Introduction
Case Presentation
Acute Renal Failure
History of low-dose Dopamine
Effect of low-dose Dopamine in Healthy Subjects
Effect of low-dose Dopamine in Experimental Acute Renal Failure
Efficacy of low-dose Dopamine for the Prevention of ARF in High-risk Patients
Effect of low-dose Dopamine in Critically Ill Patients with Mild Renal Insufficiencies
Effect of low-dose Dopamine in Acute Renal Failure
Potential harmful effects of Dopamine
Conclusion and take-home points
References
Appendices I-III

Oliver Tu, MD
Resident Grand Rounds
Internal Medicine Residency
North Carolina Baptist Hospital

October 3, 2000

"Renal-dose" dopamine or low-dose dopamine (LDD) (< 3 m g/kg/min) is employed worldwide by its proponents in the treatment of acute renal failure (ARF). However, despite decades of application of low dose dopamine, there are few randomized controlled trials and long-term clinical outcome studies in acute renal failure. I seek to present some of the studies and evidences leading up to and surrounding the use of low dose dopamine in the treatment of ARF. In general, studies on the use of low dose dopamine can be categorized into:

1. effect of dopamine on normal human subjects,
2. animal studies with experimental acute renal failure,
3. prevention of acute renal failure with prophylactic low dose dopamine in high risk patients,
4. prevention of acute renal failure with prophylactic low dose dopamine in critically ill patients with mild renal insufficiency, and
5. effect of low dose dopamine in acute renal failure.

Among all the categories, there is the least amount of evidence in the use of low dose dopamine for the treatment of acute renal failure.

Mr. Moore is a 65 year old white male with history of hypercholesterolemia and Type 2 diabetes mellitus on insulin who initially presented to an outside hospital emergency room a few days ago with two days history of progressive shortness of breath, general malaise, subjective fever, chills, and cough productive of greenish sputum. At the outside hospital emergency room, he was found to be febrile with temperature of 101^oF, hypoxic on room air with oxygen saturation in the mid-80%, pulse of 106, and blood pressure of 105/73. Physical exam showed decreased breath sound over right lower lobe. Laboratory data was significant for elevated white count of 17,000 with 6% bands. Chemistry profile was mostly normal with BUN of 18 and creatinine of 1.3. Chest X-rays showed right lower lobe infiltrate. Mr. Moore was admitted to the outside hospital with diagnosis of right lower lobe pneumonia and started on levofloxacin and was placed on 3 liters of oxygen through nasal canula. In the next 48 hours, patient developed hypoxic respiratory failure, and was intubated. He also became hypotensive with systolic blood pressure in the 70s. He was given fluid boluses and started on dopamine drip. Patient was then transferred to medical intensive care unit at North Carolina Baptist Hospital.

On arrival, patient was sedated on an Ativan drip. Pt. was febrile with temperature of 102.6^oF with BP of 82/55 and HR of 102 on dopamine at 10 m g/kg/min. He was given more fluids and dopamine was titrated to 20 m g/kg/min. Norepinephrine was eventually started and blood pressure then became stable with SBP in the low 100s. His admission lab was significant for WBC of 30,000 with 8% bands with BUN of 30 and creatinine of 1.8. CXR showed bilateral lower lobe infiltrate. His antibiotics were changed to vancomycin, ceftriaxone and azithromycin. Over the next 24 hours, patient became afebrile with WBC coming down. Norepinephrine was weaned off and dopamine was weaned to 5 m g/kg/min with blood pressure remaining stable. However, BUN and creatinine increased to 37 over 2.5 and urine output dropped to 40mL/hr. The following day, BUN and creatinine increased to 45 over 2.9. Despite many fluid boluses, urine output decreased to 10mL/hr. Microscopic exam of the spun urine showed many granular casts. The clinical picture is consistent with acute renal failure secondary to acute tubular necrosis due to septic shock. The intern was told by the upper level resident to wean the dopamine to "renal-dose" at 3 m g/kg/min.

ARF is generally defined as an abrupt and sustained decline in renal glomerular filtration rate and retention of nitrogenous waste products such as blood urea nitrogen (BUN) and creatinine¹. ARF is usually diagnosed when routine laboratory test showed a recent increase in BUN and serum creatinine level. Oliguria is defined by urine output (UO) less than 400 mL per day and is a frequent ( ~ 50%) but not invariable clinical feature¹. The Acute Physiology and Chronic Health Evaluation (APACHE) III defines ARF as serum creatinine elevation > 1.5 mg/dL/day with UO < 410 mL/day without pre-existing chronic dialysis². ARF complicates nearly 5% of all hospital admissions and up to 30% of admission to intensive care units¹. The mortality rate of an isolated episode of ARF is about 10 to 15% and, when associated with multiple organ dysfunction, is between 50 – 90%³.

