Postoperative Infusion of Amino Acids Induces a Positive Protein Balance Independently of the Type of Analgesia Used

Sixteen patients scheduled to undergo elective colon resection for benign and malignant lesions were studied between October 2004 and March 2005. Exclusion criteria were as follows: more than 20% loss of body weight in the past 6 months, evidence of metastatic disease, severe cardiac and respiratory diseases, diabetes and albumin below 35 g/l, and anemia (hemoglobin less that 100 g/l). The Ethics Committee of the McGill University Health Center, Montreal, Quebec, Canada, approved the study (REC#03-039), and informed consent was obtained from all patients. The patients were assigned to two groups, A (PCA) and B (epidural analgesia), using a computer-generated randomization schedule.

No premedication was administered. General anesthesia in both groups consisted of propofol, nitrous oxide in 40% oxygen, desflurane, fentanyl, and rocuronium. In group B, an epidural catheter was inserted between T9 and T11 before induction of general anesthesia. Neuraxial blockade was established with 15 ml bupivacaine, 0.5%, to achieve a bilateral sensory block (to ice and pinprick) from T4 to S5 and maintained with intermittent boluses of 5 ml bupivacaine, 0.25%, every hour.

At the end of surgery, analgesia in group A was maintained with PCA with intravenous morphine, adjusted to obtain a visual analog scale score less than 4 at rest (scale: 0 = no pain to 10 = worst pain imaginable). Group B received a continuous epidural infusion with a mixture of 0.1% bupivacaine and 2 μg/ml fentanyl administered at a rate between 8 and 15 ml/h with supplemental top-ups of 0.125% bupivacaine to maintain a sensory block from T7 to L3 and a visual analog scale score less than 4 at rest. In both groups, pain was assessed twice a day, at 8:00 am and 8:00 pm.

During surgery, patients were kept normothermic, using a warming blanket spread over the body, and well hydrated with 0.9% isotonic sodium chloride solution infused at a rate of 6 ml · kg⁻¹· h⁻¹. After surgery, patients received only clear fluids till the end of isotope infusion. Lactated Ringer's solution was infused at a rate of 1.7 ml · kg⁻¹· h⁻¹as patients arrived in the ward. All fluid administration was stopped at the beginning of the isotope infusion.

All patients were studied on the second postoperative day beginning at 8:00 am. The protocol included two periods: a fasted state of 3 h followed by a 3-h fed state during which patients received a solution of 10% amino acids without electrolytes (Travasol; Baxter, Montreal, Canada) infused for 3 h at a rate of 0.02 ml · kg⁻¹· min⁻¹, equivalent to 2.9 g · kg⁻¹· day⁻¹.

The kinetics of whole body leucine and glucose were measured using an isotope dilution technique and the stable isotope tracers l-[1-¹³C]leucine and [6,6-²H²]glucose (Cambridge Isotope Laboratories, Cambridge, MA). A superficial vein in the dorsum of the hand was cannulated, and the catheter was kept patent with heparinized saline to withdraw blood samples. A second catheter was placed in a vein of the contralateral arm to provide access for the infusion of the tracers. After collecting blood and expired-air samples to determine baseline enrichments, priming doses of NaH¹³CO³(1 μmol/kg), l-[1-¹³C]leucine (4 μmol/kg) and [6,6-²H²]glucose (22 μmol/kg) were administered and followed by a continuous infusion of L-[1-¹³C]leucine (0.06 μmol · kg⁻¹· min⁻¹) and [6,6-²H²]glucose (0.22 μmol · kg⁻¹· min⁻¹) for a total period of 6 h (3 h of fasted state and 3 h of fed state). During the latter period, the dose of L-[1-¹³C]leucine was increased to 0.12 μmol · kg⁻¹· min⁻¹. Toward the end of the fasted and fed states, four blood and expired-air samples were collected at 10-min intervals to determine isotope enrichments. Blood samples for the analysis of plasma concentrations of glucose and hormones (cortisol, glucagon, and insulin) were collected once during each state, at 150 and 330 min into the isotopic infusion. Each blood sample was immediately transferred to a heparinized tube and centrifuged at 4°C (3,000 rpm for 15 min) and then stored at −70°C until analysis. Expired-air samples were collected in a 2-l latex bag and then transferred immediately to 10-ml tubes (Vacutainer; Becton Dickinson, Franklin Lakes, NJ) for carbon dioxide isotope enrichment analysis.

