Frontiers in Psychiatry | www.frontiersin.or

Edited by:Yanhui Liao,

Sir Run Run Shaw Hospital, China

Reviewed by:Yu-Tao Xiang,

University of Macau, ChinaXiao Jun Xiang,

Central South University, ChinaYanbo Zhang,

University of Alberta, Canada

*Correspondence:Yanqing Tang

[email protected]

Specialty section:This article was submitted to

Addictive Disorders,a section of the journalFrontiers in Psychiatry

Received: 21 May 2020Accepted: 16 July 2020

Published: 17 September 2020

Citation:Liu L, Jia L, Jian P, Zhou Y, Zhou J,

Wu F and Tang Y (2020) The Effects ofBenzodiazepine Use and Abuse on

Cognition in the Elders: A SystematicReview and Meta-Analysisof Comparative Studies.

Front. Psychiatry 11:00755.doi: 10.3389/fpsyt.2020.00755

SYSTEMATIC REVIEWpublished: 17 September 2020doi: 10.3389/fpsyt.2020.00755

The Effects of Benzodiazepine Useand Abuse on Cognition in theElders: A Systematic Review andMeta-Analysis of ComparativeStudiesLinzi Liu1, Linna Jia1, Peiying Jian2, Yifang Zhou3, Jian Zhou1, Feng Wu1

and Yanqing Tang1,3*

1 Department of Psychiatry, The First Affiliated Hospital, China Medical University, Shenyang, China, 2 Department ofPsychology, Queen’s University, Kingston, ON, Canada, 3 Department of Geriatrics, The First Affiliated Hospital, ChinaMedical University, Shenyang, China

Objective: Benzodiazepines (BZD) are one of the most frequently prescribed drugsworldwide. However, the cognitive effects of benzodiazepines in the elderly are highlydebated. This systematic review and meta-analysis aims to explore the following twoquestions in the elderly population: (i) Do BZD lead to any impairments in cognitivefunctions in elderly users? and (ii) Which specific cognitive domains are most affected byBZD use and abuse?

Methods: First, we performed a literature search following the PRISMA guidelines.Electronic databases, including PubMed, PsycINFO, EMBASE, Cochrane Library, andWeb of Science were searched until May 14th, 2020. After selecting the relevant articles,we integrated the results of the selected studies with a standardized cognitiveclassification method. Next, we performed meta-analyses with the random-effectsmodel on the cognitive results. Finally, we specifically examined the cognitiveimpairments of BZD in the abuse subgroup.

Results: Of the included studies, eight of the thirteen had meta-analyzable data.Compared to the controls, elderly BZD users had significantly lower digital symbol testscores (n=253; SMD: -0.61, 95% CI: -0.91 to 0.31, I² = 0%, p < 0.0001). There was nosignificant difference in Mini-Mental State Examination, Auditory Verbal Learning Test, andStroop Color and Word Test scores between BZD users and controls. According to thesubgroup analyses, BZD abusers performed significantly worse than controls in Mini-Mental State Examination (n=7726; SMD: -0.23, 95% CI: -0.44 to -0.03, I² = 86%, p =0.02), while there was no significant difference between the regular BZD users and thecontrols (n=1536; SMD: -0.05, 95% CI: -0.59 to 0.48, I² = 92%, p =0.85).

Conclusion: In the elderly population, the processing speed (digital symbol test scores)was significantly impaired in BZD users; global cognition (Mini-Mental State Examinationscores) was significantly impaired in BZD abusers but not in BZD regular users. This study

g September 2020 | Volume 11 | Article 007551

Liu et al. Cognitive Effect of BZD

Frontiers in Psychiatry | www.frontiersin.or

provides insight into the factors that interact with BZD cognitive effects, such as aging,testing tools, and abuse. Clinicians should be cautious when prescribing BZD for theelderly.

Systematic Review Registration: PROSPERO, identifier CRD42019124711.

Keywords: benzodiazepines, cognitive function, substance addiction, aged, meta-analysis, cognitive dysfuction

INTRODUCTION

Benzodiazepines (BZD) are two-ring heterocyclic compoundsconsisting of a benzene ring fused with a diazepine ring. Since itsdiscovery in the 1950s, BZD’s sedative, hypnotic, anti-anxiety,and anti-convulsive effects have been increasingly accepted,making BZD use highly prevalent among adults (1) andespecially in the elders (2). The prevalence of BZD use inelders varies between 10% and 42% worldwide (3). Forexample, BZD and related drugs are the third most abusedprescription drug in America, with roughly 1-3% of the worldpopulation being subject to abuse (4). However, inappropriateBZD prescriptions can promote BZD misuse, facilitate thedevelopment of BZD addiction, and significantly affect theusers’ overall quality of life (5, 6). Therefore, it is critical forpharmacists, clinicians, and patients to be informed on the latestresearch regarding the adverse effects of BZD use and abuse.

Since the 1970s, research has found negative effects of BZD onrecipients’ cognitive functions (7). A meta-analysis in 2017investigated the long-term cognitive impacts of BZD amongadults. This analysis reported impairments in working memory,language, and processing speed, but not in executive function(reasoning and planning) (8). The participants in the meta-analysis, however, included adults of all ages. Compared to youngadults, elderly populations require more cautiousness whenundertaking BZD therapy. The elderly are more susceptible tocognitive impairment than young adults (9). During the progressof aging, cognitive function continues to decline with structural andfunctional neurological changes (10, 11). The pharmacokinetics ofBZD in the elderly are different than in young adults, so the effects ofBZD in older adults may have unique characteristics than in anyother age group (12).

In recent years, several systematic reviews have found thatBZD use was significantly associated with a higher risk ofdementia and mild cognitive impairments (MCI). Dementiaand MCI introduced a significant growth in mortality andfinancial burdens worldwide (13, 14). The most prominentsymptom of these neurological disorders is a decline incognitive function, measured by tests such as the Mini-MentalState Examination (MMSE). Sufficient evidence has shown thatthe risk of dementia correlates with cumulative dose, treatmentduration, and long-acting effects of BZD molecules (15–18).However, previous literature failed to provide a causal accountfor the link between BZD use and the risk of dementia (18). Tounderstand the mechanism behind the adverse effects of BZD, itis crucial to investigate the specific areas of cognitive functionsthat are impacted by BZD use and abuse.

g 2

Literature in the past 50 years presents contradicting evidenceon whether BZD impairs cognitive functioning in the elderly(19–24). Among the studies that suggest an association betweenBZD use and cognitive impairment in elders, the type and degreeof cognitive impairment reported are inconsistent across studies(19, 20, 22). Moreover, the neurological alterations due to thelong-term effects of BZD use remains unclear (9). In 2018, Pictonand colleagues summarized the cognitive effects of BZD in theelderly from published evidence. They found mixed findings ofthe association between BZD therapy and cognitive decline inelderly users (25). Given the lack of consensus in the currentliterature, a meta-analysis study may help reveal the critical effectsof BZD use in the elderly and identify areas that require furtherresearch. To our knowledge, there are no systematic reviews with ameta-analysis that summarizes the current status of the cognitiveeffect of BZD use and abuse in the elderly population.

This systematic review and meta-analysis aims to explore thefollowing two questions in the elderly: (i) Is BZD use associatedwith impairment in cognitive functions in the elderly? and (ii)Which cognitive domains have declined functionality associatedwith BZD use and abuse? The answers could help characterizethe specific cognition impairments associated with BZD use, andidentify individuals vulnerable to the negative effects of BZD.This meta-analysis may also help identify and monitor thecognitive effects associated with BZD use and abuse to preventBZD addiction. With the high prevalence of BZD beingprescribed to older populations worldwide, it is essential toinform clinicians and patients about the possible cognitiveimpairments associated with BZD use and abuse. Given therefractory rate and adverse effects of dementia and MCI (26), theresults may also help reduce inappropriate BZD prescriptions toattenuate dementia risk in the elderly population.

