INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CONGENITAL HEART DISEASE, with a left-to-right shunt (such as a ventricular septal defect or a patent ductus arteriosus), results in volume overload of the left ventricle. Due to diversion of blood via the defect into the pulmonary arteries, the systemic blood flow may be compromised. Inadequate systemic blood flow has been described in preterm lambs with a patent ductus arteriosus and in rats with an arteriovenous fistula (9, 16). In previous studies, Gratama et al. (20) and Toorop et al. (36) demonstrated that in lambs, in a fed state and with a left-to-right shunt due to an aortopulmonary shunt, the left ventricular output at rest was so much increased that their systemic blood flow did not differ significantly from that of control lambs. However, a lower systemic blood flow was found in shunt than in control lambs in a fasted state (4) or during exercise (20). A lower rate of systemic O₂ supply was also found, consistent with the lower systemic blood flow. An impaired systemic O₂ supply may lead to an O₂ deficit in peripheral tissues, such as skeletal muscles, with a consequent increase in anaerobic metabolism. An increase in lactate production may then be expected.

In previous studies, Gratama et al. (18) did not find a difference in arterial lactate concentrations between shunt and control lambs. However, the arterial lactate concentration is the result of production and utilization, both of which can be increased while the arterial lactate concentration remains unchanged. The aim of this study, therefore, was to investigate, with the aid of ¹³C-labeled lactate, lactate production and utilization in conscious lambs with an aortopulmonary shunt and in control lambs, both at rest and during moderate exercise on a treadmill.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We studied 16 lambs, of mixed breed, with documented dates of birth (7 wk old). Lambs were assigned to two groups: eight lambs with an aortopulmonary shunt and eight lambs without a shunt. Until the day of study, each lamb remained with its mother. Surgical preparation, catheter care, and antibiotic administration were performed as described previously (36). In the shunt lambs, a Goretex conduit (6 mm ID; W. L. Gore, Flagstaff, AZ) was sutured between the descending aorta and the main pulmonary artery. Catheters were inserted into the aorta, the pulmonary artery, the right ventricle (only in the shunt lambs), and the right and left atria. Precalibrated electromagnetic flow transducers (10-15 mm ID; Skalar Medical, Delft, The Netherlands) were placed around the ascending aorta just above the coronary arteries and around the pulmonary artery proximal to the conduit in the shunt lambs, and around the pulmonary artery only in the control lambs. The experiments were performed with the approval of the Ethics Committee on Animal Experiments of our university.

In the week before surgery and from 2 days after surgery, the lambs were familiarized with running on a motor-driven treadmill (Laufergotest Junior, Erich-Jaeger, Hoechberg, Germany) during one short daily run. No training effect was pursued. The lambs ran freely on the treadmill without coercive measures. During the experiments, an external workload corresponding to 50% of the peak O₂ consumption (O_{2 peak}) was used. The O_{2 peak} of each lamb was determined during a graded treadmill test 1 wk after surgery, as described previously (21). In brief, resting values were determined with the lamb freely standing on the treadmill. Systemic and pulmonary blood flows were measured with the electromagnetic flow transducers. At the same time, aortic, pulmonary arterial, and left atrial pressures were measured. Furthermore, blood samples were withdrawn from the aortic and mixed venous catheters, i.e., from the pulmonary arterial catheter in the control lambs and the right ventricular catheter in the shunt lamb. O₂ saturation was determined in both samples, and hemoglobin concentration was determined only in the sample from the aorta. After O₂ consumption (O₂) had been calculated with the Fick formula, the lamb was subjected to a running speed of 3.5 km/h. After 3 min, the measurements were repeated and O₂ was calculated again. Immediately after the blood samples had been collected, the workload was increased by setting the incline at 4%. After another 3 min, the measurements were repeated, and so on until the maximum incline of 15% was reached. Thereafter, the treadmill speed was increased by 0.5 km/h while the incline of 15% was maintained. The graded treadmill test in the shunt lambs was modified by starting at a running speed of 2.5 instead of 3.5 km/h. This modification was necessary because otherwise the speed and inclination corresponding to 50% of O_{2 peak} could not be determined.

