Are Fibromyalgia Patients Cognitively Impaired? Objective and Subjective Neuropsychological Evidence

Abstract

Objective

Patients with fibromyalgia (FM) syndrome often report a cluster of cognitive disorders that strongly interferes with their work and daily life, but the relationship between impaired cognitive function and self-reported dysfunction remains unclear. We aimed to investigate the presence of cognitive impairments in patients with FM and to analyze the relationship between the impairments and their evaluation by the patients through a comparison with a group of healthy controls.

Methods

In total, 30 FM patients and 30 healthy controls performed a neuropsychological and clinical evaluation of short-term, long-term, and working memory; executive function; and self-evaluation of cognitive impairment and depressive and anxiety symptoms. To thoroughly investigate executive function, we adopted the Miyake model that identifies 4 domains: shifting, inhibition, updating, and access.

Results

Our results confirmed the presence of impairments of attention, long-term memory, working memory, and shifting and updating executive functions in FM patients compared with healthy controls. These impairments are reflected in patient reports independently of depressive symptoms.

Conclusion

The use of a self-reported questionnaire in clinical practice would provide a first and easy screen for the presence of cognitive impairment in FM patients and, in most cases, obviate the need for a time-consuming full neuropsychological test battery.

INTRODUCTION

Fibromyalgia (FM) is a chronic pain syndrome characterized by widespread musculoskeletal pain associated with a heterogeneous series of other symptoms, including fatigue, stiffness, disrupted or nonrestorative sleep, and psychological distress, particularly mood depression ([1, 2]). Patients with FM often experience the so-called “fibro fog,” a cluster of cognitive disorders not always reflected in poor test-based performance but that strongly interferes with work and daily life ([3, 4]). Bertolucci and de Oliveira reported that 50–80% of FM patients show a decline in working memory, attention, and executive function (EF) ([5]). However, EF represents a multifaceted construct composed of separable factors that tap into different cognitive mechanisms and possibly involve the activity of different brain structures. Considering EF as a whole would therefore not allow identification of subtle differences in cognitive experiences.

Although the presence of cognitive impairment has recently been added to the diagnostic criteria of the American College of Rheumatology (ACR) ([1]), cognitive dysfunction remains one of the least assessed and treated FM domains in general clinical practice because of the expertise and time required for neuropsychological tests. One possible strategy to avoid these hindrances is to use self-report tools, which may, however, be biased by the concomitant presence of depressive symptoms ([6]). Indeed, previous studies have found that depressive symptoms are the strongest single contributor to reports of cognitive deficits in chronic pain patients ([7, 8]).

Some important questions regarding the exact nature of cognitive deficits and the relationship between impaired function and self-reported dysfunction are therefore still unanswered. The present study aimed to address 2 issues: to analyze the neuropsychological performance of FM patients in short- and long-term memory, working memory, and EF by means of a comparison with pain-free healthy controls (HCs) and to investigate the relationship between objective performance on standard neuropsychological tests and subjective self-perception of cognitive status in everyday life through a specific questionnaire by means of explorative correlational analyses.

To thoroughly investigate EF, we referred to the model elaborated by Miyake and colleagues ([9]) and revised by Fisk and Sharp ([10]) that identifies 4 correlated but partially separable main factors: shifting, which involves the ability to engage and disengage attention from different tasks; inhibition, which implies holding back automatic or preponderant responses; updating, which is the ability to monitor and code information and replace old nonrelevant information with newly relevant information; and access, which mediates access to long-term memory representations and is involved in verbal fluency tasks.

Box 1. Significance & Innovations

  • Our study investigated the neuropsychological performance of fibromyalgia patients in short- and long-term memory, working memory, and executive function (EF), focusing on a multiple-domains approach for EF.
  • Our results showed that, in fibromyalgia patients, cognitive impairments are reflected in subjective complaints independently of depressive symptoms.

