Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2008 Oct 20.
Published in final edited form as: Neuropsychologia. 2007 Jan 9;45(8):1767–1776. doi: 10.1016/j.neuropsychologia.2006.12.015

Verbal Episodic Memory Declines Prior to Diagnosis in Huntington’s Disease

Andrea C Solomon 1, Julie C Stout 1, Shannon A Johnson 1, Douglas R Langbehn 2, Elizabeth H Aylward 3, Jason Brandt 4, Christopher A Ross 4, Leigh Beglinger 5, Michael R Hayden 6, Karl Kieburtz 7, Elise Kayson 8, Elaine Julian-Baros 8, Kevin Duff 5, Mark Guttman 9, Martha Nance 10, David Oakes 11, Ira Shoulson 12, Elizabeth Penziner 5, Jane S Paulsen 13, the Predict-HD Investigators of the Huntington Study Group
PMCID: PMC2570315  NIHMSID: NIHMS20754  PMID: 17303196

Introduction

Huntington’s disease (HD) is a neurodegenerative disorder caused by the expansion of a normal CAG repeat within a gene on the short arm of chromosome 4 that codes for the protein huntingtin (Ho et al., 2001; The Huntington’s Disease Collaborative Research Group, 1993). HD is characterized by the gradual onset of motor signs, cognitive symptoms, and psychiatric disturbances. No single time-point captures the transition from health to disease; instead, minor abnormalities in motor, cognitive, and psychiatric functions gradually appear and worsen. Eventually, individuals with the HD CAG-expansion are given a clinical diagnosis of HD, which is based on findings in the neurological examination of unequivocal motor signs of chorea and/or bradykinesia. However, the earliest impact on a person’s functioning may be in the realm of cognitive symptoms, and these may develop independently of motor signs. This study examines verbal episodic memory in a sample of CAG-expanded individuals who have not yet been clinically diagnosed, and who represent a wide range of points along the continuum from health to disease. In particular, we relate memory performance to several neurobiological indices of disease progression and pathology: age- and genetically-based estimates of the number of years expected to diagnosis of HD, striatal volumes, and the presence of motor findings on neurological examination. For the sake of brevity, we use the term pre-HD to refer to these CAG-expanded individuals who may have some signs and symptoms consistent with HD, but who have not yet met traditional criteria for an HD diagnosis.

In individuals diagnosed with HD, impairments in verbal episodic memory have been well documented and are consistent with a memory profile common to a variety of subcortical dementias (Beatty, 1992; Butters, Salmon, & Heindel, 1994; Delis et al., 1991; Peavy et al., 1994; Zakzanis, 1998). Specifically, HD is associated with impairments in learning and retrieval, with relatively spared storage (Beatty, 1992; Butters et al., 1994; Hodges, 2000; Peavy et al., 1994; Perry & Hodges, 1996; Pillon, Deweer, Agid, & Dubois, 1993; Pillon et al., 1994; but see Brandt, Corwin, & Krafft, 1992). Comparatively few studies have directly assessed verbal episodic memory in pre-HD, and the results have been mixed. In several studies, memory impairments have not been detected (de Boo et al., 1999; Giordani et al., 1995; Rothlind, Brandt, Zee, Codori, & Folstein, 1993; Strauss & Brandt, 1990), yet others have found evidence of memory decline in the prodromal period. For example, Diamond and colleagues (1992) found that individuals with pre-HD had poorer memory performance on several tests when compared to non-CAG expanded individuals. Similarly, a pre-HD group demonstrated selective impairments in verbal memory on a word list learning task (Lundervold & Reinvang, 1995). In addition, longitudinal data from a sample of pre-HD participants indicates that declines in verbal learning and memory are detectable prior to clinical symptom onset (Lemiere, Decruyenaere, Evers-Kiebooms, Vandenbussche, & Dom, 2004). Specifically, studies of memory in pre-HD have documented significantly poorer performance, relative to non-CAG expanded individuals, on measures of learning (Rosenberg, Sorensen, & Christensen, 1995), semantic clustering (Rosenberg et al., 1995), recall (Hahn-Barma et al., 1998; Rosenberg et al., 1995), and recognition discriminability (Berrios et al., 2002; Lanto, Riege, Mazziotta, Pahl, & Phelps, 1990).

Several studies have documented structural and functional brain changes in CAG-expanded individuals who have not yet been diagnosed (Aylward et al., 1996; Aylward et al., 2000; Aylward et al., 2004; Kipps et al., 2005; Paulsen et al., 2006b; Reading et al., 2004). Although relatively few studies have investigated the association between cognition and brain changes in HD, a recent study showed that cognitive task performance correlated with cerebral white matter and striatal volumes in early HD (Beglinger et al., 2005). In pre-HD, striatal atrophy has been linked with poorer performance on measures of verbal learning and memory (Campodonico et al., 1998; Starkstein et al., 1992). Consistent with these findings, in a sample of individuals who were considered to be at increased risk for HD by virtue of having an affected parent (but not confirmed by linkage or DNA analysis), PET imaging displayed reduced metabolic ratios in frontal, striatal, and insular regions that predicted higher false positive rates on a measure of recognition memory (Lanto et al., 1990). In the current study, we investigated relationships between MRI striatal volumes and verbal episodic memory performance.

Inconsistency in the foregoing studies of learning and memory in pre-HD may be attributable to several possible causes which continue to provide challenges for this line of research. First, study samples were constituted in varied ways. For example, some studies have included individuals that are at-risk for HD based on family history, but for whom genetic confirmation of the HD gene expansion was not available at the time; inclusion of non gene-expanded individuals in these samples would clearly decrease the sensitivity for detecting decrements in memory performance. In addition, although subjects may, technically, have been pre-diagnosis, inconsistencies in inclusion criteria have resulted in some studies including individuals with minor neurological signs consistent with HD (but not sufficient for diagnosis), whereas other studies have excluded such individuals. Similarly, as reliable methods for estimating proximity to traditionally-defined disease onset were not developed until relatively recently, this factor was not carefully controlled within and/or across many of the early studies; as such, some samples may have included individuals very close to onset, whereas others may have been comprised of individuals far from onset. Inconsistent findings may also reflect the fact that many studies have included too few subjects to yield sufficient power to reliably detect all but the largest effects. Finally, the memory tests used in previous studies have varying degrees of sensitivity due to the level of test difficulty and other psychometric properties. Each of these study factors contributes to the lack of clarity regarding the presence, severity, and nature of verbal episodic memory declines prior to the diagnosis of HD.

Thus, although it is clear that verbal episodic memory is affected in individuals with diagnosed HD, and evidence suggests that subtle changes in memory may actually begin during the prodromal phase of HD, many questions remain. The current study examines memory in pre-HD gene carriers both with and without minor neurological signs, and with a sample large enough to provide adequate power for examining several covariates or individual difference factors that may help account for variability in memory performance in pre-HD. This study is part of the Predict-HD project (Paulsen et al., 2006a), an ongoing longitudinal study aimed at identifying cognitive, neurological, and psychiatric markers that track disease progression in early HD. Consistent with the broader goals of Predict-HD, the current study provides a focused look at memory function in pre-HD and examines the sensitivity of a common measure of memory for detecting subtle declines early in the disease. We anticipated that estimates of closer proximity to disease diagnosis based on age, CAG repeat length, and motor signs would predict poorer memory performance. In addition, given previous findings that suggest a link between striatal atrophy and memory decline in pre-HD, we expected that memory performance would be related to striatal volumes.