METHODS

Systematic Review ProtocolThe process of this systematic review follows the PreferredReporting Items for Systematic Reviews and Meta-Analyses(PRISMA) guidelines, 2009 (27).

Databases and Search StrategyWe searched relevant articles in electronic databases, PubMed,PsycINFO, EMBASE, Cochrane Library, and Web of Science,from their inception to May 14, 2020. Specific search keywordsincluded “benzodiazepines”, “cognition”, and “aged”. The queryused for PubMed is ((Benzodiazepines[mh]) OR (Benzodiazepines

September 2020 | Volume 11 | Article 00755

Liu et al. Cognitive Effect of BZD

Compounds[Title/Abstract]) OR (Benzodiazep*[Title/Abstract]))AND (cognit *’ OR memory OR attention OR visual-spatial ORvisuospatial OR recall OR recognition OR problem solving ORreaction time OR vigilance OR executive function*’ OR reasoningOR psychomotor OR motor OR processing OR planning ORverbal fluency OR inhibit *’) AND ((Aged[mh]) OR (Elder*[Title/Abstract]) OR (older adults[Title/Abstract])).

Additionally, we searched in Google Scholar and searched thereference list for relevant articles to ensure that no studieswere missed.

Inclusion Criteria and the Process toIdentify StudiesWe developed the inclusion criteria according to the PICOSguideline. Appropriate papers met the following criteria: 1)Participants were human adults, older than 60 years, meansample age over 65 years; 2) the treatment groups were BZDusers; 3) the studies had placebo or non-users of BZD as controls;4) the outcomes included performances of cognitive functionsmeasured using any standardized neuropsychological instruments;and 5) any type of published clinical studies except for case reportsor conference abstracts were used. We only included articlesreported in English.

Endnote was used to delete the duplicated articles. Tworeviewers (LL and JL) independently scanned the references bytitle/abstract to exclude irrelevant articles, then read the full textto identify the appropriate studies based on the above inclusioncriteria. Finally, the debated studies were determined throughdiscussions with a third reviewer (YT).

Risk of Bias and Quality AssessmentTwo reviewers assessed the quality of each included studyindependently with the modified Newcastle-Ottawa Scale(mNOS) (28). The mNOS examines the quality of non-randomized studies by four bias reduction items: (1) selectionbias, (2) performance bias, (3) detection bias, and (4) reportingbias. Each item is graded from 0 (low quality) to 3 (high quality)according to the example given for each risk level. Disagreementswere resolved by discussions with a third reviewer (Y.T.).Publication bias was tested by using the funnel plot and Begg’stests on the outcome which synthesizes more than five studies.

Data ExtractionAfter generating a list of the included articles, two reviewersmade the data extraction form collectively. The two reviewersthen extracted the data independently. The data extraction formcontained the following information: (1) author and publicationyear, (2) study design, (3) setting (country), (4) study design, (5)mean age, (6) gender distribution, (7) education, (8) participantssources, (9) details of BZD use (BZD types, using time, dosage,and the definition of BZD use), and (10) outcomes (cognitivetask, cognitive domain, and main findings). The authors werecontacted in order to obtain any missing data.

Data Synthesis and Statistical AnalysesThe included studies utilized a variety of psychometricmeasurements, rendering it challenging to produce generalizable

Frontiers in Psychiatry | www.frontiersin.org 3

and informative conclusions. In this paper, we organize cognitivedomains based on a commonly used framework (29, 30). Accordingto this framework, we summarized the outcomes of the cognitivetask from each included study to include a balanced andcomprehensive view. Due to the discrepancy in psychometricscreening tools between studies, the meta-analysis would onlysynthesis the data from the same neuropsychological instruments,such as the MMSE and the Auditory Verbal Learning Test (AVLT).

The RevMan software (31) was used to perform the meta-analysis. We adopted the random-effects model in our meta-analysis because it is more conservative than the fixed-effectsmodel. Standardized mean difference (SMD) with 95% confidenceintervals (CIs) was used for continuous outcomes. The study’sheterogeneity was measured using I2. I2>50% indicated significantheterogeneity (32). The meta-analytic outcomes were two-tailed,with a significance level of 0.05. Based on BZD use informationreported in each study, we divided all the BZD users into twocategories: regular BZD users and BZD abusers. BZD regular use isan appropriate pattern of BZD use that follows the prescriptioninstruction. BZD abuse behaviors include meeting sedative,hypnotic, and anxiolytic use disorder criteria according to theDiagnostic and Statistical Manual of Mental Disorders, FifthEdition, and using a higher frequency or dose, or longer use timethan prescribed, or without a prescription (1, 4, 33). We thenconducted a subgroup analysis to explore the cognitive changes inBZD abusers and regular BZD users.

RESULTS

Search Results and Studies IncludedThe study selection process of this systematic review issummarized in Figure 1. A total of 5072 references werereturned by the initial search and scan, with 44 from relevantreviews and Google Scholar. After removing duplicated andirrelevant articles by title and abstract, the full texts of theremaining 79 were screened and 15 articles finally met theinclusion criteria. Two articles (34, 35) were from the samecohort study named The Canadian Study of Health and Aging(36). Another two articles had repeated participants (22, 37).Overall, 13 studies were included in the literature review. In the13 studies, eight had meta-analyzable data.

Risk of Bias and Quality AssessmentThe risk bias of the included studies was assessed by the mNOSand detailed in Table S1 in the Supplementary Materials. Thescores ranged from 13-19 out of 21. Nine studies (34, 38–45) hada sample size of more than one thousand participants. Ninestudies (34, 38, 40–42, 44–47) used instruments that measuredmulti cognition domains. Eleven studies matched or adjustededucation level as covariates and 12 studies matched or adjustedage as covariates. Therefore, the overall risk bias of the includedstudies is reasonably low. We tested the publication bias of themeta-analysis results of MMSE scores; the funnel plot (Figure S1in Supplementary Materials) and Begg’s tests (p=0.673) did notfind publication bias.

September 2020 | Volume 11 | Article 00755

Liu et al. Cognitive Effect of BZD

Studies CharacteristicsThere were 26033 participants (6374 BZD users) included in thissystematic review and 10666 (2318 BZD users) included in themeta-analysis. As shown in Table 1, all studies were carried outin America or Europe. Six out of thirteen were published in thelast five years. Eight were cohort studies; the remaining five werecross-sectional studies. The mean age of the participants is 72.5in the cohort studies at baseline, and 83.8 in the cross-sectionalstudies. Therefore, both users and controls included in thisreview were elderly participants. In the 13 studies, a total of 36tasks were used to measure cognitive functions; each studyutilized one to nine cognitive tasks.

Synthesized FindingsBenzodiazepine Use and Cognitive Declinein the EldersTable 2 provides an overview of the effects on cognitive functionfrom BZD use consolidated from all the included studies. Theclassification cognitive function division was adapted from acommonly used method of understanding and measuringcognitive domains (29). Eight of the thirteen studies withcognitive performance data had meta-analyzable data. Theyused the same cognitive measurements and reported thequantitative results required for meta-analysis (Table 1). Ourmeta-analysis synthesized data obtained with the same cognitive

Frontiers in Psychiatry | www.frontiersin.org 4

tasks across the included studies. Overall, the meta-analysisincluded the following cognitive functions: global cognitionmeasured with the MMSE, processing speed measured with thedigital symbol test, memory (recognition) measured with theAVLT, and executive functions (inhibitory control) measuredwith the Stroop Color and Word Test (SCWT) (Figures 2A–D).