Whole body oxygen consumption and carbon dioxide production were measured using indirect calorimetry (Vmax 29N; SensorMedics, Yorba Linda, CA) in the last hour of the fasted and fed states. Measurements were performed for 20 min on each occasion, and average values of whole body oxygen consumption, carbon dioxide production, and calculated respiratory quotient were calculated, with a coefficient of variation less than 10%. A graphic illustration of the study protocol is presented in figure 1.

Plasma ketoisocaproate was analyzed to represent intracellular leucine enrichment, and it was determined by positive chemical ionization gas chromatography–mass spectrometry, as previously described.11Expired ¹³CO²enrichment was analyzed by means of isotope ratio mass spectrometry (Analytical Precision AP2003; Manchester, United Kingdom). Plasma glucose was derivatized to its penta-acetate compound, and the [6,6-²H²] glucose enrichment was determined by gas chromatography–mass spectrometry using electron impact ionization. In each analysis run, duplicate injections were performed, and their means of enrichment at four time points were taken to represent enrichment at isotopic steady state.

Plasma concentration of glucose was measured by a glucose oxidase method using a glucose analyzer 2 (Beckman Instruments, Fullerton, CA). Circulating concentrations of insulin and glucagon were measured by sensitive and specific double-antibody radioimmunoassays (Amersham International, Bucks, United Kingdom). Cortisol plasma concentration was measured using the Ciba Corning ACS 180 automated immunoassay (Ciba Corning Diagnostic, East Walpole, MA).

Whole body leucine kinetic was calculated by conventional isotope dilution practice using a two-pool stochastic model during steady state conditions, obtained at each study phase (fasted or fed). When an isotopic steady state exists, the rate of appearance (R^a) of a substrate in plasma can be derived from the plasma enrichment (atom percent excess [APE]) calculated by R^a= (APE^inf/APE^pl− 1) · F, where F is the infusion rate of the labeled tracer (μmol · kg⁻¹· min⁻¹), APE^infis the isotopic enrichment in the infusate, and APE^plis the tracer enrichment in plasma at steady state. The APE value used in this calculation represents the mean of the APE values determined during each isotopic plateau. The accuracy of the isotopic enrichments at isotopic plateau was tested by evaluating the scatter of the APE values above their mean, expressed as a coefficient of variation. A coefficient of variation less than 5% was used as a confirmation of a valid plateau. Under steady state conditions, leucine flux (Q) is defined by the equation Q = S + O = B + I, where S is the rate at which leucine is incorporated into body protein, O is the rate of leucine oxidation, B is the rate at which unlabeled leucine enters the free amino acid pool from endogenous protein breakdown, and I is the rate of dietary intake or the rate of infusion of L-[1-¹³C]leucine (μmol · kg⁻¹· h⁻¹) or both. Inspection of that formula indicates that, when studies are conducted in the postabsorptive state, flux is equal to breakdown. Enrichment of plasma ketoisocaproate during infusion of L-[1-¹³C]leucine has been used to determine whole body leucine kinetics. During amino acid infusion, net leucine flux was calculated by subtracting the leucine infusion rate from the total R^aof leucine. This steady state reciprocal pool model is considered to represent the intracellular precursor pool enrichment more precisely than leucine itself.22In the calculation of oxidation, a factor of 0.76 was applied during the fasted state and accounts for the fraction of ¹³C-carbon dioxide released from leucine but retained within slow turnover rate pools of the body. A factor of 0.92 was used for the fed state.22 

In the fasted state, the R^aglucose was equal to the endogenous production of glucose. In the physiologic steady state, whole body glucose uptake equals the rate of endogenous glucose production. The glucose clearance, an index of the ability of the tissues to take up glucose, was calculated as R^aglucose divided by the corresponding plasma glucose concentration.

Based on expected difference in protein balance of 5 μmol · kg⁻¹· min⁻¹between the two groups (SD = 3 μmol · kg⁻¹· min⁻¹; power 80% and P = 0.05), a total of 16 patients was calculated to be sufficient.10All data are presented as mean ± SD. Analysis of dependent variables was performed using two-factorial analysis of variance with repeated measures. Significant effects induced by parenteral nutrition were assumed when P  values for time dependency were less than 0.05. Influences by analgesic regimen were accepted as significant when the analgesic or the interaction term of the analysis of variance was less than 0.05.

