Exploratory Open Label, Randomized Study of Acetyl- and Propionylcarnitine
in Chronic Fatigue Syndrome
RUUD C. W. VERMEULEN, MD, PHD, AND HANS R. SCHOLTE, PHD
Objectives: We compared the effects of acetylcarnitine, propionylcarnitine and both compounds on the symptoms of chronic
fatigue syndrome (CFS). Methods: In an open, randomized fashion we compared 2 g/d acetyl-L-carnitine, 2 g/d propionyl-L-
carnitine, and its combination in 3 groups of 30 CFS patients during 24 weeks. Effects were rated by clinical global impression of
change. Secondary endpoints were the Multidimensional Fatigue Inventory, McGill Pain Questionnaire, and the Stroop attention
concentration test. Scores were assessed 8 weeks before treatment; at randomization; after 8, 16, and 24 weeks of treatment; and
2 weeks later. Results: Clinical global impression of change after treatment showed considerable improvement in 59% of the
patients in the acetylcarnitine group and 63% in the propionylcarnitine group, but less in the acetylcarnitine plus propionylcarnitine
group (37%). Acetylcarnitine significantly improved mental fatigue (p ? .015) and propionylcarnitine improved general fatigue
(p ? .004). Attention concentration improved in all groups, whereas pain complaints did not decrease in any group. Two weeks
after treatment, worsening of fatigue was experienced by 52%, 50%, and 37% in the acetylcarnitine, propionylcarnitine, and
combined group, respectively. In the acetylcarnitine group, but not in the other groups, the changes in plasma carnitine levels
correlated with clinical improvement. Conclusions: Acetylcarnitine and propionylcarnitine showed beneficial effect on fatigue and
attention concentration. Less improvement was found by the combined treatment. Acetylcarnitine had main effect on mental fatigue
and propionylcarnitine on general fatigue. Key words: acetylcarnitine, attention concentration, carnitine, chronic fatigue syndrome,
ALC ? acetyl-L-carnitine; CFS ? chronic fatigue syndrome; CGI ?
clinical global impression of change; CDC ? Centers for Disease
Control and Prevention; DSM-IV ? Diagnostic and Statistical Man-
ual; MFI-20 ? multidimensional fatigue inventory; MPQ-DLV ?
McGill Pain Questionnaire–Dutch language version; OCTN2 ? or-
PLC ? propionyl-L-carnitine; RBE4 ? rat brain endothelial cells;
VAS ? visual analogue scale.
CFS is not known, sets of criteria for diagnosis of the condi-
tion have been published. The most widely accepted criteria
were formulated in the consensus meeting supervised by the
Centers for Disease Control and Prevention (CDC) (3). Pa-
tients with CFS are defined by the presence of clinically
evaluated, unexplained persistent chronic fatigue of new or
definite onset (not lifelong) resulting in substantial reduction
in previous levels of activities, and the occurrence of 4 or
more of the following symptoms (all of which must have
persisted for ?6 months and must not have predated the
fatigue): self-reported severe impairment in short-term mem-
ory or concentration, sore throat, tender lymph nodes, muscle
pain, multijoint pain without swelling or redness, headache of
a new type, unrefreshing sleep, and postexertional malaise
lasting more than 24 hours. Exclusions include a clear under-
lying organic cause, substance misuse, and severe psychiatric
disorder such as psychotic depression. Less severe psychiatric
disorders such as major depression without Diagnostic and
Statistical Manual (DSM)-IV-defined melancholic features or
stimations of the prevalence of chronic fatigue syndrome
(CFS) range from 1‰ to 4‰ (1,2). Although the cause of
anxiety disorders are not exclusionary diagnoses and are fre-
quently comorbid with CFS (4).
Kuratsune et al. (5) and Plioplys and Plioplys (6) reported
a decrease in plasma acylcarnitine in CFS, but this was not
confirmed by others (7–9). The first report on carnitine treat-
ment in CFS was by Grau et al. (10), who found no effect.
Plioplys and Plioplys (11) reported significant improvement
of CFS symptoms after 2 months of 3 g daily orally admin-
In a preliminary open label study, we treated 150 CFS
patients with 1 g oral L-carnitine bid. After 6 months, 104
patients (69%) reported marked improvement by clinical
global impression of change (CGI) [“I feel (very) much bet-
ter”]. Another 18 CFS patients were included in a randomized
double-blind study. Six were treated with oral acetyl-L-carni-
tine 1 g/d plus L-carnitine 1 g/d (low dose), 6 received twice
the dosage (high dose), and 6 received placebos. After 6
months, marked improvement was reported by 4 patients in
the low dosage group, by none in the high dosage group, and
by 1 in the placebo group (unpublished results).