Method

Participants

Data were collected at 24 sites in the United States, Canada, and Australia, as part of the baseline cognitive assessment of the Predict-HD project. All procedures were approved by the appropriate ethics committees and have therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. All participants gave informed consent prior to inclusion in the study. At the time this manuscript was written, 490 subjects had been recruited using contacts through neurological clinics and genetic testing programs, as well as by regional informational talks, brochure distributions, advertisement through HD lay organizations, mail solicitations, and recruiter attendance at conferences. Inclusion criteria required participants to have undergone genetic testing for the presence of the CAG expansion in the HD gene. Exclusion criteria included: clinical evidence of unstable medical or psychiatric illness; alcohol or drug abuse within the previous year; learning disability or mental retardation requiring special education; history of other central nervous system disease or event, such as seizures or head trauma; pacemaker or metallic implants; age less than 26 years; prescribed antipsychotic medications within the past six months; and use of phenothiazine-derivative anti-emetic medications for at least three months. Other prescribed, over-the-counter and natural remedies were not restricted.

Participant Characterization

DNA testing and estimates of years to diagnosis

A previous voluntary decision to undergo HD gene testing was a prerequisite for entry into the Predict-HD study, and confirmatory DNA testing was completed for the study using blood drawn at the baseline visit. CAG repeat numbers were obtained using a polymerase chain reaction (PCR) method. HD-specific oligonucleotide primers, flanking the HD CAG repeat, were used to specifically amplify the CAG repeat from template DNA samples. The resultant radiolabeled HD-specific repeat PCR product was displayed on a DNA sequencing gel, exposed to X-ray film, and compared to a quantified standard gel to determine the CAG repeat numbers. Based on these results, participants were designated as either CAG-Expanded (CAG-Exp; CAG length ≥ 39, n = 439) or CAG-Normal (CAG-Norm; CAG length < 30, n = 51). Control (CAG-Norm) participants were intentionally sampled at a rate of 1 to 7.

Two prognostic variables were used to estimate proximity to traditionally defined disease onset (i.e., clinical diagnosis based on unequivocal motor signs of HD): 1) probability of clinical diagnosis in 5 years, and 2) difference from parent age of onset. We estimated the proximity to clinical diagnosis of HD for each CAG-Exp participant based on current age and CAG repeat length as per the formulas described by Langbehn et al. (2004). The probability of diagnosis within the next 5 years can be estimated from a participant’s survival-analysis-based formulation. We use this probability rather than estimated years to diagnosis because preliminary analyses of relationships with striatal volumes and other clinical variables suggests that more linear and thus more easily modeled relationships are likely using the probability scale (Paulsen et al., under review). The difference from parent onset age was calculated by subtracting the participant’s current age from the parent’s age at disease onset; this number serves as an indirect control for possible additional environmental or minor genetic influences on the age of HD onset (Djousse et al., 2003; Li et al., 2003). Thus, we include parent onset age as a possible source of variance in HVLT-R performance independent of CAG length.

Neurological examination

Participants underwent a standardized neurological examination (UHDRS; Huntington Study Group, 1996) by a trained examiner who then assigned a rating based on the examiner’s confidence that the observed abnormalities in motor function (if any) were a manifestation of HD. Ratings corresponded to the following characterizations: 0 = normal (no abnormalities), 1 = non-specific motor abnormalities (less than 50% confidence), 2 = motor abnormalities that may be signs of HD (50–89% confidence), 3 = motor abnormalities that are likely signs of HD (90–98% confidence), 4 = motor abnormalities that are unequivocal signs of HD (≥ 99% confidence). In the CAG-Norm group, we observed the following base rates: 61% (31 of 51) had confidence level ratings of 0; 33% (17 of 51) had confidence level ratings of 1; and 6% (3 of 51) had confidence level ratings of 2. As expected, multivariate analysis of variance (MANOVA) revealed no effect of diagnostic confidence ratings within the CAG-Norm sample on memory performance, F(10, 88) = .89, p >.1. Thus, for all subsequent analyses, we collapsed across motor ratings within the CAG-Norm group, resulting in five subgroups: CAG-Norm (n = 51), and 4 CAG-Exp groups based on confidence level ratings: CL0 (n = 144), CL1 (n = 193), CL2 (n = 67), and CL3 (n = 24). Given the small number of participants with the confidence level rating of 4 (n = 11), and given our focus on the period prior to diagnosis of HD, participants with a motor rating of 4 were excluded from all analyses. Thus, the overall sample size was 479 (428 CAG-Exp and 51 CAG-Norm). Demographic and other participant characteristics for the CAG-Norm and CAG-Exp subgroups are included in Table 1. MANOVA revealed significant group differences for age, F(3, 424) = 3.49, p < .05, probability of diagnosis in 5 years, F(3, 424) = 19.77, p < .001, and estimated general intellectual function, F(3, 424) = 2.74, p < .05.

Table 1.

Participant characteristics and HVLT-R summary statistics for the CAG-Norm and four CAG-Exp groups.

CAG-Norm (n = 51) CL0 (n = 144) CL1 (n = 193) CL2 (n = 67) CL3 (n = 24)
Participant Characteristics
Age * [Mean (SD)] 43.2 (10.3) 39.9 (10.0) 41.4 (9.1) 44.2 (10.4) 43.8 (11.1)
Prob5yr † (Mean) n/a 15.1% 22% 28% 39%
Gender (% Female) 60.8% 66.0% 64.8% 52.2% 66.7%
Education in years [Mean (SD)] 15.1 (2.4) 14.7 (2.7) 14.4 (2.5) 14.5 (2.6) 14.1 (3.1)
ANART errors * [Mean (SD)] 15.4 (7.1) Mean Est IQ = 115 15.7 (7.4) Mean Est IQ = 114 16.7 (7.8) Mean Est IQ = 113 18.6 (8.7) Mean Est IQ = 111 19.1 (10.6) Mean Est IQ = 111
Striatal volume [Mean (SD)] 15.4 (2.8) (n = 11) 14.4 (2.3) (n = 61) 14.5 (2.7) (n = 84) 13.3 (2.6) (n = 29) 11.9 (2.3) (n = 8)
HVLT-R Variables
Mean (SD) Range Mean (SD) Range Mean (SD) Range Mean (SD) Range Mean (SD) Range
Learn 1 7.8 (1.7) 4 – 11 7.3 (1.9) 2 – 11 6.7 (2.0) 2 – 12 6.3 (2.1) 2 – 11 5.4 (1.9) 1 – 9
Learn 2 10.4 (1.4) 6 – 12 9.8 (1.8) 4 – 12 9.2 (2.0) 4 – 12 8.5 (2.2) 3 – 12 7.7 (2.2) 5 – 12
Learn 3 10.8 (1.1) 8 – 12 10.7 (1.4) 5 – 12 10.1 (1.7) 5 – 12 9.7 (1.9) 4 – 12 9.3 (1.9) 5 – 12
Learn Total 29.0 (3.5) 20 – 35 27.7 (4.4) 15 – 35 26.0 (4.9) 13 – 35 24.4 (5.4) 9 – 35 22.3 (5.3) 13 – 33
Delay 10.3 (1.9) 4 – 12 10.1 (2.0) 4 – 12 9.4 (2.2) 4 – 12 8.6 (2.6) 1 – 12 8.5 (2.1) 5 – 12
Retention (%) 92.6 (13.9) 40 – 111 92.5 (13.2) 36 – 120 90.5 (14.7) 44 – 122 87.0 (18.2) 20 – 120 90.4 (16.5) 56 – 129
Hits 11.8 (0.8) 8 – 12 11.8 (0.6) 8 – 12 11.5 (0.9) 8 – 12 11.5 (0.8) 9 – 12 11.1 (1.1) 9 – 12
Misses 0.3 (0.8) 0 – 4 0.2 (0.6) 0 – 4 0.5 (0.9) 0 – 4 0.5 (0.8) 0 – 3 0.9 (1.1) 0 – 3
Related FP 0.7 (0.8) 0 – 4 0.9 (0.9) 0 – 5 1.2 (1.0) 0 – 5 1.2 (1.1) 0 – 4 1.6 (1.2) 0 – 4
Unrelated FP 0.0 (0.0) 0 – 0 0.02 (0.1) 0 – 1 0.07 (0.3) 0 – 2 0.1 (0.4) 0 – 2 0.04 (0.2) 0 – 1
Total FP 0.7 (0.8) 0 – 4 1.0 (1.0) 0 – 6 1.2 (1.1) 0 – 5 1.3 (1.3) 0 – 5 1.7 (1.2) 0 – 4
True Negatives 11.3 (0.8) 8 – 12 11.0 (1.0) 6 – 12 10.8 (1.1) 7 – 12 10.7 (1.3) 7 – 12 10.3 (1.2) 8 – 12
Discriminability 11.0 (1.0) 8 – 12 10.8 (1.2) 6 – 12 10.3 (1.5) 5 – 12 10.2 (1.6) 6 – 12 9.4 (1.7) 6 – 12
Open in a new tab

