| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1998;65:684-690
© 1998 The Society of Thoracic Surgeons
-Ketoglutarate Early After Coronary Operations
Department of Thoracic and Cardiovascular Surgery, Sahlgrenska University Hospital, Göteborg, Sweden
Accepted for publication August 29, 1997.
Dr Jeppsson, Department of Thoracic and Cardiovascular Surgery, Sahlgrenska University Hospital, S-413 45 Göteborg, Sweden (e-mail: rolf.ekroth@hjl.gu.se).
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
-Ketoglutarate (-KG) is a Krebs cycle intermediate and the carbon skeleton of glutamate. -Ketoglutarate has provoked interest in heart surgery because of its proposed critical role in myocardial metabolism. This study investigates the role of -KG in renal function after cardiac surgical procedures.
Methods. Twenty-two patients with normal preoperative renal function were included in a prospective, randomized, and controlled study. Eleven patients received intravenous infusion of 30 g -KG/hour after the operation. Measurements were performed before operation, immediately after operation, and after 30 minutes of -KG infusion.
Results. Renal blood flow was higher during -KG infusion, 297% ± 97% (of preoperative value), than in controls, 125% ± 20% (p < 0.05). Filtration fraction was lower (12.3% ± 0.05% versus 17.2% ± 1.1%, p < 0.01), which prevented a significant difference in glomerular filtration rate. The renal arteriovenous differences of lactate, glutamate, glutamine, and glycine changed toward a net release during -KG infusion.
Conclusions. Infusion of -KG enhances renal blood flow early after coronary surgical procedures in patients with normal renal function. The mechanism is unclear, but could be associated with primarily metabolic effects, and may potentially convey a beneficial effect for renal function.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
-Ketoglutarate (-KG) is a Krebs cycle intermediate and the carbon skeleton of glutamate and glutamine. This metabolite has provoked interest in heart surgical procedures because of its suggested critical role in myocardial energy metabolism [1]. In line with this suggestion, the provision of -KG or glutamate has reduced ischemic injury and improved myocardial function in experimental and clinical heart operations [2][3][4][5].
The present work investigates another role of -KG in heart surgical procedures, namely its effect on renal perfusion and function. Our interest originated in the finding that renal blood flow (RBF) could be increased by 50% early after coronary operation by infusion of mixed amino acids [6]. Our data were in keeping with experimental and clinical nonsurgical work indicating vasodilator effects of amino acids [7][8]. The mechanism for renal vasodilation, however, is unclear, although arginine has been shown to induce renal vasodilation, suggesting the involvement of nitric oxide [9]. Others have concluded that all amino acids participating in metabolic processes in the kidney, such as gluconeogenesis and base generation, are involved in renal vasodilation [10].
The present study explores whether infusion of -KG, which can replace amino acids for base generation, in the gluconeogenic process, and for oxidation, would enhance RBF and if so, whether it does so to a similar extent as a comparable amount of mixed amino acids. A secondary issue was to study renal metabolic and functional effects of -KG infusion.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
|
The operations were performed with standard nonpulsatile cardiopulmonary bypass technique, with moderate hypothermia (nasal temperature, 30°C) and hemodilution (hematocrit, 20% to 30%). Cardioprotection was achieved with St Thomas’ crystalloid cardioplegia. Weaning off bypass was performed after rewarming to a rectal temperature of at least 36°C.
Study Protocol
Before the surgical procedure a retrograde thermodilution catheter (Webster Laboratories, Baldwin Park, CA, USA) was inserted under fluoroscopic guidance, with the distal mixing thermistor in a central position in the left renal vein. The position of the catheter was checked postoperatively. A Swan-Ganz antegrade thermodilution catheter (Baxter Healthcare Corp., Edwards Division, Santa Ana, CA) was placed in the left pulmonary artery, and its position was checked with fluoroscopy. According to the clinical routine, a femoral artery and a central venous catheter were inserted preoperatively. These catheters were used for pressure measurements and for blood sampling.
Hemodynamic measurements were performed at three preset time points: before the surgical procedure, immediately after the operation (in the operating room after skin closure, with satisfactory clinical hemodynamic variables, ie, mean arterial pressure more than 60 mm Hg, urinary output more than 2 mL · kg-1 · h-1, and mixed venous oxygen saturation more than 60%), and after 30 minutes of infusion (which continued during the third measurements). Hemodynamic measurements preceded metabolic and functional measurements. Each measurement period lasted approximately 10 minutes.