PATIENTS AND METHODS

Subjects

Thirty consecutive FM patients attending the Città della Salute e della Scienza Hospital, University of Turin, were enrolled. Because of the high prevalence of FM in women and to avoid sex-related effects, only women were enrolled in the study. All patients had a diagnosis of FM made by an expert rheumatologist according to the ACR criteria ([1]).

The exclusion criteria were ages <18 or >70 years, low education level (<5 years of education) or insufficient knowledge of the Italian language, history of medical conditions associated with cognitive dysfunction, and neurologic and/or severe psychiatric pathologies. Five patients were not undergoing psychopharmacologic treatments for FM, whereas 25 patients were taking antidepressants for the management of pain. As a control group, 30 healthy women balanced for age and education level were enrolled (HCs). In addition to the above exclusion criteria, HCs were also required to have no history of rheumatologic pathologies or chronic pain. The demographic characteristics of the 2 groups are shown in Table 1. The mean ± SD level of pain intensity of the sample was 6.68 ± 2.59 on a visual analog scale (VAS), with a mean ± SD disease severity of 63.48 ± 14.58 as assessed by the Fibromyalgia Impact Questionnaire (FIQ). FM patients scored a mean ± SD of 9 ± 3.7 on the depression subscale of the Hospital Anxiety and Depression Scale (HADS), whereas controls scored a mean ± SD of 5.4 ± 3.1.

Table 1. Demographic and clinical characteristics of the fibromyalgia patient (FM) and healthy control (HC) groups*
Variable FM (n = 30) HC (n = 30) T(df) P
  1. HADS = Hospital Anxiety and Depression Scale; HADS-Tot = HADS total score; HADS-D = HADS depression subscale; HADS-A = HADS anxiety subscale.
Age, mean ± SD 52.8 ± 9.6 53.8 ± 12.4 −0.42 (58) 0.680
Years of education, mean ± SD 10.9 ± 3.5 12.4 ± 3.1 −1.77 (58) 0.081
Marital status, no. (%)
Single 3 (10) 0 (0)
Cohabiting 1 (3.3) 1 (3.3)
Married 22 (73.3) 23 (76.7)
Divorced 4 (13.3) 4 (13.3)
Widowed 0 (0) 2 (6.7)
Work status, no. (%)
Employed 18 (60) 24 (80)
Unemployed 3 (10) 1 (3.3)
Retired 3 (10) 3 (10)
Housewife 6 (20) 2 (6.7)
HADS, mean ± SD
HADS-Tot 18.2 ± 5.8 11.2 ± 5.7 4.7 (58) < 0.0001
HADS-D 9.0 ± 3.7 5.4 ± 3.1 4.2 (58) < 0.0001
HADS-A 9.2 ± 3.2 5.5 ± 3.4 4.3 (58) < 0.0001

Procedure

After giving written informed consent, all of the subjects participated in a 90-minute testing session during which the clinical, neuropsychological, and self-perception of cognitive dysfunction evaluations were performed. Disease severity due to FM was measured using the FIQ ([11, 12]), while pain intensity experienced in the previous week was measured on a 0–10 VAS. The presence of depressive and anxiety symptoms was evaluated through the HADS. The neuropsychological assessment was made on EF, memory (Rey Auditory Verbal Learning Test [RAVLT]), and working memory (N-Back task). In accordance with the Miyake model ([9]) later modified by Fisk and Sharp ([10]), and with reference to the article by Aboulafia-Brakha and colleagues ([13]), we used the Tower of London test (ToL) for inhibition, the Trail Making Test (TMT) for shifting, the Digit Span Test (DS) for updating, and the Verbal Fluency Test (FAS) test for the access EF domains.

