Total Protein Level in Cerebrospinal Fluid is Stable in Elderly Adults

To the Editor: Protein analysis of cerebrospinal fluid (CSF) samples from elderly people may help with the clinical diagnosis of infectious, inflammatory, or cognitive disorders, but during aging, several processes modify CSF protein levels.1 A reduction in the CSF turnover rate leads to an accumulation of nondegraded proteins.2, 3 Many neurodegenerative and neurological diseases are associated with high CSF protein concentrations,4 which blood–brain barrier dysfunction5, 6 and neuronal damage7 can explain. The combination of low CSF secretion and low CSF turnover can result in a prolonged half-life of many CSF proteins. These factors lead to accumulation of nondegraded proteins and may be related to diseases such as Alzheimer's disease or normal pressure hydrocephalus.8 It is important to understand changes in CSF total protein level during aging if such proteins are to be used as biomarkers of disorders. To the knowledge of the authors of this letter, CSF total protein levels have not been studied during aging, making it difficult to interpret this biological parameter. Age-related CSF total protein concentrations were investigated in a large cohort of individuals aged 51–100 to clarify the use of this parameter in clinical practice. Lumbar punctures were performed according to the standard procedure for routine diagnosis. All CSF samples were collected as part of clinical practice and analyzed in the University Hospital of Tours between January 2009 and November 2012. CSF samples were collected in test tubes and immediately transferred to the laboratory. Results were retrospectively computed for all CSF samples (N = 6,896). CSF samples of adults aged 51 and older on the day of the lumbar puncture were included (n = 3,408). CSF samples with normal glucose level (2.75–4.40 mmol/L; Hexokinase method), red blood cell count of less than 10/μL, and white blood cell count less than 4/μL were included (n = 1,192). Cerebrospinal fluid was centrifuged (2000 g, 10 minutes, at room temperature) and analyzed directly. Total CSF protein concentration was determined using a colorimetric method (Pyrogallol red/molybdate) using an analyzer with an absorbance at 600 nm (urinary/CSF protein reagent, AU 2700; Beckman Coulter, Krefeld, Germany). The laboratory participates in external quality control programs. Descriptive data are presented as medians and interquartile ranges. A quantile box plot was used to describe the mean distributions of age groups. CSF protein concentrations for each age group were compared using the Kruskal–Wallis nonparametric test because of right-skewed distributions in each age group. Analyses used R Software (v 2.12.2, R Foundation for Statistical Computing, Vienna, Austria). Total protein concentrations were measured in 1,192 CSF samples. CSF protein levels ranged from 0.10 to 3.34 g/L (mean 0.48 ± 0.24 g/L). Samples were divided into five groups corresponding to decades of age (51–60, n = 301; 61–70, n = 294; 71–80, n = 383; 81–90, n = 181; 91–100, n = 33). Distribution according to age group is shown in Figure 1. The means, medians, and dispersions did not differ significantly between age groups. Cerebrospinal fluid total protein level was measured in a large number of subjects aged 51–100. Total protein levels did not vary greatly between groups. This study shows that, despite processes related to age possibly influencing the half-life of many proteins, total protein concentration is not significantly affected. It is likely that changes in rates are low (in the context of total protein concentration) and thus are not detected by the determination of CSF total protein concentration, suggesting that it is not necessary to provide age-specific reference ranges when examining CSF total protein concentration during aging. Total protein level in the CSF of an elderly adult found to be greater than the normal threshold should be regarded as pathological and be investigated further. Results were analyzed from individuals with a medical indication for a lumbar puncture because it was difficult to obtain CSF from a large number of healthy volunteers. To minimize the effect of diseases, CSF samples with cellularity (to avoid diseases such as infectious or carcinomatous meningitis, or samples with blood) and those with high glucose levels (to avoid diseases such as diabetes mellitus, with which there may also be an isolated increase in total protein level) were excluded. The effect of the other rare causes of high total protein level was minimized by including a large number of subjects in the study. Total protein level in the CSF of elderly individuals is stable, but the individual distribution of proteins may vary because of their lower turnover and changes in blood–brain barrier integrity, for example. A study of protein composition using methods with high sensitivity could be of great interest for the understanding of neurodegenerative diseases.9 The authors would like to thank Elodie Boidras and Emilie Villeneuve for their technical assistance. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper.