The demographics and clinical characteristics of the 400 patients (299 females, 101 males) registered in the data base are presented in Table 1 and Table 2. Overall, the sex ratio female: male was 2.9, the mean patients age was 39,86 (95%CI= (30.16 – 37.36)) years. The mean age at MS clinical onset was 31.2 (IC 95% = (30.27- 32.14)) years, and the mean duration of follow-up from multiple sclerosis clinical onset to the last visit was 33.76 (IC 95%= (30.16 – 37.36)) months. The initial course of the disease was relapsing-remitting in 362 patients (90.5%), whereas it was progressive in 38 (9.5%) patients. In the first group, 309 patients (77.2%) had residual deficit after the first relapse, and 197 patients (49.2%) had two relapses or more within the first 2 years after clinical onset (mean number of relapses during the first 2 years of MS was: 1.5 ± 0.8). The main results for relapsing-remitting and progressive initial courses of MS are compared in table 2. The two groups differ in several respects: the first group was characterized by a greater female preponderance (p=0.001), an earlier onset of MS (p< 0.0001). Review of the patient’s initial symptoms found that the primary progressive group experienced motor weakness, cerebellar and polysymptomati symptoms more frequently (95%- 47.4%- 50%) than those with relapsing-remitting MS (52%- 33.4% - 28%) (P= 0.001). OCB were analyzed in 396 patients. OCB in the CSF were found in 78% of patients. Initial cerebral MRI was studied in all patients and spinal MRI in 144 patients. Three to four Barkhof criteria were fulfilled in 93.5% of patients and spinal lesions were found in 84.7%.

During the follow-up of the 400 patients, a total of 144 (36%) and 83 (20%) patients reached EDSS scores 4 and 6 respectively. The median times from the onset of MS to assignment of a score 4 and 6 was 10 (95% confidence interval (CI) = 8.6- 11.4) years and 19 (95% confidence interval (CI)= 14.2- 23.7) years, respectively. The median interval from onset of disease to the assignment of each of these scores was significantly shorter in the group of patients with a progressive initial course of disease than among those who had relapsing–remitting disease at onset ( Figure 1. a, b, Table 2). It was 12 vs 4 years and 20 vs 9 years P<0, 0001, respectively. 57 patients with relapsing onset (15,7%) had converted to secondary progressive form. The mean time to the SP form was 22 years ((95% CI = (19.9 - 24.1)).

Considering EDSS and disease duration, we calculated 2 scores which are the progression index (EDSS to disease duration ratio) and the MSSS score (algorithm also considering EDSS and disease duration). The median progression index in our cohort was 0.33. The progression index was significantly shorter in patients having a progressive form than in patients with a relapsing–remitting onset (p = 0.001). The median MSSS score was 4.26. 42% of patients had rapid disease progression with a MSSS score > 5. The onset of the disease with pyramidal, cerebellar, and poly-symptomatic involvement was correlated with a MSSS score >5 (P<0.05).

The variables included in the multivariate Cox models to estimate the most predictive risk factor(s) for disability (EDSS 4 and EDSS 6) were: age of onset, sex, pyramidal involvement, cerebellar involvement, poly-symptomatic involvement, clinical form, number of relapses in the first two years, interval between the first and second years (P1-P2), and incomplete recovery or sequelae after the first relapse. Combinations of all variables were performed for at least two predictors. Variables included in the logistic regression were significant at P<0.05. This analysis showed that 04 factors are predictive of disability (Table 3): age of onset over 40 years, pyramidal onset, incomplete recovery after the first attack, progressive form at onset.

Using the MSSS severity score, and after univariate analysis we identified several risk factors predictive of disease severity. These included male sex (P= 0.002), age of onset ≥ 40 years (P= 0.001), pyramidal (P= 0.0001), cerebellar (P= 0.002), and poly-symptomatic (P=0.0001) onset of the disease, incomplete recovery after the first attack (P= 0.0001), a high number of attacks during the first 2 years (P= 0.04) and the progressive form (P=0.001), and all these factors were correlated with a high MSSS score. However, prognostic factors in the multivariate binomial logistic regression analysis were limited to pyramidal onset, incomplete recovery after the first attack, short interval between the first and second attack, and progressive onset.

A total of 330 (82.5%) received disease modifying drugs for at least 6 months. Treatment was started 25.1 ± 13.4 months after diagnosis of multiple sclerosis on average, and consisted of beta-interferon (63.5%), Natalizumab (5.7%), cyclophosphamid (10.5%) and azathioprine (2.7%). The mean duration of interferon treatment was 3.16 years (95 CI %= 2.90 - 3.41).

