Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Nov 1.
Published in final edited form as: Alcohol Clin Exp Res. 2019 Oct 8;43(11):2354–2366. doi: 10.1111/acer.14201

Morphometric Biomarkers of Adolescents with Familial Risk for Alcohol Use Disorder

Arkadiy L Maksimovskiy 1,4, Emily N Oot 1,7, Anna M Seraikas 1, Maya Rieselbach 1, Carolyn Caine 1, Jennifer T Sneider 1,4, Julia E Cohen-Gilbert 1,4, Sion K Harris 5,6, Lisa D Nickerson 2,4, Michael L Rohan 3,4, Marisa M Silveri 1,4,7
PMCID: PMC6824946  NIHMSID: NIHMS1050666  PMID: 31529792

Abstract

Background:

While many adolescents exhibit risky behavior, teenagers with a family history (FH+) of an alcohol use disorder (AUD) are at a heightened risk for earlier initiation of alcohol use, a more rapid escalation in frequency and quantity of alcohol consumption and developing a subsequent AUD in comparison to youth without such family history (FH-). Neuroanatomically, developmentally normative risk-taking behavior parallels an imbalance between more protracted development of the pre-frontal cortex (PFC) and earlier development of limbic regions. Magnetic resonance imaging (MRI) derived volumetric properties were obtained for these structures in FH+ and FH- adolescents.

Methods:

Forty-two substance-naïve adolescents (13–14-year-olds), stratified into FH+ (N=19, 13 girls) and FH- (N=23, 11 girls) age/handedness-matched groups, completed MRI scanning at 3.0T, as well as cognitive and clinical testing. T1 images were processed using Freesurfer to measure PFC and hippocampi/amygdalae subfields/nuclei volumes.

Results:

FH+ status was associated with larger hippocampal/amygdala volumes (p<.05), relative to FH- adolescents, with right amygdala results appearing to be driven by FH+ boys. Volumetric differences also were positively associated with family history density (p<.05) of having an AUD. Larger subfields/nuclei volumes were associated with higher anxiety levels and worse auditory verbal learning performance (p<.05).

Conclusion:

FH+ risk for AUD is detectable via neuromorphometric characteristics, which precede alcohol use onset and the potential onset of a later AUD, that are associated with emotional and cognitive measures. It is plausible that the development of limbic regions might be altered in FH+ youth, even prior to the onset of alcohol use, which could increase later risk. Thus, targeted preventative measures are warranted that serve to delay the onset of alcohol use in youth, particularly in those who are FH+ for an AUD.

Keywords: Adolescence, Development, Hippocampus, Amygdala, Alcohol, Risk

1. Introduction

Family history of alcohol use disorder (AUD; FH+) is a major risk factor for the emergence of alcohol problems in adolescence (Kilpatrick et al, 2000). Neuromorphometric differences in FH+ adolescents’ brain structure, as measured by structural MRI (sMRI), have been identified and could provide a type of biomarker for an elevated risk for problem drinking (Bava & Tapert, 2010). The heightened risk in FH+ youth is relative to an already elevated risk in young people without such family histories (FH-), in comparison with other age groups (Jordan & Andersen, 2017). The trajectory of healthy brain development involves a temporary imbalance between more protracted maturation (via pruning and other processes) of the pre-frontal cortex (PFC) and relatively faster and earlier limbic maturation (Casey, Jones, & Hare, 2008). This state is accompanied by a combination of heightened urges and emotions with immature PFC--mediated executive control (i.e., the imbalance hypothesis; (Somerville & Casey, 2010)). Adolescents are thus more prone to risk-taking which might involve misusing alcohol (Fuhrmann, Knoll, & Blakemore, 2015). However, behavioral (such as maladaptive coping strategies) and neuromorphometric deviations from this trajectory, as evident in FH+ youth, may be indicative of a heightened risk for alcohol problems that persist into adulthood (Kilpatrick et al, 2000).

Developmental changes within the amygdala and the hippocampus, which are involved in learning and decision making, follow an inverted-U pattern and still approach their peaks during early adolescence (Wierenga et al, 2014). Structural maturation of these regions parallels emotional development as well as the maturation of the capacity to learn from the consequences of one’s actions (Somerville & Casey, 2010; Whitford et al, 2006). Adolescents with a FH+ status show heightened reactivity towards emotional stimuli and exhibit maladaptive learning in comparison to their FH- counterparts (Ellis, Zucker, & Fitzgerald, 1997; Johnson, Cohen, Kasen, Smailes, & Brook, 2001; Swartz, Williamson, & Hariri, 2015). This combination may contribute to an increased cognitive susceptibility to misusing alcohol or other drugs (Casey & Jones, 2010). An exacerbated risk hypothesis is proposed for the investigation of neural and cognitive phenotype of heightened risk for an AUD in FH+ adolescents, which differentiates this group from other adolescents by building on the imbalance hypothesis model (Mills, Goddings, Clasen, Giedd, & Blakemore, 2014). Unlike findings pertaining to neurocognitive domains, the nature of neuromorphometric alterations that constitute exacerbated risk are not as clear.

While there has been a considerable amount of work on neuromorphometric alterations within limbic regions (see (Cservenka, 2016) for a review), less is known about the heterogeneous nature of these structures. Given that both the hippocampus and amygdala consist of structurally and functionally distinct subdivisions (for examples, see (Acsády & Káli, 2007; Blum, Moore, Adams, & Dash, 1999; Dimsdale-Zucker, Ritchey, Ekstrom, Yonelinas, & Ranganath, 2018; Fox, Oler, Tromp, Fudge, & Kalin, 2015; Ramirez, Moscarello, LeDoux, & Sears, 2015; Schmidt, Marrone, & Markus, 2012)), identifying smaller scale alterations in high-risk adolescents would not only provide more accurate morphometric information, but also offer insight as to the types of cognitive domains that might be affected. Specifically, while both structures are generally involved in memory processing and emotion (Phelps, 2004), alterations within amygdala nuclei and hippocampal subfields could have important implications for linking altered neuromorphometry to affected cognitive and emotional domains.

The current study examined hippocampal and amygdala structures, within a sample of 42 substance-naïve FH+ and FH- age-matched adolescents. Volumetric measures consist of hippocampal subfields, amygdala nuclei, and PFC volumes. Grey matter morphometry of limbic structures was predicted to be associated with familial risk due to its rapid maturation during the participants’ age range (Somerville & Casey, 2010) and selective sensitivity to familial risk for an AUD (Hanson et al, 2010; Hill et al, 2001). Adolescent limbic grey matter morphometry was shown to be affected in association with emerging pathological conditions (Hanson et al, 2010; MacMillan et al, 2004; Phillips et al, 2002). PFC changes, on the other hand, were mostly reported to occur within functional (Herting, Fair, & Nagel, 2011; Schweinsburg et al, 2004; Vaidya et al, 2019) and white matter connectivity domains (Herting, Schwartz, Mitchell, & Nagel, 2010). These findings indicate that adolescent PFC alterations become apparent when communication with limbic areas is taken into account (such as via functional MRI or white matter connectivity). Limbic regional morphometry, on the other hand, might be independently affected, as measured by selective grey matter structural alterations. Based on the mean age (13.6 years old) of study participants, it was predicted that PFC volume would not differ between FH+ and FH- groups and hippocampal/amygdala volumes would be significantly altered in the FH+ group, in comparison to the FH- participants.

No specific predictions regarding subfields and nuclei were made, as this examination was novel. Finally, FH+ status was predicted to be associated with higher levels of anxiety and lower memory scores, since such alterations within these domains were shown to be associated with alcohol problems (Goldman, 2016; Kendler, Prescott, Myers, & Neale, 2003).

Cognitive measure consists of auditory verbal learning performance and emotional measures consist of anxiety indices. This learning measure has been selected because of the association of memory and learning with adolescent alcohol problems (Goldman, 2016; Squeglia & Gray, 2016) as well as its negative relationship with hippocampal volume (“smaller is better” in adolescents; (Van Petten, 2004). Anxiety has been measured because of its positive association with motivations for problematic alcohol use (Comeau, Stewart, & Loba, 2001), as well as its association with larger amygdala volume (larger amygdala volume is associated with anxiety (De Bellis et al, 2000)). All measures were compared between FH+ and FH- groups and preliminary (due to small groups) follow-up analyses examined sex differences.