Carnitine is essential for the mitochondrial oxidation of
long-chain fatty acids (12), because it acts as a carrier of acyl
groups across the inner mitochondrial membrane to the matrix
for ?-oxidation. Later, many more functions were detected in
detoxification of acyl-CoA and acylcarnitine export from the
cell, increasing free CoA/acyl-CoA in the cell affecting lipid,
protein, and carbohydrate metabolism [review (13)], mem-
brane synthesis and repair (14), and ac(et)ylation of amino
acids, histones, and proteins. The latter process is involved in
posttranslational modification, membrane binding, and G-pro-
tein signaling. Carnitine promoted perfusion in ischemia (15),
decreased oxidative stress (16,17), and attenuated damaging
effects of mitochondrial inhibitors and uncoupler in neurons
(18). Mitochondria were not the only organelles participating
in the actions of the carnitine system; also peroxisomes, endo-
and sarcoplasmic reticulum, and nuclear membrane metabo-
lized carnitine (19). Moreover, every human cell (including
the erythrocyte) contains carnitine palmitoyltransferase activ-
From the Research Center Amsterdam, Amsterdam, Netherlands.
Address correspondence and reprint requests to Ruud Vermeulen, MD,
PhD, CFS Research Center Amsterdam, Waalstraat 25, Amsterdam, 1078 BR,
Netherlands. E-mail: email@example.com
Received for publication January 22, 2003; revision received August 5,
Sigma-Tau Ethifarma, Assen, The Netherlands, sponsored the study and
provided the study medication.
Copyright © 2004 by the American Psychosomatic Society
Psychosomatic Medicine 66:276–282 (2004)
ity, and carnitine and acylcarnitines were detected in every
cell compartment and body fluid (13). Many biochemical,
pharmacological, and clinical studies have investigated the
action of carnitine and its esters acetylcarnitine and propio-
nylcarnitine. Acetylcarnitine is a universal mitochondrial en-
ergy source and acetyl donor. It promoted energy metabolism
and neurotransmission in the aged or neurotoxin-treated or
ischemic reperfused rodent brain [reviews (18,20)]. Promising
results were obtained in human degenerative brain diseases
[review (21)]. Propionylcarnitine is also an energy source,
stimulated the Krebs cycle as a precursor of succinyl-CoA,
decreased oxidative stress in various systems, and improved
cardiac dysfunction in rodent models (20,23). It was found to
be effective in human cardiovascular disorders [reviews
(24,25)]. Acetyl- and propionylcarnitine appeared generally
somewhat more effective than L-carnitine.
SUBJECTS AND METHODS
We compared the effect of acetyl-L-carnitine (ALC) and propionyl-L-
carnitine (PLC) on complaints of patients with CFS. We chose to conduct an
open exploratory study with a discussion with the participants about the
experienced change in their clinical condition, supported by objective ques-
tionnaires. The sample size was based on the results of the 2 preliminary
studies mentioned above, indicating improvement in 50% of the patients in
the low dose groups and less than 20% in the high dosage group. The sample
size was not based on differences between the ALC and the PLC groups
because data were not available. The present trial included 90 patients with
CFS according to the CDC criteria, in an open randomized study. Patients
were recruited from the polyclinic at the CFS Research Centre Amsterdam.
Excluded were patients with an evident underlying organic cause, substance
misuse, and severe psychiatric disorder such as psychotic depression. Pres-
ence of exclusion criteria according to the CDC (3,4) was revealed by
structured interview, physical examination, and extensive laboratory tests.
