Ocena dokazov, da zloraba metamfetamina povzroča kognitivni upad pri ljudeh | nevropsihoparmakologija

Ocena dokazov, da zloraba metamfetamina povzroča kognitivni upad pri ljudeh | nevropsihoparmakologija

Anonim

Predmeti

  • Kognitivna nevroznanost
  • Farmakologija
  • Psihiatrične motnje

Izvleček

Metamfetamin (MA) je ena najpogosteje zlorabljenih prepovedanih snovi po vsem svetu. Med drugimi težavami je bila zloraba drog povezana z zmanjšanim kognitivnim delovanjem na več področjih. Vendar pa večina literature ni poskušala razlikovati kognitivnih težav, ki jih povzroča zloraba MA, od obstoječih kognitivnih težav, ki jih verjetno povzročajo drugi dejavniki. Tu se lotevamo tega vprašanja in ocenjujemo dokaze za a priori hipoteze, ki se nanašajo na šest raziskav: (a) študije na živalih; (b) študije na človeku v presečnem delu; (c) dvojna študija; (d) študije sprememb kognicije z abstinenco od MA; (e) študije sprememb v zgradbi in delovanju možganov z abstinenco od MA; in (f) študije razmerja med resnostjo zlorabe MA in obsegom opaženih kognitivnih primanjkljajev. Ugotovitve so bile na splošno mešane, pri čemer je bila nekatera podpora vzročni povezanosti med zlorabo MA in upadom kognitiv, in druge ugotovitve kažejo, da ni nobene zveze. Prevladovanje podatkov pa podpira možnost, da zloraba MA povzroči kognitivni upad neznanega trajanja pri vsaj nekaterih uporabnikih drog. Če v povprečju pri posameznih osebah ta padec verjetno postane blag v zgodnji do srednji odrasli dobi. Vendar pa bodo moderatorke spremenljivke verjetno prispevale k prisotnosti in / ali resnosti kognitivnega upada, ki ga je pokazal določen posameznik.

UVOD

Zloraba in odvisnost od metamfetamina (MA) sta največji javnozdravstveni težavi (Rawson in Condon, 2007), amfetamin pa je drugi po marihuani po razširjenosti uporabe prepovedanih drog po vsem svetu (Urad Združenih narodov za droge in kriminal (UNODC, 2011)). Čeprav se je začetek uporabe MA v Združenih državah Amerike od leta 2002 nekoliko zmanjšal (SAMHSA, 2009), se je število zdravil za MA več kot podvojilo med leti 1998 in 2007 (SAMHSA, 2009). Poleg tega se ne glede na povečanje uporabe zdravljenja, ko se razvije odvisnost od MA, preneha uporaba pogosto težko. Stopnja recidivov po psihosocialnih in farmakoloških zdravljenjih je visoka (Baker in sod., 2005; Elkashef in sod., 2008; Rawson in sod., 2004; Shoptaw in sod., 2008; Zorick in sod., 2011), za tiste, ki iščejo zdravljenje, pa večkratno zdravljenje poskusi so pogosto norma in ne izjema (Anglin in sod., 1997; Hillhouse in sod., 2007).

Poleg psihiatričnih in družbenih težav, povezanih z zlorabo MA, je vse več raziskav raziskovalo, ali je zloraba MA povezana s kognitivnimi primanjkljaji. Metaanaliza 17 presečnih raziskav je pokazala, da imajo ljudje, ki so zlorabili MA, bistveno nižje kognitivne rezultate kot kontrolni udeleženci, ki niso zlorabili drog (Scott et al, 2007). Učinki so bili največji pri ukrepih učenja ( d = –0, 66), izvršilnih funkcij ( d = –0, 63), pomnilnika ( d = –0, 59) in hitrosti obdelave ( d = –0, 52), čeprav se je večina kognitivnih domen bistveno razlikovala med skupinami. Presečne študije pa ne morejo razlikovati kognitivnih slabosti, ki so lahko bile pred zlorabo MA od tistih, ki so posledica tega. Zlasti longitudinalne študije so pokazale, da lahko otroški primanjkljaj izvršilne funkcije napoveduje zlorabo drog v mladostništvu (Tarter in sod., 2003, 2004), kar kaže na to, da so vsaj nekatere kognitivne pomanjkljivosti, opažene pri udeležencih, odvisnih od MA, lahko premorbidne. Poleg tega je večina razpoložljivih razsežnih raziskav uporabila pomanjkljive zasnove, brez ustreznega primerjanja preskusnih primerov in nadzora nad potencialno pomembnimi spremenljivkami, kot so ocene premorbidne inteligence, izobraževanja in drugih zlorab drog in alkohola. Te in druge omejitve so izzvale ugotovitev, da so dokazi o kognitivnih primanjkljajih pri osebah, odvisnih od MA, šibki (Hart in sod., 2011; za komentar pa glej Payer in sod. (2012)).

Kljub pomembnim posledicam potencialnega upada kognitiv, ki jih povzroča MA, za javno zdravje in zlasti za zdravje uporabnikov MA, kritični pregled dokazov o vzročni povezavi med zlorabo MA in kognitivnimi funkcijami ni. Ta rokopis je odziv na to vrzel v literaturi. Ker so vzročne eksperimentalne zasnove, ki vključujejo naključno dodelitev kroničnega MA ali dajanje placeba, pri ljudeh etično prepovedane, so dokazi, ki jih ponuja ta pregled, po naravi posredni in inferencialni. V tem pregledu so predstavljeni dokazi o naslednjih šestih hipotezah, ki podpirajo hipoteze, ki nakazujejo, da MA povzroča upad kognitiv:

  1. Pri živalih, ki so izpostavljene MA, bo prišlo do kognitivnega upada, zlasti pri dajanju režimov odmerjanja, ki posnemajo vzorce zlorabe MA pri ljudeh.

  2. Posamezniki, ki zlorabljajo MA, bodo imeli slabše kognitivne učinke kot dobro usklajeni posamezniki, ki MA ne zlorabljajo.

  3. Dvojčki, ki zlorabljajo MA, bodo imeli manjši kognitivni učinek kot njihovi dvojčki, ki ne zlorabljajo MA.

  4. Posamezniki, ki zlorabljajo MA, bodo pokazali izboljšanje kognicije s trajno abstinenco.

  5. Zloraba MA bo povezana s spremembami v človeških možganih. V idealnem primeru bodo te možganske spremembe povezane s kognitivnimi spremembami.

  6. Kognitivni primanjkljaji pri osebah, ki zlorabljajo MA, bodo odvisni od odmerka.

ŽIVALI, IZPOSTAVLJENE MA, POKAŽEJO KOGNITIVNI DECLINE, POSEBNO, KI SO UPRAVLJENI ODDELITVENI REŽIMI, KI SO MIMIČNE PATTERNIJE ČLOVEKOVA ZLOČILA

Pri ljudeh, odvisnih od MA, se vnos MA običajno giblje med 0, 5 do 1, 4 g na dan (Hoffman in sod., 2006; Kim in sod., 2006; King in sod., 2010; Simon in sod., 2002), pri čemer je najverjetneje omejen dnevni vnos. s finančnimi stroški MA in / ali fiziološkimi / psihološkimi posledicami uporabe (npr. hiperaktivnost, psihoza). Če predpostavimo, da povprečna odrasla oseba tehta 80 kg (Ogden in sod., 2004), to pomeni vnos 6–17, 5 mg / kg na dan, čeprav čistost vira in način uporabe lahko vplivata na natančno količino zaužitih MA. Glodalci, ki jim je dovoljeno samozdravljenje infuzij MA, se prav tako približajo temu obsegu uživanja, saj dnevni vnos MA presega 6 mg / kg po samo 21 dneh dostopa (Parsegian et al, 2011; Reichel et al, 2011; Rogers et al., 2008). Ker odrasli, odvisni od MA, zdravilo običajno uporabljajo med 1 in 5-krat na dan (McKetin in sod., 2008; Simon in sod., 2002), ocenjujemo, da se večina odmerkov MA giblje med 0, 60 do 3, 5 mg / kg na odmerek, čeprav je toleranten Uporabniki injiciranja lahko presežejo 4 mg / kg na odmerek, pri čemer so največji dovoljeni odmerki presegli 12 mg / kg na injekcijo (Buffum in Shulgin, 2001; Kramer et al, 1967). Ker podgane in druge živali presnavljajo MA lažje kot ljudje (Caldwell in sod., 1972), ni jasno, ali so enakovredni odmerki mg / kg fiziološko enaki med ljudmi in živalmi. Kljub temu smo v tem pregledu posebej pozorni na živali, ki so jim prejeli odmerke MA, ki so v mejah možne prehrane ljudi (npr. <3 mg / kg na odmerek).