Demographic characteristics and clinical data were similar in both groups (table 1). A plateau in the enrichments of plasma [1-¹³C]α-ketoisocaproate, expired ¹³C-carbon dioxide, and [6,6-²H²]glucose was achieved in the fasted and fed states (coefficient of variation < 5%), allowing the application of steady state equation to calculate glucose and protein kinetics.

Pain scores at rest were similar between groups during the first and second day (table 2). Patients with epidural analgesia had a lower visual analog scale score on coughing during the first and second day than did patients with PCA (table 2). Consumption of bupivacaine, fentanyl, and morphine are reported in table 2.

In the fasted state, endogenous R^aglucose was higher in the PCA group compared with the epidural group (table 3). The infusion of amino acids caused a decrease in endogenous R^aglucose in both groups, with greater changes in the PCA group (table 3). Glucose clearance decreased in both groups by more than 50%, and there was no difference between the two groups studied (table 3).

Rate of appearance of leucine (equivalent to protein breakdown in the fasted state) leucine oxidation, protein synthesis, and protein balance during the fasted state were similar in both groups (table 3). Administration of amino acids suppressed leucine appearance from protein breakdown in both groups, although the decrease was greater in the PCA group (table 3). Leucine oxidation increased in both groups, with greater change in the epidural group (table 3). Protein synthesis increased to a similar extent in both groups (table 3). Protein balance was negative in the fasted state and became positive during the infusion of amino acids (table 3). This net anabolic effect was greater in the PCA group (table 3).

Circulating concentration of glucose during the fasted state was similar in both groups, and the small increase observed after the amino acid infusion was not significant (table 4).

During the fed state, plasma concentration of insulin and glucagon increased by the same magnitude, whereas no changes were observed in plasma cortisol and insulin/glucagon ratio (table 4).

Consumption of oxygen, production of carbon dioxide, and respiratory quotient were not affected by the infusion of amino acids in either group (table 5). Production of carbon dioxide was higher in the epidural group during both the fasted and fed states (table 5).

The results of the current study indicate that the infusion of amino acids during the second postoperative day induced a positive protein balance, regardless of the type of analgesia provided, although balance was greater in the PCA group. At the same time, amino acid infusion decreased endogenous glucose production in both groups without affecting plasma glucose concentrations.

The data provided by the isotopic analysis allow us to dissect the different components of whole body protein metabolism and understand the principal metabolic mechanism governing protein economy. From the current findings, a 3-h infusion of amino acids, equivalent to 0.36 g · kg⁻¹· day⁻¹, during the second postoperative day, suppressed protein breakdown by more than 25%. Using leucine as a presentative amino acid, approximately 30–40% of the amino acids made available from proteolysis were oxidized, whereas 12–16% were redirected toward protein synthesis.

In a previous investigation using a similar protocol and infusing the same amount of amino acids supplemented by dextrose, at a rate of 4 mg · kg⁻¹· min⁻¹, it was found that protein balance was positive in both groups, but patients who received epidural analgesia had a slightly greater increase (27 μmol · kg⁻¹· h⁻¹) compared with patients treated with PCA (25 μmol · kg⁻¹· h⁻¹).10It was proposed that the greater insulin sensitivity in patients with epidural analgesia, as reflected by a greater increase in glucose clearance in these patients, could be responsible for such difference. Such difference was not evident in the current study, as shown by the same variation in glucose clearance in both groups. In addition, the modifications of the hormonal levels were the same whether patients had epidural or PCA. Therefore, it is not possible to explain the observed anabolic differences between the two analgesic techniques on the basis of the underlying hormonal mechanisms.

It has been shown consistently that epidural analgesia compared with intravenous opioid analgesia attenuates the postoperative nitrogen loss in patients undergoing upper abdominal surgery.5,7,8However, nitrogen balance cannot differentiate the contribution from changes in protein synthesis and in protein degradation. Therefore, it cannot provide any information about how changes in protein balance are achieved. In other studies conducted using the stable isotope technique, epidural analgesia was found to attenuate protein breakdown or decrease leucine oxidation.6,11These beneficial effects should be considered in relation to the type of nutritional support. In fact, in a previous study, conducted after an overnight fast and using the same measurement methodology, epidural analgesia attenuated postoperative protein breakdown without affecting protein synthesis, resulting in patients being in negative protein balance.23When dextrose was infused, epidural analgesia compared with PCA induced a decrease in leucine oxidation during the second postoperative day, but protein balance remained negative in both groups, with no differences between the analgesic techniques, indicating that hypocaloric amounts of dextrose could not reverse negative postoperative protein balance.11 

