Use of fenoldopam to control renal dysfunction early after liver transplantation

Authors


Abstract

With the aim of assessing whether fenoldopam can help to preserve renal function after liver transplantation, we randomized 140 consecutive recipients with comparable preoperative renal function to receive fenoldopam 0.1 μg/kg/minute (group F, 46 patients), dopamine 3 μg/kg/minute (group D, 48 patients), or placebo (group P, 46 patients) from the time of anesthesia induction to 96 hours postoperatively. There were no differences between the groups in intraoperative urinary output or furosemide administration (both P = .1). Daily recordings made during the first 4 postoperative days revealed no significant differences in urinary output (P = .1), serum creatinine (P = .5), the incidence of renal insufficiency (P = .7), the need for loop diuretics (P = .9) or vasoactive drugs (P = .8). In comparison with preoperative levels, creatinine clearance at the end of the study in the patients receiving fenoldopam remained substantially unchanged, whereas it decreased by 39 and 12.3%, respectively, in the subjects receiving placebo or dopamine (P < .001); blood cyclosporine A (CsA) levels were similar in the 3 groups (P = .1). Three subjects died in the intensive care unit (1 in each group, P = .9), 2 of them had renal failure. In conclusion, our results confirm the inefficacy of dopamine in preventing or limiting early renal dysfunction after liver transplantation, and suggest that fenoldopam may preserve creatinine clearance by counterbalancing the renal vasoconstrictive effect of CsA, as it has been reported in previous experimental studies. (Liver Transpl 2004;10:986–992.)

Acute renal failure (ARF) is defined as a sudden and marked reduction in the amount of renal glomerular filtrate, leading to an accumulation of nitrogen and other toxins that complicates the management of fluids and electrolytes;1 it affects about 5% of hospitalized patients and approximately 15% of those who are critically ill, thus prolonging their length of hospitalization, increasing the related costs, and considerably affecting their outcomes.2

Orthotopic liver transplantation (OLT) is burdened by an incidence of ARF that ranges from 10 to 25%, with a mortality rate that exceeds 50% when renal replacement therapy is required;3–5 in this setting, even moderate renal dysfunction is significantly associated with shortened graft and recipient survival.6

As renal failure is a major risk factor capable of seriously compromising OLT outcomes, various strategies have been adopted in an attempt to prevent or at least limit its deleterious effects, but none have yet proved to be really efficacious.7–10 In order to refine the effects of dopamine on renal function, a number of analogs have been synthesized. One of these, fenoldopam, has attracted the attention of clinicians because of its ability to selectively vasodilate renal vessels, and has thus renewed interest in the possibility of combating the onset of perioperative ARF by means of intravenous drug infusion.

The aim of this study was to evaluate the efficacy of fenoldopam as a “renoprotective” agent in the early post–liver transplantation phase—a period for which no data concerning its possible use are yet available.

Abbreviations

OLT, orthotopic liver transplantation; CsA, cyclosporine A; ARF, acute renal failure; ICU, intensive care unit.

Materials and Methods

This study was approved by our Institutional Review Board, and involved a population of subjects consecutively undergoing OLT at our Centre who gave their informed consent. The only exclusion criterion was a preoperative diagnosis of renal dysfunction, defined as a serum creatinine level of >1.5 mg/dl.

Intraoperative Management

All of the patients underwent the same anesthetic procedure, which included fentanyl (0.2 mg), thiopental sodium (4 mg/kg), and succinylcholine (1 mg/kg) for induction and, after tracheal intubation, sevofluorane at inhalation doses of 1 to 2.5% in a respiratory mixture of 50% oxygen and air, delivered using low fresh gas flows (2 L/minute); remifentanil (0.2–0.3 μg/kg/minute), and cisatracurium (3 μg/kg/minute) were also administered. The native liver was removed after supra- and subhepatic clamping of the inferior vena cava, preceded by the positioning and initiation of a venovenous extracorporeal bypass. All of the transplants were performed by the same surgeon (Filipponi, F). At the end of surgery, all of the patients were transferred to the Intensive Care Unit (ICU). Hemodynamic monitoring included invasive systemic arterial, pulmonary artery, and capillary occlusion pressures, and the measurement of cardiac performance by means of a pulmonary artery catheter (CCO/SVO2 Thermodilution Catheter; Edwards Life Sciences, Irvine, CA). Decisions regarding the administration of fluids and blood products were made on the basis of standards of care in order to ensure hemodynamic stability and to correct any coagulation abnormalities or bleeding.

