Abstract
Background Cardiopulmonary exercise testing can be considered the gold standard for assessing cardiorespiratory fitness. Little is known about the criteria for maximal exercise testing in people with multiple sclerosis (MS) and how these criteria behave across different levels of neurological disability.
Objective The study objectives were to determine the criteria for maximal exercise testing across various levels of disability and to assess concomitant subgroup differences in measures related to the participant, disease, and function.
Design This was a cross-sectional study.
Methods Cardiopulmonary exercise testing was conducted with a sample of 56 participants with MS. Analysis of variance was used to assess the criteria in participants with MS and low, mild, and moderate levels of disability.
Results Mean peak oxygen consumption (V̇o2peak) was 21.4 (SD=7.1) mL·kg−1·min−1. An oxygen consumption (V̇o2) plateau was seen in 37.5% of participants. A respiratory exchange ratio of 1.10 or greater was achieved by 69.6% of the participants, a maximal heart rate within 90% of their age-predicted maximal heart rate was achieved by 48.2% of the participants, and 23.2% of the participants perceived their exertion to be 18 or greater on the Borg Scale of Perceived Exertion (scores of 6–20). The values for achieved heart rate and incidence of a V̇o2 plateau were significantly lower in participants with moderate levels of disability than in those with mild levels of disability.
Limitations The primary limitations of this study were its cross-sectional nature and relatively small sample of participants with moderate levels of disability.
Conclusion The findings suggest that the outcome of cardiopulmonary exercise testing in people with MS and low to mild levels of disability (Expanded Disability Status Scale scores of ≤4.0) is a valid measure of cardiorespiratory fitness, whereas the outcome in people with moderate levels of disability (Expanded Disability Status Scale scores of >4.0) is most likely symptom limited.
Multiple sclerosis (MS) is a chronic demyelinating inflammatory disorder of the central nervous system with an unknown etiology.1 Progressive demyelination within the central nervous system may result in a range of motor, sensory, cerebellar, and cognitive dysfunctions that may directly or indirectly limit the cardiorespiratory fitness of patients with MS. Subsequent deconditioning may lead to reduced activity levels and concurrent comorbidities similar to those in people who do not have MS.2 This vicious circle may result in lower levels of perceived health-related quality of life.3 As a consequence, maintaining or improving cardiorespiratory fitness in people with MS is paramount. In the past 2 decades, several intervention studies investigated the feasibility and potential benefits of various types of exercise training for people with MS.4,5 These studies showed that people with MS can safely use endurance, resistance, or combined training regimens. It has also been suggested that these training regimens may prevent deconditioning or even improve cardiorespiratory fitness in people with MS.4,5
Measuring cardiorespiratory fitness in the context of clinical practice and research is important for evaluating the health status of patients and the effects of exercise regimens or other interventions.6 One potentially useful outcome measure for the quantification of exercise tolerance is peak oxygen consumption (V̇o2peak), and the preferred modality for testing V̇o2peak is cardiopulmonary exercise testing (CPET) on a cycle ergometer.7 Cardiopulmonary exercise testing preferably is accompanied by measurements of ventilation and gas exchange (ie, V̇o2peak) for the direct assessment of cardiorespiratory fitness.6 Although CPET can be considered the gold standard for the assessment of V̇o2peak,8 little is known about the validity, reliability, and responsiveness of maximal exercise testing in people with MS. Nonetheless, several cross-sectional and intervention studies have used CPET as a measure of cardiorespiratory fitness in people with MS.9–17 Studies in which CPET outcomes were used to determine the effects of various training regimens showed that V̇o2peak was markedly lower in people with MS than in people who were healthy (healthy controls) or reference data.10,11 Moreover, these studies also demonstrated that with increasing neurological disability, the cardiorespiratory fitness of people with MS decreased.15 However, in people with MS and disability, cardiorespiratory fitness tests such as CPET may be confounded by existing neurological symptoms, such as ataxia and spasticity, that affect test performance.