Najmanj 1 teden po dajanju MA, podgane in miši, ki jim dajemo velike odmerke MA (4 mg / kg na odmerek, pogosto večkrat na dan), kažejo primanjkljaj na več kognitivnih področjih, vključno s spominskim prepoznavanjem predmetov (Belcher in sod., 2005; Siegel in sod., 2010), spomin za prepoznavanje vonja (O'Dell in sod., 2011), prostorsko učenje (Acevedo in sod., 2007; Vorhees in sod., 2009), zaporedno učenje (Chapman in sod., 2001; Daberkow in sod., 2005), učenje integracije poti (Herring in sod., 2008), delovni spomin (Mizoguchi in sod., 2011), diskontiranje napora (podobno kot zamuda pri popuščanju) (Kosheleff in sod., 2012) in obratno učenje (Izquierdo idr. al, 2010).

Podobno kot pri ljudeh lahko akutno dajanje zmernih odmerkov MA (2 mg / kg) na živalih izboljša kognitivne zmogljivosti, kot so ukrepi obratnega učenja (Kulig in Calhoun, 1972), delovni spomin (Shoblock et al, 2003 ) in prednost večjim, zapoznelim nagradam pred manjšimi, takojšnjimi nagradami (Richards in sod., 1999). Vendar lahko kronično, namesto akutno, dajanje zmernih odmerkov (2 mg / kg dnevno) poslabša kognitivne zmogljivosti, vključno z ukrepi delovnega spomina (Lee in sod., 2011; Nagai in sod., 2007) in spomina za prepoznavanje predmetov (Arai et al, 2009; Ito in sod., 2007; Noda in sod., 2010). Dejansko dajanje mišem 1 mg / kg MA dnevno 7 dni povzroči zanesljiv primanjkljaj v spominu za prepoznavanje predmetov, tudi če ga ocenimo vsaj 1 teden po prenehanju uživanja drog (Kamei in sod., 2006; Lu et al, 2010; Mizoguchi et al, 2011 ).

Čeprav prejšnje študije dokumentirajo, da lahko kognitivni upad nastane kot posledica kroničnega dajanja zmernih odmerkov MA naivnim živalim, ta vzorec uporabe drog ne odraža vzorca uporabe, značilnega za ljudi, pri katerem lahko uporaba MA narašča s časom odmerjanje in pogostost (Sommers et al, 2006). Na primer, Segal et al (2003) so pokazali, da lahko naraščajoči režim odmerjanja MA pri podganah (npr. Od 0, 1 mg / kg do 4 mg / kg v 14 dneh) lahko povzroči toleranco do kasnejših odmerkov MA, tako, da se vedenjski in biološki primanjkljaji, ki nastanejo zaradi odmerka poživljanja, zmanjšajo. S pomočjo te paradigme je nekaj raziskav ugotovilo, da primanjkljaj v pomnilniku za prepoznavanje predmetov (Belcher in sod., 2008; Clark in sod., 2007) in delovnem spominu (Simoes et al., 2007), ki ga povzroča odmerek preganjanja pri podganah, v veliki meri prepreči predhodno stopnjevanje odmerka. Vendar pa je nedavna študija opic vervet uporabila naraščajoči odmerek MA (od 0, 1 mg / kg do 4 mg / kg na dan v 31-dnevnem obdobju) ugotovila, da so opice razvile selektivni primanjkljaj v sposobnosti zaviranja odziva na prej nagrajeno spodbudo (Groman in sod., 2012). Ti primanjkljaji so se izboljšali z abstinenco, kar kaže na to, da lahko pride do okrevanja s prekinitvijo dajanja MA.

Poleg stopnjevanja odmerjanja ljudje sčasoma samozdravljajo MA, namesto da bi zdravilo dajali neprostovoljno. Da bi posnemali to vedenje, je več raziskav modeliralo stopnjevanje odmerjanja, v katerem se MA daje samim podganam, ki dobivajo infuzijo po prostovoljnem pritisku na ročico, običajno z uporabo 0, 02 mg / 50 μl MA na infuzijo. Da bi odmerjanje stopnjevalo s časom, je podganam prvih 5–7 dni dajanja dovoljen omejen dostop do infuzij (npr. 1 ha na dan), nato pa nadaljnji dostop do infuzij (npr. 6 ha na dan) za nadaljnjih 14–21 dnevi. V tem režimu podgane sčasoma povečajo njegovo dajanje, navadno prvih nekaj dni prejemajo 1 mg / kg MA na dan, vendar do konca zdravljenja presegajo 6 mg / kg na dan. Z uporabo te paradigme je bilo pokazano, da pri podganah ocenjujejo 1 dan po dajanju MA, primanjkljaj v namernem premiku (Setsean et al, 2011). Podgane, ki so jih ocenili vsaj en teden po dajanju MA, opažajo primanjkljaj v spominu za prepoznavanje predmetov (Reichel in sod., 2011; Rogers et al., 2008) in ohranili vizualno pozornost in zaviralni nadzor (Dalley in sod., 2007). V teh študijah uprava MA ni pomembno vplivala na druge ukrepe pomnilnika prostorske konfiguracije in nekatere indekse pozornosti (npr. Znotrajdimenzionalni premik).

Zaključek

Kronično dajanje odmerkov MA, ki bodo verjetno v območju človeške porabe, pri naivnih živalih povzroči kognitivni primanjkljaj. Čeprav dokazi kažejo, da uporaba povečevalnih odmerkov MA lahko prepreči kognitivni primanjkljaj, ki ga sicer povzročijo zmerni do visoki odmerki MA, kljub temu pa številne raziskave odkrivajo, da lahko samostojni in eksperimentalni povečevalni odmerki povzročajo kognitivni primanjkljaj pri opicah in podganah ( Dalley in sod., 2007; Groman in sod., 2012; Parsegian in sod., 2011; Reichel in sod., 2011; Rogers in sod., 2008).

INDIVIDUALI, KI SO ZLORABLJENI, LAHKO DELUJO KOGNITIVNO ZMOGLJIVOST OD DOBRO UMETNIH INDIVIDUALOV, KI JIH NE zlorabljajo

Ker literatura, ki primerja kognitivno uspešnost ljudi, odvisnih od MA, in zdravih udeležencev nadzora pogosto trpi zaradi slabega raziskovalnega načrta, v katerem skupine niso primerljive z zunanjimi spremenljivkami (Hart et al, 2011), tukaj pregledujemo samo tiste študije, ki so primerjale kognicijo pri udeležencih, ki so odvisni od MA in jih nadzorujejo, in minimalno nadzorovani glede na starost, spol, izobrazbo in ocene IQ, običajno ocenjeni z meritvami branja / izgovorjave. Ta pregled je omejen tudi na študije, ki so dobile psihiatrično diagnozo odvisnosti od MA (običajno s strukturiranim kliničnim intervjujem za DSM (First et al, 1996)) in poskušale izključiti komorbidna psihiatrična in zdravstvena stanja, vključno s trenutno odvisnostjo od drog kot MA. Na koncu smo vključili le študije, v katerih je bila analiza urina potrjena abstinenca od drog v času kognitivnih testov.

