Abstract
Objectives: To establish the prognostic role of clinical and demographic factors in a hospital-based cohort of MS patients categorized by age at clinical onset and clinical course.
Methods: Eighty-three patients with MS had a clinical onset of the disease in childhood (age <16 years; early-onset MS [EOMS]) and 710 in adult age (between 16 and 65 years; adult-onset MS [AOMS]). Patients were followed for a mean period of observation of 5 years. Univariate and multivariate analyses of clinical and demographic predictors for rapid progression and disability were performed using a stepwise Cox regression model with time-dependent covariates.
Results: In EOMS, the Expanded Disability Status Scale (EDSS) evaluated at last clinical examination was lower than in AOMS, despite a longer disease duration. The probability to reach growth disability and progression was significantly lower in EOMS than in AOMS. Median times to reach EDSS score of 4 and secondary progression were longer in EOMS than in AOMS, but the age at both endpoints was significantly lower in EOMS. In EOMS and AOMS, an irreversible disability was related to a secondary progressive course, a sphincteric system involvement at onset, and an older age at onset (in EOMS only for the group >14 years); in AOMS, other unfavorable factors were a pyramidal involvement at onset and a high relapse frequency in the first 2 years. The risk of entering secondary progression was significantly influenced by a high number of relapses in EOMS and by a higher age at onset and a short interattack interval in AOMS.
Conclusion: A slower rate of progression of disease characterized EOMS patients, suggesting more plasticity to recover in developing CNS, but the early clinical manifestation cannot be considered a positive prognostic factor.
Retrospective analyses on MS populations have shown that the clinical onset of the disease is often insidious and its occurrence before the age of 10 years and after 50 years is considered rare. The employment of current diagnostic criteria1,2⇓ along with advances in biochemistry, immunology, radiology, and molecular genetics are proving to be useful to define MS diagnosis in childhood and elderly age3 as well as to differentiate several diseases mimicking MS. In particular, the distinction between the first clinical attack of MS and acute disseminated encephalomyelitis (ADEM) is often difficult.4,5⇓ Most studies described single cases or small series of patients with early-onset MS (EOMS)6-10⇓⇓⇓⇓; the youngest of them was a 10-month-old girl.11 A very early onset of the disease has been described in children under 6 years of age.12 More complete information on MS in childhood derives from collaborative studies, which reported that EOMS occurs in 2.7 to 5% of all MS cases.13-16⇓⇓⇓ Similarly, there are very few studies on late MS clinical onset, which showed a highly variable frequency, ranging from 0.6 to 9.4%.3,17,18⇓⇓
Many clinical studies analyzing the prognostic factors involved in predicting the outcome of MS were carried out over the last years. In spite of some discordance, demographic and clinical variables such as age at clinical evidence of the disease, degree of first remission, interval between the first two relapses, and functional system involvement at onset were considered as reliable indicators of the disease course in different studies.19-26⇓⇓⇓⇓⇓⇓⇓ An older age at clinical onset is frequently associated with a poor prognosis; by contrast, the implication of EOMS with regard to long-term outcome is not entirely clarified. The prognostic factors in EOMS have been evaluated in few studies, performed with various methodologic approaches,14,16,27⇓⇓ and the discussion about the existence of clinical courses different from that of adult-onset MS (AOMS) is still open. In a recent prospective study performed in 78 patients with EOMS, a worse prognosis was predicted by high clinical disability at onset.28
Based on these assumptions, we analyzed a large hospital-based historical cohort of MS patients differentiated by age at clinical onset and course to assess the long-term prognosis of selected clinical and demographic features.
Patients and methods.
Patients.