The American College of Sports Medicine (ACSM)18 has described criteria to determine whether a CPET outcome represents maximal or symptom-limited exercise testing. Cardiopulmonary exercise testing can be considered maximal if a plateau in oxygen consumption (V̇o2) is reached while workload is still increasing, the respiratory exchange ratio (RER) exceeds 1.10, the heart rate fails to increase with increasing exercise intensity, the posttest lactate level exceeds 8.0 mmol, and the perceived exertion exceeds 17 on the Borg Scale of Perceived Exertion (with scores of 6–20).19
To our knowledge, the few studies that have been done to assess the application of these criteria to people with MS have yielded inconsistent results.6,16 This inconsistency may reflect neglect of the severity of the neurological disability in the application of CPET for people with MS. In the present study, we addressed 4 of these criteria, namely, the V̇o2 plateau, RER, maximal heart rate, and perceived exertion. In addition, we assessed parallel differences in participant-related outcomes and measures of body functions. These outcomes may provide insight into the underlying symptoms that may limit maximal exercise performance.
The objectives of the present study were: (1) to determine the criteria for maximal exercise testing across various levels of disability and (2) to assess concomitant subgroup differences in measures related to the participant, disease, and function.
Method
Participant Selection
A convenience sample of 56 participants was selected from the Treating Fatigue in Multiple Sclerosis–Aerobic Training, Cognitive Behavioral Therapy, and Energy Conservation Management (TREFAMS-ACE) study (ISRCTN69520623).20 These participants were diagnosed with definite MS according to the revised McDonald criteria.21 Potential participants were included if they experienced substantial fatigue (with scores on the Checklist of Individual Strength fatigue subscale of >35),22 were ambulatory (with Kurtzke Expanded Disability Status Scale [EDSS] scores of ≤6.0),23 and had no clinical depression (with scores on the depression subscale of the Hospital Anxiety and Depression Scale of ≤11).24 Potential participants were excluded if they had been participating in a professionally guided exercise program, had started pharmacological fatigue therapy, or had started any other program to alleviate fatigue in the 3 months before inclusion. Potential participants also were excluded if they had comorbidities precluding maximal exercise participation. Written informed consent was obtained from each participant.
CPET
Unfortunately, there is no uniform definition for cardiorespiratory fitness. However, cardiorespiratory fitness can be defined as the ability of the body's circulatory and respiratory systems to supply fuel and oxygen during sustained physical activity.25 Incremental exercise testing was performed on a programmable, electromagnetically braked cycle ergometer (Kettler X7, Kettler, Ense-Parsit, Germany) to determine maximal exercise capacity. The handlebars and saddle were adjusted to match each participant's anthropometrics. After a 3-minute rest phase, participants started cycling at 25 W, with the power output being increased by 10 W (women) or 15 W (men) each minute. During exercise testing, participants were asked to maintain a cadence of 60 to 80 rpm. The electromagnetic braking system allowed for constant power, independent of cadence. Participants were vigorously encouraged only beyond an RER of greater than 1.00 by protocol. Exercise testing was terminated by volitional exhaustion, a cadence of less than 45 rpm, or safety reasons, in compliance with the ACSM's guidelines for clinical exercise testing.18 The protocol used was intended to produce voluntary exhaustion after 8 to 12 minutes of exercise.26
The gas exchange for each participant was measured with a portable mixing-chamber monitoring system (MetaMax 3B, Cortex Medical, Leipzig, Germany). The MetaMax is a valid and reliable system for measuring ventilation parameters during exercise.27–29 The V̇o2peak was defined as the highest recorded 10-second average during the final 2 minutes of exercise. In addition to V̇o2peak, we recorded maximal power during the incremental protocol, maximal ventilation, and exercise duration (time to V̇o2peak).
Criteria for Maximal Exercise Testing
We assessed 4 criteria for maximal exercise testing. First, we determined whether, with increasing work rate, a V̇o2 plateau was reached; this plateau was defined by a difference of ≤50 mL of V̇o2 per minute between the 30-second average before the last power increment and that after the last power increment.30 Second, the participants' maximal heart rate relative to the predicted heart rate was studied (220−age, in years; %HRmax). The participants' actual heart rate was measured continuously with a wireless monitoring system (Polar, Kempele, Finland). Third, the RER (ratio of carbon dioxide exhalation to V̇o2) at the time of V̇o2peak was considered. Fourth, perceived exertion, as rated with the Borg Scale of Perceived Exertion, was scored within 1 minute after voluntary exhaustion.19
Disability
Neurological disability was determined by a certified neurologist using the Kurtzke EDSS.23 The EDSS rating was determined before enrollment. Only participants with low to moderate levels of disability, defined as EDSS scores of 6.0 or less, were included; this criterion implies that the participants in the present study were able to walk at least 100 m using no more than 1 single-side walking aid. Disability levels were adapted from recent work by Dalgas et al.31 Participants with EDSS scores of 2.0 or less were categorized as having low levels of disability, those with EDSS scores of 2.5 to 4.0 were categorized as having mild levels of disability, and those with EDSS scores of 4.5 or greater were categorized as having moderate levels of disability. Because of the progressive nature of MS, time since diagnosis also was included as a disability-related measure.
