Abstract
Background Both good physical and cognitive functioning have a positive influence on the execution of activities of daily living. Patients with chronic fatigue syndrome (CFS) as well as patients with fibromyalgia have marked cognitive deficits. Furthermore, a good physical and functional health status may have a positive impact on a variety of cognitive skills—a link that has been observed in young and old individuals who are healthy, although evidence is limited in patients with CFS.
Objective The purpose of this study was to examine whether recovery of upper limb muscle function could be a significant predictor of cognitive performance in patients with CFS and in patients with CFS and comorbid fibromyalgia. Furthermore, this study determined whether cognitive performance is different between these patient groups.
Design A case-control design was used.
Methods Seventy-eight participants were included in the study: 18 patients with CFS only (CFS group), 30 patients with CFS and comorbid fibromyalgia (CFS+FM group), and 30 individuals who were healthy and inactive (control group) were studied. Participants first completed 3 performance-based cognitive tests designed to assess selective and sustained attention, cognitive inhibition, and working memory capacity. Seven days later, they performed a fatiguing upper limb exercise test, with subsequent recovery measures.
Results Recovery of upper limb muscle function was found to be a significant predictor of cognitive performance in patients with CFS. Participants in the CFS+FM group but not those in the CFS group showed significantly decreased cognitive performance compared with the control group.
Limitations The cross-sectional nature of this study does not allow for inferences of causation.
Conclusions The results suggest that better physical health status could predict better mental health in patients with CFS. Furthermore, they underline disease heterogeneity, suggesting that reducing this factor in future research is important to better understand and uncover mechanisms regarding the nature of diverse impairments in these patients.
Chronic fatigue syndrome (CFS) is a debilitating clinical disorder characterized by at least 6 consecutive months of self-reported, unexplained, persistent or relapsing fatigue that is accompanied by 4 or more of the following symptoms (also present for at least 6 months): impaired memory or concentration, sore throat, new-onset headaches, tender cervical or axillary lymph nodes, muscle pain, multijoint pain, unrefreshing sleep, and postexertional malaise.1 Some authors2,3 have reported that 43% to 70% of patients with CFS also meet the diagnostic criteria for fibromyalgia, another chronic debilitating condition. Fibromyalgia is characterized by widespread musculoskeletal pain, often associated with symptoms of fatigue, sleep disturbances, cognitive dysfunction, and mood disturbances.4,5
Indeed, a substantial overlap exists between fibromyalgia and CFS.2,6 Both disorders have similar symptoms,2 and an increasing amount of scientific evidence indicates that overlapping conditions such as CFS and fibromyalgia are bound by a common pathophysiological mechanism of central sensitization.7–10 Central sensitization represents an enhanced responsiveness of central neurons to nociceptive and non-nociceptive stimuli.11 It manifests as an increased response to various peripheral stimuli (eg, pressure, light, sound, cold, heat) inducing hyperalgesia and allodynia.9,11
A potential characteristic of central sensitization as a dysfunction of the central nervous system is the cognitive impairment that is reported very frequently by both patients with CFS and patients with fibromyalgia. Slowed response speed appears to be the most prominent and consistently affected cognitive function in patients with CFS.12–16 However, inconsistencies about deficits in other cognitive domains have been reported,17–23 and these inconsistencies, among others, may be due mainly to the heterogeneity of this disorder. Indeed, similar to CFS, cognitive deficits are experienced as very troublesome complaints by patients with fibromyalgia as well. Cognitive impairments seen in these patients, however, are different from those seen in patients with CFS, with impaired working memory performance being most consistently reported.5 Disease heterogeneity and the effect of comorbid illnesses such as fibromyalgia have not yet been studied extensively in relation to cognitive function or dysfunction in CFS and thus may explain, in part, the inconsistent findings in the literature.
In addition, good physical and functional health status may have a positive impact on a variety of cognitive skills. This positive link between physical and mental health has been observed in young and old individuals who were healthy.24,25 Although the little research that has been conducted supports a positive relationship between physical and mental health,12,26,27 evidence is limited in patients with CFS. Furthermore, both good physical and cognitive functioning have a positive influence on the execution of activities of daily living.28
The literature suggests the existence of feedback systems starting from peripheral structures (eg, muscles and joints) that influence brain activity through signaling systems.29 Because motor or physical activity is linked to cognitive functioning and to neuroplasticity,29 and obviously also is connected to muscle fatigue and recovery, examining cognitive performance and its association with recovery of peripheral muscle function (as 2 possible features of central sensitization) could indirectly reflect neural health in patients with CFS.
