Article Text

This article has a correction. Please see:


The effects of the 5-HT2C agonist m-chlorophenylpiperazine on elite athletes with unexplained underperformance syndrome (overtraining)
  1. R Budgett1,
  2. N Hiscock2,
  3. R Arida2,
  4. L M Castell2
  1. 1Olympic Medical Institute, Northwick Park Hospital, Watford, UK
  2. 2Department of Biochemistry, University of Oxford, Oxford, UK
  1. Correspondence to: Linda M Castell, Green Templeton College, University of Oxford, Oxford OX2 6HG, UK; lindy.castell{at}


A possible link between the neurotransmitter, 5-hydroxytryptamine (5-HT), plasma tryptophan, and branched chain amino acids concentration and exercise-induced fatigue is described by the central fatigue hypothesis. 5-HT receptors and neuroendocrine “challenge” tests, using prolactin release as an indirect measure of 5-HT activity were studied by recent investigations. In the present study, the original hypothesis about the role of amino acids in increasing brain 5-HT with a neuroendocrine challenge test on elite athletes diagnosed with unexplained, underperformance syndrome (UUPS) was combined. There was an apparent increased sensitivity of 5-HT receptors in athletes with UUPS compared with fit, well-trained controls, as measured via increased prolactin release following a bolus dose of m-chlorophenylpiperazine , a 5-HT agonist. No changes were observed in plasma amino acid concentrations in either group. There is evidence that well-trained athletes have a reduced sensitivity of 5-HT receptors. The present study suggests that this adaptation may be lost in athletes with UUPS: this might explain some of their observed symptoms.

Statistics from

Between 5% and 10% of endurance athletes break down each year with unexplained underperformance syndrome (UUPS). This is best defined objectively by measurable underperformance and not by symptoms; however, fatigue and an unexpected sense of effort are the most important and consistent symptoms. Athletes also regularly experience mood and sleep disturbance, as well as frequent minor infections. Many symptoms (eg, sleep disturbances, frequent fatigue, lack of motivation) overlap with chronic fatigue syndrome (CFS). Central fatigue emanates from the brain; if sustained, it can lead to chronic fatigue.

Definition of UUPS

Substandard performance for which no explanation can be found is best defined simply as “unexplained underperformance syndrome”. Alternative terms such as “postviral fatigue” or “overtraining syndrome” imply causation, which is often incorrect. Not all athletes with frequent upper respiratory tract infections (URTI), depressed mood state, fatigue or underperformance are actually “overtrained”. In order to allow researchers and clinicians to investigate the problem, a broader definition has been created.1 UUPS is precisely defined as “unexplained underperformance agreed by coach and athlete despite two weeks of relative rest”. The underperformance may occur in competition, in training, on an appropriate ergometer or in the laboratory. Two weeks of relative rest, that is, a period of light training, such as would be used for a normal taper leading up to a major competition, is necessary to manage this.2 3

Most athletes presenting with UUPS had fatigue and an increased sense of effort.3 4 Athletes often describe underperforming when the pace picks up at the end of a race. There is usually a history of very heavy training and competition. Many of these athletes sustain frequent or almost continuous minor infections, particularly URTI. In addition, symptoms such as loss of energy, unexplained heavy, stiff and sore muscles, mood disturbance including anxiety, depression, irritability and emotional liability are reported.2 5 There appears to be considerable similarity between overtraining and clinical depression.6 7 Additional symptoms include loss of competitive drive, libido and appetite. A change in expected sleep quality is often observed: this is a particularly important symptom that, on its own, may lead to the rest of the clinical picture.

Any combination of symptoms is possible: this variability in presentation and lack of any diagnostic test helps to explain the contradictions in research into UUPS.8 Fatigue emanating from the brain (central fatigue) as opposed to peripheral fatigue (from muscle) may play an important role in UUPS.

Central fatigue hypothesis

The central fatigue hypothesis links the neurotransmitter, 5-hydroxytryptamine (5-HT), with its precursor, plasma free tryptophan (p[FT]), plasma branched chain amino acids (p[BCAA]), and exercise-induced fatigue.9 Uniquely among amino acids, tryptophan binds to albumin in the blood, with competition from fatty acids.9 10 An exerciseinduced increase in plasma free fatty acids (p[NEFA]) leads to an increase in unbound (free) tryptophan.11 Free tryptophan competes with BCAA for the same port of entry into the brain across the blood–brain barrier.12 The p[FT] increases in humans after prolonged exhaustive exercise: this may increase the rate of synthesis in 5-HT, which is involved in sleep and fatigue. Despite the rate-limiting enzyme tryptophan hydroxylase,13 an increased level of brain tryptophan can lead to increased 5-HT in some areas of the brain.11 The plasma concentration ratio of free tryptophan to BCAA (p[FT/BCAA]) governs the rate of tryptophan entry into the brain and hence the rate of synthesis of 5-HT. This may lead to premature fatigue in endurance athletes.9 Provision of BCAA may help to combat the increase in p[FT/BCAA]. Some exercise supplementation studies found a positive effect of BCAA feeding, in particular on mental performance; some found no effect.14 15 Most studies have measured only physical, rather than mental, performance. The mental component of maintaining a given power output is clearly important.