To assess self-evaluation of cognitive impairments, we used a self-report questionnaire originally developed to assess perceived changes in cognitive functioning in cancer patients: the Functional Assessment of Cancer Therapy cognition scale (FACT-Cog 2) ([14]). Although specifically constructed to evaluate the so-called “chemo fog” in cancer patients ([15, 16]), because the scale contained no references to oncologic pathology or chemotherapy, it may be used with other populations ([17, 18]). The FACT-Cog was specifically constructed to minimize the impact of distress on patient reports by means of behaviorally based items ([14]) that are less susceptible to depressive and anxiety symptoms. For example, the occurrence of subjective cognitive disturbances in the FACT-Cog is quantified by reference to behavioral frequencies (once a week) instead of standard system responses (somewhat) because the latter can be more prone to psychological bias. This is of particular importance in FM, which has been described as a stress-related disorder ([19]) with a high occurrence of psychological comorbidities ([20]).

Clinical description

FIQ

The questionnaire includes 20 items assessing the severity of the disease on a scale of 0–100, with a higher score corresponding to a higher level of impairment ([11, 12]).

HADS

The HADS consists of 14 items divided into 2 subscales: 1 for depression and 1 for anxiety. Each subscale score ranges from 0–21, with a score of 8 or more suggesting a clinically relevant level of depression/anxiety symptoms ([21, 22]).

Neuropsychological evaluation

Inhibition

The ToL evaluates the planning and inhibition EFs ([23]). The subject has to move 3 different colored disks from a prearranged sequence on 3 different pegs to match 12 predetermined goals. The task has to be done as quickly as possible, following several specific rules and in a given number of moves, a number that increases with the difficulty of the task. The execution time is registered as a dependent variable and then converted and compared with the normative data ([24]).

Shifting

The TMT provides information on visual search, scanning, processing speed, mental flexibility, and attention shifting ([25, 26]). It is composed of 2 parts: in the first part (TMT-A), the subject has to connect numbers ascending from 1 to 25, and in the second part (TMT-B), the subject has to alternate between numbers and letters (1, A, 2, B, 3, C, etc.). The test provides 3 scores: the time (in seconds) of parts A and B and the difference between them (TMT-BA).

Updating

The DS assesses short-term verbal memory span and the ability to manipulate and update verbal information while in temporary storage ([27, 28]). Subjects are required to repeat, immediately following presentation, increasingly longer strings of single digit numbers in either forward (DS-F) or reverse (DS-B) order. For both trials, the final score is the number of digits correctly repeated.

Access

The FAS test assesses phonemic verbal fluency by requiring the subjects to orally produce as many words as possible beginning with the letters F, A, or S within a timeframe of 1 minute ([29, 30]).

Memory

The RAVLT ([30, 31]) is a word list learning task assessing short- and long-term memory. Subjects freely recall a 15-word list 5 times (immediate recall [RAVLT-I]) after oral presentation. Fifteen minutes later, subjects are asked to recall the 15-word list again (delayed recall [RAVLT-D]).

Working memory

Our N-Back paradigm was gathered from the study by Legrain and colleagues ([32]). Subjects are presented with 3 blocks of 61 trials each. A cross remains at the center of the monitor for the entire duration of a block. Every 2,400 milliseconds, 2 circles are presented, 1 on the left and 1 on the right of the central cross. The circles may be both blue or both yellow and remain on the screen for 600 milliseconds. The subject has to match the color of the current visual target to the color of the preceding one, judging whether the color is the same or not and pressing a certain key on the keyboard as quickly as possible (1-Back task). The dependent variables are reaction time of correct responses (TRc) and type of response (correct, incorrect, omitted, or anticipated [<240 milliseconds]).

Self-perception of cognitive dysfunction evaluation

The FACT-Cog 2 ([14]) is a 50-item measure designed to assess cognitive experiences after receiving chemotherapies in cancer patients. On a 5-point Likert scale, subjects rate the frequency with which each statement had occurred in the previous week, with higher scores reflecting fewer cognitive problems and a better quality of life. The FACT-Cog yields 7 subscale scores (mental acuity, concentration, verbal and nonverbal memory, verbal fluency, functional interference, other people noticed deficits, and impact on quality of life) and a total score.