2. Materials and Methods

2.1. Participants

The analyzed sample consisted of 42 healthy adolescent participants (24 girls, 18 boys) who completed a baseline visit as part of a 3-year longitudinal neuroimaging study. Participants were stratified into groups of high (FH+, N=19, 13 girls) and low (FH-, N=23, 11 girls) familial risk for an AUD. All subjects were selected based on the availability of the data and familial risk groups were equated on age. The analyzed sample was enriched for FH+ youth, which yielded a higher rate of adolescents with familial risk for drug abuse than reported within the general population (i.e. Nurnberger, Wiegand, and Bucholz, 2004). We took this approach in order to equate enrollment in each group and achieve sufficient power for statistical analyses. The Child Behavior Checklist (CBCL; (Achenbach & Rescorla, 2001)) was used to assess behavioral and mental health problems within the sample. Socioeconomic status was measured using the Hollingshead method (Hollingshead, 1975).

Participants were recruited through Boston Children Hospital (BCH) research participant registries and via local advertisements. Interested individuals completed an online eligibility screen, and if inclusion criteria were met, follow-up verification and scheduling were completed via telephone. All participants and their parent(s)/guardian(s) provided written informed assent and consent, respectively, after receiving a complete description of the study. Participants received monetary compensation for study completion. The study was approved by the Partners Healthcare Institutional Review Board. Trained staff conducted a structured clinical interview using the Mini International Neuropsychiatric Interview for Children and Adolescents (MINI-KID) (Sheehan et al, 2010) to rule out Axis I diagnoses. Prior to scanning, all participants underwent urine screening to rule out psychoactive substance use (Clarity Diagnostics Drugs of Abuse Panel, Boca Raton, FL) and in girls, pregnancy (QuPID One-Step Pregnancy, Stanbio Laboratory, Inc., San Antonio, TX). Participants were excluded if they had prior head trauma with loss of consciousness, radiologically detected brain abnormalities, MR scanning contraindications, lifetime psychoactive substance use (including alcohol, marijuana or nicotine), or a diagnosed psychiatric condition (current or lifetime). Prenatal alcohol and cigarette exposure was assessed using the obstetrics history questionnaire during the baseline visit.

2.2. Assessment of family history status

Accompanying parents/guardians of participants underwent a Family History–Epidemiologic (FHE) interview to determine family history of AUD (Zucker, 1994). This information was used to stratify participants into FH- and FH+ groups. Each FH+ participant had at least one biological parent and/or grandparent with a diagnosed AUD. Although it was not a formal component of the FHE assessment, all participating parents reported information about their alcohol use. Duration since their last drink was estimated based on information reported and the date of this assessment. Family expression of alcoholism, or family history density (FHD) of alcoholism, also was calculated for the FH+ participants, with a single parent with a history of an AUD contributing .5 and a single grandparent contributing a .25, for a possible range from 0 (FH- group) to a maximum of 2.0 (.25–2.0, FH+ group) (Stoltenberg, Mudd, Blow, & Hill, 1998).

2.3. Clinical and Cognitive Measures

The Wechsler Abbreviated Scale of Intelligence (WASI:(Stano, 2004)) vocabulary and matrix reasoning subtests were administered to participants for the purposes of obtaining an estimate of general intelligence. The California Verbal Learning Test-Children’s Version (CVLT-C; (Delis, Kramer, Kaplan, & Ober, 1994)) was administered to assess working memory/auditory attention span, auditory verbal learning, and verbal recognition. Behavioral and clinical survey data were collected using Research Electronic Data Capture (REDCap; (Harris, Taylor, Thielke, Payne, & Gonzalez, 2009), a secure, web-based electronic data capture tool hosted through McLean Hospital, Partners Health Care. Surveys included the Spielberger State—Trait Anxiety Inventory (STAI; (Spielberger, Gorsuch, & Lushene, 1970) to assess anxiety symptoms. The Child Behavioral Checklist (CBL; (Achenbach & Ruffle, 2000)) was used to assess internalizing, externalizing, and total behavioral problem scores.

2.4. Magnetic Resonance Imaging Acquisition/Processing

Imaging data were acquired on a Siemens TIM Trio 3.0 Tesla MRI system (Erlangen, Germany) using a 32-channel head coil. High-resolution structural images were collected using a T1-weighted multi-echo magnetization prepared rapid acquisition gradient echo (ME-MPRAGE) 3D sequence in 4 echoes using the following parameters: TE=1.64/3.5/5.36/7.22 msec, TR=2.1 sec, TI=1.1 sec, FA=12°, 176 slices, 1×1×1.3 mm voxel, acquisition time=5 min.

All T1-ME-MPRAGE images were segmented, labeled and analyzed using the semi-automated FreeSurfer version 6.0 reconstruction pipeline (Fischl, 2012; Fischl et al, 2002). All data were carefully inspected and manually edited. Manual edits were conducted by the first author (AM) and applied to the brainmask file. Edits predominantly involved adjusting pial surfaces to exclude dura matter (all files), with a minimal number of edits involving adjusting pial surfaces to expand the white matter surface. All volumetric files (aseg and subfield volumes) were visually inspected for accuracy of reconstruction; no edits were necessary to those files. Bilateral volumes of the hippocampal and amygdala subfields were extracted using Freesurfer OSX El Capitan Dev Version (Iglesias et al, 2015; Saygin et al, 2017). This method has been shown to provide robust segmentation of 1mm resolution T1 images with high accuracy (Iglesias et al, 2015) and applied to the AUD population (Zahr, Pohl, Saranathan, Sullivan, & Pfefferbaum, 2019). Extracted hippocampal subfields included the parasubiculum, presubiculum (divided into the presubiculum-head and the presubiculum-body), subiculum (divided into the subiculum-head and the subiculum-body), CA1 region (divided into the head and body), CA3 region (divided into the head and body), CA4 (divided into the head, body, the granule cell layer of dentate gyrus head, and granule cell layer of dentate gyrus body), molecular layers of the hippocampal head and body, the hippocampus-amygdala transition area, fibria, tail, and the hippocampal fissure. Amygdalar nuclei included the anterior amygdaloid area, cortico-amygdaloid transition area, basal nucleus, lateral nucleus, accessory basal nucleus, central nucleus, cortical nucleus, medial nucleus, and the paralaminar nucleus. PFC volume was calculated as a sum of individual PFC structures (including the superior frontal gyrus, rostral and caudal middle frontal gyri, pars opercularis, pars triangularis, pars orbitalis, lateral and medial orbitofrontal gyrus, precentral gyrus, paracentral gyrus, and the frontal pole) using the Desikan atlas. All neural regions were adjusted for each participant’s respective head size (using eTIV, a measure of estimated total intracranial volume generated by Freesurfer).

2.5. Statistical Analyses

Analyses of covariance (ANCOVA) models (using Fit Model functionality in JMP; (SAS Institute Inc., 1989)) were used to compare effect sizes and mean differences between FH+ and FH- groups. The first round of analyses to test for group differences used FH status as an independent variable, eTIV and age as covariates, and hippocampal/amygdala and PFC volumes as dependent measures. Each hemisphere was examined separately. These analyses consisted of linear, quadratic, and cubic effects. Furthermore, in order to test for a volumetric developmental imbalance, ratios of limbic volumes (hippocampus and amygdala, separately) to the PFC volume were calculated for each hemisphere (O’Brien & Hill, 2017). These ratios were analyzed using the same models as described above. A second round of analyses were conducted for significant effects (p < .05, adjusted for multiple comparisons), using respective subfields (for hippocampi), nuclei (for amygdalae), and Desikan regions (for the PFC) as dependent measures. The second round of analyses were conducted to identify smaller-scale areas that might have been driving the main effects. A third round of analyses were conducted to examine whether the effects of family history have an impact on significant brain regions (generated by the first two analyses). General linear models were employed for these analyses, with FHD scores as independent variables, age and intracranial volume as covariates, and each brain region as a dependent variable. Further, in cases of significant findings from the first two rounds of analyses, preliminary analyses (due to small groups) were conducted in order to examine for the potential effects of sex. To examine the interaction effect between sex and family history, we created and entered into the model a combined variable with the following categories: FH+ boys, FH- boys, FH+ girls, FH- girls. Other covariates in the model remained the same as in previous models.

Group differences in state and trait anxiety, and memory function were examined using ANCOVA models, which included each respective measure as a dependent variable, FH status as an independent variable, and age as a covariate. Measures of emotional measures of state and trait anxiety consisted of two STAI adjusted t-scores, for each respective measure. The cognitive measure of auditory verbal learning consisted of CVLT total number of correct recall responses, recorded across trials, as well as short delay free recall, total number of correct responses, short delay cued recall total number of correct responses, long delay free recall total number of correct responses, long delay cued recall total number of correct responses, and long delay cued recall, discrimination scores.