To quantify the difference in magnitude in anabolism between nutritional support and type of analgesia, the infusion of amino acids in the current study caused an average increase in protein balance of 36.7 μmol · kg⁻¹· h⁻¹, whereas the increase in protein balance was 7.3 μmol · kg⁻¹· h⁻¹as a result of the analgesia technique. Similarly, in the study where glucose and amino acids were infused together, the protein balance increased by 26 μmol · kg⁻¹· h⁻¹, whereas epidural accounted for only an increase of 2.1 μmol · kg⁻¹· h⁻¹.10Therefore, in both studies, the effects of amino acids on protein balance were 5–10 times more powerful than the type of analgesia used.

In the current study, the rate of infusion of amino acids was 2.9 g · kg⁻¹· day⁻¹, such that the plasma amino acid concentration was maintained twofold to threefold above basal value.24Three hours of amino acid infusion was found to be sufficient for maximal incorporation of amino acids into whole body and tissue compartments.25The quantity of amino acids administered was 0.36 g · kg⁻¹· day⁻¹, less than the daily recommended intake of 1.5 g · kg⁻¹· day⁻¹;26nevertheless, a consistent anabolic effect was shown. This is in agreement with previous findings demonstrating that amino acids are more efficiently utilized for maintaining lean body mass when given in divided doses rather than with continuous infusion.25 

Suppression of gluconeogenesis reduces the need for muscle protein breakdown to supply gluconeogenic amino acids. If the rate of gluconeogenesis from amino acids is decreased, that amount of nitrogen is available for reincorporation into protein rather than for excretion as urea. In the current study, EGP was slightly decreased (15–30%), whereas in the previously described study,10where the nutritional support was amino acids plus glucose, EGP was almost totally suppressed (80–90%). Paradoxically, amino acid oxidation was similar in both studies, but protein balance was greater in the current study. Therefore, even if the inhibition of EGP should spare amino acids, these do not become automatically available for the synthetic pathways.

Glucose is produced endogenously by both glycogenolysis and gluconeogenesis. Under normal overnight postabsorptive conditions, glycogenolysis constitutes approximately 50% of whole body glucose production, with the remainder being derived from gluconeogenesis.27Gluconeogenesis progressively increases with the duration of fasting, contributing to more than 90% of glucose production after 42 h of fasting.28Considering the long perioperative fasting, the endogenous glucose production was primarily of gluconeogenic origin in our study.

The observed reduction in EGP during an amino acid infusion is in agreement with previous studies in the postoperative period and on the third day after trauma where an infusion of an amino acids mixture at a similar rate decreased endogenous glucose production by 8–12%.20,29However, in contrast with these observations in patients, infusion of amino acids in volunteers increased endogenous glucose production. Tappy et al.  18showed increases in EGP and gluconeogenesis by 84% and 235%, respectively, after an infusion of amino acids at a rate of 4.8 g · kg⁻¹· day⁻¹. In contrast, Krebs et al.  19measured EGP by stable isotopes and glycogenolysis by ¹³C nuclear magnetic resonance spectrometer in volunteers after overnight fasting and found that gluconeogenesis increased by 100% after the amino acid infusion at a rate of 4.8 g · kg⁻¹· day⁻¹. Such increase had been explained as a result of a direct effect of amino acids acting as substrate for the gluconeogenic pathway and indirectly by increased plasma glucagon concentration, which stimulates gluconeogenesis.

Despite the decrease in EGP in both groups, a decrease in glucose clearance was observed, implying that glucose uptake was decreased. The amino acids effect of decreasing glucose consumption has been found also in studies of volunteers, where it has been demonstrated that they inhibit glucose transport/phosphorylation resulting in a decrease in intracellular utilization of glucose.30 

In conclusion, a short-term infusion of amino acids after colorectal surgery inhibits protein breakdown and stimulates protein synthesis, thus rendering protein balance positive in both groups, although balance was greater in the PCA group. The effect of amino acids on postoperative glucose metabolism was characterized by a decrease glucose clearance indicating a state of insulin resistance and by a decrease in endogenous glucose production.

The authors thank Patrick Charlebois, M.D., and Barry Stein, M.D. (Assistant Professors, Department of Surgery, McGill University Health Centre, Montreal, Quebec, Canada), for studying the two doctors' patients.