ABSTRACT
Background
Smoking and obesity interact to exacerbate the risk of hypertension, diabetes, and cardiovascular disease, but their potential synergistic effects on outcomes in multiple sclerosis (MS) have not been well studied. We aimed to study whether smoking and obesity interact to affect disease progression and cognitive function in patients with MS.
Methods
Incident cases from the population‐based case–control study Epidemiological Investigation of MS (EIMS) were categorized by smoking and obesity status at diagnosis and followed up to 15 years postdiagnosis through the Swedish MS registry (n = 3336). Cox regression was used to analyze outcomes, including clinical disease worsening (CDW), progression to Expanded Disability Status Scale (EDSS) levels 3 and 4, physical worsening as measured by a 7.5‐point increase in the MS Impact Scale (MSIS) physical score, and cognitive decline, defined as an 8‐point or greater reduction on the Symbol Digit Modalities Test (SDMT). Interaction effects on the additive scale were assessed by combining dichotomous variables for smoking (nonsmoker = 0, smoker = 1) and obesity (nonobese = 0, obese = 1), yielding four categories: 0/0 (reference category), 0/1, 1/0, and 1/1.
Results
Additive interactions between smoking and obesity were identified for CDW (attributable proportion due to interaction [AP] 0.18, 95% CI 0.03–0.30), progression to EDSS 4 (AP 0.18, 95% CI 0.08–0.26), MSIS‐Physical score worsening (AP 0.32, 95% CI 0.21–0.42), and cognitive decline (AP 0.27, 95% CI 0.19–0.35).
Conclusions
Smoking and obesity appear to synergistically worsen MS progression and cognitive functioning, with the observed additive interactions across most outcomes suggesting that these factors partly share common biological pathways contributing to disease progression.
Keywords: disability progression, expanded disability status scale, life quality, multiple sclerosis, obesity, smoking
1. Background
Multiple sclerosis (MS) is the most common chronic inflammatory disease of the central nervous system, mainly affecting young people. Typically, a phase of treatable bouts of inflammatory demyelination is followed by gradually increasing disability and brain atrophy [1]. The disease results from a combination of genetic predisposition, environmental exposures, and lifestyle habits, including smoking and adolescent obesity [1].
In the context of MS, increasing evidence also links smoking and obesity with increased disease activity and faster disease progression [2, 3, 4, 5] and with lower cognitive performance [3, 5, 6]. Furthermore, smoking and obesity have been associated with reduced efficacy of immune‐modifying therapy in MS [7, 8, 9, 10]. Smoking and obesity seem to interact to exacerbate the risk of hypertension, diabetes, and cardiovascular disease [11, 12]. However, synergistic effects between smoking and obesity have not been investigated with regard to outcomes in MS.
Considering the important socioeconomic impact of MS, it is of utmost importance to understand how environmental exposures and modifiable lifestyle habits, and their possible interactions, affect disease outcomes and identify targets for secondary preventive measures. In the current study, we followed up 3336 individuals with MS from an incident case–control study to investigate the possible additive interaction between smoking and obesity on disease progression and cognitive performance.
2. Methods
Epidemiologic Investigation of Multiple Sclerosis (EIMS) is a population‐based case–control study with incident cases of MS. Cases for the present study were recruited from hospital‐based and privately run neurology units between April 2005 and 2019 (n = 3567) with a response rate of 93%. Each case was diagnosed by a local neurologist according to the McDonald criteria [13, 14]. Study design and methods are described in detail elsewhere [15].
We follow up patients from EIMS through the Swedish MS registry, which is part of the clinical documentation system and used in all Swedish neurology departments. Information is continuously updated regarding medical treatment, disease activity, cognitive functioning, and health‐related quality of life. Of the 3567 patients, 3336 (94%) were followed up with EDSS scores in the Swedish MS registry and were included in the study. The study was approved by the Regional Ethical Review Board at Karolinska Institute and has been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.
2.1. Definition of Exposures
EIMS used a standardized questionnaire to collect information regarding environmental exposures and lifestyle habits. BMI at diagnosis was calculated using self‐reported height and weight. Since obesity but not overweight has been associated with unfavorable disease outcomes in MS [5], we dichotomized BMI into obese and nonobese. Similarly, current smoking, but not past smoking, at diagnosis has been associated with worse disease outcomes [3], and smoking was therefore dichotomized into current smokers and nonsmokers.