ARF is generally divided into three categories (Table 1¹): 1) pre-renal azotemia, characterized by renal hypoperfusion with preserved renal parenchymal tissue (~ 55-60%), 2) intrinsic renal azotemia, characterized by disease of the renal parenchymal tissue ( ~ 5-40%), and 3) post-renal azotemia, characterized by acute obstruction of the urinary tracts ( < 5%). Most acute intrinsic renal azotemia is caused by ischemia or nephrotoxins and is often associated with acute tubular necrosis¹.

It should be noted that BUN and creatinine are relatively insensitive indices of glomerular function and renal glomerular filtration rate (GFR)¹. GFR may fall by approximately 50% before serum creatinine rise (Figure 1¹). However, given our current clinical practices, measurement of BUN and creatinine will likely be our principle diagnostic tool for ARF.

Table 1. Classification and major disease categories causing acute renal failure

Figure 1. Relationship between serum creatinine and GFR¹.

The synthesis of dopamine (3, 4-dihydroxyphenylethylamine) was first reported in 1910 independently by Mannich and Jacobsohn⁴ and Barger and Ewins⁵. As one of the catecholamines, dopamine is derived from tyrosine (Figure 2⁶). Tyrosine first undergoes hydroxylations via tetrahydrobiopterin-dependent hydroxylase, forming dopa. Dopa then undergoes decarboxylation via dopa decarboxylase, forming dopamine. Dopamine can then be transformed into norepinephrine via a copper-containing monooxygenase, dopamine b –hydroxylase. Finally, epinephrine is synthesized from norepinephrine after methylation by S-adenosylmethionine.

Figure 2. Biosynthesis of the catecholamines: dopamine, norepinephrine, and epinephrine from tyrosine.

For the next 30 years, research on dopamine was limited to comparisons of the action of dopamine with other sympathomimetic amines⁷. In 1942, Holtz et al⁸. reported that dopamine could decrease blood pressure in guinea pigs and rabbits, raising the suspicion for the first time that dopamine acts on receptors distinguishable from the classical a and b adrenoceptors. In 1963, McDonald et al. found that intravenous dopamine in normal human subjects markedly reduced renal arterial resistance^9,10. The following year, McDonald et al. found that intravenous infusions of dopamine (2.6-7.1 m g/kg/min) to 7 normal subjects increased the average p-aminohippurate clearance (a measure of renal plasma flow) from 507 to 798 ml/min, inulin clearance (a measure of glomerular filtration rate) from 109 to 136 ml/min¹⁰. In 1970, Talley et al. published one of the first studies on the use of dopamine, alone or combined with diuretics, in the treatment of acute renal failure¹¹. In five patients with ARF due to dehydration, iodinated contrast materials, vigorous diuretic therapy, or major surgery, it was shown that the combination of dopamine and diuretic, but not either alone, were able to increase urine flow and reduce blood urea nitrogen or serum creatinine levels towards preoliguric levels.

In 1972, Dr. Leon Golberg of Emory University School of Medicine published the landmark review⁷, analyzing the many studies published by then, and found that dopamine induced vasodilation is not blocked by b -adrenergic blocking agents, atropine, and antihistamines, nor was attenuated by prior treatment with reserpine, monoamine oxidase inhibitors, or compound 48/80. With these evidences and others, he concluded that there must be a specific dopamine receptor. Later studies identified the presence of dopamine receptors DA-1 on renal, mesenteric, coronary, and cerebral arteries and dopamine DA-2 receptors on autonomic ganglia and sympathetic nerve endings with the help of specific agonists (fenoldopam and quinpriole) and antagonist (SCH 23390 and domperidone) for DA-1 and DA-2 receptors, respectively¹².

In healthy experimental animals or humans, dopamine infusion induces a dose-dependent increase in renal blood flow (RBF), natriuresis, and diuresis¹³(Table 2¹⁴).

Table 2. Some effects of dopamine on renal physiology

At low doses (~0.5 - ~3.0 m g/kg/min), dopamine augments RBF predominantly through intrarenal vasodilatations, mediated predominantly through the DA-1 receptors located on the renal vasculature¹². Engagement of DA-2 receptors on presynaptic sympathetic nerve terminals with inhibition of norepinephrine release and possibly post-synaptic DA-2 receptors many also contribute¹³. Dopamine induces natriuresis by binding to DA-1 and probably DA-2 receptors, causing inhibitions of basolateral Na, K-ATPase activity in proximal tubule, medullary thick ascending limb of the loop of Henle, and cortical collecting duct epithelial cells, thus attenuating tubular reabsorption of sodium¹³. Dopamine also probably inhibits the central antidiuretic hormone (ADH) release and antagonizing the actions of ADH on collecting duct cells, thus promoting diuresis¹³.

Several studies have evaluated the influence of LDD on renal function in laboratory animals with experimental ARF^13,15. Classic models included inducing ischemic ARF by occlusion of renal artery or aorta or inducing nephrotoxic ARF by uranyl nitrate or glycerol. Most studies were controlled and compared infusion of dopamine at 0.4 – 6 m g/kg/min to diluent, started prior to the induction of ARF. RBF and GFR are usually determined by clearance of para-aminohippurate (PAH) and inulin, respectively. In one study, Conger et al clamped bilateral renal artery in 12 rats for 55 minutes and infused dopamine at 0.4 – 1.5 m g/kg/min in 6 rats and normal saline in the other 6 for 4 hours starting post-clamp. GFR increased significantly more in rats treated with dopamine (0.496 vs. 0.351 mL/min)¹⁶.

