In 2008, an estimated 1.4 million Americans met the criteria for cocaine abuse/dependence in the past 12 months (National Institute on Drug Abuse, 2010). Excessive and repeated cocaine consumption poses a major public health concern, as it is associated with psychological (eg, panic, mania, dysphoria, and paranoia) and social (eg, crime) consequences and has the potential of provoking physiological problems, including cardiovascular and neurological complications and human immunodeficiency virus (National Institute on Drug Abuse, 2010; Schwartz et al., 2010). Cocaine is a potent dopamine transporter inhibitor, which has been shown, in animals, to acutely increase dopamine release in the mesocorticolimbic brain reward system, and conversely, to cause hypodopaminergic activity in the prefrontal cortex when administered chronically (Volkow et al., 1997; Dackis and O'Brien, 2001; Goldstein and Volkow, 2011). Consistently, cocaine use disorder (CUD; abuse/dependence) has been shown, albeit inconsistently, to be associated with deficits in attention, working memory, problem solving, and impulse-related cognitive functions (Moeller et al., 2005; Kalapatapu et al., 2011; Madoz-Gurpide et al., 2011; Albein-Urios et al., 2012). In addition, CUD has been regularly associated with episodic memory deficits and, less consistently, with impaired visuospatial abilities (van Gorp et al., 1999; Selby and Azrin, 1998; Simon et al., 2002; Cunha et al., 2004; Sanchez de Leon et al., 2009). However, it remains to be determined which of these cognitive functions are the most prominently compromised by cocaine. Finally, there is a lack of consensus on the length of the residual cognitive effects of cocaine persisting after acute withdrawal (Beatty et al., 1995; Roselli and Ardila, 1996; Roselli et al., 2001; Browndyke et al., 2004; Van der Plas et al., 2009).
A previous meta-analysis (Jovanovski et al., 2005) attempted to demystify the relationship between cocaine and cognitive dysfunction. Results from this meta-analysis revealed greatest impairment on specific subscores from various tests measuring attention, visual memory, and working memory. Minimal dysfunction was found among tests assessing verbal fluency and visuospatial functions. Finally, tests of executive function revealed mixed findings. Although these findings clearly helped clarify how cocaine affects cognitive function, a number of methodological limitations from this meta-analysis make it difficult to draw definite conclusions on the relationship between cocaine and cognition.
The above-mentioned meta-analysis thoroughly assessed cognition by calculating how different subscores from various neuropsychological tests were affected by cocaine. However, by regrouping studies based on test subscores instead of regrouping studies based on cognitive domains (eg, executive functions, attention, working memory, etc), the analysis produced an abundant number of effect-size estimates based on a limited number of studies. With the publication of numerous studies on the cognitive consequences of cocaine since 2005, it is now feasible to regroup neuropsychological subscores into cognitive domains, an approach that may be methodologically advantageous, as it confers greater statistical power. Another methodological limitation of the previous meta-analysis on cocaine is the lack of consideration for length of abstinence. The analysis included cocaine abusers with a wide range of abstinence duration, varying from zero days to many months. Comparing subjects with variable abstinence durations on neuropsychological test performance makes it difficult to interpret study findings, particularly in the case of cognitive processes associated with small effect-size estimates, as these results may be due to performance normalization after long-term abstinence. A recent meta-analysis on alcohol and cognition by our group showed that the partitioning of studies based on length of abstinence is a valid approach that helps reducing heterogeneity, and that the length of abstinence can greatly affect cognitive performance, with greater abstinence duration revealing amelioration of cognitive function (Stavro et al., 2013). Last, the previous cocaine meta-analysis did not include studies that were primarily designed to assess impulse- or reward-related functions in cocaine abusers by using cognitive tasks measuring response inhibition, decision-making, or delayed-reward discounting.
This current meta-analysis aimed to clarify the relationship between cocaine and cognitive dysfunction by performing a systematic quantitative review of the literature. Particular focus was placed on conducting an analysis based on cognitive domains rather than individual cognitive test subscores, on taking into consideration the effects of abstinence duration on cognitive dysfunction, and honing in on the relationship between cocaine and impulse-/reward-related functions, as measured with cognitive tasks.