| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
*Department of Anesthesiology, Yale University School of Medicine, New Haven, Connecticut;
Department of Anesthesiology, Duke University Medical Center, Durham, North Carolina;
Department of Internal Medicine, Vanderbilt University Medical Center, Nashville, TN; and
Department of Internal Medicine (Section of Nephrology), Yale University School of Medicine, New Haven, Connecticut
Address correspondence and reprint requests to Susan Garwood, MB, ChB, Department of Anesthesiology, Yale University School of Medicine, PO Box 208051, New Haven, CT 06520-8051. Address e-mail to susan.garwood{at}yale.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
1.5 mg/dL were prospectively randomized into 4 groups: 1) placebo, 2) mannitol 1 g/kg added to the CPB prime, 3) dopamine 2 µg · kg-1 · min-1 from the induction of anesthesia to 1 h post-CPB, or 4) mannitol plus dopamine. The primary outcome measure was ß2M excretion rate at 1 h post-CPB. Secondary outcome measures included ß2M excretion rate at 6 and 24 h post-CPB; urinary flow rate and creatinine clearance at 1, 6, and 24 h post-CPB; and the highest postoperative serum creatinine level. Length of intensive care stay and hospitalization, as well as adverse events, were also considered secondary outcomes. Dopamine significantly increased ß2M excretion rate at 1 h post-CPB (2.48 ± 3.61 µg/min) compared with placebo (0.59 ± 1.04 µg/min; P = 0.001). This effect was not ameliorated by the addition of mannitol (ß2M excretion rate, 2.05 ± 2.77 µg/min; P = 0.007 compared with placebo). ß2M excretion rate was similar in patients given placebo or mannitol alone (P = 0.831). Rather than being a protective drug in the setting of CPB, dopamine alone or in combination with mannitol increases ß2M excretion rate, which may be a measure of renal tubular dysfunction. The clinical implications of this increase and whether it is also seen in patients with established renal dysfunction undergoing CPB require additional investigation. IMPLICATIONS: In many clinical settings, an increased beta-2-microglobulin (ß2M) excretion rate indicates renal tubular injury. In this cardiopulmonary bypass (CPB) study, a dopamine infusion (alone or with mannitol) resulted in an increased ß2M excretion rate. It is unclear whether this dopamine-related increase implies renal injury after CPB, and further investigations are required to examine the mechanism/clinical relevance of this observation.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
DA increases renal blood flow and glomerular filtration in healthy adults (2,3), as well as in animal models of septic, ischemic, and nephrotoxic renal failure (4,5). These improvements in renal performance have not been translated into clinical benefits, possibly because of the fundamental difference between most experimental models of acute renal insufficiency and clinical renal dysfunction. Most experimental insults are single interventions, whereas clinical renal failure is considered multifactorial. Even though studies in cardiac surgery patients have demonstrated that DA increases urinary output (6–8), no concurrent improvement in renal function has been detected (7,8). Perioperative urinary flow, in particular, does not appear to correlate with renal outcome (9). Likewise, mannitol has had protective properties in experimental models, but studies in adult cardiac surgery patients have been unable to replicate the animal findings. The discrepancies between the animal and clinical data may reflect a true lack of efficacy in renal protection or, because of shortcomings in sample size or choice of indicator of renal function (10), an inability to detect an actual improvement in renal performance. Most investigations have focused on diuresis (6–8,11) or glomerular function, as measured by serum creatinine or creatinine clearance (8,11), to evaluate renal performance in patients undergoing cardiac surgery. However, urinary output, serum creatinine levels, and creatinine clearances are insensitive monitors of renal function (9,12). Consequently, unless the anticipated change in glomerular function is substantial, detection would require such a large sample size that multicenter trials would be required (10).