Note: Striatal volumes were available for 11 CAG-Norm, 61 CL0, 84 CL1, 29 CL2, and 8 CL3 participants. Prob5yr is the probability of diagnosis in 5 years. Mean Est IQ is the average estimated intelligence quotient. Total learning score (Learn Total) is the sum of words correctly recalled on trials 1, 2, and 3. Delayed recall score (Delay) is the number of words correctly recalled on the delayed recall trial. Retention is the percentage of words correctly recalled on the higher of trial 2 or 3 that were correctly recalled on the delayed recall trial. The recognition test yielded numbers of hits, misses, true negatives, semantically-related false positives (Related FP), and semantically-unrelated false positives (Unrelated FP). Related and unrelated false positives were summed to obtain total false positives (Total FP). A discriminability index was calculated by subtracting the total number of false positives from the number of hits. MANOVA revealed significant group differences (* p < .05, † p < .001; denoted in table) for age, F(3, 424) = 3.49, p < .05, prob5yr, F(3, 424) = 19.77, p < .001, and estimated general intellectual function, F(3, 424) = 2.74, p < .05.

Estimate of pre-morbid intellectual function

The American National Adult Reading Test (ANART; Gladsjo, Heaton, Palmer, Taylor, & Jeste, 1999) is a 50-word oral reading test used to estimate general intellectual functioning. ANART performance was covaried in all analyses to control for the influence of general intellectual function on memory performance, and an estimated intelligence quotient (IQ) was calculated from the raw score (number of errors) on this task (see Table 1).

Volumetric analyses of striatum

MRIs were obtained using a standard protocol designed to optimize visualization of the basal ganglia. All MRIs were obtained using a 1.5 Tesla magnetic field; 23 sites used a General Electric model and one site used a Siemens. In addition to a sagittal localizing series, we obtained an axial 3D volumetric spoiled gradient echo series, with a flip angle of 20 degrees, TE = 3, TR = 18, FOV = 24cm, 124 slices at 1.5mm/slice, matrix 256 × 192, ¾ phase FOV, NEX = 2. Total scanning time was approximately 15 minutes. A single rater, who had been trained to >.90 for inter-rater reliability with an expert rater (E.H.A.), was blinded to participant characteristics and completed all measurements for the study. Measurements were made by manually drawing boundaries of the caudate (head and body) and putamen, as previously described (Aylward et al., 1996; Aylward et al., 1997), and volumes were calculated based on the number of identified pixels. Caudate and putamen volumes were summed to obtain a measure of total striatal volume. The data were analyzed in the order received. MRI data from 193 of the Predict participants had been analyzed at the time of this report (see Table 1). The MRI sample is smaller because of a lag induced by the extra processing steps needed to produce the volumetric data and does not reflect any other known sampling bias.

Procedures

As part of the Predict-HD study, the Hopkins Verbal Learning Test – Revised (HVLT-R; Brandt & Benedict, 2001), a 12-item learning task comprised of words from three semantic categories, was administered according to standardized instructions. The list was presented orally for three learning trials, and immediately following each presentation, the participant was asked to recall as many words from the list as possible. The participant was not informed that additional trials would follow. After 20–25 minutes, the participant was again asked to recall as many list words as possible. Following this delayed recall trial, a recognition test was administered, in which the participant was read a list of 24 words (the 12 words from the learning list and 12 distractors) and asked to identify words from the learning list. Two versions of the HVLT-R (Forms 2 [n=284] and 4 [n=191]) were used in the study, counterbalanced by site. In Table 1, the 13 variables characterizing HVLT-R performance are listed and defined, and their descriptive statistics, including means, standard deviations, and minimum and maximum values, are shown separately for the CAG-Norm and the four CAG-Exp groups. We opted to focus our analyses on 5 target HVLT-R variables, which were selected according to the predominant neuropsychological model of learning and memory, based on information processing models (Atkinson & Shiffrin, 1968; Atkinson & Shiffrin, 1971): to represent encoding, we selected trial 1 learning and total learning; for retrieval we used delayed recall; and for storage, the variables were recognition discriminability and retention. These variables are all significantly correlated with each other (r values ranging from .35 to .87, p < .001) and contain some redundant raw measures. As described below, we statistically controlled for this redundancy by completing an initial multivariate test.

Statistical Analyses

For the CAG-Exp participants, the five target variables from the HVLT-R were examined in two separate hierarchical regression models, one to examine the influence of estimates of proximity to disease onset (the prognostic model), and the other to assess the influence of striatal volumes (the striatal model). To control for individual differences, demographic variables (age, sex, years of education) and estimated IQ (ANART score) were entered into the models first. In the prognostic model, probability of diagnosis in 5 years and the difference from parent age of onset were then entered together in a separate step to examine the incremental contribution of these variables. In the striatal model, the incremental contribution of striatal volume was examined by adding this variable to the model in a separate step after demographic variables and estimated IQ. The CAG-Norm sample was excluded from these analyses; the prognostic (and probably the striatal) model is irrelevant as CAG-Norm participants have no estimated proximity to onset. Furthermore, we were unable to examine the striatal model in the CAG-Norm sample because the sample size was too small at the time these data were analyzed.

To control for multiple hypothesis testing across the five HVLT-R measures, each of the above models was fit in a multivariate outcome framework (i.e., theoretically identical to a MANOVA, but with a mixture of categorical and continuous predictor variables). The statistical significance of relationships between prognostic or striatal measures and specific HVLT-R outcomes was only examined if Wilks’ lambda test of joint significance across all measures was significant at p ≤ .05 (Rencher & Scott, 1990). Multivariate outcome regressions were also completed for the CL0 and CL1 groups separately to assess whether findings were present in groups with no or minimal motor signs. Finally, MANCOVA (controlling for age and estimated IQ) was used to compare HVLT-R performance across five groups: CAG-Norm and CAG-Exp with confidence level ratings of 0, 1, 2, and 3.

Preliminary Analyses

There was no significant effect of test form (HVLT Form 2 compared to Form 4) for trial 1 learning, total learning, delayed recall, or retention; however, for recognition discriminability, participants tested with Form 4 had slightly higher discriminability scores, F(1, 473) = 8.3, p < .01. Given the small absolute difference (less than ½ point) in the scores on the two forms, we opted to collapse across forms. To check for collinearity in the data, we examined tolerance levels, which were all well above the recommended cutoff of .10 (Cohen, Cohen, West, & Aiken, 2003).