In general, with ischemic renal insult, animals receiving LDD had lesser compromise of GFR than controls. Though the differences are usually statistically significant, degree of protection varies greatly among different studies. In animals with ARF from nephrotoxic agents, LDD usually afforded little or no protection¹³. Overall, no consistent positive or negative responses were demonstrated¹⁵; furthermore, there are insufficient data regarding the long-term effect on renal function or animal survival¹³. It is therefore difficult to conclude from animal studies as to the possible effect of dopamine on acute renal failure in human beings¹⁵.

These trials evaluated LDD in patients at high risk of developing ARF, usually after elective abdominal aortic surgery, coronary artery bypass graft surgery, liver transplantation, renal transplantation, or coronary angiogram. Appendix I presented the highlights of some of the major studies. Most studies were controlled prospective trials comparing the effect of dopamine at 2-3 m g/kg/min to saline infusions. Many were randomized-blinded controlled studies. However, all studies suffer from small number of patients with different protocols and patient populations, thus precluding meta-analysis. There were few cases of acute renal failure in either groups in many of these studies, making it very difficult to detect a benefit of low dose dopamine as a prophylaxis.

In a prospective randomized double-blinded study, Baldwin et al. randomized 37 consecutive patients to receive dopamine at 3 m g/kg/min (18 patients) or saline placebo (19 patients) after elective infra-renal abdominal aortic aneurysm repair or aortobifemoral grafting¹⁷. Patients with acute renal failure were excluded from the study. The two groups were statistically similar in terms of age, preoperative morbidity, prior diuretic use, ejection fraction, type of surgery, duration of aortic cross-clamping, blood loss, preoperative creatinine, and preoperative 2 hour creatinine clearance. The infusions were continued for 24 hours. Crystalloids were given to maintain urine output at 1-1.5 mL/kg/hr. Mannitol or diuretics were not used. Urine output was measured for 24 hours and creatinine clearance measured at 24 hours and at day 5 post-surgery. One patient from the placebo group and three from the dopamine group had myocardial infarction. Two patients died during the study (one MI, one renal failure) and both were in the dopamine group. Intention to treat analysis were performed for day 1 result but the two patients who died were excluded from the 5 day analysis, presumably since the parameters could not be measured. At 24 hours, urine output is slightly more in the dopamine group at 1.83 mL/kg/hr compared to the placebo group at 1.6 mL/kg/min. However, the difference of 0.23 was not statistically significant with 95% CI of –0.18 to 0.64. In terms of creatinine clearance, there was no statistical difference at 24 hours or at day 5 (Figure 3 and 4). However, there was no acute renal failure in the control groups, making it difficult to detect a benefit. Thus, within the limitations of this short-term small study, it showed that LDD offered no benefit in terms of renal function (CrCl and UO) in well hydrated patients after elective abdominal aortic surgery. In addition, there are more morbidity and mortality in the dopamine group but statistical significance was not calculated.

Figure 3. Creatinine clearance according to dopamine use

Figure 4. Difference in creatinine clearance between the two groups

Three major studies evaluated the efficacy of low dose dopamine in preventing acute renal failure after major cardiac surgery. In a prospective randomized double blinded study by Myles et al., 52 patients undergoing elective CABG without previous renal failure were randomized into two groups to either receive dopamine at 200 m g/min (N = 25) or 5% dextrose (N = 24) following induction of general anaesthesia¹⁸. Three patients (6%) withdrew from the study. The infusion were continued for 24 hours and patients were given intravenous furosemide according to a protocol if patient becomes oliguric despite achieving a wedge pressure of greater than 12 with colloid infusion. Creatinine clearance of the control and dopamine group at the first four hour (0-4 hrs) of the study (127 vs. 104 mL/min, p = 0.27) and the last four hours (20-24 hrs) of the study (107 vs. 91.2 mL/min, p = 0.48) were no statistically different. Urine output in the control group was higher in the first 4 hours (1231 vs. 917 mL, p = 0.066) but did not reach statistical significance. At the end of the 24 hour study, there was no different in total urine output (3304 vs. 3659 mL, p = 0.36). Three patients in each group were given a single dose of furosemide during the study (P = 0.65). Ten patients in the control group and nine in the dopamine group were placed on either epinephrine or norepinephrine during the study. No patient developed acute renal failure during the hospitalization. Nevertheless, this study showed that 24 hours of prophylactic low dose dopamine in patients with normal renal function undergoing CABG does not increase creatinine clearance or urine output.

