Prognostic factors in ALS: different approaches to the same problem

Abstract

The natural history of amyotrophic lateral sclerosis (ALS), the prognoses, and the survival times are fields of considerable interest that are scarcely studied in South American countries.To describe the survival of a representative cohort of Uruguayan ALS patients, and to identify covariates associated with survival using different analyses.Survival was assessed using the Kaplan-Meier method. Different Cox proportional hazards functions were used to identify independent prognostic predictors since the diagnosis: classic, stratified, and truncated.We included 166 definite and probable ALS patients. The median follow-up was of 13.6 years. An analysis was performed according to the recruitment groups: prevalent, exhaustive incident, and non-exhaustive incident cases. The median survival since the diagnosis was longer in the prevalent group (33 months) than in the exhaustive incident (22 months) and non-exhaustive incident (14 months) groups. The median survival time of the entire cohort from onset to death was 37 months and 23 months from the diagnosis. Factors related to survival from diagnosis to death were: age at onset, bulbar region onset, clinical form, and progression rate.The present study described the role of clinical and demographic factors in ALS survival in the Uruguayan population and shed light on differences involving survival models and the temporal bias produced by the lack of precision in determining the onset of the disease.

Figure 1

Types of left truncation. Type-I left truncation can affect the representativeness of the sample, while type-II affects the exposure time.

Figure 2

( A ) Overall Kaplan-Meier survival estimates (survival since diagnosis); ( B ) overall Kaplan-Meier survival estimates (survival since symptom onset); ( C ) Kaplan-Meier survival estimates (since diagnosis) according to the recruitment group.