Statistical analyses

The data were analyzed using SPSS, version 20. Independent-samples t-tests were used to compare continuous variables between the FM and the HC groups. Pearson’s bivariate correlations were used to analyze the relationship between clinical, neuropsychological, and metacognition variables in the FM group. To reduce Type I error, only the significantly different variables between the 2 groups were inserted into the correlational analyses. P values less than 0.05 were considered statistically significant.

RESULTS

Clinical characterization

The FM patients and controls differed significantly in the level of depressive and anxiety symptoms, with the FM patients having higher scores on the HADS (Table 1). In particular, 22 FM patients (73.3%) versus 10 HCs (33.3%) had a higher score than the cutoff on the depression subscale, and 19 FM patients (63.3%) versus 9 HCs (30%) had a higher score than the cutoff on the anxiety subscale. Notably, even when HCs had above-cutoff scores, the score was close to the cutoff.

Neuropsychological variables

Detailed statistical values are shown in Table 2. Group means showed a significant difference between the 2 groups in the DS-B (updating/working memory), with the patients showing a worse performance compared with the HCs, whereas only a trend toward a significant difference was found in the DS-F subtest (short-term memory). A significant difference was found in the RAVLT-D test (episodic memory), with the patients showing a lower number of words recalled compared with the HCs. In addition, significant differences were found in the TMT-B test (attention shifting), showing that the patients required more time than the HCs to complete the task. Regarding the 1-Back task, the results showed no difference in the accuracy between the 2 groups, but the FM patients had longer reaction times than the HCs (891 and 722 milliseconds, respectively).

Table 2. Neuropsychological performance of the fibromyalgia patient (FM) and healthy control (HC) groups*
FM HC T(df) P
  1. Values are the mean ± SD unless indicated otherwise. FAS = Verbal Fluency Test; DS-F = Digit Span Test forward; DS-B = Digit Span Test backward; RAVLT-I = Rey Auditory Verbal Learning Test immediate recall; RAVLT-D = RAVLT delayed recall; TMT-A = Trail Making Test A; TMT-B = Trail Making Test B; TMT-BA = TMT B minus A score; ToL = Tower of London test; TRc = reaction time of correct responses; Ne = number of errors; NV = anticipations and omissions.
Verbal fluency/Access
FAS 37.7 ± 13.9 42.1 ± 7.6 −1.54 (44.9) 0.131
Short-term memory
DS-F 5.1 ± 1.0 5.7 ± 1.2 −1.98 (58) 0.053
Updating/working memory
DS-B 3.8 ± 1.1 4.4 ± 0.9 −2.21 (58) 0.031
Episodic memory
RAVLT-I 47.7 ± 11.6 52.0 ± 7.5 −1.72 (49.7) 0.092
RAVLT-D 9.9 ± 3.6 11.7 ± 2.4 −2.19 (50.4) 0.033
Attentional shifting
TMT-A 42.4 ± 19.4 35.8 ± 12.0 1.60 (48.4) 0.117
TMT-B 97.3 ± 39.9 75.7 ± 28.6 2.40 (52.5) 0.020
TMT-BA 53.7 ± 29.9 39.7 ± 21.5 2.08 (52.6) 0.042
Problem solving/inhibition
ToL 26.9 ± 3.7 27.5 ± 2.9 −0.57 (58) 0.567
Working memory
1-Back
TRc 891.2 ± 185.0 722.4 ± 131.9 4.07 (58) < 0.0001
Ne 17.9 ± 20.6 14.1 ± 7.9 0.95 (37.51) 0.349
NV 3.8 ± 5.4 4.6 ± 4.7 −0.61 (58) 0.543

In order to bring out the individual differences that could be flattened by group analyses, we analyzed the individual scores by comparing for each test the number of subjects with a clinically deficient performance according to the age- and education-corrected scores (equivalent score ≤1). The results showed that a significantly higher number of FM patients compared with HCs had a deficient performance in both subscales of the RAVLT (episodic memory) and the ToL (problem solving/EF) (Figure 1).