Immunosuppression

The immunosuppressive protocol included an oral formulation of cyclosporine A (CsA) (Sandimmun Neoral; Novartis Pharma S.A., Haningue, France) administered at a starting dose of 15 mg/kg; 10 mg/kg methylprednisolone (Solu-Medrol; Pharmacia & Upjon, Puurs, Belgium) intraoperatively (subsequently reduced by 50% per day to a prednisolone dose of 20 mg/day), and 20 mg intravenous basiliximab (Simulect; Novartis Pharma S.A., Haningue, France) intraoperatively, during the anhepatic phase and on the fourth postoperative day.

Study Protocol

Before undergoing surgery, the patients were randomized to one of the following intravenous treatments using unmarked, sealed envelopes containing the group allocation (the order of the envelopes was determined blindly by a member of the Department's secretarial staff who played no other role in the study):

  • Group F: fenoldopam (Corlopam; Segix SpA, Milan, Italy) 0.1 μg/kg/minute

  • Group D: dopamine (PHT; Pharma Srl, Milan, Italy) 3 μg/kg/minute

  • Group P: placebo saline

All of the treatments were started upon the induction of anesthesia, and then continued for 96 hours postoperatively.

The following variables were recorded during the study (the postoperative data collection lasted 96 hours):

  • Hourly intraoperative and postoperative diuresis. We targeted a minimal urine output of 1.5 ml/kg/hour during the intraoperative phase and 1 ml/kg/hour throughout the ICU stay; if these levels were not reached, furosemide was administered at a starting dose of 20 mg and, if necessary, followed by stepwise increases to 75, 125, 250, and 500 mg (bolus or continuous intravenous infusion). Before giving the loop diuretic, the attending clinicians were required to evaluate the patient's volemic balance carefully.

  • Daily postoperative serum creatinine levels.

  • Postoperative creatinine clearance (calculated every 12 hours by means of the simultaneous collection of blood and urine samples, the latter from continuous collections).

  • Absolute changes from baseline in serum creatinine and creatinine clearance levels 24, 48, 72 and 96 hours after the end of surgery.

  • Cumulative fluid balances throughout the study period.

  • The incidence of renal dysfunction.

  • The need for postoperative renal replacement therapies.

The total amount of administered diuretics, the occurrence of hypotension (defined as the need to administer intravenous vasoactive drugs during the study period), trough blood CsA levels, and patient outcome were also recorded. The ICU physicians were not aware of which group the patients had been randomized to, and the study was not financially supported by any pharmaceutical company.

Definitions

Renal dysfunction was defined as serum creatinine levels of ≥1.5 mg/dl, the doubling of serum creatinine within 24 hours, or the onset of acute oliguria requiring the use of continuous renal replacement therapy.11, 12 The function of the transplanted organs was evaluated on the basis of standard criteria.13, 14

Statistical Analysis

The data are presented as mean values +SD. After the analysis of variance, the groups were compared using repeated measures and one-way analysis of variance. The categorical variables were analyzed using the chi-squared test according to Brandt-Snedecor. The statistical analyses were performed using STATA software (Release 7.0; Stata, College Station, TX); a P value of <.05 was considered significant.

Results

This study involved a population of 144 consecutive subjects who underwent OLT at our Centre between January 2001 and July 2003; 4 were excluded because of incomplete sample collections. The analysis was therefore based on 140 patients: 46 in group P (placebo), 48 in group D (dopamine), and 46 in group F (fenoldopam). The main pre-, intra-, and postoperative data were similar in the 3 treatment groups (Table 1). There was no therapy crossover between the groups, nor was it necessary to discontinue any of the study drugs at any time because of the occurrence of adverse events potentially attributable to them.