After the first postoperative measurement the patients were randomly allocated to -KG treatment or to serve as control subjects. The -KG group (n = 11) received infusion of approximately 66 mL (100 mL/h for 40 minutes: 30 minutes before hemodynamic measurements started plus 10 minutes required for completion of all measurements) of the -KG solution (300 g/L), while the control group (n = 11) did not receive any infusion.
Hemodynamic Measurements
Renal blood flow was determined by retrograde thermodilution technique, originally described by Hornych and associates [11] and modified by Tidgren and Hjemdahl [12]. At each measurement, RBF was measured during at least 45 seconds and the mean was calculated. Renal vascular resistance was calculated according to the formula 
, where CVP (central venous pressure) was assumed to be representative of renal venous pressure, and MAP is mean arterial pressure. All flows were related to body surface area.
Renal Function Measurements
Preoperative and Postoperative Renal Function
Serum creatinine level was measured before the operation, on days 1 through 3 after the operation, and on the day of discharge from the hospital (Table 1).
Glomerular Function
Glomerular filtration rate was calculated with the formula 
, where RPF is the renal plasma flow and FF is filtration fraction. Renal plasma flow is calculated with the formula 
, where Hct is the hematocrit of renal venous blood. 
, which was calculated according to the formula: 
. Concentrations of 51Cr-EDTA were measured in arterial blood and renal venous blood, according to a method described previously [13].
Tubular Function
Extraction of p-aminohippurate (PAH) was calculated according to the following formula: 
. p-Aminohippurate was measured in arterial and renal venous blood according to a method described previously [13].
Fractional excretion of sodium was calculated with sodium and 51Cr-EDTA concentrations according to the following formula: 
. Sodium was analyzed in plasma with an ABL510 analyzer (Radiometer, Copenhagen, Denmark) and in urine with an FLM 3 analyzer (Radiometer; Copenhagen, Denmark). 51Cr-EDTA was measured in urine and plasma according to a method previously described [13]. Urine was sampled during 20 minutes, with the endpoint coinciding with blood sampling. Before each urine sampling, the bladder was rinsed.
Metabolic Measurements
The concentrations of amino acids in arterial and in renal venous blood were determined by ion exchange chromatography using an automated amino acid analyzer (Alpha Plus, LKG Products, Bromma, Sweden). Free fatty acids were analyzed in serum with an enzymatic colorimetric method (NEFA C, Wako Chemicals GmbH, Neuss, Germany). Glucose and lactate concentrations were measured in whole blood in an automated analyzer, (2300 Stat Plus, Yellow Springs Instruments Co, Yellow Springs, OH). Renal arteriovenous differences of amino acids, free fatty acids, glucose, and lactate were calculated according to the following formula: 
.
Blood Gases
Blood oxygen saturation (HbSO2), hemoglobin content (Hb), and oxygen tension (PO2) were measured using an ABL510 analyzer (Radiometer, Copenhagen, Denmark). The blood oxygen content was calculated according to the following formula: 
. Renal arteriovenous difference of oxygen were calculated according to the formula: arterial oxygen content - renal venous oxygen content.
Statistical Analysis
The patients were randomly allocated to either of the two study groups by a computerized procedure of sequential allocation [14]. To establish whether the randomization had provided groups that were comparable before treatment, intergroup comparisons of the presurgical values and the first postoperative values were performed with the nonparametric Mann-Whitney U test. To evaluate effects of -KG, intergroup comparisons were done during treatment with Mann-Whitney U test. The values of RBF and RBF-derived variables (renal vascular resistance, glomerular filtration rate) were measured with thermodilution technique, which is influenced by individual venous anatomy and catheter position. Both factors may differ between individuals, which may compromise comparisons between individuals. One way of enhancing comparisons between individuals is to normalize values of each patient. We did this by expressing the postoperative values of each patient in relation to his presurgical value. It was reasoned that the presurgical value should better reflect the normal renal blood flow than measurements after surgical trauma, hemodilution, hypothermia, and extracorporeal perfusion.
The results are expressed as means ± standard error of the mean. Statistical significance was defined as p less than 0.05.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
-KG group had a transient elevation in serum creatinine up to 30 mg/L, which returned to the normal range before discharge from hospital.
