Professor Julia L. Newton, Institute for Ageing and Health, Medical School, Framlington Place, Newcastle-upon-Tyne, NE2 4HH, UK. (fax: 00 44 191 282 5338; e-mail: email@example.com).
Abstract. Jones DEJ, Hollingsworth KG, Taylor R, Blamire AM, Newton JL (From the Institute of Cellular Medicine, Newcastle Magnetic Resonance Centre, and Institute for Ageing and Health, Newcastle University, UK). Abnormalities in pH handling by peripheral muscle and potential regulation by the autonomic nervous system in chronic fatigue syndrome. J Intern Med 2010; 267: 394–401.
Objectives. To examine muscle acid handling following exercise in chronic fatigue syndrome (CFS/ME) and the relationship with autonomic dysfunction.
Design. Observational study.
Setting. Regional fatigue service.
Subjects & interventions. Chronic fatigue syndrome (n = 16) and age and sex matched normal controls (n = 8) underwent phosphorus magnetic resonance spectroscopy (MRS) to evaluate pH handling during exercise. Subjects performed plantar flexion at fixed 35% load maximum voluntary contraction. Heart rate variability was performed during 10 min supine rest using digital photophlethysmography as a measure of autonomic function.
Results. Compared to normal controls, the CFS/ME group had significant suppression of proton efflux both immediately postexercise (CFS: 1.1 ± 0.5 mmol L−1 min−1 vs. normal: 3.6 ± 1.5 mmol L−1 min−1, P < 0.001) and maximally (CFS: 2.7 ± 3.4 mmol L−1 min−1 vs. control: 3.8 ± 1.6 mmol L−1 min−1, P < 0.05). Furthermore, the time taken to reach maximum proton efflux was significantly prolonged in patients (CFS: 25.6 ± 36.1 s vs. normal: 3.8 ± 5.2 s, P < 0.05). In controls the rate of maximum proton efflux showed a strong inverse correlation with nadir muscle pH following exercise (r2 = 0.6; P < 0.01). In CFS patients, in contrast, this significant normal relationship was lost (r2 = 0.003; P = ns). In normal individuals, the maximum proton efflux following exercise were closely correlated with total heart rate variability (r2 = 0.7; P = 0.007) this relationship was lost in CFS/ME patients (r2 < 0.001; P = ns).
Conclusion. Patients with CFS/ME have abnormalities in recovery of intramuscular pH following standardised exercise degree of which is related to autonomic dysfunction. This study identifies a novel biological abnormality in patients with CFS/ME which is potentially open to modification.
Chronic fatigue syndrome (CFS/ME) is a major clinical problem affecting a substantial number of predominantly young individuals through its impact on quality of life and social function . To date, little progress has been made in terms of identifying aetiological processes in CFS/ME. This failure to elucidate key mechanisms has impaired the development of successful therapeutic approaches for this important condition. There has also been an inevitable tendency to invoke psychological explanations for the origin of fatigue in CFS/ME patients. This psychological model is at odds with patient perceptions of the nature of their problems, which are very much related to difficulties in the translation of intention to undertake physical activity into capacity to perform that activity. The perception of many patients is that their condition has a ‘peripheral’ origin as opposed to the ‘central’ one favoured by majority expert opinion in the field. The consequence of these combined effects has been to frustrate the CFS/ME patient body and to limit therapeutic progress.
The strongest body of evidence for organic dysfunction in CFS/ME relates to impairment of autonomic nervous system function [2–5]. Historically, studies regarding autonomic dysfunction in CFS/ME have been limited by the restrictions inherent in the experimental modalities used. A recent study performed by our group, using the comprehensive autonomic symptom scoring assessment tool the Composite Autonomic Symptom Scale (COMPASS), has identified a clear association with autonomic dysfunction related symptoms in CFS/ME, with a particularly strong apparent link with symptoms suggestive of vasomotor instability . This finding is supported by the data regarding objective assessment of vasomotor autonomic function in CFS/ME [7–17]. Although there appears to be an association between autonomic dysfunction and fatigue in CFS/ME at present the underlying mechanism is unclear.
