Neuromuscular fatigue and exercise capacity in fibromyalgia syndrome




To assess quadriceps strength and fatigability by using femoral nerve magnetic stimulation (FNMS) and their relationship to exercise capacity in patients with fibromyalgia syndrome (FMS) and healthy controls.


Twenty-two women (11 with FMS, 11 controls) performed a maximal incremental cycling test and a quadriceps fatigue test on 2 separate visits. For quadriceps assessment, we used FNMS during and after maximum voluntary contraction (MVC) to evaluate central and peripheral factors of neuromuscular fatigue. Subjects performed sets of 10 intermittent (5 seconds on/5 seconds off) isometric contractions starting at 10% MVC, in 10% MVC increments from one set to another until exhaustion. Neuromuscular fatigue was assessed with FNMS after each set.


FMS patients had reduced initial MVC compared to controls (mean ± SD 102 ± 18 versus 120 ± 24 Nm; P < 0.05) without significant impairment of voluntary activation (mean ± SD 93.5% ± 3.0% versus 93.1% ± 3.4%; P = 0.74). During the fatigue task, FMS patients exhibited a greater fall in evoked muscular responses (mean ± SD −26% ± 6% versus −16% ± 8% at set 50% MVC; P < 0.05), but not in MVC (mean ± SD −24% ± 7% versus −19% ± 4% at set 50% MVC; P = 0.12). During the cycling test, FMS patients had lowered maximal exercise capacity and an enhanced rate of perceived exertion (RPE) compared to controls. The percent reduction in evoked muscular responses during the quadriceps fatigue test correlated with maximum oxygen consumption (r = 0.56, P < 0.05) and RPE at submaximal intensity (r = 0.84, P < 0.05) during cycling.


Greater impairment in muscle contractility is associated with enhanced perception of exertion and reduced maximal exercise capacity in FMS patients. Neuromuscular impairments should be considered as an important factor underlying functional limitations in FMS patients.


Fibromyalgia syndrome (FMS) is a multifactorial chronic disorder combining multifocal pain, fatigue, sleep disturbances, and various other symptoms (1). Fear of pain and kinesiophobia (2) contribute to decreased physical activity (3, 4) and reduced exercise capacity (5–7). Physical fitness in FMS has mostly been characterized by reduced maximal performance and increased perceived exertion during exercise such as stationary cycling (6–8). In addition to alteration in sensory perception, reduced muscle strength and increased fatigability may contribute to exercise limitation and promote hyperalgesia (9, 10).

Impairment of locomotor muscles has been identified as a relevant contributing factor to reduced exercise capacity and disability in numerous chronic diseases such as rheumatoid conditions (11), cardiovascular diseases (12), and respiratory disorders (13). In patients with FMS, quadriceps weakness has been repeatedly reported (7, 14, 15), while impaired endurance and higher fatigability compared to controls have also been suggested (14, 16, 17). However, the relative contribution of motivation, pain, and deconditioning in neuromuscular impairment remains unclear. Moreover, the role of muscle performance (e.g., strength and fatigability) impairments regarding exercise capacity in FMS patients remains to be determined.

Neuromuscular function can be assessed with artificial muscle stimulation, which allows for distinguishing peripheral (e.g., action potential propagation, excitation–contraction failure, contractility) and central (activation failure due to pain and spinal or supraspinal mechanisms) factors responsible for attenuated strength and fatigue (18). To our knowledge, only 2 studies (19, 20) have evaluated quadriceps function in FMS by using transcutaneous electrical muscle stimulation. Both reported reduced maximal voluntary strength in FMS patients, but a reduction in maximal voluntary activation (VA) was reported by Jacobsen et al (19) only. Norregaard et al (20) explored peripheral fatigability during intermittent contractions and reported no difference between FMS patients and controls. Both studies are questionable regarding the methodology used, involving submaximal muscle rather than supramaximal nerve stimulation and leading to abnormally low VA in controls (e.g., ∼80% [20]).

