Carnitine-Related Alterations in Patients With Intermittent Claudication
Indication for a Focused Carnitine Therapy
Background Carnitine metabolism is altered in peripheral arterial disease. l-carnitine supplementation may correct these alterations and improve walking performance.
Methods and Results Plasma levels of carnitine and its esters were measured at rest and after maximally tolerated exercise in 22 claudicant patients and 8 normal subjects. One week later, this protocol was repeated in patients after random administration of placebo or l-carnitine (500 mg IV as a single bolus). Two groups of patients emerged. In 10 patients (group IC1), the plasma level of acetylcarnitine at rest was 3.7±0.2 μmol/L and increased significantly (P<.01) at maximally tolerated exercise. In 12 patients (group IC2), the resting level of plasma acetylcarnitine was elevated (7.9±0.7 μmol/L, P<.01) and decreased with exercise. Furthermore, group IC2 patients had a significantly lower walking capacity than group IC1 patients. In both groups, placebo did not affect the metabolic profile, nor did it improve exercise performance. Conversely, after l-carnitine administration, all but one patient in group IC2 (n=7) showed an increase in plasma acetylcarnitine concentration during exercise versus the decrease observed without l-carnitine. This metabolic effect was accompanied by a significant increase (P<.01) in walking capacity. Interestingly, in group IC1 patients (n=5), l-carnitine neither improved walking capacity nor modified the metabolic profile. Statistical analysis showed that changes in walking capacity with l-carnitine treatment were influenced exclusively by exercise-induced changes in plasma acetylcarnitine.
Conclusions In patients with intermittent claudication, assessment of plasma acetylcarnitine at rest and after exercise may be a means to select a target population for l-carnitine therapy.
l-Carnitine administration improves walking capacity in patients with peripheral arterial disease without affecting regional hemodynamics.1 This effect is associated with an increase in total carnitine content in the ischemic muscle and a decrease in plasma lactate concentration on exercise, the latter probably resulting from a reduction in the ratio of acetyl–coenzyme A (acetyl-CoA) to CoA. Indeed, carnitine is a physiological modulator of the mitochondrial pool of acetyl-CoA,2 and acetyl-CoA is an end-product inhibitor of pyruvate dehydrogenase.3 Through the action of carnitine acetyltransferase (CAT), carnitine depletes acetyl-CoA and releases free CoA and acetyl carnitine, which, unlike CoA esters, may be transported out of the cell and released in the bloodstream. This acyl scavenging process, which requires adequate availability of carnitine, becomes crucial under conditions of limited oxygen availability, when the shortage of free CoA limits the mitochondrial oxidation of both pyruvate and α-ketoglutarate, and the concurrent accumulation of CoA esters results in inhibition of the enzymes involved.4 5 6 Indeed, increased levels of short-chain acylcarnitines, mostly acetylcarnitine, occur in muscle and plasma of normal subjects performing maximal exercise.7 8 9 10
In patients with peripheral arterial insufficiency at stage II of Fontaine’s classification (ie, claudication on effort, no pain at rest, and/or trophic lesions in the affected leg), increased carnitine esterification may occur even at rest.11 12 In particular, patients with the lowest walking capacity exhibit the highest concentration of short-chain acylcarnitines at rest in both plasma11 and ischemic skeletal muscle.12 This suggests that the more severe the ischemic disease, the higher the accumulation of CoA esters in the affected tissues and consequently the greater the amount of carnitine required for their removal. Although this does not affect carnitine levels in muscle, it may lead in some cases to a reduced availability of free carnitine to meet the increased metabolic demand produced by walking.
The exercise-induced changes in plasma levels of carnitine and its esters observed in the present study before and after l-carnitine seem to support this hypothesis and indicate that in a subgroup of patients with altered carnitine metabolism, l-carnitine supplementation restores a normal response of plasma acetylcarnitine to exercise with a concomitant improvement in walking capacity.
