Severity of neuropsychological impairment in cocaine and alcohol addiction: association with metabolism in the prefrontal cortex

https://doi.org/10.1016/j.neuropsychologia.2004.04.002Get rights and content

Abstract

We used exploratory and confirmatory statistical approaches to study the severity of neuropsychological (NP) impairment in 42 crack/cocaine addicted subjects and in 112 comparison subjects (40 alcoholics and 72 controls). Twenty neuropsychological test indices most reliably defining predetermined cognitive domains were submitted to exploratory factor analysis. A four-dimensional model of neurocognitive function was derived: Verbal Knowledge, Visual Memory, Verbal Memory, and Attention/Executive functioning accounted for 63% of the variance. We then examined this model’s association with resting glucose metabolism in the brain reward circuit measured with 2-deoxy-2[18F]fluoro-d-glucose positron emission tomography. Results revealed that (1) cocaine addicted individuals had a generalized mild level of neurocognitive impairment (<1 S.D. below control mean); and (2) controlling for age and education, relative metabolism in the dorsolateral prefrontal cortex significantly predicted the Visual Memory and Verbal Memory factors and relative metabolism in the anterior cingulate gyrus significantly predicted the Attention/Executive factor. Nevertheless, it remains to be determined whether metabolic changes in these regions are associated with addiction. Our results also suggest that compared to cocaine, alcohol has a more detrimental effect on Attention/Executive functioning, as assessed with traditional NP measures. We conclude that relative to other psychopathological disorders (such as schizophrenia), the severity of neuropsychological impairment in cocaine addiction is modest, albeit not indicative of the absence of neurocognitive dysfunction. The impact of such small differences in performance on quality of life, and possibly on craving and relapse, may be substantial. Tasks that simulate real-life decision-making or that target specific putative cognitive-behavioral or motivational–emotional mechanisms might offer greater sensitivity in characterizing the changes that accompany addiction to drugs. Obtaining valid estimates of alcohol use in cocaine addicted subjects is essential in characterizing neurocognitive functioning in individuals addicted to drugs.

Introduction

Despite the growth of the research on the cognitive deficits in cocaine abusers (reviewed in Rogers and Robbins, 2001), the nature of these deficits is uncertain and the study of their putative neuropathological mechanisms is still in its infancy. Variability in research findings comparing cocaine users to non-users contributes to the delay in establishing a consensus on the magnitude and pattern of the neuropsychological (NP) deficits in cocaine addiction. Thus, although significant decrements are revealed in many of the NP studies (e.g., Beatty, Katzung, Moreland, & Nixon, 1995; Gillen et al., 1998; Rosselli & Ardila, 1996), lack of significant differences between cocaine users and non-users are common (e.g., Bolla, Rothman, & Rothman, 1999; Selby & Azrin, 1998) and counterintuitive results, where cocaine users outperform controls, are also frequently reported (e.g., Bolla et al., 1999, Gillen et al., 1998, Hoff et al., 1996, van Gorp et al., 1999).

Characterizing the underlying cognitive domains instead of analyzing multiple single test indices of cognitive functioning, might minimize some of this variability. Factor analytic techniques are ideal for reducing the redundant information contained within multiple single measures into a few separate dimensions. This is important because most tests in common clinical NP use place simultaneous demands on several cognitive functions. Analyzing this smaller set of variables would also increase sample-to-variable ratio, thereby reducing chance findings while preserving the measured content.

The analysis of cognitive domains in drug addiction has been previously undertaken in our laboratory (Hoff et al., 1996) and by others (e.g., Beatty et al., 1995; Di Sclafani, Tolou-Shams, Price, & Fein, 2002; Gillen et al., 1998; Robinson, Heaton, & O’Malley, 1999; Selby & Azrin, 1998). However, factor analytic techniques were not employed and no effort was made to empirically test the a priori assignment of test measures into the predetermined cognitive clusters.

Moreover, the relationship of these cognitive domains with brain function (e.g., Hoff et al., 1996, Holman et al., 1991; Mittenberg & Motta, 1993; Strickland et al., 1993) has not been determined. Associations between cerebral perfusion and neurocognitive performance in drug addicted individuals have been previously deducted from the co-occurrence of perfusion abnormalities and NP impairments in the same individuals (see for example Gottschalk, Beauvais, Hart, & Kosten, 2001; Holman et al., 1991, Strickland et al., 1993). When the analysis of correlations between functional measures of brain and behavior was conducted, sample sizes were limited (Wang et al., 1993), specific cognitive domains were not examined (Woods et al., 1991), or regional brain measures were not used (e.g., Di Sclafani et al., 1998).