Results

Primary Analyses: Tests of Major Hypotheses

Memory & proximity to disease onset

The multivariate test across HVLT-R variables revealed an overall effect of probability of diagnosis, F(5, 385) = 8.53, p < .0001. Univariate analyses indicated that higher probability of diagnosis in 5 years was independently associated with poorer performance on trial 1 learning, total learning, delayed recall, and recognition discriminability, p < .001 (see Table 2), but not retention, after controlling for demographic variables and estimated IQ. The difference from parent age of onset was unrelated to performance on any of the HVLT-R measures. The full regression model, including prognostic, demographic, and estimated IQ variables, accounted for a significant portion of the variance in all 5 HVLT-R indices (see Table 3), with adjusted R2 values ranging from .06 (retention) to .23 (total learning). Together, the prognostic variables (probability of diagnosis in 5 years and difference from parent age of onset) were significantly associated with all of the HVLT-R measures except retention, after controlling for demographics and estimated IQ, p < .001.

Table 2.

Standardized beta coefficients and semi-partial correlation coefficients for IVs in full prognostic and striatal models.

Prognostic/Striatal Variables Demographic Variables
Prob5yr Parent-onset Age Sex Educ ANART errors
Std β sr Std β sr Std β sr Std β sr Std β sr Std β Sr
Learn 1 −.25† −.21 −.05 −.04 −.03 −.03 −.15** −.15 .03 .02 −.30† −.24
Learn Total −.32† −.27 −.04 −.03 .01 .004 −.15† −.14 .06 .05 −.33† −.26
Delay −.27† −.23 .02 .02 .002 .002 −.10* −.10 .05 .04 −.32† −.26
Retention −.08 −.07 .04 .03 −.01 −.01 −.08 −.08 −.02 −.02 −.24† −.19
Discrim −.23† −.19 −.01 −.01 −.02 −.02 −.04 −.04 .05 .04 −.24† −.19
Striatum Age Sex Educ ANART errors
Std β sr Std β sr Std β sr Std β sr Std β sr
Learn 1 .06 .06 −.10 −.09 −.27† −.26 .06 .05 −.33† −.27
Learn Total .16* .15 −.13 −.12 −.27† −.26 .06 .05 −.38† −.31
Delay .09 .09 −.17* −.16 −.22** −.21 .05 .05 −.38† −.31
Retention .00 .00 −.11 −.10 −.13 −.13 .01 .01 −.26** −.21
Discrim .18* .17 −.10 −.09 −.13 −.12 .01 .01 −.36† −.29
Open in a new tab

Note: Standardized beta coefficients (β) express the linear relationship between individual predictor variables and the dependent variable in terms of predicted standard deviations of change of the dependent variable for each standard deviation of change of the predictor variable. Semi-partial correlation coefficients (sr) represent the correlation between a predictor variable and the dependent variable after the predictor variable has been regressed on all other predictors. The square of the semi-partial correlation represents the predictor variable’s unique contribution to the full model R2. Educ is years of education; prob5yr is the probability of diagnosis in 5 years; parent-onset is the difference from parent age of onset.

*

p < .05

**

p ≤ .01

†

p ≤ .001

Table 3.

Summary and change statistics for full prognostic and striatal models.

Prognostic Model (n = 400)
F R2 adj R2 change F change
Learn 1 13.13† .15 .04 10.46†
Learn Total 20.64† .23 .08 20.48†
Delay 16.96† .19 .06 15.28†
Retention 5.07† .06 .01 1.77
Discrim 9.20† .11 .04 9.38†
Striatal Model (n = 182)
F R2 adj R2 change F change
Learn 1 8.35† .17 .00 .65
Learn Total 12.45† .24 .02 5.10*
Delay 10.11† .20 .01 1.67
Retention 3.14** .06 .00 .00
Discrim 7.78† .16 .03 5.99*
Open in a new tab

Note: Change statistics represent the degree to which the final step of the model (prob5yr and parent-onset for the prognostic model; striatal volume for the striatal model) increased the model’s ability to predict HVLT-R performance.

*

p < .05

**

p ≤ .01

†

p ≤ .001

Memory & striatal volumes

The multivariate test revealed an overall effect of striatal volume across the five HVLT-R variables, F(5, 171) = 3.11, p = .01. Univariate analyses showed that smaller striatal volumes were independently associated with lower scores on total learning and recognition discriminability, after controlling for demographic variables and estimated IQ, p < .05 (see Table 2). The full striatal regression model, including demographic variables, estimated IQ, and striatal volumes, accounted for a significant portion of variance in all 5 HVLT-R variables (see Table 3), with adjusted R2 values ranging from .06 (retention) to .24 (total learning).

Secondary Analyses

The influence of individual difference factors

Multivariate tests across the five HVLT-R variables revealed significant effects of sex, F(5, 385) = 3.72, p < .01, and estimated premorbid intelligence, F(5, 385) = 7.88, p < .0001, in the prognostic model as well as in the striatal model, sex, F(5, 171) = 3.45, p < .01, estimated premorbid intelligence, F(5, 171) = 5.55, p < .001. Univariate analyses indicated that higher estimates of intellectual functioning (i.e., lower number of ANART errors) were independently associated with better scores on all five HVLT-R indices in both models, p ≤ .001 for prognostic model, p ≤ .01 for striatal model (see Table 2). Sex was also independently related to HVLT-R performance in both models; women performed better than men on measures of learning (trial 1 learning, total learning) and retrieval (delayed recall) in both models, p ≤ .05 for prognostic model, p ≤ .01 for striatal model. These data are consistent with the well-documented link between general intellectual functioning and performance on cognitive tasks, as well as evidence suggesting that women typically outperform men on selective verbal tests (Lezak, 1995). These findings highlight the need to control for individual differences when examining cognitive performance.

Findings in confidence level rating groups 0 and 1

The relationship between memory performance and probability of diagnosis in 5 years was observed even in the subgroup with a confidence level rating of 0 (i.e., the subgroup with no motor signs), multivariate F(5, 125) = 2.75, p < .05. Specifically, univariate analyses revealed that higher probability of diagnosis in 5 years was independently associated with lower scores on total learning and delayed recall in the CL0 group, p < .01. There was also a trend for higher probability of diagnosis in 5 years to be associated with lower HVLT-R scores in the CL1 group, although the multivariate test did not reach statistical significance, F(5, 170) = 2.1, p = .07.

In the second model, smaller striatal volumes were associated with lower HVLT-R scores in the CL0 group, multivariate F(5, 51) = 4.25, p < .01. Specifically, univariate analyses indicated that smaller striatal volumes were independently associated with lower recognition discriminability scores in the CL0 group, p ≤ .001. Striatal volumes were unrelated to memory performance in the CL1 group.

Memory performance across confidence level ratings

To determine which HVLT-R measure(s) are most sensitive to early changes in pre-HD and to examine the pattern of memory performance at different stages in the pre-HD period, we compared HVLT-R performance across groups: CAG-Norm and CAG-Exp with confidence ratings of 0, 1, 2, and 3. The multivariate analysis revealed an overall effect of group on the set of HVLT-R variables, F(20, 1540) = 2.88, p < .001. Univariate tests for individual HVLT-R measures revealed a main effect of group on 4 HVLT-R indices: trial 1 learning, F(4, 468) = 6.54, p < .001; total learning, F(4, 468) = 10.69, p < .001; delayed recall, F(4, 468) = 5.90, p < .001; and recognition discriminability, F(4, 468) = 6.43, p < .001.