Sedem študij, ki so primerjale kognitivno uspešnost odvisnih od MA in zdravih udeležencev nadzora, je ustrezalo tem kriterijem (tabela 1). Od teh študij le ena majhna študija (Leland in sod., 2008) ni zaznala pomembnih razlik v kognitivni uspešnosti med skupinami. V drugih šestih raziskavah (Gonzalez in sod., 2004; Henry in sod., 2009, 2010; Kalechstein in sod., 2003; Rendell in sod., 2009; Woods in sod., 2005) so ugotovili, da so udeleženci, odvisni od MA, na nekaterih kognitivnih testih opravili slabše rezultate kot zdravih udeležencev nadzora, vključno s študijo funkcionalnih sposobnosti v resničnem svetu (npr. ukrepi finančne funkcije, komunikacije, prevoza, upravljanja z zdravili itd.) (Henry et al, 2010). V teh študijah so se odvisne od MA in kontrolne skupine značilno razlikovale le pri podskupini opravljenih testov. Kot je bilo podobno ugotovljeno v metaanalizi, ki sta jo izvedla Scott et al (2007), sta verbalno učenje in spomin predstavljala dokaj skladno področje šibkosti za udeležence, odvisne od MA.

Tabela polne velikosti

Poleg študij, ki so na ključnih demografskih spremenljivkah ustrezale MA in nadzirale udeležence, je bilo več raziskav statistično nadzorovano glede na demografske razlike med skupinami. Od teh osmih raziskav (glej tabelo 1) je pet ugotovilo, da so udeleženci, odvisni od MA, opravili slabše rezultate kot kontrolni udeleženci na podskupini izvajanih kognitivnih testov (Hoffman in sod., 2006; Kim in sod., 2006; Rippeth in sod., 2004 ; Salo in sod., 2007, 2009). Druge tri študije niso našle pomembnih razlik med skupinami na večkratnih kognitivnih testih (Chang in sod., 2005; King in sod., 2010; Simon in sod., 2010), čeprav sta dve izmed teh študij uporabili Bonferronijev popravek za statistično pomembnost, in to metoda je lahko preveč konzervativna, kadar so odvisne spremenljivke medsebojno povezane (Miller, 1981), kot je to primer z uspešnostjo na večini kognitivnih testov (Warner in sod., 1987).

Zaključek

Pri pregledu kognitivnih podatkov iz razumno dobro ujemajočih se skupin odvisnih od MA in zdravih udeležencev v nadzoru je večina študij ugotovila, da imajo posamezniki, odvisni od MA, vsaj nekatere kognitivne teste nižje ocene od kontrolnih oseb, čeprav so nekatere študije izjeme pri povsem nepomembne razlike (Chang in sod., 2005; King in sod., 2010; Leland in sod., 2008; Simon in sod., 2010). Opozarjamo, da nismo našli nobenih dobro ujemajočih se študij, v katerih bi udeleženci, odvisni od MA, opravili bistveno boljše rezultate kot kontrolni subjekti na katerem koli kognitivnem testu, kar bi lahko pričakovali, če bi bile razlike med skupinami zgolj artefakt statistične napake tipa I. Dokazi torej kažejo, da imajo vsaj nekateri posamezniki, odvisni od MA, nižje kognitivne funkcije, kot bi bilo pričakovati zaradi njihovih demografskih značilnosti.

Nobena od razpoložljivih študij ni omogočila razkroja ploskve njihovih kognitivnih podatkov, tako da je bilo mogoče opaziti prekrivanje učinkovitosti med osebami, odvisnimi od MA in kontrolnimi osebami. Vendar pa je razlika v uspešnosti med udeleženci, odvisnimi od MA in kontrolnimi udeleženci, v povprečju majhna, saj so najbolj pomembne razlike med skupinami znotraj (in pogosto nižje od) 1 SD uspešnosti (na podlagi standardnih odstopanj za poročene skupine v študijah). Ob predpostavki, da so porazdelitve rezultatov približno normalne, sumimo, da je običajno, da odvisne od MA in kontrolne skupine kažejo pošteno stopnjo prekrivanja kognitivnih zmogljivosti.

Primanjkljaji, ki jih odkrijejo udeleženci v presečnih študijah, odvisni od MA, lahko še vedno odražajo dejavnike, ki so udeležence nagibali k zlorabi drog, ne pa da so posledica zlorabe, v presečnih študijah pa je nemogoče nadzorovati vse morebitne zmede. Čeprav so na primer skoraj vse pregledane študije izključile trenutno odvisnost od drugih zdravil razen MA (z izjemo nikotina), je pri posameznikih, ki zlorabljajo MA, izjemno zlorabljena polisciplina. Glede na to, da so taki, če jih obravnavamo ločeno, so dokazi o presežku kognitivnega upada, ki jih povzroča MA, omejeni.

Blizanci, ki zlorabljajo MA, bodo imeli SPODNJO KOGNITIVNO UČINKOVITOST NJIHOVIH DVOJNIH PAR, KI NE zlorabljajo MA

Ker imajo gensko povezani posamezniki ponavadi podobne kognitivne sposobnosti (Bouchard, 1998; Winterer in Goldman, 2003) in so pogosto vzgojeni skupaj, dvojčki, ki se ne strinjajo z zlorabo MA, nudijo dragocen navidezno eksperimentalni načrt za nadzor nad tujimi zmešnjavami. Le ena raziskava je preučila uspešnost dvojčkov, ki niso bili zlorabljeni zaradi zlorabe MA (Toomey in sod., 2003; čeprav sta Ersche in sod., 2012 primerjala tudi brate in sestre, ki niso bili v skladu s stimulativno odvisnostjo od zaviralnega nadzornega ukrepa, je velika večina teh udeležencev zlorabila kokain, ne MA). Toomey in sod. (2003) so pregledali nevropsihološko delovanje 50 moških parov dvojčkov iz Vietnam Era Twin Registry (31 monozygotic; 19 dizygotic), v katerih je imel samo eden izmed članov zgodovino uporabe močnih stimulansov. Uporaba močnih stimulansov je bila opredeljena kot tedenska uporaba spodbujevalcev vsaj 1 leto (dvojnik, ki ne zlorablja, nikoli uporabil spodbujevalcev tedensko), vendar noben udeleženec v letu pred testiranjem ni podprl uporabe stimulansov ali marihuane. Med tistimi, ki so zlorabili stimulanse, jih je večina (80%) zlorabila amfetamine (devet v kombinaciji s kokainom), preostali pa so v prvi vrsti zlorabili kokain. Par dvojčkov se je na splošno dobro ujemal in se ni bistveno razlikoval v izobraževanju, učnih težavah v šoli, zgodovini poškodbe glave, življenjskih diagnozah osi I, uživanju alkohola, uživanju cigaret, odvisnosti od marihuane ali IQ celotne lestvice, ki jo je ocenil revidirani Wechsler Inteligenčna lestvica za odrasle (WAIS-R; obe skupini pomeni = 103).

Rezultati so pokazali, da sta imela dvojčka, ki sta zlorabljala stimulante, bistveno slabše rezultate kot njihovi partnerji, ki ne zlorabljajo, na dveh časovnih testih pozornosti (odpoved in sled pri delu A) in dveh testih psihomotorne hitrosti / spretnosti (Finger Taping in Grooved Pegboard; p <0, 05 ). Vendar sta dvojčka, ki zlorabljajo stimulante, na enem preizkusu trajne pozornosti (test neprekinjene učinkovitosti: pravilna številka, opustitve in občutljivost) opravila bistveno bolje kot nenapadniki. Nobena od spremenljivk iz drugih devetih preizkušanih testov se med skupinami ni bistveno razlikovala, vključno s preskusi izvršilne funkcije, neverbalnim sklepanjem, branjem, vizualnim spominom ali verbalnim spominom.