For the reasons outlined above, this study aimed at examining whether recovery of upper limb muscle function could be a significant predictor of cognitive performance in patients with CFS and patients with CFS and comorbid fibromyalgia. A second aim was to investigate whether cognitive performance is different between these patient groups.
Method
Study Design and Setting
This study was designed as a blinded case-control study in line with the STROBE Statement.30 All assessments took place at the Pain in Motion research laboratory of Artesis University College Antwerp and Vrije Universiteit Brussel between February 2011 and December 2012. All participants gave written informed consent prior to commencement of the study.
Participants and Assessments
The study sample consisted of 3 groups of participants. Patients with a diagnosis of CFS were split into a group of patients with CFS only (CFS group) and a group of patients with CFS and comorbid fibromyalgia (CFS+FM group). The third group consisted of individuals who were healthy and inactive (control group). Each study participant had to be Dutch speaking and aged between 18 and 65 years.
Patients with CFS were recruited from a private practice for internal medicine through calls during patient information sessions and advertisements placed in the newsletter of a local patient support group. Written confirmation of a CFS diagnosis, as defined by the Centers for Disease Control and Prevention (CDC) 1994 criteria for CFS,1 was required from each participant's physician before study participation. The comorbid diagnosis of FM was identified according to the American College of Rheumatology (ACR) 2010 criteria for fibromyalgia.31 To satisfy these diagnostic criteria for fibromyalgia, the following 3 conditions have to be met: (1) Widespread Pain Index (WPI) score ≥7 and symptom severity (SS) scale score ≥5, or WPI score 3–6 and SS scale score ≥9; (2) symptoms have been present at a similar level for at least 3 months; and (3) the patient does not have a disorder that would otherwise explain the pain.31 Condition 3 also is required in the diagnosis of CFS and has consequently been ruled out by the patient's physician. As mentioned in the original article by Wolfe et al,31 this simple clinical case definition of fibromyalgia correctly classifies 88.1% of cases classified by the ACR classification criteria and does not require a physical or tender point examination. Therefore, and to ensure blinding of the assessor (physical therapist), the other conditions (WPI, SS, and symptom duration) were identified using a self-reported questionnaire. The control group comprised individuals who were healthy (pain-free and without any chronic disease) and inactive and was drawn from relatives, friends, or acquaintances of researchers, students, university personnel, or patients participating in the study. Inactive was defined as working in an occupation that did not require moderate to intense physical labor and performing a maximum of 3 hours of moderate physical activity per week. Moderate physical activity is defined as activity demanding at least threefold the energy spent passively.32
In order to preclude confounding factors, participants could not have intellectual disabilities and could not be pregnant or less than 1 year postnatal. Furthermore, all participants, if applicable, were asked to stop taking antidepressive, antiepileptic, and pain medications 2 weeks prior to study participation and were asked not to undertake physical exertion and to refrain from consuming caffeine, alcohol, or nicotine on the days of the assessments. For ethical reasons, participants were able to take nonopioid analgesics as described in the first step of the World Health Organization analgesic ladder (nonsteroidal anti-inflammatory drugs, acetylsalicylic acid, and paracetamol) during the study period but not on the days of the assessments.
The study consisted of 2 assessment sessions, separated by 7 days and performed by the same assessor, who was blinded to whether participants were patients or controls (Fig. 1). On the first day, after signing the informed consent form, collecting data on personal characteristics (age, sex, height, weight, disease duration, fibromyalgia criteria, and occupational situation) and checking for the presence of possible confounders, all participants completed 3 performance-based cognitive tests on a computer. Seven days later, on the second testing day, they performed a fatiguing exercise test and subsequent recovery measures with a hydraulic handheld dynamometer. Finally, at the end of the second assessment session, the success of assessor blinding was examined by asking whether the assessor thought the participant belonged to a patient group (CFS or CFS+FM) or to the control group.
Flow chart of the study. CDC=Centers for Disease Control and Prevention, CFS=chronic fatigue syndrome, FM=fibromyalgia.
