The association of different acute manifestations of multiple sclerosis on functional outcome

Introduction

Multiple sclerosis (MS) is an autoimmune demyelinating disease of the central nervous system (CNS), typically presenting in young adults (20–50 years old). MS affects around two to three times more women than men. There are 2.8 million people in the world living with MS, with a higher prevalence in North America and Europe. Clinical manifestations are heterogeneous since demyelination can happen in any part of the CNS (1). It may cause physical and cognitive impairment during the disease's development, such as fatigue and loss of self-efficacy. Therefore, MS provokes significant disability and harmfully affects patients’ functional independence.

The current literature lacks studies regarding clinical factors associated with patients’ personal experiences with the disease. Therefore, health professionals could direct care to minimize the impact of MS symptomology on the patient's life and improve functionality. Data regarding the effect of specific symptoms on functionality and outcomes are often controversial, for example regarding the effect of paralysis and hypoesthesia in functionality (2–4). Thus, this study aims to clarify the association between these clinical features and patients’ functional outcomes, evaluated through the Extended Disability Status Scale (EDSS).

Results

A total of 195 patients were diagnosed with relapsing-remitting (173 patients, 88.7%) and primary progressive MS (19 patients, 9.7%) and included in the final analysis. Three (1.6%) patients had the clinical course of MS that was still under investigation at admission to the study. A total of 140 patients were women (71.8%), and the average age was 39.89 ± 11.57 years. Their median EDSS at admission was 2 points (IQR 1–4) (Table 1). General information about the symptoms of the first acute manifestation of the disease is also described in Table 1.

An univariate analysis was performed using the symptoms of the first acute manifestation of the disease, as well as demographic information such as sex and age. From these, five variables showed statistically significant correlation with worse EDSS at admission, were they: acute blurry vision (U = 3204.5; p < 0.001; r = −0.251), acute paralysis (U = 1326.5; p = 0.021; r = −0.165), acute hypoesthesia (U = 3444.5; p = 0.005; r = −0.204), autonomic syndrome (U = 2696.5; p < 0.001; r = −0.325), and Lhermitte sign (U = 2617.5; p = 0.017; r = −0.181). All variables showed a small effect size, excluding the autonomic syndrome, which showed a moderate effect size.

The linear regression model included the five variables that showed a correlation with the EDSS. The final model was statistically significant (R² = 0.117; p < 0.001), although with a low R², indicating poor model fit to data. Two variables were shown to be independent predictors of worse EDSS at admission: acute blurry vision (Beta = 0.183; p = 0.010; IC95% = 0.219–1.563; OR = 1.201) and autonomic syndrome (Beta = 0.219; p = 0.003; IC95% = 0.371–1.806; OR = 1.245). Therefore, patients with acute blurry vision or autonomic syndrome at presentation have, respectively, 20% and 24.5% more chance of developing worse functional status. The test statistics for each variable and the model are described in Table 2.

Discussion

This study highlights how some symptoms at diagnosis are associated with worse prognosis based on EDSS, such as acute blurry vision, acute paralysis, acute hypoesthesia, autonomic syndrome, and Lhermitte sign, although with a small effect size, except for autonomic syndrome, which showed a moderate effect size. In multivariate analysis, acute blurry vision and autonomic syndrome were shown to be independent predictors of worse prognosis.

As expected, all symptoms at first presentation associated with higher EDSS are evaluated by the scale, except for the Lhermitte sign. Lhermitte's sign typically presents in cervical spine lesions or low brainstem lesions (5). This could help to explain why this sign might be associated with worse outcomes while other brainstem signs (such as cranial nerve impairments) are not. A previous study tried to assess the correlation between Lhermitte's sign and prognostic factors but bared unfruitful results (6). However, the main difference between the presenting study and the previous is the moment when LS was present, which in the latter was in any moment of disease.

Acute blurry vision at presentation was considered to be an independent predictor of worse EDSS in our study. Besides visual acuity being evaluated in EDSS, blurry vision is a symptom directly related to possible affection in pathways, such as the optic nerve. Some studies have associated alterations in the retina and optic nerve, and even acuity, with worse functionally measured by EDSS (7, 8). This can be explained by the association of these features with central inflammation, which plays a significant role in the pathophysiology of MS (9). Furthermore, a study showed that increasing annual rates of atrophy of the inner retinal layers are associated with worsening ambulation, significantly increasing the EDSS score (10).