Experimental Protocol

To determine the lactate turnover rate, [1-¹³C]lactate was administered according to the prime-dose, constant-rate infusion technique (34). Before the infusion of [1-¹³C]lactate was started, blood samples were withdrawn from the aorta for determination of the lactate concentrations and the isotope ratio (¹³C-to-¹²C ratio) of lactate to determine the natural abundance of ¹³C in lactate. A priming dose of 15.6 mg/kg [1-¹³C]lactate (99 atom% ¹³C; Tracer Technologies, Somerville, MA) was administered over 10 min into the right atrial catheter, followed by a constant-rate infusion (model 2620 pump; Harvard, Millis, MA) of 0.156 mg · min¹ · kg¹ [1-¹³C]lactate (34, 39). During a steady state, two blood samples (at 30 and 40 min after the start of the infusion of the priming dose) were obtained from the aorta for determination of the lactate concentration and the isotope ratio of lactate. At the same time points, blood samples were withdrawn with a heparinized syringe from the aortic and mixed venous catheters. O₂ saturation was determined in all samples. Hemoglobin concentration, pH, PCO₂, PO₂, and plasma HCO₃ concentration, as well as epinephrine and norepinephrine concentrations, were determined in the aortic sample. Immediately after collection of these blood samples at rest, blood flow to the myocardium was determined by using radioactive microspheres labeled with either ¹⁴¹Ce, ¹¹³Sn, ¹⁰³Ru, or ⁹⁵Nb (NEN-Trac; Du Pont, Biotechnology Systems, Wilmington, DE). These were injected into the left atrium while a reference sample was withdrawn over 1.25 min at a rate of 6 ml/min with a Harvard pump from the aortic catheter into a preweighed, heparinized syringe (24, 36).

Ten minutes after the injection of the microspheres, speed and inclination of the treadmill were set to values that would impose a workload corresponding to ~50% of O_{2 peak}. The lambs had to run at this workload for 30 min. At 10-min intervals, blood flows and pressures were measured, and blood samples were withdrawn for determination of O₂ saturation, hemoglobin, pH, PCO₂, PO₂, and plasma HCO₃, lactate, and isotope ratio of lactate, epinephrine, and norepinephrine, as described for the resting period. At a time between 20 and 30 min of exercise, microspheres labeled with an isotope different from the one used during the resting period were injected. After the last blood sample had been withdrawn, the treadmill was stopped and the lamb was allowed to recover.

Measurements and Calculations

Immediately after they were withdrawn, the blood samples were mixed with sodium fluoride to stop glycolysis and were then put in ice. For the determination of the concentration of lactate, a part of the blood was deproteinized with cold 18% perchloric acid (2:1, vol/vol) and centrifuged. The protein-free supernatant was removed and neutralized with a potassium hydroxide-morpholinopropanesulfonic acid mixture. Lactate was determined in duplicate by an enzymatic method (5).

For the determination of the isotope ratio of lactate, fatty acids were removed from the plasma by extraction with chloroform. The lactate was then extracted with diethyl ether-ethyl acetate and dried under nitrogen. [1-¹³C]lactate was determined as a derivative of n-heptafluorobutyrate anhydride-n-butylamine (3) by gas chromatography-mass spectrometry. We used a model 5890 gas chromatograph (Hewlett-Packard, Palo Alto, CA) interfaced to a VG Trio-2 quadrupole mass spectrometer (Fisons Instruments, Manchester, UK). The mass spectrometer was used in the chemical ionization mode. Single-ion monitoring was carried out at m/e 359 (m + 0) and m/e 360 (m + 1), corresponding to [m + NH₄]⁺ of the unlabeled and the labeled lactate, respectively. Standards containing 0.0, 2.5, 5.0, and 7.5% [1-¹³C]lactate were prepared by diluting natural lactate with [1-¹³C]lactate to obtain a calibration graph of isotope ratio vs. molar fraction (F; r = 0.998). The F of [1-¹³C]lactate of the blood samples was calculated from this calibration graph.

Plasma epinephrine and norepinephrine concentrations were determined by HPLC with electrochemical detection (32). After sample collection, the blood was centrifuged at 4°C. The thrombocyte-poor plasma was fortified with the antioxidant glutathion and stored at 20°C pending determination of epinephrine and norepinephrine levels.

