Multiple sclerosis (MS) is an autoimmune and demyelination disease of the central nervous system that causes progressive accumulation of disability over time [1
]. Overweight and obesity typify common comorbid conditions found with MS. Marck et al. found that among 2399 people with MS (PwMS), 22.5% were overweight and 19.4% were obese. Notably, overweight and obese PwMS have higher rates of diabetes, elevated blood pressure, increased insulin resistance, blood lipid issues, depression, anxiety, and arthritis compared with people of normal weight [2
Although physical activity (PA) is a standard recommendation for weight control in the general population [3
], conflicting evidence exists regarding the relationship between body weight status and PA in PwMS. One study investigated the effects of a 6-month internet-delivered PA behavioral intervention on the body composition of 82 PwMS. The researchers reported that although the intervention resulted in higher intensities of PA, there were no significant differences in body composition [4
]. A recent study reported that 12 weeks of high-intensity aerobic and strength training did not alter the total fat and lean body mass in the MS sample [5
]. In contrast, a long-term (12 months) endurance exercise program resulted in a significant decrease in body fat in 89 PwMS individuals [6
]. Moreover, 24 weeks of combined endurance and resistance training demonstrated improvements in body composition, especially in the lean tissue mass in a group of 22 PwMS [7
]. This inconclusive evidence reinforces the need for new information in order to clarify the relationship between body weight and PA in the MS population.
In the present study, we focused on leisure-time PA (i.e., walking, dancing, hiking, and swimming) that people typically engage in during their free time. It is worth noting that the World Health Organization’s (WHO) global recommendations to perform PA to benefit one’s health are mainly based on leisure-time PA. Therefore, the primary aim of our study was to examine the relationship between weight status (via body mass index) and leisure-time PA in a cohort of PwMS. Furthermore, we examined this relationship according to the patient’s level of neurological disability, represented by the Expanded Disability Status Scale (EDSS) score. Our hypothesis was that obesity would be related to decreased participation in leisure time PA and a higher level of disability.
The median EDSS for the entire study group was 2.0 (range 0.0–6.5), mean disease duration was 6.4 (standard deviation [SD] = 8.2) years, and mean age was 40.5 (SD = 12.9) years. The mean BMI and GLTEQ scores of the total group were 24.7 (SD = 4.7) and 16.4 (SD = 18.0), respectively. Demographic and clinical characteristics according to weight groups are presented in Table 1
. The odds ratio (OR) (95% confidence interval [CI]) between leisure-time PA and BMI according to the WHO definition was 1.070 (0.548–2.086; p
= 0.844) for overweight and 1.648 (0.698–3.888; p
= 0.254) for obesity without adjustment, and 1.126 (0.521–2.436; p
= 0.763) for overweight and 1.093 (0.442–2.702; p
= 0.847) for obesity after adjusting for age, gender, and disability status (Table 2
). According to the chi-squared analysis, no correlation was found (x2
= 0.420, p
= 0.564) between leisure-time PA and obesity (based on the BMI threshold for PwMS provided by Pilutti and Motl ) (Table 3
The OR (95% CI) between disability level and BMI according to WHO’s definition was 1.674 (0.735–3.810; p
= 0.220) for overweight and 0.618 (0.172–2.221; p
= 0.460) for obesity without adjustment, and 1.130 (0.466-2.738; p
= 0.787) for overweight and 0.447 (0.119-1.682; p
= 0.234) for obesity after adjustment for age, sex, and leisure-time PA (Table 4
). The chi-squared analysis showed no relationship between level of disability and obesity (x2
= 0.179, p
= 0.701) according to the BMI thresholds for obesity presented by Pilutti and Motl (2016) (Table 5
). Figure 1
illustrates the relationship between leisure-time PA and BMI according to the disability subgroups.
Our main objective was to investigate the relationship between PA and obesity in PwMS. Accordingly, no associations were found between leisure-time PA and BMI in PwMS. This conclusion was confirmed by WHO’s criteria for obesity and the cutoff scores of Pilluti and Motl (2016) for assessing obesity in PwMS. Our outcome is in agreement with previous studies that did not demonstrate a relationship between PA with being overweight or obese, based on BMI [14
]. Furthermore, the recent review of Ewanchuk et al. strengthens this line of evidence [17
]. The authors found that 11 (out of 17) exercise interventional studies did not demonstrate any effect of PA programs (aerobic, resistance, and combined) on obesity outcomes, including BMI.