In a controlled randomized unblinded study, Tang et al (1999) randomized 40 patients without pre-existing renal disease undergoing elective CABG into two equal groups to receive either dopamine at 2.5 – 4 m g/kg/min or no infusion for 48 hours starting from induction of general anaesthesia¹⁹. Central venous pressure was maintained between 8 and 12 mmHg. No significant differences were detected between the two groups in serum creatinine (Figure 5) and weight adjusted urine output (Figure 6) in any day of the first post-operative week. Thus, similar to the previous study, this study showed that low dose dopamine for 48 hours for patients with normal renal function undergoing elective CABG makes no impact in renal function. However, creatinine clearance was not obtained in this study and there was no acute renal failure in either group.

Figure 5. Serum creatinine (m mol/L) of control and low dose dopamine (LDD) group during the first postoperative week. Day 0-7: D0-7. P < 0.05 between control and LDD group.

Figure 6. Urine output (mL/kg) per day of control and low dose dopamine group (LDD) during the first postoperative week. Day 0-7: D0-7. P < 0.05 between control and LDD group.

In another controlled randomized double-blinded study, Lassnigg et al. randomized 132 patients with normal renal function undergoing elective cardiac surgery to receive saline placebo (N = 40), dopamine at 200 m g/kg/min (N = 42), or furosemide 0.5 m g/kg/min (N = 41) after induction of anesthesia²⁰. Only data from the placebo and dopamine group are presented here. The infusions were continued until discharge from intensive care unit, 48 hours after surgery, or when urine output was greater than 4 liters in 4 hours. Furosemide 20 mg injections were needed in 16 patients in the dopamine group and 20 patients in the placebo group for urine output < 0.5 mL/kg/hr. There was one death (severe sepsis) and no acute renal injury (creatinine increase of > 0.5 mg/dL between baseline and maximum value within 48 hrs) in the placebo group and no death but one acute renal injury in the dopamine group. None in either group required dialysis. Creatinine clearance and hourly urine output was reportedly similar between the dopamine and placebo group at all periods, though no specific p values were given (Figure 7, 8).

Figure 7. Creatinine clearance (mL/min) of the control and low-dose dopamine (LDD) group during the peri-operative study periods.

Figure 8. Urine output (mL/hr) of the control and low dose dopamine group during the peri-operative study periods.

A randomized controlled trial by Parks et al. showed no difference in creatinine clearance and urine output from day 1 to 5 with dopamine infusion at 3 m g/kg/min for 48 hours starting pre-op for patients undergoing elective surgery for obstructive jaundice²¹. There were no acute renal failure in either group. In randomized study of 48 patients undergoing liver transplant, Swygert et al. showed that low dose dopamine did not improve BUN and serum creatinine at day 7 and glomerular filtration rate at 1 month post-operation after 30 days of cyclosporine²². Incidence of acute renal failure is 4% in both groups.

There are a few controlled randomized studies in patients undergoing coronary angiogram. These studies mostly used creatinine as a measure of renal function, not creatinine clearance. In a study of 40 patients by Kapoor et al.²³, serum creatinine at 24 hours after the procedure did not change in patients who received dopamine at 5 m g/kg/min starting 30 minutes before and till 6 to 8 hours after the procedure while it increased in patients not receiving dopamine. (1.52 to 1.37 mg%, p > 0.05 vs. 1.52 to 1.96 mg %, p =0.01). There were no acute renal failure or change in urine output in either group. In another study, Abizaid et al.²⁴ randomized 40 high-risk patients with creatinine greater than or equal to 1.5 mg/dL to receive saline hydration or saline hydration plus dopamine at 2.5 m g/kg/min infused for 12 hours starting 2 hours before the procedure. Peak creatinine, change in creatinine, and percentage of patients with contrast-induced acute renal failure (30% in control vs. 50% in LDD, p = 0.6) are similar in both group, though there is a trend toward slightly more increase in creatinine in the LDD group (0.5 mg/dL in control vs. 0.6 mg/dL in LDD group, p = 0.06). Lastly, in a study of high-risk patients with chronic renal failure with baseline creatinine greater than 130 m mol/L but less than 200 m mol/L or with diabetes²⁵, 68 patients were randomized to receive dopamine at 2 m g/kg/min hours (N = 34) or saline (N = 34) for 48 hours. There were no statistical difference in baseline creatinine (100.6 vs. 100.3 m mol/L), peak creatinine (112.3 vs. 117.5 m mol/L), or change in creatinine (11.7 vs. 19.3 m mol/L) between the control and LDD group. One patient from each group withdrew from the study. Six patients developed acute renal failure, two in the control group (6%) and four (12%) in the LDD group. The investigators did not provide the statistical significance of this difference. Furthermore, there was one death in the control group, none in the LDD group. Together, these three studies do not provide convincing evidence to show that low dose dopamine decrease the risk of acute renal failure following coronary angiogram. On the contrary, low dose dopamine may even increase that risk in patients with diabetes or mild renal insufficiency, since a higher percentage of patients developed acute renal failure in the later two studies. However, the result did not reach statistical difference in the second study and the third study did not provide the p-value. Though the first study by Kapoor et al. showed that creatinine did not increase following angiogram in the dopamine treated group, its significance is unclear as there was no acute renal failure in either group.