Left-to-right shunt flow was obtained by subtracting systemic from pulmonary blood flow. Left-to-right shunt fraction was calculated by dividing left-to-right shunt flow by pulmonary blood flow. Blood O₂ concentration was calculated as the product of O₂ saturation, hemoglobin concentration, and a hemoglobin O₂-binding capacity of 1.36 ml/g (29). Systemic O₂ supply was calculated as the product of arterial O₂ concentration and systemic blood flow. Whole body O₂ was calculated by multiplying the mixed arteriovenous O₂ concentration difference by systemic blood flow.

At rest, the lactate turnover rate [rate of appearance (R_a), in micromoles per minute per kilogram] was calculated with a steady-state equation according to Steele et al. (34) where F_i is the molar fraction of [1-¹³C]lactate in the infusate, F_ao the molar fraction of [1-¹³C]lactate in the aorta at steady state, and I is the rate of tracer lactate infusion (in micromoles per minute per kilogram). At rest, the rate of disappearance (R_d) equals R_a.

During exercise, R_a was calculated with a non-steady-state equation according to Steele (33) where V is the volume of the lactate pool, which is assumed to be 50% of total body water (26). Total body water, determined by means of a single-injection dilution technique with deuterium oxide (²H₂O), was 780 and 809 ml/kg for control and shunt lambs, respectively (19). _ao is the mean concentration of lactate (in micromoles per milliliter) of the consecutive aortic samples, F_ao is the difference in F of [1-¹³C]lactate of the consecutive aortic samples, t is the time (in minutes) between the two samples, and _ao is the mean molar fraction of [1-¹³C]lactate of the consecutive aortic samples.

R_d then follows from where C_ao is the difference in concentration of lactate in the consecutive aortic samples. The metabolic clearance rate (MCR, in milliliters per minute per kilogram) was calculated by dividing R_d by the blood lactate concentration.

Statistical Analysis

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Hemodynamic Data

O_{2 peak} experiment. The maximal absolute external workload achieved by the shunt lambs was significantly lower than that of control lambs (speed, 3.6 ± 0.1 vs. 3.9 ± 0.1 km/h, respectively; P < 0.05; inclination, 12 ± 1 vs. 15 ± 0 %, respectively; P < 0.001). Hemodynamic data, O₂-related variables, and plasma catecholamine concentrations at rest and during maximal exercise (O_{2 peak}) are shown in Table 1. Systemic blood flow, systemic O₂ supply, and O₂ at rest were lower in shunt than in control lambs. O_{2 peak} tended to be lower in the shunt than in the control lambs. There was no significant difference in plasma epinephrine and norepinephrine concentrations between the two groups of lambs, either at rest or during maximal exercise.

Moderate exercise (50% of O_{2 peak}). On the day of the study, there were no statistically significant differences between the control and the shunt lambs in age (45 ± 1 vs. 44 ± 1 days, respectively) and weight (13.4 ± 0.8 vs. 12.6 ± 0.6 kg, respectively). Blood gases were normal in both groups of lambs (Table 2). During moderate exercise, the relative workload was similar in control and shunt lambs (Fig. 1G). The absolute external workload, however, was lower in shunt than in control lambs (speed, 2.9 ± 0.1 vs. 3.4 ± 0.1 km/h, respectively; P < 0.01; 0% inclination in both groups of lambs). The mean left-to-right shunt fraction in the shunt lambs was 55 ± 3% of pulmonary blood flow at rest and decreased to 52 ± 3 (P < 0.05), 46 ± 3, and 42 ± 4% of pulmonary blood flow at 10, 20, and 30 min of moderate exercise. The left-to-right shunt led to significant hemodynamic and O₂-related differences between shunt and control lambs (Table 2, Fig. 1). The hemodynamic response to exercise was similar in shunt and control lambs, and most of the differences between the two groups at rest persisted during exercise (Table 2, Fig. 1). Systemic blood flow, systemic O₂ supply, and O₂ were lower in shunt than in control lambs, both at rest and during moderate exercise (Fig. 1, B-D). There was no difference in arterial O₂ saturation or arteriovenous O₂ concentration difference between shunt and control lambs either at rest (2,012 ± 116 vs. 2,193 ± 131 µmol/l, respectively) or during exercise. There was no difference in plasma epinephrine or norepinephrine concentrations between shunt and control lambs. In both groups of lambs, plasma epinephrine and norepinephrine concentrations increased during exercise compared with resting levels (Fig. 1, E and F).