Eleven out of 14 experiments on global cognition in the 13studies showed no relationship between BZD use and decreasedperformance in global cognition. However, two experimental results(42, 47) showed that BZD users with higher socioeconomic statusand BZD abusers performed worse in the MMSE than non-usingcontrols. The remaining study by Bierman et al. (38) provided athird alternative result. The population-based 9-year cohort studyconcluded that, although there is a significant negative correlationbetween the MMSE scores and the accumulated BZD, the decreasein the MMSE scores with BZD use was relatively small. Moreover,the authors reported no correlation between dosage of BZD andMMSE scores. Taken together, the conclusions drawn from theliterature are consistent with our meta-analysis results (Figure 2A);BZD users did not show significantly worse performance in theMMSE compared to the controls (n=9262; SMD: -0.18, 95% CI:-0.36 to 0.00, I² = 87%, p=0.05).

Seven studies tested the processing speed of BZD users. Fourstudies concluded that BZD use in the elderly population mayresult in significant impairment (41, 42, 47). They tested this

FIGURE 1 | Flow diagram of the search and study identification process.

September 2020 | Volume 11 | Article 00755

TABLE 1 | Summary of included articles.

itive task Cognitive domain

General CognitionGeneral Cognition

RAVLT; FTT General Cognition; Processing speed;Immediate recall: verbal/visual; Gross motorspeed

g Task; RAVLT; General Cognition; Processing speed;Immediate recall: verbal/visual; Reasoning/planning

; TMT-A; BVRT; General Cognition; Processing speed;Immediate recall: verbal/visual; Verbal fluency

SB General Cognition

I; SDMT; RCFT;T FAS; IGT; Towert; N-Back; SCWT

General Cognition; Vigilance/focus; Processingspeed; Immediate recall: verbal/visual; Delayedrecall: verbal/visual; Recognition: verbal/visual;Verbal fluency; Reasoning/planning; Workingmemory; Inhibitory control

T; semantic verballs in one minute);

General Cognition; Immediate recall: verbal/visual; Delayed recall: verbal/visual; Verbalfluency; Reasoning/planning; Processing speed

SCWT; DVT General Cognition; Vigilance/focus; Immediaterecall: verbal/visual; Delayed recall: verbal/visual;Recognition: verbal/visual; Inhibitory control

esign; BuschkeVLT Trial 6; Verbaln Test; WAISAIS Similarities;hension

Processing speed; Immediate recall: verbal/visual; Recognition: verbal/visual; Fine motorspeed; Verbal Fluency; Semantic processing;Reasoning/planning

LT-I; PLT-d;rd Chart

General Cognition; Processing speed;Immediate recall: verbal/visual; Delayed recall:verbal/visual; Inhibitory controlGeneral Cognition

General Cognition

Mental State Examination; TMT-B, The Trail Making Test, part B; DSS, Therail Making Test, part A; BVRT, The Benton Visual Retention Test; IST, IsaacsT, Rey Complex Figure Test; CVLT, California Verbal Learning Test; COWATe and Cued Selective Reminding Test; CDT, clock drawing test; DST, digiture Learning Test-delay.

Liuet

al.Cognitive

EffectofB

ZD

Frontiersin

Psychiatry

|www.frontiersin.org

Septem

ber2020

|Volum

e11

|Article

007555

Study Study design Country Samples (BZDusers)

Mean Age(BZD users)

Male per-centage

(BZD users)

Participants sources Cogn

Gray et al. (39) cohort study U.S. 3434(1018) 74.4(74.5) 40.4(33.1) population based CASIHanlon et al. (40) cohort study U.S. 2765(400) range(65-105) 33.5 community SPMSQ; OMPaterniti et al. (42)a cohort study France 1176(159) 65(65.3) 41.6(23.7) population based TMT-B; DSS;

Bierman et al. (38)b cohort study Netherlands 2105(1189) 69.2 47.5 population based MMSE; CodinRCPM

Mura et al. (41)a,b cohort study France 5195(969) 73.5(74.6) 40.1(23.1) community MMSE; TMT-IST

Zhang et al. (43)b cohort study U.S. 5423(405) 73.0(73.6) 34.1(30.6) Alzheimer’s diseasecenter

MMSE; CDR-

Ros-Cucurull et al.(47)a,b

cohort study Spain 64(33) 73.2(73.5) 28.13(21.9) BZD users fromhospital; controls arevolunteers

MMSE; CPT-CVLT; COWAof London Te

Del Ser et al. (44) cohort study Spain 1087(810) 74.7 36.1 community MMSE; FCSRfluency (animCDT; DST

Hoiseth et al. (46)a cross-sectionalstudy

Norway 241(168) 78.6(78.1) 27.8(25.0) hospital MMSE; HVLT

Helmes and Ostbye(34)a

cross-sectionalstudy

Canada 1754(408) 79.7(79.6) 38.7 community WAIS Block DFree Recall; AFluency; TokeInformation; WWAIS Compr

van Vliet et al. (45)a,b

cross-sectionalstudy

Netherlands 2275(702) range(85-90) 28.0(20.0) community MMSE; LDT;SCWT the Th

Puustinen et al.(37)a

cross-sectionalstudy

Finland 119(64) 81.6(82.1) 23.53(15.6) hospital MMSE

Hessmann et al.(48)a,b

cross-sectionalstudy

Germany 395(49) 78.8(84.0) 31.9(28.6) hospital MMSE

CASI, Cognitive Abilities Screening Instrument; SPMSQ, Short Portable Mental Status Questionnaire; OMC, Orientation-Memory-Concentration Test; MMSE, MinDigit Symbol Substitution Test; RAVLT, Rey Auditory Verbal Learning Test; FTT, Finger Tapping Test; RCPM, Raven’s Coloured Progressive Matrices; TMT-A, The TSet Test; CDR-SB, Clinical Dementia Rating Sum of Boxes; CPT-II, Conners Continuous Performance Test II-Omissions; SDMT, Symbol Digits Modalities Test; RCFAS, Controlled Oral Word Association Test; IGT, Iowa Gambling Task; SCWT, Stroop Color and Word Test; FCSRT, total immediate and delayed recall in the Frsymbol test; HVLT, Hopkins Verbal Learning Test; DVT, Digit Vigilance Test; LDT, Letter Digit Coding Test; PLT-I, Picture Learning Test-immediately; PLT-d, PicParticipants’ characteristics were collected according to baseline information.aThe authors report meta-analyzable data.bThe BZD abuse studies.

C

B

I

s

a

;

ePi

i

Fet

Liu et al. Cognitive Effect of BZD

TABLE 2 | Overview of tasks used to assess cognitive functioning in benzodiazepine users across different cognitive domains.

Cognitive domain Task Studies Sensitivitya

Attention and processing speedVigilance/focus Conners Continuous Performance Test II-Omissions (CPT-II) Ros-Cucurull et al. (47)* 1/1

Digit Vigilance Test (DVT) Hoiseth et al. (46) 0/1Processing speed TMT-B Paterniti et al. (42); Mura et al. (41)* 2/2

TMT-A Mura et al. (41)* 1/1The Digit Symbol Substitution (DSS) test Paterniti et al. (42) 1/1Coding task Bierman et al. (38)* 0/1Symbol Digits Modalities Test (SDMT) Ros-Cucurull et al. (47)* 1/1WAIS Block Design Helmes and Ostbye (34) 0/1Letter Digit Coding Test (LDT) van Vliet et al. (45)* 0/1Digit Symbol Test (DST) Del Ser et al. (44) 1/1

7/11Memory and learningImmediate recall: verbal/visual

Rey Auditory Verbal Learning Test (RAVLT) Paterniti et al. (42); Bierman et al. (38) 0/2

The Benton Visual Retention Test (BVRT) Mura et al. (41)* 1/1Rey Complex Figure Test (RCFT) Immediate recall Ros-Cucurull et al. (47)* 1/1California Verbal Learning Test (CVLT) Ros-Cucurull et al. (47)* 1/1Hopkins verbal learning test (HVLT) Hoiseth et al. (46) 0/1Buschke free recall Helmes and Ostbye (34) 1/1Picture Learning Test (PLT-i) van Vliet et al. (45)* 0/1Free and Cued Selective Reminding Test (FCSRT)-immediaterecall