Associations between anatomical regions (that were affected by FH status) and cognitive measures were examined using linear regression. Separate regression models included each of the anatomical regions (adjusted for FH status, age, and head size) as predictors, and each of the cognitive measures as dependent variables.

All results were corrected for multiple comparisons using the False Discovery Rate (FDR) correction.

3. Results

3.1. Demographics

There were no significant group differences in age, handedness, internalizing, externalizing, or total CBCL scores (ps >.05) between FH+ and FH- groups. Furthermore, when FH+ and FH- groups were broken down by sex, they did not differ on mean age (p > .05). As expected, mean FHD scores also were significantly higher within FH+ participants, in comparison to FH- participants (FH+: 0.46 ± 0.23; FH-: 0.0 ± 0.0; p = .001). Only five of the 19 FH+ adolescents had a parent with AUD (the remaining 14 were FH+ due to one or more grandparents with a lifetime AUD). Three of those parents reported active alcohol consumption at the time of testing, and two parents reported drinking 4 and 24 years prior to testing, respectively. Demographic characteristics for each FH group are presented in Table 1, and clinical measures are presented in Table 2.

Table 1.

Demographic Characteristics. This table summarizes the demographic characteristic within the examined sample. The FH+ and FH- groups did not differ on sex, age, or handedness characteristics (ps > 0.05). As expected, FHD scores were significantly higher within FH+ participants, in comparison to FH- participants (p < .05). Socioeconomic status was measured using the Hollingshead method (Hollingshead, 1975). FH+: Family History Positive participants; FH-: Family History Negative participants.

Family History Status (Total N=42) Sex (N) Age in years (Mean) Handedness (Number of Left-Handed participants) Family History Density (Mean) Race Ethnicity: Hispanic vs. Non-Hispanic Socioeconomic Status
FH+
(N=18)
 Boys: 6
Girls: 13
 Boys: 13.36 ± .32
Girls: 13.67 ± .56
Everyone: 13.6 ± .51		 1 0.46 Asian/White: 1
Unknown: 1
White: 15
Other: 1		 Non-Hispanic: 16
Hispanic: 2		 49.42
FH-
(N=24)
 Boys: 12
Girls: 11
 Boys: 13.43 ± .23
Girls: 13.78 ± .48
Everyone: 13.6 ± .40		 2 0 African American: 1
Asian: 2
Asian/White: 2
White: 19		 Non-Hispanic: 24
Hispanic: 0		 53.24
Open in a new tab

Table 2.

Clinical and Cognitive Measures. This table summarizes cognitive and emotional measures for FH groups. FH groups do not significantly differ on any of the presented scores (ps > .05). FH+: Family History Positive participants; FH-: Family History Negative participants; WASI: Wechsler Abbreviated Scale of Intelligence; CVLT: The California Verbal Learning Test-Children’s Version; MASC: Multidimensional Anxiety Scale for Children; STAI: The State-Trait Anxiety Inventory. *Internalizing, externalizing, and total problems mean scores have been assessed using the Child Behavioral Checklist (Achenbach & Ruffle, 2000).

Family History Status (Total N=42) WASI IQ CVLT: Auditory Verbal Learning MASC Total t-score STAI State t-score STAI Trait t-score Internalizing Mean Score* Externalizing Mean Score* Total Problems Mean Score*
FH+
(N=18) 		M: 113.15
S.D.: 7.66
 M: 54.72
S.D.: 9.29		 M: 53.33
S.D.: 7.99		 M: 45.11
S.D.: 10.9		 M: 39.06
S.D.: 7.95		 M: 3.78
S.D.: 4.6		 M: 3.0
S.D.: 4.68		 M: 12.06
S.D.: 12.55
FH-
(N=24) 		M: 114.47
S.D.: 10.83		 M: 57.08
S.D.: 7.52		 M: 48.63
S.D.: 14.67		 M: 43.96
S.D.: 9.13		 M: 36.04
S.D.: 4.95		 M: 5.27
S.D.: 4.29		 M: 3.7
S.D.: 4.23		 M:14.78
S.D.: 10.98
Open in a new tab

3.2. Family History Status and Volumetric Measures

Global Hippocampal and Amygdala Volumes

ANCOVA models indicated that FH+ participants exhibited significantly larger bilateral hippocampal and amygdala volumes relative to FH- participants. This linear effect was found within the right (Adj. R2 = .47, F1,40 = 13.12, p = .0001) and left (Adj. R2 = .5, F1,40 = 14.45, p = .0001) hippocampi, as well as right (Adj. R2 = .47, F1,40 = 13.12, p = .0001) and left (Adj. R2 = .49, F1,40 = 12.55, p = .0001) amygdalae (Figure 1, 2, respectively). After testing, linear, quadratic, and cubic effects of each structure, only linear effects were significant. Results of these analyses are as follows. Left hippocampus: Linear: β = .46, t40 = 2.71, p = .03; Quadratic: β = .57, t40 = 1.18, p = .31; Cubic: β = −.79, t40 = −1.58, p = .2. Right hippocampus: Linear: β = .47, t40 = 2.66, p = .03, Quadratic: β = .06, t40 = .13, p = .9, Cubic: β = −.32, t40 = −.62, p = .67. Left Amygdala: Linear: β = .51, t40 = 2.96, p = .01; Quadratic: β = .61, t40 = 1.24, p = .22; Cubic: β = −.88, t40 = −1.72, p = .15. Right Amygdala: Linear: β = .45, t40 = 2.49, p = .04, Quadratic: β = .02, t40 = .04, p = .97, Cubic: β = −.07, t40 = −.14, p = .97. Table 3 summarizes the mean values as well as adjusted and unadjusted p-values of all linear volumetric measures..

Figure 1. Hippocampal Volume and Family History.

Figure 1.

Open in a new tab

This figure presents the mean (indicated by the height of each column) and individual values (individual data points) for left and right hippocampal volumes of adolescents with FH+ and FH- status . FH+ volume is significantly larger than FH- volume of the left and right hippocampus. Left Hippocampal Volume: FH- adjusted mean: 3,504.6217 mm3; FH+ adjusted mean: 3,642.3524 mm3; Right Hippocampal Volume: FH- adjusted mean: 3,609.7129 mm3; FH+ adjusted mean: 3,790.5399 mm3. FH+: Family History Positive; FH-: Family History Negative. Error bars indicate standard error mean (1 standard error from the mean).

Figure 2. Amygdala Volume and Family History.

Figure 2.

Open in a new tab

This figure presents the mean and individual values for left and right amygdala volumes of adolescents with FH+ and FH- status. FH+ volume is significantly larger than FH- volume of the left and right amygdala. Left Amygdala Volume: FH+ adjusted mean: 1,780.7323 mm3; FH- adjusted mean: 1,690.9297mm3. Right Amygdala Volume: FH+ adjusted mean: 1,860.9523 mm3; FH- adjusted mean: 1,732.6264 mm3. FH+: Family History Positive; FH-: Family History Negative. Error bars indicate standard error mean (1 standard error from the mean).

Table 3.

Statistical Results Summary. This table displays the corrected and uncorrected p-values as well as mean values for grey matter volume in hippocampal subfields and nuclei of the amygdala. All mean values have been adjusted for covariates. Analyses were conducted using the ANCOVA model with grey matter regions as dependent measures and family history status as independent variable, covaried for age and head-size (using intra-cranial volume). Abbreviations: FH+: Family history of alcohol abuse; FH-: No family history of alcohol abuse; U.C. p: Uncorrected (raw) p-values; C. p: Corrected p-values; PFC: Pre-Frontal Cortex.