Tubular excretion of low-molecular-weight proteins has been proposed as a more sensitive marker of renal injury (10,13,14). One such marker is ß2-microglobulin (ß2M), an 11,800-Da protein that is expressed on the surface of all nucleated cells as part of the major histocompatibility complex. It is 95% filtered by the glomerulus and 99.9% reabsorbed in the proximal tubule. A decrease in tubular reabsorption and an increase in urinary excretion of ß2M accompany even minimal changes in proximal tubular function. ß2M excretion has a sensitivity of 84.6% and a specificity of 100% as a marker for tubular lesions (15). It has been extensively used in the diagnosis and monitoring of chronic disease processes, in the acute onset of clinical renal dysfunction, in detecting renal changes after therapeutic protocols, and in acute experimental clinical protocols. Furthermore, it has been established by biopsy that ß2M excretion rate correlates with clinically important renal tubular damage (16,17) and can track changes in tubular pathophysiology (16,18,19). Finally, ß2M excretion rate has been included in an international database as a useful screening tool for nephrotoxic exposure (20). The purpose of this study was to assess the effects of mannitol and DA, the two drugs used most often as renal protective adjuncts in CPB, on renal tubular function as measured by changes in ß2M excretion.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Informed, written consent was obtained from 135 patients, who were enrolled in the study between September 1995 and March 1998. Male and nonpregnant female patients aged 21 to 79 yr, with a preoperative serum creatinine level of 1.5 mg/dL, who were scheduled for elective, primary coronary artery bypass graft surgery requiring CPB, were included in this study. Patients were excluded who had had cardiac catheterization within 5 days of surgery; preoperative hypotension, defined as a systolic blood pressure <90 mm Hg, at any time; use of an intraaortic balloon pump at any time during the current hospitalization; the administration of dopaminergic or antidopaminergic drugs; a contraindication to the use of DA; and chronic inflammatory disease states, lymphoproliferative disorders, or carcinoma, which are associated with increased serum levels of ß2M. Because of the diurnal variation in ß2M excretion, patients undergoing surgery in the late afternoon or evening were also excluded.
All patients received their routine cardiac medications before surgery and were premedicated with morphine sulfate 0.1 mg/kg and 0.2–0.4 mg of scopolamine administered IM 1 h before surgery. Standard ASA noninvasive monitors and peripheral IV and intraarterial catheters were placed before the induction. Anesthesia was induced with sufentanil (2–5 µg/kg) and midazolam (0.03–0.05 mg/kg). Intubation was facilitated with pancuronium. Patients were ventilated with a volume-controlled ventilator to achieve normocapnia (PaCO2, 36–44 mm Hg). A flow-directed pulmonary arterial catheter, transesophageal echocardiography probe, urinary bladder catheter, and temperature probes (esophageal and bladder) were inserted after the induction. Anesthesia was maintained with a continuous infusion of sufentanil (1 µg · kg-1 · h-1 before CPB; 0.15 µg · kg-1 · h-1 during and after CPB) and midazolam (0.2 µg · kg-1 · min-1 before CPB; 0.05 µg · kg-1 · min-1 during and after CPB). No volatile anesthetics were used, and muscle relaxation was achieved with pancuronium.
Systolic blood pressure was maintained within 20% of the patient’s baseline pressure (the average of three preoperative readings) during the pre-CPB period and at 90–130 mm Hg systolic after CPB. Perfusion pressure was maintained between 50 and 80 mm Hg during CPB. Episodes of hypertension and hypotension (defined as pressures outside these specified ranges) were treated, as clinically appropriate, to remain within the specified ranges during the intraoperative period. If urinary output remained less than 0.5 mL · kg-1 · h-1 for 2 or more consecutive collection periods (1 h pre- and post-CPB and 30 min during CPB), 20 mg of furosemide was administered.
The bypass circuit was primed with 2 L of lactated Ringer’s solution, 5000 U of heparin, 50 mEq of sodium bicarbonate, and mannitol or placebo, as previously described. Crystalloid cardioplegia consisted of sodium 142 mEq/L, potassium 20 mEq/L, chloride 159 mEq/L, and dextrose 5.6 g/L, with a pH of 7.6–7.8 adjusted with sodium bicarbonate in a normal saline base. Cardioplegia was mannitol free, with or without blood, in a 4:1 proportion, according to surgeon preference. A membrane oxygenator and arterial filter were used, along with alpha-stat pH management, moderate hypothermia (26°C–32°C), and a pump flow rate of 1.8–2.4 L · min-1 · m-2. The hematocrit was maintained greater than 15% during CPB and 25% after CPB. No inhaled anesthetics were used during the CPB period.