Del Ser et al. (44) 1/1

5/9Delayed recall: verbal/visual

Rey Auditory Verbal Learning Test (RAVLT) Paterniti et al. (42); Bierman et al. (38)* 1/2

Rey Complex Figure Test (RCFT) Delayed recall Ros-Cucurull et al. (47)* 1/1California Verbal Learning Test (CVLT) Ros-Cucurull et al. (47)* 0/1Hopkins verbal learning test (HVLT) Hoiseth et al. (46) 0/1Picture Learning Test (PLT-d) van Vliet et al. (45)* 0/1the Orientation- Memory-Concentration Test (OMC) hanlon1998* (40) 1/1Buschke free recall Helmes and Ostbye (34) 1/1Free and Cued Selective Reminding Test (FCSRT)-delayedrecall

Del Ser et al. (44) 0/1

4/9Recognition: verbal/visual Rey Auditory Verbal Learning Test (RAVLT) Paterniti et al. (42); Bierman et al. (38)*; Helmes and

Ostbye (34)1/3

Rey Complex Figure Test (RCFT) recognition Ros-Cucurull et al. (47)* 0/1California Verbal Learning Test (CVLT) Ros-Cucurull et al. (47)* 0/1Hopkins Verbal Learning test (HVLT) Hoiseth et al. (46) 0/1

1/69/23

MotorGross motor speed the Finger Tapping Test (FTT) Paterniti et al. (42) 1/1Fine motor speed WAIS Block Design Helmes and Ostbye (34) 0/1

1/2LanguageVerbal fluency The Isaacs Set Test (IST) Mura et al. (41)* 1/1

Controlled Oral Word Association Test (COWAT FAS) Ros-Cucurull et al. (47)* 1/1verbal fluency Helmes and Ostbye (34); Del Ser et al. (44) 0/2

Semantic processing Token Test Helmes and Ostbye (34) 1/1WAIS Information Helmes and Ostbye (34) 0/1

3/6Executive functionsReasoning/planning Raven’s Colored Progressive Matrices (RCPM) Bierman et al. (38)* 1/1

Iowa Gambling Task(IGT) Ros-Cucurull et al. (47)* 0/1Tower of London Test Ros-Cucurull et al. (47)* 1/1WAIS Similarities Helmes and Ostbye (34) 0/1WAIS Comprehension Helmes and Ostbye (34) 1/1clock drawing test Del Ser et al. (44) 0/1

Working memory N-Back Ros-Cucurull et al. (47)* 1/1Inhibitory control SCWT Ros-Cucurull et al. (47)*; Hoiseth et al. (46) 1/2

the third chart of the 40-item SCWT van Vliet et al. (45)* 0/15/10

Frontiers in Psychiatry | www.fro

ntiersin.org 6

September 2020 | Volume 11 | A

* for BZD abuse studies.

rticle 00755

Liu et al. Cognitive Effect of BZD

domain mainly with the Trail Making Test (TMT) or the digitsymbol test. The results showed that BZD users performedsignificantly worse across all the abovementioned tests than thecontrols. A meta-analysis (Figure 2B) was performed usingthe two studies that utilized the TMT. The results showed thatin the elderly, the BZD users performed significantly worse thancontrols in digit symbol tests (n=253; SMD: -0.61, 95% CI: -0.91 to0.31, I² = 0%, p < 0.0001). However, three other studies (34, 38, 45)suggested that the BZD users’ processing speed was notsignificantly impaired during a coding task and block design task.

An overview of the eight studies that tested memory andlearning ability is presented in Table 2. Five out of nine tasksshowed impairment in verbal/visual immediate recall amongBZD users. More specifically, three out of five verbal immediaterecall tasks and two out of four visual immediate recall tasksshowed worse performances in BZD users than controls. Whenmeasuring with verbal/visual delayed recall tasks, four out ofnine studies showed less accurate responses among BZD users.In these tasks, three out of seven verbal delayed recall tasks andone out of two visual delayed recall tasks showed BZD usersperformed worse than controls. The meta-analysis of the AVLTresults (Figure 2C) is high in heterogeneity and non-significant(n=2580; SMD: -0.04, 95% CI: -0.15 to 0.06, I² = 84%, p=0.41).

Results of the tasks for motor, language ability, and executivefunctions are controversial (Table 2). In the domains of language

Frontiers in Psychiatry | www.frontiersin.org 7

ability and reasoning/planning function, three (34, 38, 47) out offour studies (44) that tested the respective domain showed asignificantly worse performance in BZD users. Only one papertested working memory (47) and showed that BZD users had asignificantly worse performance than controls. In the domain ofinhibitory control, the SCWT data sourced from two studies canbe used for meta-analysis (Figure 2D). The results were non-significant and had high heterogeneity (n=1802; SMD: -0.07,95% CI: -0.43 to 0.30, I² = 84%, p =0.72).

Benzodiazepine Abuse and Cognitive Decline in theEldersWe conducted subgroup analyses of BZD abusers (Figure 3).Seven studies included participants with BZD abuse (38, 40, 41,43, 45, 47, 48) (Table 1). In the BZD abuse subgroup, the abusersreceived significantly lower MMSE scores compared to thecontrols (n=7726; SMD: -0.23, 95% CI: -0.44 to -0.03, I² =86%, p =0.02). Meanwhile, in the BZD regular use subgroup,the MMSE scores of BZD users versus controls were notsignificantly different (n=1536; SMD: -0.05, 95% CI: -0.59 to0.48, I² = 92%, p =0.85). Considering I² >50%, the heterogeneitybetween the studies cannot be ignored.

Due to the discrepancy in the type of cognitive tasks used inthe included studies, we could not obtain meta-analyzable data toexamine the effects of BZD abuse on specific cognitive functions.

A

B

D

C

FIGURE 2 | (A) Effect of BZD use on Mini Mental State Examination in the elderly: forest plot. (B) Effect of BZD use on Digital Symbol test in the elderly: forest plot.(C) Effect of BZD use on Auditory Verbal Learning Test in the elderly: forest plot. (D) Effect of BZD use on Stroop Color and Word Test in the elderly: forest plot.

September 2020 | Volume 11 | Article 00755

Liu et al. Cognitive Effect of BZD

Nevertheless, the results of the experiments in these studies weresummarized in Table 2. Notably, for language ability, Mura et al.used the Isaacs Set Test to explore the effect of BZD abuse (41),and Ros-Cucurull used the Controlled Oral Word AssociationTest (47). Both results showed that BZD abusers performedworse than the control groups. In the domain of recognition,BZD abusers performed worse than controls in three differenttasks in two studies (38, 47). Experiments testing performance onother cognitive divisions showed mixed results.

DISCUSSION

Main Findings of BZD UseThis meta-analysis and systematic review included 6374 BZDusers and 19,659 controls to comprehensively investigate theaffected cognition domains by BZD use and abuse in elders. Atotal of 13 papers met the inclusion criteria and were included inthe literature review. Eight out of the 13 papers had appropriatetests and sufficient information for the meta-analysis.

Consistent with previous systematic reviews, our meta-analysissuggests no impairment in global cognition among elderly BZDusers (49, 50). Interestingly, studies with young adults showedopposing results (51, 52), suggesting BZD use significantlyimpairs participants’ global cognitive functioning. One reason forthese results may be the fact that the negative effect of various riskfactors on global cognitive decline decreases with age (53).