Seven hundred ninety-three MS patients, followed at the MS Center of the Department of Neurologic and Psychiatric Sciences of the University of Bari, Italy, were retrospectively reviewed and included in the analysis. All data were collected in a standardized European database for MS (EDMUS).29 With use of the widely applied cutoff of age at clinical onset of the disease,13-16,27⇓⇓⇓⇓ 83 patients had the clinical onset of the disease in early age (EOMS; <16 years) and 710 patients in adult age (AOMS; between 16 and 65 years), and 63 of them had their first symptoms in elderly age (after 45 years). Based on medical history and clinical–paraclinical evaluations, cases of early and adult onset of clinical symptoms were classified as definite MS (360 clinically definite and 433 laboratory-supported definite MS).1 Patients were observed over a mean follow-up time of 5 years. General demographic and clinical features were evaluated in all patients (table 1). Clinical course was classified according to standardized definitions.30 Residual neurologic disability after the first clinical attack was not considered in the analysis because it was not available in all MS patients. The time to reach an irreversible clinical disability corresponding to a degree of Kurtzke’s Expanded Disability Status Scale (EDSS) score 4 (limited walking ability but able to walk without aid or rest for >500 m)31 and the time to start a secondary progression were considered as endpoints. We focused on the disability score of 4 because only six EOMS patients reached an irreversible EDSS score of 6.
Table 1 Clinical and demographic features in MS population
Diagnostic MRI was performed in 90% of MS patients at the time of the first clinical attack or during the clinical course, and serial MRI were carried out in several patients to evaluate the disease activity. MR findings were strongly suggestive of demyelinating disease; however, in this retrospective study, we did not analyze the MRI data because of the lack of standardization of MRI procedures.
Family history and associated diseases were investigated in all MS patients. Twenty-eight percent of AOMS and 5% of EOMS patients were under disease-modifying treatments; the most used were interferon-β (148 with relapsing–remitting or secondary progressive AOMS and 2 with relapsing–remitting EOMS) and azathioprine (35 with secondary progressive or primary progressive AOMS); 2 EOMS patients were treated with IV immunoglobulins. The relatively low percentage of treated patients, heterogeneity of therapy, period of treatment, and clinical course of MS patients did not allow us to analyze treated patients separately.
Statistical analysis.
For statistical analysis, MS patients were categorized into groups based on 1) age at clinical onset (EOMS and AOMS groups, as described above) and 2) clinical course defined as “not primary progressive” (including relapsing-remitting and secondary progressive patients) and “primary progressive.”
Comparison between clinical and demographic variables in MS groups was made by 2 × κ contingency tables χ2 tests for categorical variables and Kruskall–Wallis χ2 tests for continuous ones. Considering the large difference between primary progressive and other MS clinical courses, we analyzed this group separately.
With use of time to reach irreversible EDSS 4 and time to secondary progression as survival endpoints, separate Kaplan–Meier survival curves were presented for both endpoints; differences in survival times were assessed with a log-rank test. Time-dependent Cox proportional hazards regression models were used to assess possible predictors of survival.32 Cox models with time-dependent covariates allowed us to take into account, along with covariates at baseline, also the effect on survival of covariates changing over time (such as the clinical course, first interattack interval, and number of relapses in the first 2 years). Univariate and stepwise multivariate Cox models were tested for the two survival endpoints. In the tables⇓⇓, the results were expressed in terms of hazard ratios and Wald χ2 statistics p values. All the analyses were performed using SAS/STAT (version 6.12) and some routines written in SAS Macro Language (SAS software release 6.12; Cary, NC, 1989 to 1996).
Results.
Demographic and clinical details.
The distribution by age at clinical onset in 793 MS patients showed a peak between 21 and 30 years, including in this range 367 patients (46.3%), whereas the frequency of onset in early age was 10.5%. In the EOMS group, 13 patients showed a clinical onset at <10 years (median age 8.25 years, range 1.3 to 9.88 years), 26 patients between 10 and 14 years (median age 12.78 years, range 10.24 to 13.87 years), and 44 patients at >14 years (median age 15.04 years, range 14.09 to 15.95 years). According to clinical course, 679 cases were identified as “not primary progressive” (including all of the EOMS group and 596 AOMS patients) and 114 cases as “primary progressive” (including only AOMS patients, 33 of them with onset of the disease in elderly age).