In summary, these studies do not support the use low dose dopamine for the prevention of acute renal failure in high risk patients. In patients undergoing coronary angiogram with diabetes or mild renal insufficiency, there may be a trend towards more acute renal failure with the use of low dose dopamine. However, small number of patients and low incidence of acute renal failure in either group limits many studies. Small differences may not be detected in these studies due to insufficient power. If we assume 20% risk of acute renal failure in high risk patients, it would take about 400 patients to detect even a 10% reduction in risk in the low dose dopamine group with a two-tail error of 0.05 and 80% power¹³. Given the low percentage of acute renal failure in many of these studies, it would take even more patients in a study to detect the difference. The variety of patient population and protocols also make meta-analysis difficult. Larger randomized control trials are needed.

Most of these studies attempt to assess whether low dose dopamine improves renal function in critically ill patients with mild renal insufficiency in the intensive care unit. Five studies were presented in Appendix II. All of these studies are case-series with small number of patients and using patients themselves as their control. Since there are other aspects to these studies, only data related to low dose dopamine are presented here. There is also one retrospective study with more patients (N = 268).

In a case series by Duke et al., 23 critically ill patients were enrolled²⁶. Entry criteria include no indication of acute renal failure, creatinine clearance greater than 30 mL/min, serum creatinine less than 3.4 mg/dL, urine output > 0.5 mL/kg/hr. Patients with unstable cardiopulmonary status, oliguria, or diuretic use were excluded. Baseline serum creatinine was 0.99 + 0.51 mg/dL. Diagnosis of these patients include pneumonia, intra-abdominal sepsis, multiple trauma, pancreatitis, fat embolus syndrome, pulmonary thromboembolism, abdominal aortic aneurysm, ARDS, and multi-organ system failure. Eleven (61%) patients were on other pressors. Patients received 5 hour infusion of D5W placebo, dopamine 200 m g/min (mean of 2.9 m g/kg/min), and dobutamine 175 m g/kg/min in random sequence. Creatine clearance increased from 79 mL/min on placebo to 88 mL/min on dopamine but this was not statistically significant (p = 0.25). Urine output increased significantly from 90 to 145 mL/hr (p < 0.01). Five patients (22%) withdrew from the study, one of which had acute renal failure. ICU death were 28%.

Lherm et al. studied 29 nonoliguric patients with sepsis (N = 14) on no pressors or septic shock on pressors (N = 15)²⁷. All patients had clinical fluid retention. Dopamine at 2 m g/kg/min were infused for 2 hours. Mean arterial pressure (MAP) were maintained within 20% of baseline values by fluids or pressors. With dopamine infusion, creatinine clearance significantly increased from 76 to 120 mL/min (p < 0.05) in the sepsis group but remain statistically unchanged from 60 to 52 mL/min (NS) in the septic shock group. With low dose dopamine, urine volume also increased significantly from 113 to 216 mL/2 hr (p <0.05) in the sepsis group but remained unchanged from 201 to 184 mL/min (NS) in the septic shock group. As a second part of the study, the infusion of low dose dopamine was extended to 48 hours in the sepsis group. At 48 hours, 2 hour data on urine output and creatinine clearance were collected. The dopamine was then discontinued and urine output and creatinine clearance off dopamine (new base line on day 2) were measured for two hours (Figure 9). Though diuresis on dopamine at day 2 is still more than that off dopamine, it is significantly less (p = 0.004) than diuresis on dopamine on the first day, suggesting that the diuretic effect of dopamine may gradually decrease with time. Mortality rate was 50% in the sepsis group and 67% in the septic shock group.

Figure 9. Variation of diuresis and creatinine clearance

Olson et al. also studied 16 patients with sepsis with 31% of the patients on dobutamine²⁸. Baseline creatinine was 1.1 + 0.35 mg/dL. Dopamine at 3 m g/kg/min were infused for 2 hours. Patient were volume resuscitated (mostly to keep pulmonary artery wedge pressure (PAWP) > 10 mm Hg and cardiac output > 5 L/min) prior to dopamine infusion. Cimetidine drip was used to improve the accuracy of creatinine clearance estimation since it decreases tubular creatinine secretion. Creatinine clearance remained unchanged (80.5 vs. 83.1 mL/min, p > 0.05) but urine output increased (88.4 vs. 115.2 mL/hr, p = 0.02) with dopamine infusion. Mortality rate was 63%.