Table 1. Demographic and Clinical Data of the Study Population
 Group P (placebo)Group D (dopamine)Group F (fenoldopam)
  • Abbreviations: PBRC, packed red blood cells; ICU, intensive care unit; PGD, primary graft dysfunction; CTP, Child-Turcotte-Pugh; B/C, grade B and C of cirrhosis severity according to Child classification; NC, not classified.

  • *

    The considered aPPT value is the worst observed.

  • Numbers expressed as n (%) or ± standard deviation.

Patients, n (%)46 (32.8)48 (34.3)46 (32.8)
Median age, years (25–75th percentiles)51 (42–57.5)50 (46–54)45 (42–50)
Gender, M/F34/1237/1135/11
CTP, B/C/NC20/22/424/17/719/19/8
Underlying liver disease, n (%)   
 Post-HBV cirrhosis4 (8.7)6 (12.5)5 (10.9)
 Post-HCV cirrhosis18 (39.1)16 (33.3)15 (32.6)
 Multiviral cirrhosis15 (32.6)18 (37.5)18 (39.1)
 Ethanol cirrhosis2 (4.3)3 (6.3)3 (6.5)
 Other7 (15.3)5 (10.4)5 (10.8)
Preoperative serum creatinine mg/dL0.88 ± 0.20.85 ± 0.30.81 ± 0.2
Preoperative creatinine clearance (ml/min)110.7 ± 43102.5 ± 41.6110 ± 42.8
Cold ischemia time, minutes549.3 ± 128.8521.6 ± 195.3535 ± 154.7
Duration of surgery, minutes399.8 ± 101.7371.5 ± 59.6394 ± 57.1
Intraoperative PRBC units2.3 ± 2.12.1 ± 2.32.4 ± 2.0
Postoperative PRBC units2.4 ± 2.62.7 ± 2.92.2 ± 2.3
Patients with aPTT ratio* ≤1.22/>1.22, n39/743/540/6
Patients with PGD0 (0)1 (2)1 (2.1)
ICU outcome, alive/dead45/147/145/1
ICU stay, days5.6 ± 6.74.7 ± 4.74.2 ± 3.3
Hospital stay, days24.3 ± 19.122.2 ± 17.322.8 ± 18.1
Hospital outcome, alive/dead45/147/145/1

Intraoperative Phase

Total intraoperative diuresis was 4.3 ± 2.8 ml/kg/hour in group F, 5.5 ± 2.3 in group D, and 4.8 ± 2.5 in group P (P = .7). The anhepatic phase lasted 65.8 ± 18.5 min in group F, 69.7 ± 29.3 in group D, and 63.7 ± 17.8 in group P (P = .5), with urine outputs of 8.5 ±4.4 ml/kg/hour, 7.5 ± 6.6 ml/kg/hour, and 8.1 ± 5.0 ml/kg/hour, respectively (P = .8). There was no between-group difference in the amount of furosemide required to ensure the target minimum urine production of 1.5 ml/kg/hour (33.4 ± 47 in group F, 25 ± 46.1 in group D, and 17.3 ± 38 mg in group P; P = .1), or in the use of vasoactive drugs (3 subjects in groups P and F, and 4 in group D; P = .8).

Postoperative Phase

The ANOVA for repeated measures did not reveal any significant variations in blood cyclosporine levels (P = .1), diuresis (P = .1), serum creatinine (P = .5), or blood furosemide levels (P = .9) (Table 2). There was no difference in the need for vasoactive drugs (3 patients in group P, 2 in group D, and 1 in group F; P = .8).

Table 2. Postoperative Data
 Group P (placebo) 46 patientsGroup D (dopamine) 48 patientsGroup F (fenoldopam) 46 patients
  1. Abbreviation: CsA, cyclosporine A.