Figure 1.

Figure 1.

  • Open in figure viewer
  • Download Powerpoint slide

Percentage of fibromyalgia patients (FP) and healthy controls (HC) with a clinically impaired performance (equivalent score = 0–1). FAS = Verbal Fluency Test; DS-F = Digit Span Test forward; DS-B = Digit Span Test backward; RAVLT-I = Rey Auditory Verbal Learning Test immediate recall; RAVLT-D = RAVLT delayed recall; TMT-A = Trail Making Test A; TMT-B = Trail Making Test B; TMT-BA = TMT B minus A score; ToL = Tower of London test; N.S. = not significant.

Self-perception of cognitive dysfunction

The FM patients reported a significantly worse judgment compared with the HCs in the total score and in all subscales of the FACT-Cog (Table 3). Not only did patients declare deficits in all of the cognitive abilities investigated, but they also had lower scores in the other people noticed deficits subscale, reporting that other people noticed their inadequate cognitive performance. In addition, FM patients reported lower scores on the quality of life subscale, which evaluates the impact of cognitive performance on their quality of life.

Table 3. Self-perception of cognitive dysfunction (Functional Assessment of Cancer Therapy cognition scale)*
FM HC T(df) P
  1. Values are the mean ± SD unless indicated otherwise. FM = fibromyalgia patient group; HC = healthy control group; MA = mental acuity; Con = concentration; Mem = verbal and nonverbal memory; VF = verbal fluency; FI = functional interference; OnD = other people noticed deficits; QoL = impact on quality of life; Tot = total score.
MA 2.6 (0.8) 3.3 (0.7) −3.25 (58) 0.002
Con 2.3 (0.8) 3.1 (0.7) −4.18 (58) < 0.0001
Mem 2.4 (0.7) 3.1 (0.6) −3.79 (58) < 0.0001
VF 2.2 (0.8) 2.9 (0.6) −3.97 (52.1) < 0.0001
FI 2.4 (0.7) 3.2 (0.5) −5.89 (51.6) < 0.0001
OnD 2.7 (1.0) 3.4 (0.4) −3.60 (39.3) 0.001
QoL 2.1 (1.1) 3.5 (0.6) −5.93 (45.6) < 0.0001
Tot 2.2 (0.6) 2.9 (0.4) −5.27 (52.9) < 0.0001

Correlations between clinical and neuropsychological variables in FM

Given the exploratory nature of these analyses, we did not apply corrections for multiple comparisons ([33]). The severity of FM, as assessed by the FIQ, showed a statistically significant positive correlation with the TMT-B (r = 0.434, P = 0.021) and a trend toward a significant correlation with TRc (r = 0.375, P= 0.05) (Table 4); the higher the severity of FM, the lower the ability to shift attention (TMT-B) and the higher the reaction times in the 1-Back task. No other significant correlations were found.

Table 4. Correlational analyses in fibromyalgia patients*
FIQ VAS HADS-D HADS-A
  1. Values are the Pearson’s correlation coefficients. FIQ = Fibromyalgia Impact Questionnaire; VAS = visual analog scale; HADS-D = Hospital Anxiety and Depression Scale depression subscale; HADS-A = HADS anxiety subscale; DS-B = Digit Span Test backward; RAVLT-D = Rey Auditory Verbal Learning Test delayed recall; TMT-B = Trail Making Test B; TMT-BA = TMT B minus A score; TRc = reaction time of correct responses; FACT-Cog = Functional Assessment of Cancer Therapy cognition scale; MA = mental acuity; Con = concentration; Mem = verbal and nonverbal memory; VF = verbal fluency; FI = functional interference; OnD = other people noticed deficits; QoL = impact on quality of life; Tot = total score.
  2. aP < 0.05.
  3. bP < 0.01.
DS-B −0.173 −0.150 −0.236 −0.065
RAVLT-D −0.169 −0.223 −0.189 −0.028
TMT-B 0.434a 0.336 0.046 0.082
TMT-BA 0.369 0.207 −0.038 0.055
TRc 0.375 0.244 0.139 0.181
FACT-Cog
MA −0.262 −0.162 −0.182 0.017
Con −0.493b −0.314 −0.272 −0.233
Mem 0.079 −0.144 −0.256 0.094
VF −0.355 −0.245 −0.35 −0.181
FI −0.430a −0.214 −0.296 −0.119
OnD −0.467a −0.390a −0.401a −0.24
QoL −0.568b −0.254 −0.245 −0.171
Tot −0.523b −0.325 −0.315 −0.162