The perception of fatigue is not always abnormal. It is appropriate for such perception to occur following significant exercise. Indeed, central impairment of the initiation of physical activity has been recently reported in patients induced to experience peripheral fatigue . This suggests that the potential presence of a feedback signal from fatigued peripheral muscle to the central nervous system may prevent initiation of further exercise which may be damaging to the individual. Hence, pathological fatigue could be a consequence of inappropriate or excessive signal feedback, with the effect of precluding the initiation of exercise in people comparatively early in the fatiguing process, leading to inappropriate perception of fatigue. Recently, understanding of the metabolic basis of fatigue in muscle has been permitted by in vivo studies of exercising muscle. The role played by acidosis within muscle has been identified as being of particular importance in fatigue associated diseases  and CFS/ME . Recovery following exercise-induced fatigue is associated with activation of transporter mechanisms able to reduce intracellular acidosis . Critical amongst these are active transporters such as the sodium/proton anti-porter, monocarboxolate transporters (able to cotransport lactate and protons) and the chloride/bicarbonate transporter. Active proton excretion following fatigue inducing exercise is associated with the resolution of that fatigue and perception of the ability to return to exercise .
In this study, we test the hypothesis that exercise-induced acid handling is altered in CFS/ME leading to a perception of fatigue which is inappropriate to the degree of exercise undertaken. Furthermore, we have also explored the possibility that autonomic dysfunction may be a contributing factor to abnormality in muscle proton homeostasis.
Subjects and methods
Patients with CFS/ME (n = 16) were recruited through our local CFS/ME Clinical Service. All subjects fulfilled the Fukuda clinical diagnostic criteria for CFS . A comparator group of age and sex matched healthy sedentary controls recruited from the local population (n = 8) underwent assessment. Sedentary was defined as having a sedentary job and performing <3 h of physical activity per week . Subjects were excluded, if they had potential secondary causes for fatigue such as diabetes mellitus or hypothyroidism, or were taking medications that could lead to fatigue or interfere with muscle function. This included antihypertensives, antianginals, diuretics and antidepressants. Subjects refrained from caffeine or vigorous exercise on the morning of the experiment. This study was approved by the Newcastle and North Tyneside local research ethics committee and all subjects provided written informed consent.
Assessment of severity of fatigue
In order to quantify fatigue, all participants completed the generic fatigue measure, the Fatigue Impact Scale (FIS). This tool has been validated for self-completion in both normal populations and CFS patients [25, 26]. The range of potential scores is 0–160 with higher scores indicating increased fatigue.
Magnetic resonance spectroscopy (MRS) acquisition
Magnetic resonance spectroscopy data were acquired using a 3T Intera Achieva scanner (Philips, Andover, MA, USA) with a 14-cm diameter  P surface coil for transmission/reception of signal and the in-built body coil for anatomical imaging. A purpose-built exercise apparatus was developed for operation within the magnetic resonance imaging scanner. This apparatus permitted a controlled plantar flexion between 0o (foot vertical) and 30o to exercise the soleus and gastrocnemius muscles with the patient lying supine: restraining straps prevented the recruitment of other muscle groups (e.g. quadriceps). Subjects performed an exercise protocol, consisting of 3 min rest, 3 min of plantar flexion at 0.5 Hz and 3 min of rest to measure recovery to equilibrium. Exercise involved a fixed load of 35% of the maximum voluntary contraction (determined prior to spectroscopy) to produce anaerobic metabolism and allow evaluation of pH handling. Phosphorus spectra were collected at 10 s intervals throughout the exercises using an adiabatic 1D-ISIS sequence to localise signal to gastrocnemius and soleus muscles. Quantification of phosphocreatine (PCr), inorganic phosphate and ATP was performed using the AMARES time domain fit routine in the jMRUI processing software . The changes in PCr and inorganic phosphate were used to calculate parameters of oxidative metabolism during the exercise period: a single exponential fit was used to estimate the half time for PCr recovery. The chemical shift (i.e. the horizontal displacement) of the inorganic phosphate peak from the phosphocreatine peak is dependent on the muscle pH, and therefore by measuring this distance, muscle pH at every time-point can be determined . The pH at three key time-points is examined: (i) the baseline muscle pH which can demonstrate resting metabolic imbalances (e.g. in conditions with severe mitochondrial myopathies), (ii) the pH immediately postexercise which reflects the degree to which anaerobic pathways have been activated during exercise, and (iii) the minimum pH postexercise (referred to as ‘nadir pH’) which indicates the maximum burden of acid clearance imposed by anaerobic metabolism and the resynthesis of phosphocreatine . In addition, time to pH recovery was assessed by measuring the time from the cessation of exercise until pH returned to within 0.01 units of its preexercise value. If pH is higher or equal to its initial value at cessation of exercise, due to the pH-raising effect of PCr hydrolysis, pH recovery time is taken to be zero.