The purpose of the present study was to investigate quadriceps strength and fatigability and their relationship to exercise capacity in FMS patients. To assess neuromuscular function, we used femoral nerve magnetic stimulation (FNMS), recently validated by our team (21) and more suitable for clinical practice than standard electrical neurostimulation (22, 23). To assess endurance and fatigue, we used an isolated muscle exercise test involving intermittent isometric contractions at increasing force levels with iterative measurements of fatigue in order to limit the confounding effects of pain and motivation. We hypothesized that patients with FMS would have reduced maximal strength and increased muscle fatigability, due to both peripheral and central mechanisms. We also hypothesized that these neuromuscular impairments would be associated, at least in part, with reduced exercise capacity and increased sensations of fatigue during stationary cycling.

Significance & Innovations

  • Reduced quadriceps strength in patients with fibromyalgia syndrome is not associated with voluntary activation deficit, but most probably with smaller muscle mass.

  • During repeated isometric quadriceps contractions, patients with fibromyalgia syndrome exhibit larger contractility impairment as assessed with magnetic neurostimulation.

  • Larger impairment in muscle contractility may be involved in enhanced perceived exertion and reduced maximal exercise capacity in patients with fibromyalgia syndrome.



Twenty-two women (11 FMS patients and 11 healthy controls) volunteered to participate in this study and gave written informed consent. Their main characteristics are shown in Table 1. At initial screening, all patients underwent physical examination by a rheumatologist (MG) and met the American College of Rheumatology criteria for FMS (1, 24). Patients were excluded if they presented contraindications to magnetic stimulation, a pathologic knee condition, femoral nerve compression history, disorders of the thyroid gland, and a current psychiatric condition. We included normal weight FMS patients (body mass index [BMI] <25 kg/m2) to avoid the confounding effect of body mass excess and to guarantee supramaximal FNMS (25). Seven patients took analgesics for pain (5 took first- and/or second-class analgesics and 2 took third-class analgesics). Three took specific serotoninergic reuptake inhibitors, 3 took benzodiazepines, and 4 took first-class antidepressants for sleep disorders. Healthy sedentary (<3 hours of physical activity per week) controls were paired with FMS patients by age and BMI. The study was conducted according to the Declaration of Helsinki with approval from the local Committee on Human Research (Comité de Protection des Personnes Sud-EST V).

Table 1. Subject characteristics and questionnaires*
 FMS patientsControlsP
  • *

    Values are the mean ± SD. FMS = fibromyalgia syndrome; BMI = body mass index; FIQ = Fibromyalgia Impact Questionnaire; VAS = visual analog scale; FSS = Fatigue Severity Scale; PCS = Pain Catastrophizing Scale; SF-36 = Short Form 36 health survey.

  • Habitual physical activity assessed by the Ricci-Gagnon questionnaire.

Age, years44 ± 948 ± 80.28
BMI, kg/m221.2 ± 2.521.6 ± 1.30.66
Body fat, %32 ± 532 ± 40.68
Pain duration, years6 ± 5
FIQ55 ± 94 ± 3< 0.001
VAS pain in last 24 hours, mm49 ± 202 ± 1< 0.001
VAS pain in last week, mm51 ± 184 ± 2< 0.001
FSS49 ± 1622 ± 12< 0.001
PCS25 ± 9
Physical activity18 ± 722 ± 50.15
SF-36 total45 ± 1088 ± 6< 0.001
SF-36 physical38 ± 1090 ± 6< 0.001
SF-36 psychological52 ± 1186 ± 8< 0.001

Experimental design.

During the first visit, after answering questionnaires, subjects performed a maximal incremental exercise test on a cycle ergometer. During the second session, subjects performed quadriceps neuromuscular assessment.

Anthropometric measurements.

Body fat mass percentage was assessed from 4 skinfold measurements (26). Quadriceps volume estimation was based on a truncated cone calculation using 3 thigh perimeters and thigh skinfold (27).