Left Right
Structure Subfield/Nucleus FH-
mm3 		FH+
mm3 		U.C.
p		 C.
p		 Cohen’s
d		 FH-
mm3 		FH+
mm3 		U.C.
p		 C.
p		 Cohen’s
d
Hippocampus Parasabiculum 66.33 ±8.47 67.25 ±12.3 .78 .78 −0.29 68.69 ±11.74 68.54 ±12.9 .97 .97 0.04
Presubiculum-head 144.47 ±19.81 150.56 ±21.44 .25 .50 −1.34 142.64 ±19.81 154.85 ±21.44 .05 .05 −2.69
Subiculum-head 198.61 ±26.02 214.62 ±36.34 .08 .16 −2.87 188.64 ±25.48 206.07 ±32.96 .03 .07 −3.22
CA1-head 537.74 ±50.02 583.41 ±72.74 .01 .02 −5.83 555.22 ±57.12 596.40 ±74.52 .03 .05 −5.08
CA3-head 127.48 ±15.7 126.10 ±24.3 .82 .98 0.31 136.84 ±20.18 135.79 ±16.09 .85 .85 0.25
CA4-head 135.01 ±15.4 139.28 ±20.17 .39 .72 −1.01 139.99 ±16.24 143.14 ±17.26 .49 .49 −0.77
CC-ML-DG-head 161.86 ±17.52 167.52 ±20.88 .28 .72 −1.29 168.84 ±20.18 173.18 ±22 .46 .49 −0.95
HP-head 348.82 ±32.88 368.60 ±38.44 .03 .06 −3.31 351.84 ±36.63 372.37 ±41.41 .05 .09 −3.29
HATA 54.59 ±9.28 55.95 ±6.98 .60 .60 −0.48 56.94 ±8.54 6.36 ±6.35 .14 .14 18.54
Presubiculum body 178.39 ±27.83 180.46 ±22.92 .77 .77 −0.41 171.02 ±18.34 184.23 ±28.04 .04 .05 −2.74
Subiculum-body 245.20 ±27.93 247.59 ±26.85 .77 .77 −0.46 246.93 ±26.57 253.62 ±26.12 .38 .38 −1.3
CA1-body 125.65 ±26.22 132.73 ±39.68 .46 .46 −1.23 128.71 ±22.75 136.49 ±29.42 .33 .33 −1.52
CA3-body 80.86 ±14.26 80.77 ±13.89 .98 .98 0.02 91.01 ±17.69 90.05 ±16.73 .84 .85 0.23
CA4-body 118.08 ±10.84 116.99 ±11.8 .72 .72 0.32 122.66 ±13.3 125.45 ±11.12 .34 .49 −0.8
CC-ML-DG-body 136.39 ±12.9 134.44 ±11.69 .54 .72 0.56 139.81 ±14.61 144.00 ±12.07 .21 .49 −1.15
HP-body 231.26 ±24.02 232.62 ±19.64 .81 .81 −0.29 235.24 ±25 243.13 ±24.64 .23 .23 −1.58
fimbria 81.39 ±20.68 79.35 ±13.67 .71 .71 0.49 73.57 ±13.15 75.72 ±16.03 .64 .64 −0.56
Hippocampal Tail 532.49 ±53.23 564.10 ±54.37 .06 .06 −4.31 591.13 ±60.72 627.14 ±77.43 .07 .07 −4.33
Hippocampal Fissure 132.53 ±24.56 140.41 ±25.76 .30 .30 −1.57 138.02 ±22.15 142.85 ±27.46 .50 .50 −0.97
Whole Hippocampus 3504.62 ±273.74 3642.35 ±297.6 .05 .05 −8.15 3609.71 ±318.23 379.54 ±324.5 .02 .02 180.19
Amygdala Lateral 668.89 ±71.27 707.85 ±78.71 .04 .11 −4.5 67.56 ±75.43 719.83 ±90.34 .03 .07 −71.65
Basal 430.16 ±44.35 456.86 ±47.51 .03 .11 −3.94 441.82 ±36.76 481.60 ±50.68 .00 .01 −6.02
Accessory 243.33 ±23.1 253.87 ±35.18 .20 .35 −1.95 257.61 ±24.1 273.76 ±32 .04 .07 −3.05
Central 43.44 ±7.86 45.60 ±10.47 .44 .56 −0.71 44.46 ±9.34 46.61 ±8.28 .44 .52 −0.72
Medial 17.93 ±4.68 17.59 ±5.32 .83 .83 0.15 19.34 ±6.23 19.46 ±6.59 .95 .95 −0.05
Cortical 21.72 ±5 22.11 ±7 .83 .83 −0.16 23.77 ±5.16 25.01 ±5.67 .47 .52 −0.53
Paralaminar 46.57 ±5.4 49.98 ±5.74 .03 .11 −1.44 48.04 ±4.51 52.56 ±5.75 .00 .01 −2
Cortico-amygdaloid 162.02 ±17.92 166.51 ±19.5 .38 .56 −1.04 168.49 ±17.24 179.02 ±20.57 .04 .07 −2.42
Anterior 56.87 ±7.3 60.36 ±9.6 .14 .31 −1.2 58.54 ±8.68 63.09 ±10.05 .10 .15 −1.49
Whole Amygdala 1690.93 ±152.56 1780.73 ±191.6 .04 .04 −6.85 1732.63±153.4 1860.95 ±201.21 .01 .01 −9.64
PFC Whole PFC 118382.27 ±10136.21 119443.73 ±9784.19 .73 .73 −10.64 116318.52 ±10437.1 118631.74 ±11056.52 .48 .48 −22.31
Open in a new tab

Two FH+ participants were identified to be at risk for the effects of prenatal alcohol exposure based on parental reports. In both cases, the potential effects were likely minimal or indirect, since participants’ mothers did not have an AUD. One of these participants was also at risk for cigarette exposure due to missing maternal cigarette use data. We therefore visually inspected their bilateral limbic volume (hippocampus and amygdala) data and confirmed that they were consistent with the rest of the FH+ group.

Ratios of Hippocampal and Amygdala Volumes to the PFC Volume

ANCOVA models did not reveal any significant differences between FH+ and FH- participants on the ratios of the hippocampal and amygdala volumes to the PFC volume. The results are follows. Left Hippocampus to Left PFC Ratio: Adj. R2 = .06, F1,41 = .26, p = .99, β = −.1, t40 = −.63, p = .53. Right Hippocampus to Right PFC Ratio: Adj. R2 = .07, F1,41 = .33, p = .97, β = −.09, t40 = −.58, p = .57. Left Amygdala to Left PFC ratio: Adj. R2 = −.02, F1,41 = .56, p = .58, β = −.14, t40 = −.88, p = .39. Right Amygdala to Right PFC ratio: Adj. R2 = −.01, F1,41 = .88, p = .43, β = −.15, t40 = −.96, p = .34.

Hippocampal Subfields

Within the right hippocampus, the right presubiculum was significantly larger in volume in the FH+ group compared to the FH- group (right presubiculum head: FH+ adjusted mean: 154.854 mm3; FH- adjusted mean: 142.64 mm3; F1,40 = 4.18, p = .048; right presubiculum body: FH+ adjusted mean: 184.23 mm3; FH- adjusted mean: 171.02 mm3; F1,40 = 4.33, p = .048). CA1 head was also larger in FH+ participants (marginal on the right), in comparison to FH- participants, within the bilateral hippocampus (left CA1 head: FH+ adjusted mean: 537.7 mm3; FH- adjusted mean: 583.4 mm3; F1,40 = 7.64, p = .018; marginal within the right CA1 head: FH+ adjusted mean: 596.4 mm3; FH- adjusted mean: 555.23 mm3; F1,40 = 5.44, p = .051).

Amygdaloid Nuclei

Within the right amygdala, the basal nucleus (FH+ adjusted mean: 481.6 mm3; FH- adjusted mean: 441.82 mm3; F1,40 = 9.98, p = .012) and the paralaminar nucleus (FH+ adjusted mean: 52.56 mm3; FH- adjusted mean: 48.04 mm3; F1,40 = 7.96, p = .012) were larger in volume in FH+ participants, in comparison to FH- participants.

Family History Density

General linear models, have revealed significant associations between FHD measures and global as well as subfield/nuclear volumetric measures. Hippocampal and amygdalae volumes are positively correlated with family density scores, that were adjusted for age and intracranial volume (Left hippocampus: β = .28, t40 = 2.5, p = .017, Adj. R2 = .5, F1,41 = 14.45, p = .0001; Right hippocampus: β = .27, t40 = 2.35, p = .02, Adj. R2 = .47, F1,41 = 13.12, p = . 0001; Left Amygdala: β = .3., t40 = 2.55, p = .02, Adj. R2 = .49, F1,41 = 12.55, p = .0001; Right Amygdala: β = .41, t40 = 3.56, p = .001, Adj. R2 = .47, F1,41 = 13.12, p = .0001). See Figure 3. These patterns remained consistent for most subfield and nuclei, except for the right presubiculum head. All regions which were significant as a main effect of family history, were positively correlated with FHD measures. Statistical details for the subfield/nuclei are as follows. Right presubiculum head: β = .15, t40 = 1.05, p = .15, Adj. R2 = .21, F1,41 = 4.6, p = .008. Left CA1 head: β = .39, t40 = 3.14, p = .003, Adj. R2 = .38, F1,41 = 9.34, p = .0001. Right CA1 head: β = .33, t40 = 2.56, p = .015, Adj. R2 = .34, F1,41 = 8.04, p = .0003. Right presubiculum body: β = .28, t40 = 2.06, p = .047, Adj. R2 = .29, F1,41 = 6.47, p = .001. Right basal nucleus of the amygdala: β = .48, t40 = 4.06, p = .0003, Adj. R2 = .45, F1,41 = 12.09, p = .0001. Right paralaminar nucleus of the amygdala: β = .46, t40 = 3.76, p = .0006, Adj. R2 = .4, F1,41 = 10.3, p = .0001.