We completed post-hoc Bonferroni comparisons (see Figure 1) to examine differences between specific CAG-Exp groups and the CAG-Norm group; effect sizes (Cohen’s d) are shown in Table 4. There were no differences between the CL0 and the CAG-Norm groups on any HVLT-R measure. Learning (trial l learning, total learning) was significantly worse in the remaining CAG-Exp groups (CL1, CL2 and CL3), relative to the CAG-Norm group. Delayed recall was also worse in the CAG-Exp groups with more pronounced motor signs (CL2 and CL3). Finally, recognition discriminability was worse in the CL1 and CL3 groups (a similar trend in the CL2 group did not reach significance). The retention variable from the HVLT-R did not distinguish any of the CAG-Exp groups from the CAG-Norm group. As shown in Table 4, the effects we observed were generally small to medium; however, large effects were more consistently found in the group with the most pronounced motor signs.

Figure 1.

Figure 1

Open in a new tab

Bargraphs of HVLT-R performance by confidence level group. Confidence level ratings correspond to the following characterizations: 0 = normal (no abnormalities), 1 = non-specific motor abnormalities (less than 50% confidence), 2 = motor abnormalities that may be signs of HD (50–89% confidence), 3 = motor abnormalities that are likely signs of HD (90–98% confidence), 4 = motor abnormalities that are unequivocal signs of HD (≥ 99% confidence). As noted in the method section, individuals with confidence level ratings of 4 were excluded from all analyses. Bars represent means; error bars show standard error of the mean. Asterisks indicate differences from CAG-Norm group; * p ≤ .05, ** p ≤ .01, *** p ≤ .001

Table 4.

Effect sizes reflecting the magnitude of the difference between the CAG-Norm group and CAG-Exp groups on HVLT-R variables.

CL0 CL1 CL2 CL3
Learn 1 .30 .52 .63 1.09
Learn Total .30 .62 .84 1.30
Delay .13 .43 .66 .73
Retention .03 .12 .29 .05
Discrim .18 .49 .46 1.05
Open in a new tab

Note: Effect sizes (Cohen’s d) are the absolute value of the difference in adjusted means (after controlling for age and estimated IQ) divided by the pooled standard deviation. Cohen (1992) offers the following rough conventions for interpreting these effect sizes: small ≈ .2, medium ≈ .5, and large ≈ .8.

Discussion

This is the first study in pre-HD to demonstrate relationships between estimated proximity to traditionally defined onset of HD and verbal episodic memory; results indicate that performance on a task of verbal episodic memory, the Hopkins Verbal Learning Test – Revised, is associated with DNA-based estimates of probability of diagnosis within 5 years. These findings replicate similar previous results (Berrios et al., 2002; Campodonico et al., 1998; Hahn-Barma et al., 1998; Lemiere et al., 2004), but are in contrast to others (de Boo et al., 1999; Giordani et al., 1995; Rothlind et al., 1993; Strauss & Brandt, 1990). Given that effect sizes were small in the CL0 group and medium in the CL1 group, previous null findings may well reflect inadequate power to detect these subtle effects, especially in samples that include only individuals with a near or total absence of motor signs or who are very far from estimated age of onset. Nonetheless, our data clearly show that, with ample power, it is possible to detect early changes in verbal learning and memory prior to clinical diagnosis of HD. Importantly, the differences in performance observed here are very subtle and do not indicate memory ‘impairment’ in the clinical sense. Instead, these findings are useful in documenting that memory changes begin, in at least some CAG-expanded individuals, very early in the continuum from health to disease. Such information helps to describe how the burdens associated with cognitive decline unfold in CAG-expanded individuals, and points to the idea that forestalling such changes will be an important goal of preventive treatments.

Our results indicate that subtle declines in verbal learning and memory occur early in the prodromal phase of HD. In particular, the CAG-Exp group with minimal motor abnormalities (CL1) had significantly lower HVLT-R scores than the control group, particularly on measures of learning. Given the high base rate of confidence level ratings of 1 in the CAG-Norm sample (33%), which obviously cannot reflect motor signs of HD, it is also possible that the CAG-Exp CL1 group includes both individuals with minimal motor signs of HD as well as individuals who are asymptomatic and received a rating of 1 for other reasons. Thus, the finding that the CL1 group differed from the control group on several HVLT-R indices indicates that cognitive abilities begin to decline in HD well in advance of clinical diagnosis and perhaps prior to the onset of even minor neurological signs of HD. In addition, although the CL0 group did not differ from the CAG-Norm group on any of the HVLT-R measures, it is important to note that the significant relationships between memory performance and probability of diagnosis in 5 years and between memory and striatal volumes were observed even in this subgroup of individuals with no motor signs, indicating that our findings were not driven solely by participants with pronounced motor signs.

Interestingly, we found that groups with more pronounced motor signs had lower HVLT-R scores than did groups with no or minimal motor signs, as shown in Figure 1. Although speculative until confirmed by longitudinal analyses, the observation of poorer HVLT-R scores in those with higher motor ratings may suggest that verbal episodic memory function declines steadily as individuals approach clinical diagnosis (i.e., as they move farther along the continuum from health to disease). Future longitudinal analyses will allow us to examine the progression of cognitive symptoms and motor signs over time in this sample.

The finding that smaller striatal volumes were associated with lower HVLT-R scores is consistent with previous studies linking cognition and structural brain changes in HD (Beglinger et al., 2005; Campodonico et al., 1998; Starkstein et al., 1992) and suggests that cognitive symptoms in pre-HD are related to neuronal death or dysfunction in the basal ganglia. Neuropathology in the striatum interferes with the functioning of fronto-striatal circuitry, which results in cognitive (as well as motor and psychiatric) symptoms (Lichter, 2001). The dorsolateral prefrontal circuit in particular has been implicated in cognitive functions (Lichter & Cummings, 2001); in HD and pre-HD, executive dysfunction associated with dorsolateral prefrontal circuit damage is thought to interfere with strategic memory functions that facilitate efficient encoding and retrieval processes (Butters et al., 1994; Caine, Hunt, Weingartner, & Ebert, 1978; Delis et al., 1991; Hodges, 2000; Lichter, 2001; Pillon et al., 1994). Our observation of a relationship between learning and striatal volume is consistent with this model, although we did not find a relationship between our measure of retrieval (delayed recall) and striatal volume. Future studies might further examine relationships between striatal volume and memory by separately considering caudate and putamen volumes, as well as other parts of the fronto-striatal circuitry to add resolution to these findings. Future work might also examine the relationships between specific aspects of episodic memory and volumetric or functional measures of particular brain regions. For example, given the link between the dorsolateral prefrontal cortex and episodic memory retrieval (e.g., Buckner & Wheeler, 2001; Cabeza & Nyberg, 2000), as well as the well documented role of the hippocampus in memory encoding (e.g., Kirchhoff, Wagner, Maril, & Stern, 2000; Squire & Zola-Morgan, 1991), additional work is needed to elucidate the ways in which these various brain regions work in concert to support episodic memory function.

Our results point to several considerations for future use of the HVLT-R in pre-HD. Our target measures of encoding (i.e., trial one learning and total learning), yielded well-distributed data, but there were ceiling effects for our target measures of retrieval and storage (i.e., delayed recall, retention, and recognition discriminability). Nonetheless, our measure of retrieval and one of our storage measures were still sufficiently sensitive to show associations with estimated proximity to onset (delayed recall and recognition discriminability) and striatal volumes (recognition discriminability). Future research in pre-HD could adopt more difficult memory assessment tools in an attempt to add sensitivity in the assessment of retrieval and storage processes in pre-HD. Alternatively, because the HVLT-R has known sensitivity in manifest HD, our data suggest how this test can be appropriately applied in studies that include a wide range of CAG-expanded participants, ranging from far from onset to manifest HD, by taking into account the limitations in sensitivity of this subset of measures in individuals at the far from onset end of the disease spectrum. It is also important to note that in our sample, the 5 target HVLT-R measures were highly correlated with one another, and are, in some cases, partially redundant (e.g., learning trial 1 is included in the calculation of total learning score); thus, the interpretation of individual variables, as well as for specific memory processes, must be tempered. This likely reflects the nature of memory function in pre-HD, which is affected only subtly, and perhaps more uniformly across these memory processes; however, such strong interrelationships should not be assumed to be present in manifest HD or in other neurodegenerative memory disorders, in which examination of these variables separately may be better justified.

