Intended for healthcare professionals


Exercise responses and psychiatric disorder in chronic fatigue syndrome

BMJ 1995; 311 doi: (Published 26 August 1995) Cite this as: BMJ 1995;311:544
  1. Russell J M Lane, consultant neurologista,
  2. Adrian P Burgess, lecturer in psychologyb,
  3. Janis Flint, senior clinical psychologistb,
  4. Massimo Riccio, consultant psychiatristb,
  5. Leonard C Archard, reader in molecular pathologyc
  1. a Academic Unit of Neuroscience, Charing Cross and Westminster Medical School, London W6 8RF
  2. b Department of Psychiatry, Charing Cross and Westminster Medical School, London W6 8RF
  3. c Department of Biochemistry, Charing Cross and Westminster Medical School, London W6 8RF
  1. Correspondence to: Dr Lane.
  • Accepted 29 June 1995

Fatigue, exercise intolerance, and myalgia are cardinal symptoms of the chronic fatigue syndrome, but whether they reflect neuromuscular dysfunction or are a manifestation of depression or other psychiatric or psychological disorders diagnosed in a high proportion of fatigued patients in the community is unclear.1 In previous studies patients with the chronic fatigue syndrome showed exercise intolerance in incremental exercise tests, which seemed to be related to an increased perception of effort; also, blood lactate concentrations in some patients tended to increase more rapidly than normal at low work rates, implying inefficient aerobic muscle metabolism.2 We examined venous blood lactate responses to exercise at a work rate below the anaerobic threshold in relation to psychiatric disorder.

Patients, methods, and results

We studied 96 consecutive patients meeting the Oxford criteria for diagnosis of the chronic fatigue syndrome3 by using the subanaerobic threshold exercise test.4 Subjects pedalled an electronically braked bicycle ergometer for 15 minutes at 90% of the predicted work rate at their anaerobic threshold, based on age, weight, and sex; venous blood lactate concentrations were measured before, immediately after, and 30 minutes after exercise. Continuous electrocardiographic monitoring allowed measurement of mean heart rate during exercise. An abnormal result was defined as lactate concentrations exceeding the upper 99% reference limit for normal control subjects4 at two or more time points. We screened a convenience sample of 43 of the 96 patients using 11 established neuropsychological and psychiatric instruments, including the present state examination, for assessment of psychiatric caseness and prediction of psychiatric diagnoses.5

The study group comprised 41 men and 55 women of similar age range. Duration of symptoms ranged from six months to 20 years (mean 43.5 (SD 11.6) months). All patients were ambulant.

Thirty one (12 men, 19 women) of the 96 patients (32%) had abnormal lactate responses to exercise (subanaerobic threshold exercise test positive). These patients did not differ significantly from those with normal lactate responses in age, sex, or duration of symptoms. Twenty nine patients (30%) had mean heart rates above the predicted upper 95% reference limit for normal controls.4 However, there was no significant difference between the proportions with abnormal heart rates among those with abnormal or normal lactate responses (7/31 v 22/63 (no data in two cases); odds ratio 0.54 (95% confidence interval 0.2 to 1.46)).

The 43 patients studied by psychological and psychiatric tests did not differ significantly from the 53 other patients in terms of age (mean 34.6 v 34.4 years); mean duration of symptoms (43.6 v 37.1 months); sex distribution (19 men v 24 men); whether they were working (27 unemployed v 27 unemployed (not recorded in two cases)); or proportion with abnormal lactate responses (15/43 v 16/53). Eighteen of the 43 (42%) patients fulfilled criteria for psychiatric caseness. Diagnoses were neurotic depression (12 patients), manic depression (3), phobic anxiety (2), and anxiety state (1). Patients with normal lactate responses were more likely to fulfil criteria for psychiatric caseness than those with abnormal lactate responses (15/28 v 3/15; odds ratio 4.6 (1.06 to 20.1)).


Patients with symptoms that precluded or severely restricted physical activity might be expected to show increased lactate responses to exercise as a result of “deconditioning.” However, abnormal lactate responses were seen in only a third of our cases, and we found no relation between lactate responses and duration of symptoms or any other demographic variable. Furthermore, although 30% of all patients had abnormally high heart rates during exercise, a high heart rate was as common in patients with normal lactate responses as it was in those with abnormal responses. Patients with the chronic fatigue syndrome who also had psychiatric disorders such as depression, which is commonly associated with fatigue and inactivity, might also be expected to be unfit and to have abnormal exercise results; but the converse proved to be true.

Our results suggest that some patients with the chronic fatigue syndrome have impaired muscle energy metabolism that is not readily explained by physical inactivity or psychiatric disorder. This adds to the growing body of evidence that the syndrome is heterogeneous.


We thank the Staines Myalgic Encephalomyelitis Self Help group and the Myalgic Encephalomyelitis Association for their support.


  • Funding No additional funding.

  • Conflict of interest None.