Regarding the relationship between the FACT-Cog subscales and clinical parameters (Table 4), the analyses showed statistically significant negative correlations between the FIQ and the concentration (P = 0.008), functional interference (P = 0.022), and quality of life (P = 0.002) subscales and the total score (P = 0.004). Patients with more severe FM reported greater self-perception of cognitive impairments in concentration and attention and reported experiencing a greater negative impact of cognitive deficits on their quality of life. Significant correlations also emerged between the FIQ and the other people noticed deficits subscale of the FACT-Cog (P = 0.012) and between the other people noticed deficits subscale and pain intensity (P = 0.040). The other people noticed deficits subscale measures how severe, in the patients’ opinion, cognitive deficits are perceived by other people. Therefore, greater severity of pain and clinical symptoms were positively correlated with the patients’ perception that other people tend to notice their cognitive impairments.

The correlational analyses between cognitive impairment perception and the psychological variables showed that the depression subscale of the HADS and the other people noticed deficits subscale of the FACT-Cog were significantly and negatively correlated (r = −0.401, P = 0.028), suggesting that patients with more severe depressive symptoms claimed more that their cognitive deficits are noted by other people.

Finally, the correlational analyses between the neuropsychological tests and the self-perception of cognitive dysfunction showed statistically significant correlations between the verbal fluency subscale of the FACT-Cog and the DS-B (r = 0.396, P = 0.03), the TMT-B (r = −0.361, P = 0.049), and the TMT-BA (r = −0.396, P = 0.03) scores, suggesting that a worse performance in these tests was correlated with a higher number of patient reports of cognitive dysfunction in the verbal fluency domain (EF) (Table 5).

Table 5. Correlational analyses between the Functional Assessment of Cancer Therapy cognition scale and the neuropsychological tests in fibromyalgia patients*
DS-B RAVLT-D TMT-B TMT-BA TRc
  1. Values are the Pearson’s correlation coefficients. DS-B = Digit Span Test backward; RAVLT-D = Rey Auditory Verbal Learning Test delayed recall; TMT-B = Trail Making Test B; TMT-BA = TMT B minus A score; TRc = reaction time of correct responses; MA = mental acuity; Con = concentration; Mem = verbal and nonverbal memory; VF = verbal fluency; FI = functional interference; OnD = other people noticed deficits; QoL = impact on quality of life; Tot = total score.
  2. aP < 0.05.
MA 0.073 0.058 −0.153 −0.11 −0.1
Con 0.163 0.071 −0.215 −0.258 −0.124
Mem 0.21 0.08 −0.122 −0.144 −0.006
VF 0.396a −0.033 −0.361a −0.396a −0.116
FI 0.143 0.119 −0.201 −0.196 −0.214
OnD 0.313 0.023 −0.293 −0.266 0.081
QoL 0.119 0.177 −0.307 −0.357 −0.377a
Tot 0.137 0.13 −0.239 −0.28 −0.304

A significant correlation was also observed between the quality of life subscale of the FACT-Cog and the reaction times in the correct responses of the 1-Back task (r = −0.377, P = 0.04); patients with faster reaction times judged their quality of life to be better (Table 5).