Perfusion and Renal Function
Cardiac output, central venous pressure, pulmonary capillary wedge pressure, and systemic vascular resistance were not affected by treatment (Table 2).
|
-KG infusion, (297% ± 97% of presurgical value), than in control patients (125% ± 20%, p < 0.05, Fig 1). There was no significant difference in renal vascular resistance between groups (82% ± 27% versus 109% ± 16%). Filtration fraction was lower in the -KG group (12.3% ± 0.5% versus 17.2% ± 1.1%, p < 0.01), and although glomerular filtration rate tended to be higher during -KG infusion (260% ± 80% versus 180% ± 42%), no statistically significant difference was seen. p-Aminohippurate extraction was lower in the -KG group than in the control group (0.52 ± 0.04 versus 0.87 ± 0.01; p < 0.01). Fractional excretion of sodium tended to be higher in the -KG group (14.1% ± 3.1% versus 8.1% ± 1.4%), but the statistical analyses did not indicate a difference (p = 0.09).
|
-KG group than in the control group, (-0.13 ± 0.02 versus 0.01 ± 0.01 mmol/L; p < 0.01) (Fig 2). Renal arteriovenous differences of glutamine, glutamate, glycine, and taurine moved in a negative direction in the -KG group toward a net release (Fig 2; Table 4).
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
-KG causes a significant increase in renal blood flow in the postoperative period after a coronary operation. The effect is comparable with what is seen during infusion of mixed amino acids. In addition, it is suggested that exogenous -KG participates in renal metabolism, because renal metabolic changes were observed.
Amino Acids and Renal Blood Flow
Recently our group demonstrated that a mixed amino acid infusion enhances RBF early after cardiac surgical procedures [6]. Our finding was in line with a body of evidence from animal and clinical, nonsurgical work showing that amino acids reduce renal vascular resistance and increase RBF [7][8]. The mechanism for amino acid-induced renal vasodilation has been debated. Various alternatives have been proposed, with the main options being either primarily metabolic or a direct effect on the renal vasculature. It has also been discussed whether the effects are mediated by systemic or intrarenal agents. The finding that the effect can be accomplished in the isolated renal preparation argues for intrarenal mechanisms [10].
-Ketoglutarate and Renal Metabolism
One intrarenal alternative relates to renal metabolic activity of amino acids. Glutamine, which dominates renal amino acid uptake, is metabolized for three purposes: oxidation, base generation, and gluconeogenesis. All three processes include deamination of glutamine to -KG, which is further metabolized to glucose or H2CO3. Although the kidney can utilize a number of substrates, the current data point to glutamine as the dominating substrate, judging from balance studies. According to Brezis and coworkers [10], glutamine is particularly efficient in causing renal vasodilation, suggesting a coupling between metabolic and vascular effects.
The present results are compatible with the concept that amino acid-induced renal vasodilation is linked to metabolic events, possibly to Krebs cycle activity. Thus, infusion of -KG enhanced RBF, comparable to what has been observed during amino acid infusion, and initiated marked changes in renal balances of glutamate, glutamine, and lactate. Our data cannot determine which biochemical event precedes changes in renal net balances. However, according to experimental work, -KG is used for base generation, and it inhibits renal glutamine degradation [15][16]. This leads to accumulation of glutamine in the renal cortex, followed by reduced uptake of glutamine. Concomitantly, glutamate synthesis is promoted, with subsequently increased renal release of glutamate.
Normal renal lactate metabolism is complex, with cortical glucogenic lactate consumption occurring simultaneously with medullar anaerobic lactate production. The infusion of -KG induced lactate release. From theoretical considerations, this could imply that -KG replaced lactate as substrate for glucogenesis, while lactate production continued unchanged. This is conceivable, because renal glucose release was unchanged (suggesting continuing gluconeogenesis). Alternatively, the release of lactate could represent renal hypoxia, with increased anaerobic glycolysis, although the markedly enhanced RBF argues against such a mechanism.
Renal Blood Flow and Renal Function
A motive for increasing RBF is to improve the relationship between renal energy supply and demand. However, increased RBF could lead to increased energy demand if glomerular filtration rate is increased, with a resultant increase in tubular workload. In the present study, filtration fraction decreased, which attenuated the increase in workload. This in turn implies that renal perfusion increased more than tubular load, suggesting an improved energy supply–demand balance.