Patient demographics, medical and surgical history, preoperative medications, routine laboratory values, and cardiac catheterization data were recorded during the preoperative anesthesia evaluation. During surgery, hemodynamic measurements, derived indices, and temperatures were downloaded at 1-min intervals into the Data Logger (Premise Development Corp., Hartford, CT). Data were analyzed offline to identify episodes of hypotension and hypertension, either of which was considered significant if it continued for five or more consecutive minutes. If the disturbance lasted longer than 5 min, then each complete 5-min period was regarded as a separate episode. Drug and fluid administration, the number and type of grafts, pump prime and cardioplegia volumes, and CPB and aortic cross-clamp times were also recorded. Each patient was assigned a "trauma score" on the basis of the number of conduits harvested: one point per graft harvested. Urinary output was noted hourly before and after CPB and at half-hourly intervals during CPB.
Systemic and pulmonary artery blood pressures and core body temperatures were downloaded every minute from the time of intensive care unit (ICU) admission until 6 h post-CPB. Episodes of hypotension and hypertension were defined and recorded as in the intraoperative period. From 6 to 24 h post-CPB, hemodynamic data were downloaded every 15 min. During this period, any single systemic blood pressure reading <90 mm Hg was regarded as a single episode of hypotension, and any reading more than 130 mm Hg was regarded as hypertension. Central pressures and cardiac index were recorded every hour from ICU admission until 24 h post-CPB (or until removal of the pulmonary artery catheter, whichever came first). All medications and fluids, hourly urinary output and chest tube drainage, routine daily serum chemistries, and any significant hemodynamic, respiratory, cardiac, neurological, or metabolic events were also noted.
For the purposes of applying inclusion and exclusion criteria and comparing demographics among groups, preoperative serum creatinine was taken as the most recent value documented in the patient’s record before patient consent and randomization. Serum creatinine and ß2M, as well as urinary ß2M, were measured at baseline (immediately after the induction of anesthesia and insertion of invasive monitors and urinary bladder catheter) and at 1, 6, and 24 h after CPB. All blood and urine samples were drawn from the indwelling arterial and urinary bladder catheter, respectively. Samples were taken from accurately timed urine collections (20 min), and the pH of the sample for ß2M determination was corrected to >6.5 with sodium hydroxide if necessary. Urine samples were stored at -20°F and serum samples at 4°F until analysis. Serum creatinine was analyzed by spectrophotometry, and serum and urinary ß2M were analyzed by radioimmunoassay. The creatinine clearance (Cockcroft-Gault formula) and ß2M excretion rate were calculated according to standard formulas.
The primary end-point for this study was ß2M excretion rate at 1 h post-CPB. Secondary outcomes were ß2M excretion rates at 6 and 24 h post-CPB, creatinine clearances, postoperative serum creatinine levels, and urinary flow rates. Length of ICU stay, length of hospitalization, and significant clinical events were also considered secondary outcomes.
