Retest effects and cognitive decline in longitudinal follow-up of patients with early HD A.–C. Bachoud–Lévi, MD, PhD; P. Maison, MD, PhD; P. Bartolomeo, MD, PhD; M.–F. Boissé, MSc; G. Dalla Barba, MD, PhD; A.–M. Ergis, PhD; S. Baudic, PhD; J.–D. Degos, MD; P. Cesaro, MD, PhD; and M. Peschanski, MD, PhD

Article abstract—Objective: To assess the natural progression of cognitive impairment in Huntington’s disease (HD) and to reveal factors that may mask this progression. Background: Although numerous cross-sectional studies reported cognitive deterioration at different stages of the disease, progressive cognitive deterioration has been, up to now, difficult to demonstrate in neuropsychological longitudinal studies. Methods: The authors assessed 22 patients in early stages of HD at yearly intervals for 2 to 4 years (average, 31.2 ⫾ 10 months), using a comprehensive neuropsychological battery based on the Core Assessment Program for Intracerebral Transplantation in Huntington’s Disease (CAPIT-HD). Results: The authors observed a significant decline in different cognitive functions over time: these involved primarily attention and executive functions but also involved language comprehension, and visuospatial immediate memory. Episodic memory impairment that was already present at the time of enrollment did not show significant decline. The authors found a significant retest effect at the second assessment in many tasks. Conclusion: Many attention and executive tasks adequately assess the progression of the disease at an early stage. For other functions, the overlapping of retest effects and disease progression may confuse the results. High interindividual and intraindividual variability seem to be hallmarks of the disease. NEUROLOGY 2001;56:1052–1058

The natural history of Huntington’s disease (HD) is ill-defined. Although most cross-sectional studies reported an early impairment in attention, planning,1 and memory,2-4 and a late impairment in visuospatial abilities5,6 and language2,7 (see Brandt8 for review), progressive cognitive deterioration has been, up to now, difficult to demonstrate in neuropsychological longitudinal studies. Whereas clinical trials are currently ongoing, more longitudinal data are required to understand the source of these difficulties to enable design of better protocols. Three published longitudinal studies have attempted to define a slope of decline. In a cohort of mild HD patients (stages II and III), performance at episodic and semantic memory remained stable over 1 year9; decline was observed only in the Selective Reminding Test and letter fluency.10 Using four grouping tests (“psychomotor” [reflecting attention and planning], “memory,” “visual,” and “semantic knowledge”), a second prospective study conducted for 42 months in early Huntington’s patients (mostly at Stages I and II) showed a significant decline only in the “psychomotor” tests.11 The heterogeneity of the

cohort at entry and the use of composite indices, however, may have blurred the picture. Nevertheless, in the lamotrigine trial,12 which was based on a more homogeneous patient population, significant cognitive effects appeared not as a decline but rather as an improvement in performance for the Wechsler Adult Intelligence Scale (WAIS; full scale), word fluency, and immediate paragraph recall both in the placebo and in the lamotrigine group. The authors concluded that most of the cognitive tasks were “unable adequately to reflect disease progression.” This relative stability in longitudinal studies suggests that methodologic biases masked the effects of disease progression. We have readdressed this issue within the framework of studies preparing for clinical trials that involve cell and gene therapy in patients with HD, exploring higher brain functions by use of a comprehensive neuropsychological battery. We conducted an annual evaluation of a cohort of patients at early stages of HD over a period of up to 4 years to assess the cognitive deterioration and to explain difficulties encountered in longitudinal follow-up in HD.

From the Département de Neurosciences Médicales, CHU Henri Mondor; AP/HP, Créteil; INSERM U. 421 (Drs. Bachoud-Lévi, Baudic, Cesaro, and Peschanski), IM3, Créteil; LSCP (CNRS, EHESS) (Dr. Bachoud-Lévi), Paris; INSERM U. 324 (Drs. Bartolomeo, Dalla Barba, and Ergis), Paris; and Service de Pharmacologie Clinique (Dr. Maison), CHU Henri Mondor, Créteil, France. Supported by INSERM, Program Hospitalier de Recherche Clinique (PHRC No. AMO96175), Association Française contre les Myopathies, and Association France Huntington. Received July 5, 2000. Accepted in final form December 30, 2000. Address correspondence and reprint requests to Dr. A.-C. Bachoud-Lévi, Service de Neurologie, Hôpital Henri Mondor, 94010 Créteil, France; e-mail: [email protected] 1052 Copyright © 2001 by AAN Enterprises, Inc.