  2. Values expressed as ± standard deviation.

Diuresis, ml/kg/h   
 day 11.7 ± 0.41.4 ± 0.71.6 ± 0.6
 day 21.2 ± 0.61.2 ± 0.31.3 ± 0.7
 day 31.4 ± 0.81.3 ± 0.31.3 ± 0.8
 day 41.6 ± 0.61.3 ± 0.51.6 ± 0.7
Serum creatinine, mg/dL   
 day 10.7 ± 0.10.7 ± 0.10.8 ± 0.4
 day 20.8 ± 0.30.8 ± 0.31 ± 0.7
 day 31 ± 0.90.9 ± 0.41 ± 0.7
 day 40.9 ± 10.9 ± 0.41 ± 0.8
Serum CsA (trough) ng/mL   
 day 1167 ± 188190 ± 112171 ± 213
 day 2353 ± 202441 ± 202440 ± 236
 day 3379 ± 127310 ± 101408 ± 153
 day 4328 ± 136308 ± 78369 ± 142
Furosemide, mg   
 day 115.7 ± 30.319.4 ± 24.926.5 ± 102
 day 255.9 ± 173.159.2 ± 173.255.7 ± 187.3
 day 373.5 ± 246.286 ± 23989 ± 221.9
 day 476 ± 155.881.8 ± 137.171.1 ± 174.9

Creatinine clearance (measured every 12 hours) decreased in all of the patients, but the reduction was less marked in the patients treated with fenoldopam; this difference became statistically significant from the 72nd hour (Fig. 1). Moreover, at the end of the study (96 hours after OLT), the preoperative creatinine clearance levels of the patients receiving fenoldopam were substantially unchanged, showing an increase of 3%, as opposed to decreases of 39% in group P and 12.3% in group D (Fig. 2). The analysis of variance for repeated measures revealed significant changes from baseline in creatinine clearance (P < .001), which correlated with the use of fenoldopam (P < .001). There were no significant differences in cumulative fluid balances during the study period: 149 ± 1659 ml in group F, 135.5 ± 1051 ml in group D, and 126.6 ± 888.9 ml in group P (P = .2).

Figure 1.

Creatinine clearance during the study. Mean values +SE. a: P < .05 vs. placebo. b: P < .01 vs. placebo. c: P < .001 vs. placebo. d: P < .05 vs. dopamine.

Figure 2.

Changes in creatinine clearance after surgery.

The incidence of renal failure in the study population as a whole was 10.7% (15 subjects): 6 patients (13%) in group P, 5 (10.4%) in group D, and 4 (8.7%) in group F (P = .7). Three patients required continuous venovenous hemofiltration: 1 in each group (P = .9). In 10 cases, ARF was associated with other postoperative complications, including cardiac failure, septic shock, the need for surgical reintervention, and respiratory failure. Two subjects (1 in group D and 1 in group F) showed primary graft dysfunction, but neither required retransplantation. Three subjects died in the ICU (1 in each group, P = .9), leading to a total ICU mortality rate of 2.7%: 2 of them had ARF (66.6%).

Discussion

As renal failure is relatively frequent in subjects with end-stage liver disease, and may increase mortality regardless of transplantation,3–5, 12, 16 OLT clinicians have always been interested in strategies and drugs capable of preventing or at least reducing its harmful consequences. The administration of small and frequent doses of furosemide has been found to have some protective effect,17 but there is still no evidence that loop diuretics reduce mortality or the need for replacement therapy in critically ill patients.18 The use of “renal” doses of dopamine during the intraoperative period of OLT was associated with a lower incidence of ARF in one study,8 but this was not confirmed in a subsequent study19 and, in fact, “low-dose” dopamine is now generally considered to be not only of unproven benefit but perhaps even harmful.20. Despite these disappointing results, the above-mentioned drugs are still widely used in clinical practice to prevent and/or treat ARF.18

At doses of between 0.1 and 0.3 μg/kg/minute, fenoldopam (which was originally developed in oral form as a drug for long-term use in congestive heart failure) acts as a selective dopamine-1 (DA1) receptor agonist that dilates the renal artery, thus increasing total renal blood flow (as a result of the activation of adenylate cyclase) and natriuresis, as a result of the inhibition of sodium-potassium adenosine triphosphatase–dependent processes in the proximal convoluted tubule and the thick part of the ascending loop of Henle.21 Its characteristic rapid and predictable onset of action, half-life, cytochrome P450-independent metabolism (which avoids overdosing in hepatopathic subjects), and linear dose/effect relationship,21 make it easy to use even in hemodynamically unstable subjects, such as cirrhotic patients undergoing OLT. It has a number of theoretical advantages: tubules may become more resistant to ischemic injury because the dopaminergic-1 blockade of the tubular Na±-K+–adenosine triphosphatase pump reduces medullary oxygen demand; renal vasodilation may increase the availability of oxygen for the kidneys; the maintenance of a high tubular flow may reduce the risk of obstruction and consequent back-leakage; and limiting the vasoconstrictive reflex may favor renal blood flow.21 However, some caveats have been reported: even small doses of fenoldopam can cause severe hypotension (due to a possible coupling of its effect with that of the administered anesthetic drugs); furthermore, clinicians may be led into the error of considering it a substitute for adequate patient rehydration.22