Zaključek

Edina ustrezna razpoložljiva študija dvojčka je pokazala, da so dvojčki, ki zlorabljajo stimulante, imeli nekaj kognitivnih ukrepov slabše od zlorabe dvojčkov, čeprav je bil ta vzorec obrnjen na enem testu, v katerem so zlorabi spodbujevalcev delovali bolje kot nenapadniki. Vendar pa ta študija ni uporabila urinov, da bi zagotovili, da so bili udeleženci v času testiranja brez drog, kar je pomembno glede na to, da lahko akutni MA izboljša osnovno učinkovitost (Hart in sod., 2008; Mahoney in sod., 2011). Nadalje ni jasno, ali tedenska uporaba MA, ki je bila preučena v tej študiji, predstavlja odvisnost ali zlorabo, kot jo določa DSM-IV. Rezultati kažejo, da je tedenska uporaba stimulansov lahko povezana z nekaj kognitivnimi slabostmi, ne pa tudi z zmanjšanjem svetovnega kognitivnega sistema.

ČLOVEKI, KI JIH ZLOČUJEJO, POKAŽE IZBOLJŠANJE POZNAVANJA S TRAJINO VZDRŽEVANJEM

Možno je, da je vsak padec kognitivnih funkcij, ki ga povzroči zloraba MA, trajen in nepopravljiv. Če zloraba MA povzroči nepopravljiv kognitivni upad, odsotnost razmerja med kognicijo in trajno abstinenco od MA ne vpliva na vzročno zvezo med zlorabo MA in kognitivno funkcijo. Kljub temu bi izboljšanje kognicije oseb, ki zlorabljajo MA, s trajno abstinenco zagotovilo posredne dokaze, da zloraba MA zavira normalno stanje kognitivnih funkcij.

Nekaj ​​raziskav je pokazalo, da so posamezniki, odvisni od MA, bolje opravili kognitivne teste, ko so bili testirani v zgodnji abstinenci in spet pri poznejši abstinenci (Jaffe in sod., 2005; Volkow in sod., 2001; Wang in sod., 2004). Vendar pa te študije niso vključevale kontrolne skupine, ki je bila ponovno testirana v primerljivih intervalih, zato je bilo nemogoče ugotoviti, ali so bile izboljšave testnih rezultatov artefakt prejšnje izpostavljenosti testiranju. Izboljšave, povezane s predhodno izpostavljenostjo preskusom, so pogoste in so razvidne več kot eno leto po prvem testiranju (Basso in sod., 1999; Beglinger in sod., 2005; Dikmen in sod., 1999).

Tri presečne študije so primerjale kognitivno uspešnost udeležencev, ki so zlorabljali MA, ki so bili abstinentni za različna časovna obdobja. S pomočjo Stroop Task, Salo in sod. (2009) so primerjali uspešnost udeležencev, odvisnih od MA, ki so bili na kratko abstinentni (od 3 tednov do 6 mesecev), in udeležencev, odvisnih od MA, ki so bili abstinentni vsaj 1 leto. Dve skupini udeležencev MA se nista razlikovali po starosti, spolu, izobrazbi ali premorbidnem IQ-ju. Rezultati so pokazali, da so udeleženci, ki so bili na kratko abstinentni, opravili bistveno slabše naloge Stroop kot udeleženci, ki so bili abstinentni vsaj eno leto.

Kim et al (2006) je udeležencem, odvisnim od MA, ki so bili abstinentni v obdobju manj kot 6 mesecev, izvedli tri nevropsihološke teste (razvrščanje s kartami Wisconsina, test stroop in sledilski test) in primerjali njihovo uspešnost z udeleženci, odvisnimi od MA, ki so bili abstinent več kot 6 mesecev. Skupine se niso bistveno razlikovale po starosti, spolu, socialno-ekonomskem statusu, izobrazbi, uživanju alkohola, kajenju cigaret ali trajanju uživanja MA. Udeleženci, ki so bili abstinentni dlje od 6 mesecev, so bili na Wisconsinovem razvrščanju kartice (WCST) bistveno boljši od tistih, ki so bili abstinentni manj kot 6 mesecev. Učinkovitost Stroop in Trailmaking se med skupinama ni bistveno razlikovala. V tej študiji so bili dobljeni tudi strukturni MRI pregledi in osebe, odvisne od MA, z daljšo abstinenco so imele večje količine sive snovi v desnem srednjem čelnem girusu (rMFG) kot tiste s krajšo abstinenco. Poleg tega so bile celotne in vztrajne napake na WCST negativno povezane s količino sive snovi v rMFG.

Simon et al (2004) je nevropsihološki akumulator upravljal z udeleženci, ki so z njim zlorabljali MA, ki so med zdravljenjem neprekinjeno vzdržali zdravljenje, in njihovo delovanje primerjali z dvema skupinama, ki so zlorabljali MA: (1) Zlorabniki MA, ki so imeli med zdravljenjem so bili recidivirani, vendar so bili abstinentni najmanj 4 tedne pred kognitivnim testiranjem; in (2) zlorabe MA, ki med zdravljenjem niso prenehale uporabljati MA in so bile pozitivne na MA na vsaki seji zdravljenja. Rezultati so razkrili, da je skupina, ki je med testiranjem še naprej uporabljala MA, bistveno boljša od skupin z relapsom in abstinenco na testih spomina. Ta rezultat je lahko posledica akutnih učinkov MA na uspešnost. Poleg tega so bili v tej raziskavi uporabljeni nadomestni preskusni obrazci in ni jasno, da so bile raziskovalne skupine primerljive glede na uporabljene testne obrazce.

Samo dve vzdolžni študiji sta primerjali nevropsihološko uspešnost posameznikov, odvisnih od MA, v obdobju trajne abstinence z uspešnostjo zdravih kontrolnih oseb, ki so bili testirani v primerljivih intervalih. Simon et al (2010) je nevropsihološko akumulatorje udeležencem, odvisnim od MA, dodelil v 4–9 dneh abstinence in ponovno po 1 mesecu trajne abstinence. Kontrolna skupina je bila testirana v primerljivem intervalu. Po ponovnem preizkušanju se skupina, odvisna od MA, ni bistveno izboljšala od kontrolne skupine na nobeni kognitivni domeni; izboljšali so 0, 14 standardnih odstopanj več kot kontrolna skupina na celotni kognitivni bateriji, vendar se ta razlika ni približala statističnemu pomenu ( p = 0, 33).

Iudicello in sod. (2010) so na nevropsihološkem akumulatorju dvakrat - na začetku in ponovno približno 1 leto pozneje - na nevropsihološki bateriji, odvisni od MA. Nekateri udeleženci, odvisni od MA, so v tem letu ostali abstinentni ( n = 25), nekateri pa so se zmenili ( n = 58), vendar so bili na testiranju čisti od MA. Rezultati so pokazali, da se pri ponovnem testiranju abstinentne in neobstojne skupine MA niso spremenile bistveno bolj kot kontrolne skupine na kognitivni bateriji. Toda post-hoc analize so pokazale trend ( p = 0, 06) pri abstinentnih udeležencih, odvisnih od MA, ki so bili na začetku oslabljeni, da bi se pri ponovnem testiranju izboljšali bolj kot pri ponovnem vključevanju udeležencev MA in zdravem nadzoru (ne glede na to, ali so bili oslabljeni na začetku ali niso oslabljeni). Vendar je bilo nekaj testiranih podskupin zelo majhnih (npr. N = 6).