Table 1 Baseline characteristics of the cohort of patients with Huntington’s disease Characteristics

Baseline

No. of patients

22

Age, y, mean ⫾ SD Sex, % F/M Laterality, % R/L Education, y, mean ⫾ SD Duration of HD, y, mean ⫾ SD Codon repeats, mean ⫾ SD

42.0 ⫾ 6.4 41/59 95/5 13.7 ⫾ 4.2 4.0 ⫾ 2.8 45.3 ⫾ 4.2

First onset manifestation,* % Cognitive

22.7

Motor

59.0

Psychiatric

45.4

* Because two types of symptoms may have co-occurred at onset in single patients, the percentage of first-onset manifestation exceeded 100%.

Subjects and methods. Twenty-two adult patients (9 women, 13 men) at an early stage of HD gave informed consent and were included in this study. They were recruited from the HD clinic of the Neurologic Department of the Henri Mondor Hospital in Créteil (France). A general description of the cohort is provided in table 1. Twenty patients were in Stage I, and two were in Stage II in the Total Functional Capacity score (TFC13). All patients were followed-up for at least 2 years (three assessments), and some up to 4 years (five assessments). CAG repeats, age at entry, duration of illness, and educational levels were not controlled, because they do not directly correlate with functional staging.13-16 Twelve patients were free from associated disease. In the 10 others, previous medical events were diagnosed before inclusion in this study (S-protein deficit:1; headache: 2; nephritis: 2; idiopathic chronic diarrhea: 2; colic polyp: 1; uterine fibroma: 2; hypertension: 1), whereas one case of hypothyroidism, one of diabetes, and two of dyslipidemias were discovered and treated during the study. During the course of the study, no specific trend was observed in the administration of medications. A single patient was totally free from treatment. Administration of psychotropic medication was variable, including administration for repeated short periods (antidepressants in seven patients; sedative or anxiolytic benzodiazepines in two); administration during the whole study (antidepressants in 10; neuroleptics in 10 [dose increase: 3; dose decrease: 2; stable dose: 5]; mood regulators in 5; benzodiazepines in 11). In one patient, neuroleptics were introduced during the course of the study. Neuropsychological assessment. Neuropsychological tasks were administered during a 4-day hospitalization in split sessions of less than 2 hours each. One session was performed per day, and the order of the tasks and their daily time of presentation were kept constant. Complete neuropsychological testing lasted 6 to 8 hours. A large number of standardized tests were used to ensure comprehensiveness of the assessment. The battery included evaluations of attention, planning, language, memory, and visuospatial abilities. The selection of the tasks was guided by the Core Assessment Protocol for In-

tracerebral Transplantation in Huntington’s Disease (CAPIT-HD17), a battery that was designed to detect potential effects of neurosurgical therapeutic strategies in small cohorts of patients. French standardized tests were used when no standardized adaptations of the recommended English tests were available. Some additional tasks of interest also were included, such as the digit cancellation task. The delay between two assessments was increased to 12 months, rather than the recommended 6 months, to limit the potential bias introduced by retest effects. Because no parallel forms are available for all of the tasks, the same form of each task was used at each session to facilitate interpretation of results. Global cognitive efficiency. General cognitive decline was assessed by using the Mini-Mental State Examination (MMSE18), the Raven’s Coloured Progressive Matrices (RCPM19), and the Mattis Dementia-Rating Scale (MDRS20). The premorbid IQ was computed from the score obtained on the Bonnardel test.21 Attention and executive functions. The battery included the Stroop Test,22 the Trail Making Test form A (TMT A23), as well as the digit and figure cancellation tasks24 and the symbol digit codes task (WAIS-R25). In the digit and figure cancellation tasks, patients are requested to cancel one, two, and then three digits or figures in a panel of strings. Planning difficulties were evaluated by the Trail Making Test form B (TMT B26) and the Modified Wisconsin Card Sorting Test (WCST27). Also included were the categorical verbal fluency (for animals) and the literal fluency (for letters P and M) tasks. Verbal fluency was computed for 2 minutes, with an intermediate measurement at 1 minute.2,28 Language. Various tests were used to disentangle the different components of language impairment. The Token Test (short form29) evaluated oral comprehension. The picture naming test30 and the Articulatory Rate evaluation31 assessed speech production. For the picture naming test, a set of 20 pictures was selected from a French set of 80 standardized pictures.32 The assessment of articulatory rate involved 10 repetitions of three disyllabic, three trisyllabic, and three quadrisyllabic words. Memory. Short-term memory was assessed by using the Corsi block span,33 the visual span,34 and the forward digit span test.35 Long-term memory was examined by using backward digit span,6 a French adaptation of the Free and Cued Selective Reminding Test (FCSRT),36 and the Rey Auditory Verbal Learning Task37 (RAVLT). Visuospatial abilities. Visuospatial abilities were tested by using the Mental Rotation Test38 and the Judgment of Line Orientation Task39 (JLOT). Motor, psychiatric, and functional assessments. To place neuropsychological results in the perspective of patients’ general state of health, a series of tests aimed at defining diverse functional, psychiatric, and motor parameters were performed in parallel to the neuropsychological assessment. Daily living ability was evaluated by using the Total Functional Capacity scale (TFC).13 Severity of depression, which may affect performance, was quantified systematically by using the Montgomery and Åsberg Depression Rating Scale (MADRS).40 Motor responses were examined by reaction time tasks (RT).41 In these tasks, performed with Brain Scan software 1.3.1. (Centre de revalidation April (2 of 2) 2001