Thirty male subjects were enrolled in this study, 22 subjects (mean age, 59.0±1.1 years) with peripheral arterial disease at stage II of Fontaine’s classification and 8 healthy, untrained, age-matched control subjects. All subjects gave informed consent before the study, which was approved by the institutional review committee. The diagnosis of peripheral arterial disease was established in advance from history and physical and Doppler examination. Patients who presented an ankle-to-brachial systolic pressure index at rest <0.90, which decreased by at least 15% after maximal treadmill exercise, were selected. Patients with diabetes mellitus and those whose exercise was limited primarily by coronary artery disease, congestive heart failure, pulmonary disease, and severe hypertension were excluded from the study. A pretrial washout period of 2 weeks was allowed in all cases. Diuretics and oral antiplatelet agents were the only drugs allowed. All subjects were on a low-fat, low-cholesterol diet. Tests were carried out in the morning, after an overnight fast in a quiet room at a constant temperature of 21±1°C. All subjects underwent graded treadmill exercise, starting from an initial stage of 2.5 mph and 3% grade with subsequent 3% grade increases every 2 minutes up to a maximum of 15%. Patients with peripheral arterial disease walked until claudication pain became intolerable (maximal walking capacity); control subjects exercised until they reached their maximal heart rate–targeted end points. Arm blood pressure by auscultation and heart rate by ECG were measured at 1-minute intervals. The systolic pressures in the posterior tibial artery of the claudicant limb and the right brachial artery were obtained by Doppler ultrasound. Ankle-to-brachial systolic pressure ratio was measured at rest and 5 minutes after the end of exercise. In patients, the initial claudication time, ie, the time walked until the onset of typical claudication pain in the affected leg (in seconds), also was measured. To ensure that the patients had a stable walking capacity, two treadmill tests were conducted during the washout period, and only those patients exhibiting a change in maximal walking capacity of no more than 20% were enrolled in the study. Of the original 25 patients, three did not meet this inclusion criterion and thus were excluded from the study.
Venous blood samples (5 mL) for the assay of carnitine and its esters were obtained by a catheter placed in an antecubital vein. Resting blood samples were collected after subjects had been standing for 5 minutes. Subjects then performed a treadmill test as described above, and the time to initial claudication pain was measured. The test terminated when patients could not continue because of unbearable pain in the affected leg (maximal walking time). At this point, blood samples were drawn for the determination of carnitine and its esters. A week later, claudicant patients randomly received placebo or l-carnitine (500 mg IV as a single bolus). Resting blood samples were drawn 30 minutes after the injection, exercise was started, and the time to initial claudication pain was measured. During this second test, each patient was stopped at exactly the same maximal walking time as before treatment, and blood samples were drawn, except for 3 patients in the placebo group who did not reach the same maximal walking time during the second test as under control conditions. This protocol ensured that after treatment, plasma concentrations of carnitine and its esters were measured for each patient at the same walking time and under exactly the same experimental conditions as before treatment (ie, when the patient stopped exercise). This procedure, however, did not allow us to measure the maximal walking time during the second treadmill test (ie, after treatment). Therefore, we used the time to initial claudication pain as a performance index. This parameter, often used for functional evaluation of claudicant patients, has been considered a more reliable indicator of walking performance than maximal walking time.13 14
Both assessment of treadmill performance and carnitine assay were performed by physicians unaware of the treatment.
Plasma carnitine and acylcarnitines were measured as follows. Sodium heparin (1.4 USP U/mL) was added to blood samples, which were then centrifuged. The resultant plasma was stored at −70°C until required for carnitine determinations, which were carried out by a radioenzymatic procedure.15 Perchlorate fractionation of plasma was used initially to separate long-chain acylcarnitines (pellet) from acid-soluble carnitine comprising free carnitine and acylcarnitines with short and medium chain lengths (carbons <10). The protein precipitate was washed three times to avoid contamination by retained acid-soluble carnitine.16 Long-chain acylcarnitine content was estimated as free carnitine released after alkaline hydrolysis of the pellet. Free carnitine and acetylcarnitine were assayed directly in the pooled supernatants by radioenzymatic assays.15 17 The direct assay of acetylcarnitine proved to be necessary for the analysis of the subjects treated with l-carnitine. In fact, owing to the large concentrations of free carnitine, the esterified fraction became too small to be estimated as the difference between total and free carnitine. The acid-soluble fraction was further characterized by the high-performance liquid chromatography separation and quantification of the different short-chain acyl esters.18
Values are reported as mean±SEM. Differences between groups were compared with Student’s t test for unpaired data. Comparison of time to initial claudication pain before and after l-carnitine administration was done with Student’s t test for paired data. Stepwise multiple regression analysis was used to determine which variable predicts the response to l-carnitine therapy. In this context, change in time to initial claudication pain with drug therapy was the dependent variable; the plasma resting concentration of acetylcarnitine and the change in acetylcarnitine with exercise served as independent variables.