Habitual physical activity was evaluated with the Ricci-Gagnon French questionnaire (28). We used the Fibromyalgia Impact Questionnaire (FIQ) to assess the status of FMS patients (29). Pain was assessed with a 100-mm visual analog scale (VAS) during the last week and the last 24 hours. The influence of FMS on quality of life was evaluated with the Medical Outcomes Study Short Form 36 (30). Catastrophizing was assessed with the Pain Catastrophizing Scale (31).

Maximal cycling test.

Subjects performed a maximal incremental exercise test on a computer-controlled electrically braked cycle ergometer (Ergometrics 800, Ergoline) with expired gas analysis and electrocardiogram (Medisoft) (32). Stationary cycling was used because it is the most frequently used exercise modality in FMS (5–8) and it strongly involves the quadriceps muscle, which is a major locomotor muscle easily assessed with FNMS. Based on usual recommendations (i.e., initial power and increments chosen to obtain an exercise duration ranging from 8–12 minutes [33]), exercise protocols were defined as follows: 15W initial power in FMS patients and 30W in controls for 2 minutes followed by 2-minute increments (15W in FMS patients and 25W in controls). The rate of perceived exertion (RPE) was assessed each minute all along the test with a standard 100-mm VAS. A fingertip blood sample was obtained 3 minutes after exhaustion and analyzed for lactate concentration (NOVA+, Nova Biomedical).

Quadriceps neuromuscular assessment.

Experimental setup.

All measurements were conducted on the right lower extremity under isometric conditions. The subject lay supine on a customized quadriceps chair. The knee was flexed at 90° and the hip angle was 130° for proper access to the femoral triangle during FNMS. Voluntary strength and evoked responses to FNMS were measured with an inextensible ankle strap connected to a strain gauge (SBB 200 kg, Tempo Technologies). Compensatory movement of the upper body was limited by 2 belts across the thorax and abdomen. Subjects were asked to keep their hands on their abdomen. A visual feedback of both the force produced and the target force levels (see below) was provided to the subjects.


FNMS was performed with a 45-mm figure-eight coil powered by 2 Magstim 200 stimulators (peak magnetic field 2.5T, stimulation duration 0.1 msec) linked to a Bistim Module (Magstim), as previously described (21). Single (twitch) and paired stimulations (10-Hz and 100-Hz doublets) were performed at the maximum stimulator output. The coil was positioned high in the femoral triangle in regard to the femoral nerve. The optimal stimulation site allowing maximal unpotentiated quadriceps peak strength (unpotentiated twitch [Twu]) and maximum vastus lateralis M-wave amplitude was determined. After 20 minutes of rest, stimulation supramaximality was carefully checked with decreasing stimulator power output (100%, 95%, 90%, 85%, and 80%). Twu and M-wave amplitude were not significantly reduced until 80% of the maximal power output was reached in both groups, confirming FNMS supramaximality, as previously reported in our laboratory (21).

The quadriceps surface electromyogram (EMG) signal was recorded from the vastus lateralis (as a surrogate for the entire quadriceps muscle [34]), as described in detail previously (21). EMG signals were amplified (BioAmp, ADInstruments) with a bandwidth from 5–500 Hz. EMG data together with force signals were digitized online at a sampling frequency of 2,000 Hz and recorded on a dedicated device (PowerLab, ADInstruments).

Quadriceps fatigue test description.

Before starting the initial neuromuscular assessment, subjects performed ten 5-second submaximal isometric quadriceps contractions in order to warm up the quadriceps muscle and to familiarize themselves with the visual feedback and the soundtrack instructions (see below). Then, subjects performed 3 maximum voluntary contractions (MVCs) with 1 minute of rest between each MVC (34). Following these MVCs allowing full muscle potentiation (35), the initial neuromuscular assessment was performed. It included a 5-second MVC superimposed with a 100-Hz doublet (Db100s) and followed after 2 seconds (i.e., on a relaxed muscle) by 2 potentiated doublets at 100 Hz (Db100) and 10 Hz (Db10) performed 4 seconds apart. After 15 seconds of rest, the subject performed a second MVC followed after 2 seconds by one potentiated twitch (Twp). During all MVCs, subjects were vigorously encouraged by the experimenter (DB). Evoked high- and low-frequency paired stimulations provide an extensive understanding of peripheral fatigue (e.g., high- and low-frequency peripheral fatigue) (21), and high-frequency superimposed stimulation provides optimal resolution for central activation assessment (34).