Before entry into the study, the nature of the study was explained to the
patients and written consent was obtained. An independent ethics committee
approved the study. The trial was conducted in accordance with the Decla-
ration of Helsinki (1996 revision) and under the principals of good clinical
practice, as laid out in the International Conference on Harmonisation docu-
ment Good Clinical Practice Consolidated Guideline. All analyses were
conducted on an intent-to-treat basis, with the last recorded value on each
outcome measure being carried forward to the end of the trial, unless stated
otherwise. Randomization was performed after opening of sealed envelopes
with the randomization codes and by blocks of 6 patients, stratified for gender,
and assigned the patients to an ALC group, a PLC group, and an ALC?PLC
group. After randomization, patients received 2 g acetyl-L-carnitine per day in
the ALC group, 2 g propionyl-L-carnitine in the PLC group, and 2 g ACL plus
2 g PLC in the ALC?PLC group. The medication was taken after breakfast
and dinner for 24 weeks. The dropouts were analyzed during the whole trial,
and their data were included in the results.
The trial profile is presented in Figure 1. Assessed for eligibility were 114
patients, and 90 were enrolled in the study. Table 1 shows the gender
distribution (78% females), the average age (37–42 years), the duration of
CFS (median 3 years or more), and the plasma carnitine levels at random-
ization, that were (about) the same in the ALC, the PLC, and the ALC?PLC
group. The plasma carnitine and carnitine esters levels showed no abnormal-
ities and were similar to those in controls (8).
Eight patients stopped because of side effects—3 in the ALC group, 2 in
the PLC group, and 3 in the ALC?PLC group. They experienced an over-
stimulated feeling and sleeplessness. Another 8 patients stopped, because they
did not experience any effect from the treatment—4 in the ALC group, 1 in
Figure 1.Trial profile
ACETYL- AND PROPIONYLCARNITINE IN CFS
277Psychosomatic Medicine 66:276–282 (2004)
the PLC group, and 3 in the ALC?PLC group. Two patients stopped for
reasons unrelated to the treatment.
At screening patients’ complaints, cognitive performance and pain were
assessed. This was repeated after a no-treatment period of 8 weeks, and then
the patients were randomly distributed in 3 groups and treated for 24 weeks.
During treatment, patients were reassessed at 8, 16, and 24 weeks. Patients
reported after 2 weeks of no-treatment whether the withdrawal of medication
had influenced their condition compared with 2 weeks earlier. The primary
endpoint of the study was the CGI (26). CFS is a condition without objective
criteria. Accordingly, in this study of CFS, the CGI was rated according to the
subjective experience of the patient, who had to answer the question: Do you
experience a change in your condition compared with the time of random-
ization? The CGI was rated on a 5-point scale, ranging from ?3 very much
worse, to ?3 very much better. Secondary endpoints were the fatigue axes of
the Multidimensional Fatigue Inventory (MFI-20) (27), the Stroop test for
attention concentration (28), and the McGill Pain Questionnaire-Dutch Lan-
guage Version (MPQ-DLV) for pain (29). The MFI-20 consists of 20 ques-
tions, the general, physical, and mental fatigue axes were used in the study.
The score on each axis is 5 to 20 points, indicating no fatigue to extreme
fatigue. The Stroop test estimates attention concentration. The result of the
reading time of the third card minus that of the second card of the Stroop test
is shown because in the preliminary studies (mentioned above) this gave the
best results to differentiate patients from controls. The score given is the time
needed to read aloud the names of 100 words of colors printed in another color
minus the time needed to read aloud the colors of 100 stripes. The Stroop test
can be used more than 1 time without affecting the interference score by a
practice or learning effect (30). The McGill Pain Questionnaire-Dutch Lan-
guage Version (MPQ-DLV) was used for the assessment of pain. The result
of the visual analogue scale (VAS) pain score is expressed as a distance of 0
to 100 mm, indicating no pain to extreme pain. A trained psychologist
administered all questionnaires and tests.
Blood was sampled by venipuncture in heparin tubes at randomization and
after the 24-week treatment period, for the assay of free carnitine and carnitine
esters in plasma, using tandem mass spectrometry as described by Vreken et
Data are expressed as mean and SD, unless indicated otherwise. Statistical
analysis was done with SPSS 11.0. Correlations were calculated by Pearson’s
correlation test for continuous data with a normal distribution. Data with a
not-normal distribution were tested with Spearman rank correlation test.
Changes of scores during therapy were tested by Friedman’s test for multiple
related variables. Significant levels were assumed at p ? .05. Bonferroni
correction was applied for multiple observations.
Role of the Funding Source
The funding source had no role in the decision to publish the data and in
the writing of the paper.