The results consistently showed impairment in elderly BZDusers’ processing speed, but not inhibitory control. Processingspeed, defined by the time it takes for a person to complete amental task, has been found to be associated with caudate activityin neuroimaging studies (54). Meanwhile, a higher BZD dose isassociated with volume reductions in the caudate nucleus (55).This imaging evidence supports our results of decreasedprocessing speed among BZD users. Our result in memoryfunctions, however, is inconsistent with previous studies. While

Frontiers in Psychiatry | www.frontiersin.org 8

the meta-analysis (8) with adults of all ages found that BZD usershad a significantly impaired working memory and immediatememory ability, our systematic review and meta-analysis did notfind sufficient support for the effect. However, the interactionbetween age and BZD’s effect on memory requires morerobust validation.

Main Findings of BZD AbuseThe BZD abuse subgroup in this review included 1673 BZD abusersand 6053 controls. The subgroup analysis showed that BZD abusersreceived significantly lower MMSE scores than the controls, whilethe BZD regular users’ scores were not significantly different fromcontrols. These results demonstrated that impairment in globalcognition occurs after the BZD user develops into abuse. However,the high heterogeneity in the results cannot be ignored. Accordingto the results of the subgroup analysis, the confidence intervals ofthe two subgroups overlapped, and the difference between the twogroups was not statistically significant (56). However, the subgroupanalyses results, especially when not pre-specified at the beginningof the trial, cannot reflect the differences between the effect ofinterventions in subgroups (57). Future studies that directly contrastthe cognition of BZD users and abusers are necessary. Only twostudies tested language ability in BZD abusers; both foundsignificantly worse language performance in BZD abuserscompared to controls. Meanwhile, only two studies testedrecognition ability; neither found a significant difference in theperformance between BZD abusers and controls. These findings,however, are not conclusive because of the heterogeneity in thestatistical results, and the small number of studies in the analyses.Therefore, more experiments are needed to reach morereliable conclusions.

After searching the databases, we did not find a meta-analysisinvestigating the cognitive effects of BZD abuse. The most relevantstudies are two systematic reviews and meta-analyses papers ondementia risk in the elderly with long-duration and high dosageBZD users (58, 59). In 2015, Zhong and colleagues summarized sixnested case-control or prospective cohort studies and concluded

FIGURE 3 | Subgroup analysis of the effect of BZD abuse and regular use on mini mental state examination in the elderly.

September 2020 | Volume 11 | Article 00755

Liu et al. Cognitive Effect of BZD

that higher dosage BZD users had an increased risk of dementia(59). In 2019, He et al. found that the risk of dementia was higher inpatients taking BZD for a longer duration (>3 years) among sixcase-control and four cohort studies (58). Although long-term orhigh dosage use is not equivalent to dependence, BZD addiction ismore likely to occur in long-term and high-dose users (60). Thesestudies support the results of our subgroup analysis to some extent,but the mechanism and effects behind BZD addiction and abuseneeds further investigation.

Strengths and Limitations of This ReviewThis study is, to our knowledge, the first systematic review andmeta-analysis of the cognitive effects of BZD use and abuse inelders. We attempted to consolidate results from studies testingdifferent cognitive functions and data from a variety of cognitivetasks by utilizing a mature cognitive domain classificationcatalogue. The subgroup analysis of BZD abusers allowed us topreliminarily compare the effects of BZD use and abuse. Thisanalysis encourages further studies to examine the qualitativedifference of BZD use and abuse. These results can help identifyand monitor the cognitive effects of BZD use and abuse, sheddinglight on awareness and prevention of BZD addiction at earlystages, providing evidence for clinical decision-making, andimproving the life quality of the elderly.

There are some limitations to this meta-analysis and systematicreview. First, the number of studies and cognitive domainsexamined in this meta-analysis was limited. Five of the thirteenstudies did not report cognitive tests data, and the domains ofmotor and language did not have meta-analysis results. Second,although the n is quite large, the meta-analysis results of the digitalsymbol test, AVLT, and SCWT were drawn from two smallsample size studies, which is difficult to justify or interpret,considering the biased nature of the original studies. In thesubgroup analysis of MMSE, the sample size of studies in theBZD regular users’ subgroup varied from hundreds to 2500, whichare much smaller than the sample size of the abusers’ subgroup(>7000). Although the random effects model was used to reducethe impact caused by the difference between the sample size of thetwo subgroups, there was no analyzable data for further analysis.Third, due to the limitation of meta-analysis methodology, wecould not directly compare the cognitive differences betweenBZD users and abusers. Fourth, other variabilities existed in the13 studies in terms of the participants’ source (hospital orcommunity), sample size, sex ratio, and cognitive measurements.These factors could also bias the findings to an uncertain extent.Fifth, users and controls were not well matched in the analysis.Participants in the experimental groups of the studies had differentpreexisting conditions such as anxiety, depression, insomnia, andAPOE e4 status. One study did not age match the interventiongroup with the control group or as a covariable, which means thestudy could not completely distinguish the cognition impaired byBZD and by natural aging. Other factors related to cognition, suchas the use of other psychotropic medications, certain physicaldiseases, and educational level, were not examined in moststudies. Finally, due to the limited number of studies, sources ofheterogeneity were not examined.

Frontiers in Psychiatry | www.frontiersin.org 9

Implications and Future DirectionsOur meta-analysis confirmed the negative effects of BZD onelderly users’ processing speed. Therefore, doctors should becautious when prescribing BZD drugs to elderly patients,especially those with family histories of dementia, Alzheimer’sdisease, and other aging-related cognitive deficits. Additionally,although global cognition was not impaired in BZD regularusers, BZD abusers had significantly worse performance inglobal cognition. This research can inform more individualizedprescription decisions. For example, elderly patients whosedaily activities require higher cognitive processing should beinformed of BZD’s potential side effects on their cognitiveprocessing speed. Patients with a history of addiction shouldprioritize alternative treatments to BZD therapy to prevent BZDdependence and abuse.

Another important finding in this study is that the results ofcognitive performance are highly dependent on the type of cognitivemeasurements in the study. For example, as previously reported,BZD users had significantly lower processing speeds when testedwith the TMT. However, studies measuring processing speed withthe coding task or block design task did not reveal any significantfindings. Therefore, clinical practitioners should be mindful whenselecting cognitive tests. It might be reasonable to use tests withhigher sensitivity to reduce missed diagnoses.

In addition, through our exploration in the literature on thecognitive effects of BZD use, few studies paid attention to BZDabuse and addiction in participants (47, 61, 62). BZD use andabuse can be qualitatively different from BZD use. A surveyconducted in 2015-2016 showed that BZD abusers accountedfor about 17% of BZD users (6). Moreover, approximately halfof the patients who used BZD for longer than 1 month aresubject to BZD abuse or addiction (60). The neglect of the BZDabuse subgroup may be accountable for the mixed conclusionsfrom studies on the cognitive effects of BZD use. Therefore, weencourage future researchers to separate BZD regular users andBZD abusers to achieve more precise and rigid conclusions.

CONCLUSIONS

In conclusion, this meta-analysis indicated no significant globalcognition deficit (MMSE scores) in BZD users, but did revealdeficits in elders with BZD abuse behaviors. BZD users performedsignificantly worse in the cognition domain of processing speed(digit symbol test scores) than the controls, but not in memory andlearning (AVLT scores) or inhibitory control (SCWT scores).Studies that tested the other cognitive domains, however, showedconflicting results. Unfortunately, these cognitive domains’measurements varied across studies, rendering it unavailable to bemerged into meta-analysis. Clinicians should be cautious whenprescribing BZD for the elderly, especially to patients with a familyhistory of age-related cognitive deficits. Moreover, the majority ofthe included studies did not clearly distinguish between the use andabuse of BZD, making it challenging to evaluate the effects of BZDabuse. Future well-designed studies are needed in order to verify thecognitive effects of BZD use and abuse.

September 2020 | Volume 11 | Article 00755

Liu et al. Cognitive Effect of BZD

DATA AVAILABILITY STATEMENT

The raw data supporting the conclusions of this article will bemade available by the authors, without undue reservation, to anyqualified researcher.