Central fatigue: exercise and clinical studies

Using plasma prolactin release (p[PRL]) as an indirect measure of 5-HT activity, Wilson and Maughan16 found that the administration of a 5-HT reuptake inhibitor impaired the ability of athletes to perform prolonged exercise. Jakeman et al17 suggested that central serotoninergic receptor function was decreased in response to endurance training: however, Strachan and Maughan18 saw no change in endurance athletes in response to D-fenfluramine. Struder et al19 provided evidence that welltrained athletes have a reduced sensitivity of 5-HT receptors.

In patients with CFS, evidence of increased sensitivity of 5-HT receptors was observed compared with controls and depressed patients.20 Vassallo et al.21 administered a 5-HT2C agonist versus placebo to CFS and sedentary volunteers: no change in p[PRL] between the groups was observed. In CFS patients versus sedentary controls, Castell et al22 observed both a higher resting p[FT] and p[FT/BCAA]. In a previous study by the present group on postoperative fatigue before and after major surgery, an increase in fatigue was markedly correlated with an increase in p[FT] and p[FT/BCAA].23

The present study tested the hypothesis that p[PRL] increases in response to 5-HT releasing agents in athletes suffering from UUPS compared with healthy, non-fatigued athletes training at a similar level. This study used p[PRL] as an indirect measure of 5-HT activity in the brain, specifically because of its strong link with 5-HT2C receptors. Prolactin is one of several factors involved but is easy to measure and has a long history in this role.24 The sensitivity of 5-HT receptors was tested with a 5-HT2C agonist, m-chlorophenylpiperazine (a metabolite of the antidepressant, trazodone, which bypasses presynaptic release). The hypothesis was also tested that p[FT] is higher and p[BCAA] lower at rest in athletes with UUPS cf. controls, leading to an increased p[FT/BCAA].


Ethical permission was obtained from Northwick Park Hospital Ethical Committee. Olympic athletes with UUPS were recruited over 1 year (n=19): 10 rowers, two cyclists, two triathletes, three middle distance runners and two swimmers (age range 20–35 years) all presented at the British Olympic Medical Centre fulfilling the criteria for diagnosis with no identifiable medical cause. In eight cases, an athlete from the same sport, with a similar level of training but without UUPS, was recruited as a matched control. Four rowers from top university squads were also recruited as healthy controls (total n=12, age range 19–42 years). Participants were predominantly men: too few women athletes were recruited to study sex differences. All were allocated a study number and tested in random order to preserve anonymity and blinding at assay.

Participants arrived at the laboratory, fasted, at 12 noon. They rested supine for 15 min, remaining so for the duration of the study. Thirty minutes later, after the first two blood samples, mCPP was administered orally as a liquid drawn up in a syringe in a dose of 0.025 mg/kg. The elixir was made up in solution at a concentration of 2 mg/ml, at Northwick Park Hospital Pharmacy, on the day of the experiment.

Blood samples

After a 15-min rest, a 10-ml blood sample was taken to measure prolactin, p[FT] and p[BCAA]. A catheter was left in a forearm vein and flushed with heparinised saline to ensure patency. The first sample was taken 30 min before the true baseline sample, to overcome problems with the surge in prolactin release, which occurs immediately after venepuncture, returning to normal 30 min later. For each sample, after discarding ca. 2 ml of blood, a further 10 ml was taken, immediately centrifuged and stored frozen at −80°C for subsequent assay. The catheter was then flushed again with heparinised saline. This procedure was repeated every 30 min for a total of 180 min after administration of mCPP.


Free tryptophan was measured in supernatant obtained via the standard method of ultrafiltration (Ultra-filtration vials TGC00, Millipore, Watford, UK). The p[BCAA] and p[FT] were assayed enzymatically25 26 and the p[FT/BCAA] ratio assessed. The p[PRL] was measured using an antibody sandwich immunoassay (Chiron Diagnostics, Halstead, UK). All samples were assayed blind via random renumbering.