2.2. Outcome Measures
Baseline was defined as the date of the first recorded EDSS. Confirmed disability worsening (CDW) was defined as an increase in the EDSS score of at least 1 point from baseline sustained between two follow‐up visits separated in time by no < 6 months (1.5 point if EDSS at baseline was 0, 0.5 points if the baseline EDSS ≥ 5.5). Time to milestones EDSS 3 and 4 were studied as secondary outcomes and were limited to subgroups of patients with a baseline EDSS of < 3 and < 4, respectively. Other secondary outcomes were changes in the physical components of the MS Impact Scale 29, consisting of 20 questions about the subjective physical impacts of MS [16]. An increased score of 7.5 points or more in the MSIS‐29 component was defined as a worsening from the patient's perspective, based on prior research in the SELECT trial [17]. In this trial, an increased score of 7.5 points or more was identified as a clinically significant threshold for physical worsening, supported by both anchor‐ and distribution‐based methods and aligned with changes observed in EDSS scores. The Symbol Digit Modalities Test (SDMT) was used as a measure of cognition performance outcome [18]. An 8‐point or greater worsening on the SDMT was defined as cognitive decline [19].
2.3. Statistical Analysis
Categorical variables were summarized using frequency and percentage. Continuous variables were summarized using mean and standard deviation (SD) or median and interquartile range (IQR) as appropriate. We analyzed time to 24‐week CDW, the secondary milestones EDSS 3 and 4 end‐points and physical decline as measured by MSIS‐29 and cognitive decline by using multivariable Cox proportional hazard regression.
A potential interaction between smoking and obesity was assessed using the departure of additivity of effects as the criterion of interaction and quantified by calculating the attributable proportion due to interaction (AP), the relative excess risk due to interaction (RERI), and the synergy index (SI), together with 95% CI. The interaction was tested by combining the dichotomous variables for smoking (nonsmoker = 0, smoker = 1) and obesity (nonobese = 0, obese = 1), resulting in the following categories: 0/0: neither smoker nor obese (reference category), 0/1: nonsmoker and obese, 1/0: smoker and nonobese, and 1/1: smoker and obese.
Stratified analyses were performed to evaluate potential effect modification on the relative scale. We stratified the analyses by smoking status to examine whether obesity influences the impact of smoking on MS outcomes and by BMI status to investigate whether smoking influences outcomes differently in obese and nonobese individuals.
All analyses were adjusted for age at diagnosis, sex, disease phenotype (relapsing or progressive onset), disease duration (between onset and baseline), baseline EDSS, and disease‐modifying therapy. Treatment was accounted for by using the proportion of the follow‐up time on treatment, distinguishing between low‐ and high‐efficacy disease‐modifying therapies.
In a sensitivity analysis, we further adjusted for physical activity and sun exposure. Physical activity was categorized into sedentary leisure time, moderate exercise, regular exercise 1–2 times/week, and regular exercise three or more times/week. We calculated an index, ranging between 3 and 12, for sun exposure based on questions regarding the frequency of sunbathing, traveling to sunnier countries, and the use of sunbeds. Sun exposure was then dichotomized into low or high exposure using the median value as a cutoff (6). All analyses were conducted in Stata version 16.1 and Statistical Analysis System (SAS) version 9.4.
3. Results
Our study comprised 3336 individuals with MS, of which 71% were females. Mean age at baseline was 37.8 years, and the mean duration since the onset of disease, defined as the first clinical symptoms, was 3.2 years. Baseline characteristics of cases, by smoking and BMI status, are presented in Table 1.
TABLE 1.
Baseline characteristics by smoking and BMI status at diagnosis.