AUTHOR CONTRIBUTIONS

LL and YT: conceptualization. LL and LJ: data curation andanalysis. FW and YT: project administration. LL, YZ, JZ, and PJ:supervision, writing—review, and editing. LL and PJ: writing—original draft.

FUNDING

National Key Research and Development Program of China(2018YFC1311604 and 2016YFC1306900 to YT), NationalScience Fund for Distinguished Young Scholars (81725005 toFei Wang), Liaoning Education Foundation (Pandeng Scholar to

Frontiers in Psychiatry | www.frontiersin.org 10

Fei Wang), Innovation Team Support Plan of Higher Educationof Liaoning Province (LT2017007 to Fei Wang), Major SpecialConstruction Plan of China Medical University (3110117059 and3110118055 to Fei Wang), Joint fund of National Natural ScienceFoundation of China (U1808204 to FW), Natural ScienceFoundation of Liaoning Province (2019-MS-05 to FW).

ACKNOWLEDGMENTS

The authors thank Xiang YT and Guo WB for providingtheoretic guidance.

SUPPLEMENTARY MATERIAL

The Supplementary Material for this article can be found online at:https://www.frontiersin.org/articles/10.3389/fpsyt.2020.00755/full#supplementary-material

REFERENCES

1. Blanco C, Han B, Jones CM, Johnson K, Compton WM. Prevalence andCorrelates of Benzodiazepine Use, Misuse, and Use Disorders Among Adults inthe United States. J Clin Psychiatry (2018) 79(6). doi: 10.4088/JCP.18m12174

2. Marra EM, Mazer-Amirshahi M, Brooks G, van den Anker J, May L, Pines JM.Benzodiazepine Prescribing in Older Adults in U.S. Ambulatory Clinics andEmergency Departments (2001-10). J Am Geriatr Soc (2015) 63(10):2074–81.doi: 10.1111/jgs.13666

3. Voshaar RC, Couvée JE, van Balkom AJ, Mulder PG, Zitman FG. Strategiesfor discontinuing long-term benzodiazepine use: meta-analysis. Br JPsychiatry J Ment Sci (2006) 189:213–20. doi: 10.1192/bjp.189.3.213

4. Votaw VR, Geyer R, Rieselbach MM, McHugh RK. The epidemiology ofbenzodiazepine misuse: A systematic review. Drug Alcohol Depend (2019)200:95–114. doi: 10.1016/j.drugalcdep.2019.02.033

5. Airagnes G, Pelissolo A, Lavallee M, Flament M, Limosin F. BenzodiazepineMisuse in the Elderly: Risk Factors, Consequences, and Management. CurrPsychiatry Rep (2016) 18(10):89. doi: 10.1007/s11920-016-0727-9

6. Maust DT, Lin LA, Blow FC. Benzodiazepine Use and Misuse Among Adultsin the United States. Psychiatr Serv (2019) 70(2):97–106. doi: 10.1176/appi.ps.201800321

7. Salzman C, Shader RI, Harmatz J, Robertson L. Psychopharmacologicinvestigations in elderly volunteers: Effect of diazepam in males. J AmGeriatr Soc (1975) 23(10):451–7. doi: 10.1111/j.1532-5415.1975.tb00929.x

8. Crowe SF, Stranks EK. The Residual Medium and Long-term Cognitive Effects ofBenzodiazepineUse: AnUpdatedMeta-analysis.Arch Clin Neuropsychol Off J NatlAcad Neuropsychol (2018) 33(7):901–11. doi: 10.1093/arclin/acx120

9. Lipnicki DM, Crawford JD, Dutta R, Thalamuthu A, Kochan NA, Andrews G,et al. Age-related cognitive decline and associations with sex, education andapolipoprotein E genotype across ethnocultural groups and geographicregions: a collaborative cohort study. PloS Med (2017) 14(3):e1002261.doi: 10.1371/journal.pmed.1002261

10. Goyal MS, Vlassenko AG, Blazey TM, Su Y, Couture LE, Durbin TJ, et al. Lossof Brain Aerobic Glycolysis in Normal Human Aging. Cell Metab (2017) 26(2):353–60.e3. doi: 10.1016/j.cmet.2017.07.010

11. Kovacs GG, Milenkovic I, Wöhrer A, Höftberger R, Gelpi E, Haberler C, et al.Non-Alzheimer neurodegenerative pathologies and their combinations are morefrequent than commonly believed in the elderly brain: a community-based autopsyseries. Acta Neuropathol (2013) 126(3):365–84. doi: 10.1007/s00401-013-1157-y

12. Dingemanse J, Häussler J, Hering W, Ihmsen H, Albrecht S, Zell M, et al.Pharmacokinetic-pharmacodynamic modelling of the EEG effects of Ro 48-6791,

a new short-acting benzodiazepine, in young and elderly subjects. Br J anaesth(1997) 79(5):567–74. doi: 10.1093/bja/79.5.567

13. Wimo A, Guerchet M, Ali GC, Wu YT, Prina AM, Winblad B, et al. Theworldwide costs of dementia 2015 and comparisons with 2010. Alzheimer’sDementia J Alzheimer’s Assoc (2017) 13(1):1–7. doi: 10.1016/j.jalz.2016.07.150

14. Alzheimer's Association. 2016 Alzheimer’s disease facts and figures. Alzheimer’sDementia J Alzheimer’s Assoc (2016) 12(4):459–509. doi: 10.1016/j.jalz.2016.03.001

15. Penninkilampi R, EslickGD.A Systematic Review andMeta-Analysis of the Risk ofDementia Associated with Benzodiazepine Use, After Controlling for ProtopathicBias. CNS Drugs (2018) 32(6):485–97. doi: 10.1007/s40263-018-0535-3

16. Islam MM, Iqbal U, Walther B, Atique S, Dubey NK, Nguyen PA, et al.Benzodiazepine Use and Risk of Dementia in the Elderly Population: ASystematic Review and Meta-Analysis. Neuroepidemiology (2016) 47(3-4):181–91. doi: 10.1159/000454881

17. Gomm W, von Holt K, Thome F, Broich K, Maier W, Weckbecker K, et al.Regular Benzodiazepine and Z-Substance Use and Risk of Dementia: AnAnalysis of German Claims Data. J Alzheimer’s Dis JAD (2016) 54(2):801–8.doi: 10.3233/jad-151006

18. de Gage SB, Pariente A, Begaud B. Is there really a link betweenbenzodiazepine use and the risk of dementia? Expert Opin Drug Saf (2015)14(5):733–47. doi: 10.1517/14740338.2015.1014796

19. Panel BtAGSBCUE. American Geriatrics Society 2019 Updated AGS BeersCriteria® for Potentially Inappropriate Medication Use in Older Adults. J AmGeriatr Soc (2019) 67(4):674–94. doi: 10.1111/jgs.15767

20. Coyle-Gilchrist IT, Peck LF, Rowe JB. Research paper does not show causallink between benzodiazepine use and diagnosis of dementia. BMJ (Clin Res ed)(2012) 345:e7984; author reply e93. doi: 10.1136/bmj.e7984

21. Basu R, Dodge H, Stoehr GP, Ganguli M. Sedative-hypnotic use ofdiphenhydramine in a rural, older adult, community-based cohort: effectson cognition. Am J Geriatr Psychiatry Off J Am Assoc Geriatr Psychiatry(2003) 11(2):205–13. doi: 10.1097/00019442-200303000-00011

22. Vaapio S, Puustinen J, Salminen MJ, Vahlberg T, Salonoja M, Lyles A, et al.Symptoms associated with long-term benzodiazepine use in elderlyindividuals aged 65 years and older: a longitudinal descriptive study. Int JGerontol (2015) 9(1):34–9. doi: 10.1016/j.ijge.2014.03.009