Control subjects walked longer (1640.9±101.1 versus 342.4±51.8 seconds, P<.01) and reached higher values of heart rate (157.6±1.7 versus 98.8±3.3 beats per minute; P<.01) than patients with intermittent claudication. At peak exercise, systolic and diastolic pressures were similar in the two groups (162.5±9.9 versus 169.8±4.7 and 98.1±2.8 versus 95.7±2.4 mm Hg, respectively).
In patients with intermittent claudication, the ankle-to-arm systolic pressure ratio was 0.65±0.02 at rest and decreased to 0.52±0.02 (P<.01) after exercise. Time to initial claudication pain was 181.4±26.1 seconds, and maximal walking capacity was 342.4±51.8 seconds. Control subjects did not exhibit any symptoms that limited exercise.
In normal subjects, plasma free carnitine and long-chain acylcarnitine concentration did not change with exercise, whereas acetylcarnitine increased from the resting value of 3.0±0.4 to 4.5±0.3 μmol/L (P<.05). Patients with intermittent claudication showed resting values of free carnitine and long-chain acylcarnitines similar to those observed in control subjects, whereas acetylcarnitine was higher than in the control group (6.0±0.6 μmol/L, P<.01). In the patient group, there was no change in the plasma concentration of carnitine and its esters with exercise. However, a correlation analysis between the resting levels of acetylcarnitine and its changes with exercise revealed a negative relationship (Fig 1⇓), indicating that patients with the lowest plasma levels of acetylcarnitine at rest had the greatest increase of this ester with exercise.
As Fig 2⇓ shows, all control subjects showed an increase in plasma acetylcarnitine concentration at peak exercise. An analogous response occurred only in 10 claudicant patients (designated group IC1); these showed a significant increase in plasma acetylcarnitine from 3.7±0.2 to 5.2±0.5 μmol/L (P<.01). On the contrary, all 12 remaining patients showed a decrease in plasma acetylcarnitine at the maximal tolerated walking time, although the decrease was minimal in some cases. In these patients (designated group IC2), plasma acetylcarnitine decreased from the resting value of 7.9±0.7 to 6.1±0.6 μmol/L (P<.01) with exercise. When the patients were divided into groups IC1 and IC2, we observed that the former had resting levels of acetylcarnitine similar to those of control subjects, whereas in group IC2 patients, the resting levels of this ester were higher than those observed in both control subjects (P<.01) and group IC1 patients (P<.01). Consequently, the ratio of acetylcarnitine to free carnitine (an index of the distribution of total carnitine between free and acylcarnitine) was significantly higher in group IC2 than IC1 patients (0.19±0.01 and 0.11±0.01, respectively; P<.01). Furthermore, group IC1 patients had a higher time to initial claudication pain and a higher maximal walking capacity than group IC2 patients. The clinical characteristics and plasma venous concentrations of carnitine and its esters in control subjects and patients in groups IC1 and IC2 are reported in Tables 1⇓ and 2⇓, respectively.