Participant-Related Outcomes
We included age, sex, and use of heart rate–regulating medications as participant-related factors that might be associated with the various disability subgroups. In addition, participant-reported height and weight before exercise testing were recorded, and the body mass index was calculated.
Measures of Function
Ambulation was assessed with a 2-minute walk test.32 In a wide corridor, participants were asked to walk as far as they could, around 2 pylons spaced 30 m apart, for 2 minutes by following standardized instructions. No encouragements were given during the test; only a 1-minute halfway mark was provided.
Fatigue was measured with the Checklist of Individual Strength fatigue subscale.33 This checklist assesses the subjective experience of fatigue during the preceding 2 weeks and consists of 8 questions answered on a 7-point Likert scale.
Data Analysis
Baseline characteristics were described with descriptive statistics (means and standard deviations) and are presented for each level of disability. The Shapiro-Wilk test, in conjunction with visual inspection, indicated a normal distribution for all measures except EDSS and time to diagnosis. A 1-way analysis of variance with Bonferroni adjustment for multiple comparisons was used to determine differences between subgroups. For EDSS and time to diagnosis, the nonparametric Kruskal-Wallis and Mann-Whitney U tests were used. A chi-square test was used to determine differences in sex distribution and medication between subgroups. For describing the association between variables, the Spearman rank correlation coefficient was applied. All tests were performed 2-tailed with a significance level of P<.05. SPSS version 19.0 (SPSS Inc, Chicago, Illinois) was used for the statistical analyses.
Role of the Funding Source
The TREFAMS-ACE study is funded by the Fonds NutsOhra (ZonMw 89000005).
Results
A total of 56 participants (37 women and 19 men) with definite MS were assessed after enrollment into the TREFAMS-ACE study. Table 1 shows the baseline characteristics for the total sample and for each disability level. Participants had been diagnosed with MS 10.3 (SD=7.2) years before inclusion in the present study, and the average disability level (score) on the EDSS was 2.8 (SD=1.6). The mean V̇o2peak, which was 21.4 (SD=7.1) mL·kg−1·min−1, decreased significantly (Spearman rho=−.601, P<.01) with increasing disability (Figure). On average, the V̇o2peak was achieved after 11.3 (SD=3.9) minutes of exercise. Five participants were taking beta-blocking medication during the exercise testing. However, a sensitivity analysis excluding participants taking beta-blocking medication did not change the results.
Baseline Characteristics and Group Differences for Defined Levels of Neurological Disabilitya
Changes in peak oxygen consumption (in mL·kg−1·min−1; Spearman rho=−.601, P<.01; ▲) and percent maximal heart rate (Spearman rho=−.396, P<.01; ●) achieved during incremental cycling in relation to disability level (scores on the Kurtzke Expanded Disability Status Scale [EDSS]).
Criteria for Maximal Exercise Testing
Table 1 shows that in 37.5% of participants, a plateau in V̇o2 occurred while exercise intensity was still increasing. The incidence of this plateau was significantly lower in participants with moderate levels of disability (0.0%) than in those with low levels of disability (52.0%) or mild levels of disability (36.4%). Maximal heart rate findings showed that 48.2% of participants reached a maximal heart rate within 90% of their age-predicted maximal heart rate. The RER results showed that 69.6% of participants reached an RER of 1.10 or greater. Finally, 23.2% of participants perceived their exertion to be 18 or greater on the Borg Scale of Perceived Exertion. Tables 1 and 2 show that %HRmax was significantly lower in participants with moderate levels of disability than in those with mild levels of disability (P<.01). In addition, a significant correlation (Spearman rho=−.396, P<.01) was found between %HRmax and EDSS scores (Figure). Other factors for which values were significantly lower in participants with moderate levels of disability than in those with mild levels of disability were the 2-minute walk test (P<.01), maximal power during the incremental protocol (P<.01), and time to V̇o2peak (P<.01).