Table 2 summarizes the CGIs after 8 weeks of no treatment
and after 8, 16, and 24 weeks of treatment. Improvement was
reported after 24 weeks of treatment by 59% of the ALC
group, 63% of the PLC group, and 37% of the ALC?PLC
group, whereas 10% in the ALC group, 3% in the PLC group,
and 16% in the ALC?PLC group felt worse.
After 24 weeks of therapy, the medication was stopped and
all patients returned 2 weeks later for follow-up. Worsening of
CFS in the 2 weeks was reported by 52% in the ALC group,
50% in the PLC group, and by 37% in the ALC?PLC group.
No patients improved (Table 2). In the ALC group, 4 of the 17
patients who improved at 24 weeks experienced no change at
follow-up, 2 weeks later. In the PLC group, improvement
continued in 7 patients and in the ALC?PLC group in 3
patients (not shown). In the ALC group, 2 patients had not
improved at 24 weeks, but worsened at follow-up. This oc-
TABLE 1.Gender, Year of Birth, Duration of CFS, and Plasma Carnitine Levels at Randomization
Group ALCPLC ALC ? PLC
Duration CFS, median (y)
Duration CFS, range (y)
Plasma LC (?M)
Plasma ALC (?M)
Plasma PLC (?M)
37 ? 11
34 ? 10
4.9 ? 1.4
0.32 ? 0.14
38 ? 11
35 ? 9
5.3 ? 1.7
0.31 ? 0.14
42 ? 12
36 ? 8
5.3 ? 1.3
0.32 ? 0.14
Age and carnitine levels were expressed as means ? SD.
CFS ? chronic fatigue syndrome; ALC ? acetyl-L-carnitine; PLC ? propionyl-L-carnitine; LC ? L-carnitine.
TABLE 2a.Experienced Benefit From the Medication Number of Patients (%)
Treatment (wks)08 1624 Follow-up
ALC ? PLC
R. C. W. VERMEULEN AND H. R. SCHOLTE
278Psychosomatic Medicine 66:276–282 (2004)
curred in 3 patients in the PLC group and 2 patients in the
ALC?PLC group (not shown).
Table 3 summarizes the general fatigue, physical fatigue,
and mental fatigue scores. Significant improvement of general
fatigue score was found in the PLC (p ? .004) and in the
ALC?PLC group (p ? .000), and that of mental fatigue in the
ALC group (p ? .015). The physical fatigue score was not
significantly improved in the PLC group (p ? .069).
Table 4 show the scores of the attention concentration test
and the pain test. The score of the Stroop test is presented as
median and quartiles because the distribution of the data in the
PLC group at randomization showed a not-normal distribu-
tion. The attention concentration score improved significantly
in all groups (Table 4b). The pain score was unexpectedly low
in the ALC group (Table 4a). None of the treatments had
significant effect on the pain scores (Table 4b).
The CGI was correlated with the improvement in the gen-
eral fatigue, physical and mental fatigue score in all the groups
(r ? 0.362, p ? .05), and with the attention concentration
score in the ALC (r ? 0.476, p ? .010) and the ALC?PLC
(r ? 0.489, p ? .006), but not in the PLC group. The CGI was
not correlated with the pain score in any of the groups.
The levels of plasma L-carnitine increased significantly in
all groups of patients, but somewhat less in the responders.
Surprisingly, carnitine levels in the group receiving ALC plus
PLC became not considerably higher than in the other 2
groups. The levels of the carnitine esters increased in all
groups, but remained low compared with L-carnitine (Table 5).
The increase of carnitine levels in women was not different
from that in the male participants.
The change of the plasma L-carnitine concentration in the
ALC group was inversely related to clinical improvement, and
the least increase of carnitine was related to most improve-
ment. This was not found in the PLC and ALC?PLC group.
Change of plasma carnitine was related to improvement of
general and mental fatigue in the ALC group, but not in the
PLC and ALC?PLC group. Change of plasma carnitine was
not related to change in attention-concentration or pain score.
Plasma ALC and PLC were not related to clinical outcome
(correlations not shown).
The first aim of the study was to compare the different
influences of acetylcarnitine and propionylcarnitine on symp-
toms of CFS. The second goal was to obtain an indication of
the percentage of responders. Therefore, the study was de-
signed as an observation period of 8 weeks, followed by a
randomized trial and an observation period of 2 weeks.