Cognitive Tests
To investigate cognitive function, the Stroop task, the psychomotor vigilance task (PVT), and the operation span (OSPAN) task with concomitant mathematical processing were used. All 3 tests were conducted on the same computer and in the same order (Stroop task, PVT, and OSPAN) by every participant. To ensure standardization of the procedure, each test began with the presentation of written instructions for that particular test. Short breaks (±5 minutes) were allowed between tests. Completion of the entire test battery took about 50 minutes. Each of the 3 tests has been used and described in detail in 2 of our previous studies of female patients with CFS.12,33
The Stroop task34 was used to assess selective attention, cognitive inhibition, and choice reaction time (RT). In this test, different stimuli (words or “XXX”) appeared in different colors (yellow, green, red, or blue) in the middle of the computer screen. The meaning of the stimulus is the task-irrelevant dimension, and the color in which the stimulus is presented is the task-relevant dimension. Accordingly, participants were instructed to respond to the presented ink color in which stimuli were written by pressing the corresponding color key on the keyboard as quickly and as accurately as possible. The presented stimuli could be classified under 2 different conditions: “incongruent” (word and color are different [eg, the word “red” displayed in green]) and “no word” (“XXX” presented in one color). Mean response RT and accuracy were recorded for each condition. In order to quantify cognitive inhibition, the RT of the no-word condition was subtracted from the RT of the incongruent condition. Thus, an interference score is provided, which can be seen as an indicator of the inhibition subcomponent of executive functioning. Deficits in cognitive inhibition should result in an increased Stroop interference score.
The PVT35 was used to assess sustained attention (or vigilance) and simple RT. Participants were instructed to respond to a visual stimulus (red spot on black background) that appeared in the middle of the screen at random interstimulus intervals (2–10 seconds). They were required to press the mouse button as quickly as possible whenever they perceived the appearance of the stimulus on the screen. If the participant did not respond within 500 milliseconds, the trial was recorded as a lapse. The PVT ran for a total time of 10 minutes. The mean RT of correct responses (<500 milliseconds) and the number of lapses were recorded.
Working memory capacity was assessed using the OSPAN task with concomitant mathematical processing as described by Conway and Engle.36 The task began with a practice block (divided into 3 sections). First, participants had the chance to practice the simple letter span. They saw letters appear on the screen one at a time. After having seen the whole letter span, they had to recall these letters in the same order in which they saw them.
Next, participants practiced the mathematical portion of the OSPAN task. They first saw a mathematical operation appear on the screen (eg, [7 × 3]−3=?). Then, a number (eg, 18) was presented on the screen, and participants were instructed to respond if the number was the correct solution by clicking on “True” or “False.” The final practice session consisted of performing both the letter recall and the mathematical operations together. Participants first had to solve the mathematical operation and only then saw the letter to be recalled. The dual-task design with the mathematical processing was used to keep the task-relevant information (letter span) active and accessible in memory during the execution of complex cognitive tasks (mathematical operations). After the completion of the 3 practice sessions, the program automatically proceeded to the experimental block, which was the same as the final practice session (mathematical operations + letter recall). The experimental block consisted of 3 sets of each set size (ranging from 3 to 7). Thus, a total of 75 letters and 75 mathematical problems were presented. At the end of the experiment, the OSPAN total score was registered and used for further data analyses. This score indicates the number of letters recalled in the correct position (regardless of whether the whole letter set was correct) and is a measure of working memory capacity.
Fatiguing Exercise Test and Recovery Measurements
The procedure utilized to assess recovery from fatiguing exercise in this study was used and described in a previous study by our group.37 The method is briefly explained here. For details, the reader is referred to the previous study.37 A hand exercise was used because our clinical experience indicates that many patients with CFS (with or without comorbid fibromyalgia) report muscle fatigue and abnormally slow recovery, especially of the upper limb muscles during and after functional activities of daily living such as combing and washing hair, ironing, cooking, and cleaning (eg, windows).
