Prevalence of comorbidities in Multiple Sclerosis and impact on physical disability according to disease phenotypes

Background: Comorbidities are prevalent among Multiple Sclerosis (MS) patients. Few studies have characterized their prevalence and impact in Latin American populations.
Objective: We aim to assess the prevalence of comorbidities and their impact on the risk of physical disability across different MS phenotypes.
Methods: Cross-sectional multicenter study of patients under regular clinical care at the Programa de Esclerosis Múltiple UC and Hospital Dr. Sótero del Río in Chile. Prevalence of comorbidities was estimated from the retrospective assessment of electronic medical charts. Disease phenotypes were categorized into two groups: clinically isolated syndrome/relapsing-remitting (inflammatory group) and primary/secondary progressive MS patients (progressive group). A multivariable analysis using binary logistic regression for assessing the risk of EDSS ≥ 6.0 in each group was performed.
Results: A total of 453 patients was included, 71% female, mean age at onset 31 years, mean disease duration 10 years, and median EDSS 2.0 (range 0–10). In the whole sample, most prevalent comorbidities were ever-smoking (42.2%), depression/anxiety (34.9%), thyroid disease (15.7%), hypertension (11.3%) and insulin resistance/ type 2 diabetes mellitus (11.0%). When assessing the risk of EDSS ≥ 6, in the inflammatory group (N = 366), age at onset (OR 1.06, 95%CI(1.02–1.11), p = 0.008), disease duration (OR 1.06, 95%CI(1.00–1.12), p = 0.039) and epilepsy comorbidity (OR 5.36, 95%CI(1.33–21.5), p = 0.018) were associated with a higher risk of dis- ability. In the progressive group (N = 87), disease duration was a risk factor (OR 1.08 95%CI(1.02–1.16), p = 0.014), while shorter diagnostic delay (OR 0.91 95%CI(0.85–0.99), p = 0.025) and insulin resistance/type 2 diabetes mellitus comorbidity were protective factors (OR 0.18 95%CI(0.04–0.83), p = 0.028), 72% of these patients were receiving metformin.
Conclusions: Comorbidities are common across different MS disease phenotypes. Epilepsy seems particularly related with a higher risk of physical disability in relapsing-remitting patients, while the role of insulin re- sistance/type 2 diabetes mellitus or the impact of metformin use as a protective factor should be further studied. Prospective and larger studies are still needed in order to assess the real impact of comorbidities and their management in MS outcomes.

1. Introduction
Comorbidities have been commonly described among Multiple Sclerosis (MS) patients. Higher prevalence compared to healthy con- trols include affective disorders, cardiovascular disease, systemic au- toimmune disorders and epilepsy, among others (Marrie et al., 2015; Hauer et al., 2020]. It is suggested that comorbidities may affect MS disease course, but this interaction is complex, including genetic pre- disposition, environmental exposure, systemic inflammation, and re- lationship with disease-modifying therapies (Marrie, 2019). Studies evaluating the impact of comorbidities, with an emphasis on modifiable factors that may affect MS-related outcomes, such as physical disability, cognitive impairment, brain atrophy and quality of life, and differences among different populations across the globe are still needed (Hauer et al., 2020) .
In the present study, we aimed to describe the clinical phenotypes and prevalence of comorbidities of patients under regular clinical care in two reference centers in Chile, and to assess the impact of co- morbidities on disability outcomes, according to different MS pheno- types.

2. Material and methods
2.1. Patients
A multicenter study including patients diagnosed with Multiple Sclerosis according to McDonald 2017 criteria (Thompson et al., 2018) (relapsing- remitting RR, secondary-progressive SP, primary-progressive PP) or Clini- cally Isolated Syndrome (CIS), under regular clinical care at the MS Units of the Universidad Católica de Chile and Hospital Dr. Sótero del Río.
Demographic and MS-related variables were gathered from a pro- spective database collected from January 2008 to January 2019 in- cluding Expanded Disability Status Scale (EDSS) assessed at least 3 months after a relapse.
Relevant comorbidities were assessed retrospectively from elec- tronic medical charts. Patients from both reference centers receive neurological care by the same treating physician, who records every new comorbidity and/or new medication of every patient. Also, most patients also receive their general medical care in the same health system, being able to review medical charts from other specialists, re- ducing ascertainment bias. All patients signed an informed consent approved by the local ethics committee.