Placebo, administered to 5 group IC1 and 5 group IC2 patients, did not modify exercise-induced changes in carnitine and its esters. In particular, in group IC1 patients, plasma acetylcarnitine increased from the resting value of 3.7±0.3 to 5.1±0.7 μmol/L (P<.05) with exercise under control conditions and from 3.6±0.3 to 5.2±0.6 μmol/L (P<.05) after placebo. In group IC2 patients, plasma acetylcarnitine decreased with exercise from 8.5±1.2 to 6.8±1.2 μmol/L (P<.05) under control conditions and from 9.6±1.0 to 7.7±1.1 μmol/L (P<.05) after placebo. Similarly, placebo did not modify the time to initial claudication pain in either group.
l-Carnitine was given to 5 group IC1 and 7 group IC2 patients. In group IC1, treatment did not affect the exercise-induced plasma changes in carnitine and its esters, nor did it modify the exercise performance. In particular, 3 patients showed a decrease in the time to initial claudication pain compared with pretreatment values, and 2 showed an increase not exceeding 4.6%. In group IC2 patients, after l-carnitine administration, resting plasma concentrations of free carnitine and long-chain acylcarnitines, which were 147.2±22 and 3.7±0.4 μmol/L, respectively, were not modified by exercise. On the contrary, plasma acetylcarnitine levels increased with exercise from 6.2±1.2 to 7.3±1.1 μmol/L (P<.05). Indeed, at the same workload as before l-carnitine administration, all but 1 group IC2 patient showed an increase in plasma short-chain acylcarnitines with exercise as opposed to the decrease observed without l-carnitine. In these patients, normalization of the plasma acetylcarnitine response to exercise induced by l-carnitine paralleled an improvement in walking performance. After l-carnitine administration, no group IC2 patient experienced unbearable claudication pain at the same walking time as before treatment. Moreover, in the 6 patients in whom l-carnitine normalized the plasma acetylcarnitine response to exercise, we observed an improvement in time to initial claudication pain ranging from 11.1% to 85.8%. In the remaining patient, treatment did not modify exercise performance. Therefore, in all of group IC2, l-carnitine improved the time to initial claudication pain from the control value of 94.8±18 to 125.0±24 seconds (P<.01). These data strongly suggest that in patients with intermittent claudication, changes in plasma acetylcarnitine concentration with exercise may predict the response to l-carnitine. However, we also applied a stepwise multiple regression analysis and found that changes in the time to initial claudication pain with l-carnitine were influenced only by exercise-induced changes in acetylcarnitine (F=6.700, r=.633, P=.027), not by the plasma concentration of these esters at rest (F=1.558, r=.500, P=.262).
l-Carnitine treatment had no effect on exercise-induced changes in heart rate and blood pressure in either group.
Carnitine metabolism is altered in patients with peripheral arterial disease.11 12 19 The results of the present study indicate that patients with intermittent claudication can be divided into two groups that are distinguished by differences in plasma carnitine metabolism. Like control subjects, patients in one group (IC1) showed a significant increase in plasma acetylcarnitine concentration at peak exercise. Patients in the other group (IC2) had a decrease in this ester with exercise. Furthermore, group IC1 patients had a plasma concentration of acetylcarnitine at rest similar to that in control subjects, whereas group IC2 patients had high resting levels of this ester. These metabolic findings seem to be related to the severity of functional impairment because group IC2 patients had a significantly lower walking capacity than group IC1 patients.