Recovery from fatiguing exercise was measured using a hydraulic handheld dynamometer (Saehan Corp, Masan, Japan). Participants were asked to sit on a chair while holding the dynamometer in their nondominant hand. First, participants were instructed to grip the instrument as hard as possible (baseline isometric maximum voluntary contraction [MVC]). Next, every participant performed a fatiguing exercise test consisting of 18 maximum contractions using a 50% duty cycle (5-second contraction, 5-second rest). After the fatiguing exercise test (recovery phase), participants were instructed to make single 5-second isometric MVCs at time intervals of 0, 5, 10, 15, 20, 30, and 45 minutes postexercise. These values were converted into percentages of baseline MVC, with the latter being taken as 100%. For statistical analyses, the recovery data were split into 3 sections: MVCstart (MVC at 0 minutes postexercise), MVCmid (mean of the MVCs from 5 to 30 minutes postexercise), and MVCend (MVC at 45 minutes postexercise).
Hydraulic instruments are the most widely used devices that measure grip strength. The handheld dynamometer is a reliable instrument and an objective way of measuring strength in elderly people and people who are physically impaired. It is reported to have a good test-retest reliability and interrater reliability for clinical use (intraclass correlation coefficient >.90) in people who are healthy38–40 as well as in people with muscular dystrophy, cumulative trauma disorders, and lateral epicondylitis.38
Data Analysis
Data analyses were performed using the Statistical Package for the Social Sciences 20.0 for Windows (IBM Corp, Armonk, New York). Normality of the variables was tested using the Kolmogorov-Smirnov goodness-of-fit test and through visual inspection of the histograms and distribution graphs. Comparability of the groups was studied with a Pearson chi-square test for sex and occupational situation and with a one-way independent analysis of variance (ANOVA) for age, body mass, height, body mass index, and disease duration.
Cognitive performance was compared among the groups with a one-way independent ANOVA for variables that were normally distributed and variables that were normalized through logarithmic transformation. When a significant main effect was found, Bonferroni post hoc comparisons were performed to identify significant differences among the 3 groups. Variables still lacking normal distribution after transformation were compared with a nonparametric Kruskal-Wallis test. Post hoc paired comparisons were performed when a main effect was found. To remove the bias of possible confounding variables such as time of cognitive testing and occupational situation, these analyses were repeated once again using analysis of covariance (ANCOVA). Time of cognitive testing and occupational situation were separately entered as covariates into the analyses.
To determine the association between recovery of upper limb muscle function and cognitive performance correlation analyses (Pearson and Spearman) were performed. When calculating the statistical significance of the correlations, no correction for multiple comparisons was done due to the exploratory nature of this study and to minimize the risk of type II errors. For all correlation analyses, only the MVC during recovery that revealed the strongest association with performance on cognitive tests (based on P values and on correlation coefficients) was considered for further regression analyses. Hence, the outcome of the correlation analyses was used only for identifying the appropriate variables for the regression analysis.
Simple linear regression analyses were performed to determine whether recovery of upper limb muscle function could be a significant predictor of cognitive performance in the CFS and CFS+FM groups. Because age may significantly influence cognitive function as well as recovery of peripheral muscle function, the analysis also was performed with age entered as additional predictor variable (covariate).
For all comparisons (except for the correlation analyses), P<.05 (2-sided) was considered statistically significant. Data are reported as means (SD) within the text and in Table 1 and as means (standard error of the mean) in Figure 2.
Demographic Data of the Study Samplesa
Cognitive performance in patients with chronic fatigue syndrome (CFS group, n=18), patients with chronic fatigue syndrome + fibromyalgia (CFS+FM group, n=30), and participants who were healthy and inactive (control group, n=30). Values are means±standard errors of the mean. RT IC=reaction time in incongruent condition, RT NWC=reaction time in no-word condition, PVT=psychomotor vigilance task, OSPAN=operation span. (A) Reaction times and accuracy percentages for the incongruent condition of the Stroop task. (B) Reaction times and accuracy percentages for the no-word condition of the Stroop task. (C) Stroop interference scores. (D) Reaction times and number of lapses on the PVT. (E) Total scores on the OSPAN task. Statistically significant differences between the CFS+FM or CFS group and the control group are indicated with asterisks (*P<.05, **P<.01, ***P<.001). Statistically significant differences in Stroop reaction times (graphs A and B) between the CFS+FM group and the control group became nonsignificant after controlling for occupational situation (P>.05).