Zaključek

Dokazi za izboljšanje kognitiv, povezani z abstinenco od uporabe MA pri posameznikih, odvisnih od MA, so mešani. Dve od treh presečnih študij abstinence kažejo, da se nekatere kognitivne funkcije (npr. Izvršilno delovanje) izboljšajo z abstinenco. Vendar je edina longitudinalna študija z močno eksperimentalno zasnovo, ki je ocenila 1 leto abstinence (Iudicello in sod., 2010), ugotovila, da se le določena podvrsta oseb, odvisnih od MA (tisti, ki so oslabljeni na začetku), izboljša z abstinenco, medtem ko druge odvisne od MA posamezniki se ne izboljšajo.

MA ZLORABE SE PRIDRUŽUJEMO SPREMEMBE V ČLOVEKOVEM ZAVORU

Ker struktura in funkcija možganov vplivata na kognitivno uspešnost (glej Lezak in sod. (2004)), spremembe v možganih, povezane z zlorabo MA, lahko pomenijo, da se pojavijo tudi kognitivne spremembe, kar najbolje kažejo študije, ki hkrati ocenjujejo oba možgana zgradba / funkcija in kognitivna uspešnost. Pri živalih so poleg kognitivnega upada, ki ga povzročajo kronični zmerni (2 mg / kg dnevno) ali stopnjevajoči se odmerki MA, opažene sočasne spremembe pri D2-podobnem dopaminskem receptorju in vezavi prenašalcev dopamina (GAT) in Groman et al, 2012 ), raven dopamina in serotonina v tkivih (Lu et al, 2010), vezava na NMDA receptorje (Lee in sod., 2011), izražanje glutamatnih receptorjev (mGluR5) (Reichel in sod., 2011), odstranjevanje piramidalnih nevronskih celic (Parsegian et al, 2011) in z novostjo povzročeno hiperfosforilacijo zunajcelične signalne kinaze 1/2 (Ito in sod., 2007; Kamei in sod., 2006; Nagai in sod., 2007). Presečne študije pri ljudeh prav tako kažejo razlike v možganski strukturi in funkciji med odvisnimi od MA in zdravih udeležencev nadzora (za preglede glej Berman in sod. (2008); Chang in sod. (2007); Salo in Fassbender (2012)) . Vendar pa je pri ljudeh mogoče, da bi nekatere ali vse te razlike v naravi lahko bile premorbidne. Tako smo tukaj osredotočili svoj pregled na spremembe v možganih, ki so jih opazili v longitudinalnih študijah, saj imajo te spremembe večjo verjetnost, da odražajo posledico in ne zgolj korelacijo zlorabe MA. Trenutno so bile vse vzdolžne študije možganov opravljene med abstinenco od zlorabe MA, ne pa med aktivno uporabo. Kljub temu bodo spremembe, ki se pojavijo v možganih med abstinenco, verjetno odražale kompenzacijske odzive na učinke zlorabe MA.

S pozitronsko emisijsko tomografijo (PET) in radioaktivno označenim [ 11 C] d-treo- metilfenidatom, Volkow in drugi (2001) so preučili spremembe v transporterju dopamina (DAT) pri strijah petih udeležencev, odvisnih od MA, ki so jih ovrednotili enkrat po 6 ali manj mesecev abstinence in spet po 9 mesecih abstinence. Po dolgotrajni abstinenci se je razpoložljivost DAT znatno povečala v jedru hudata (+ 19%) in v možganih (+ 16%); in ta povečanja so bila močno povezana s trajanjem abstinence ( r = 0, 92). Nevropsihološke teste smo izvajali v obeh časovnih točkah. Izboljšanje nevropsihološke učinkovitosti in povečanje razpoložljivosti DAT sta pokazali pozitivna gibanja v dveh preskusih (časovno sprehod in pozno priklic, p <0, 18), vendar nobeno od teh razmer ni doseglo pomembnosti. Čeprav v tej študiji ni bilo nobenih kontrol za večkratno merjenje, je verjetno, da povečanje razpoložljivosti DAT ne bo zgolj artefakt ponovnega testiranja (glej Meyer in sod., 2002; Nurmi in sod., 2000).

V drugi študiji razpoložljivosti DAT v striatumu z uporabo računalniške tomografije z enim protonskim emisijam (SPECT) s Tc-99m TRODAT (Chou in sod., 2007) je bilo pet osamosvojljenih udeležencev MA ocenjeno na izhodiščni ravni (medtem ko je bila pozitivna na MA v analizi urina) in spet po 2 tednih abstinence. Po 1-tedenskem zamujanju je bila ponovno testirana tudi kontrolna skupina ( n = 7). Čeprav skupine niso bile statistično primerjane znotraj subjektov, se je razpolovna razpoložljivost DAT pri udeležencih MA povečala s 5 na 38% (povprečno 20%), medtem ko so se kontrolni udeleženci s ponovnim testiranjem spremenili iz -14 na 13%. WCST je bil uporabljen tudi na obeh preskusnih sejah, izboljšanje tega testa pa je bilo povezano s spremembo razpoložljivosti DAT. Vendar kontrolna skupina ni bila uporabljena za izračun učinkov prakse z večkratnim merjenjem, zato ni jasno, ali se povezava med DAT in WCST sčasoma izboljšuje v kognitivnih funkcijah, sposobnosti, da izkoristijo prejšnjo izpostavljenost preskusu ali oboje. Poleg tega dejstvo, da so bili udeleženci pozitivni na MA med prvo oceno SPECT, znatno oteži razlago sprememb DAT.

Wang in sod. (2004) so ​​izmerili presnovo cerebralne glukoze z uporabo [ 18 F] fluorodeoksiglukoze in PET pri petih udeležencih, odvisnih od MA, ki so bili testirani, ko so bili abstinentni za 1 leto abstinence. Čeprav ni bilo sprememb v absolutni regionalni presnovi glukoze, se je relativna presnova (regionalne vrednosti možganov normalizirala na globalne vrednosti) znatno podaljšala s podaljšano abstinenco pri talamu (+ 12%), ne pa tudi v striatumu ali okcipitalni skorji. V primerjavi z nevropsihološko zmogljivostjo, izmerjeno tudi v obeh časovnih točkah, je bila sprememba talamičnega metabolizma pozitivno povezana z izboljšanjem na testih časovne hoje, hitrosti obdelave in zapoznelega odpoklica (vendar ne s takojšnjim odpoklicem ali žlebičastimi ploščicami). Vendar ni bil uporabljen noben postopek za nadzor nad učinki ponovnega testiranja.

V drugi študiji presnove cerebralne glukoze z uporabo [ 18 F] fluorodeoksiglukoze in PET so ocenili 10 MA-odvisnih udeležencev po 5–9 dneh abstinence in ponovno po 4 tednih trajne abstinence (Berman in sod., 2007). V primerljivih časih je bilo testiranih dvanajst zdravih kontrolnih oseb. Compared with the control subjects, the MA participants had significantly greater increases in global glucose metabolism across testing sessions, with the largest increases in the bilateral parietal lobes, orbitofrontal cortex, insula, and cingulate gyrus. The authors also administered a test of auditory vigilance during both timeframes. The MA participants tended to have slower reaction times after a month of abstinence, while the control subjects were faster at retest. However, the group by reaction time interaction was not significant ( p =0.53). When reaction time was related to cerebral metabolic rate for glucose (CMRglc), reaction time slowing for MA subjects was significantly correlated with increased CMRglc in the left and right parietal lobes. In contrast, there was no relationship between reaction time and CMRglc in the control subjects.