In the first observation period the scores for fatigue, con-
centration, and pain of the patients proved to be stable. The
medication was well accepted by the patients, with a dropout
number of 8 because of side effects. The nature of the side
effects—overstimulation and sleeplessness—is different from
TABLE 2b. Friedman Test of CGI at 0 (Randomization), 8, 16, and 24 Weeks of Treatment
GroupALC PLC ALC ? PLC
20.24 p ? 0.000 30.25 p ? 0.0006.60 p ? 0.084
The first part of the table summarizes the clinical global impression of change (CGI) during treatment and 2 weeks later. Improved: CGI ? 2; no change: ?1 ?
CGI ? ?2; worse: CGI ? ?2. The CGI at randomization (0 weeks treatment) was relative to 8 weeks before. The CGI at 8, 16, and 24 weeks were relative
to the situation at 0 weeks (randomization). The CGI at follow-up (24 weeks of treatment, followed by 2 weeks without treatment) was relative to the condition
at 24 weeks.
ALC ? acetyl-L-carnitine; PLC ? propionyl-L-carnitine.
TABLE 3a.Clinical Measures of Fatigue Before and During Treatment
Treatment ALCPLC ALC ? PLC
General fatigue (MFI-20)Screening
18.6 ? 1.9
17.6 ? 2.1
16.7 ? 3.5
16.5 ? 4.1
15.9 ? 4.2
18.1 ? 2.6
16.9 ? 2.6
16.5 ? 3.6
15.8 ? 4.4
15.7 ? 4.4
17.0 ? 3.3
16.4 ? 2.8
15.1 ? 3.2
15.0 ? 2.9
15.1 ? 3.6
18.4 ? 1.8
18.0 ? 2.4
17.0 ? 2.9
15.7 ? 4.0
16.5 ? 3.1
17.8 ? 2.3
17.4 ? 3.0
16.5 ? 3.0
15.8 ? 4.0
16.4 ? 3.2
16.3 ? 2.5
15.1 ? 3.4
15.1 ? 3.2
13.8 ? 4.1
13.9 ? 3.5
19.1 ? 1.4
19.0 ? 1.5
18.0 ? 2.8
16.9 ? 3.2
17.3 ? 3.3
18.5 ? 1.6
17.9 ? 2.2
17.3 ? 2.9
16.1 ? 3.5
16.5 ? 3.4
15.7 ? 3.9
15.3 ? 3.7
14.3 ? 4.1
14.2 ? 4.0
14.6 ? 4.0
Physical fatigue (MFI-20)
Mental fatigue (MFI-20)
Mean ? SD. MFI-20 ? multidimensional fatigue inventory; ALC ? acetyl-L-carnitine; PLC ? propionyl-L-carnitine.
ACETYL- AND PROPIONYLCARNITINE IN CFS
279Psychosomatic Medicine 66:276–282 (2004)
those reported in most other studies, describing trimethyl-
amine formation or diarrhea in a few patients.
In advance, we had expected that ALC would have most
effect on attention concentration, PLC on physical fatigue, and
ALC?PLC on both. The CGI improved in the ALC and the
PLC group, but not significantly in the ALC?PLC group,
likely because of the higher total carnitine ester dose and
indicating an inversed U-form dose–response curve. An indi-
cation for the same effect was found in the preliminary dou-
ble-blind pilot study. The effect of ALC on mental fatigue and
attention concentration was significant. The PLC group
showed most improvement in general and physical fatigue and
slightly less in attention concentration. The ALC?PLC group
improved very well on general fatigue and also, but not
significantly, on physical and mental fatigue. The pain score
was not influenced in a significant way by any medication.
The CGI as an overall impression correlated significantly
with the change of the 3 separate fatigue scores, thus validat-
ing the concept (not shown).
In the present study, all patients had normal plasma free
carnitine levels, and a normal spectrum of the acylcarnitine
species, excluding primary and secondary carnitine deficiency
In the second observation period, 50% of patients deterio-
rated in the ALC and PLC groups. For most the relapse was
hardly acceptable, because they had adapted their activity to
TABLE 3b.Friedman Test of Fatigue Scores at 0, 8, 16, and 24 Weeks Treatment
Group ALCPLCALC ? PLC
General fatigue ?2
Physical fatigue ?2
Mental fatigue ?2
4.41 p ? 0.218
3.62 p ? 0.313
10.13 p ? 0.015
13.02 p ? 0.004
7.13 p ? 0.069
2.78 p ? 0.428
18.70 p ? 0.000
6.24 p ? 0.102
6.51 p ? 0.083
TABLE 4a. Clinical Measures of Attention Concentration and Pain
GroupALC PLC ALC ? PLC
Median (25–75 percentile). ALC ? acetyl-L-carnitine; PLC ? propionyl-L-carnitine.