Power calculations were performed with the program G*Power 3.1.5 (Franz Faul, Kiel, Germany).41 The a priori sample size calculation was based on the results of our previous study on the association between cognitive performance and physical fitness and activity in women with CFS.12 The calculation revealed that a total sample size of at least 39 participants would provide 82% power with α=.05 to detect a difference in cognitive performance among the 3 groups. The post hoc power analysis also indicated a power of >99% (N=78, α=.05). Because the relationship between upper limb muscle recovery and cognitive performance has not been investigated previously, no data were available to provide a basis for an a priori power analysis. Hence, a post hoc power analysis was performed for the linear regression analyses (simple and multiple). For the simple linear regression analyses, the post hoc power analysis revealed 97% power (n=30, α=.05) in the CFS+FM group and 75% power (n=18, α=.05) in the CFS group. Adding an extra predictor (age) to the model yielded 94% and 71% power in the CFS+FM and CFS groups, respectively.
Results
This report is the second from a study examining, among other factors, the association between cognitive performance and various aspects of central sensitization in patients with CFS. For the results on recovery of upper limb muscle function in patients with CFS with and without comorbid fibromyalgia, we refer the reader to our first report.37
Group Characteristics
Forty-eight patients with CFS (30 with fibromyalgia and 18 without fibromyalgia) and 30 control participants were included. Hence, 62.5% of the patients with CFS also met the ACR 2010 criteria for fibromyalgia. Demographic data of the 3 study samples are listed in Table 1. The groups were comparable for sex distribution, age, body weight, and body mass index (P>.05) but not for body height (F2,75=4.69, P=.012) and occupational situation (χ26=17.74, P=.007). Disease duration was not significantly different (P=1.00) between the 2 patient groups.
Cognitive Performance
Cognitive performance of the CFS+FM, CFS, and control groups is shown in Figure 2. The 1-way independent ANOVA indicated significant main effects for Stroop RT of the incongruent condition (F2,75=4.85, P=.01), Stroop RT of the no-word condition (F2,75=4.53, P=.014), PVT RT (F2,75=21.18, P<.001), and PVT lapses (F2,75=9.39, P<.001). Subsequent post hoc analyses revealed that participants in the CFS+FM group were significantly slower (longer RTs) than controls in both the incongruent condition (P=.008) and the no-word condition (P=.011) of the Stroop task (Figs. 2A and 2B, respectively). Furthermore, both the CFS+FM group and the CFS group showed significantly longer RTs on the PVT than controls (P<.001 and P=.002, respectively). The CFS+FM group also showed a significantly higher percentage of lapses on the PVT compared with the control group (P<.001) (Fig. 2D). No significant main effects for OSPAN total score were found (F2,75=1.62, P=.204) (Fig. 2E). The nonparametric Kruskal-Wallis test revealed significant main effects for Stroop accuracy percentage of the incongruent condition (H2=8.58, P=.014) and for Stroop interference scores (H2=8.25, P=.016). No significant main effect was indicated for Stroop accuracy percentage of the no-word condition (H2=0.11, P=.945) (Fig. 2B). Post hoc paired comparisons indicated a significantly lower accuracy percentage of the incongruent Stroop condition (P=.011) and a significantly greater Stroop interference score (stronger Stroop effect) (P=.013) in the CFS+FM group compared with the control group (Figs. 2A and 2C, respectively). In all post hoc comparisons, no statistically significant differences (P>.05) were found between the CFS and CFS+FM groups.
The ANCOVA revealed that the covariate time of cognitive testing (early morning, late morning, early afternoon, late afternoon) did not significantly predict any of the cognitive test results (P>.05). Hence, all results remained the same after controlling for time of cognitive testing. When “occupational situation” was entered as a covariate in the model, the ANCOVA indicated that the covariate significantly predicts OSPAN total score (P=.001) but not the other cognitive test results (P>.05). However, when the effect of occupational situation was removed (when controlling for occupational situation), the earlier mentioned significant differences in Stroop RTs became nonsignificant (P=.074 and P=.94 for the incongruent and no-word conditions, respectively). All other results remained the same after controlling for occupational situation.
Association Between Recovery of Upper Limb Muscle Function and Cognitive Performance
Correlation analyses.