The generalizability of our results should be considered bearing in mind that the Predict-HD sample is more educated than the average population, and thus it is unknown whether the findings reported here generalize to groups with lower levels of education. Participants also know that they have the HD CAG expansion, which could potentially affect the results of cognitive and motor testing. Also, we note that our estimates of pre-morbid intellectual function varied significantly, but minimally, as a function of diagnostic confidence level ratings, indicating that the ANART may have some sensitivity to disease-related changes in cognitive ability. Thus, the inclusion of estimated premorbid intelligence as a covariate almost certainly makes our analysis of other cognitive associations with HD progression slightly conservative.

Our results suggest that subtle changes in learning and memory, and potentially other cognitive abilities, may serve as sensitive clinical markers of the earliest changes associated with the Huntington’s disease process. As such, these findings indicate that measures of verbal episodic memory may augment predictive models of HD. Ultimately, however, the relative importance of these memory findings must be assessed in the context of other cognitive abilities to determine their contribution within a more complete model of cognitive function in pre-HD. Longitudinal data on cognitive function in a substantially sized sample, which will soon be available from the Predict-HD study, will help to define the strength of memory predictors relative to other indications of cognitive change in individuals with the HD CAG-expansion.

Acknowledgments

National Institute of Neurological Disorders and Stroke grant # 40068, the National Institutes of Mental Health grant # 01579, Roy J. Carver Trust Medicine Research Initiative, Howard Hughes Medical Institute, and High Q Foundation grants to Jane S. Paulsen, the Huntington’s Disease Society of America, the Huntington’s Society of Canada, the Hereditary Disease Foundation, and the High Q Foundation grants to the Huntington Study Group, National Science Foundation Graduate Research Fellowship to Andrea C. Solomon.