Nevertheless, the first (albeit limited) clinical experiences involving the use of fenoldopam as a means of preventing ARF were encouraging in patients undergoing coronary bypass grafting,23 or major abdominal24 or elective aortic surgery,25 and in patients at risk for radiocontrast nephropathy,26 although a subsequent study found that it offered no additional protection over intravenous saline when administered after cardiovascular interventions with radiocontrast dye.27

The clinical situation of OLT recipients is frequently more complicated than that of other patients because renal dysfunction in the perioperative period may be due to various mechanisms. Before transplantation, subjects with end-stage liver disease may be affected by renal hypoperfusion caused by reduced perfusion pressure or increased renal vascular resistance.16 Furthermore, kidney function may be threatened in every phase of OLT surgery: during the preanhepatic phase, there may be even severe hemodynamic instability due to the surgical maneuvers and sometimes considerable blood loss;28 during the anhepatic phase, renal venous outflow may be hampered as a result of the vena cava and portal vein cross-clamping (thus leading to an increase in subdiaphragmatic venous pressure) and renal artery perfusion pressure may be reduced because of the simultaneous decrease in systemic arterial pressure and the cardiac index;28 during the neohepatic phase, a reperfusion syndrome may occur, with hypotension, arrhythmias, lactic acidosis, hypothermia, electrolyte disturbances and coagulopathy.28 Finally, a series of events capable of causing renal dysfunction may occur during the post-OLT phase as a result of arterial hypotension, coagulopathy, hypovolemia, the administration of nephrotoxic drugs (e.g., antibiotics, radiological contrast media, immunosuppressants),28 or abdominal hypertension.29

In such a complex setting, it may be wishful thinking to expect significant preventive or therapeutic effects on often impending renal dysfunction by administering a single “magic” protective agent such as fenoldopam, however promising it may seem.22 However, the results of our study show a substantial preservation of renal function in fenoldopam-treated patients in comparison with preoperative values: creatinine clearance increased by 3%, as opposed to the decrease of 12.3% in group D and 39% in group P. Furthermore, comparison of the 12-hourly creatinine clearance values in the 3 groups showed that they were significantly higher in the fenoldopam-treated patients from the 72nd postoperative hour. This, together with the absence of any between-group difference in blood CsA levels, may be related to the possibly greater efficacy of fenoldopam against the renal activity of CsA, which is capable of causing even intense renal vasoconstriction, potentially leading to renal dysfunction.13, 28

In this regard, and in comparison with placebo, it was found that fenoldopam administered at a dose of 0.1 μg/kg/minute increased renal blood flow in a dose-dependent manner in a population of normotensive subjects, without any changes in systemic blood pressure or heart rate.21 Furthermore, in a rat model, fenoldopam not only prevented CsA-induced renal vasoconstriction, but was also capable of treating the renal damage caused by chronic CsA administration.30 These findings have been reproduced in kidney transplant recipients treated with an oral fenoldopam formulation.31

In conclusion, patients undergoing OLT may experience renal dysfunction as a result of a wide range of causes and mechanisms that may be interdependent and may vary in importance on an individual basis.12, 16–28, 32 The results of our study confirm that dopamine has no real renoprotective action in this setting, and that fenoldopam is capable of preserving creatinine clearance. They also suggest that it may counteract the renal vasoconstrictive effect of CsA. The use of fenoldopam could therefore be considered particularly appropriate in the case of recipients at risk of renal dysfunction when blood CsA levels tend to be high. As the design of this study excluded high-risk subjects (such as those already affected by renal dysfunction before surgery), it is essential to test fenoldopam in such patients.

Ancillary