DISCUSSION

FM patients often experience a cluster of cognitive disorders that strongly interfere with their work and daily life ([3]); however, cognitive dysfunction remains underrecognized and undertreated. This is because neuropsychological tests require neuropsychological expertise and are time consuming and the relationship between patient cognitive reports and objective neuropsychological tests in chronic pain patients is still controversial ([6, 7]). Recently, Landrø and colleagues ([6]) reopened this debate, finding that cognitive reports in chronic nonmalignant pain subjects were validated on objective neuropsychological assessment. Because the authors, as highlighted by McGuire ([34]), used norm-referenced neuropsychological tests and did not have a pain-free comparison group, we investigated the presence of cognitive impairment in FM patients and analyzed the relationship between the cognitive deficit and the patients’ self-evaluation of cognitive impairment by means of a comparison with a group of HCs.

The finding that FM patients have poorer performances in working memory, attention, and EF has been reported by Bertolucci and de Oliveira ([5]). However, EF represents a multifaceted construct composed of separable factors that tap into different cognitive mechanisms that possibly involve the activity of different brain structures. Considering EF as a whole would not allow the identification of subtle differences in cognitive experiences. In the present study, we selected neuropsychological tests in order to cover the different EF aspects (inhibition, shifting, updating, and access) ([9, 10]).

Our results indicated that FM patients are more impaired in the long-term memory (delayed subscale of the RAVLT), attention shifting (TMT-B), and updating (DS-B) components of EF. In addition, although the accuracy of FM patients in visual working memory was comparable to that of HCs, the former showed a significant slowdown in reaction times (1-Back task).

Reports about working memory functions are common across a wide variety of chronic pain states and, as reviewed by Berryman et al, moderate impairment in working memory can be consistently observed across studies and paradigms ([35]). Consistent with previous studies showing a slowdown in the response processing of FM patients ([36]), our patients were slower in selecting whether the target was similar to or different than a previous stimulus. Seo and colleagues also observed that when a greater amount of manipulation (e.g., 2-Back condition) was required, FM patients also showed reduced accuracy ([36]). It is therefore possible that our task was sensitive enough to detect differences in reaction times but not difficult enough to require a great amount of manipulation ([37]) to bring out differences in accuracy.

Regarding EF, the patients showed impairment in the shifting subcomponent, which has already been highlighted by a previous study ([38]). Poor performances on complex tests that involve interference or attention switching have also been found in other chronic pain states ([39]). Notably, although the performance on the TMT-B is usually considered a measure of shifting abilities ([13]), it also requires updating abilities. Therefore, it is possible that low performances on the TMT-B resulted from both shifting and updating deficits. In fact, the updating subcomponent, as measured by the DS-B, also appeared to be impaired. Although the patients’ mean score on the DS-B was slightly higher than normative data, FM patients showed a significantly lower performance compared with HCs. This result is in line with previous studies by Cherry and colleagues ([40]) and De Melo and Da-Silva ([41]) that showed that FM patients performed worse compared to patients with other chronic rheumatic diseases. In line with studies in the literature, we did not observe deficits in the subcomponents of inhibition or access of EF. Veldhuijzen and colleagues ([42]), in investigating the ability of FM patients to inhibit preponderant information, evidenced a slower performance than that of the controls but equal accuracy. This finding parallels our results on the N-Back task and may then point to an underlying problem of mental processing speed and/or psychomotor speed ([42]). Evidence that FM patients are not impaired in the access domain is accumulating; FM patients did not show impairments in verbal fluency, either phonemic ([43]) or semantic ([40]). In an apparent contradiction, in a review of cognitive dysfunction in FM ([44]), Glass reported that several studies observed fluency disturbances in FM patients. Methodological problems or differences could account for contrasting results reported in published studies on cognitive deficits in FM. In fact, in some of the studies reviewed by Glass ([44]), the groups of FM patients and HCs were not balanced in sample size ([45]) or education level ([46]). Concerning the shifting function, other studies failed to show impairment in FM when comparing performances with the normative data ([47, 48]).