In the present study, 20 NP test indices most reliably defining six predetermined cognitive domains in 42 crack/cocaine addicted subjects and 72 comparison subjects were submitted to exploratory factor analysis. A group of 40 alcoholics was additionally included for comparison and to increase statistical power. We then examined the association of the empirically (factor-analytically) derived four cognitive scales with measures of regional cerebral glucose metabolism obtained at resting baseline using positron emission tomography with 2-deoxy-2[18F]fluoro-d-glucose (PET 18FDG). Six regions of interest (ROIs) were selected for these correlation analyses: the orbitofrontal gyrus, anterior cingulate gyrus, dorsolateral prefrontal cortex, hippocampus, thalamus, and basal ganglia. These regions comprise the brain reward circuit most frequently implicated in the neurobiology of drug addiction (see Goldstein & Volkow, 2002; Volkow & Fowler, 2000).

Section snippets

Subjects

A comprehensive battery of NP tests was administered to 42 cocaine addicted subjects, 40 alcoholics, and 72 comparison subjects who participated in PET studies at Brookhaven National Laboratory between 1988 and 1996. The NP functioning of subgroups of the current study’s population was previously described (e.g., Hoff et al., 1996, Wang et al., 1993). The NP battery was administered a mean of 60 days (range −13–1466 days; 90% of the subjects were tested within 6 months and 98% within 1 year of

Demographics, drug use, and PET 18FDG measures

Means and standard deviations for demographics, drug use variables, and measures of regional absolute and relative metabolism in the reward ROIs for the three study groups are presented in Table 1. There were significant differences between the cocaine addicted subjects and comparison subjects in distribution of sex and handedness, and in age and education. For the alcohol group, there were significant differences from the comparison subjects in distribution of handedness, in education, and in

Discussion

The main objective of this study was to characterize the nature of the NP impairment in crack/cocaine addicted subjects using a group of alcoholics and non-addicted subjects for comparison. The present study: (1) developed a four-dimensional model of neurocognitive functioning from multiple standard NP test indices in 42 crack/cocaine addicted subjects, 40 alcoholics, and 72 comparison subjects participating in PET neuroimaging studies at Brookhaven National Laboratory between 1988 and 1996;

Conclusions

In conclusion, this study documented a relatively mild NP impairment in cocaine addiction which should not be misinterpreted as indicative of the absence of neurocognitive dysfunction in this group. Demonstrating clear associations between cognitive functioning as assessed by NP testing and neuronal functioning as estimated by PET 18FDG, we furthered the fledgling research on the brain–behavior relationship in drug addiction.

Acknowledgements

This study was supported by grants from the National Institute on Drug Abuse (to NDV: DA06891-06; and to RZG: 1K23 DA15517-01), US Department of Energy (OBER), National Institute on Alcohol Abuse and Alcoholism (AA/ODO9481-04), and ONDCP.

References (43)

  • K. Rubia et al.

    Functional frontalisation with age: Mapping neurodevelopmental trajectories with fMRI

    Neuroscience and Biobehavioral Reviews

    (2000)
  • M.J. Selby et al.

    Neuropsychological functioning in drug abusers

    Drug and Alcohol Dependence

    (1998)
  • N.D. Volkow et al.

    Role of dopamine and the frontal cortex in drug addiction in humans

    Neurobiology of Learning and Memory

    (2002)
  • S.W. Woods et al.

    SPECT regional cerebral blood flow and neuropsychological testing in non-demented HIV-positive drug abusers: Preliminary results

    Progress in Neuro-psychopharmacology & Biological Psychiatry

    (1991)
  • E.J. Bartlett et al.

    Stability of resting deoxyglucose metabolic values in PET studies of schizophrenia

    Psychiatry Research

    (1991)
  • R.M. Bilder et al.

    Neuropsychology of first-episode schizophrenia: Initial characterization and clinical correlates

    American Journal of Psychiatry

    (2000)
  • K.I. Bolla et al.

    Dose-related neurobehavioral effects of chronic cocaine use

    Journal of Neuropsychiatry and Clinical Neurosciences

    (1999)
  • R. Cabeza et al.

    Imaging cognition II: An empirical review of 275 PET and fMRI studies

    Journal of Cognitive Neuroscience

    (2000)
  • C.S. Carter et al.

    Anterior cingulate cortex, error detection, and the online monitoring of performance

    Science

    (1998)
  • C.S. Carter et al.

    Parsing executive processes: Strategic vs. evaluative functions of the anterior cingulate cortex

    Proceedings of the National Academy of Sciences of the United States of America

    (2000)
  • B.J. Casey et al.

    The role of the anterior cingulate in automatic and controlled processes: A developmental neuroanatomical study

    Developmental Psychobiology

    (1997)
  • Cited by (294)

    View all citing articles on Scopus
    View full text