The female/male ratio did not differ in the MS groups; however, EOMS patients showed an increase of woman predominance (ratio 2.6:1) in the pubertal period (between 10 and 14 years), confirmed in the group with a clinical onset at >14 years of age (ratio 2.4:1), in contrast to a significant male predominance in the group <10 years (ratio 0.6:1). In EOMS, the relapsing–remitting course was significantly more frequent, whereas no cases was primary progressive. Cerebellar and brainstem systems at onset were involved with significantly high frequency in EOMS. The EDSS evaluated at last clinical examination was significantly lower in EOMS than in the other MS groups, despite a significant longer disease duration (see table 1).
Five patients with EOMS had a first-degree relative with MS (three sib pairs and two parent–child pairs); the frequency (6%) did not differ from that in the adult group (3.1%).
Prognostic factors.
When the time to reach irreversible disability score, corresponding to EDSS 4, was considered as the endpoint, Kaplan–Meier survival curves stratified by clinical course and age at onset showed that a higher percentage of EOMS patients (75%) did not reach this disability stage at follow-up times than other MS groups (57% of AOMS “not primary progressive” and 5% of “primary progressive” course) (log-rank test, p < 0.0001) (figure 1). The estimated median time to reach EDSS 4 was significantly longer in the EOMS than the AOMS groups (see table 1), but the age at EDSS 4 was lower in EOMS patients (median 31.6 years, range 18.15 to 58.28 years) than in AOMS “not primary progressive” (median 41.16 years, range 21.78 to 69.22 years) and “primary progressive” (median 48.18 years, range 19.96 to 71.55 years) (p < 0.0001).
Figure 1. Kaplan–Meier survival curve for time to reach Expanded Disability Status Scale (EDSS) 4. The proportion of patients who reached EDSS 4 in each group is reported in parentheses. Log-rank test, χ2 = 109.51, p < 0.0001. No-PP = not primary progressive; PPMS = primary progressive MS.
The probability to reach a secondary progression in a short time, evaluated only in the group of “not primary progressive” patients, was lower in EOMS patients (14.45% of events) than in AOMS patients (24% of events) (log-rank test, p < 0.0003) (figure 2), with an estimated median time of developing the secondary progressive phase significantly longer in EOMS (see table 1). The median age at secondary progression was lower in EOMS patients (30.7 years, range 17.6 to 52.6 years) than in AOMS patients (37.55 years, range 22.7 to 60.23) (p < 0.002).
Figure 2. Kaplan–Meier survival curve for time to reach progression. The proportion of patients who reached secondary progression in each group is reported in parentheses. Log-rank test, χ2 = 12.84, p < 0.0003.
In EOMS patients, a secondary progressive course and a short interval between the first and second attack were significantly unfavorable factors associated univariately with a greater risk of developing over time high clinical disability; furthermore, borderline significance was found for sphincteric system involved at onset. Testing the effects of age at onset in EOMS subgroups, only the age >14 subgroup for time to reach EDSS 4 was significant. The further step of multivariate Cox confirmed the significant association between growth disability, secondary progressive course, and sphincteric system involvement. The analysis did not retain any age subgroup for the outcome (table 2)
Table 2 Univariate and multivariate survival analysis for time to reach EDSS 4 and time to reach progression in EOMS patients
The risk of developing a secondary progression in EOMS was univariately associated with a first interattack interval shorter than 1 year and a high number of relapses in the first 2 years of the disease. The adverse prognostic role of a high number of relapses in the first 2 years of the disease was confirmed by multivariate analysis. The age at onset did not influence this outcome (see table 2).
In the group of AOMS “not primary progressive” patients, the risk to reach irreversible disability in a short time was greatly influenced by several factors including a higher age at onset of >45 years, an involvement of pyramidal and sphincteric systems at onset, as well as a secondary progressive course and a high number of clinical relapses in the first 2 years of the disease. The significant association between irreversible growth disability and involvement of pyramidal and sphincteric system at onset was confirmed in AOMS “primary progressive” patients (table 3).
Table 3 Univariate and multivariate survival analysis for time to reach EDSS 4 and time to reach progression in AOMS patients
In AOMS patients, the critical factors associated with a secondary progression were a higher age at onset and a short interval between the first and second relapse (see table 3).
Discussion.