In the CFS+FM group, the correlation analysis revealed significant negative correlations between recovery capacity at the beginning of the recovery phase (MVCstart) and Stroop RT of the incongruent and no-word conditions (r=−.43, P=.018, and r=−.45, P=.012, respectively) and Stroop interference (r=−.43, P=.017). Furthermore, significant associations were found between recovery capacity during the middle part of the recovery phase (MVCmid) and Stroop RT of the incongruent condition (r=−.36, P=.048), Stroop RT of the no-word condition (r=−.64, P<.001), PVT RT (r=−.48, P=.007), PVT lapses (r=−.59, P=.001), and OSPAN total score (r=.59, P=.001). Finally, in this patient group, recovery capacity at the end of the recovery phase (MVCend) was significantly inversely related to PVT RT (r=−.39, P=.032).
In the CFS group, MVCstart was not significantly related to cognitive performance (P>.05). There was a significant negative association between MVCmid and PVT lapses (r=−.55, P=.017), and both MVCmid and MVCend were significantly inversely related to Stroop RT of the no-word condition (r=−.58, P=.012, and r=−.47, P=.048, respectively).
No significant correlations between recovery capacity of upper limb muscle function and cognitive performance were found in the control group (P>.05).
Regression analyses.
Based on the outcome of the correlation analyses, recovery capacity during MVCmid was considered for further regression analyses. To determine whether recovery of upper limb muscle function could be a significant predictor of cognitive performance in patients with CFS and in patients with CFS and comorbid fibromyalgia, simple linear regression analyses were performed with MVCmid entered as the predictor variable. Table 2 presents the results of the regression analyses in both patient groups (CFS and CFS+FM) with and without age entered as a covariate. The analyses revealed that, in patients with CFS+FM, MVCmid was a significant predictor of cognitive performance in all investigated cognitive domains (P<.01). These results remained significant (P<.01) when age was included as a covariate in the analyses. In the CFS group, MVCmid was a significant predictor of Stroop RT performance (R2=.27, β=−.52, P=.026 for the incongruent condition and R2=.33, β=−.58, P=.012 for the no-word condition), Stroop interference score (R2=.24, β=−.49, P=.038), and number of lapses on the PVT (R2=.31, β=−.55, P=.017) but not for PVT RT (R2=.20, β=−.45, P=.061) and performance on the OSPAN task (R2=.12, β=.34, P=.169). When age was entered as a covariate, MVCmid remained a significant predictor for Stroop RT on the no-word condition (R2=.47, β=−.47, P=.030) and for PVT lapses (R2=.32, β=−.59, P=.017) and became a significant predictor for PVT RT (R2=.24, β=−.50, P=.049).
Results of Simple Linear Regression Analyses Predicting Cognitive Performance on Recovery of Upper Limb Muscle Function in Patients With Chronic Fatigue Syndrome and Fibromyalgia (CFS+FM Group) and Patients With Chronic Fatigue Syndrome Only (CFS Group)a
Success of Assessor Blinding
In the CFS and CFS+FM groups, the assessors' assumption regarding disease status was correct in 62.5% (30 out of 48) of the cases. In the control group, the assessors guessed correctly in 90% (27 out of 30) of the cases.
Discussion
This study was the first to investigate whether recovery of upper limb muscle function could be a significant predictor of cognitive performance in patients with CFS. In addition, to extend previous investigations, patients with CFS were divided into 2 groups (CFS only and CFS+FM) to determine the influence of comorbid fibromyalgia on cognitive performance of patients with CFS.
We found that recovery of upper limb muscle function was a significant predictor of cognitive performance in the CFS+FM group. Interestingly, recovery of upper limb muscle function seemed to be a significant predictor of cognitive performance in the CFS group but only for the cognitive domains of simple and choice RT and sustained attention.
The results of this study demonstrated that the CFS+FM group showed significantly decreased RT (simple and choice), cognitive inhibition, and selective and sustained attention in comparison with the control group. Although the CFS group also showed worse results on cognitive tests compared with the control group, these differences were significant only for simple RT.
Cognitive Performance and Recovery of Upper Limb Muscle Function
This study revealed that recovery of upper limb muscle function is a significant predictor of cognitive performance in patients with CFS, with and without comorbid fibromyalgia. There are several ways in which a better recovery of muscle function could lead to better cognitive performance. In a previous study,12 we demonstrated the presence of a significant positive association between physical fitness and cognitive performance in female patients with CFS. Indeed, the benefits of better physical performance on a wide range of cognitive domains have been demonstrated in people who were healthy as well in patients with chronic pain.42–44 However, the biological mechanisms that underlie these beneficial effects are still not fully elucidated. Different systems (eg, nervous, cardiovascular, and endocrinological) and mechanisms (eg, supramolecular, molecular, and cellular) may interact at different levels.