| Total | Smoking | Nonsmoking | p | |||
|---|---|---|---|---|---|---|
| Obese | Nonobese in2 | Obese | Nonobese in1 | |||
| N | 3336 | 112 | 624 | 362 | 2238 | |
| Age at diagnosis (SD) | 37.7 (11.1) | 35.1 (9.4) | 36.2 (11.1) | 40.6 (10.4) | 37.7 (11.2) | < 0.0001 |
| Female, n (%) | 2361 (71) | 86 (77) | 438 (70) | 274 (76) | 1563 (70) | 0.03 |
| Nordic origin, n (%) | 2637 (80) | 82 (75) | 465 (75) | 300 (84) | 1790 (81) | 0.0003 |
| Treatment, n (%) | 3178 (95) | 110 (98) | 624 (96) | 341 (94) | 2128 (95) | 0.29 |
| MS phenotype | ||||||
| Relapsing onset, n (%) | 3138 (94) | 107 (96) | 589 (94) | 339 (94) | 2103 (94) | 0.04 |
| Progressive onset, n (%) | 152 (4.6) | 1 (0.6) | 28 (4.5) | 15 (4.1) | 108 (4.8) | |
| Unknown, n (%) | 46 (1.4) | 5 (3.6) | 7 (1.1) | 8 (2.2) | 27 (1.2) | |
| Disease duration at diagnosis, years (SD) | 3.2 (5.3) | 2.2 (3.2) | 3.1 (5.5) | 3.1 (5.3) | 3.2 (5.3) | 0.14 |
| Baseline EDSS (SD) | 1.8 (1.4) | 1.8 (1.3) | 1.9 (1.5) | 2.0 (1.6) | 1.7 (1.4) | 0.002 |
| Baseline MSIS‐PHYS (SD) | 20 (21) | 27 (22) | 25 (23) | 26 (23) | 18 (20) | < 0.0001 |
| Baseline SDMT (SD) | 52 (12) | 52 (11) | 50 (12) | 51 (11) | 53 (12) | < 0.0001 |
All outcomes were negatively affected by current smoking, with adjusted ORs of adverse outcomes ranging between 1.21 and 1.27 (Table 2). Compared to having a BMI < 30 kg/m2, obesity was associated with an increased risk of reaching EDSS 3 (adjusted HR 1.45, 95% CI 1.19–1.78) and EDSS 4 (adjusted HR 1.37, 95% CI 1.08–1.69), and with an increased risk of physical worsening according to the MSIS‐Physical score (HR 1.40, 95% CI 1.17–1.66 Table 3).
TABLE 2.
HR with 95% CI of having unfavorable outcomes postdiagnosis by smoking status at diagnosis.
| Current smoking | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b |
|---|---|---|---|---|---|
| First clinical disease worsening (CDW) | |||||
| No | 2600 | 6.0 (4.2) | 1154 (44) | 1.0 (reference) | 1.0 (reference) |
| Yes | 736 | 5.9 (4.4) | 362 (49) | 1.13 (1.00–1.27) | 1.21 (1.07–1.36) |
| Current smoking | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b |
|---|---|---|---|---|---|
| EDSS 3 | |||||
| No | 2044 | 7.0 (4.5) | 575 (28) | 1.0 (reference) | 1.0 (reference) |
| Yes | 561 | 7.1 (4.8) | 187 (33) | 1.20 (1.02–1.42) | 1.22 (1.04–1.45) |
| Current smoking | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b |
|---|---|---|---|---|---|
| EDSS 4 | |||||
| No | 2411 | 7.8 (4.6) | 419 (17) | 1.0 (reference) | 1.0 (reference) |
| Yes | 663 | 8.1 (4.8) | 141 (21) | 1.21 (1.00–1.47) | 1.27 (1.04–1.54) |
| Current smoking | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b , c |
|---|---|---|---|---|---|
| Physical decline (increased MSIS‐29 physical score by 7.5 or more) | |||||
| No | 2055 | 5.0 (4.4) | 826 (40) | 1.0 (reference) | 1.0 (reference) |
| Yes | 574 | 4.9 (4.8) | 259 (45) | 1.15 (1.00–1.32) | 1.22 (1.05–1.40) |
| Current smoking | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b , d |
|---|---|---|---|---|---|
| Cognitive decline (decreased SDMT score by 8 or more) | |||||
| No | 2124 | 5.2 (3.2) | 524 (25) | 1.0 (reference) | 1.0 (reference) |
| Yes | 600 | 5.4 (3.3) | 154 (26) | 1.02 (0.85–1.22) | 1.21 (1.01–1.45) |
Crude.
Adjusted for age at diagnosis, sex, disease phenotype, disease duration, baseline EDSS, disease‐modifying therapy, and obesity.
Adjusted for baseline MSIS‐PHYS.
Adjusted for baseline SDMT score.
TABLE 3.