Appendix

We gratefully thank the study participants for their time and effort. We also wish to acknowledge the investigators, coordinators, cognitive raters, and motor raters that participated in data collection for this study: Henry Paulson, MD, Rachel Conybeare, BS, Becky Reese, BS, and Ania Mikos, BA (University of Iowa Hospitals and Clinics); Adam Rosenblatt, MD, Christopher Ross, MD, PhD, Lisa Gourley, Barnett Shpritz, MS, and Arnold Bakker, MA (Johns Hopkins University); Lynn Raymond, MD, PhD, and Joji Decolongon, MSC (University of British Columbia); Randi Jones, PhD, Joan Harrison, RN, and Claudia Testa, MD, PhD (Emory University School of Medicine, Atlanta, Ga); Mark Guttman, MD(Centre for Addiction and Mental Health, University of Toronto); Elizabeth McCusker, MD, Jane Griffith, RN, Bernadette Bibb, PhD, and Catherine Lawson, PhD (Westmead Hospital, Wentworthville, Australia); Ali Samii, MD, Hillary Lipe, ARNP, and Rebecca Logsdon, PhD (University of Washington and VA Puget Sound Health Care System); Edmond Chiu, MD, Phyllis Chua, Olga Yastrubetskaya, PhD, and Phillip Dingjan, BA (The University of Melbourne, Kew, Victoria, Australia); Susan Perlman, MD, and Russell Carroll, PhD (University of California, Los Angeles Medical Center); Kimberly Quaid, PhD, Melissa Wesson, MS, and Joanne Wojcieszek, MD (Indiana University School of Medicine, Indianapolis); Michael D. Geschwind, MD, PhD, Katherine Rose, BA, and Christina Wyss-Coray, RN (University of California San Francisco); Joseph Jankovic, MD, William Ondo, MD, and Christine Hunter, RN, CCRC (Baylor College of Medicine, Houston, TX); Diana Rosas, MD, Lindsay Muir, and Alexandra Zaleta, BA (Massachusetts General Hospital, Boston); Oksana Suchowersky, MD, Mary Lou Klimek, RN, and Dolen Kirstein, BSC (University of Calgary, Calgary, Alberta); Vicki Wheelock, MD, Terry Tempkin, RNC, MSN, and Kathleen Baynes, PhD (University of California Davis, Sacramento); William M. Mallonne, MD, and Greg Suter, BA (Hereditary Neurological Disease Centre, Wichita, Kan); Pietro Mazzoni, MD, PhD, Jennifer Williamson, MS, and Paula Leber, MA (Columbia University Medical Center, New York, NY); Martha Nance, MD, Dawn Radtke, RN, and David Tupper, PhD (Hennepin County Medical Center, Minneapolis); Peter Panegyres, MB, BS, PhD, and Rachel Zombor, BSc(Hons) (Neurosciences Unit, Graylands, Selby-Lemnos & Special Care Health Services, Perth, Australia); Peter Como, PhD, Amy Chesire, CSW-R, Charlyne Hickey, RN, CMS, and Frederick Marshall, MD (University of Rochester); Lauren Seeberger, MD, and Sherrie Montellano, MA (Colorado Neurological Institute, Englewood); Richard Dubinsky, MD, Carolyn Gray, RN, and Phyllis Switzer, MBBS (University of Kansas Medical Center, Kansas City); Wayne Martin, MD, and Marguerite Wieler, MSc, PT (University of Alberta); and Joel Perlmutter, MD, and Melinda Kavanaugh, MSW, LCSW (Washington University, St Louis, Mo). University of Rochester (PI: Como, P.), Columbia-Presbyterian Medical Center (Co-I: Marder, K. & Mazzoni, P.), Baylor College of Medicine (PI: Jankovic, J.), Massachusetts General Hospital (PI: Rosas, D.), University of Iowa (PI: Paulson, H.), Washington University (PI: Perlmutter, J. S.), Johns Hopkins University (Co-I: Rosenblatt, A. & Ross, C.), University of Kansas Medical Center (PI: Dubinsky, R. M.), University of Calgary (Co-I: Furtado, S. & Suchowersky, O.), Emory University School of Medicine (PI: Jones, R.), The Centre for Addiction and Mental Health (PI: Guttman, M.), University of Alberta (PI: Martin, W.), Indiana University School of Medicine (PI: Quaid, K.), University of British Columbia (PI: Raymond, L. A.), UCLA Medical Center (PI: Perlman, S.), Colorado Neurological Institute (PI: Seeberger, L.), Westmead Hospital (PI: McCusker, E.), University of California Davis (PI: Wheelock, V. L.), University of Minnesota/Minnesota VA Medical Center (PI: Nance, M.), University of California San Francisco (PI: Geschwind, M. D.), Hereditary Neurological Disease Centre (HNDC) (PI: Mallonee, W. M.), Univ of Wash and VA Puget Sound Health Care System (PI: Samii, A.), St George’s Health Service (PI: Chiu, E.), Graylands, Selby-Lemnos & Special Care Health Services (Co-I: Connor, C. & Panegyres, P.), Royal Melbourne Hospital (PI: Chua, P.).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Atkinson RC, Shiffrin RM. Human memory: A proposed system and its control processes. In: Spence KW, Spense JT, editors. The psychology of learning and motivation: Advances in research and theory. Vol. 2. New York: Academic Press; 1968. pp. 89–195. [Google Scholar]
  2. Atkinson RC, Shiffrin RM. The control of short-term memory. Scientific American. 1971;225(2):82–90. doi: 10.1038/scientificamerican0871-82. [DOI] [PubMed] [Google Scholar]
  3. Aylward EH, Codori AM, Barta PE, Pearlson GD, Harris GJ, Brandt J. Basal ganglia volume and proximity to onset in presymptomatic Huntington disease. Archives of Neurology. 1996;53(12):1293–1296. doi: 10.1001/archneur.1996.00550120105023. [DOI] [PubMed] [Google Scholar]
  4. Aylward EH, Codori AM, Rosenblatt A, Sherr M, Brandt J, Stine OC. Rate of caudate atrophy in presymptomatic and symptomatic stages of Huntington’s disease. Movement Disorders. 2000;15(3):552–560. doi: 10.1002/1531-8257(200005)15:3<552::AID-MDS1020>3.0.CO;2-P. [DOI] [PubMed] [Google Scholar]
  5. Aylward EH, Li Q, Stine OC, Ranen N, Sherr M, Barta PE, et al. Longitudinal change in basal ganglia volume in patients with Huntington’s disease. Neurology. 1997;48(2):394–399. doi: 10.1212/wnl.48.2.394. [DOI] [PubMed] [Google Scholar]
  6. Aylward EH, Sparks BF, Field KM, Yallapragada V, Shpritz BD, Rosenblatt A, et al. Onset and rate of striatal atrophy in preclinical Huntington. Neurology. 2004;63(1):66–72. doi: 10.1212/01.wnl.0000132965.14653.d1. [DOI] [PubMed] [Google Scholar]
  7. Beatty WW. Memory dysfunction in the subcortical dementias. In: Backman L, editor. Memory functioning in dementia. Amsterdam: Elsevier Science; 1992. pp. 27–43. [Google Scholar]
  8. Beglinger LJ, Nopoulos PC, Jorge RE, Langbehn DR, Mikos AE, Moser DJ, et al. White matter volume and cognitive dysfunction in early Huntington’s disease. Cognitive and Behavioral Neurology. 2005;18(2):102–107. doi: 10.1097/01.wnn.0000152205.79033.73. [DOI] [PubMed] [Google Scholar]
  9. Berrios GE, Wagle AC, Markova IS, Wagle SA, Rosser A, Hodges JR. Psychiatric symptoms in neurologically asymptomatic Huntington’s disease gene carriers: A comparison with gene negative at risk subjects. Acta Psychiatrica Scandinavica. 2002;105(3):224–230. doi: 10.1034/j.1600-0447.2002.0o456.x. [DOI] [PubMed] [Google Scholar]
  10. Brandt J, Benedict RHB. Hopkins Verbal Learning Test-Revised. Lutz, FL: Psychological Assessment Resources; 2001. [Google Scholar]
  11. Brandt J, Corwin J, Krafft L. Is verbal recognition memory really different in Huntington’s and Alzheimer’s disease. Journal of Clinical and Experimental Neuropsychology. 1992;14(5):773–784. doi: 10.1080/01688639208402862. [DOI] [PubMed] [Google Scholar]
  12. Buckner RL, Wheeler ME. The cognitive neuroscience of remembering. Nature Reviews: Neuroscience. 2001;2(9):624–634. doi: 10.1038/35090048. [DOI] [PubMed] [Google Scholar]
  13. Butters N, Salmon D, Heindel WC. Specificity of the memory deficits associated with basal ganglia dysfunction. Revue Neurologique (Paris) 1994;150(8–9):580–587. [PubMed] [Google Scholar]
  14. Cabeza R, Nyberg L. Imaging cognition II: An empirical review of 275 PET and fMRI studies. Journal of Cognitive Neuroscience. 2000;12(1):1–47. doi: 10.1162/08989290051137585. [DOI] [PubMed] [Google Scholar]
  15. Caine ED, Hunt RD, Weingartner H, Ebert MH. Huntington’s dementia. Clinical and neuropsychological features. Archives of General Psychiatry. 1978;35(3):377–384. doi: 10.1001/archpsyc.1978.01770270127013. [DOI] [PubMed] [Google Scholar]
  16. Campodonico JR, Aylward E, Codori AM, Young C, Krafft L, Magdalinski M, et al. When does Huntington’s disease begin? Journal of the International Neuropsychological Society. 1998;4(5):467–473. doi: 10.1017/s1355617798455061. [DOI] [PubMed] [Google Scholar]
  17. Cohen J. A power primer. Psychological Bulletin. 1992;112(1):155–159. doi: 10.1037//0033-2909.112.1.155. [DOI] [PubMed] [Google Scholar]
  18. Cohen J, Cohen P, West SG, Aiken LS. Applied multiple regression/correlation analysis for the behavioral sciences. 3. Mahwah, NJ: Lawrence Erlbaum; 2003. [Google Scholar]
  19. de Boo GM, Tibben AA, Hermans JA, Jennekens-Schinkel A, Maat-Kievit A, Roos RA. Memory and learning are not impaired in presymptomatic individuals with an increased risk of Huntington’s disease. Journal of Clinical and Experimental Neuropsychology. 1999;21(6):831–836. doi: 10.1076/jcen.21.6.831.854. [DOI] [PubMed] [Google Scholar]