Because the change in muscle pH and the change in phosphocreatine concentration is measured for every time-point proton efflux (PE) from the muscle can be calculated for every time-point after cessation of exercise, using the formula:
where V is the PCr synthesis rate and the buffer capacity β is taken as 20 slykes . The first term estimates the proton generation caused by the resynthesis of PCr following exercise: the second term estimates the protons accounted by the change in muscle pH. Sum of these two terms are the protons estimated to be leaving the muscle studied. For healthy controls, the time course of proton efflux postexercise is a monotonic decay from a maximum at end-exercise, with the maximum proton efflux proportional to nadir pH. Therefore, the proton efflux behaviour can be characterised by measuring: (i) the initial proton efflux postexercise, the immediate postexercise response to acidosis, which should be maximal in control subjects and reduced if there is impairment of proton clearance, (ii) the time to maximum proton efflux which will be zero in control subjects and greater than zero if efflux is delayed, (iii) the maximum proton efflux which should be proportional to the nadir pH under normal metabolic regulation.
Autonomic nervous system assessment
Subjects also attended for assessment of the integrity of the autonomic nervous system. This was performed at the same time of the day in all subjects in a room with consistent ambient temperature. Subjects rested supine for 10 min whilst heart rate and blood pressure were monitored using continuous beat to beat digital photophlethysmography (Taskforce; CNSystems, Graz, Austria). Heart rate variability (HRV) was derived using spectral analysis  to derive total heart rate variability (Total HRV), low frequency HRV (LF; predominantly sympathetic), high frequency HRV (HF; predominantly parasympathetic) and very low frequency HRV (VLF).
The CFS/ME patients were substantially more fatigued than normal controls (median FIS scores 94 [range 48–137] vs. 2 [0–11], P < 0.0001). Results of the resting heart rate variability are shown in Table 1. Maximum voluntary contraction and muscle volume were; however, similar in CFS/ME patients and in normal controls (MVC: control 18.4 ± 4.9 kg, CFS/ME, 14.7 ± 3.9 kg, P = ns. Muscle volume: controls, 541 ± 71 cm3, CFS/ME, 601 ± 144 cm3, P = ns). All subjects were able to fully comply with the demands of the protocol, and to deplete PCr concentrations sufficiently to calculate proton efflux, even where pH changes were small.
Table 1. Heart rate variability (HRV) in the CFS patients compared to controls expressed as mean ± SD. Low frequency (LF), high frequency (HF) and very low frequency (VLF) HRV
CFS/ME, chronic fatigue syndrome.
1216 ± 1382
1150 ± 1023
440 ± 404
420 ± 276
241 ± 167
357 ± 339
570 ± 1373
353 ± 316
2.6 ± 2.1
1.5 ± 0.5
Following exercise at 35% of maximum voluntary contraction, magnetic resonance spectroscopy was used to explore the intramuscular acid handling in the postexercise recovery period. The muscle pH of the control and CFS/ME groups was not significantly different either (i) at baseline (CFS: 7.03 ± 0.03, control: 7.04 ± 0.02), (ii) immediately postexercise (CFS: 7.04 ± 0.04, control: 7.01 ± 0.02) or (iii) at the minimum value seen during recovery (CFS: 6.98 ± 0.09, control: 6.99 ± 0.02). Whilst there is a common monotonic decrease in proton efflux with time in all the controls, in agreement with previous work in healthy controls , the time course of efflux in CFS/ME patients is not uniform and does not decrease monotonically. CFS/ME patients had significant suppression of proton efflux immediately postexercise (Fig. 1a), the time taken to reach maximum proton efflux was significantly prolonged in CFS/ME patients (Fig. 1b), and the magnitude of maximum proton efflux was reduced compared to the controls (Fig. 1c). These findings suggest that there is significant impairment of both the level of proton excretion in recovery phase following exercise, and a prolongation of the kinetics of that recovery. In simple terms, the CFS/ME patients recover substantially more slowly than normal controls. As has been previously described in healthy controls of all ages , the rate of maximum proton efflux showed a strong inverse correlation with nadir pH following exercise (Fig. 2). This is a physiological response whereby the level of acidosis regulates the resulting stimulation of proton efflux tailoring to the need for that recovery. In CFS/ME patients, in contrast, this close relationship is absent with no relationship being seen between stimulus to pH recovery (pH) and the rate of that recovery (even with removal of extreme outliers, no correlation persists). In normal individuals the time for pH recovery following exercise is closely correlated with autonomic nervous system parameters: there is a strong correlation between total HRV and time to pH recovery (r2 = 0.7; P = 0.007, Fig. 3a). This relationship was particularly strong with very low frequency (VLF) HRV (r2 = 0.8; P = 0.006) although the relationship was still present (in a weaker form) with both low frequency (LF) and high frequency (HF) HRV (r2>0.6; P = 0.01). The association between autonomic nervous system regulation and pH recovery appears to be absent in CFS/ME patients (Fig. 3b) with none of the autonomic parameters correlating with time to pH recovery. There was no significant differences in the postexercise recovery rate of PCr (CFS: 32.7 ± 9.0 s, controls: 27.2 ± 7.1 s), indicating no significant difference in mitochondrial oxidative function.