23. Murphy P, Hindmarch I, Hyland CM. Aspects of short-term use of twobenzodiazepine hypnotics in the elderly. Age Ageing (1982) 11(4):222–8.doi: 10.1093/ageing/11.4.222

24. Nikaido AM, Ellinwood EHJr., Heatherly DG, Gupta SK. Age-related increasein CNS sensitivity to benzodiazepines as assessed by task difficulty.Psychopharmacology (1990) 100(1):90–7. doi: 10.1007/bf02245796

September 2020 | Volume 11 | Article 00755

Liu et al. Cognitive Effect of BZD

25. Picton JD, Marino AB, Nealy KL. Benzodiazepine use and cognitive decline in theelderly. Am J Health-Syst Pharm (2018) 75(1):E6–E12. doi: 10.2146/ajhp160381

26. Grasset L, Matthews FE, Pérès K, Foubert-Samier A, Helmer C, Dartigues JF, et al.Evolution of dementia diagnosis over time (1988-2013): Evidence from French andEnglish cohorts. Implication for secular trends analyses. Alzheimer’s Dementia(Amsterdam Netherlands) (2018) 10:490–7. doi: 10.1016/j.dadm.2018.07.005

27. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items forsystematic reviews and meta-analyses: the PRISMA statement. PloS Med(2009) 6(7):e1000097. doi: 10.1371/journal.pmed.1000097

28. Bawor M, Dennis BB, Anglin R, Steiner M, Thabane L, Samaan Z. Sex differencesin outcomes of methadone maintenance treatment for opioid addiction: asystematic review protocol. System Rev (2014) 3:45. doi: 10.1186/2046-4053-3-45

29. Galioto R, Spitznagel MB. The Effects of Breakfast and Breakfast Composition onCognition in Adults. Adv Nutr (Bethesda Md) (2016) 7(3):576s–89s. doi: 10.3945/an.115.010231

30. Lezak MD HD, Bigler ED, Tranel D. Neuropsychological assessment. 4th ed ed.Oxford University Press: New York (2012).

31. Review Manager (RevMan). The Cochrane Collaboration. 2019. Copenhagen:The Nordic Cochrane Centre TCC (2014). Available at: revman.cochrane.org

32. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency inmeta-analyses. BMJ (Clin Res ed) (2003) 327(7414):557–60. doi: 10.1136/bmj.327.7414.557

33. Yen CF, Ko CH, Chang YP, Yu CY, Huang MF, Yeh YC, et al. Dependence,misuse, and beliefs regarding use of hypnotics by elderly psychiatric patientstaking zolpidem, estazolam, or flunitrazepam. Asia-Pacific Psychiatry Off JPacific Rim Coll Psychiatrists (2015) 7(3):298–305. doi: 10.1111/appy.12147

34. Helmes E, Ostbye T. Associations between Benzodiazepine Use andNeuropsychological Test Scores in Older Adults. Can J Aging = La Rev Candu vieillissement (2015) 34(2):207–14. doi: 10.1017/s0714980815000082

35. Ebly EM, Hogan DB, Fung TS. Potential adverse outcomes of psychotropicand narcotic drug use in Canadian seniors. J Clin Epidemiol (1997) 50(7):857–63. doi: 10.1016/S0895-4356(97)00118-2

36. Canadian Study of Health and Aging working Group. Canadian study of healthand aging: study methods and prevalence of dementia. CMAJ CanMed Assoc J = Jl’Association Med Can (1994) 150(6):899–913. doi: 10.1001/jama.271.11.872

37. Puustinen J, Nurminen J, Kukola M, Vahlberg T, Laine K, Kivela SL. Associationsbetween use of benzodiazepines or related drugs and health, physical abilities andcognitive function – A non-randomised clinical study in the elderly. Drugs Aging(2007) 24(12):1045–59. doi: 10.2165/00002512-200724120-00007

38. Bierman EJM, Comijs HC, Gundy CM, Sonnenberg C, Jonker C, BeekmanATF. The effect of chronic benzodiazepine use on cognitive functioning inolder persons: Good, bad or indifferent? Int J Geriatr Psychiatry (2007) 22(12):1194–200. doi: 10.1002/gps.1811

39. Gray SL, Dublin S, Yu O, Walker R, Anderson M, Hubbard RA, et al.Benzodiazepine use and risk of incident dementia or cognitive decline:Prospective population based study. BMJ: Br Med J (2016) 352:i90. doi:10.1136/bmj.i90

40. Hanlon JT, Horner RD, Schmader KE, Fillenbaum GG, Lewis IK, Wall WE Jr.,et al. Benzodiazepine use and cognitive function among community-dwellingelderly. Clin Pharmacol Ther (1998) 64(6):684–92. doi: 10.1016/S0009-9236(98)90059-5

41. Mura T, Proust-Lima C, Akbaraly T, Amieva H, Tzourio C, Chevassus H, et al.Chronic use of benzodiazepines and latent cognitive decline in the elderly:Results from the Three-city study. Eur Neuropsychopharmacol (2013) 23(3):212–23. doi: 10.1016/j.euroneuro.2012.05.004

42. Paterniti S, Dufouil C, Alperovitch A. Long-term benzodiazepine use and cognitivedecline in the elderly: the Epidemiology of Vascular Aging Study. J ClinPsychopharmacol (2002) 22(3):285–93. doi: 10.1097/00004714-200206000-00009

43. Zhang Y, Zhou XH, Meranus DH, Wang L, Kukull WA. Benzodiazepine Useand Cognitive Decline in Elderly with Normal Cognition. Alzheimer Dis AssocDisord (2016) 30(2):113–7. doi: 10.1097/WAD.0000000000000099

44. Del Ser T, Zea MA, Valenti M, Olazaran J, Lopez-Alvarez J, Rebollo-VazquezA, et al. Effects of commonly prescribed drugs on cognition and mild cognitiveimpairment in healthy elderly people. J Psychopharmacol (2019) 33(8):965–74. doi: 10.1177/0269881119857206

45. van Vliet P, van der Mast RC, van den Broek M, Westendorp RGJ, de CraenAJM. Use of benzodiazepines, depressive symptoms and cognitive function inold age. Int J Geriatr Psychiatry (2009) 24(5):500–8. doi: 10.1002/gps.2143

Frontiers in Psychiatry | www.frontiersin.org 11

46. Hoiseth G, Tanum L, Tveito M, Kristiansen KM, Kvande K, Lorentzen B, et al. AClinical Study of the Cognitive Effects of Benzodiazepines in PsychogeriatricPatients. Pharmacopsychiatry (2013) 46(6):209–13. doi: 10.1055/s-0033-1349131

47. Ros-Cucurull E, Palma-Álvarez RF, Garcıá-Raboso E, Cardona-Rubira C,Jacas C, Grau-López L, et al. Benzodiazepine Use Disorder and CognitiveImpairment in Older Patients: A Six-Month-Follow-Up Study in anOutpatient Unit in Barcelona. J Stud Alcohol Drugs (2018) 79(6):844–52.doi: 10.15288/jsad.2018.79.844

48. Hessmann P, Dodel R, Baum E, Muller MJ, Paschke G, Kis B, et al.Prescription of Benzodiazepines and Related Drugs in Patients with MildCognitive Deficits and Alzheimer’s Disease. Pharmacopsychiatry (2019) 52(2):84–91. doi: 10.1055/s-0044-100523

49. Nader D, Gowing L. Is Long-TermBenzodiazepine Use a Risk Factor for CognitiveDecline? Results of a Systematic Review. J Addict (2020) 2020:1569456.doi: 10.1155/2020/1569456

50. Lapeyre-Mestre M. [Benzodiazepines, cognitive decline and dementia: Areview of causality criteria from published observational studies]. Therapie(2019) 74(3):407–19. doi: 10.1016/j.therap.2018.09.071

