The primary finding in this study was a similar reduction in prefrontal gray matter volume in the CrA and Cr subjects. The existence of prefrontal volume reduction at 6-week abstinence establishes that it is not a result of acute cocaine exposure. The negative association of prefrontal cortical volume with performance on frontal cortex mediated abilities in the substance dependent samples suggests that the prefrontal volume deficits have functional consequences. Consistent with our a priori hypothesis, Cr dependence is associated with structural and functional damage to the prefrontal cortex. Contrary to our a priori hypothesis, however, white matter signal hyperintensity was not increased in the substance dependent samples.
There are only a few studies that examine regional volume changes in cocaine abusers. Liu et al. (1998)
reported findings that mirror the results presented here. They studied 25 male polydrug abusers (age 21–44) compared to 14 normal male controls of comparable age. Twenty-three of the 25 individuals in the substance abuse group indicated that cocaine was their drug of choice. All subjects were housed on a closed research ward where at least 15 days abstinence from drugs and alcohol was documented. The substance dependent sample showed reduced normalized gray matter volume in the prefrontal lobe only.
Bartzokis et al. (2000)
did not find a difference in frontal cortical volume in their study of age-related brain volume reductions in amphetamine and cocaine addicts compared to normal controls. The cocaine dependent sample (n
= 10) showed a reduced temporal lobe volume and a negative age versus temporal gray matter association in comparison to the amphetamine (n
= 9) and normal control (n
= 16) samples. The disagreement in our results for the frontal cortex may reflect differing anatomical definitions of frontal versus prefrontal lobe. In addition, they did not address recent drug use in their samples (which consisted of recent admissions to inpatient and outpatient treatment centers); thus acute effects of cocaine or other drugs could have contributed to their findings.
We fully expected that concomitant cocaine and alcohol dependence would be associated with greater structural brain abnormalities than cocaine dependence alone, but such was not the case. However, O’Neill et al. (2001)
had a similar result in their study of 16-week abstinent individuals dependent on cocaine only, alcohol only, and cocaine and alcohol. They found smaller normalized prefrontal gray matter volumes on quantified structural MRIs in the combined cocaine dependent samples. Within the prefrontal gray matter, significant volume loss was localized to the dorsolateral prefrontal and lateral prefrontal regions. Similar to our study, O’Neill et al., did not find a difference in frontal cortical volume between the sample dependent on cocaine only and the sample dependent on both cocaine and alcohol.
Impairments in prefrontal lobe mediated abilities have been noted in studies of alcohol abuse (Oscar-Berman, 2000
), and in the scant literature on cocaine abuse (Di Sclafani et al., in press). MicroCog incorporates an Analogies and an Object Match test in the executive function domain. The Analogies test is based on Luria’s (1980)
argument that patients with frontal lesions are impaired in the ability to transfer meaning (such as in the understanding and use of analogies, proverbs, metaphors, etc.). The Object Match test requires subjects to group the same objects according to different principles. Individuals with prefrontal lobe damage find it difficult to switch mental set once a concept has been derived, which may reflect rigidity, inflexibility, or perseveration (traits characteristic of substance abusers, particularly perseveration). In light of the documented effects of both chronic cocaine dependence and chronic alcohol dependence on pre-frontal lobe mediated abilities, it is surprising that concomitant cocaine and alcohol dependence in this sample was not associated with greater neuropsychological impairments on these prefrontal tests. However, the data consistently fail to indicate even additive effects of cocaine and alcohol abuse on frontal mediated functional impairments. This is consistent with our failure to find greater prefrontal cortical volume deficits in the subjects dependent on both cocaine and alcohol versus those dependent on cocaine only.
A cautionary note must be sounded vis-a-vis the attribution of abnormal brain structure and function to substance exposure. There is evidence of premorbid frontal lobe abnormalities in individuals at risk for substance abuse that may be associated with the predisposition to substance abuse. These structural and functional abnormalities are actually premorbid and comorbid, in that they are also associated with the presence of comorbid mood and externalizing disorder symptoms and traits (e.g. Berman and Noble, 1995
; Hesselbrock et al., 1991
; Pihl and Peterson, 1991
; Poon et al., 2000
). Moreover, the comorbidity of psychiatric disorders and substance use is substantial. Therefore, we cannot rule out a contribution from premorbid (and comorbid) differences to the prefrontal cortical volume reduction in the substance abusers versus normal controls.
Finally, we were surprised that we did not find an increase in the volume of white matter signal hyperintensities in the substance dependent samples. It may be that the relatively young age of our substance dependent sample was responsible for our failure to find evidence for ischemic brain disease in this group.
In summary, the current study suggests that chronic Cr dependence is associated with prefrontal cortex volume reductions that are present at 6-week abstinence (and thus are not acute effects of cocaine use). Comorbid alcohol dependence had no additional effect on prefrontal cortical volumes or on neuropsychological deficits in cocaine dependent individuals. The prefrontal cortex volume reductions are associated with poorer performance on tests of executive function, abilities known to be mediated by the frontal lobe.