2.2. Statistical analysis
Two groups were created and analyzed separately: the in- flammatory group (CIS/RR), and the progressive group (PP/SP). A univariate analysis was performed using Fisher’s exact test, Chi- squared, Mann-Whitney and Student T-test for assessing exploratory differences between both groups. For the assessment of the impact of comorbidities on disability, a multivariable binary logistic regression for EDSS ≥ 6 as the dependent variable included all variables with p<0.2 was performed. Statistical significance was set at p<0.05 and analyses were conducted using Statistical Package for the Social Sciences (SPSS, IBM, Chicago, Illinois) v21. 3. Results 3.1. Characteristics of included patients A total of 453 patients were included, 366 in the inflammatory group and 87 in the progressive group. In the whole cohort, most prevalent comorbidities observed were ever-smoking (42.2%), depres- sion/anxiety (34.9%), thyroid disease (15.7%), hypertension (11.3%), insulin resistance/type 2 diabetes mellitus (11%), migraine (7.9%), dyslipidemia (7.3%), epilepsy (4.6%) and bipolar disorder (4.4%). Patients with progressive MS had a higher prevalence of hypertension (20.7% progressive MS vs. 9% inflammatory MS, p = 0.004). Characteristics of included patients, comorbidities and differences between disease phenotypes are summarized in Table 1. 3.2. Impact of comorbidities on EDSS according to MS phenotype In the inflammatory group, 20/366 patients (6%) had EDSS≥6.0 and in the progressive group, 57/87 (66%) had EDSS≥6.0.In the in- flammatory group, multivariable analysis showed that older age at onset (OR 1.06 95%CI(1.02–1.11), p = 0.008), longer disease duration (OR 1.06 95%CI(1.00–1.12), p = 0.039) and comorbidity with epilepsy (OR 5.36 95%CI(1.33–21.5), p = 0.018) were risk factors for EDSS≥6. No protective factors were found in this group. In the progressive group (N = 87), multivariable analysis showed that longer disease duration was a risk factor for EDSS ≥ 6.0 (OR 1.08 95%CI(1.02–1.16), p = 0.014), while shorter diagnostic delay (OR 0.91 95%CI(0.85–0.99), p = 0.025) and comorbidity with insulin re- sistance/type 2 diabetes mellitus were protective factors (OR 0.18 95%CI(0.04–0.83), p = 0.028), 72% of the progressive MS patients with insulin resistance/type 2 patients were receiving metformin. Summary of the variables included in the binary logistic regression is shown in Table 2. 4. Discussion In this multicenter cross-sectional study, we found that clinical variables, such as older age at onset for the inflammatory group and longer disease duration in both groups, were non-modifiable in- dependent risk factors for disability accrual, whereas shorter diagnostic delay was a protective factor for the progressive MS group. Comorbidity with epilepsy was an independent risk factor for EDSS ≥ 6 in the in- flammatory group, while insulin resistance/type 2 diabetes, appeared to be a protective factor among progressive patients. Epilepsy has been previously reported as a risk factor for poor outcomes in MS. These include worse cognitive performance, higher rates of global and regional brain atrophy (Uribe-San-Martin, 2014, Uribe-San-Martin, 2018), and even a 2 fold increase in all-cause mor- tality in patients with epilepsy and MS (Chou et al., 2019). We report a prevalence of 4.6% (same in both phenotypes), which is in line with a meta-analysis estimate of 3.09 95%CI(2.01–4.16) (Marrie et al., 2015). The slightly higher prevalence reported in the present study may be explained by the fact that both of the participating hospitals are epi- lepsy reference centers. Our finding that comorbidity with epilepsy was an independent risk factor for EDSS ≥ 6.0 in the inflammatory group should be emphasized. This is consistent with a meta-analysis including 2845 MS patients, showing that epilepsy patients had higher EDSS scores after similar disease duration (Gasparini et al., 2017), and a retrospective study including 3166 patients with incident MS from Canada, showing that comorbidity with epilepsy was associated with higher EDSS scores (Zhang et al., 2018). This association seems far from anecdotal and a possible causal relationship is suggested (Schorner, 2019). Taking into account our previous cross-sectional and longitudinal studies in patients with epilepsy and MS, poor epilepsy control or the development of refractory epilepsy seems to be related to worse cognitive and brain volume outcomes compared to patients with good epilepsy control (Uribe-San-Martin, 2014; Uribe-San- Martin, 2018). The latter highlights the relevance of identification and aggressive management of relevant comorbidities in MS. Larger pro- spective studies assessing the impact of epilepsy management in the course of MS are still needed. The case of insulin resistance/type 2 diabetes mellitus as a protec- tive factor in the progressive group deserves further discussion. One explanation for this finding could be that treatment of insulin re- sistance/type 2 diabetes mellitus, for example, with metformin or pio- glitazone, has reported having a beneficial impact on T2/gadolinium enhancing lesions, as well as an increase in the number of regulatory T cells in patients with MS (), and may have had a positive impact on disability in this selected group of patients. This finding is particularly interesting in the light of a recent study in adult rodent oligodendrocyte progenitor cells (OPC), showing that aged OPCs become unresponsive to pro-differentiation signals, and that fasting or treatment with metformin can reverse these changes and restore the regenerative ca- pacity of aged OPCs, improving remyelination in aged animals fol- lowing focal demyelination (Neumann et al., 2019). Further studies are still needed to explore the promising role of metformin as a potential remyelinating therapy.
Overall, the prevalence of comorbidities in MS patients found in the present study is similar to those previously reported in larger cohorts and systematic reviews (Marrie et al., 2015; Hauer et al., 2020). Taking into account the Chilean National Health Survey (ENS, 2018), higher prevalences of depression/anxiety (34.9% vs. 6.2%), bipolar disorder (4.4% vs. 2.2%) and epilepsy (4.6 vs. 1.7%) can be observed among patients with MS compared with the general population, which high- lights the relevance for increased surveillance of this disorders during routine clinical care.
In Chile, patients receive health cared in one of two health systems, public or private. Every patient who is suspected to have Multiple Sclerosis is mandatory notified through a centralized law (Ley de Garantías Explícitas en Salud – GES – “Explicit Guarantees in Health”), in order to guarantee universal access to diagnosis and treatment, in- dependent of their health insurance. When notified, each patient is referred to a specialized health provider center for MS in a public or private system. No differences were observed comparing Hospital Dr. Sótero del Río (public health) vs. Universidad Católica de Chile (private health), so every MS case diagnosed in these two settings, were in- cluded in the study (independent from disease severity), making our sample highly representative of our national reality.
Limitations of the present study are mainly because of the retro- spective assessment of electronic medical charts, the selection of only one disability outcome measure (EDSS) and not including magnetic resonance imaging or patient-reported outcomes, such as quality of life or impact of the disease scales.
In conclusion, comorbidities are common across different MS dis- ease phenotypes and have an impact on physical disability in this population. Prospective and larger studies are still needed in order to assess the real impact of comorbidities and their management in MS outcomes.