Statistical analysis

Data is expressed as mean (SE). Repeated measures ANOVA and paired t tests were used to test for significance.


Analysis of p[PRL] showed (1) overall a marked increase (45%, p<0.001), peaking at 150 min, compared with baseline levels; (2) at every point except two (90 and 180 min), p[PRL] was higher (either significantly or close to significance) in fatigued athletes with UUPS than in the controls (fig 1). The assay coefficient of variance was 5%. There was no difference (either significantly or close to significance) in p[BCAA], p[FT] or p[FT/BCAA] in either group at any time point or between groups (data not shown).

Figure 1

Plasma prolactin concentration (mM) (SE). Time (min) before and after administration of a bolus dose of mCPP at 0. p Denotes significance between control (n=12) and UUPS (n=19) groups.


The measurements of p[PRL] over 3 h enabled 5-HT receptor sensitivity to be tested directly since mCPP bypasses presynaptic release. At every time point, except at 90 and 180 min, p[PRL] was significantly (or close to significance) higher in athletes with UUPS than in well-trained, fit athletes. This suggests that these individuals may have more sensitive 5HT2C receptors. There is evidence that well-trained athletes have a reduced sensitivity of 5-HT receptors.19 Based on findings in the present study, adaptation appears to be lost in athletes with UUPS.

If underperformance can be explained in terms of a major disease, the diagnosis cannot be made. Thus, all athletes with a diagnosis of UUPS should have a history and physical examination performed by an experienced physician. Endurance athletes present with fatigue and underperformance with secondary changes in mood specific to the sport and individual (Budgett, 1998). Those athletes with frequent minor infections (particularly URTIs) may form a separate overlapping subgroup.27 At present, management of athletes with UUPS involves gentle exercise regimes with relative rest over 6–12 weeks.3 Clinically depressed UUPS patients are treated with antidepressants.

Lac and Maso28 surveyed data from several studies on possible biological markers of overtraining during training: blood parameters alone do not appear to be sufficiently specific to make the diagnosis. Future diagnoses may be helped by measurements of decreases in plasma glutamine;29 30 changes in catecholamines;31 inflammatory cytokines, leptin32 and the hypothalamic pituitary axis,33 34 together with alterations in eccentric/concentric muscle force.35 Meeusen et al36 observed a dysfunctional hypothalamic–pituitary response in an overtrained athlete undertaking two bouts of maximal exercise. Using the same exercise tests, they observed a 14% increase in plasma prolactin in another athlete diagnosed with UUPS, whose ACTH was concomitantly decreased by 7%. Other neurotransmitters, including dopamine, may also have a role in UUPS.37

Resting p[FT] was 30% higher in CFS patients than sedentary controls;22 this might indicate persistent fatigue in individuals with CFS. In the present study, resting p[FT] did not differ between UUPS athletes and controls. However, there was a similar magnitude of difference in PRL (p<0.01), suggesting that increased PRL at rest may also be indicative of persistent fatigue in these individuals with UUPS. Uusitalo et al38 saw neither change in brain 5-HT re-uptake in overtrained athletes in the midbrain (which regulates 5-HT uptake) 1 year after baseline samples nor any correlation between 5-HT re-uptake and depression scores. However, more progressive monitoring of these subjects during their return to health might have provided better insight.

The causes for UUPS are multifactorial but lead to a common outcome. Fatigue is the most consistent symptom, although it is underperformance that is measurable and objective. A major part of this fatigue seems central in origin. Thus, changes in the brain, such as increased 5-HT activity, could be an important part of the syndrome.


The present study showed an apparent increased sensitivity of 5-HT receptors in athletes with UUPS compared with fit, welltrained controls. This, combined with the increase in 5-HT levels commonly observed during prolonged endurance exercise, might explain some symptoms seen in UUPS. Nevertheless, it does not explain increased vulnerability to infection and peripheral changes, which also occur in these individuals.

The full 5-HT sensitivity test is not suitable for regular monitoring of athletes. However, prolactin levels, as one of several blood parameters, could prove useful in monitoring the individual response to training and recovery. It is important to note that, because of the wide intra-individual variation for many parameters, accurate deductions cannot be made from a single sample taken during illness.


The authors are grateful to the participants, to the Linbury Trust for funding this study and to Dr Mike Franklin for helping with prolactin analysis.


View Abstract


  • Competing interests None.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • Correction
    BMJ Publishing Group Ltd and British Association of Sport and Exercise Medicine