Mortality In Multiple Sclerosis: A Systematic Literature Review and Meta-Analysis

Background: Mortality is an important aspect of the longitudinal course of multiple sclerosis (MS) as a chronic disease with relatively early onset. While some evidence suggests that mortality may be increased in MS, the rate relative to non-MS populations is not well-established, and the factors affecting mortality are not clear. The standardized mortality ratio (SMR), a rate ratio used to compare mortality between a specific group and the general population, is a useful tool for evaluating the overall mortality burden in MS.

Objectives: To conduct a systematic review of the literature on the relative mortality rates in MS and to quantitatively synthesize the available evidence by performing meta-analysis of the SMRs.

Methods: Using keywords for mortality and MS, we systematically searched the English language, MEDLINE- and EMBASE-indexed literature published between 1/1/1980–8/8/2012. Included articles were studies evaluating at least 100 MS patients and reporting data pertaining to SMRs or similar measures (e.g., hazard ratio [HR] or relative risk [RR]). Random effects meta-analyses were performed to estimate a mean SMR, and heterogeneity was explored through meta-regression.

Results: Twenty articles reporting SMRs, HRs, or RRs were retained after an extensive systematic review, of which four were additional analyses on previously published MS populations. The majority of the included studies were on North American and European MS populations with data collection periods between 1940 and 2008. SMRs identified in the literature ranged from 1.3 to 4.4, and the contribution of study methodologies and population-based factors (e.g., geography and age at MS onset) to SMR variability is unclear. Based on estimates from the meta-analysis, the overall SMR in MS is 2.52 (95% confidence interval: 2.15–2.89), representing a death rate approximately 2.5 times that of the general population.

Conclusions: MS is associated with an increased risk of mortality; however, country, treatment and age effects are not clear, warranting additional research. This study is limited by heterogeneity of studies included in the analysis. Future analyses of longitudinal population-based data may provide a more complete account of mortality in MS populations.