View this table: [in this window] [in a new window]   Table 2.   Hemodynamic data, blood gases, and oxygen saturation at rest and during moderate exercise (50% of O2 peak)

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Hemodynamic data, O2-related variables, and plasma catecholamine concentrations of control (, n = 8) and shunt (, n = 8) lambs at rest and during moderate exercise [50% of peak O2 consumption (O2 peak)]. A: heart rate; B: systemic blood flow; C: systemic O2 supply; D: O2 consumption (O2); E: epinephrine; F: norepinephrine; G: O2 peak. Data are means ± SE. * Control vs. shunt, P < 0.05. § Rest vs. exercise, P < 0.05.

Lactate Kinetics

O_{2 peak} experiment. The arterial lactate concentrations at rest and during exercise at each level of a similar absolute external workload were much the same in shunt and control lambs (Fig. 2). The lactate concentration at O_{2 peak} was lower in shunt than in control lambs.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Top: arterial lactate concentration at rest and at different levels of exercise (levels 1-8; duration at each level, 3 min) during O2 peak experiment of control (, n = 8) and shunt (, n = 8) lambs. Data are means ± SE. * Control vs. shunt, P < 0.05. Note that first measurements in control lambs start at level 3. Bottom: speed and incline of treadmill at each level of exercise.

Moderate exercise (50% of O_{2 peak}). The arterial lactate concentrations at rest and during moderate exercise were similar in shunt and control lambs (Fig. 3). In both groups of lambs, there was an increase in arterial lactate concentration during exercise in comparison with resting levels. Lactate R_a and R_d did not differ between shunt and control lambs, either at rest (12.97 ± 1.71 vs. 12.55 ± 1.25 µmol · min¹ · kg¹, respectively) or during exercise at a similar relative workload (Fig. 3). R_d was always lower than R_a during exercise, which is consistent with the increase in arterial lactate concentration. MCR of lactate was similar in shunt and control lambs, both at rest (11.35 ± 0.95 vs. 10.94 ± 1.09 ml · min¹ · kg¹, respectively) and during exercise.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   A: arterial lactate concentration (), rate of appearance (Ra; ), and rate of disappearance (Rd; ) in control lambs (n = 8) at rest and during exercise at 50% O2 peak. B: arterial lactate concentration (), Ra (), and Rd () in shunt lambs (n = 8) at rest and during exercise at 50% O2 peak. Data are means ± SE. § Rest vs. exercise, P < 0.05.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   A: correlation between Ra and O2 at rest and during exercise in control () and shunt () lambs; r = 0.42; P < 0.01; n = 39; y = 9.12 + 0.03 · x. B: correlation between Ra and systemic blood flow; r = 0.57; P < 0.0001; n = 39; y = 1.44 + 0.13 · x. C: correlation between Ra and systemic O2 supply; r = 0.61; P < 0.0001; n = 39; y = 2.10 + 0.03 · x. Both rest and exercise data are included.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We found that the arterial lactate concentration and the lactate production rate were similar in shunt and control lambs, both at rest and during exercise on a treadmill at a similar relative workload of 50% of their O_{2 peak}. Also, the R_d and the MCR of lactate were similar in both groups of lambs. Hence, the shunt lambs have similar lactate kinetics as do the control lambs but at a lower level of absolute external workload. In both groups of lambs, the arterial concentration of lactate increased during exercise, compared with resting levels, due to a larger increase in the R_a than in the R_d. Despite a lower systemic O₂ supply in the shunt lambs compared with that in the control lambs, both at rest and in a more pronounced manner during exercise, the lactate production rate in shunt lambs was not higher than in control lambs. At rest, the shunt lambs have apparently adapted themselves to the lower systemic O₂ supply by consuming less O₂; this is in agreement with a metabolism that is turned down. The lower systemic O₂ in the shunt lambs may be caused by less growth or by restrictions of external work. A similar relationship between a low systemic blood flow and O₂ supply, and a limited use of skeletal muscles, has been described in humans and in beagles for situations such as immobilization, bed rest, or senescence (6, 23).

We found a direct relationship between systemic O₂ and lactate production in both groups of lambs; this is in agreement with earlier studies in dogs (27), rats (13), and humans (31). Lactate production was higher when more work was done and more energy was expended. Shunt and control lambs have equal lactate production and R_d at rest and during exercise at a similar fraction of the exercise capacity (Fig. 3). However, from the direct relationship between lactate production and systemic O₂, it may be speculated that, at a similar absolute external workload, the lactate production rate will be higher in shunt than in control lambs. A higher lactate production in the shunt lambs may be the result of a different state of fitness, since it is known that, in human beings with different levels of training, lactate kinetics are similar when exercising at a similar relative workload but are indeed different when exercising at a similar absolute external workload (12).

A reduced systemic O₂ supply and a higher lactate production by skeletal muscles has been observed in human beings with congestive heart failure in comparison with healthy men when exercising at a similar absolute external workload (1, 8, 14, 30). The reduced systemic O₂ supply and the expected higher lactate production in shunt lambs, when they exercise at a similar absolute external workload, may have been caused by the inability of the heart to perfuse skeletal muscles adequately or by abnormalities in the skeletal muscles themselves. In a previous study, Gratama et al. (20) did indeed find a decreased blood flow to skeletal muscles in shunt compared with control lambs during exercise at 80% of their O_{2 peak}. This is in agreement with the lower skeletal muscle blood flow as described in rats, with an arteriovenous fistula, during exhaustive exercise on a treadmill (16) and in patients with congestive heart failure (38). Abnormalities in the skeletal muscles themselves were found in human beings with congestive heart failure, who have reduced exercise capacity (30, 37), depletion of oxidative enzymes (8, 14), fibrosis (2), and myofibrillar breakdown (8, 14), probably as a result of a limited use of skeletal muscles.

One should consider whether the maximal exercise capacity was achieved in both groups of lambs while they exercised at a different O_{2 peak} protocol. Some determinants of maximal exercise point to a similar maximal exercise level, whereas others suggest that the exercise at O_{2 peak} might have been lower in the shunt than in the control lambs. The heart rates at O_{2 peak} in the shunt lambs were lower than in the control lambs, although not statistically significant different from the results described by Gratama et al. (21) in a study in which both control and shunt lambs exercised at a similar O_{2 peak} protocol. Furthermore, in both our groups of lambs, the maximal heart rates during O_{2 peak} were even higher than the heart rates achieved during infusion with 0.1 µg · min¹ · kg¹ isoproterenol, which are a good approximation of the heart rate at maximal exercise (22). The similar mixed venous O₂ saturations in the shunt and control lambs give evidence that the shunt lambs did exercise until exhaustion. On the other hand, the arterial lactate concentrations and the plasma concentrations of epinephrine and norepinephrine tended to be lower in the shunt than in the control lambs. Therefore, we cannot rule out the possibility that the shunt lambs had exercised at a lower level than the control lambs, both during the maximal and the moderate exercise tests. A difference in the moderate-exercise experiments between shunt and control lambs might have occurred to the extent that the shunt lambs were exercising at somewhat <50% of O_{2 peak} and the control lambs were exercising at somewhat >50% of O_{2 peak}.

We found a decreased exercise capacity in shunt lambs compared with control lambs, as has also been shown in previous studies (21). In children with congenital heart defects, a decreased exercise capacity has also been described (11, 15, 17). A decreased exercise capacity may be caused by a lower systemic blood flow, a lower O₂ transport capacity to the muscles, and/or a lower arteriovenous O₂ concentration difference over the skeletal muscles. We found a similar arteriovenous O₂ concentration difference, but we did indeed find a lower systemic blood flow in shunt lambs compared with control lambs and, as a consequence, a lower systemic O₂ supply.

Whether the decreased exercise capacity in shunt lambs can be improved by physical training remains to be investigated. Physical training leads to cardiovascular adaptations with an increase in maximal O₂, stroke volume, and systemic blood flow (6, 7). Furthermore, endurance training may produce major adaptations in skeletal muscles (25). These adaptations to training in skeletal muscles result in an increased oxidative capacity and a lower lactate production during exercise of any given intensity (25). In patients with congestive heart failure, bicycle training has been shown to improve exercise tolerance and to decrease their complaints (10, 35).

In conclusion, we have found that lambs with an aortopulmonary shunt have the same lactate kinetics as control lambs at rest and during moderate exercise performed at a similar fraction of their O_{2 peak}, despite a lower systemic O₂ supply. We speculate that the shunt lambs have adapted themselves to the decreased systemic O₂ supply through consuming less O₂ by limiting the performance of external work.