The clinical expression of MS before adulthood is rare and occurs predominantly after a subclinical phase, as indicated by MRI findings. Several studies showed that 50 to 80% of patients with clinically isolated syndromes suggestive of MS have multifocal white matter lesions on MRI of the brain, suggesting that the “true” onset of the disease may precede the appearance of clinical symptoms.33 Nevertheless, an early predisposition with clinical evidence during adolescence is demonstrated. In the current study, we found a higher frequency of EOMS (10.5%) than other authors, who reported percentages from 2.7 to 5%.13,15,16⇓⇓ We think that this increase reflects a recruiting bias in a neurologic center highly specialized in MS diagnosis and treatment. Conversely, the absence of prodromal illness as well as of seizures, bilateral optic neuritis, and menin-gism, the finding of multiphasic clinical course during the follow-up, and the progressive deterioration over time in these patients were strongly indicative of MS and excluded a possible diagnosis of ADEM. According to previous studies, the MS diagnosis can be adequately defined only by observing the subsequent natural history of the disease; in our study, the observation was prolonged for a mean period of 5 years.4,5⇓
An increased susceptibility of women was confirmed in EOMS, without significant difference in ratio with AOMS.14-16⇓⇓ Nevertheless, a prevalence of women was observed in pubertal age (between 10 and 14 years), even if lower than the female overrepresentation found in our recent collaborative study.28 Several authors reported a higher frequency of women in EOMS and in late-onset MS, emphasizing the role of hormonal factors in increasing female MS susceptibility during puberty and menopause.34,35⇓
Consistent with previous studies, in EOMS, the clinical course was predominantly relapsing-remitting,13,15,16⇓⇓ whereas no patient had the primary progressive course,27,28⇓ which was rare in other reports.12-15⇓⇓⇓ In AOMS, the frequency of the primary progressive and secondary progressive course was similar to the range in other studies,19,20,24,25⇓⇓⇓ and our data confirm that the primary progressive form occurs more frequently in later ages.21,24,25⇓⇓
In agreement with our previous collaborative studies,16,28⇓ symptoms suggesting an involvement of brainstem and cerebellar systems at onset were preponderant in EOMS patients, whereas the frequency of motor dysfunction was low13-16,28⇓⇓⇓⇓ and increased with an increase of age at clinical onset, in particular, in “primary progressive” AOMS.3,18,24,25,36-38⇓⇓⇓⇓⇓⇓
In EOMS, a secondary progressive course and the involvement of sphincteric system at onset were critical elements that greatly influenced the risk of an irreversible disability score of EDSS 4. In disagreement with a recent collaborative study, which reported no association between age and disability, a higher age at onset in EOMS was univariately associated with a worse clinical disability.28 In turn, the risk of entering secondary progression was associated with high number of relapses in the first 2 years.
In patients with “not primary progressive” AOMS, a secondary progressive course was confirmed to be associated with growth disability; however, other critical factors were an older age at onset and a short interval between the first and second attack. In line with several authors, motor and sphincter symptoms strongly influenced the risk to reach irreversible disability both in adult “not primary” and in “primary progressive” patients.25,26,39,40⇓⇓⇓ Finally, the risk of entering a secondary progressive course in AOMS patients was significantly associated with an older age at onset and a short first interattack interval.
A poor prognosis was found in few patients with EOMS, associated with a fewer number of clinical factors than AOMS. The lack of primary progressive course and a lesser incidence of motor dysfunction at onset in EOMS patients could be a favorable outcome regarding long-term disability, whereas a negative prognostic factor was a higher age at onset (>14 years). EOMS patients reached a worse clinical disability and a progression phase after long-term disease. Nevertheless, these patients get sick at a young age, and although the clinical course could be favorable for several years, they reach a more severe disease at a younger age than AOMS patients. This result is consistent with a recent report that suggests that early onset of the disease is not a favorable predictor of the outcome in all patients.41
Acknowledgments
Supported by a grant from “MIUR–Piano Biomedicina, Progetto 1.”
- Received February 8, 2002.
- Accepted August 28, 2002.
References
Letters: Rapid online correspondence
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
