A longitudinal twin study of effects of adolescent alcohol abuse on the neurophysiology of attention and orienting

Longitudinal Functional Brain effects resulting from adolescent alcohol abuse are difficult to definitively document. But there is evidence that the adolescent brain may be more vulnerable to the effects of addictive substances, because extensive neuromaturational processes are occurring during this developmental period (Clark et al., 2008; Lubman et al., 2007). Brain constituents actively developing during adolescence include the prefrontal cortex, limbic system areas, and white matter myelin. Deficits or developmental delays in these structures and their functions are suggestive of disrupted developmental trajectories in early-onset substance users, although there is growing evidence that high-risk youths have premorbid neurobiological vulnerabilities (Clark et al., 2008; Lubman et al., 2007). One magnetic resonance imaging (MRI) study of adolescent-onset alcohol users suggested that the total hippocampal volume correlates positively with the age at onset and negatively with the duration of the alcohol use disorder (De Bellis et al., 2000). Another MRI study (De Bellis et al., 2005) found a smaller prefrontal cortex associated with early-onset drinking. Genetic factors contribute to individual differences in trajectories of adolescent drinking (Jackson et al., 2009; Madden et al., 2000; McGue et al., 2001; Rose et al., 2001; Viken et al., 2007) to the development of adolescent-onset alcohol disorders (Pagan et al., 2006) and to neurophysiological abnormalities in adolescents at elevated risk for alcoholism (Begleiter et al., 1984; Hill et al., 2000, 2001). Early initiation of alcohol use is robustly predictive of later alcoholism disorder (DeWit et al., 2000) and to certain attentional deficits observed in adult alcoholics (Ahveninen et al., 2000). Grant and colleagues (2006) found that the genetic factors influencing early-onset regular drinking also contribute to later alcohol dependence and drug abuse, but the association was not entirely explained by shared genes (nor by shared family environments). Effects of alcohol on brain function can be studied by using auditory event-related potentials (ERPs), stimulus-averaged electroencephalogram (EEG) epochs. The ERP component P3, elicited by targets or by unexpected “deviants” embedded within a train of repetitive nontarget stimuli, has been employed extensively in alcohol research. Both auditory and visual studies suggest that P3 is reduced and/ or delayed in chronic alcoholics (Porjesz and Begleiter, 1997). But the interpretation of such evidence is uncertain: previous family (Begleiter et al., 1987; Patterson et al., 1987; Perlman et al., 2009; Pfefferbaum et al., 1991) and twin studies (Carlson et al., 2002; Perlman et al., 2009; Yoon et al., 2006) suggest that these P3 phenomena could be genetic markers of vulnerability to alcohol abuse, rather than consequences of it. That interpretation is supported by evidence (Almasy et al., 1999; Katsanis et al., 2007; O'Connor et al., 1994; Wright et al., 2001) that P3 amplitude is moderately heritable, with heritability estimates ranging from 39 to 79%, and by analogous evidence of inheritance of P3 latency (Almasy et al., 1999; Polich and Burns, 1987; Rogers and Deary, 1991). A meta-analysis by van Beijsterveldt and van Baal (2002) found meta-heritability estimates of 60% for P3 amplitude and 51% for P3 latency. And longitudinal twin studies demonstrate familial factors substantially contribute to the age-to-age phenotypic stability of P3 across 18 months of adolescence (van Beijsterveldt et al., 2001). P3 consists of 2 major subcomponents: novel-sound P3 (P3a), a subcomponent thought to be associated with involuntary attention switching to stimulus changes (Escera et al., 2000), with further study suggesting it is a complex signal that comprises alerting, orienting, and executive control processes triggered by an unexpected stimulus (SanMiguel et al., 2010). P3a is thought to originate from frontal areas where stimulus-driven disruption of attention arouses activation related to dopaminergic processes (Polich, 2007; Polich and Criado, 2006). Target-tone P3 (P3b), the other subcomponent, putatively reflects conscious stimulus evaluation, target detection, and working memory functions, and is thought to originate from temporal-parietal activity related to norepinephrine processes (Polich, 2007; Polich and Criado, 2006; Polich and Herbst, 2000). Both P3a and P3b are diminished and delayed in chronic alcoholics (Hada et al., 2000; Porjesz and Begleiter, 1997), but these abnormalities could be markers of inherited risk for alcoholism, as well (Almasy et al., 1999; O'Connor et al., 1994; Wright et al., 2001). P3a and P3b variations are hypothesized to relate how alcohol and other substance use affects differential neurotransmitter levels within and between individuals (Polich and Criado, 2006). High-risk drinking patterns and alcohol abuse are associated with other types of substance use and abuse, especially tobacco use and nicotine dependence. Alcohol dependence and regular tobacco use share genetic covariance, even when sociodemographic and personality variables, as well as histories of other psychopathologies are taken into account (Bierut et al., 2000; Li et al., 2007; Madden et al., 2000). Furthermore, the joint trajectories of smoking and drinking across adolescence appear to be more heritable than either one alone (Jackson et al., 2009). Accordingly, P3 measures of alcoholism risk in young adults may be modulated by concurrent smoking status. Studies of young adults suggest that the reduction of visual P3 amplitude in those at elevated risk for alcoholism is enhanced by current smoking patterns (Anokhin et al., 2000; Polich and Ochoa, 2004). Moreover, in these studies tobacco smoking accounted for a larger proportion of the total variance than did alcoholism risk per se. Alcohol abuse and dependence in adolescence is often associated with behavioral psychopathology: behavioral disorders that commonly co-occur with alcohol use disorders in adolescents include conduct disorders and antisocial personality, and mood and anxiety disorders. Behavioral disorders may both precipitate alcohol use disorders and result from them (Clark and Bukstein, 1998). Here, we addressed the longitudinal effects of adolescent alcohol abuse on brain function in co-twins highly concordant and discordant for their self-reported alcohol problems in late adolescence. Alcohol use was assessed on 4 occasions between ages 16 and 25. We hypothesized that even when genetic and common environmental factors are controlled, a correlation between alcohol abuse and P3 measures will remain, and we tested that hypothesis in both between-family and within-family associations.