Figure 3. Family History Density and Global Limbic Volume.

Figure 3.

Open in a new tab

This figure depicts relationships between global hippocampal/amygdalae volumes with family history density (FHD). Blue dots indicate FH- and red dots indicate FH+.

Sex Differences

Preliminary analyses of sex differences indicated that the association between FH+ status and higher right amygdala volume was stronger in adolescent boys, with FH+ adolescent boys having higher mean right amygdala volume compared to all other groups (F3,38 = 4.73, p = .028) (Figure 4). No sex differences were detected within the left amygdala (F3,38 = 2.07, p = .171) or the hippocampus (right: F3,38 = 2.02, p = .171; left: F3,38 = 1.41, p = .255).

Figure 4. Family history of AUD and sex differences.

Figure 4.

Open in a new tab

This figure presents the mean and individual values for individuals of different sex and family history status. FH+ male participants have significantly higher right amygdala volume in comparison to each of the other groups. FH- Participants: Girls adjusted mean: 1,700.29 mm3; Boys adjusted mean: 1,757.81 mm3; FH+ Participants: Girls adjusted mean: 1,805.99 mm3; Boys adjusted mean: 1,988.9597 mm3. FH+: Family History Positive; FH-: Family History Negative. Error bars indicate standard error mean (1 standard error from the mean).

3.1.6. Pre-frontal Cortex

There were no significant group differences in PFC volume between FH+ and FH- participants (F1,40 = .05, p = .17). FH- Participants adjusted mean: 0.15 mm3; FH+ Participants adjusted mean: 0.15 mm3.

3.3. Cognitive/Emotional Scores and Volumetric Measures

There were no direct effects of FH status on the examined anxiety and auditory verbal learning measures. Results of these tests are as follows. STAI trait anxiety t-scores: F1,40 = .69, p = .687; STAI state anxiety t-scores: F1,40 = .16, p = .753; CVLT total number of correct responses: F1,40 = .79, p = .46. CVLT short delay free recall, total number of correct responses: F1,40 = .35, p = .96; CVLT short delay cued recall, total number of correct responses: F1,40 = 1.54, p = .43; CVLT long delay free recall, total number of correct responses: F1,40 = .68, p = .77; CVLT long delay cued recall, total number of correct responses: F1,40 = 1.08, p = .35; CVLT long delay cued recall, discrimination: F1,40 = .05, p = .95.

Trait and State Anxiety

Both hippocampal and amygdala volumes significantly associated with anxiety trait t-scores. Left hippocampal volume was associated with anxiety trait score (β = .38, t40 = 2.59, p = .027) and significantly explained a portion of its variance (Adj. R2 = .12, F1,41 = 6.69, p = .027). Right hippocampal volume was associated with anxiety trait score (β = .48, t40 = 3.44, p = .006) and significantly explained a portion of its variance (Adj. R2 = .21, F1,41 = 11.83, p = .006). The hippocampal effects were driven by CA1 head subfield volumes, with a marginal effect on the left subfield (Left CA1 head: β = .30, t40 = 2, p = .053; Adj. R2 = .07, F1,41 = 3.98, p = .053; Right CA1 head: β = .34, t40 = 2.26, p = .029; Adj. R2 = .09, F1,41 = 5.12, p = .029). Right amygdala volume was also associated with anxiety trait score (β = .35, t40 = 2.38, p = .03) and significantly explained a portion of its variance (Adj. R2 = .1, F1,41 = 5.64, p = .03). This effect was driven by the right basal nucleus (β = .31, t40 = 2.03, p = .049; Adj. R2 = .07, F1,41 = 4.12, p = .049).

STAI state anxiety t-scores were associated with bilateral hippocampal volume indices. Left hippocampal volume was associated with STAI state anxiety (β = .38, t40 = 2.58, p = .028), explaining) and explained a significant portion of the variance (Adj. R2 = .12, F1,41 = 6.64, p = .028). Right hippocampal volume was related to STAI state anxiety (β = .38, t40 = 2.62, p = .028) and explained a significant portion of the state anxiety variance (Adj. R2 = .13, F1,41 = 6.88, p = .028).

Subfield analyses provide evidence that these effects were driven by the bilateral (marginal on the left) CA1 head hippocampal subfields. Left CA1 head explained a portion of the state anxiety at a level that approached significance (β = .30, t40 = 2, p = .053) and explained the variance at a marginal level (Adj. R2 = .07, F1,41 = 3.99, p = .053). Right CA1 head was associated with the state anxiety t-scores at a significant level (β = .32, t40 = 2.11, p = .041) and explained a significant portion of the variance as well (Adj. R2 = .08, F1,41 = 4.47, p = .041).

Visual inspections indicate that both the directionality and magnitude of the described effects are consistent between FH+ and FH- groups.

Auditory Verbal Learning Performance

Higher volume of the right presubiculum head hippocampal subfield was associated with lower total number of correct responses on the CVLT test (β = −.33, t40 = −2.21, p = .033) and significantly explained a portion of the scores’ variance (Adj. R2 = .11, F1,41 = 4.89, p = .033). None of the other tested CVLT measures were significantly associated with hippocampal or amygdala volumes (p’s > .05).

4. Discussion

Consistent with the study hypotheses, FH+ status was associated with larger volumetric measures in comparison to FH- status within the hippocampi and amygdalae but not the PFC. This finding is supportive of the exacerbated risk model, which superimposes an additional level of neurodevelopmental risk for AUD for FH+ adolescents in comparison to FH- youth. In healthy adolescents, a temporary neurodevelopmental fronto-limbic imbalance occurs when the PFC matures more slowly than subcortical regions (Koolschijn & Crone, 2013; Mills et al, 2014; van Duijvenvoorde, Achterberg, Braams, Peters, & Crone, 2016). This period is linked to heightened risk-taking behaviors, including substance use (Mills et al, 2014). Current findings suggest that 13–14-year-old adolescents with a FH+ status do not show neuromorphometric differences within the PFC, but present with altered limbic volumes. The structural alteration within the limbic system could be linked to a heightened risk for AUD.

Familial risk for AUD also appears to predict limbic volume in a linear pattern. The linear effect was verified via significant linear tests and non-significant quadratic and cubic models. Specifically, all volumetric measures which significantly differed between FH+ and FH- groups, were also positively correlated with FHD scores. This pattern was significant for all global volumes (hippocampal and amygdalae) and most subfield/nuclear regions, with the exception of the right presubiculum head. These findings extend the impact of family history on AUD risk, indicating that it is not only a binary predictor, but also a continuous measure of elevated risk, which increases with a rising number of relatives who have histories of AUD.

While current findings are consistent with prior work on a similar age-range (i.e. Hanson et al, 2010), it is intriguing that older individuals with familial risk for substance use disorders (late adolescents and young adults) were shown to have smaller limbic volume, relative to participants without elevated risk (see Benegal, Antony, Venkatasubramanian, & Jayakumar, 2007; Dager et al, 2015, for examples). These findings highlight the need for an extensive longitudinal study of neuromorphometric alterations of heightened risk for substance use disorders across a broad developmental spectrum. Although speculative, it is possible that neural tissue in those at risk follows an altered developmental trajectory: showing a delay in pruning (thus, higher volume) at a younger age, and increased pruning (thus, lower volume) at later points of development. Given the non-linear trajectory of neural development, it is plausible that exacerbated risk is expressed non-linearly as well.

In addition to age related considerations, a persuasive explanation of the current findings is that amygdala volume is decreased in association with a higher risk for alcohol disorders but increased in association with anxiety characteristics. Consistent with this theory, it was reported that offsprings from families with high familial risk for alcohol problems, had smaller amygdala volumes (Hill et al, 2001; Hill et al, 2013; Dager et al, 2015). On the contrary, adolescents who exhibited high anxiety traits, relative to their counterparts with lower anxiety characteristics, were shown to have larger amygdala volumes (Hill, Tessner, Wang, Carter, & McDermott, 2010; Roth, Humphreys, King, & Gotlib, 2018; Albaugh, 2017). Given that most participants in the current sample did not have a first degree relative with an alcohol disorder, it is likely that amygdala morphometry reflects the presence of anxiety characteristics. The positive association of amygdala volume with anxiety measures in the current sample provides further support for this interpretation. Furthermore, the association of amygdala volume with anxiety characteristics introduces an additional factor of complexity to using neuromorphometry for detecting exacerbated risk for alcohol problems in young adolescents.