HR with 95% CI of having unfavorable outcomes postdiagnosis by BMI status at diagnosis.
| BMI status (kg/m2) | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b |
|---|---|---|---|---|---|
| First clinical disease worsening (CDW) | |||||
| 18.5–30 | 2862 | 6.0 (4.4) | 1303 (46) | 1.0 (reference) | 1.0 (reference) |
| > 30 | 474 | 5.3 (3.8) | 213 (45) | 1.14 (0.98–1.33) | 1.16 (1.00–1.36) |
| BMI status (kg/m2) | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b |
|---|---|---|---|---|---|
| EDSS 3 | |||||
| 18.5–30 | 2250 | 7.2 (4.6) | 638 (28) | 1.0 (reference) | 1.0 (reference) |
| > 30 | 355 | 5.9 (4.2) | 124 (35) | 1.49 (1.22–1.82) | 1.45 (1.19–1.78) |
| BMI status (kg/m2) | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b |
|---|---|---|---|---|---|
| EDSS 4 | |||||
| 18.5–30 | 2655 | 8.0 (4.7) | 473 (18) | 1.0 (reference) | 1.0 (reference) |
| > 30 | 419 | 7.1 (4.3) | 87 (21) | 1.38 (1.08–1.74) | 1.37 (1.08–1.69) |
| BMI status (kg/m2) | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b , c |
|---|---|---|---|---|---|
| Physical decline (increased MSIS‐29 physical score by 7.5 or more) | |||||
| 18.5–30 | 2265 | 5.0 (4.4) | 912 (40) | 1.0 (reference) | 1.0 (reference) |
| > 30 | 364 | 4.3 (4.4) | 173 (48) | 1.42 (1.20–1.69) | 1.40 (1.17–1.66) |
| BMI status (kg/m2) | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b , d |
|---|---|---|---|---|---|
| Cognitive decline (decreased SDMT score by 8 or more) | |||||
| 18.5–30 | 2338 | 5.3 (3.3) | 583 (25) | 1.0 (reference) | 1.0 (reference) |
| > 30 | 386 | 4.8 (2.8) | 95 (25) | 1.16 (0.92–1.45) | 1.15 (0.95–1.42) |
Crude.
Adjusted for age at diagnosis, sex, disease phenotype, disease duration, baseline EDSS, disease‐modifying therapy, and smoking.
Adjusted for baseline MSIS‐PHYS.
Adjusted for baseline SDMT score.
Categorizing the patients based on current smoking and obesity, using nonobese nonsmokers as the reference group, revealed significant additive interactions between smoking and obesity regarding the risk of CDW, reaching EDSS 4, MSIS‐Physical score worsening, and cognitive disability worsening, with APs ranging from 0.18 to 0.32 (Table 4). For progression to EDSS 3, the AP was 0.14 (95% CI −0.02 to 0.21), indicating a potential interaction; however, this did not reach statistical significance.
TABLE 4.
HR with 95% CI of having unfavorable outcomes postdiagnosis by smoking and obesity status at diagnosis.
| BMI status | Smoking | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b | Interaction measure (95% CI) |
|---|---|---|---|---|---|---|---|
| First clinical disease worsening (CDW) | |||||||
| Nonobese | No | 2238 | 6.0 (4.3) | 991 (44) | 1.0 (reference) | 1.0 (reference) | |
| Nonobese | Yes | 624 | 6.1 (4.5) | 312 (50) | 1.13 (0.99–1.28) | 1.15 (1.01–1.31) | |
| Obese | No | 362 | 5.4 (3.8) | 163 (45) | 1.11 (0.94–1.31) | 1.13 (0.97–1.34) | |
| Obese | Yes | 112 | 4.9 (3.9) | 50 (45) | 1.25 (0.94–1.67) | 1.47 (1.10–1.96) |
AP 0.18 (0.03–0.30) RERI 0.22 (0.00–0.44) SI 1.82 (0.49–5.28) |
| BMI status | Smoking | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b | Interaction measure (95% CI) |
|---|---|---|---|---|---|---|---|
| EDSS 3 | |||||||
| Nonobese | No | 1780 | 7.2 (4.6) | 481 (27) | 1.0 (reference) | 1.0 (reference) | |
| Nonobese | Yes | 470 | 7.4 (4.8) | 157 (33) | 1.20 (1.01–1.45) | 1.16 (0.97–1.40) | |
| Obese | No | 264 | 6.1 (4.1) | 94 (36) | 1.49 (1.19–1.86) | 1.39 (1.13–1.78) | |
| Obese | Yes | 91 | 5.4 (4.7) | 30 (33) | 1.63 (1.13–2.36) | 1.76 (1.21–2.56) |