After the initial neuromuscular assessment, sets of 10 intermittent (5 seconds on/5 seconds off) contractions at submaximal target forces were performed, starting at 10% MVC for the first set, 20% MVC for the second set, and so on until task failure. Subjects had visual force feedback providing the target level and listened to a soundtrack indicating the contraction–relaxation rhythm. Task failure was defined as 2 consecutive contractions 10N below the target force for more than 2.5 seconds. Five seconds after the end of each 10-contraction set and at exhaustion, neuromuscular assessments similar to the initial neuromuscular assessment were performed.

Data analysis.

From the maximal incremental cycling test, we assessed maximal work load, maximum oxygen consumption (VO2 max), minute ventilation (VE), heart rate (HR), respiratory exchange ratio (RER), and RPE. We also assessed submaximal responses for the same variables at 50% and 75% of both measured and predicted VO2 max (36) in order to account for the potential submaximality of the maximal cycling test in FMS patients, as previously reported (6). One FMS patient was excluded from this analysis because she did not reach 75% of predicted VO2 max.

The following parameters were calculated from the mechanical responses to FNMS during quadriceps neuromuscular assessment: peak force for Twu, Twp, Db100, and Db100s, and the ratio of Db10 over Db100 (Db10:100). M-wave amplitude was calculated from a potentiated single stimulation. For MVCs, we considered the maximum value of the 2 trials performed at the initial neuromuscular assessment, after each submaximal contraction set, and at exhaustion. Maximum VA was calculated from Db100s and Db100 as follows:

equation image

A correction was applied to the original equation when the superimposed stimulation was administrated slightly before or after the real peak MVC (34).

We analyzed data from the 10-contraction sets at 10–50% MVC (set 10%, set 20%, set 30%, set 40%, and set 50%) because they were completed by all of the subjects. Changes in maximal voluntary and evoked quadriceps responses from the initial neuromuscular assessment to set 50% were used as indexes of fatigability.

Statistical analysis.

All variables are reported as the mean ± SD within the text and tables. Normal distribution and homogeneity of variances analysis were confirmed using the Kolmogorov-Smirnov and skewness tests, respectively. Unpaired t-tests were conducted to compare FMS patients and controls for the following variables: subject characteristics, questionnaire scores, and neuromuscular function at the initial neuromuscular assessment. To compare changes in variables during the quadriceps fatigue test, we used two-way repeated-measures analyses of variance (time × group) and t-tests with Bonferroni correction for post hoc analysis. Linear regression analysis was used to determine the relationship between variables (i.e., between quadriceps neuromuscular function, cycling exercise responses, and questionnaires). The alpha level was set at 0.05 for all tests. Statistical analysis was performed with a statistical software package (NCSS).


Questionnaire scores.

Questionnaire scores are shown in Table 1. FMS patients had moderate to severe FMS, as indicated by the FIQ score and VAS for pain. They showed higher fatigue perception than controls and a moderate pain catastrophizing score. Reported physical activity was not significantly different between FMS patients and controls. FMS patients had an altered quality of life in both the physical and psychological domains.

Neuromuscular assessment of the quadriceps.

FMS patients had lower MVC but similar VA compared to controls (Table 2). Twp and Db100 were also significantly lower in FMS patients. Estimated quadriceps volume was significantly reduced in FMS patients and, consequently, both voluntary and evoked strength normalized to volume were not different between the groups.