In another study from the same laboratory, cerebral gray-matter volume of 12 MA-dependent participants was measured when they were abstinent for 4–7 days, and again after1 month of sustained abstinence (Morales et al, 2012). Twelve control subjects were also assessed at similar timeframes. Compared with the control participants, the MA participants showed significant increases in gray matter in the temporal gyrus, right angular gyrus, right insula, left precuneus, left inferior frontal gyrus, and left occipital pole (with decreased gray matter in the cerebellum). At the same statistical threshold, the control subjects did not show any significant changes in gray matter with retesting (see Figure 1). Repeat neuropsychological testing was not available for these participants.

Image

Changes in gray matter for methamphetamine (MA)-dependent individuals after 1 month of abstinence compared to control subjects re-tested after an 1 month interval. (a) Control subjects did not show any significant changes in gray matter with retesting. (b) MA-dependent participants show increases in gray matter with 1 month abstinence. Note: N =12 per group; statistical threshold: p 100 voxels.

Slika v polni velikosti

  • Prenesite diapozitiv PowerPoint

Zaključek

Longitudinal studies of abstinence show convincing changes in the neurochemical markers (ie, DAT, glucose metabolism) and gray-matter structure of MA-dependent subjects compared with control subjects who were tested at similar intervals. Although three of four studies that included cognitive data found relationships between changes in the brain and changes in cognitive performance, it is unclear whether the changes in cognitive performance reflect practice effects from repeated testing, longitudinal changes in cognitive function, or both. However, these findings do suggest that changes in the brain during abstinence may be linked with individual differences in cognition.

COGNITIVE DEFICITS IN MA-ABUSING INDIVIDUALS WILL BE DOSE-RELATED

If MA abuse causes cognitive decline in humans, individuals who are exposed to higher amounts of MA might exhibit greater cognitive deficits than those exposed to lower amounts of MA. Although an imprecise measure of cumulative dose, several studies have obtained self-reported duration of MA use (in years, months or days) as a proxy for cumulative MA exposure. One study found that years of MA use was associated with worse performance on the Stroop Task in MA-dependent adults (Salo et al, 2009). However, the vast majority of studies correlating duration of MA use with cognitive performance in MA-abusing participants (Chang et al, 2002; Henry et al, 2010; Hoffman et al, 2006; Iudicello et al, 2011; Johanson et al, 2006; Monterosso et al, 2005; Salo et al, 2005, 2011; Simon et al, 2000; Woods et al, 2005), including a meta-analysis which included MA-abusing participants from 17 different studies (Scott et al, 2007), found nonsignificant results. In addition, in the study of twins who had used stimulants weekly (Toomey et al, 2003), better performance on a few tests of dexterity and memory was correlated with greater total days of stimulant use.

Some studies have used self-reported frequency of MA use as an estimate of MA exposure. Simon et al (2000) found that MA abusers using more frequently (both in terms of days per week and times per day) performed worse than those using less frequently on tests of memory, abstract reasoning, and executive functioning. Another study (Henry et al, 2010) found that a functional measure of financial abilities was negatively related to the number of times MA-dependent participants used per month. However, in this same study, the six other functional abilities measured were unrelated to frequency of use. Likewise, Rippeth et al (2004) found no relationship between global cognition and daily vs less than daily MA use, and Price et al (2011) found that recent frequency of MA use was associated with better performance on the Grooved Pegboard Test by male MA-dependent individuals, but worse performance by females. Similar mixed findings have been observed for measures of addiction severity, with one study finding that increased severity of amphetamine addiction was associated with worse memory and attention (McKetin and Mattick, 1998), while another study found no relationship between cognition and addiction severity (Hoffman et al, 2006).

Perhaps the best measure of exposure to MA consists of reports of the total dosage consumed, either recently or estimated across the lifespan. Two studies found that the amount of recent MA use (grams per day or per week) was associated with worse motor response inhibition (Monterosso et al, 2005) and nonverbal reasoning (King et al, 2010) in MA-dependent or -abusing individuals (King and colleagues also found a negative relationship between executive functioning and 'lifetime METH use', but it is unclear how this was measured). In contrast, several other studies have not found a relationship between multiple cognitive measures and recent amount of MA consumed (Hoffman et al, 2006; Rippeth et al, 2004) or estimates of total lifetime quantity consumed (Chang et al, 2002; Cherner et al, 2010a; Henry et al, 2010; Iudicello et al, 2011; Woods et al, 2005).

All of the aforementioned studies evaluated relationships between parameters of MA use (often several parameters simultaneously) and cognitive function in post - hoc analyses, following up primary analyses that typically compared MA-abusing to healthy control subjects. To our knowledge, only one study has examined the relationship between parameters of MA use and cognitive function in a primary analysis. In that study (Cherner et al, 2010b), parameters of MA use for MA-dependent participants who were cognitively impaired (at least mild impairment in two or more cognitive domains based on demographically adjusted norms) were compared with those of MA-dependent participants who were not impaired on a comprehensive cognitive battery. The results revealed that the cognitively impaired and non-impaired groups did not differ on any index of MA use, including years of use, lifetime grams consumed, average grams used per year, length of abstinence, method of administration, bingeing pattern, or age of initiation.

Zaključek

The vast majority of research has not found a relationship between cognitive performance and duration of MA use. Findings from studies utilizing potentially more accurate measures of MA administration such as frequency of use or amount of recent use have been mixed, with the majority of studies not finding a relationship between cognition and estimates of cumulative lifetime dose. In addition, most of the statistically significant relationships cited previously were found in post-hoc analyses that analyzed the relationship between multiple MA use parameters and multiple cognitive tests, without consideration of Type I error rate or confounding variables. As such, the available evidence for a linear relationship between self-reported MA usage and cognitive performance is weak.

POTENTIAL MODERATORS OF THE RELATIONSHIP BETWEEN COGNITION AND MA ABUSE

Given that MA exposure does not appear to be linearly related to cognitive function in humans, but animal studies and other sources of data do suggest that MA causes cognitive decline in some individuals, authors have suggested that other factors may moderate the relationship between MA exposure and cognitive decline (Cherner et al, 2010b; Dean and London, 2010). Moderating variables may also help to explain variability in the cognitive performance of MA-dependent participants. For example, although some MA-dependent participants show significant impairment relative to demographic normative data, others perform well within expected levels (Dean and London, 2010). Likewise, animal studies show that not all animals exposed to MA decline in cognitive function, or decline to the same degree (Clark et al, 2007; Daberkow et al, 2005; Groman et al, 2012). The susceptibility to the development of cognitive decline may thus depend on moderating variables.

Starost

It is well known in the neuropsychological literature that similar neurological insults do not confer the same degree of cognitive decline in all individuals. In particular, the age at which the insult occurs can significantly affect the clinical outcome. Research suggests that, with other factors controlled, older individuals generally have a worse outcome from various types of neurological injury than younger individuals (Hukkelhoven et al, 2003; Lanzino et al, 1996; Luerssen et al, 1988; Weimar et al, 2004). Such findings have supported the hypothesis that younger individuals have greater brain reserve and neural plasticity to withstand neurological injury than older adults (for review, see Satz (1993)).

Studies in drug abuse have likewise suggested that the aging brain is less able to compensate for repeated drug exposure than the younger brain (see Dowling et al (2008)). For example, research suggests that alcohol abuse may have greater neurotoxic and cognitive effects in older individuals than younger individuals, particularly after the age of 50 (Rourke and Grant, 1999; Schottenbauer et al, 2007). Likewise, in animals, MA exposure has been shown to have a greater effect on markers of neuronal damage (Teuchert-Noodt and Dawirs, 1991) and extracellular dopamine levels (Bowyer et al, 1993) in older animals than younger animals. In humans, relatively little research has examined the interaction between age and MA abuse. The meta-analysis by Scott et al (2007) found that the effect size differences in cognitive performance between MA participants and control participants increased with increasing age, but it is unclear whether this effect was due to age effects alone, or interactions between age and MA use. However, these authors also noted that most of the participants in cognitive studies of MA abuse have been relatively young, with a mean age in the 30s or younger. To the extent that deficits in cognitive function become more pronounced with increasing age, the current literature may thus underestimate the lifetime prevalence of cognitive deficits in MA-abusing individuals.