Ichai et. al recently presented two studies on critically ill patients in ICU^29,30. In the first study, 12 patients with mild renal insufficiency (CrCl 30 – 80 mL/min) randomly received four different doses of dopamine and dobutamine (0, 3, 7, 12 m g/kg/min) for four hours at each infusion. Only patients with mean arterial pressure > 70 mm Hg, cardiac index > 2.5 L/min/m², and pulmonary artery wedge pressure greater than 12 mm Hg are included in the study. Patients on other pressors were excluded. For dopamine at 3 m g/kg/min, creatinine clearance increased significantly from 61.1 to 80.2 mL/min (p < 0.05) and urine output increased from 48 to 79 mL/hr (p < 0.05). Mortality rate was 50%. In the second study, 8 critically ill patients with similar criteria as the previous study received placebo (D5W) for 4 hours, then dopamine at 3 m g/kg/min for 48 hours, and then another placebo infusion for 4 hours. Measurements were made 6 times: at initial baseline after 4 hours of placebo infusion (C1), at 4 hrs after dopamine was initiated(H4), 8 hrs (H8), 24 hrs (H24), 48 hrs (H48), and then after the second 4 hour placebo infusion (C2). During the study, volume loading with hydroxyethyl starch were performed to maintain optimal pulmonary artery wedge pressure. With dopamine, creatinine clearance increased from baseline of 67 mL/min and peaked at 92.2 mL/min (p < 0.01) at 8 hours (Figure 10). Urine output increased from 88 mL/hr and peaked at 176 mL/hr at 8 hrs as well (Figure 11). Improvement in creatinine clearance and urine output disappeared by 48 hours, with no significant difference between C1 and H48 with p > 0.05. ICU mortality rate was 0%.

Figure 10. Creatinine clearance at various time interval. P < 0.05 vs. C1 at H4, H8, H24

Figure 11. Urine output (mL/L) at various time interval. P < 0.05 at H4, H8

In a retrospective, observational study using the data from the placebo arm of NORASEPT II Study (multicenter, randomized, double-blind trials comparing monoclonal antibody to human tumor necrosis factor-a with placebo), 395 of the total 1,879 study patients with septic shock had oliguria (UO < 30 mL or <0.5 mL/kg for at least 1 hour that was unresponsive to a 500 mL fluid challenge) at the time of randomization³¹. Inclusion criteria include 1) clinical evidence of acute infection, 2) hyperthermia or hypothermia, 3) tachycardia, 4) need for mechanical ventilation or tachypnea, 5) evidence of organ dysfunction within 12 hours before enrollment. Exclusion criteria included chronic renal failure or the need for dialysis. Like the previous study, the 395 patients were divided into 3 groups: 1) ND – no dopamine [N=94, 24%] , 2) LDD - < 3 m g/kg/min [N=174, 44%], and 3) HDD > 3 m g/kg/min [N=127, 32%]. Only the ND and LDD are included in this discussion. The average dose of dopamine in LDD group was 2.5 + 0.5 m g/kg/min. Baseline creatinines were similar in both groups (1.6 + 0.8 mg/dL for ND and 1.7 + 1.0 mg/dL for LDD). Norepinephrine was used in 83 (88%) and 148 (85%) of the ND and LDD group, respectively. There were no known differences in baseline characteristics. Of these patients, diuretics, dopamine, and all other drugs were administered at the discretion of the physicians. Outcome measurement (acute renal failure, need for dialysis, 28-day survival) were assessed up to 28 days. There were no statistical differences in these outcomes between the two groups (Table 3)..

Table 3. Study outcome by dopamine category. P> 0.05 for all pairs

Given that many of the critically ill patients with sepsis are on pressors in clinical practice, this studies showed that there are insufficient evidence that low dose dopamine improves survival or decrease need for dialysis and incidence of acute renal failure in septic oliguric patients on pressors.

There are only a few studies on the renal effect of low dose dopamine alone in acute renal failure. Most studies are small uncontrolled case series. Furthermore, most of these case series uses oliguria as entry criteria, a frequent but not invariable feature of acute renal failure, and patients enrolled in these studies probably do not have acute renal failure had more stringent criteria been used. In addition, some studies do not measure creatinine clearance or other estimation of glomerular filtration rate. Four larger case series (11 to 52 patients) are presented below (Appendix III). Abizaid et al²⁴, whose first phase of the study was mentioned earlier, also performed a randomized controlled study in the second phase. There is also one larger retrospective studies assessing mortality and dialysis requirement when low dose dopamine are used.

In a case series by Davis et al. with 15 post cardiac surgical patients (CABG, valve surgery, etc.) having urine output less than 0.5 mL/kg/hr for two hours despite pulmonary arterial wedge pressure greater than 12 mm Hg, dopamine were infused at 100 m g/min for one hour³². Eight patients were on other vasoactive drugs (pressors and vasodilators). Urine output increased from 22 to 54 mL/hr (p < 0.05) and creatinine clearance increased from 70 mL/min to 115 mL/min (p < 0.05). In 9 patients, urine output remain less than 0.5 mL/kg/hr and dopamine was increased to 200 m g/min for 5 hours. For these 9 patients, creatinine clearance increased from 63 mL/min pre-dopamine to 108 mL/min (p < 0.05) on 200 m g/min of dopamine. Urine output also increased from 22 mL/hr to 50 mL/hr ( p < 0.05). There was one hospital death and one patient later required dialysis. This study showed that in post-cardiac surgical patients with oliguria and mild renal insufficiency, short term LDD may improve urine output and creatinine clearance.