Data from a pilot study1 indicated that 25 patients per group would be required to detect a significant change in ß2M excretion rate at 1 h post-CPB, assuming an 
of 0.05 and a statistical power of 80%. Data were analyzed with one-way analysis of variance, the Scheffé multiple comparison procedure, the Kruskal-Wallis H test, the Mann-Whitney U-test, Wilcoxon’s signed rank test, or the 
2 test, where appropriate. All tests were two tailed. Univariate analysis was performed for all perioperative variables, and ß2M excretion rate at 1 h was modeled by using linear and logistic regression analysis. The statistical package SPSS (SPSS Inc., Chicago, IL) was used for all analyses. Values are reported as mean and SD unless otherwise noted. P < 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
-aminocaproic acid or aprotinin. Patients who had received -aminocaproic acid were noted to experience a profound baseline microglobulinuria that was 10–20 times the normal value of 0.26 µg/min. Patients receiving -aminocaproic acid after the 1-h post-CPB time point were removed from subsequent ß2M analysis (n = 9). The remaining 100 patients were randomized into 4 groups: placebo (n = 24), mannitol (n = 26), DA (n = 25), and mannitol plus DA (n = 25). All 100 patients completed the study protocol, and there were no in-hospital mortalities. The demographic characteristics of the four groups were similar (Table 1). Except for a significantly more frequent occurrence of pulmonary disease in the placebo group, there were no differences in comorbid disease states among the four groups. Anesthetic drugs, fluids administered, and the use of intraoperative vasoactive drugs did not differ among the four groups. Intraoperative hemodynamic control was also similar, with no differences in episodes of hypotension or hypertension. Although postoperative care was determined by individual surgical preference, there were no differences in hemodynamic control, inotropes, vasodilators, fluids administered, or chest tube drainage during ICU stay.
|
Baseline renal function was similar (Table 2). Thirteen patients were given furosemide for a urinary output <0.5 mL · kg-1 · h-1 during surgery: placebo, 5 of 24; mannitol, 2 of 26; DA, 4 of 25; and mannitol plus DA, 2 of 25 (P = 0.440). All four groups had a significantly increased urinary flow rate at 1 h post-CPB compared with baseline (Table 2), with the DA group and the DA plus mannitol groups having significantly faster urinary flow rates than placebo (P = 0.008). By 24 h post-CPB, only the DA group had a significantly increased urinary flow rate compared with baseline.
|
There is some evidence that furosemide increases the excretion rate of ß2M for up to 4 h (see below). The 13 patients who received furosemide were excluded from the 1-h post-CPB ß2M analysis, and ß2M excretion rates were recalculated for that time point (Table 2). There were no statistical differences within each group regarding whether the furosemide-treated patients were included at 1 h after CPB (i.e., comparing the entire group of 100 patients with the 87 patients who did not receive furosemide at 1 h after CPB). However, the between-group differences remained in both the 100-patient analysis and the 87-non-furosemide-treated patient analysis. In those patients who did not receive intraoperative furosemide, ß2M excretion rates remained significantly increased at 1 h after CPB compared with baseline in all four groups. With the 13 furosemide-treated patients excluded, ß2M excretion rates at 1 h after CPB remained significantly increased in both the DA group and the DA plus mannitol group compared with the placebo group (P = 0.001 and P = 0.005, respectively). Because the diuretic regimen in the postoperative period was determined by routine surgical care and because ß2M excretion rates at 6 and 24 h after CPB were secondary outcomes, patients receiving furosemide in the postoperative period were not excluded from further ß2M excretion rate analyses (at 6 and 24 h post-CPB). ß2M excretion rates remained significantly more than baseline at 6 and 24 h post-CPB for all groups except for DA plus mannitol at 24 h after CPB. There were no differences among groups at 6 and 24 h after CPB.
There were no differences in creatinine clearances among groups at any time during the study (Table 2). The postoperative serum creatinine level was significantly higher than the preoperative level in all four groups (Table 3). There were, however, no differences in postoperative serum creatinine among groups (P = 0.783). Dialysis was not required for any of the patients. There were no differences in the incidence of significant postoperative events or length of stay among the four groups: placebo, 5.3 ± 1.2 days; mannitol, 5.8 ± 2.0 days; DA, 6.0 ± 2.1 days; and mannitol plus DA, 5.9 ± 3.2 days (P = 0.726).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Because DA is traditionally considered a renal vasodilator, the rationale for its use in the peribypass period is to increase renal blood flow at a time when renal vascular resistance is increased and when total and, probably, regional renal blood flow are reduced. DA is a mixed agonist, with DA-1 and DA-2 activity. Carey et al. (22) demonstrated that selective inhibition of DA-2 activity in the dog increases renal plasma flow and glomerular filtration rate, an effect that was reversed by the co-infusion of a DA-2 agonist. These investigators concluded that DA-2 activity causes renal vasoconstriction, an effect that may be mediated by the release of endothelin. Thus, on the basis of animal studies (22,23), the physiological expression of DA action at DA-2 receptors may be renal vasoconstriction and decreased glomerular filtration.