Izobraževanje

Individuals with higher levels of education tend to have a better cognitive outcome in the context of a neurological injury than those with lower levels of education (Barnett et al, 2006; Elkins et al, 2006; Jones et al, 2006; Kesler et al, 2003; McDowell et al, 2007; Stern, 2002). Although brain reserve related to educational level has also been implicated as a protective factor in the development of cognitive deficits in substance use (Fein and Di Sclafani, 2004), to our knowledge, no research has examined the interaction between educational level and cognitive deficits in MA abuse.

Unlike the educational level attained by adults who suffer a neurological injury later in life, MA abuse typically begins during adolescence or young adulthood, at an age in which it is common for individuals to be in school. Thus, MA abuse and exposure to schooling may not be independent phenomena. Dean et al (2012) investigated the possibility that MA usage in young adulthood interferes with the amount of education one receives. These authors hypothesized the MA use in early adulthood may interfere with educational attainment, such that some MA-dependent individuals may have the requisite cognitive abilities to go further in school, but because of MA use (and/or its psychosocial correlates), this potential is cut short. Data supported this hypothesis: not only was age of first MA use negatively related to years of education attained ( r =−0.48, p <0.01), but also when the cognitive battery scores of MA-dependent participants were placed into a normative model of educational attainment, the MA participants had significantly fewer years of education than predicted by their cognitive scores. In addition, the discrepancy between actual and predicted educational levels of the MA participants was negatively correlated with the age of first MA use ( r =−0.42, p =0.01; see Figure 2), indicating that those who began to use MA at an earlier age had larger discrepancy between their real and predicted educational level. Importantly, these data suggest that the years of education attained by MA-dependent individuals tends to underestimate overall cognitive function. To the extent that this phenomenon generalizes to other MA users, this suggests that studies that match MA participants and healthy control participants for educational level may actually underestimate the degree of cognitive deficit present (see Dean et al (2012) for recommendations regarding this issue).

Image

Relationship between age of first use of methamphetamine (MA) and the difference between predicted vs actual years of education. Note: N =36 MA-dependent participants. Pearson correlation. Predicted education was predicted from cognitive battery scores and demographic characteristics (age, gender, ethnicity), using a regression model developed in healthy comparison subjects ( N =42). Positive difference scores indicate that predicted education was greater than actual years of education, whereas negative difference scores indicate that predicted education was less than actual years of education. Results show that larger (positive) difference scores were associated with a younger onset of MA use. Reprinted from Dean et al (2012) with permission from Elsevier.

Slika v polni velikosti

  • Prenesite diapozitiv PowerPoint

Genetika

As genetic makeup contributes significantly to cognitive function (Bouchard, 1998; Winterer and Goldman, 2003), some of the variation in the cognitive abilities of MA-abusing individuals reflects natural variation in the human genome. However, emerging data also suggest that genetic variability may moderate the effect of MA abuse on cognitive function.

Cherner et al (2010a) hypothesized that genetic variability in the metabolism of MA influences neurotoxicity and cognitive function in MA abusers. These authors tested MA-dependent participants for common functional variants of a gene that codes for cytochrome P450, family 2, subfamily D, polypeptide 6 (CYP2D6), which catalyzes hydroxylation and demethylation of MA. MA-dependent participants with two functional alleles coding for the enzyme (extensive metabolizers of MA) had significantly worse performance than MA-dependent participants with partially functional or nonfunctional alleles (intermediate and poor metabolizers of MA, respectively) in multiple cognitive domains. As extensive metabolizers had worse cognition than the other groups, the authors theorized that the metabolic byproducts of MA may be more neurotoxic than the parent compound itself.

DISKUSIJA

As chronic MA abuse cannot be manipulated experimentally in humans, the nature of the evidence regarding the causal relationship between MA abuse and cognition in humans is inherently indirect. In addition, the ability of the literature to elucidate the nature of the relationship between MA abuse and cognitive function is inevitably constrained by the quality and availability of the current state of the literature. In this respect, when evaluating the evidence related to our a priori hypotheses, we note that relatively few studies were available to draw conclusions for some of our hypotheses. Within the studies that were available, limitations were common and in part reflect natural weaknesses of quasi-experimental designs, which rely on the self-report and behavior of individuals with diverse life histories. Given these limitations, we believe that the evaluation of whether MA abuse causes cognitive decline in humans should be based on the preponderance of all the available evidence, rather than reference to one or more individual studies which do, or do not, support a particular viewpoint.

On the basis of the aforementioned perspective, we believe that the preponderance of the data suggests that MA abuse does cause mild cognitive decline in at least some individuals. We base this conclusion on the analysis of the evidence available for our six a priori hypotheses, which we summarize below:

(1) MA abuse will cause cognitive decline in animals, particularly when administered dosing regimens similar to human consumption. This hypothesis was supported. Chronic administration of doses of MA that are likely within the range of human consumption produce decline on various cognitive measures in animals. Although the use of escalating doses of MA can sometimes prevent cognitive decline otherwise observed by single moderate to high doses of MA, both experimenter and self-administered escalating dosing regimens have produced cognitive decline in rats and monkeys. As animal studies typically utilize experimental designs that can support causal inferences, evidence from these studies should likely be weighed more heavily than other forms of evidence in the overall evaluation of MA-induced cognitive decline.

(2) Individuals who abuse MA will have lower cognitive scores than well-matched individuals who do not abuse MA. This hypothesis was supported. Despite a few exceptions, the majority of studies that minimally controlled for age, gender, education, and premorbid IQ found that MA-dependent participants performed worse on at least some of the cognitive tests administered when compared with healthy participants who did not use drugs. However, because it is impossible to control fully for premorbid function and all possible confounds in cross-sectional studies, support for this hypothesis alone provides limited evidence regarding the causal relationship between MA abuse and cognitive function.

(3) Twins who abuse MA will have lower cognitive performance than their twin pairs who do not abuse MA. This hypothesis received mild support. In the only study available (Toomey et al, 2003), twins with a history of using stimulants on a weekly basis performed worse on four cognitive tests than their twin pairs who did not use stimulants (out of 14 tests), although this pattern was reversed on one test of attentional vigilance (stimulant users performed better than non-users on this test). However, it is unclear whether the weekly stimulant use endorsed in this study constitutes either dependence or abuse of stimulants based on psychiatric diagnostic criteria.

(4) Humans who abuse MA will show improvements in cognition with sustained abstinence. This hypothesis received mild support. Two of three cross-sectional studies suggested that MA-dependent participants who are abstinent for longer periods of time perform better on cognitive tests than those who are abstinent for shorter periods of time. However, the only relevant longitudinal study with a strong research design that examined 1 year of abstinence (Iudicello et al, 2010) found that only a small subset of MA-dependent individuals improve with abstinence. As a whole, the data suggest that some MA-dependent individuals do have a suppression of cognition that improves with abstinence. In addition, to the extent that decline from MA abuse is irreversible, evidence regarding this hypothesis cannot clarify whether MA abuse causes cognitive decline or not.

(5) MA abuse will be associated with changes in the human brain. This hypothesis was supported. Longitudinal studies show convincing changes in the neurochemical function (ie, DAT, glucose metabolism) and gray-matter structure of MA-dependent subjects during abstinence. These changes in the brain during abstinence are highly likely to be related to the brain alterations that occur as a direct result of MA use (eg, reflecting a compensatory reaction or recovery). Changes in the brain during abstinence have also been shown to be related to individual differences in cognitive function. Although it is unclear whether these changes in the brain during abstinence result in true cognitive improvements over time, the demonstration that the brain changes from MA abuse increases the odds that MA abuse also alters cognitive function to some degree.

