Association between Elevated Plasma Vitamin B12 and Short-Term Mortality in Elderly Patients Hospitalized in an Internal Medicine Unit

Abstract

Background. The prognostic value of vitamin B12 blood levels remains controversial. An association between elevated vitamin B12 and mortality has been reported, particularly among elderly patients with cancers and liver or blood diseases. The present study explored the relationship between mortality and elevated vitamin B12 levels in a population of unscheduled inpatients in an internal medicine unit. Methods. This retrospective observational analysis was conducted between August 2014 and December 2018. We compared 165 patients with elevated plasma vitamin B12 levels (>600 pmol/l) with a random sample of 165 patients with normal B12 levels who were hospitalized during the same period. Demographic, clinical, and biological characteristics were assessed during hospitalization. The primary endpoint was all-cause death at 1 year. Results. Patients with elevated B12 were younger, with a lower body mass index and lower plasma albumin than those with normal B12 (75 ± 16 years vs 79 ± 13 years, p = 0.047; 23 ± 5 vs 26 ± 7 kg/m2, $p<0.001$ ; and 33 ± 5 vs 35 ± 5 g/l, $p<0.001$ , respectively). The prevalence of auto-immune disease and referral from an intensive care unit was higher among patients with elevated B12 (11% vs 5%, p = 0.043 and 36% vs 10%, $p<0.001$ , respectively). After 1 year of follow-up, 64 (39%) patients with elevated B12 had died compared to 43 (26%) patients with normal B12 (p = 0.018). Multivariate analysis using the Cox proportional hazards regression model adjusted for age, gender, body mass index, intensive care unit hospitalization, albumin level, and the presence of solid cancer or autoimmune disease revealed elevated B12 to be associated with a significant risk of death in the first year of follow-up (hazard ratio: 1.71 [1.08–2.7], p = 0.022). Conclusion. Elevated B12 is an early warning indicator of increased short-term mortality, such as independently of age, cancer, or comorbidities, in patients hospitalized in an internal medicine department.