TABLE 4b.Friedman Test of Attention Concentration and Pain Scores at 0, 8, 16, and 24 Weeks Treatment
Group ALCPLC ALC ? PLC
Attention concentration ?2
Pain score ?2
17.46 p ? 0.000
0.86 p ? 0.840
10.84 p ? 0.011
3.09 p ? 0.380
12.97 p ? 0.004
0.74 p ? 0.877
TABLE 5.Plasma Carnitine Levels in Patients Benefiting From the Treatment (Responders) and Nonresponders
Group ALC PLCALC ? PLC
LC before (?M)
LC after (?M)
ALC before (?M)
ALC after (?M)
PLC before (?M)
PLC after (?M)
33 ? 7
46 ? 5b
5.0 ? 1.6
5.5 ? 1.3
0.31 ? 0.11
0.38 ? 0.08b
36 ? 11
42 ? 11b
5.0 ? 1.1
6.5 ? 1.7b
0.34 ? 0.17
0.36 ? 0.17b
29 ? 5
43 ? 6
4.6 ? 1.1
5.7 ? 2.2
0.23 ? 0.05
0.44 ? 0.10
37 ? 9a
47 ? 12b
5.7 ? 1.8a
6.5 ? 2.4
0.34 ? 0.16a
0.93 ? 1.07b
34 ? 5
41 ? 9b
5.0 ? 1.2
6.0 ? 1.4b
0.29 ? 0.06
0.52 ? 0.50
39 ? 12
44 ? 12b
5.7 ? 1.5
7.1 ? 2.4
0.37 ? 0.22
0.58 ? 0.41
Data on L-carnitine (LC), acetyl-L-carnitine ALC, and propionyl-L-carnitine (PLC) at randomization and after 24 weeks therapy in nonresponders [clinical global
impression (CGI) ? 1] and responders (CGI ? 2). “After” and “before” refer to the treatment.
aSignificant difference between nonresponders vs. responders (p ? 0.05).
bSignificant difference between before treatment vs. after treatment (p ? 0.05).
R. C. W. VERMEULEN AND H. R. SCHOLTE
280Psychosomatic Medicine 66:276–282 (2004)
the improvement. The improved patients in the ALC?PLC
group (37%) all deteriorated in the second observation period.
The different effects of acetylcarnitine and propionylcarni-
tine could be explained by a different transport of the esters
from plasma to the tissues and the brain. If so, the observation
that the concentration of carnitine in the plasma in the ALC
group increased more in patients with less therapeutic effect,
indicates lower transport. It is conceivable that ALC and PLC
are not transported as such, because the plasma ALC and PLC
levels were relatively low and showed no correlation with
improvement. The precise mode of action of the carnitine
esters in alleviating the CFS symptoms is not clear. In this
study, acetylcarnitine apparently had a more “central” action
and propionylcarnitine a more “peripheral” action, which is in
line with other human studies. It is generally thought that
acetylcarnitine can easily pass the blood–brain barrier, and
that propionylcarnitine preferentially enters the heart. The fate
of double-labeled acetylcarnitine has been studied by Kurat-
sune et al. (32). They showed that only the acetyl moiety is
rapidly taken up by the brain, leaving the carnitine moiety in
the circulation. This was in line with older studies [reviewed
in (13)] establishing that carnitine import in the brain is a slow
process with low affinity. Recently, however, with immortal-
ized RBE4 cells, an in vitro model of the blood–brain barrier,
it was shown that these cells expressed the high-affinity car-
nitine transporter OCTN2, and transported carnitine, acetyl-
carnitine, and propionylcarnitine (33). When this is the case, a
preferential uptake of acetylcarnitine by brain and a preferen-
tial uptake of propionylcarnitine by heart are less likely,
because OCTN2 is also present in muscle and several other
cells. Perhaps another messenger contributes to, or is involved
in, the action of the carnitine esters, such as purines (34) or
This study indicates that both ALC and PLC are successful
in the treatment of symptoms of CFS in a major subset of CFS
patients. The fact that the pattern of improvement by acetyl-
carnitine, propionylcarnitine, and their combination was not
identical suggests that differences in transport and/or metab-
olism are involved in their action.