AP 0.14 (−0.02–0.21) RERI 0.21 (−0.04–0.48) SI 1.38 (0.68–2.37) |
| BMI status | Smoking | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b | Interaction measure (95% CI) |
|---|---|---|---|---|---|---|---|
| EDSS 4 | |||||||
| Nonobese | No | 2096 | 7.9 (4.7) | 352 (17) | 1.0 (reference) | 1.0 (reference) | |
| Nonobese | Yes | 559 | 8.3 (4.9) | 121 (22) | 1.24 (1.01–1.54) | 1.21 (1.00–1.50) | |
| Obese | No | 315 | 7.2 (4.2) | 67 (21) | 1.39 (1.07–1.81) | 1.33 (1.03–1.74) | |
| Obese | Yes | 104 | 6.7 (4.4) | 20 (19) | 1.41 (0.89–2.20) | 1.86 (1.17–2.91) |
AP 0.18 (0.08–0.26) RERI 0.31 (0.01–0.50) SI 1.64 (0.44–2.99) |
| BMI status | Smoking | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b , c | Interaction measure (95% CI) |
|---|---|---|---|---|---|---|---|
| Physical decline (increased MSIS‐29 physical score by 7.5 or more) | |||||||
| Nonobese | No | 1723 | 5.0 (4.2) | 674 (39) | 1.0 (reference) | 1.0 (reference) | |
| Nonobese | Yes | 467 | 5.2 (5.1) | 202 (43) | 1.09 (0.94–1.27) | 1.14 (0.98–1.34) | |
| Obese | No | 273 | 4.6 (4.8) | 124 (45) | 1.28 (1.05–1.55) | 1.27 (1.04–1.55) | |
| Obese | Yes | 91 | 3.4 (2.4) | 49 (54) | 1.87 (1.40–1.50) | 2.02 (1.51–2.72) |
AP 0.32 (0.21–0.42) RERI 0.66 (0.44–0.87) SI 2.85 (0.98–5.13) |
| BMI status | Smoking | N | Years (SD) | Outcome (%) | HR (95% CI) a | HR (95% CI) b , d | Interaction measure (95% CI) |
|---|---|---|---|---|---|---|---|
| Cognitive decline (decreased SDMT score by 8 or more) | |||||||
| Nonobese | No | 1834 | 5.3 (3.3) | 457 (25) | 1.0 (reference) | 1.0 (reference) | |
| Nonobese | Yes | 504 | 5.5 (3.4) | 126 (25) | 0.97 (0.80–1.18) | 1.15 (0.96–1.40) | |
| Obese | No | 290 | 4.8 (2.8) | 67 (23) | 1.02 (0.79–1.32) | 1.05 (0.81–1.36) | |
| Obese | Yes | 96 | 4.7 (2.8) | 28 (29) | 1.34 (0.91–1.96) | 1.60 (1.09–2.36) |
AP 0.27 (0.19–0.35) RERI 0.42 (0.06–0.67) SI 2.38 (0.91–6.55) |
Abbreviations: AP, attributable proportion due to interaction; RERI, relative excess risk due to interaction; SI, synergy index.
Crude.
Adjusted for age at diagnosis, sex, disease phenotype, disease duration, baseline EDSS, and disease‐modifying therapy.
Adjusted for baseline MSIS‐PHYS.
Adjusted for baseline SDMT score.
The HR for reaching EDSS 4 was 1.21 (95% CI 1.00–1.50) among nonobese patients who smoked, and 1.33 (95% CI 1.03–1.74) among obese nonsmokers, whereas the double‐exposed group had an HR of 1.86, 95% CI 1.17–2.91. The AP was 0.18 (95% CI 0.05–0.26) and the RERI was 0.32, 95% CI 0.01–0.50, suggesting an additive interaction between smoking and obesity with respect to the risk of reaching EDSS 4. Synergistic effects of smoking and obesity appeared most pronounced for MSIS‐Physical score worsening and cognitive disability worsening. While the SI did not reach statistical significance in our analysis, the estimates suggested an interaction consistent with the findings from AP and RERI (Table 4). Our results remained similar when we further adjusted the analysis for physical activity and sun exposure habits. Stratified analyses by smoking status showed no consistent amplification of the effects of obesity by smoking (Table S1). When stratified by BMI status, smoking was more strongly associated with adverse outcomes in individuals with obesity than in those without obesity (Table S2).
4. Discussion
In our study, we observed that smoking and obesity significantly interacted to increase the risk of disability worsening and cognitive decline in MS.