Nonetheless, there is evidence suggesting that PwMS who perform more PA have a lower BMI and are less likely to be obese [18
]. This discrepancy may be explained by the parameters used to assess obesity. Although BMI is the most common measurement for obesity, it does have several limitations. Mainly, it does not distinguish between body fat, lean tissue mass, and bone mineral density. There is the possibility that including additional measures of body composition in the present study may have resulted in different conclusions.
Some may argue that the decision to focus solely on leisure-time PA may possibly be the reason that no significant associations were found with obesity. It is worth noting that according to the American College of Sports Medicine’s weight loss and controlling guidelines: (i) at least 150 min per week of moderate intense PA is needed to prevent weight gain; (ii) 150–250 min per week (approximately 1200–2000 kcal per week) of moderate intense PA is needed to prevent weight gain >3% and is associated with modest weight loss, and (iii) approximately 250–300 min per week (approximately 2000 kcal per week) of moderate intense PA is needed for greater weight loss and prevention of weight regain [3
An additional finding of our study was the absence of any association between obesity and disability level in the MS cohort. Although this topic has been thoroughly investigated, the existing evidence is contradictory. Several previous studies have found that weight status is related to mobility and associated disability levels, indicating that PwMS who were less mobile (higher EDSS) had a higher BMI [20
]. In contrast, Pilutti et al. (2012) reported that over a 24-month time period, BMI was not predictive of disability in a sample of 269 people with relapsing-remitting MS [14
]. In a study of mild-moderate PwMS (median EDSS = 4.0) [21
], an association between disability and weight status was not identified. Collectively, the research does not support a strong and consistent association between disability and BMI in PwMS, and our data strengthen this line of evidence. Furthermore, although not directly connected with disability, it is worth noting that the risk of cardiovascular disease, which is increased in PwMS, is not related to obesity or changes in body composition [22
]. Therefore, we believe that BMI is not as important of a factor in disability as other demographic and clinical characteristics in PwMS. Moreover, it is questionable if BMI should be targeted as a means for reducing disability in MS.
Regardless of our results, we find it important to notice that the benefits of PA in PwMS have been well documented [23
]. PA beneficially affects a variety of MS symptoms (i.e., fatigue, mobility, depressive symptoms, muscle strength, and quality of life). Moreover, recent evidence suggests that exercise may also have disease-modifying effects and possibly preventive effects by lowering disease risk.
The main strengths of this study include the relatively large sample and the use of two criteria for obesity. Nevertheless, our study has several limitations that warrant attention. First, dietary patterns, smoking, and alcohol consumption, all of which affect weight status, were not examined. Second, leisure time PA was quantified solely by a self-report questionnaire and not by direct measurement of the participants’ PA levels. Although the GLTEQ is a recommended measuring tool for evaluating PA in PwMS [11
], it does have certain disadvantages. One disadvantage of this tool is that it does not take into account the length of time it takes to perform the PA, only that it lasts for at least 15 consecutive minutes. For instance, there was no difference in the score between a person who performed fast walking 3 times a week for 20 minutes or 3 times a week for one hour. Therefore, although the GLTEQ is recommended to differentiate between active and insufficiently active people, there is a possibility that it lacks the sensitivity to differentiate between the activity levels recommended for weight control. We speculate that PA assessed with other tools such as accelerometers and pedometers might demonstrate different relationships with BMI compared with leisure-time PA in the MS population. Additionally, there seems to be a discrepancy between objectively vs. subjectively assessed PA in PwMS [24
]. Moreover, in a recent validation study of the GLTEQ, it was found that the GLTEQ scores primarily reflect moderate-to-vigorous PA rather than light PA and sedentary behavior in PwMS [25
]. Finally, this was a cross-sectional study; consequently, we cannot determine causality (although no relationships were demonstrated).