Table 2. Quadriceps strength and VA in FMS patients and controls*
 FMS patientsControlsP
  • *

    Values are the mean ± SD. VA = voluntary activation; FMS = fibromyalgia syndrome; MVC = maximum voluntary contraction; Db100 = peak potentiated 100-Hz doublet; Twp = peak potentiated single twitch; Db10:100 = ratio of the peak potentiated doublets at 10:100 Hz.

MVC, Nm102 ± 18120 ± 24< 0.05
Db100, Nm51 ± 859 ± 11< 0.05
Twp, Nm35 ± 640 ± 7< 0.05
Quadriceps volume, cm3539 ± 93650 ± 109< 0.05
Db10:1000.95 ± 0.070.90 ± 0.070.11
MVC/volume, Nm/cm30.195 ± 0.0430.185 ± 0.0430.66
Db100/volume, Nm/cm30.098 ± 0.0280.094 ± 0.0280.78
Twp/volume, Nm/cm30.068 ± 0.0200.063 ± 0.0190.62
VA level, %93.5 ± 3.093.1 ± 3.40.74

During the quadriceps fatigue test, the total number of submaximal contractions was not significantly different between the groups (mean ± SD 54 ± 8 versus 58 ± 8 for FMS patients and controls; P > 0.05). MVC decline did not significantly differ between the groups (at set 50%, mean ± SD −24% ± 7% in FMS patients versus −19% ± 4% in controls and at exhaustion, −21% ± 7% in FMS patients versus −21% ± 5% in controls; F = 1.52, P = 0.12). Changes in Twp, Db100, and VA during the quadriceps fatigue test are shown in Figure 1. Reductions in Twp and Db100 were significantly greater in FMS patients. The reduction in Db10:100 did not differ between the groups (data not shown; P > 0.05). The drop in VA showed a tendency to be larger in FMS patients (interaction: F = 2.06, P = 0.06). No significant change was found for M-wave characteristics over time in both groups (P > 0.05; data not shown).

Figure 1.

Evoked quadriceps strength and voluntary activation (VA) during the quadriceps fatigue test in fibromyalgia syndrome (FMS) patients and controls. A, Potentiated peak twitch (Twp). B, Potentiated peak 100-Hz doublets (Db100). C, VA level (VADb). Pre = initial measurement; 10–50 = measurements after sets of 10 contractions corresponding to 10–50% of maximum voluntary contraction (MVC); Exh = measurement immediately after exhaustion; # = significantly different from Pre (P < 0.05); † = significant difference between FMS and controls (P < 0.05).

Maximal incremental cycling test.

The total mean ± SD duration for the cycling test was 12 ± 1 minutes for FMS and 11 ± 2 minutes for controls (P > 0.05). FMS patients had a lower maximum workload than controls (mean ± SD 90 ± 15W versus 153 ± 40W; P < 0.05). Similarly, VO2 max was lower in FMS patients (Table 3). The percentage of maximal theoretical HR reached at exhaustion was lower in FMS patients compared to controls (mean ± SD 86% ± 11% versus 100% ± 7%; P < 0.05). RER (Table 3) and blood lactate concentration (mean ± SD 5.3 ± 2.0 versus 8.6 ± 1.7 mmoles/liter; P < 0.05) at exhaustion were also lower in FMS patients compared to controls.

Table 3. Whole-body exercise responses in FMS patients and controls*
 50% VO2 max75% VO2 max100% VO2 max, measured
  • *

    Values are the mean ± SD. FMS = fibromyalgia syndrome; VO2 max = maximum oxygen consumption; RER = respiratory exchange ratio; HR = heart rate; VE = minute ventilation; RPE = rate of perceived exertion.

  • Significant difference between FMS and controls (P < 0.01).

  • Significant difference between FMS and controls (P < 0.05).