  20. Delis DC, Massman PJ, Butters N, Salmon D, Cermak LS, Kramer JH. Profiles of demented and amnesic patients on the California Verbal Learning Test: Implications for the assessment of memory disorders. Psychological Assessment. 1991;3(1):19–26. [Google Scholar]
  21. Diamond R, White RF, Myers RH, Mastromauro C, Koroshetz WJ, Butters N, et al. Evidence of presymptomatic cognitive decline in Huntington’s disease. Journal of Clinical and Experimental Neuropsychology. 1992;14(6):961–975. doi: 10.1080/01688639208402547. [DOI] [PubMed] [Google Scholar]
  22. Djousse L, Knowlton B, Hayden M, Almqvist EW, Brinkman R, Ross C, et al. Interaction of normal and expanded CAG repeat sizes influences age at onset of Huntington disease. American Journal of Medical Genetics, Part A. 2003;119(3):279–282. doi: 10.1002/ajmg.a.20190. [DOI] [PubMed] [Google Scholar]
  23. Giordani B, Berent S, Boivin MJ, Penney JB, Lehtinen S, Markel DS, et al. Longitudinal neuropsychological and genetic linkage analysis of persons at risk for Huntington’s disease. Archives of Neurology. 1995;52(1):59–64. doi: 10.1001/archneur.1995.00540250063014. [DOI] [PubMed] [Google Scholar]
  24. Gladsjo JA, Heaton RK, Palmer BW, Taylor MJ, Jeste DV. Use of oral reading to estimate premorbid intellectual and neuropsychological functioning. Journal of the International Neuropsychological Society. 1999;5(3):247–254. doi: 10.1017/s1355617799533079. [DOI] [PubMed] [Google Scholar]
  25. Hahn-Barma V, Deweer B, Durr A, Dode C, Feingold J, Pillon B, et al. Are cognitive changes the first symptoms of Huntington’s disease? A study of gene carriers. Journal of Neurology, Neurosurgery, and Psychiatry. 1998;64(2):172–177. doi: 10.1136/jnnp.64.2.172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Ho LW, Carmichael J, Swartz J, Wyttenbach A, Rankin J, Rubinsztein DC. The molecular biology of Huntington’s disease. Psychological Medicine. 2001;31(1):3–14. doi: 10.1017/s0033291799002871. [DOI] [PubMed] [Google Scholar]
  27. Hodges JR. Memory in the dementias. In: Tulving E, Craik FIM, editors. The Oxford handbook of memory. Oxford: University Press; 2000. pp. 441–459. [Google Scholar]
  28. Huntington Study Group. Unified Huntington’s Disease Rating Scale: Reliability and consistency. Movement Disorders. 1996;11(2):136–142. doi: 10.1002/mds.870110204. [DOI] [PubMed] [Google Scholar]
  29. Kipps CM, Duggins AJ, Mahant N, Gomes L, Ashburner J, McCusker EA. Progression of structural neuropathology in preclinical Huntington’s disease: A tensor based morphometry study. Journal of Neurology, Neurosurgery, and Psychiatry. 2005;76(5):650–655. doi: 10.1136/jnnp.2004.047993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kirchhoff BA, Wagner AD, Maril A, Stern CE. Prefrontal-temporal circuitry for episodic encoding and subsequent memory. Journal of Neuroscience. 2000;20(16):6173–6180. doi: 10.1523/JNEUROSCI.20-16-06173.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Langbehn DR, Brinkman RR, Falush D, Paulsen JS, Hayden MR on behalf of an International Huntington’s Disease Collaborative Group. A new model for prediction of the age of onset and penetrance for Huntington’s disease based on CAG length. Clinical Genetics. 2004;65(4):267–277. doi: 10.1111/j.1399-0004.2004.00241.x. [DOI] [PubMed] [Google Scholar]
  32. Lanto AB, Riege WH, Mazziotta JC, Pahl JJ, Phelps ME. Increased false alarms in a subset of persons at-risk for Huntington’s disease. Archives of Clinical Neuropsychology. 1990;5(4):393–404. [PubMed] [Google Scholar]
  33. Lemiere J, Decruyenaere M, Evers-Kiebooms G, Vandenbussche E, Dom R. Cognitive changes in patients with Huntington’s disease (HD) and asymptomatic carriers of the HD mutation--a longitudinal follow-up study. Journal of Neurology. 2004;251(8):935–942. doi: 10.1007/s00415-004-0461-9. [DOI] [PubMed] [Google Scholar]
  34. Lezak MD. Neuropsychological assessment. 3. New York: Oxford; 1995. [Google Scholar]
  35. Li JL, Hayden MR, Almqvist EW, Brinkman RR, Durr A, Dode C, et al. A genome scan for modifiers of age at onset in Huntington disease: The HD MAPS study. American Journal of Human Genetics. 2003;73(3):682–687. doi: 10.1086/378133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lichter DG. Movement disorders and frontal-subcortical circuits. In: Lichter DG, Cummings JL, editors. Frontal-subcortical circuits in psychiatric and neurological disorders. New York: Guilford; 2001. pp. 260–313. [Google Scholar]
  37. Lichter DG, Cummings JL. Introduction and overview. In: Lichter DG, Cummings JL, editors. Frontal-subcortical circuits in psychiatric and neurological disorders. New York: Guilford; 2001. pp. 1–43. [Google Scholar]
  38. Lundervold AJ, Reinvang I. Variability in cognitive function among persons at high genetic risk of Huntington’s disease. Acta Neurologica Scandinavica. 1995;91(6):462–469. doi: 10.1111/j.1600-0404.1995.tb00447.x. [DOI] [PubMed] [Google Scholar]
  39. Paulsen JS, Hayden M, Stout JC, Langbehn DR, Aylward E, Ross CA, et al. Preparing for preventive clinical trials: The Predict-HD study. Archives of Neurology. 2006a;63(6):883–890. doi: 10.1001/archneur.63.6.883. [DOI] [PubMed] [Google Scholar]
  40. Paulsen JS, Langbehn DR, Stout JC, Aylward E, Ross CA, Nance M, et al. Detection of brain disease decades before diagnosis: The Predict HD study. doi: 10.1136/jnnp.2007.128728. under review. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Paulsen JS, Magnotta VA, Mikos AE, Paulson HL, Penziner E, Andreasen NC, et al. Brain structure in preclinical Huntington’s disease. Biological Psychiatry. 2006b;59(1):57–63. doi: 10.1016/j.biopsych.2005.06.003. [DOI] [PubMed] [Google Scholar]
  42. Peavy G, Jacobs D, Salmon DP, Butters N, Delis DC, Taylor M, et al. Verbal memory performance of patients with human immunodeficiency virus infection: Evidence of subcortical dysfunction. The HNRC Group. Journal of Clinical and Experimental Neuropsychology. 1994;16(4):508–523. doi: 10.1080/01688639408402662. [DOI] [PubMed] [Google Scholar]
  43. Perry RJ, Hodges JR. Spectrum of memory dysfunction in degenerative disease. Current Opinion in Neurology. 1996;9(4):281–285. doi: 10.1097/00019052-199608000-00007. [DOI] [PubMed] [Google Scholar]
  44. Pillon B, Deweer B, Agid Y, Dubois B. Explicit memory in Alzheimer’s, Huntington’s, and Parkinson’s diseases. Archives of Neurology. 1993;50(4):374–379. doi: 10.1001/archneur.1993.00540040036010. [DOI] [PubMed] [Google Scholar]
  45. Pillon B, Deweer B, Michon A, Malapani C, Agid Y, Dubois B. Are explicit memory disorders of progressive supranuclear palsy related to damage to striatofrontal circuits? Comparison with Alzheimer’s, Parkinson’s, and Huntington’s diseases. Neurology. 1994;44(7):1264–1270. doi: 10.1212/wnl.44.7.1264. [DOI] [PubMed] [Google Scholar]
  46. Reading SA, Dziorny AC, Peroutka LA, Schreiber M, Gourley LM, Yallapragada V, et al. Functional brain changes in presymptomatic Huntington’s disease. Annals of Neurology. 2004;55(6):879–883. doi: 10.1002/ana.20121. [DOI] [PubMed] [Google Scholar]
  47. Rencher AC, Scott DT. Assessing the contribution of individual variables following rejection of a multivariate hypothesis. Communications in Statistics, Part B: Simulation and Computation. 1990;19:535–553. [Google Scholar]
  48. Rosenberg NK, Sorensen SA, Christensen AL. Neuropsychological characteristics of Huntington’s disease carriers: A double blind study. Journal of Medical Genetics. 1995;32(8):600–604. doi: 10.1136/jmg.32.8.600. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Rothlind JC, Brandt J, Zee D, Codori AM, Folstein S. Unimpaired verbal memory and oculomotor control in asymptomatic adults with the genetic marker for Huntington’s disease. Archives of Neurology. 1993;50(8):799–802. doi: 10.1001/archneur.1993.00540080010005. [DOI] [PubMed] [Google Scholar]
  50. Squire LR, Zola-Morgan S. The medial temporal lobe memory system. Science. 1991;253(5026):1380–1386. doi: 10.1126/science.1896849. [DOI] [PubMed] [Google Scholar]
  51. Starkstein SE, Brandt J, Bylsma F, Peyser C, Folstein M, Folstein SE. Neuropsychological correlates of brain atrophy in Huntington’s disease: A magnetic resonance imaging study. Neuroradiology. 1992;34(6):487–489. doi: 10.1007/BF00598956. [DOI] [PubMed] [Google Scholar]
  52. Strauss ME, Brandt J. Are there neuropsychologic manifestations of the gene for Huntington’s disease in asymptomatic, at-risk individuals? Archives of Neurology. 1990;47(8):905–908. doi: 10.1001/archneur.1990.00530080093016. [DOI] [PubMed] [Google Scholar]
  53. The Huntington’s Disease Collaborative Research Group. A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes. Cell. 1993;72(6):971–983. doi: 10.1016/0092-8674(93)90585-e. [DOI] [PubMed] [Google Scholar]
  54. Zakzanis KK. The subcortical dementia of Huntington’s disease. Journal of Clinical and Experimental Neuropsychology. 1998;20(4):565–578. doi: 10.1076/jcen.20.4.565.1468. [DOI] [PubMed] [Google Scholar]

RESOURCES