In a case series by Flancbaum et al. with 19 patients in the surgical ICU with urine output less than 0.5 mL/kg/hr for at least one hour despite achieving a pulmonary arterial wedge pressure of > 10 mm Hg, mean arterial pressure of > 65 mm Hg, and cardiac index of > 2.0 L/min/m², dopamine at 2.8 + 0.37 m g/kg/min were infused for 4 hours³³. Urine output increased from 0.37 mL/kg/hr to 1.04 mL/kg/hr (p < 0.001) over the treatment periods. Creatinine at surgical ICU admission was 1.08 + 0.5 mg/dL and did not change during hospitalization. No patient developed acute renal failure.

Henderson et al. studied 11 patients with oliguria (UO < 20 mL/hr x 3 hrs) in the intensive care units³⁴. Diagnosis of the patients included perforated colonic diverticulum, septicemia, severe burns, acute pancreatitis, and barbiturate overdose. Central venous pressure was kept between 5 to 10 mm H₂O with intravenous fluid replacement. Patients received dopamine 1 m g/kg/min for 12 hours. Urine output was measured every 3 hours (Quarter 1 through 4: Q1 - 4). Serum creatinine were not significantly different before and after infusion (334 + 180 m mol/L vs. 376 + 185 m mol/L). Urine output increased significantly from 10 mL/hr up to 118 mL/hr (p < 0.005) during the second quarter of the study (Figure 12). Four patients died and three required dialysis for acute renal failure. The study showed that in oliguric patients with moderate renal insufficiency, short term LDD appear to increase urine output.

In a case series by Parker et al, 52 patients with 1) creatinine clearance less than 40 mL/min/1.73 m² or a fall in CrCl by more than 30 mL/min/1.73 m² compared to value at ICU admission, 2) urine output < 1 mL/kg/hr despite achieving pulmonary arterial wedge pressure > 12 mm Hg and/or cardiac index > 3.5 L/min/M² with fluids were given dopamine at 1.5 – 2.5 m g/kg/min³⁵. In 18 patients, urine output did not increase to at least 1 mg/kg/hr with dopamine infusion and furosemide 3-5 mg/kg were given intravenously over 24 hours. Paired 3 hour data collection on and off dopamine for urine output and creatinine clearance were collected as follow (Figure 13):

Figure 13. Study design and collection protocol.

A total of 199 paired collections were obtained. Average creatinine clearance increased by 3.8 mL/min/1.73/m² (p < 0.05), likely of no clinical significance. Urine output increased by 30 mL/hr (p < 0.05). Four patients eventually needed dialysis.

In a second phase of the study by Abizaid et al²⁴ mentioned above, 72 patients with contrast nephropathy (creatinine increased > 25% above baseline) were randomized to receive 0.45% saline at 1 mL/kg/hr or dopamine at 2.5 m g/kg/hr plus saline hydration till serum creatine returned to baseline. Both group received similar amount of contrast. Patients on LDD had high peak creatinine (3.7 vs. 2.7, p = 0.0.1) and four patients in the group received dialysis compared to none in the control group (p = 0.04) (Table 4 & Figure 14²⁴). Therefore, LDD may actually increase the risk of requiring dialysis. However, the result may be skewed by the very sensitive nature of the inclusion criteria of contrast nephropathy.

Figure 14. Peak serum creatinine levels and time course for both treatment groups

Table 4. Comparison between control and LDD group after contrast-induce renal failure.

In a retrospective, observational study using the data from Auriculin Anaritide Acute Renal Failure Study Group (a randomized, placebo-controlled clinical trial comparing the effect of synthetic atrial natriuretic peptide on the need of dialysis and mortality among patients with acute tubular necrosis of ischemic or toxic origin), 256 patients in the placebo arm of the study were identified³⁶. All patients had ATN with progressive renal insufficiency with a rise in serum creatinine of at least 1.0 mg/dL over the previous 24 to 48 hours. Patients who have chronic renal failure, other causes of ARF, prior transplant, or shock (SBP < 90 mm Hg with pressor support), or anticipated dialysis within 24 hours are excluded. These 256 patients were divided into 3 groups according to the maximum dopamine dosage they received.: 1) ND – no dopamine [N=79], 2) LDD - < 3 m g/kg/min [N=86], and 3) HDD > 3 m g/kg/min [N=91]. Only data from the ND and LDD group will be presented. Prevalence of chronic conditions (diabetes, hypertension, coronary artery disease, cirrhosis, metastatic malignancy, chronic congestive heart failure, immunosuppression, and chronic obstructive pulmonary disease) were similar between the two groups. The LDD group had a higher percentage of patients with mechanical ventilation, sepsis, or arrhythmia but had a smaller percentage of patients with oliguria (Fig. 15). The baseline creatinine in the LDD group is also lower (4.2 vs. 5.2 mg/dL, p < 0.05).