Although many exciting findings have been made in the past 2 decades, it remains a challenge to understand the complex relationship between these different systems and mechanisms and the way in which they mediate the benefits of physical health on cognitive function.29 Taking into account what is already known, improvements in physical and functional health status could possibly lead to an up-regulation of brain plasticity. De Lange et al45 showed a significant relationship between gray matter volume and the amount of general physical activity in patients with CFS. In this perspective, it is plausible that the link between recovery of muscle function and cognitive function in patients is driven by neurogenesis, angiogenesis, or both, possibly involving at the molecular level the increased availability of several growth factors such as brain-derived neurotrophic factor, insulin-like growth factor 1, and vascular endothelial-derived growth factor. However, we do not exclude the possibility that other mechanisms also are involved and that different biological mechanisms underlie exercise, physical status, and cognitive function in patients with CFS, with and without comorbidities such as fibromyalgia.
Cognitive Performance
The CFS+FM group exhibited cognitive deficits that are consistent with previous findings in patients with CFS.12–16 Additionally, the cognitive deficits demonstrated by the CFS group, although less pronounced than in the CFS+FM group, are in accordance with these findings. The CFS group demonstrated significantly worse performance on simple RT (PVT) compared with the control group, whereas the CFS+FM group performed worse on all cognitive domains except on working memory performance. The finding that cognitive deficits were more pronounced in patients with CFS and fibromyalgia in comparison with those without fibromyalgia, although not in accordance with the findings of Cook et al,16 is in line with our expectations. Pain has been shown to impair cognitive performance in patients with chronic pain in general46 and in patients with fibromyalgia in particular.47 Furthermore, recently our group was the first to investigate the association between cognitive performance and pain severity in CFS patients. Interestingly, the results of this study revealed no association between pain severity and cognitive functioning in female patients with CFS.33 It has been shown that both patients with fibromyalgia and those with CFS exhibit reduced cerebral gray matter volume, which is associated with worse cognitive performance.45,48,49
A possible explanation for our findings could be that this gray matter volume reduction causes, either directly or indirectly, cognitive impairment in these patients. However, we suggest a different neurobiological basis in fibromyalgia and CFS. In both conditions, it could be hypothesized that the presence of central sensitization might explain the reduced gray matter volume observed with persistent nociceptive input to the brain, in this way causing neuroplastic adaptations in terms of cortical plasticity. Subsequently, in patients with comorbid fibromyalgia, cognitive pathways could be influenced indirectly following changes in pain pathways, whereas cognitive performance might be influenced more directly by the persistent nociceptive input in patients with CFS.
Limitations, Strengths, and Suggestions for Further Research
The present study had a few limitations as well as several strengths that should be mentioned. First, the number of participants in the CFS group was rather small for the regression analysis. This limitation could possibly explain why recovery of muscle function was not found to be an equally strong predictor for cognitive function in both patient groups. However, for the comparisons among the groups, the study appeared to be sufficiently powered. Second, we did not include a test for malingering in our study protocol. However, we attempted to tackle this shortcoming by using the Stroop task as a diagnostic tool for identifying malingerers. This technique is based on the knowledge that the presence of a Stroop interference effect is a normal finding in people who are healthy as well as in patients with cognitive impairment. Patients with a high suspicion of malingering show an inversion of the Stroop interference effect, usually in combination with longer RTs, increased error percentages, or both.50 No malingerers were identified in this study. Finally, the cross-sectional and exploratory nature of this study should be highlighted. The cross-sectional nature of the study does not allow for inferences of causation. Given the absence of previous data regarding the association between cognitive performance and recovery of muscle function, an exploratory approach was used to test several hypotheses. Although we accounted adequately for possible type I and II errors in the statistical analyses, the latter entails that findings must be considered cautiously.
