Heavy alcohol use in adolescence is linked to disruptions in white matter integrity, new research suggests.
In a case control study of more than 400 participants, the association was more pronounced in younger adolescents and in the anterior and middle corpus callosum, which serve the interhemispheric integration of frontal networking and communication.
The results provide clinicians with yet another reason to ask adolescents about their alcohol use, said investigator Adolf Pfefferbaum, MD, Center for Health Sciences, SRI International, Menlo Park, California, and professor emeritus at Stanford University School of Medicine.
However, when questioning adolescents about their alcohol use, “sometimes it’s better to ask ‘How much alcohol do you drink?’ ” instead of just asking if they drink, Pfefferbaum told Medscape Medical News. That’s because they may be more willing to answer the first question honestly.
It’s also important for clinicians to nonjudgmentally tell teens there is evidence “that heavy drinking is bad for their brain,” he added.
The findings were published online December 30 in JAMA Psychiatry.
Adolescence is a critical period of physiologic and social maturation accompanied by significant structural, functional, and neurochemical brain changes, the investigators note.
Diffusion tensor imaging (DTI) produces a measure called fractional anisotropy (FA), which characterizes some of these brain changes by measuring molecular water diffusion in the brain.
“FA is a measure of the integrity of brain white matter; so, the part of the brain that connects neurons with each other,” Pfefferbaum said. He added that FA decreases in diseases such as multiple sclerosis (MS), reflecting “some kind of pathology.”
Affected fiber systems include the corpus callosum, superior longitudinal fasciculus, internal and external capsule, brainstem, and cortical projection fibers. Disruption of these neural systems may degrade neural signal transmission and affect certain cognitive functions, possibly resulting in enhanced impulsivity, poor inhibitory control, and restricted working memory capacity, the researchers write.
FA follows an inverted U-shaped pattern. “The natural trajectory is to increase from infancy up to middle adolescence and then, as we get older, from about age 25 to 30 years, starts to go down. Our brains are starting to show signs of aging a bit by then,” said Pfefferbaum.
The current analysis assessed 451 adolescents (228 boys and 223 girls) from the NCANDA study, for whom researchers had four years of longitudinal DTI data. All were age 12 to 21 years at baseline.
The NCANDA cohort was recruited across five US sites. Participants are assessed yearly on psychobiologic measures, including brain maturation. The cohort, which did not have any significant substance abuse upon entry, is balanced in terms of gender and ethnicity.
The investigators quantified the developmental change of white matter (WM) integrity within each individual as the slope of FA over visits. They also examined altered developmental trajectories associated with drinking onset during adolescence and the differential alcohol associations by age with specific regional WM fiber tracts.
Researchers assessed drinking on a scale of 1 to 4, based on the youth-adjusted Cahalan score. The scale considers quantity and frequency to classify drinking levels based on past-year self-reported patterns.
Results showed that 291 participants (37.2%) remained at no-to-low drinking levels (youth-adjusted Cahalan score, 0) throughout the time points examined, and 160 (20.5%) were classified as heavy drinkers for at least two consecutive visits (youth-adjusted Cahalan score > 1).
Among the no-to-low drinkers, 48.4% were boys with a mean age of 16.5 years and 51.2% were girls with a mean age of 16.5 years. About two thirds of the group (66%) were White.
Among heavy drinkers, 53.8% were boys with a mean age of 20.1 years and 46.3% were girls with a mean age of 20.5 years. In this group, 88.8% were White.
Investigators did not analyze moderate drinkers or those who initiated heavy drinking for only one visit.
The findings also showed that heavy drinkers exhibited significant reduction of whole-brain FA. The slopes of the 78 heavy drinkers were significantly more negative than the 78 matched no-to-low drinkers (mean, –0.0013 vs 0.0001; P = .008).
“The concept of the slopes is really important here because it’s the trajectory that seems to be the most sensitive measure,” Pfefferbaum said. “Probably what’s happening is the exposure to alcohol is interfering with the normal myelination and normal development of the adolescent’s white matter.”
The no-to-low drinkers had relatively stable FA measures across all visits.
A reduction in FA was significantly linked to heavy drinking. An analysis of 63 youth who transitioned from being a no-to-low drinker to a heavy drinker showed that before the transition, they had significantly increased FA over visits (95% CI of slope, 0.0011 – 0.0024; P < .001). In addition, their corresponding slopes were not different from other no-to-low drinkers of the same age range.
However, this group’s FA declined significantly after they reported heavy drinking, resulting in slopes significantly below zero (95% CI of slope, –0.0036 to –0.0014; P < .001) and that were lower than the no-to-low participants of the same age range.