Current findings also indicate a lack of significant difference in the ratios of limbic-to-PFC volume between young adolescents with a heightened familial risk and their normal risk counterparts. Thus, although the limbic-to-PFC morphometric ratio might become a useful metric for predicting the onset of a substance use disorder within an older sample of adolescents, it is not appropriate for identifying exacerbated risk in a younger sample of adolescents. It is also important to note that limbic-to-PFC ratio differences might become pronounced in adolescents with a larger familial loading, as analyzed in prior work (see O’Brien and Hill, 2017). Thus, although adolescents in the current sample are at an exacerbated risk for developing addiction disorders, the limbic-to-PFC ratio differences might be present in even higher risk samples, results from greater familial predispositions with multiplex factors. Consideration of these characteristics highlight the continuous, rather than dichotomous, nature of familial risk.

Smaller scale nuclei and subfield analyses revealed that amygdala-related group differences were driven by bilateral basal and paralaminar nuclei and hippocampal related effects were driven by bilateral CA1 head and right presubiculum subfields. Higher hippocampal subfield volume was associated with higher trait anxiety scores and lower auditory verbal learning performance, across all participants. Specifically, higher bilateral CA1 volume measures were related to elevated state and trait anxiety measures and higher right presubiculum head volume indices were associated with lower auditory verbal learning scores. These findings are consistent with prior reports which indicated that CA1 head volume plays a role in anxiety regulation (Freeman-Daniels, Beck, & Kirby, 2010) and the presubiculum was involved in memory formation (Bannerman et al, 2004; Fanselow & Dong, 2010; Zeidman & Maguire, 2016).

In contrast to hippocampal findings, volumetric measures of the right basal amygdala nucleus were associated only with higher trait, and not state, anxiety indices. This is consistent with prior work, which reported on the role of basal amygdala nucleus in fear conditioning (Cartoni, Puglisi-Allegra, & Baldassarre, 2013), a long-term learning mechanism linked to trait (and not state) measures (Indovina, Robbins, Núñez-Elizalde, Dunn, & Bishop, 2011). In addition to offering more fine-tuned neuromorphometric information, identifying these regions provides specific avenues of research for identifying concrete neurocognitive phenotypes that might be affected in FH+ adolescents.

Given the mean participant age of 13.6 years in the current sample and the deliberate selection of substance naïve participants, FH associated behavioral differences, such as more alcohol use, were not expected to emerge until a few years later (Kessler et al, 2005; O’Brien & Hill, 2017). Thus, the presence of cognitively and emotionally linked anatomical characteristics in the absence of direct behavioral differences might reveal early biomarkers of risk for AUD at an older age.

Although the direct effect of familial risk on cognitive measures is not significant, it might be possible that limbic subfields and nuclei mediate the relationship between familial risk and cognitive function. In that scenario, limbic differences might serve as biomarkers of familial risk that translates into cognitive differences. Given that mediation analyses require a relatively large sample size (see Fritz & MacKinnon, 2007), one that exceeds the magnitude of the current dataset, it would be worthwhile for future studies to examine the role of limbic subfields/nuclei as mediators of familial risk and cognitive susceptibility towards substance abuse.

Preliminary analyses of sex differences indicate that FH associated neuromorphometric differences within the right amygdala are selectively higher in FH+ adolescent boys in comparison to all other sub-groups. A replication of this finding with a larger sample would be indicative of a heightened developmental risk within the FH+ adolescent male population. This is consistent with prior work showing that men are at a higher risk than women for alcohol problems (Kessler et al, 2005).

It is important to consider the possibility that current findings differ from prior reports due to discrepancies in segmentation methodology of neuroimaging data (i.e. manual tracing versus automatic segmentation). Automatic Freesurfer segmentation was selected for delineating subregions because of its time-saving advantages over manual tracing methods and because of the demonstrated accuracy in identifying hippocampal and amygdala volume (see Morey et al, 2005 and Iglesias et al, 2015). Nevertheless, although correlations between Freesurfer’s limbic segmentation and the manual tracing methods were reported to be high (hippocampus R2=.82 and amygdala R2=.56, as reported in Morey et al, 2005), they were not perfect (the correlations are particularly lower for amygdala volume; see Morey et al, 2005). It is thus possible that current findings contradict prior work (which is based on manual tracing; see Hill et al, 2001, for an example) due to methodological differences in subcortical segmentation.

Despite important implications of these findings, this study has some limitations. Although the hippocampal subfield segmentation method is the best available (Iglesias et al, 2015) and has been applied to study AUD-related morphometry using scans of similar resolution (Zahr et al, 2019), the nature of this measure remains probabilistic and therefore requires further verification. As mentioned earlier, sex-subgroups were small and thus need to be replicated within a larger sample for more definitive conclusions. Additionally, the cross-sectional nature of the study design limits the interpretation to non-causal associative findings. A longitudinal follow-up with larger groups is underway to investigate more robust conclusions regarding the developmental risk factors for AUD. Given the current association of morphometric differences with auditory verbal learning, future studies should explore additional memory domains as well as their relationship to grey matter differences. Since the hypotheses of this study focused on measuring limbic structural volume, relative to PFC, PFC was treated as one structure. However, because PFC is composed of multiple functionally heterogeneous regions (see Miller & Cohen, 2001), future work should investigate developmental differences within regions of the PFC. Furthermore, the link between morphometry and cognitive function would be strengthened by a functional MRI examination of the respective subfield and nuclei activation during relevant tasks. It is also important to highlight the fact that the current sample does not include participants with Axis I disorders. Although this exclusion criteria decreases the influence of comorbid conditions, it also restricts the interpretability of the findings to more high functioning adolescents. Limbic subfields and nuclei should be analyzed in samples with comorbid conditions in future studies.

5. Conclusion

These findings indicate a neuromorphometric phenotype that may suggest a heightened risk for AUD in healthy FH+ adolescents. Larger volumetric measures of hippocampi and amygdalae might be indicative of an altered limbic development and constitute a risk factor for future alcohol-related problems due to comorbid anxiety. This risk increases with the number of relatives who were affected by AUD, as measured using FHD. The association of larger volumetric measures with higher levels of risk-inducing cognitive traits (state and trait anxiety) and lower levels of protective emotional components (memory function), presents a phenotype of an emotional and cognitive vulnerability for maladaptive behavior. These findings offer an approach for detecting a conferred risk for alcohol problems in early adolescence. Successful early identification of alcohol-related risk factors might identify individuals with a need for interventive measures, and consequently help decrease the rate of alcohol problems from developing in later years.

6. Acknowledgements

The first author would like to express his gratitude to the senior author, Dr. Marisa Silveri, for providing him with the opportunity to work with this rich and informative dataset which was collected as part of Dr. Silveri’s R01 award. The first author is grateful for Dr. Silveri’s continued support and helpful feedback during the analysis and writing stages of this manuscript. Additionally, the first author would like to thank Dr. Scott Lukas for allocating the funding (via the T32 award) and salary support during the analyses and preparation of this manuscript. Thanks to Emily Oot, Anna Seraikas, Maya Rieselbach, and Carolyn Caine, for their help with scheduling participants and collecting data, Dr. Jennifer Sneider and Dr. Julia Cohen-Gilbert for management and coordination of imaging data collection and pre-processing, Dr. Sion Harris for referring participants for this study from the Boston Children Hospital’s (BCH) Adolescent Substance use and Addiction Research (CeASAR) Center, and all authors for thorough review of this manuscript. Furthermore, the authors would like to thank all reviewers and the editor, for their thorough and helpful comments. The authors have no conflict of interest to declare.

Sources of Support: This work was supported by the following awards from the National Institutes of Health: National Institute on Alcohol Abuse and Alcoholism (NIAAA) R01 AA022493 (awarded to Dr. Silveri) and National Institute on Drug Abuse (NIDA) T32 DA015036 (awarded to Dr. Lukas).