VO2, ml/minute/kg     
 FMS11.4 ± 1.313.1 ± 1.317.1 ± 2.119.6 ± 2.323.7 ± 2.7
 Controls17.5 ± 3.613.1 ± 2.126.1 ± 5.020.3 ± 3.736.1 ± 6.3
 FMS0.89 ± 0.100.91 ± 0.101.01 ± 0.111.06 ± 0.131.12 ± 0.13
 Controls0.90 ± 0.060.88 ± 0.151.04 ± 0.080.96 ± 0.101.22 ± 0.05
HR, beats per minute     
 FMS100 ± 16107 ± 19126 ± 23139 ± 25152 ± 20
 Controls116 ± 14107 ± 15148 ± 15130 ± 23172 ± 10
VE, liters/minute     
 FMS19 ± 322 ± 531 ± 637 ± 652 ± 15
 Controls27 ± 621 ± 545 ± 1134 ± 1180 ± 16
RPE, points     
 FMS2.2 ± 2.52.7 ± 2.15.5 ± 1.47.4 ± 1.88.7 ± 1.1
 Controls1.6 ± 0.80.7 ± 1.14.6 ± 1.62.9 ± 2.68.2 ± 1.8

HR, RPE, RER, and VE as a function of VO2 are shown in Figure 2 and Table 3. The slopes of HR VO2 (mean ± SD 4.4 ± 1.1 versus 3.1 ± 0.9 beats per minute per ml/minute/kg; P < 0.05) and RPE VO2 (mean ± SD 0.57 ± 0.22 versus 0.35 ± 0.11 points per ml/minute/kg; P < 0.05) linear regressions were significantly larger in FMS patients compared to controls.

Figure 2.

Cycling exercise response in fibromyalgia syndrome (FMS) patients (shaded symbols) and controls (open symbols). A, Heart rate (HR) as a function of oxygen consumption (VO2). B, Minute ventilation (VE) as a function of VO2. C, Rate of perceived exertion (RPE) as a function of VO2. Squares = value at 50% of measured VO2 max; triangles = value at 50% of predicted VO2 max; circles = value at 75% of measured VO2 max; diamonds = value at 75% of predicted VO2 max; rectangles = value at 100% of measured VO2 max. The linear HR VO2 relationship was extrapolated in FMS patients to maximal theoretical HR, providing a theoretical quantification of exercise submaximality and metabolic impairments as factors responsible for reduced VO2 max in FMS patients compared to controls (see main text for further details). bpm = beats per minute.

VO2 max correlated with the percentage reduction in Twp (Figure 3A) and Db100 (r = 0.34, P < 0.05) at set 50% during the quadriceps fatigue test. The RPE at 75% of predicted VO2 max correlated with the percentage reduction in Twp (Figure 3B) and Db100 (r = 0.66, P < 0.05) at set 50%. VO2 max correlated with self-reported physical activity (r = 0.51, P < 0.05). No other correlation was found between questionnaire scores and exercise capacity or quadriceps strength and fatigability (all P > 0.05).

Figure 3.

Linear regression between the reduction in potentiated twitch (Twp) at set 50% during the quadriceps fatigue test and maximum oxygen consumption (VO2 max; A) and the rate of perceived exertion (RPE) at 75% of predicted VO2 max (B) in fibromyalgia syndrome (FMS) patients and controls. One FMS patient was excluded from these analyses because she did not reach 75% of predicted VO2 max. Pre = initial measurement.


The present study aimed at evaluating quadriceps function (i.e., strength and fatigability) and its relationship to exercise capacity in a group of FMS patients and healthy controls. Our results showed that FMS patients have reduced volitional and evoked quadriceps strength without significant impairment of VA and a greater percentage of reduction in evoked strength and similar falls in maximal voluntary strength and activation during an isolated quadriceps exercise. Also, FMS showed a reduced maximal exercise capacity and an increased RPE during cycling, which correlate with the amount of evoked quadriceps strength reduction during the quadriceps fatigue test. Therefore, these results suggest that peripheral neuromuscular fatigue might contribute to exercise limitations in FMS patients.