Gaudio et al. (24) demonstrated that the enhancement of renal perfusion by DA in the postischemic, energy-depleted state (which may be similar to the bypass period) did not improve histopathologic scores of tubular injury. The increased renal blood flow did not, apparently, adequately replenish adenosine triphosphate levels (24). In another study, continued inactivation of the Na+/K+ and Na+/H+ pumps under conditions of increased sodium delivery allowed sodium and water to passively diffuse into the tubular cells, with subsequent accumulation and tubular cell swelling (25).
We have demonstrated that the use of DA is associated with increased ß2M excretion rates immediately after CPB. We have also shown that the addition of mannitol to DA during CPB does not mitigate this increase in ß2M excretion rate. Logistic regression modeling revealed that immediately after CPB, patients receiving DA were almost eight times more likely to manifest an increase in ß2M excretion rate. In a prospective and randomized, but not blinded, trial of small-dose DA during CPB, Tang et al. (26) also determined that DA caused a significant increase in the excretion of retinal-binding protein, another marker validated as a sensitive and accurate indicator of early renal tubular injury.
ß2M excretion rate is used as a marker for renal tubular injury in chronic and acute diseases and in experimental and therapeutic protocols. It is an early and sensitive marker for the development of renal dysfunction in the post-CPB period (27–29). Although ß2M excretion rate has been generally described as a sensitive measure of tubular function, it has been suggested that during surgery, the increase in ß2M excretion rates may be secondary to surgical trauma and the shedding of ß2M from cell surfaces (30). In a study of only nine patients, Walenkamp et al. (30) reported that the excretion of ß2M was increased during surgery and related this increased excretion rate in their series of orthopedic patients to the level of surgical trauma. Although the results were confounded because five of the nine patients received gentamicin, which has been clearly demonstrated to increase ß2M excretion rate (31,32), Walenkamp et al. (30) postulated that surgery per se stimulates the release of ß2M from nucleated cells and that the increased urinary excretion rates were a result of exceeding ß2M reabsorption capacity of the proximal tubule. ß2M is in steady-state production, with a short half-life (33), and constant excretion rates are found in both normal individuals and those with increased production of the protein, e.g., rheumatoid arthritis or multiple myeloma patients. Unless there is concurrent renal tubular dysfunction (33,34), individuals with increased serum levels of ß2M do not exhibit increased urinary excretion of the protein. To date, no apparent saturation point for reabsorption of ß2M has been demonstrated either in animal studies (35) or in patients without underlying disease (33,34). Patients with reduced glomerular function exhibit a reduced ß2M clearance that is manifested by increased serum levels (33,34). In our study, there was no evidence for an increased serum ß2M during the period of investigation. Serum ß2M actually decreased in the early postoperative period, probably reflecting hemodilution. Furthermore, there was no correlation between the surgical trauma score and ß2M excretion rate. In fact, the group with the significantly highest surgical "trauma scores" (placebo) had the slowest ß2M excretion rates, suggesting that Walenkamp et al.’s conclusion is erroneous. There was also no correlation in our study between serum ß2M and ß2M excretion rates.
A limitation to the use of ß2M excretion as a marker of tubular injury is the coadministration of certain proteins or amino acids, such as lysine and arginine, which inhibit tubular reabsorption of ß2M and thus cause transient increases in ß2M excretion (36,37). The tubular reabsorption of ß2M is competitively inhibited by numerous proteins over a wide range of net charges and size in a dose-related manner (37). It is postulated that such proteins, including ß2M, are taken up by common tubular protein endocytic sites, irrespective of their physicochemical features (37). In particular, the administration of the lysine analogs, such as -aminocaproic acid and tranexamic acid, results in a profound ß2-microglobulinuria (36). In our study, dramatically increased ß2M excretion rates were detected in the patients receiving -aminocaproic acid, including the baseline measurement if the drug was given before the urine collection. Therefore, all patients treated with antifibrinolytic therapy before, during, or after surgery were excluded from this analysis.