Besides these limitations, this study also had several strengths. The most important one is that our control group was matched to both patient groups for age, sex, and body mass index. Both CFS groups also were matched for disease duration. The participants in the control group had to be inactive because patients with CFS are known to have a more sedentary lifestyle. In this way, observed differences between groups could not be due to a higher activity level of the control group. Another important strength of this study is that we reduced the heterogeneity of the disease by including patients with CFS according to the same strict diagnostic criteria and by dividing this group into subgroups based on fibromyalgia comorbidity. Furthermore, we anticipated sources of bias such as pregnancy; use of medications, caffeine, alcohol, and nicotine; and physical exertion on the days of the assessments. Finally, we attempted to blind the assessor regarding participants' disease status. However, blinding was successful in only 37.5% of the patients and in only 10% of the controls.
Future studies should try to account for the aforementioned limitations. Furthermore, they should continue to explore the relationship among exercise, indexes of physical and functional health status, and cognitive function in patients with chronic pain. Taking into account the important role that pain plays in cognitive functioning, investigating the extent to which altered pain processing and endogenous pain inhibitory mechanisms might be related to cognitive dysfunctions in patients with CFS, with and without FM, might also reveal some exciting results. These results might even open doors for further applied research as well as fundamental research.
In conclusion, recovery of upper limb muscle function is a significant predictor of cognitive performance in patients with CFS, suggesting that better physical health status could possibly lead (indirectly) to an up-regulation of brain plasticity in these patients. Furthermore, worse cognitive performance was observed in patients with CFS and comorbid fibromyalgia compared with controls. This finding also was observed in patients with CFS only but to a far lesser extent, suggesting that reducing the heterogeneity of the disorder in future research is important to better understand and uncover mechanisms regarding the nature of diverse impairments in these patients.
The Bottom Line
What do we already know about this topic?
Patients with chronic fatigue syndrome (CFS) as well as patients with fibromyalgia often report debilitating cognitive problems. As already observed in young and old individuals who are healthy, good physical and functional health may have a positive impact on a variety of cognitive skills. In patients with CFS with and without fibromyalgia, evidence regarding this interrelationship is limited.
What new information does this study offer?
This study shows that recovery of upper limb muscle function (as a measure of physical health/fitness) is a significant predictor of cognitive performance in patients with CFS with and without cormorbid fibromyalgia, suggesting that better physical health could possibly lead indirectly to an up-regulation of brain plasticity in these patients. Furthermore, the results of this study indicate that patients with CFS and comorbid fibromyalgia have worse cognitive performance compared with participants who are healthy. This finding also was observed in patients with CFS only, but to a far lesser extent.
If you are a patient/caregiver, what might these findings mean for you?
Although the cross-sectional nature of this study does not allow for inferences of causation, these findings suggest that patients with CFS with and without fibromyalgia may experience cognitive benefits from better physical fitness.
Footnotes
Ms Ickmans, Dr Meeus, and Dr Nijs provided concept/idea/research design. Ms Ickmans and Dr Nijs provided writing. Ms Ickmans, Ms De Kooning, and Dr Lambrecht provided data collection. Ms Ickmans and Dr Pattyn provided data analysis. Ms Ickmans, Ms De Kooning, Dr Lambrecht, and Dr Nijs provided project management. Dr Meeus and Dr Nijs provided fund procurement. Dr Lambrecht provided study participants. Dr Meeus, Dr Lambrecht, and Dr Nijs provided facilities/equipment. Ms Ickmans, Dr Meeus, Dr Lambrecht, Dr Pattyn, and Dr Nijs provided consultation (including review of manuscript before submission). The authors thank Tinne Boey and Laura van Weijnen for their aid in the data input.
The study was funded by ME Research UK, a national charity funding biomedical research into myalgic encephalomyelitis/chronic fatigue syndrome. Dr Meeus is awardee of the 2012 Early Research Career Grant of the International Association for the Study of Pain (IASP), funded by the Scan Design Foundation by Inger and Jens Bruun. Dr Nijs is holder of the Chair “Exercise Immunology and Chronic Fatigue in Health and Disease,” funded by the European College for Decongestive Lymphatic Therapy, the Netherlands. Ms Ickmans is a research fellow of ME Research UK.
The study protocol was approved by the ethics committees of the University Hospital Brussels/Vrije Universiteit Brussel and the University Hospital Antwerp.
- Received September 4, 2013.
- Accepted December 12, 2013.
- © 2014 American Physical Therapy Association
References
Issue highlights