The reduced quadriceps strength observed at the initial neuromuscular assessment in FMS patients emphasizes previous findings regarding muscle weakness in FMS (7, 15, 16). Our results provide important insights on potential underlying mechanisms. First, reduced strength in FMS patients is not due to impaired muscle activation, since VA was similar between both groups and comparable to values previously reported in healthy subjects (34). These results do not confirm previous results suggesting reduced quadriceps VA in FMS patients assessed with transcutaneous electrical stimulation (19, 20). The use of less painful (22, 26) and supramaximal FNMS in the present study (rather than submaximal electrical muscle stimulation in previous studies) can probably explain this difference. Second, when reported to estimated quadriceps volume, quadriceps strength did not differ anymore between the groups. Therefore, the reduced quadriceps strength observed in FMS patients appears to be mainly explained by slightly smaller muscle mass rather than by activation deficiency or altered contractility. This result needs to be confirmed with more accurate measurements of muscle mass, such as magnetic resonance imaging.

During the quadriceps fatigue test, FMS patients showed a greater decline in evoked muscular responses (Twp and Db100), but no significant difference (P = 0.12) in MVC decline compared to controls. Larger exercise-induced reduction in Twp compared to MVC has been repeatedly described (37–39). Evoked muscular responses have been reported to be more sensitive than MVC to detect interindividual differences (40) and drug (13) or hypoxia (41) effects when measuring muscle fatigue. A potential explanation for greater alterations of evoked muscular responses but not MVC in FMS patients is that Twp is affected by excitation–contraction coupling failure, which is partly overcome by a high-frequency alpha motor neurons discharge rate during MVC. In the present study, similar changes in Db10:100 in both groups indicate that FMS patients did not show more low-frequency muscle fatigue. In addition, no differences in M-wave characteristics were found; therefore, impairments of nerve potential propagation are not involved. Thereby, an alternative explanation is that some alterations may occur at the actin–myosin cross-bridge level at a low- but not high-force level in FMS patients. Muscle structural abnormalities (42, 43), alterations of phosphorylation and oxidative capacities (44), and neuroendocrine disorders (45) might be potential mechanisms to explain impaired muscle contractility in FMS patients during the quadriceps fatigue test. Finally, the number of submaximal contractions was not different between the groups, suggesting that task failure during the quadriceps fatigue test may depend on factors other than peripheral contractile fatigue, as discussed elsewhere (46).

The present results are not in accordance with the results of Norregaard et al (20), showing similar reductions of evoked muscular responses during intermittent isometric quadriceps contraction at 50% MVC in FMS patients and controls. The use of submaximal electrical muscle stimulation rather than supramaximal nerve stimulation and insufficient muscle potentiation in the study by Norregaard et al (20) may explain these results. Other studies on quadriceps fatigability in FMS patients are difficult to interpret because they used volitional force measurements only (14, 45, 47).

Our results also show normal muscle activation in FMS patients during intermittent quadriceps contractions, as previously observed for elbow flexors (48). However, the tendency for larger VA reduction in FMS patients cannot entirely rule out a role of activation deficit. Isolated isometric contractions of the quadriceps may induce less pain and, as a consequence, less central inhibition compared to multijoint dynamic movements able to induce pain in structures other than muscles (e.g., skin, joints, and enthesis). Therefore, central limitations may be task dependent (18) and further studies regarding this aspect in FMS are required (49).

Our results showed that VO2 max and maximum workload are reduced in FMS patients, as previously reported (5, 6). We also showed that FMS patients achieve submaximal effort, as shown by the lower percentage of maximal theoretical HR and the lower RER and blood lactate concentration at exhaustion. Submaximality of exercise testing in FMS patients has been reported in previous studies (6, 8). This submaximality is in contrast to normal VA during the quadriceps fatigue test. Larger afferent input during cycling (due to larger muscle mass recruitment and cardiorespiratory stimulation) might explain greater involvement of central mechanisms compared to isolated quadriceps isometric contractions, as discussed above.