7. References

  1. Achenbach TM, & Rescorla L (2001). Reliability, Internal Consistency, Cross-Informant Agreement, and Stability In Manual For the ASEBA School-Age Forms and Profiles. Burlington, VT. [Google Scholar]
  2. Achenbach TM, & Ruffle TM (2000). The Child Behavior Checklist and related forms for assessing behavioral/emotional problems and competencies. Pediatrics in Review, 21(8), 265–271. [DOI] [PubMed] [Google Scholar]
  3. Acsady L, & Kali S (2007). Models, structure, function: the transformation of cortical signals in the dentate gyrus. Progress in brain research, 163, 577–599. [DOI] [PubMed] [Google Scholar]
  4. Albaugh MD, Nguyen TV, Ducharme S, Collins DL, Botteron KN, D’Alberto N, Evans AC, Karama S, Hudziak JJ, Brain Development Cooperative Group. (2017). Age-related volumetric change of limbic structures and subclinical anxious/depressed symptomatology in typically developing children and adolescents. Biological psychology, 124, 133–140. [DOI] [PubMed] [Google Scholar]
  5. Bannerman DM, Rawlins JNP, McHugh SB, Deacon RMJ, Yee BK, Bast T, Zhang WN, Pothuizen HH, Feldon J (2004). Regional dissociations within the hippocampus—memory and anxiety. Neuroscience and Biobehavioral Reviews, 28(3), 273–283. [DOI] [PubMed] [Google Scholar]
  6. Bava S, & Tapert SF (2010). Adolescent Brain Development and the Risk for Alcohol and Other Drug Problems. Neuropsychology Review, 20(4), 398–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Benegal V, Antony G, Venkatasubramanian G, & Jayakumar PN (2007). Imaging study: gray matter volume abnormalities and externalizing symptoms in subjects at high risk for alcohol dependence. Addiction biology, 12(1), 122–132. [DOI] [PubMed] [Google Scholar]
  8. Blum S, Moore AN, Adams F, & Dash PK (1999). A Mitogen-Activated Protein Kinase Cascade in the CA1/CA2 Subfield of the Dorsal Hippocampus Is Essential for Long-Term Spatial Memory. Journal of Neuroscience, 19(9), 3535–3544. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cartoni E, Puglisi-Allegra S, & Baldassarre G (2013). The three principles of action: a Pavlovian-instrumental transfer hypothesis. Frontiers in Behavioral Neuroscience, 7(153). [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Casey BJ, & Jones RM (2010). Neurobiology of the Adolescent Brain and Behavior: Implications for Substance Use Disorders. Journal of the American Academy of Child & Adolescent Psychiatry, 49(12), 1189–1201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Casey BJ, Jones RM, & Hare TA (2008). The Adolescent Brain. Annals of the New York Academy of Sciences, 1124(1), 111–126. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Comeau N, Stewart SH, & Loba P (2001). The relations of trait anxiety, anxiety sensitivity, and sensation seeking to adolescents’ motivations for alcohol, cigarette, and marijuana use. Addictive Behaviors, 26(6), 803–825. [DOI] [PubMed] [Google Scholar]
  13. Cservenka A (2016). Neurobiological phenotypes associated with a family history of alcoholism. Drug and Alcohol Dependence, 158, 8–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dager AD, McKay DR, Kent JW Jr, Curran JE, Knowles E, Sprooten E, Göring HH, Dyer TD, Pearlson GD, Olvera RL, Fox PT, Lovallo WR, Duggirala R, Almasy L, Blangero J, Glahn DC (2015). Shared genetic factors influence amygdala volumes and risk for alcoholism. Neuropsychopharmacology, 40(2), 412. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. De Bellis MD, Casey BJ, Dahl RE, Birmaher B, Williamson DE, Thomas KM, Axelson DA, Frustaci K, Boring AM, Hall J, Ryan ND (2000). A pilot study of amygdala volumes in pediatric generalized anxiety disorder. Biological Psychiatry, 48(1), 51–57. [DOI] [PubMed] [Google Scholar]
  16. Delis DC, Kramer J, Kaplan E, & Ober B (1994). California Verbal Learning Test-Children’s Version. San Antonio, TX: Psychological Corporation. [Google Scholar]
  17. Dimsdale-Zucker HR, Ritchey M, Ekstrom AD, Yonelinas AP, & Ranganath C (2018). CA1 and CA3 differentially support spontaneous retrieval of episodic contexts within human hippocampal subfields. Nature Communications, 9(1), 693. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. van Duijvenvoorde AC, Achterberg M, Braams BR, Peters S, & Crone EA (2016). Testing a dual-systems model of adolescent brain development using resting-state connectivity analyses. Neuroimage, 124, 409–420. [DOI] [PubMed] [Google Scholar]
  19. Ellis DA, Zucker RA, & Fitzgerald HE (1997). The role of family influences in development and risk. Alcohol Health and Research World, 21, 218–226. [PMC free article] [PubMed] [Google Scholar]
  20. Fanselow MS, & Dong H-W (2010). Are the Dorsal and Ventral Hippocampus Functionally Distinct Structures? Neuron, 65(1), 7–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Fischl B (2012). FreeSurfer. NeuroImage, 62(2), 774–781. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Fischl B, Salat DH, Busa E, Albert M, Dieterich M, Haselgrove C, & Montillo A (2002). Whole Brain Segmentation: Automated Labeling of Neuroanatomical Structures in the Human Brain. Neuron, 33(3), 341–355. [DOI] [PubMed] [Google Scholar]
  23. Fox AS, Oler JA, Tromp DPM, Fudge JL, & Kalin NH (2015). Extending the amygdala in theories of threat processing. Trends in Neurosciences, 38(5), 319–329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Freeman-Daniels E, Beck SG, & Kirby LG (2010). Cellular correlates of anxiety in CA1 hippocampal pyramidal cells of 5-HT1A receptor knockout mice. Psychopharmacology, 213(2–3), 453–463. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Fritz MS, & MacKinnon DP (2007). Required sample size to detect the mediated effect. Psychological science, 18(3), 233–239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Fuhrmann D, Knoll LJ, & Blakemore S-J (2015). Adolescence as a Sensitive Period of Brain Development. Trends in Cognitive Sciences, 19(10), 558–566. [DOI] [PubMed] [Google Scholar]
  27. Goldman MS (2016). Risk for Substance Abuse: Memory as a Common Etiological Pathway:. Psychological Science, 10(3), 196–198. [Google Scholar]
  28. Hanson KL, Medina KL, Nagel BJ, Spadoni AD, Gorlick A, & Tapert SF (2010). Hippocampal Volumes in Adolescents with and without a Family History of Alcoholism. The American Journal of Drug and Alcohol Abuse, 36(3), 161–167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Harris PA, Taylor R, Thielke R, Payne J, & Gonzalez N (2009). A metadata-driven methodology and workflow process for providing translational research informatics support. Journal of Biomedical Informatics. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Herting MM, Fair D, & Nagel BJ (2011). Altered fronto-cerebellar connectivity in alcohol-naïve youth with a family history of alcoholism. NeuroImage, 54(4), 2582–2589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Herting MM, Schwartz D, Mitchell SH, & Nagel BJ (2010). Delay Discounting Behavior and White Matter Microstructure Abnormalities in Youth With a Family History of Alcoholism. Alcoholism: Clinical and Experimental Research, 34(9), 1590–1602. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Hill SY, De Bellis MD, Keshavan MS, Lowers L, Shen S, Hall J, & Pitts T (2001). Right amygdala volume in adolescent and young adult offspring from families at high risk for developing alcoholism. Biological Psychiatry, 49(11), 894–905. [DOI] [PubMed] [Google Scholar]
  33. Hill SY, Wang S, Carter H, McDermott MD, Zezza N, & Stiffler S (2013). Amygdala volume in offspring from multiplex for alcohol dependence families: the moderating influence of childhood environment and 5-HTTLPR variation. Journal of alcoholism and drug dependence [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Hill SY, Tessner K, Wang S, Carter H, & McDermott M (2010). Temperament at 5 years of age predicts amygdala and orbitofrontal volume in the right hemisphere in adolescence. Psychiatry Research: Neuroimaging, 182(1), 14–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Hollingshead AB (1975). Four factor index of social status.