Our study has shown that Algerian patients have similar characteristics to European and North Africans patients: predominantly female, average age of onset 30 years, relapsing-remitting onset in almost 90% of our cohort 202122. However, some specific features were found: Isolated motor dysfunction was the most frequent onset symptom, exceeding visual and sensory signs, an initial poly-symptomatic involvement was relatively as frequent (30%), and recovery from the initial neurological episode was found in about one-third of the patients. These phenotypic characteristics underline the particular severity of North African MS. Indeed, originating from North Africa has been described in some publications as a factor of severity for MS 9101123. The course of MS was more severe compared to European patients such that the EDSS 4 and 6 scores were readily reached presaging a more severe disease outcome. Thus, the study of Sidhom 12 revealed that the disability progression profile was more rapid in North African patients than in European patients. It was shown after statistical study by the Cox method that ethnic origin was a risk factor for reaching EDSS 6. The severity of the disease was more common in patients born in France and whose parents were of North African origin, reflecting genetic and environmental influences on the progression of the disease. For disability measures, the overall MSSS distribution in our cohort is similar to European and US data 1924. In a Tunisian and European study 12, the MSSS was 4.8 and 5.1. The same result is observed for the progression index which is similar to our cohort, it was 0.51 and 0.41. In contrast, the Moroccan study found a higher MSSS score and progression index (6.39 and 0.97) 10. The median time to reach EDSS 4 was similar to that reported in the North African studies and was shorter than in the European studies 10112224. The median time to EDSS 6 (19 years (95% CI = (14.2 - 23.7)) was relatively short compared with European cohorts. It was 27.9 years in Canada 26, 24 years in the United States 27 and between 20 and 24 years in France 2823. The evolution of MS is therefore very variable, ranging from a severe disease that is rapidly disabling to a mild form that causes minimal long-term disability. Predicting this evolution at the beginning of the disease will help in the therapeutic management. Most of the studies evaluating the predictive factors of disability have been performed in Western populations and they are identical in all these populations 2729. However, these predictive factors of long-term disability have not been studied in Arab populations, with the exception of the Yamout study in Lebanon 17.

Late age at onset is considered predictive of early progression to disability in most studies 293031. The presence of pyramidal and cerebellar signs has been described in several studies as a fact associated with a poor long-term prognosis 312125. Incomplete recovery after the first attack is also predictive of poor long-term disability and secondary progression 2832. We also confirmed that the median time to irreversible disability (EDSS 4 or 6) was shorter in patients with a progressive form than in those with a remittent form. This result is in agreement with previous studies 33. A short interval between the first and second attacks (P1-P2) was also considered to be a poor prognosis. Scalfari et al. found a statistically significant association between P1-P2 interval and disability outcome 34. However, these authors described this result as a less predictive factor than the number of relapses in the first two years. Other studies 2528 reported similar results, showing that a longer P1-P2 interval was correlated with a longer time to disability and passage to secondary progressive form.

It has been described that MSSS-based analysis identifies more risk factors for severity than when the EDSS is used alone. For example, in the two American studies using MSSS, the male gender was correlated with the severity of the disease 3536. While the studies that used the EDSS alone did not identify gender as a risk factor for severity 37. These observations support the suggestion that the MSSS was a more nuanced indicator of disease severity than the EDSS, since it incorporates two factors that are not accounted for by raw EDSS scores: duration of the disease and change in EDSS score over time. This hypothesis was confirmed by the study of Pachner et al, which showed that disease severity as measured by the MSSS was practically stable over several years of follow-up, thus predicting individual-level disease severity over time 38. Benign MS is defined by an EDSS ≤ 2 or ˂3 after 10 years or more of disease progression 3940. In a recent study the authors propose an alternative definition (EDSS ≤1.0, absence from any disability, and the ability to work after 15 years of disease duration) which might truly reflect 41. In our series we found 31 patients (8%) meeting this definition. Although the definition of benign forms remains debated, it is generally admitted that they are characterized by the absence of significant ambulatory impairment 10 to 20 years after the onset of the disease. However, the practical impact of this concept remains limited for two reasons: on the one side, it is a progressive modality that can only be determined a posteriori, and on the other side, epidemiological studies with sufficient hindsight tend to show that a significant proportion of initially "benign" patients nevertheless end up worsening with time, following the usual evolution of other MS patients 41. The suggestion that MS is progressing more slowly than previously thought may represent a change in the type of patient with MS being seen in MS clinics today, perhaps driven by an increased recognition of MS, new diagnostic techniques, availability of immunomodulatory drugs to treat MS and increased survival (related to better health care and management of chronic diseases). All these factors might contribute to the differences between older and newer natural history studies. In fact, the specific component of these factors remains to be determined, in particular the impact of immunomodulatory drugs on long term disability progression 42. There were some limitations to our study that require discussion: we did not study the impact of disease-modifying drugs therapy. Moreover, considering variability in the duration and type of treatment in our population, it may be very hard to include this component in our evaluation. The accessibility of medical care in some centres, a less awareness of MS in the general public and the low socio-economic among some patients may have influenced this time of diagnosis, the time of treatment and the severity of the disease but those data which were not assessed in this study. Nevertheless, this is a detailed study of MS patients in Algeria, and it should provide baseline characteristics of MS patients in this region, since environmental and genetic factors are thought to have significant roles in MS 6.

We wish that in the future other epidemiological studies from other centers in the country will appear to clarify the characteristics and evolution of the disease in our region where multiple sclerosis is little studied. A national registry is strongly recommended, as well as long-term follow-up epidemiological and genetic studies.