The findings of this unique before-and-after assessment “bring us one step closer to causality,” and further illustrates that heavy drinking in adolescence affects WM integrity, Pfefferbaum said.
None of the slope measures correlated with number of visits or use of tobacco or cannabis. The association of alcohol with the slope measures was more apparent in the younger cohort (< 19 years).
“The effects were seen more readily in younger adolescents because they are the ones who are still progressing along this normal developmental trajectory,” Pfefferbaum noted. “In a sense, the younger you are when you’re exposed to alcohol, probably the more vulnerable you are.”
Previous studies have suggested that damage in WM tracts is associated with heightened neural reactivity to alcohol cues in adults with alcohol use disorder. Given this evidence, the greater WM degradation at younger vs older ages might help explain why adolescents who initiate early drinking are more likely to develop addiction later in life, the investigators write.
Of the five major fiber tracts, only the commissural fibers (corpus callosum) showed a significant association with alcohol. The researchers note that white matter volume shrinkage and callosal demyelination are two of the most prominent markers in adult alcoholism and are potential markers in adolescent alcohol abuse.
Upon further extending the analysis to the four sub-regions of the corpus callosum, the investigators found that only the anterior and middle callosal regions (genu and body) showed significant age-alcohol interactions.
This could be a result of the timing of fiber myelination in these regions of the brain compared to others, Pfefferbaum said.
He noted that these fibers connect the left and right part of the anterior regions of the brain, especially the frontal lobes, which are particularly vulnerable to the effects of alcohol. “It may well be that we have this interaction of the developmental time and the sensitivity of the frontal parts of the brain,” he added.
Although the researchers did not find any sex effects, Pfefferbaum stressed that this doesn’t mean they do not exist. “We just may not have the power to see them,” he said.
The study did not look specifically at binge drinkers, defined as consuming five drinks in 2 hours for men and four drinks in 2 hours for women. Pfefferbaum noted that it is difficult to get “good quantification” of binge drinking. “We don’t have a fine enough grain analysis to separate that out,” he said.
Asked whether the altered FA trajectory in heavy drinkers affects cognition, Pfefferbaum said “those studies are still in progress,” with results hopefully available within about a year.
Pfefferbaum said he and his colleagues are continuing to follow these adolescents and hope to see if the altered FA trajectory in heavy drinkers returns to normal, adding: “The real question now is: If they stop heavy drinking, will they get back on track?”
This study is believed to be the first to suggest in vivo differential vulnerability in WM microstructure with respect to age, the authors note.
In addition to asking teens about their alcohol use, the clinician’s role should be to “counsel and refer,” said Pfefferbaum. He also suggested accessing resources from the National Institute on Alcohol Abuse and Alcoholism.
Important Data, but Several Limitations
Commenting for Medscape Medical News, Oscar G. Bukstein, MD, MPH, medical director of outpatient psychiatry service at Boston Children’s Hospital, and professor of psychiatry at Harvard Medical School, said the findings provide further evidence that alcohol affects the maturing brain.
This study, and others that have examined cannabis use, “show that you have a dynamically growing brain with certain sections, particularly in this case the anterior and middle corpus callosum, that mature later [and] that are more likely to be affected by early alcohol use,” said Bukstein, who was not involved with the research.
He stressed the importance of determining the mechanism involved and noted some study limitations. For example, the DTI technology used may “already be out of date,” he said.
Using older technology may have prevented finding an impact of heavy drinking on parts of the brain other than the anterior and middle corpus callosum, Bukstein noted.
Newer technology might provide “a finer-grain nonlinear voxel-wise analysis,” although using more updated scanning techniques may not have detected additional differences in study groups, he added.
Bukstein also noted that there were limitations: The study did not have “gradations,” but only looked at heavy drinking and no-to-low drinking. “You’d like to find out about kids who are somewhere in the middle.” It also didn’t determine a “cutoff” where deleterious effects of alcohol on the brain begin, Bukstein added.
Additionally, the study didn’t look at brain development outcomes in children with conditions such as depression and ADHD that are known to lead to substance use — something a larger study may have been able to do, he said.
Bukstein noted that a newer and much larger study, the Adolescent Brain Cognitive Development (ABCD) study, has begun assessing kids for risk factors such as substance use, starting at age 10 years.
The study was funded by grants from the US National Institute on Alcohol Abuse and Alcoholism (NIAAA), and by the National Institute on Drug Abuse, the National Institute of Mental Health, the National Institute of Child Health and Human Development, and the Stanford Institute for Human-Centered Artificial Intelligence-AWS Cloud Credits for Research. Pfefferbaum reported having received an NIAAA grant during the conduct of the study. Bukstein has disclosed no relevant financial relationships.
JAMA Psychiatry. Published online December 30, 2020. Abstract