Our findings indicate that smoking and obesity interacted on the additive scale to increase the risk of disability progression in MS across most outcomes, with each factor individually contributing to increased risk and an amplified effect when both exposures were present. In contrast, cognitive decline exhibited a somewhat different interaction pattern. The additive interaction terms indicated that each exposure on its own had minimal or nonsignificant effects on the outcome. However, when both exposures were combined, the risk increased substantially. This pattern may reflect a distinct sensitivity of cognitive outcomes to the combined exposures, or it may be due to limitations in sample size affecting the precision of these estimates.
The significant additive interactions observed across most outcomes suggest that smoking and obesity may partly share common biological pathways contributing to MS progression. Both factors induce a sustained, long‐term inflammatory response by increasing the production and release of inflammatory mediators. Smoking appears to affect the differentiation and function of adipocytes in adipose tissue, with elevated levels of cytokines further triggering adipose tissue inflammation [20, 21]. Similarly, obesity exacerbates systemic inflammation, with accumulating evidence indicating that central inflammation can trigger local inflammation within the brain and contribute to neurodegeneration [22, 23]. Smoking and obesity also induce changes in the gut microbiome and metabolic pathways, leading to increased intestinal permeability and systemic inflammation [24, 25]. Furthermore, both exposures induce epigenetic modifications that may lead to increased expression of multiple inflammatory genes [26, 27].
Our stratified analyses suggest that effect modification on the relative scale may occur, with obesity potentially amplifying the harmful effects of smoking on MS progression. This finding may reflect the role of obesity as a chronic pro‐inflammatory state, making the central nervous system more vulnerable to smoking‐induced damage. Conversely, we observed no clear evidence that smoking amplifies the effects of obesity, suggesting an asymmetry in modulation. This asymmetry could be due to the more stable nature of obesity as a risk factor compared to smoking, which is more often subject to behavioral changes over time. These results highlight a complex interplay between these exposures, where persistent inflammation from obesity may enhance the detrimental effects of intermittent or sustained smoking.
Underlying factors, associated with both obesity and smoking, such as increased susceptibility to infections, may also have an influence on the disease progression. Furthermore, obesity and smoking increase the risk of comorbid conditions, such as cardiovascular disease, metabolic conditions, and lung diseases [28]. These coexisting diseases may also contribute to increased disability in MS.
While there is currently no evidence that weight loss directly benefits MS‐specific outcomes [29, 30], smoking cessation has repeatedly been associated with less disability and slower disease progression compared to continuing smoking [31, 32]. Given the synergistic effects between smoking and obesity, secondary interventions should be particularly important for obese patients who smoke. Smoking cessation among MS patients with obesity would not only reduce the independent risks associated with smoking but also eliminate the synergistic effect between smoking and obesity, potentially leading to substantially reduced risks of unfavorable disease outcomes.
Our study had a population‐based design, and the response rate was high. Information regarding smoking and BMI was collected at study inclusion and should therefore not be subjected to recall bias. Although self‐reported weight to some extent may be underestimated, several studies have found high agreement between self‐reported and measured weight [33, 34]. The Swedish MS registry, which is continuously updated, provides a unique opportunity to follow up patients over time. We were able to consider numerous potential confounding factors; however, we cannot completely rule out residual confounding or the existence of characteristics or confounding linked to smoking and BMI that we do not adjust for.
In conclusion, we observed synergistic interactions between smoking and obesity in relation to disability progression and cognitive decline. These findings underscore the significance of targeted interventions for individuals who both smoke and are obese, highlighting the importance of secondary prevention measures within this specific group.
Author Contributions
Johansson Eva: writing – original draft, investigation, conceptualization, methodology, writing – review and editing. Tomas Olsson: funding acquisition, conceptualization, methodology, writing – review and editing, investigation. Lars Alfredsson: funding acquisition, conceptualization, methodology, writing – review and editing, investigation. Anna Karin Hedström: writing – original draft, funding acquisition, investigation, conceptualization, methodology, writing – review and editing, formal analysis, supervision.
Conflicts of Interest
Alfredsson reports grants from the Swedish Research Council, the Swedish Research Council for Health Working Life and Welfare, and the Swedish Brain Foundation during the conduct of the study; personal fees from Teva and Biogene Idec outside the submitted work. Olsson has received lecture/advisory board honoraria and unrestricted MS research grants from Biogen, Novartis, Sanofi, and Merck. Johansson and Hedström have declares no conflicts of interest.
Supporting information
Tables S1–S2.