Estimated Mean Difference Between People With Multiple Sclerosis and Low, Mild, or Moderate Disabilitya
Discussion
The V̇o2peak (21.4 [SD=7.1] mL·kg−1·min−1) achieved by participants with MS in the present study was similar to V̇o2peak values previously reported for people with MS.6,9,11–16 However, the achieved V̇o2peak was lower than reference values reported for sedentary people who are healthy (39.0 [SD=6.8] mL·kg−1·min−1 for men and 30.0 [SD=5.4] mL·kg−1·min−1 for women), indicating that physical fitness in people with MS is indeed compromised.34 This observation emphasizes the need for a valid and reliable measure for effectively evaluating physical fitness in people with MS. The present study suggests that V̇o2peak is a valid measure of cardiorespiratory fitness only in people with low to mild levels of disability (EDSS scores of ≤4.0). In people with moderate levels of disability (EDSS scores of >4.0), V̇o2peak may be predominantly symptom limited. The V̇o2peak values achieved by people with moderate levels of disability may reflect their function rather than their fitness.
Criteria for Maximal Exercise Testing
Various central and peripheral mechanisms may limit the ability of people with MS to meet the ACSM's criteria for maximal exercise testing.18 We tested this hypothesis by assessing 4 commonly used criteria based on maximal cycling ergometry, namely, the incidence of a V̇o2 plateau, %HRmax, RER, and perceived exertion.
With respect to the definition used in the present study, 37.5% of participants were able to reach a V̇o2 plateau. This incidence is comparable to the incidence of 42% recently found in over 800 people by Edvardsen et al.35 However, the V̇o2 plateau is an elusive phenomenon and a subject of long-lasting debate.36–38 In particular, determination of the V̇o2 plateau can be complicated, and the methodology for this process may greatly influence the incidence of such a plateau. Nonetheless, our results showed that the incidence of a V̇o2 plateau was significantly lower in participants with moderate levels of disability than in those with low and mild levels of disability.
Our findings also indicated that the energy cost of low-level exercise for participants with moderate levels of disability was substantially higher than that for participants with low to mild levels of disability, as shown by the relatively high RER values parallel to earlier exercise termination in participants with MS and moderate levels of disability than in participants with MS and low to mild levels of disability. These findings, for instance, may be related to low skeletal muscle oxidative capacity39 or respiratory muscle weakness.40 However, our findings also suggest that, for participants with MS, %HRmax may be a better criterion of maximal exercise than RER. The latter notion is in agreement with the results of Mostert and Kesselring,10 who found a heart rate reserve of 37 bpm and concluded that their graded exercise testing was not ventilation limited.
In contrast, Kuspinar et al16 reported a perceived exertion of 19.3 (SD=1.0), a peak RER of 1.2 (SD=0.1), and a maximal heart rate of 94%. However, the patients in their sample had slightly lower levels of disability (median EDSS score of 1.5).16 Motl and Fernhall6 reported significantly lower RER, perceived exertion, and heart rate in patients with MS than in people who were healthy (controls). They predicted V̇o2peak based on work rate only for patients who met 2 of the 3 criteria for maximal exercise and found that for those patients, V̇o2peak could be predicted to within 10% of the true value by use of established equations based on the work rate of people who were healthy.6 These data suggested that V̇o2peak values in people who have MS and meet 2 of the 3 criteria for maximal exercise closely resemble V̇o2peak values in people who are healthy. However, in our sample, only 43% of participants met 2 of these 3 criteria. The median EDSS score for the sample in the study of Motl and Fernhall6 was 1.5; in the present study, the score was 2.25, with a comparable range (1–6.0). However, when the level of disability was taken into account, only 1 of 9 participants with moderate levels of disability in our sample met 2 of the 3 criteria for maximal exercise. These discrepancies reflect the great diversity of symptoms and limitations in people with MS.
