Background and aims The precise knowledge of the natural history of Huntington's disease (HD) and the impact of variables that may modify the rate of survival are essential for counselling of patients and design of future therapeutic trials. Failure to use survival analyses methodology in previous studies may have produced serious bias in estimates of the survival rates in the disease. The aim of the present study was to examine the determinants of survival in HD.

Methods We retrospectively reviewed 309 medical records of all HD patients and at risk subjects between 1977 and 2008 followed at the Department of Neurological Sciences of ‘Federico II’ University in Naples. Inclusion criteria were manifest signs and symptoms and positive molecular testing for HD (CAG triplet number >35). Demographic and clinical data were collected at the first evaluation and at follow-up although the last day we examined the patient in 2008.

Results A total of 135 patients were included in the study. The remaining 174 subjects were excluded for the following reasons: gene test missing (n=84), lack of traceability (n=71), premanifest state (n=16) and refusal to participate in the study (n=3). At the time of data collection, 94 patients were still living and 41 patients were deceased (19 males and 22 females) with an average death age of 56.55 years±14.94 (mean±SD, range 18–83). Mean follow-up time after diagnosis was 145 months (range 12–552). Higher survival probabilities were registered in patients with CAG repeat length ≤47 (n=101, 75% of the cohort; p=0.015) and with onset ≤41 (n=74, 55% of the cohort; p=0.046). The individual contribution of each covariate was assessed using a Cox analysis and the final model was generated considering only the significant ones.

Conclusions Survival time was longer in patients with a triplet repeat size of the mutant allele ≤47 and with onset of the disease ≤41 years of age. No significant impact either of the size of the wild-type allele or its interaction with the expanded allele was found in the survival rates. Larger studies using the appropriate statistical tool are necessary to better understand the disease duration determinants in HD.