The distribution between carnitine and acylcarnitines is influenced by changes in tissue metabolism under a variety of conditions.20 21 22 During maximal exercise in normal subjects, there is a redistribution of free to esterified carnitine, as indicated by the decrease in free carnitine and the concomitant increase in short-chain acylcarnitines (represented mostly by acetylcarnitine) in both plasma and skeletal muscle.8 9 23 The increased production of acetylcarnitine by the working muscle is well documented and reflects the ability of free carnitine to buffer the excess acetyl-CoA.23 On the contrary, the tissue source of plasma acetylcarnitine is not known. It may be of muscle origin, in which case the plasma increase observed in the control group and group IC1 patients would reflect sufficient availability of free carnitine that, coupled with normal CAT activity, ensures the removal of excess acetyl-CoA during exercise. Alternatively, free carnitine could be released from muscle during exercise, acylated at a site other than the contracting muscle (presumably the liver), and then released in plasma.24
On the other hand, it is difficult to explain the reduction in plasma acetylcarnitine observed in group IC2 patients at peak exercise. Interestingly, exercise-induced reduction in short-chain acylcarnitine has been observed in the plasma and muscle of claudicant patients.11 12 In our study, group IC2 patients had a higher resting concentration of plasma acetylcarnitine and a lower walking capacity than group IC1 patients. This finding is consistent with the previous observation that the muscle concentration of short-chain acylcarnitine at rest negatively correlates with subsequent exercise performance.12 Thus, it is reasonable to assume that in the more advanced stage of intermittent claudication, there is increased formation of short-chain acylcarnitines (represented mostly by acetylcarnitine) from the muscle at rest. In other disease states, the continued generation of acylcarnitines has metabolic consequences. For example, inherited disorders of organic acid metabolism are associated with the chronic export of carnitine from tissue in the form of acylcarnitines, yet an increased amount of carnitine may be required to buffer the acetyl-CoA pool under the impaired metabolic conditions.25 Consequently, the amount of this factor available may not be able to meet the increased metabolic demand. The reduction in plasma acetylcarnitine observed in group IC2 patients with exercise could reflect such a condition. In effect, group IC2 patients had a ratio of acetylcarnitine to free carnitine that was significantly higher than that of group IC1 patients. This, in addition to the negative relationship between the resting concentration of plasma acetylcarnitine and its changes with exercise, strongly suggests that the higher the resting concentration of this ester, the lower the availability of free carnitine to generate it with exercise. That a reduction in plasma acetylcarnitine during exercise may reflect a reduced formation of this ester in the affected muscle of group IC2 patients seems to be supported by the finding that exercise-induced reduction in the muscle content of short-chain acylcarnitine has been observed in the most advanced form of intermittent claudication.12 Furthermore, more direct evidence that the altered response of acetylcarnitine to exercise may depend on a reduced availability of free carnitine to form this ester is provided by the results observed after l-carnitine administration. In group IC2 patients, l-carnitine supplementation induced an increase in plasma acetylcarnitine concentration with exercise, as opposed to the decrease before drug administration. On the contrary, in group IC1 patients who had a lower resting concentration of plasma acetylcarnitine that increased with exercise, l-carnitine treatment did not modify the metabolic profile.
However, reduced availability of carnitine could be responsible at most for the lack of increase in acetylcarnitine during exercise but not for the decrease observed in this and previous studies.11 12 Considering the key role played by CAT in the acetylcarnitine formation, the possibility that impaired activity of this enzyme may concur to alter the acetylcarnitine response to exercise may not be ruled out.
Alternatively, the reduction in plasma concentration of acetylcarnitine during exercise may reflect an increased use of this ester. This implies that the ischemic muscle acts as an acetylcarnitine scavenger during exercise. To the best of our knowledge, however, there is no evidence of such a condition in the skeletal muscle or other tissues in either the normal or the disease state. In addition, increased use of acetylcarnitine by the ischemic muscle during exercise would require optimal activity of both the Krebs’ cycle and the respiratory chain that is unlikely under the hypoxic conditions during claudication.
Although the mechanisms responsible for the metabolic findings observed in the present study remain to be clarified, changes in plasma acetylcarnitine with exercise appear to be a marker of patients who are responsive to carnitine therapy. In group IC2 patients, l-carnitine induced a significant improvement in time to initial claudication pain. In group IC1 patients, treatment did not modify exercise performance. Statistical analysis confirmed that changes in time to initial claudication pain were related to exercise-induced changes in acetylcarnitine.
A previous study demonstrated that chronic treatment with l-carnitine improves walking performance in patients with intermittent claudication.1 This beneficial effect was associated with an increase in short-chain acylcarnitine content in the ischemic muscle and a reduction in lactate concentration in the venous blood leaving the exercising affected limb.1 Thus, carnitine supplementation may be critical for removal of acetyl-CoA excess and improvement of oxidative metabolism in patients with peripheral arterial disease. The results of the present study indicate that improvement in walking performance by l-carnitine is achieved only in patients with an abnormal response of plasma levels of acetylcarnitine to exercise. Therefore, in patients with intermittent claudication, assessment of plasma levels of acetylcarnitine at rest and after exercise, by revealing a target population who could benefit from carnitine supplement, may provide a means for a focused therapy.