Of these patients, diuretics, dopamine, and all other drugs were administered at the discretion of the physicians. Independent correlates of low-dose dopamine use were identified. The relative risks (RR) of death and the combined outcome of death or dialysis were estimated using proportional hazard regression with adjustment for potential confounding and bias. Outcomes were assessed up to 60 days after randomization. Unadjusted analysis showed the relative risk of death is 1.11 (95% CI, 0.66 – 1.89) and that of combined outcome of death or dialysis is 1.10 (95% CI, 0.71 – 1.71). After adjusting for confounding variables (including male sex, oliguria, mechanical ventilation, acute myocardial infarction, stroke or seizure, chronic immunosuppression, total bilirubin, bicarbonate, serum albumin, arrhythmia), the relative risk of death is 0.82 (95% CI 0.42 – 1.6) and that of the combined outcome of death or dialysis 0.95 (0.58 – 1.58). In short, in patients with acute renal failure due to acute tubular necrosis, there is insufficient evidence to show that low dose dopamine improved survival or combined risk of death or dialysis.

Figure 15. Selected baseline characteristics by dopamine category

To summarize the studies presented on the use of low dose dopamine for acute renal failure, there are four small case series, one randomized controlled study, and one retrospective study. The case series do not provide sufficient evidence on the effect of LDD on glomerular filtration rate. Urine output did consistently increase over the study period (140-1100%). However, the studies are limited by using oliguria as the criteria and patients most likely had mild to moderate renal insufficiency, not acute renal failure. The one randomized control trial of 72 patients with contrast nephropathy showed that LDD does not confer additional benefit compared with saline alone and may increase the risk of requiring dialysis. However, it is limited by the sensitive definition of acute renal failure. A larger retrospective study show that there is insufficient evidence that LDD in patients with ARF due to ATN improves mortality or combined risk of mortality and dialysis.

Potential Harmful Effects of Low-Dose Dopamine

Few studies have evaluated rigorously the potential toxicity of low dose dopamine in critically ill patients. Besides local extravasation of dopamine which can provoke distal ischemia and gangrene, there are a number of known side-effects even at low doses (Table 5¹⁵).

Table 5. Potential harmful effect of dopamine

 

Harmful effects of dopamine

Tachyarrhythmias and myocardial ischemia

Increased left and right ventricular afterload

Depressed respiratory drive

Interpulmonary shunting

Risks of central line insertion

Over diuresis in volume depletion

Altered immune response

Possible gut ischemia

Hypokalemia and hypophosphatemia

The adverse reactions are not benign in nature and must be taken into account should low dose dopamine be used.

Low dose dopamine has been used for the last three decades in the treatment of acute renal failure. However, there is insufficient evidence to support its use.

For the prevention of acute renal failure in high risk patients, randomized controlled studies do not support the use of low dose dopamine. In patients undergoing coronary angiogram with diabetes or mild renal insufficiency, there may be a trend towards more acute renal failure with the use of low dose dopamine. Despite having different protocols and patient populations, studies are fairly consistent in showing that there is no improvement in renal function in terms of increased creatinine clearance or urine output. However, small number of patients and low incidence of acute renal failure limit many studies. Until larger randomized controlled trial or meta-analysis indicates otherwise, low dose dopamine is not recommended for prophylactic used for high risk patients, especially in patients undergoing coronary angiogram.

On the effect of low dose dopamine in critically ill patients with mild renal insufficiency (CrCl 60 - 80 mL/min), there are only a few case studies with small number of patients and measured outcomes are short-term changes in creatinine clearance and urine output. In most cases, low dose dopamine increases urine output substantially (30-100%). Patient groups with some or all of the patients on other pressors appear to derive no improvement in creatinine clearance from low dose dopamine, while patient groups with none of the patients on any pressors have improved creatinine clearance (31-58%). A larger retrospective study showed that there is insufficient evidence that low dose dopamine decreases the risk of acute renal failure, need for dialysis, and mortality in septic oliguric patients with pressors. Overall, current evidences are insufficient to determine whether low dose dopamine should be used in critically ill patients with mild renal insufficiency. Randomized controlled trials are much needed.

Lastly, on the use of low dose dopamine in acute renal failure, there is NO well designed randomized controlled studies with enough power on how low-dose dopamine affects the course of acute renal failure. A larger retrospective study show that there is insufficient evidence that LDD in patients with ARF due to ATN improves mortality or combined risk of mortality and dialysis. From the small case series on oliguric patients, LDD appears to improve urine output (140-1100%), at least temporarily, in hydrated patients who have mild to moderate renal insufficiency, consistent with other case series of patients with mild renal insufficiency without oliguria. Therefore, if a patient has mild to moderate renal insufficiency or acute renal failure and is oliguric, it is reasonable to try low dose dopamine for a few hours to see if urine output increases, assuming that the risk of LDD is minimal relative to its benefit. If urine output does not increase, LDD should be discontinued. If urine output increases, LDD can be continued for one to two days and then discontinued, as its effect appears to decrease with time.

Well designed randomized controlled studies are much needed to specifically assess whether low dose dopamine 1) decreases the need for dialysis, 2) improves fluid management, and 3) decrease mortality and other morbidity in patients with acute renal failure.

References

¹ Brenner, BM, et al. Acute renal failure. Brenner and Rector’s The Kidney, 6^th Ed. W.B. Saunders Company; 1021-1262; 2000.