(6) Cognitive deficits in MA-abusing individuals will be dose-related . Most data did not support this hypothesis. The majority of studies examining the relationship between cognitive function and duration of MA use, frequency of MA use, and/or total cumulative dose have not found significant relationships. Although some exceptions exist that have found MA usage parameters to be related to cognitive function, these findings have typically been exhibited in post-hoc analyses that examined the relationships between multiple MA use parameters and multiple cognitive tests, without consideration of Type I error rate or confounding factors.

Overall, evidence from five of our six a priori hypotheses received at least some degree of support for the notion that MA abuse causes cognitive decline. Animal studies, cross-sectional human studies, and brain studies provided fairly consistent indirect evidence for MA-induced cognitive decline. Studies of twins and cognition during abstinence provided weaker but supportive evidence, whereas studies of the dose–response relationship between MA consumption and cognition were generally not supportive. On the whole, the data support the perspective that, across individuals at the age of early-to-middle adulthood, MA abuse causes mild declines in cognitive function. These declines are likely to be observable in some, but not all, individuals who abuse MA. In addition, the extent of cognitive decline exhibited by a given individual is likely to vary depending on moderating variables such as age, educational level, and genotype.

The only hypothesis examined that received little empirical support was that relating the severity of cognitive deficits in humans to self-reported extent of MA use. It is possible that self-reported use history is too inaccurate to adequately quantify an individual's lifetime exposure to MA. In addition, given the consistent animal literature documenting the deleterious effects of binge-like dosing regimens, cumulative use may not be the best estimate of neurotoxic exposure to MA. Rather, it is possible that binge episodes differentially contribute to cognitive decline, perhaps particularly in individuals who have not yet developed considerable tolerance to the drug (Segal et al, 2003). Cherner et al (2010b) did not find a relationship between cognitive function and binge patterns in MA-dependent humans, but to our knowledge, this is the only study to have examined this issue.

In addition to challenges associated with self-reported cumulative MA use, it seems likely that MA-dependent individuals vary in their susceptibility to MA-induced cognitive decline. With other factors held constant, the literature on cognitive reserve suggests that older adults and those with less education are more susceptible to drug-induced decline than younger, more highly educated adults (Dowling et al, 2008; Satz, 1993). In addition, to the extent that drug use interferes with educational exposure, cognitive development and MA use may not be mutually exclusive phenomena (Dean et al, 2012). If cognitive development and MA use are interrelated, this may make it difficult to disentangle cognitive decline caused by MA abuse from a failure to develop particular cognitive skills (see Dean et al (2012)). Finally, if one considers that the average effect of MA abuse on cognition is relatively mild in middle adulthood (see below), it is perhaps not surprising that simple bivariate relationships between cognition and self-reported MA use would be difficult to detect in comparison to the large individual differences in cognition associated with variables such as genetic makeup (Bouchard, 1998; Winterer and Goldman, 2003), age, and educational attainment (Heaton et al, 1996). All of these factors suggest that simple linear models of MA abuse and cognition are unlikely to capture the dynamic interplay between MA abuse and a developing, adaptable organism.

As the evidence for MA-induced cognitive decline is at times ambiguous with a lack of unanimity, it may be tempting to conclude the MA abuse does not cause cognitive decline in humans. However, for this to be the case, the following conditions should be true: (a) human-like dosing of MA that causes cognitive decline in rats and monkeys does not have any cognitive effect in humans; (b) lowered cognitive scores of MA-abusing participants are entirely premorbid, despite efforts to account for premorbid function through group matching; (c) twin data on cognitive deficits in stimulant users are spurious; (d) improvements shown in cognition with abstinence are unrelated to MA use; and (e) changes in the brain likely caused by MA abuse are unrelated to cognition. Although it is possible that all of these conditions are true, we find it more reasonable to expect that MA abuse does cause at least mild cognitive decline in some people.

The notion that, on average, MA abuse causes mild rather than severe decline in cognitive function is supported by several lines of evidence (at the age of early-to-middle adulthood). First, in both animals and humans, not all studies which compared MA-exposed subjects to control subjects found cognitive deficits in the MA group (Chang et al, 2005; Clark et al, 2007; Grace et al, 2010; King et al, 2010; Leland et al, 2008; Simoes et al, 2007; Simon et al, 2010). We would suspect such null studies to be rare if MA caused severe cognitive decline. In addition, of those studies in which MA-dependent humans performed worse than control subjects, it is not uncommon for the groups to differ in one or less SD's of performance (based on data reported in the studies). According to some neuropsychological naming conventions (eg, Mitrushina et al, 2005), this would suggest that the mean performance of MA-dependent subjects falls in the low average to average range. This is consistent with studies that have referenced the cognitive performance of MA-dependent participants to published normative data (Gonzalez et al, 2004; Rippeth et al, 2004; Simon et al, 2010). Nonetheless, although MA dependence may be associated with mild cognitive decline across individuals , some susceptible individuals may exhibit considerable cognitive impairment relative to their demographically matched peers (eg, see Kalechstein et al (2003)).

As the normed cognitive performance of MA-dependent individuals frequently places outside the range of impairment according to some neuropsychological naming conventions, Hart et al (2011) concluded that the cognitive deficits of MA-dependent individuals are unlikely to be associated with functional implications. That is, because the deficits tend to be relatively mild, they are unlikely to have real-world implications. We do not share this categorical conceptualization of cognitive function. Rather, we suspect that mild deficits noted on cognitive tests likely relate to mild deficits in the cognitive construct being measured. Whether a specific cognitive deficit relates to 'real-world' impairment is likely to be complex and dependent on a number of factors, such as the social and environmental context in which the cognitive ability is needed, and the manner in which 'real-world' function is operationalized (eg, a mild deficit in short-term memory may affect memory for casual conversations, but have no observable relationship with household tasks or basic activities of daily living). Further, it should be noted that scores within the same normative range for those with different demographic characteristics are not necessarily equivalent in terms of the functional ability being measured. For example, low average performance for individuals with a high school degree likely relates to much lower functional ability than low average performance for individuals who graduated college (of which, MA-dependent individuals typically have lower educational attainment than their peers). For this reason, evidence suggests that scores normed to the average performance of healthy adults, without consideration of demographic characteristics (ie, 'absolute' scores), have a tighter relationship with everyday functioning than scores that are normed to the specific demographic characteristics of the individual (Silverberg and Millis, 2009).

Certainly, functional consequences are more likely the more that cognitive scores deviate from (absolute) normative expectations, and, as such, MA-dependent individuals with mild cognitive deficits may not have difficulties in some or all measures of everyday function. However, initial studies suggest that, despite the mild (mean) deficits described, MA-dependent individuals do have lower scores on performance measures of everyday functional ability than demographically similar control subjects (Henry et al, 2010). Similarly, MA-dependent participants with mild or greater deficits across cognitive tests are more likely to be unemployed than their cognitively intact peers (Weber et al, 2012). Given the limited data available and the complexity of the relationship between cognitive test scores and everyday function, more research is needed to ascertain the impact of cognitive performance deficits on real-world function in this population (for more information, see Marcotte and Grant, (2010)).

Zaključek

Although some findings suggest that MA abuse does not cause cognitive decline in humans, the preponderance of the evidence suggests that MA abuse does cause cognitive decline in at least some individuals. When averaged across individuals, this decline appears to be mild in early-to-middle adulthood, but moderator variables likely have a role in attenuating or exacerbating the degree of decline exhibited by a given individual. The ultimate validity of our current conclusions will depend on replications and improved future research that hopefully will provide greater leverage to address the causal relationship between MA abuse and cognitive function.