Conclusive evidence for the effects of carnitine esters on
CFS will have to await a double blind, placebo-controlled,
We thank Otto J. Bauermann (Sigma-Tau Ethifarma) for his coop-
eration and Professor J. F. Koster for reading of the manuscript.
1. Gunn WJ, Connell DB, Randall B. Epidemiology of chronic fatigue
syndrome: the Centers for Disease Control Study. Ciba Foundation
Symposium 1993;173:83–93; discussion 93.
2. Llansola M, Felipo V. Carnitine inhibits hydrolysis of inositol phospho-
lipids induced by activation of metabotropic receptors. Neurochem Res
3. Fukuda K, Straus SE, Hickie I, Sharpe MC, Dobbins JG, Komaroff A.
The chronic fatigue syndrome: a comprehensive approach to its definition
and study. International Chronic Fatigue Syndrome Study Group. Ann
Intern Med 1994;121:953–9.
4. Cleare AJ. The neuroendocrinology of chronic fatigue syndrome. Endocr
5. Kuratsune H, Yamaguti K, Takahashi M, Misaki H, Tagawa S, Kitani T.
Acylcarnitine deficiency in chronic fatigue syndrome. Clin Infect Dis
6. Plioplys AV, Plioplys S. Serum levels of carnitine in chronic fatigue
syndrome: clinical correlates. Neuropsychobiology 1995;32:132–8.
7. Majeed T, De Simone C, Famularo G, Marcellini S, Behan PO. Abnor-
malities of carnitine metabolism in chronic fatigue syndrome. Eur J Neu-
8. Scholte HR, De Jong G, De Klerk JBC, Hussaarts LM, Kurk RM,
Moesker A, Van Oudheusden LJ, Rodrigues Pereira R, Van Rossum N,
Vermeulen RCW. The meaning of plasma free and acylcarnitine levels.
When to supplement? 2nd Colloquium on Mitochondria and Myopathies,
Halle/Saale, Germany. Eur J Med Res 2000;5(suppl 1):45–6.
9. Soetekouw PMMB, Wevers RA, Vreken P, Elving LD, Janssen AJM,
Van der Veen Y, Bleijenberg G, Van der Meer JWM. Normal carnitine
levels in patients with chronic fatigue syndrome. Neth J Med 2000;57:
10. Grau JM, Casademont J, Pedrol E, Fernandez-Sola J, Cardellach F,
Barros Urbano-Marquez NA. Chronic fatigue syndrome: studies on skel-
etal muscle. Clin Neuropathol 1992;11:329–32.
11. Plioplys AV, Plioplys S. Amantadine and L-carnitine treatment of
chronic fatigue syndrome. Neuropsychobiology 1997;35:16–23.
12. Fritz IV. Action of carnitine on long-chain fatty acid oxidation by liver.
Am J Physiol 1955;197:297–304.
13. Scholte HR, Boonman AMC, Hussaarts-Odijk LM, Ross JD, Van Oud-
heusden LJ, Rodrigues Pereira R, Wallenburg HCS. New aspects of the
biochemical regulation of the carnitine system and mitochondrial fatty
acid oxidation. In: Seim H, Lo ¨ster H, eds. Carnitine pathobiochemical
basics and clinical applications. Bochum: Ponte Press; 1996:11–31.
14. Arduini A, Denisova N, Virmani A, Avrova N, Federici G, Arrigoni-
Martelli E. Evidence for the involvement of carnitine-dependent long-
chain acyltransferases in neuronal triglycerides and phospholipid fatty
acid turnover. J Neurochem 1994;62:1530–8.
15. Hulsmann WC, Peschechera A, Serafini E, Ferrari LE. Release of isch-
emia in paced rat Langendorff hearts by supply of L-carnitine: role of
endogenous long-chain acylcarnitine. Mol Cell Biochem 1996;156:
16. Pastorino JG, Snyder JW, Serroni A, Hoek JB, Farber JL. Cyclosporin
and carnitine prevent the anoxic death of cultured hepatocytes by inhib-
iting the mitochondrial permeability transition. J Biol Chem 1993;268:
17. Luo X, Reichetzer B, Trines J, Benson LN, Lehotay DC. L-carnitine
attenuates doxorubicin-induced lipid peroxidation in rats. Free Radic Biol
18. Virmani MA, Biselli R, Spadoni A, Rossi S, Corsico N, Calvani M,
Fattorossi A, De Simone C, Arrigoni-Martelli E. Protective actions of
L-carnitine and acetyl-L-carnitine on the neurotoxicity evoked by mito-
chondrial uncoupling or inhibitors. Pharmacol Res 1995;32:383–9.