Funding: The study was supported by grants from the Swedish Research Council (2016‐02349 and 2020‐01998), the Swedish Research Council for Health, Working Life and Welfare (2015‐00195 and 2019‐00697), the Swedish Brain Foundation (FO2020‐0077 and FO2024‐0344), the Swedish Medical Research Council, Margaretha af Ugglas Foundation, and the Swedish Foundation for MS Research.
Data Availability Statement
Anonymized data underlying this article will be shared on reasonable request from any qualified investigator who wants to analyze questions that are related to the published article.
References
- 1. Ward M. and Goldman M. D., “Epidemiology and Pathophysiology of Multiple Sclerosis,” Continuum 28 (2022): 988–1005. [DOI] [PubMed] [Google Scholar]
- 2. Rosso M. and Chitnis T., “Association Between Cigarette Smoking and Multiple Sclerosis. A Review,” JAMA Neurology 77 (2020): 245–253. [DOI] [PubMed] [Google Scholar]
- 3. Wu J., Alfredsson L., Olsson T., et al., “The Influence of Tobacco on Multiple Sclerosis Disease Activity and Progression,” Journal of Neurology, Neurosurgery, and Psychiatry 94 (2023): 589–596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Castro K., Ntranos A., Amatruda M., et al., “Body Mass Index in Multiple Sclerosis Modulates Ceramide‐Induced DNA Methylation and Disease Course,” eBioMedicine 43 (2019): 392–410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Wu J., Alfredsson L., Olsson T., et al., “Obesity Negatively Affects Disease Progression, Cognitive Functioning, and Quality of Life in People With Multiple Sclerosis,” Neurology(R) Neuroimmunology and Neuroinflammation 12 (2024): e200334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Owji M., Ashraf‐Ganjouei A., Sahrian M. A., et al., “The Relationship Between Cognitive Function and Body Mass Index in Multiple Sclerosis Patients,” Multiple Sclerosis and Related Disorders 32 (2019): 37–40. [DOI] [PubMed] [Google Scholar]
- 7. Huppke B., Ellenberger D., Hummel H., et al., “Association of Obesity With Multiple Sclerosis Risk and Response to First‐Line Disease Modifying Drugs in Children,” JAMA Neurology 76 (2019): 1157–1165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Kvistad S. S., Myhr K. M., Holmoy T., et al., “Body Mass Index Influence Interferon‐Beta Treatment Response in Multiple Sclerosis,” Journal of Neuroimmunology 288 (2015): 92–97. [DOI] [PubMed] [Google Scholar]
- 9. Petersen E. R., Oturai A. B., Koch‐Henriksen N. K., et al., “Smoking Affects the Interferon Beta Treatment Response in Multiple Sclerosis,” Neurology 90 (2018): e593–e600. [DOI] [PubMed] [Google Scholar]
- 10. Petersen E. R., Sondergaard H. B., Laursen J. H., et al., “Smoking Is Associated With Increased Disease Activity During Natalizumab Treatment in Multiple Sclerosis,” Multiple Sclerosis 25 (2019): 1298–1305. [DOI] [PubMed] [Google Scholar]
- 11. Akbartabartoori M., Lean M. E., and Hankey C. R., “Smoking Combined With Overweight or Obesity Markedly Elevates Cardiovascular Factors,” European Journal of Cardiovascular Prevention and Rehabilitation 13 (2006): 938–946. [DOI] [PubMed] [Google Scholar]
- 12. Nishiyama M., Kimijima M., Muto T., and Kimuera K., “Presence of an Interaction Between Smoking and Being Overweight Increases Risk of Hypertension, Diabetes, and Cardiovascular Disease in Outpatients With Mood Disorders,” Environmental Health and Preventive Medicine 17 (2012): 285–291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Polman C. H., Reingold S. C., Edan G., et al., “Diagnostic Criteria for Multiple Sclerosis: 2005 Revisions to the “McDonald Criteria”,” Annals of Neurology 58 (2005): 840–846. [DOI] [PubMed] [Google Scholar]
- 14. Polman C. H., Reingold S. C., Banwell B., et al., “Diagnostic Criteria for Multiple Sclerosis: 2010 Revisions to the McDonald Criteria,” Annals of Neurology 69 (2011): 292–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Hedström A. K., Hillert J., Olsson T., and Alfredsson L., “Smoking and Multiple Sclerosis Susceptibility,” European Journal of Epidemiology 28 (2013): 867–874. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Hobart J., “The Multiple Sclerosis Impact Scale (MSIS‐29): A New Patient‐Based Outcome Measure,” Brain 124, no. 5 (2001): 962–973. [DOI] [PubMed] [Google Scholar]
