Lack of direct involvement of a diazepam long-term treatment in the occurrence of irreversible cognitive impairment: a pre-clinical approach

Abstract

Several observational studies have found a link between the long-term use of benzodiazepines and dementia, which remains controversial. Our study was designed to assess (i) whether the long-term use of benzodiazepines, at two different doses, has an irreversible effect on cognition, (ii) and whether there is an age-dependent effect. One hundred and five C57Bl/6 male mice were randomly assigned to the 15 mg/kg/day, the 30 mg/kg/day diazepam-supplemented pellets, or the control group. Each group comprised mice aged 6 or 12 months at the beginning of the experiments and treated for 16 weeks. Two sessions of behavioral assessment were conducted: after 8 weeks of treatment and after treatment completion following a 1-week wash-out period. The mid-treatment test battery included the elevated plus maze test, the Y maze spontaneous alternation test, and the open field test. The post-treatment battery was upgraded with three additional tests: the novel object recognition task, the Barnes maze test, and the touchscreen-based paired-associated learning task. At mid-treatment, working memory was impaired in the 15 mg/kg diazepam group compared to the control group (p = 0.005). No age effect was evidenced. The post-treatment assessment of cognitive functions (working memory, visual recognition memory, spatial reference learning and memory, and visuospatial memory) did not significantly differ between groups. Despite a cognitive impact during treatment, the lack of cognitive impairment after long-term treatment discontinuation suggests that benzodiazepines alone do not cause irreversible deleterious effects on cognitive functions and supports the interest of discontinuation in chronically treated patients.

Fig. 1. Study design, timeline, and methods of the study.

The top line shows the timing of treatment and behavioral sessions. The bottom line shows the age of mice (mature and middle-aged) at the different stages of the experiment. The lower part of the figure indicates the behavioral tests performed for each assessment session and their order. EPM elevated plus maze, NORT novel object recognition test, OFT open field test, PAL paired-associate learning.

Fig. 2. Mid-treatment consequences of a chronic diazepam administration on: anxiety, spontaneous locomoto ractivity and working memory.

The effects of 2 months of a 15 mg/kg/day (D15), a 30 mg/kg/day (D30) of diazepam treatment, or control conditions (CTL) on: A the percentage of time spent in open arms of mature and middle-aged mice during the elevated plus maze (EPM) test, B the total distance traveled during the open field test (OFT) performed on middle-aged mice, C the number of entries and the exclusion rate of mature and middle-aged mice in the Y maze test, D the percentage of spontaneous alternations of mature and middle-aged mice during the Y maze test, E table for Spearman correlation coefficients, and F relationship between the number of entries in the Y maze spontaneous alternation task and plasma nordiazepam or oxazepam concentrations. Mature mice are shown in black and middle-aged mice in red. Results are expressed as mean +/− SD for the Y maze test and as median (IQR) for the EPM and OFT, *p < 0.05, **p < 0.01, ***p < 0.001. The statistical analysis was performed using one-way ANOVA followed with Bonferroni’s post hoc comparison and Spearman test for the correlation analysis.

Fig. 3. Long-term consequences of a chronic diazepam exposure after treatment withdrawal on: anxiety, spontaneous locomotor activity, working memory and visual recognition memory.

The long-term effects of a chronic diazepam treatment after drug withdrawal on A the percentage of time spent in open arms of mature and middle-aged mice during the elevated plus maze (EPM) test, B the total traveled distance during the open field test (OFT) performed on middle-aged mice, C the number of entries and the exclusion rate of mature and middle-aged mice in the Y maze test, D the percentage of spontaneous alternation of mature and middle-aged mice during the Y maze test, and E the discrimination index during the novel object recognition (NOR) test. Mature mice are shown in black and middle-aged mice in red. Results are expressed as mean +/− SD for the EPM, Y maze spontaneous alternation, and NOR tests and as median (IQR) for the OFT.

Fig. 4. Long-term consequences of a chronic diazepam exposure after treatment withdrawal on: spatial reference learning and memory and visuospatial learning.

The long-term effects of a chronic diazepam treatment after drug withdrawal on A the mean total escape latency during the learning session and B the time spent in target quadrant during the probe trial of the Barnes Maze test and C percentage of correct response performed by mice in the dPAL task. Mature mice are shown in black and middle-aged mice in red. Results are expressed as mean +/− SD.