- Received August 10, 1995.
- Revision received October 23, 1995.
- Accepted November 3, 1995.
- Copyright © 1996 by American Heart Association
Brevetti G, Chiariello M, Ferulano G, Policicchio A, Nevola E, Attisano T, Ambrosio G, Siliprandi N, Angelini C. Increases in walking distance in patients with peripheral vascular disease treated with l-carnitine: a double-blind, cross-over study. Circulation. 1988;77:767-773.
Kerney AL, Randle PJ, Cooper RH, Whithouse S, Pask HT, Denton RM. Regulation of pyruvate dehydrogenase in rat heart. Biochem J. 1976;11:327-348.
Hulsmann WC, Siliprandi D, Ciman M, Siliprandi N. Effect of carnitine on the oxidation of a-oxoglutarate to succinate in the presence of acetoacetate or pyruvate. Biochem Biophys Acta. 1966;93:166-168.
Lysiak W, Lilly K, Di Lisa F, Toth P, Bieber LL. Quantitation of the effect of l-carnitine on the levels of acid-soluble short-chain acyl-CoA and CoASH in rat heart and liver mitochondria. J Biol Chem. 1988;263:1151-1157.
Harris RC, Foster CVL, Hultman E. Acetylcarnitine formation during intense muscular contraction in humans. J Appl Physiol. 1987;63:440-442.
Lennon DLF, Stratman FW, Shirago E, Nagle FJ, Maddem M, Harrson P, Carter AL. Effects of acute moderate intensive exercise on carnitine metabolism in men and women. J Appl Physiol. 1983;55:489-494.
Hiatt WR, Regensteiner JG, Wolfel EE, Ruff L, Brass EP. Carnitine and acylcarnitine metabolism during exercise in humans: dependence on skeletal muscle metabolic state. J Clin Invest. 1989;84:1167-1173.
Spriet LL, MacLean DA, Dyck DJ, Hultman E, Cederblad G, Graham TE. Caffeine ingestion and muscle metabolism during prolonged exercise in humans. Am J Physiol. 1992;262:E891-E898.
Hiatt WR, Nawaz D, Brass EP. Carnitine metabolism during exercise in patients with peripheral vascular disease. J Appl Physiol. 1987;62:2383-2387.
Hiatt WR, Wolfel EE, Regensteiner JG, Brass EP. Skeletal muscle carnitine metabolism in patients with unilateral peripheral arterial disease. J Appl Physiol. 1992;73:346-353.
Radack K, Wydersky RJ. Conservative management of intermittent claudication. Ann Intern Med. 1990;113:135-146.
Fishlock RC, Bieber LL, Snoswell AM. Sources of error in determinations of carnitine and acylcarnitine in plasma. Clin Chem. 1984;30:316-318.
Brevetti G, Angelini C, Rosa M, Carrozzo R, Perna S, Corsi M, Matarazzo A, Marcialis M. Muscle carnitine deficiency in patients with severe peripheral vascular disease. Circulation. 1991;84:1490-1495.
Brass EP, Hoppel CL. Carnitine metabolism in the fasting rat. J Biol Chem. 1978;253:2688-2693.
Brass EP, Hoppel CL. Relationship between acid-soluble carnitine and coenzyme A pools in vivo. Biochem J. 1980;190:495-504.
Hoppel CL, Genuth SM. Carnitine metabolism in normal-weight and obese human subjects during fasting. Am J Physiol. 1980;238:E409-E415.
Friolet R, Hoppeler H, Krahenbuhl S. Relationship between the coenzyme A and the carnitine pools in human skeletal muscle at rest and after exhaustive exercise under normoxic and acutely hypoxic conditions. J Clin Invest. 1994;94:1490-1495.
Soop M, Bjorkman O, Cederblad G, Hagenfeldt L, Wahren J. Influence of carnitine supplementation on muscle substrate and carnitine metabolism during exercise. J Appl Physiol. 1988;64:2394-2399.