Landrø and colleagues found an accordance between the objective performance and the subjective experiences of cognitive impairments, suggesting that, in chronic pain patients, subjective experiences might reflect genuine cognitive deficits to a larger extent than previously believed ([6]). Consistently, Glass and colleagues reported a match between reported and objective deficits in memory capacity in FM patients ([49]).

Our results indicated a good correlation between the verbal fluency subscale of the FACT-Cog and EF and working memory tests, thus suggesting that the FACT-Cog can be used as an effective screening tool for objective cognitive deficits. However, it should also be noted that not all subscales showed strong correlations. Therefore, although patient reports can be used as an effective tool to identify objective deficits, a complete picture of the patients’ cognitive status is better achieved with a more in-depth neuropsychological battery. Another caveat is that the lack of strong correlations can be explained by the operational definition of the tested constructs. What is considered as memory in everyday life is tested at a more fine-grained level in the neuropsychological domain. As such, the objective and subjective reports do not capture the same level of definition and, consequently, may not correlate completely. One possible confounding factor in the interpretation of self-evaluation questionnaires is the concomitant presence of mood disturbances in the majority of chronic pain patients ([50]). For example, Williams and colleagues investigated the relationship between the self-perception of cognitive functioning and other symptoms commonly present in FM (pain, fatigue, sleep, and depressive and anxiety symptoms) ([3]). The authors found that the domains of mood and fatigue were strongly associated with the perceived dyscognitions in FM, whereas pain was uniquely associated with perceived language deficits and, unexpectedly, was not related to attention or concentration. In agreement with this, the results of our study showed that the perceived cognitive dysfunctions were not associated with pain intensity. We found no association between the self-evaluation of cognitive functioning and depressive or anxiety symptoms. This is not surprising considering the FACT-Cog questionnaire is specifically constructed to minimize the impact of distress on patient reports ([14]). It is possible that the presence of depressive symptoms could be partially related to an increase in the FM patients’ reports, but using the FACT-Cog could help to avoid this bias.

Patient reports of cognitive deficits were related to the severity of the disease as assessed by the FIQ. This result was expected, given that the FIQ evaluates the impact that FM symptoms have on the patients’ daily lives and that the presence of cognitive deficits is often reported as causing difficulties in everyday functioning. Demonstrating a relationship with the severity of the disease were not only patient reports of cognitive deficits, but also objective deficits and in particular a slowdown in reaction times and shifting deficits. No other relationship emerged between test performance and pain intensity and depressive and anxiety symptoms. With regard to both types of symptoms, studies in the literature have shown undefined results; a recent review reported that poor scores on EF measures did not seem related to mood disorders, but a correlation between verbal fluency and neuropsychiatric symptoms, such as hallucinations, irritability, and anxiety, was described ([5]).

Our study presents some methodological limitations. First, the results of the correlational analyses should be interpreted with caution. Indeed, these results were obtained from a relatively small sample size and were not corrected for multiple comparisons, given the exploratory nature of the analysis. Second, we did not select patients on the basis of their ongoing pharmacologic therapy. Indeed, some medications can affect cognitive function, making it difficult to discern which cognitive deficit might be attributable to FM and which to medications. Nevertheless, most of our patients taking antidepressants for pain were treated with duloxetine, a dual antidepressant that does not significantly impair cognition ([48]). In addition, this limitation did not invalidate the main result of the study, which concerns the degree of accordance between subjective and objective reports.

In conclusion, our data indicate that long-term and working memory, shifting of attention, and updating EF of FM patients are impaired compared with HCs. These impairments are reflected in patient reports independently of depressive symptoms. The use of a self-report questionnaire in clinical practice would provide a first and easy screen for the presence of cognitive impairment in FM patients and, in most cases, avoid the need for a time-consuming full neuropsychological test battery.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Ms Tesio had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

refrence.http://onlinelibrary.wiley.com/doi/10.1002/acr.22403/full

Leave a Reply

Your email address will not be published. Required fields are marked *

Close