Tubular activity was assessed in two ways in our study: by measuring fractional extraction of PAH and fractional excretion of sodium. Fractional PAH extraction decreased markedly and fractional excretion of sodium tended to increase (p = 0.09) during -KG treatment. This normally suggests reduced tubular activity. There is evidence, though, that PAH and -KG compete for active tubular transport, and the reduced PAH extraction probably implies that -KG used the PAH transport system [17].
The tendency for lower sodium extraction (p = 0.09) indicates a fractionally lower extraction, but is probably explained by the moderately higher plasma sodium levels in this group (Table 2). This in turn reflects that -KG was administered as a sodium salt. Taken together, these considerations imply that we cannot make any firm conclusions on tubular effects, although the postoperative follow-up of normal, unchanged serum creatinine levels argues against any adverse effects. Admittedly, our data in this respect are confusing. Unfortunately, there are no alternative, clinically available methods to evaluate tubular function during -KG treatment. To clarify this issue, nonclinical experimental studies are required.
Methodologic Issues
Two further methodologic issues merit discussion. First, when designing the study protocol, we discussed whether the control group should receive a volume equivalent to that of the -KG solution. We decided against this. It was reasoned that the approximately 50- to 66-mL infusion during 30 to 40 minutes would not cause any significant volume expansion. According to measurements of preload (pulmonary capillary wedge pressure, central venous pressure, Table 2) and cardiac output data, there were no differences between study groups. This argues against volume expansion as being responsible for enhanced RBF and in favor of a pharmacologic effect of -KG.
A second debatable issue concerns renal uptake and release of metabolites. Ideally, the product of blood flow and arteriovenous difference should have been used as a quantitative estimate of uptake or release. As described in the Patients and Methods section, RBF data were normalized as a percentage of the preoperative value. This procedure is valuable in the analysis of blood flows, but unfortunately, it creates problems for the evaluation of uptake and release data. The normalization procedure cannot be used (for mathematical reasons) when negative values are involved (which is the case for many of the arteriovenous difference values). Instead, arteriovenous differences only were used for the analysis of metabolic effects. This estimate is critically dependent on blood flow, and a change in arteriovenous differences can sometimes be explained totally by a change in blood flow without any change in the true uptake or release. For example, an increase in blood flow at an unchanged release will result in a reduced arteriovenous difference. This means that arteriovenous differences should be used with caution, particularly when both blood flows and arteriovenous differences change. In the present series, as RBF increased, the arteriovenous differences of lactate, glutamine, and glutamate (and a few other amino acids) widened further, strongly suggesting that the true release of these metabolites was amplified by treatment. This suggestion is further supported by Mann-Whitney tests of the product of arteriovenous difference and "nonnormalized" RBF values (data not shown). These tests confirmed the differences in Table 3 and Table 4. Still, arteriovenous difference is not an ideal quantitative measure of uptake or release, but can be useful as a qualitative estimate if blood flow changes are appreciated.
Clinical Implications
Acute renal failure, secondary to renal hypoperfusion, is a dreaded and relatively frequent complication in heart operations. It predominantly complicates operations involving an already compromised renal function or postoperative low output heart failure. In the present series we selected patients with good functional reserves to obtain an uncomplicated setting for the experimental procedure. It was reasoned that the clinical need for vasoactive and renoactive agents during -KG treatment would make the analysis of -KG effects difficult. This was a major concern, because it was not known at the time whether the treatment had any renal effects in humans, let alone in clinical heart operations. According to the present results, -KG infusion enhances RBF in low-risk patients undergoing heart surgical procedures, with the clinical potential to reduce the risk for postoperative renal failure. It is not known if the same effect exists in patients at higher risk. This patient category clearly would benefit more from an efficient renoprotective agent, and follow-up studies including high-risk patients are urgently needed.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
lberg, Anneli Ambring, Eva Brändström, Liselott Thunberg, and Kerstin Andersson, PhD, is gratefully acknowledged. The study was supported by grants from Sahlgrenska University Hospital Foundation, Göteborg Medical Society, The Swedish Medical Society, Nils Gunnarssons’s Legacy, and Pharmacia-Upjohn, Stockholm, Sweden. |
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentAcknowledgmentsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