We also must consider the possibility that the lower %HRmax in participants with MS and moderate levels of disability than in those with MS and mild levels of disability did not signal invalidity of the exercise testing but did signal invalidity of the predictive equation for estimating maximal heart rate. Theoretically, changes in autonomic function may reduce potential maximal heart rate in people with MS and make the interpretation of maximal heart rate difficult.41 However, in a more recent study, Hale et al42 were unable to identify unambiguous differences between patients who had MS and showed autonomic function before maximal exercise testing and patients who had MS and did not show autonomic function before maximal exercise testing.
In summary, physiological changes due to MS may affect the validity of common criteria for maximal exercise testing. On the basis of our findings, we argue that %HRmax may be the better indicator for maximal versus symptom-limited exercise in MS. Assessment of additional criteria may provide further insight into the validity of CPET as a measure of fitness or function.
Exercise Testing in People With MS and Moderate Levels of Disability
Our findings showed that in people with MS and moderate levels of disability and presumably also in people with MS and severe levels of disability, the outcome of maximal exercise testing may not be a valid measure of fitness. Hence, for such people, alternative measures should be considered. Recently, Skjerbaek et al43 provided one of the first results of exercise therapy for patients with severe MS (EDSS scores of 6.5–8.0). To measure their treatment effect, they used arm cycling ergometry because the patients were mostly wheelchair bound. They found a nearly significant interaction effect (P=.06) for V̇o2peak when they compared the exercise group and the control group. This finding shows that arm cycling ergometry may be responsive to training but does not show that fitness is actually being measured. Submaximal measures such as work efficiency or V̇o2 efficiency may prove to be physiological alternatives. However, these contemporary measures, like measures in which extrapolation or predictive equations are used, may be difficult to include in a training prescription.
Limitations
The present study had some limitations. First, it was a cross-sectional study with a relatively small number of participants, especially those with moderate levels of disability. Second, to our knowledge, there is no consensus on how to categorize various levels of disability on the basis of the EDSS, and the use of different cutoff values for the levels of disability may alter our conclusions. However, our findings suggest that it is feasible to combine subgroups with low and mild levels of disability and use EDSS scores of ≤4.0 as a cutoff for mild versus moderate levels of disability. Third, we included only participants with MS-related fatigue and without depressive symptoms. This fact may raise concerns regarding the generalizability of our findings to people without MS-related fatigue. However, MS-related fatigue is reported in more than 80% of all patients with MS, suggesting that the findings in the present study are representative of most people with MS.44
Future Research
Because our data are not conclusive, we believe that it is too early to advise revision of the ACSM's criteria for maximal exercise testing in people with MS. Additional information on the validity and responsiveness of CPET in people with MS is warranted, especially for people with MS and more severe levels of disability. Contemporary criteria, such as a submaximal verification phase after incremental exercise termination, as suggested by Midgley and colleagues,45,46 may be interesting alternatives for the current criteria. In addition, studies of the underlying mechanisms that lead to symptom-limited exercise termination may provide additional insight into the exercise capacity of people with MS.
In conclusion, the present study suggests that V̇o2peak in people with MS and moderate levels of disability (ie, classified as having EDSS scores of >4.0 and ≤6.0) is markedly lower than in people who are healthy. In people with moderate levels of disability, the outcome of incremental exercise testing is likely to be limited by central dysfunction, peripheral dysfunction, or both due to MS rather than by cardiorespiratory fitness.
Footnotes
Mr Heine, Dr Hacking, and Dr Kwakkel provided concept/idea/research design. All authors provided writing. Mr Heine provided data collection and analysis, project management, and institutional liaisons. Dr Hacking provided study participants and institutional liaisons. Dr Hoogervorst, Dr Hacking, Dr Verschuren, and Dr Kwakkel provided consultation (including review of manuscript before submission).
The authors thank the staff of the Multiple Sclerosis Center, St Antonius Hospital, Nieuwegein, the Netherlands, and Erasmus Medical Center, Rotterdam, the Netherlands, for their efforts to find eligible participants for the present study.
The TREFAMS-ACE study was approved by the Medical Ethical Board of VU University Medical Center, Amsterdam.
The TREFAMS-ACE study is funded by the Fonds NutsOhra (ZonMw 89000005).
- Received September 6, 2013.
- Accepted March 20, 2014.
- © 2014 American Physical Therapy Association
References
Issue highlights
