urine volumes were slightly higher in the dopamine group (1.83 mL/kg per hour for dopamine compared with 1.6 mL/kg per hour for placebo, a difference of 0.23 [CI, –0.18 to 0.64]).

Table 2 summarizes the differences between the groups in plasma creatinine levels, urea levels, and crude and indexed creatinine clearance; all variables were measured preoperatively and postoperatively (at 24 hours and 5 days). Table 3 summarizes the mean change in each variable from the preoperative level to the 24-hour and 5-day point in the dopamine and placebo groups. There was no statistical difference in plasma creatinine levels between groups at any of the time periods or in changes from baseline to 24 hours and 5 days.

Preoperative and 5-day plasma urea levels were not statistically different between groups. At 24 hours, a small decrease in urea levels was noted in both groups (5.71 mmol/L for dopamine compared with 4.99 mmol/L for placebo, a difference of 0.72 [CI, –0.62 to 2.06; P > 0.2]). The mean decrease in urea levels from baseline to 24 hours was small in both groups ( –1.07 mmol/L for dopamine compared with –1.84 mmol/L for placebo, a difference of 0.77 [CI, –0.12 to 1.67; P = 0.09]).

The preoperative creatinine clearance in the placebo group was 18% lower than in the dopamine group (P = 0.16) (Table 2). By 24 hours, the creatinine clearance increased by 0.165 mL/s (9.89 mL/min) in the dopamine group and by 0.199 mL/s (11.98 mL/min) in the placebo group (P > 0.2). By day 5, the creatinine clearance in the dopamine group had decreased to near baseline, although the increase was sustained in the placebo group. However, the difference between the group means was 0.28 mL/s ( –0.103 mL/s compared with 0.178 mL/s [CI, –0.2 to 0.21; P > 0.2]). Indexing creatinine clearance for body surface area did not change the interrelation of the major end points in either group.

Discussion

Top Methods Results Discussion Author & Article Info References

In this study, minor changes in renal function were noted postoperatively in the placebo and dopamine groups. In both groups, a small decrease was noted in plasma urea levels and a small increase was noted in creatinine clearance by 24 hours; these changes were not statistically or clinically significant. No difference was noted between the groups, suggesting that generous fluid replacement alone was as effective as fluid combined with low-dose dopamine. The difference from baseline was maintained at 5 days in the placebo group (but not in the dopamine group), but it also was not statistically significant. The study was not large enough to detect any decrease in the incidence of renal failure, which is an uncommon complication of elective abdominal aortic surgery.

Although the differences in urine production and fluid requirements were not statistically different, the trend suggested that the addition of dopamine might allow the production of more urine per unit of fluid infused. The mechanisms by which dopamine increases urine production are related to an increase in the glomerular filtration rate, which is largely attributable to increased cardiac output [14], and to a direct action on tubular sodium uptake mechanisms leading to natriuresis. Normal tubuloglomerular feedback (which regulates glomerular filtration rate in response to filtered sodium load) is blunted in the presence of dopamine, which further enhances its diuretic action [15]. This trend in favor of dopamine was of no clinical significance in the context of this study, although it does support the view that in euvolemic patients, dopamine has a mild diuretic effect. A similar diuretic effect might have deleterious consequences in hypovolemic patients by worsening the hypovolemia. It is therefore essential to exclude hypovolemia before resorting to dopamine in oliguric patients.

When considering the use of dopamine in this context, the cost and adverse effects (such as arrhythmias, depression of respiratory drive, increased intrapulmonary shunt, myocardial ischemia in patients with obstructive coronary artery disease, and increased pulmonary wedge pressure) need to be considered. Four patients in the study had postoperative myocardial infarction. The relatively high rate of myocardial infarction may reflect the inadequate cardiac resources available locally, which severely limit the number of patients who can be investigated and treated for active ischemic heart disease before aortic surgery. Although the numbers were too small for meaningful comparison, three of the patients with myocardial infarction were from the dopamine group, which raises some questions about the safety of dopamine in such patients.

Although the study was small and without the power to discriminate different rates of renal failure (an infrequent clinical end point), we did not find any compelling evidence to support the routine use of dopamine in patients receiving elective vascular surgery. Because patients with acute oliguric renal failure were not included in the study, our results cannot be extrapolated to the treatment of acute oliguric renal failure.