From the current literature, it is unclear whether the physiologic exercise response at submaximal intensities in FMS patients is impaired (7, 8). In the present study, FMS patients had significantly lower HR and VE at 50% and 75% of measured VO2 max compared to controls. However, reduced VO2 max and the issue of submaximality as discussed above may explain these results, and comparisons of physiologic responses at percentages of measured VO2 max may lead to erroneous conclusions. Therefore, we expressed submaximal physiologic responses as a percentage of predicted VO2 max, i.e., the individual theoretical maximal exercise performance for age and body mass (36). In this case, HR and VE at 50% and 75% of predicted VO2 max were no longer different between FMS patients and controls (Table 3), indicating that altered physiologic responses in FMS patients previously described at percentages of measured VO2 max were mostly due to exercise submaximality. Nevertheless, the larger HR/VO2 slope and larger RER at 75% of predicted VO2 max in FMS patients still suggests some impairment of the physiologic responses at submaximal intensities. From the individual HR/VO2 slopes, we extrapolate the VO2 max that FMS patients would display if they would have reached their maximal theoretical HR (i.e., mean ± SD 177 ± 6 beats per minute, similar to the maximal HR in controls; P = 0.24). Figure 2A showed that the extrapolated VO2 max is still lower than VO2 max measured in controls (mean ± SD 28.6 ± 8.8 versus 36.1 ± 7.1 ml/minute/kg; P < 0.05). Therefore, we proposed that the reduced VO2 max measured in FMS patients compared to controls is the consequence of exercise submaximality (e.g., due to increased pain perception [8]; see below) and metabolic impairments (44). The significant correlation between VO2 max and the percentage reduction in evoked muscular responses during the quadriceps fatigue test (Figure 3A) supports the involvement of increased quadriceps fatigability in the reduced exercise capacity of FMS patients.

Contrary to HR and VE, the RPE in FMS patients was similar at 50% and 75% of measured VO2 max and higher at 50% and 75% of predicted VO2 max, and was similar at exhaustion compared to controls. The greater increase in RPE and premature maximal RPE achievement in FMS patients appeared to be a key mechanism leading to lower maximal exercise capacity. At least part of the enhanced RPE in FMS patients may be due to greater impairment in quadriceps contractility (Figure 3B) that might increase muscle afferent input (9, 10). No relationship could be observed between symptoms reported in questionnaires and neuromuscular assessments or exercise capacity, as previously reported (6, 15). Multifactorial components of FMS as well as interindividual heterogeneity might explain these results.

It is always challenging to pair control subjects with FMS patients (6). We chose to pair patients and controls for age and BMI, and to include sedentary controls in order to avoid the effect of training. Subjects had similar self-reported physical activity, but we cannot exclude that the amount of spontaneous physical activity was lower in FMS patients because we did not measure objective physical activity with an accelerometer, for instance (4). This may explain at least part of the reduced exercise and muscle performances (8). Additionally, we chose normal weight FMS patients and those who agreed to participate in a maximal exercise test and neurostimulation. This prevents us from generalizing our results to the entire FMS population, and larger samples are needed to confirm our results. Future studies may also focus on neuromuscular fatigue assessed with nerve stimulation following locomotor exercise (e.g., cycling, walking) in order to further clarify the involvement of muscle fatigability versus other mechanisms such as pain perception in functional limitations of FMS patients.

The present investigation showed that reduced quadriceps muscle strength in FMS patients is not due to impaired muscle activation, but possibly to reduced muscle mass. Greater alteration in quadriceps contractibility in FMS patients probably contributes to an enhanced perception of exertion and lowered exercise capacity. These results highlight the importance of considering neuromuscular impairments as a factor contributing to functional limitations in FMS patients and to target muscle function in the management of these patients (50).


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 published. Dr. Verges 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.

Study conception and design. Bachasson, Guinot, Wuyam, Verges.

Acquisition of data. Bachasson, Guinot.

Analysis and interpretation of data. Bachasson, Guinot, Wuyam, Favre-Juvin, Millet, Levy, Verges.