19. Bremer J. The role of carnitine in cell metabolism. In: De Simone C,
Famularo G, eds. Carnitine today. Heidelberg: Springer-Verlag; 1997:
20. Calvani M, Arrigoni-Martelli E. Attenuation by acetyl-L-carnitine of
neurological damage and biochemical derangement following brain isch-
emia and reperfusion. Int J Tissue React 1999;21:1–6.
21. Pettegrew JW, Levine J, McClure RJ. Acetyl-L-carnitine physical-
chemical, metabolic, and therapeutic properties: relevance for its mode of
action in Alzheimer’s disease and geriatric depression. Mol Psychiatry
22. Vanella A, Russo A, Acquaviva R, Campisi A, Di Giacomo C, Sorrenti
V, Barcellona ML. L-propionyl-carnitine as superoxide scavenger, anti-
oxidant, and DNA cleavage protector. Cell Biol Toxicol 2000;16:
23. Terada R, Matsubara T, Koh N, Nakamura J, Hotta N. Effects of pro-
pionyl-L-carnitine on cardiac dysfunction in streptozotocin-diabetic rats.
Eur J Pharmacol 1998;357:185–91.
24. Retter AS. Carnitine and its role in cardiovascular disease. Heart Dis
25. Lango R, Smolenski RT, Narkiewicz M, Suchorzewska J, Lysiak-
Sydlowska W. Influence of L-carnitine and its derivatives on myocardial
metabolism and function in ischemic heart disease and during cardiopul-
monary bypass. Cardiovasc Res 2001;51:21–9.
ACETYL- AND PROPIONYLCARNITINE IN CFS
281Psychosomatic Medicine 66:276–282 (2004)
26. Gøtzsche PC. Sensitivity of effect variables in rheumatoid arthritis: a Download full-text
meta-analysis of 130 placebo controlled NSAID trials. J Clin Epidemiol
27. Smets EMA, Garssen B, Cull A, De Haes JCJM. Application of the
multidimensional fatigue inventory (MFI-20) in cancer patients receiving
radiotherapy. Br J Cancer 1996;73:241–5.
28. Dyer FH. The Stroop phenomenon and its use in the study of perceptual,
cognitive and response processes. Memory Cognit 1973;1:106–20.
29. Van Der Kloot WA, Oostendorp RAB, Van Der Meij J, Van Den Heuvel
J. The Dutch version of the McGill pain questionnaire: a reliable instru-
ment. Ned Tijdschr Geneeskd 1995;139:669–73.
30. Provost SC, Woodward R. Effects of nicotine gum on repeated admin-
istration of the Stroop test. Psychopharmacology 1991;104:536–40.
31. Vreken P, van Lint AE, Bootsma AH, Overmars H, Wanders RJ, van
Gennip AH. Rapid diagnosis of organic acidemias and fatty-acid oxida-
tion defects by quantitative electrospray tandem-MS acyl-carnitine anal-
ysis in plasma. Adv Exp Med Biol 1999;466:327–37.
32. Kuratsune H, Watanabe Y, Yamaguti K, Jacobsson G, Takahashi M,
Machii T, Onoe H, Onoe K, Matsumura K, Valind S, Kitani T, Langstrom
B. High uptake of [2–11C] acetyl-L-carnitine into the brain: a PET study.
Biochem Biophys Res Commun 1997;231:488–93.
33. Friedrich A, Prasad PD, Freyer D, Ganapathy V, Brust P. Molecular
cloning and functional characterization of the OCTN2 transporter at the
RBE4 cells, an in vitro model of the blood-brain barrier. Brain Res
34. Capecchi PL, Laghi Pasini F, Quartarolo E, Di Perri T. Carnitines
increase plasma level of adenosine and ATP in humans. Vasc Med
R. C. W. VERMEULEN AND H. R. SCHOLTE
282 Psychosomatic Medicine 66:276–282 (2004)