- 17. Phillips G. A., Wyrwich K. W., Guo S., et al., “Responder Definition of the Multiple Sclerosis Impact Scale Physical Impact Subscale for Patients With Physical Worsening,” Multiple Sclerosis Journal 20, no. 13 (2014): 1753–1760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Strober L., DeLuca J., and Benedict R. H. B., “Symbol Digit Modalities Test: A Valid Clinical Trial Endpoint for Measuring Cognition in Multiple Sclerosis,” Multiple Sclerosis 25 (2019): 1781–1790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Weinstock Z., Morrow S., Conway D., et al., “Interpreting Change on the Symbol Digit Modalities Test in People With Relapsing Multiple Sclerosis Using a Reliable Change Methodology,” Multiple Sclerosis 28 (2022): 1101–1111. [DOI] [PubMed] [Google Scholar]
- 20. Qiu F., Liang C. L., Liu H., et al., “Impacts of Cigarette Smoking on Immune Responsiveness: Up and Down or Upside Down?,” Oncotarget 8 (2017): 268–284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Gregor M. F. and Hotamisligil G. S., “Inflammatory Mechanisms in Obesity,” Annual Review of Immunology 29 (2011): 415–445. [DOI] [PubMed] [Google Scholar]
- 22. Miller A. A. and Spencer S. J., “Obesity and Neuroinflammation: A Pathway to Cognitive Impairment,” Brain, Behavior, and Immunity 42 (2014): 10–21. [DOI] [PubMed] [Google Scholar]
- 23. Hahad O., Daiber A., Michal M., et al., “Smoking and Neuropsychiatric Disease—Associations and Underlying Mechanisms,” International Journal of Molecular Sciences 22 (2021): 7272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Shahi S. K., Ghimire S., Lehman P., and Mangalam A. K., “Obesity Induced Gut Dysbiosis Contributes to Disease Severity in an Animal Model of Multiple Sclerosis,” Frontiers in Immunology 13 (2022): 966417. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Shapiro H., Goldenberg K., Ratiner K., and Elinav E., “Smoking‐Induced Microbial Dysbiosis in Health and Disease,” Clinical Science 136 (2022): 1371–1387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Xiao Y., Liu D., Cline M. A., and Gilbert E. R., “Chronic Stress, Epigenetics, and Adipose Tissue Metabolism in the Obese State,” Nutrition & Metabolism 17 (2020): 88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Joehanes R., Just A. C., Marioni R. E., et al., “Epigenetic Signatures of Cigarette Smoking,” Circulation. Cardiovascular Genetics 9 (2016): 436–447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Marrie R. A. and Horwitz R. I., “Emerging Effects of Comorbidities on Multiple Sclerosis,” Lancet Neurology 9 (2010): 820–828. [DOI] [PubMed] [Google Scholar]
- 29. Hedström A. K., Stenberg E., Spelma T., et al., “The Impact of Bariatric Surgery on Disease Activity and Progression of Multiple Sclerosis: A Nationwide Matched Cohort Study,” Multiple Sclerosis 28, no. 13 (2022): 13524585221107095, 10.1177/13524585221107095. [DOI] [PubMed] [Google Scholar]
- 30. “Obesity Negatively Affects Disease Progression, Cognitive Functioning, and Quality of Life in People With Multiple Sclerosis,” (2024). [DOI] [PMC free article] [PubMed]
- 31. Rodgers J., Friede T., and Vonberg F. W., “The Impact of Smoking Cessation on Multiple Sclerosis Disease Progression,” Brain: A Journal of Neurology 145 (2022): 1368–1378. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Wu J., Olsson T., Hillert T., et al., “Influence of Oral Tobacco Versus Smoking on Multiple Sclerosis Disease Activity and Progression,” Journal of Neurology, Neurosurgery, and Psychiatry 94 (2023): 589–596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Hodge J. M., Shah R., McCullough M. L., et al., “Validation of Self‐Reported Height and Weight in a Large, Nationwide Cohort of US Adults,” PLoS One 15 (2020): e0231229. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. JL C., LA D. R., SR J., and DP S., “Accuracy and Reliability of Self‐Reported Weight and Height in the Sister Study,” Public Health Nutrition 15 (2012): 989–999. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Tables S1–S2.
Data Availability Statement
Anonymized data underlying this article will be shared on reasonable request from any qualified investigator who wants to analyze questions that are related to the published article.