Our study defines autonomic dysfunction as urinary or fecal incontinence/retention. These symptoms are also correlated with spinal symptoms, which can also be associated with gait disturbances, pyramidal, and sensitive symptoms thus resulting in higher EDSS (11–14). Another important feature while interpreting this finding is that EDSS itself measures sphincteric dysfunction since it is a great cause of functional dependence, especially both fecal and urinary incontinence (15). Moreover, urodynamic dysfunction signs are associated with higher EDSS in previous studies (16–19). Alterations in the anal sphincter were not associated with differences in EDDS in previous studies (2). However, some older studies suggest an association of sphincter involvement with unfavorable prognosis (3).

The effects of acute blurry vision and autonomic dysfunction are not to be taken lightly. They showed 20% and 24.5% increased odds of worse functional outcomes, respectively. For instance, many clinical variables such as sex, age at onset, and family history of MS have no significant effect on functional outcomes. Moreover, known predictors such as the number of relapses have similar or even lower effect sizes on long-term disability (4).

Although hypoesthesia and paralysis were associated with worse EDSS in the univariate analysis, their effects were not significant in the multivariate analysis. As for hypoesthesia, this may be related to a possible relationship between hypoesthesia and dynamic postural control (20), maybe leading patients to subjectively associate hypoesthesia and ataxia. As ataxia is evaluated in the EDSS, the effect of hypoesthesia in the univariate analysis may be exaggerated by a possible correlation with ataxia. In a previous study, sensory symptoms did not affect long-term disability outcomes (4).

As for acute paralysis, the evidence supporting this finding is controversial in the literature (3, 4). We see it as counterintuitive since motor function is evaluated through the EDSS, and severe motor impairment has higher punctuation on the scale (15). Therefore, we expected patients with acute paralysis to have worse functional outcomes. More studies are needed to confirm or discard this correlation and to help understand why it may have happened.

In this setting, the findings of our study may potentially help physicians in defining initial disease-modifying therapies (21). For example, severe motor or cerebellar involvement is considered a criterion of severity of onset and worse prognosis. Therefore, these clinical features are taken into account when deciding on the initial therapy or changing to high-efficacy therapies. In the future, acute blurry vision and sphincteric involvement may further integrate recommendations in stratifying the risk of MS worsening, and for this reason, impact major clinical decisions.

Our study has some limitations. First, symptoms at onset are susceptible to memory bias since the patient may recall more severe symptoms than others. This is particularly important because it may erroneously overestimate or underestimate the frequency of specific symptoms. Therefore, the associations found in the study may not correspond perfectly to the daily basis clinical practice reality. Second, we only assessed sphincteric symptoms to characterize an autonomic syndrome. However, other symptoms related to dysautonomia are also described in MS. Other essential elements, such as the number of white matter lesions and locations, are not collected in this study. Also, the scale directly measures most symptoms that showed significant association with EDSS. For that reason, other functional status scales could enhance this analysis.

Despite these limitations, this study contributes significantly to investigations and understanding of MS progression. The only first acute manifestations of MS irrespectively correlated to worse functional outcomes were acute blurry vision and autonomic syndrome. Although some other symptoms were significantly associated with higher EDSS, such as acute paralysis, acute hypoesthesia, and Lhermitte sign, they were shown not to be independent predictors of worse functional outcomes, probably because of inevitable interactions with factors that had not been taken into account in the univariate analysis. For that matter, these results may help better understand the relationship between MS symptomatology and functionality and, in specific settings, may help the physician establish the patient's prognosis.

However, further studies, preferentially prospective cohorts from the time of diagnosis, are needed to help establish other potential predictors of worse functional outcomes. Also, following patients since the diagnosis and/or the first acute manifestation of the disease may mitigate the memory bias and potentially allow for objective neurological examination in this context. Furthermore, other models that better account for confounding variables, such as disease progression indicators (number of relapses, neuroimaging features), may provide a more reliable association. Interventional studies would also help explore possible interventions to mitigate the effect of these independent predictors of functionality.

Methods

Author Disclosures

All contributors have confirmed that no conflict of interest exits.

Author Contributions

Manuscript: JPFG, GSMN, MYSL, LDMA, FABM

Idea: ACSF

Statistics: JPFG, CSTA, GSMN

>Revision: CSTA, PAPJ