In this study, we have demonstrated that patients with CFS/ME have substantial abnormalities in the recovery of intramuscular pH following a standardised level of exercise. Proton efflux from muscle (critical for acidosis resolution) is substantially lower immediately postexercise: in normal controls, it is well established that this is the point of maximal proton efflux. In CFS/ME patients, this immediate fast removal of protons from the muscle does not occur. The time for proton efflux to reach a maximum is significantly prolonged compared to control subjects: the peak rate of proton efflux is also substantially impaired in CFS/ME patients. There is a close relationship between the degree of acidosis and proton efflux suggesting a closely regulated process: this has been observed in controls in other studies , and the pH changes observed in the control group are also in agreement with similar studies [27, 32]. In CFS/ME patients this relationship is lost. In addition, autonomic dysfunction assessed in the current study by total heart rate variability, is associated with dysregulation of pH recovery. There are two ways in which autonomic dysfunction could lead to abnormalities of muscle function. Studies have suggested that the autonomic nervous system (particularly the sympathetic branch) plays an important role in the drive of transporters in muscle that remove acid, in particular, the sodium/proton anti-porter . A further mechanism that requires evaluation is the effect that the autonomic nervous system has upon the calibre of vessels ‘draining’ blood from the muscle it is possible that this may also have an impact upon the muscles ability to remove acid down stream.
Previous studies of CFS/ME patients using phosphorus spectroscopy have concentrated on mitochondrial oxidative function, which has been found to be comparable to that seen in matched controls in a number of different muscles [34–37]. Our study confirms that the same is observed in the soleus and gastrocnemius. We believe that this is the first study to quantify rates of proton efflux in CFS/ME and our findings would suggest that this methodology has the potential to be used as an end-point in clinical trials of targeted interventions in this patient group.
The design of this study precludes us from exploring causality of the relationships identified and further studies are needed to identify the cause of the muscle abnormalities and whether their relationship with autonomic dysfunction is cause or consequence. One explanation for our findings would be that the impaired proton efflux following exercise reflects a de-conditioning phenomenon of the type that has been postulated to occur in CFS/ME, although the data regarding maximum voluntary contraction and muscle volume argue against this. In order to minimise the effect of inactivity, we also importantly compared our results from CFS/ME patients to those found in sedentary controls. Given the role that acidosis within muscles directly plays in the perception of fatigue this association is likely to be of biological significance to patients even if it is not the primary insult in the disease. If proton efflux is abnormal in CFS/ME as a de-conditioning phenomenon it is still likely that it plays a role in the clinical expression of the disease making it a reasonable therapeutic target. Given the role played by the autonomic nervous system in regulation of acid transporter pathways in muscle and impact upon the vascular flow around muscles, the alternative explanation is that these phenomenon represent the mechanism (or one of many mechanisms) by which autonomic dysfunction results in a clinical expression of fatigue.
The potential significance of our findings is that they identify a clear potential approach to therapy. The area of exercise based therapy in CFS/ME is controversial with many patient reports suggesting that exercise can often make them feel worse [35, 38]. Our findings would add further evidence to the suggestion that unstructured approaches to exercise which led to further proton efflux dysregulation would be expected to exacerbate the symptoms experienced by patients in the short term. There are clear data; however, to suggest that modulated exercise can be associated with substantial changes in the expression of proton transporter proteins within muscle and increased functional acid excretion [20, 21]. Taken together these observations suggest that carefully structured exercise, ideally using bio feedback and the information obtainable by dynamic studies of the type described here, can be expected lead to normalisation in proton excretion which, we would postulate, would be associated with symptomatic improvement. This hypothesis is easily testable and the development of such regimes would, if monitored against biological markers of the acid exclusion pathway, be safe in this disease.
The identification of organic processes occurring in the periphery in CFS/ME patients is an important step in understanding pathophysiology and changing perceptions of the nature of the disease. The observations could give rise to novel managed approaches to therapy which will help alleviate the symptoms experienced by this patient group.
Medical Research Council, ME Research UK, CFS/ME Northern Clinical Network.
None of the funders contributed to the design, performance or interpretation of the results of this study.
Conflict of interest statement
None of the authors have any conflict of interest.