  36. Iglesias JE, Augustinack JC, Nguyen K, Player CM, Player A, Wright M, Roy N, Frosch MP, McKee AC, Wald LL, Fischl B, Van Leemput K; Alzheimer’s Disease Neuroimaging Initiative (2015). A computational atlas of the hippocampal formation using ex vivo , ultra-high resolution MRI: Application to adaptive segmentation of in vivo MRI. NeuroImage, 115, 117–137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Indovina I, Robbins TW, Núñez-Elizalde AO, Dunn BD, & Bishop SJ (2011). Fear-Conditioning Mechanisms Associated with Trait Vulnerability to Anxiety in Humans. Neuron, 69(3), 563–571. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Johnson JG, Cohen P, Kasen S, Smailes E, & Brook JS (2001). Association of Maladaptive Parental Behavior With Psychiatric Disorder Among Parents and Their Offspring. Archives of General Psychiatry, 58(5), 453–460. [DOI] [PubMed] [Google Scholar]
  39. Jordan CJ, & Andersen SL (2017). Sensitive periods of substance abuse: Early risk for the transition to dependence. Developmental Cognitive Neuroscience, 25, 29–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Kendler KS, Prescott CA, Myers J, & Neale MC (2003). The Structure of Genetic and Environmental Risk Factors for Common Psychiatric and Substance Use Disorders in Men and Women. Archives of General Psychiatry, 60(9), 929–937. [DOI] [PubMed] [Google Scholar]
  41. Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, & Walters EE (2005). Lifetime Prevalence and Age-of-Onset Distributions of DSM-IV Disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62(6), 593–602. [DOI] [PubMed] [Google Scholar]
  42. Kilpatrick DG, Acierno R, Saunders B, Resnick HS, Best CL, & Schnurr PP (2000). Risk factors for adolescent substance abuse and dependence: Data from a national sample. Journal of Consulting and Clinical Psychology, 68(1), 19. [DOI] [PubMed] [Google Scholar]
  43. Koolschijn PC, Crone EA (2013). Sex differences and structural brain maturation from childhood to early adulthood. Developmental Cognitive Neuroscience, 5, 106–118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. MacMillan S, Szeszko PR, Moore GJ, Madden R, Lorch E, Ivey J, Banerjee SP, Rosenberg DR (2003). Increased amygdala: hippocampal volume ratios associated with severity of anxiety in pediatric major depression. Journal of child and adolescent psychopharmacology, 13(1), 65–73. [DOI] [PubMed] [Google Scholar]
  45. Miller EK, & Cohen JD (2001). An integrative theory of prefrontal cortex function. Annual review of neuroscience, 24(1), 167–202. [DOI] [PubMed] [Google Scholar]
  46. Mills KL, Goddings AL, Clasen LS, Giedd JN, & Blakemore SJ (2014). The developmental mismatch in structural brain maturation during adolescence. Developmental neuroscience, 36(3–4), 147–160. [DOI] [PubMed] [Google Scholar]
  47. Morey RA, Petty CM, Xu Y, Hayes JP, Wagner II HR, Lewis DV, LaBar KS, Styner M, McCarthy G. (2009). A comparison of automated segmentation and manual tracing for quantifying hippocampal and amygdala volumes. Neuroimage, 45(3), 855–866. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. O’Brien JW, & Hill SY (2017). Neural predictors of substance use disorders in Young adulthood. Psychiatry Research: Neuroimaging, 268, 22–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Phelps EA (2004). Human emotion and memory: interactions of the amygdala and hippocampal complex. Current Opinion in Neurobiology, 14(2), 198–202. [DOI] [PubMed] [Google Scholar]
  50. Phillips L, Velakoulis D, Pantelis C, Wood S, Yuen H, Yung A, Desmond P, Brewer W, McGorry PD (2002). Non-reduction in hippocampal volume is associated with higher risk of psychosis. Schizophrenia Research, 58(2–3), 145–158. [DOI] [PubMed] [Google Scholar]
  51. Ramirez F, Moscarello JM, LeDoux JE, & Sears RM (2015). Active Avoidance Requires a Serial Basal Amygdala to Nucleus Accumbens Shell Circuit. Journal of Neuroscience, 35(8), 3470–3477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Roth MC, Humphreys KL, King LS, & Gotlib IH (2018). Self-reported neglect, amygdala volume, and symptoms of anxiety in adolescent boys. Child abuse & neglect, 80, 80–89 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. SAS Institute Inc. (1989). JMP®. Cary, NC: SAS Institute Inc. [Google Scholar]
  54. Saygin ZM, Kliemann D, Iglesias JE, van der Kouwe AJ, Boyd E, Reuter M, Stevens A, Van Leemput K, McKee A, Frosch MP, Fischl B, Augustinack JC; Alzheimer’s Disease Neuroimaging Initiative (2017). High-resolution magnetic resonance imaging reveals nuclei of the human amygdala: manual segmentation to automatic atlas. NeuroImage, 155, 370–382. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Schmidt B, Marrone DF, & Markus EJ (2012). Disambiguating the similar: The dentate gyrus and pattern separation. Behavioural Brain Research, 226(1), 56–65. [DOI] [PubMed] [Google Scholar]
  56. Schweinsburg AD, Paulus MP, Barlett VC, Killeen LA, Caldwell LC, Pulido C, Brown SA, Tapert SF (2004). An fMRI Study of Response Inhibition in Youths with a Family History of Alcoholism. Annals of the New York Academy of Sciences, 1021(1), 391–394. [DOI] [PubMed] [Google Scholar]
  57. Sheehan DV, Sheehan KH, Shytle RD, Janavs J, Bannon Y, Rogers JE, & Wilkinson B (2010). Reliability and validity of the mini international neuropsychiatric interview for children and adolescents (MINI-KID). The Journal of Clinical Psychiatry. [DOI] [PubMed] [Google Scholar]
  58. Somerville LH, & Casey BJ (2010). Developmental neurobiology of cognitive control and motivational systems. Current Opinion in Neurobiology, 20(2), 236–241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Spielberger CD, Gorsuch RL, & Lushene RE (1970). The State-Trait Anxiety Inventory (Test Manual). Consulting Psychologists Press. [Google Scholar]
  60. Squeglia LM, & Gray KM (2016). Alcohol and Drug Use and the Developing Brain. Current Psychiatry Reports, 18(5), 2509. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Stano JF (2004). Wechsler abbreviated scale of intelligence. Rehabilitation Counseling Bulletin, 48(1). [Google Scholar]
  62. Stoltenberg SF, Mudd SA, Blow FC, & Hill EM (1998). Evaluating measures of family history of alcoholism: density versus dichotomy. Addiction, 93(10), 1511–1520. [DOI] [PubMed] [Google Scholar]
  63. Swartz JR, Williamson DE, & Hariri AR (2015). Developmental Change in Amygdala Reactivity During Adolescence: Effects of Family History of Depression and Stressful Life Events. American Journal of Psychiatry, 172(3), 276–283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Vaidya JG, Elmore AL, Wallace A, Langbehn DR, Kramer JR, Kuperman S, & O’Leary DS (2019). Association Between Age and Familial Risk for Alcoholism on Functional Connectivity in Adolescence. Journal of the American Academy of Child & Adolescent Psychiatry, 57(7), 692–701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Van Petten C (2004). Relationship between hippocampal volume and memory ability in healthy individuals across the lifespan: review and meta-analysis. Neuropsychologia, 42(10), 1394–1413. [DOI] [PubMed] [Google Scholar]
  66. Whitford TJ, Rennie CJ, Grieve SM, Clark CR, Gordon E, & Williams LM (2006). Brain maturation in adolescence: Concurrent changes in neuroanatomy and neurophysiology. Human Brain Mapping, 28(3), 228–237. 10.1002/hbm.20273 [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Wierenga L, Langen M, Ambrosino S, van Dijk S, Oranje B, & Durston S (2014). Typical development of basal ganglia, hippocampus, amygdala and cerebellum from age 7 to 24. NeuroImage, 96(C), 67–72. 10.1016/j.neuroimage.2014.03.072 [DOI] [PubMed] [Google Scholar]
  68. Zahr NM, Pohl KM, Saranathan M, Sullivan EV, & Pfefferbaum A (2019). Hippocampal subfield CA2+3 exhibits accelerated aging in Alcohol Use Disorder: A preliminary study. NeuroImage: Clinical, 22, 101764 10.1016/j.nicl.2019.101764 [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Zeidman P, & Maguire EA (2016). Anterior hippocampus: the anatomy of perception, imagination and episodic memory. Nature Reviews Neuroscience, 17(3), 173–182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Zucker RA (1994). Pathways to alcohol problems and alcoholism: A developmental account of the evidence for multiple alcoholism and for contextual contributions to risk In Zucker RA, Boyd G, & Howard J (Eds.), The development of alcohol problems: Exploring the biopsychosocial matrix of risk (NIAAA Research Monograph No. 26, Serial No. 94–3742, pp. 255–289). Bethesda, MD: National Institutes of Health. [Google Scholar]

RESOURCES