51. Barker MJ, Greenwood KM, Jackson M, Crowe SF. An evaluation of persistingcognitive effects after withdrawal from long-term benzodiazepine use. J IntNeuropsychol Soc (2005) 11(3):281–9. doi: 10.1017/s1355617705050332

52. Federico A, Tamburin S, Maier A, Faccini M, Casari R, Morbioli L, et al.Multifocal cognitive dysfunction in high-dose benzodiazepine users: a cross-sectional study. Neurol Sci Off J Ital Neurol Soc Ital Soc Clin Neurophysiol(2017) 38(1):137–42. doi: 10.1007/s10072-016-2732-5

53. Legdeur N, Heymans MW, Comijs HC, Huisman M, Maier AB, Visser PJ. Agedependency of risk factors for cognitive decline. BMC Geriatr (2018) 18(1):187. doi: 10.1186/s12877-018-0876-2

54. De Simoni S, Jenkins PO, Bourke NJ, Fleminger JJ, Hellyer PJ, Jolly AE, et al.Altered caudate connectivity is associated with executive dysfunction aftertraumatic brain injury. Brain J Neurol (2018) 141(1):148–64. doi: 10.1093/brain/awx309

55. Huhtaniska S, Jaaskelainen E, Heikka T, Moilanen JS, Lehtiniemi H, Tohka J, et al.Long-term antipsychotic and benzodiazepine use. and brain volume changes inschizophrenia: The Northern Finland Birth Cohort 1966 study. Psychiatry Resh-Neuroimaging (2017) 266:73–82. doi: 10.1016/j.pscychresns.2017.05.009

56. Higgins JPT TJ, Chandler J, Cumpston M, Li T, Page MJ, Welch VA. CochraneHandbook for Systematic Reviews of Interventions. 2nd Edition ed. Chichester(UK): John Wiley & Sons (2019).

57. Wang SS, Ou YC, Cheng CL, Dahm P. Evidence-based urology in practice: whento believe a subgroup analysis? BJU Int (2010) 105(2):162–4. doi: 10.1111/j.1464-410X.2009.09053.x

58. He Q, Chen X,Wu T, Li L, Fei X. Risk of Dementia in Long-Term BenzodiazepineUsers: Evidence from a Meta-Analysis of Observational Studies. J Clin Neurol(Seoul Korea) (2019) 15(1):9–19. doi: 10.3988/jcn.2019.15.1.9

59. Zhong G, Wang Y, Zhang Y, Zhao Y. Association between Benzodiazepine Useand Dementia: A Meta-Analysis. PloS One (2015) 10(5):e0127836. doi: 10.1371/journal.pone.0127836

60. Soyka M. Treatment of Benzodiazepine Dependence. New Engl J Med (2017)376(12):1147–57. doi: 10.1056/NEJMra1611832

61. Bergman H, Borg S, Engelbrektson K, Vikander B. Dependence on sedative-hypnotics: neuropsychological impairment, field dependence and clinical course ina 5-year follow-up study. Br J Addict (1989) 84(5):547–53. doi: 10.1111/j.1360-0443.1989.tb00612.x

62. Chu C, Sivakumar K, Melinosky C. A Case of Reversible BilateralHippocampal Injury Secondary to Benzodiazepine Abuse (P3.1-017).Neurology (2019) 92(15 Supplement):P3.1–017.

Conflict of Interest: The authors declare that the research was conducted in theabsence of any commercial or financial relationships that could be construed as apotential conflict of interest.

Copyright © 2020 Liu, Jia, Jian, Zhou, Zhou, Wu and Tang. This is an open-accessarticle distributed under the terms of the Creative Commons Attribution License(CC BY). The use, distribution or reproduction in other forums is permitted, providedthe original author(s) and the copyright owner(s) are credited and that the originalpublication in this journal is cited, in accordance with accepted academic practice. Nouse, distribution or reproduction is permitted which does not comply with these terms.

September 2020 | Volume 11 | Article 00755

<< /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /PageByPage /Binding /Left /CalGrayProfile (Dot Gain 20%) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated 50SWOP51 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Warning /CompatibilityLevel 1.4 /CompressObjects /Tags /CompressPages false /ConvertImagesToIndexed true /PassThroughJPEGImages true /CreateJDFFile false /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /DetectCurves 0.0000 /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails false /EmbedAllFonts true /EmbedOpenType false /ParseICCProfilesInComments true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams false /MaxSubsetPct 1 /Optimize true /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveDICMYKValues true /PreserveEPSInfo true /PreserveFlatness false /PreserveHalftoneInfo false /PreserveOPIComments true /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Preserve /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true ] /NeverEmbed [ true ] /AntiAliasColorImages false /CropColorImages false /ColorImageMinResolution 300 /ColorImageMinResolutionPolicy /OK /DownsampleColorImages true /ColorImageDownsampleType /Bicubic /ColorImageResolution 300 /ColorImageDepth -1 /ColorImageMinDownsampleDepth 1 /ColorImageDownsampleThreshold 1.50000 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.40 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /ColorImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasGrayImages false /CropGrayImages false /GrayImageMinResolution 300 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.40 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /GrayImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasMonoImages false /CropMonoImages false /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile () /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False /Description << /ENU (T&F settings for black and white printer PDFs 20081208) >> /ExportLayers /ExportVisibleLayers /Namespace [ (Adobe) (Common) (1.0) ] /OtherNamespaces [ << /AsReaderSpreads false /CropImagesToFrames true /ErrorControl /WarnAndContinue /FlattenerIgnoreSpreadOverrides false /IncludeGuidesGrids false /IncludeNonPrinting false /IncludeSlug false /Namespace [ (Adobe) (InDesign) (4.0) ] /OmitPlacedBitmaps false /OmitPlacedEPS false /OmitPlacedPDF false /SimulateOverprint /Legacy >> << /AddBleedMarks false /AddColorBars false /AddCropMarks false /AddPageInfo false /AddRegMarks false /BleedOffset [ 0 0 0 0 ] /ConvertColors /NoConversion /DestinationProfileName () /DestinationProfileSelector /DocumentCMYK /Downsample16BitImages true /FlattenerPreset << /ClipComplexRegions true /ConvertStrokesToOutlines false /ConvertTextToOutlines false /GradientResolution 300 /LineArtTextResolution 1200 /PresetName ([High Resolution]) /PresetSelector /HighResolution /RasterVectorBalance 1 >> /FormElements false /GenerateStructure true /IncludeBookmarks true /IncludeHyperlinks true /IncludeInteractive false /IncludeLayers false /IncludeProfiles false /MarksOffset 6 /MarksWeight 0.250000 /MultimediaHandling /UseObjectSettings /Namespace [ (Adobe) (CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /DocumentCMYK /PageMarksFile /RomanDefault /PreserveEditing true /UntaggedCMYKHandling /LeaveUntagged /UntaggedRGBHandling /UseDocumentProfile /UseDocumentBleed false >> << /AllowImageBreaks true /AllowTableBreaks true /ExpandPage false /HonorBaseURL true /HonorRolloverEffect false /IgnoreHTMLPageBreaks false /IncludeHeaderFooter false /MarginOffset [ 0 0 0 0 ] /MetadataAuthor () /MetadataKeywords () /MetadataSubject () /MetadataTitle () /MetricPageSize [ 0 0 ] /MetricUnit /inch /MobileCompatible 0 /Namespace [ (Adobe) (GoLive) (8.0) ] /OpenZoomToHTMLFontSize false /PageOrientation /Portrait /RemoveBackground false /ShrinkContent true /TreatColorsAs /MainMonitorColors /UseEmbeddedProfiles false /UseHTMLTitleAsMetadata true >> ]>> setdistillerparams<< /HWResolution [2400 2400] /PageSize [612.000 792.000]>> setpagedevice