Other drugs administered during CPB may be postulated to act in a similar fashion. Protamine sulfate, which was administered to all of our patients, causes a glomerular-type proteinuria (excretion of albumin and immunoglobulin G) but does not inhibit the reuptake of ß2M (37). Furosemide significantly increased ß2M excretion rates in a group of healthy volunteers (38), although the mechanism of action is unclear. In our study, removing the 13 patients who received furosemide in the intraoperative period did not make a significant difference in ß2M excretion rates at 1 hour after CPB. Although DA is not a protein, it is theoretically possible that it may interfere with the reuptake of ß2M by some physicochemical means. However, there is no current evidence to support that hypothesis.
Additional limitations to our study include the fact that postoperative care was dictated by surgical preference. The primary end-point of our study, ß2M excretion rate at one hour post-CPB, was a time point at which all patients were managed according to a standardized protocol. Consequently, we did not address the question of whether longer-term use of DA might continue to cause further increases in ß2M excretion rate. Finally, we were unable to determine whether patients with more significant preoperative renal dysfunction were similarly affected by the administration of DA with respect to ß2M excretion rates.
In conclusion, we have demonstrated that the administration of DA during CPB produces an increase in the excretion of ß2M immediately after CPB. The addition of mannitol to the DA does not appear to reduce this increased ß2M excretion rate. Since increases in ß2M excretion rate have been correlated with deteriorations in renal tubular function and biopsy-proven tubular injury in other clinical settings, it is possible that the increased ß2M excretion rate associated with the use of DA in this protocol may be consistent with some degree of tubular injury after CPB. Since our study was underpowered to detect long-term renal outcome, we are unable to determine whether the DA-associated ß2M excretion rate increase is in fact related to some degree of tubular injury or whether it would become clinically significant were the DA infusion to be given for a longer period of time. Further studies are required to explore the mechanism of the increased ß2M excretion rate caused by the administration of DA and whether this has any clinical relevance.
|
Footnotes |
|---|
1 Carcoana OV, Kates RA, Panagopoulos G. The effect of cardiopulmonary bypass on renal tubular function: does mannitol provide renal tubular protection? Br J Anaesth 1995;74(Suppl 2):A74. 
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. H. Rosner, D. Portilla, and M. D. Okusa Analytic Reviews: Cardiac Surgery as a Cause of Acute Kidney Injury: Pathogenesis and Potential Therapies J Intensive Care Med, January 1, 2008; 23(1): 3 - 18. [Abstract] [PDF]
|
|
|
|
 |
|
 M. H. Rosner and M. D. Okusa Acute Kidney Injury Associated with Cardiac Surgery Clin. J. Am. Soc. Nephrol., January 1, 2006; 1(1): 19 - 32. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. F. Fischer, B. Phillips-Bute, M. Swaminathan, C. Milano, and M. Stafford-Smith SymmetryTM Aortic Connector Devices and Acute Renal Injury: A Comparison of Renal Dysfunction After Three Different Aortocoronary Bypass Surgery Techniques Anesth. Analg., January 1, 2006; 102(1): 25 - 31. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. D. Rocca and L. Pompei To Dilate or Not To Dilate Anesth. Analg., October 1, 2005; 101(4): 1242 - 1242. [Full Text] [PDF]
|
|
|
|
 |
|
 J. O. Friedrich, N. Adhikari, M. S. Herridge, and J. Beyene Meta-Analysis: Low-Dose Dopamine Increases Urine Output but Does Not Prevent Renal Dysfunction or Death Ann Intern Med, April 5, 2005; 142(7): 510 - 524. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