NEUROLOGY 56

1053

neuropsychologiques Universitaire, St. Luc, Bruxelles), patients were asked to click the mouse as quickly as possible when a square appeared on a computer screen. The square was presented for 50 msec, at intervals varying between 0.1 and 5 seconds. The UHDRS,42 including the Independence Scale and the aptitude, motor, and psychiatric sections, was not available to us at the beginning of this study but was subsequently added. Statistical analysis. The results are expressed as arithmetic means and standard deviations. To assess annual progression, an individual slope was estimated for each task by using the coefficient of linear regression model, in which the independent variable was time and the dependent variable was individual scores. The mean slopes were compared to 0, which is the predicted value indicating the absence of evolution, using Student’s t-test. The retest effect, defined as an improvement in performance over time, may appear as 1) a significant improvement during the entire follow-up period, or 2) an initial improvement followed by a decline (thus, a significant decline would not be observed overall between the first and third assessment). To further explore the latter possibility, we compared the changes observed between the first and second assessments with those observed between the second and third assessments, using a paired Student’s t-test. When a retest effect was observed, the annual evolution was calculated again, excluding the first assessment. All analyses were conducted by using the SPSS 9.0 (SPSS, Chicago, IL) package for Windows. The condition of a few patients deteriorated over time such that they were unable to perform some of the tasks. In the statistical analysis, the corresponding “missing” results were replaced by the lowest possible score for each specific task (e.g., 0 points for the JLOT). When the nature of the task precluded the determination of a “lowest” score (e.g., reaction times), the data were treated as missing values. A correlation analysis using Pearson correlation coefficients was performed between the tasks that showed a significant decrease over time and baseline variables indicating disease profile at entry: the number of CAG repeats, functional ability (TFC and independence scale), behavioral aspects (MADRS), dementia level (MDRS), and motor dysfunction (RT). To detect the influence of motor impairments in tests requiring paper and pencil responses, additional correlations were performed between UHDRS motor parts and declining tasks. Results. The characteristics of the cohort of 22 patients are presented in table 1. All patients completed at least three assessments, one completed four, and seven completed five (mean follow-up, 31.2 ⫾ 10.0 months). Motor and functional impairments were present, though variable, at entry and deteriorated with time. Across the entire group, there was a significant decline in the TFC score, the independence scale, the UHDRS motor score, and reaction times. In contrast, psychiatric performance assessed by the MADRS and the UHDRS psychiatric section was normal at entry— or showed only slight impairments—and was remarkably stable over time (table 2). Average neuropsychological performance at entry. Performance on many tasks was impaired at entry (table 3), compared with normal published levels or, if unavailable, with seven control subjects— unpublished data). Among global tasks, the MDRS score was below normal values, 1054 NEUROLOGY 56

April (2 of 2) 2001

Table 2 Means (⫾SD) for the 22 patients in functional, psychiatric, and motor assessments during 3.6-year follow-up Assessments

Score

Annual slope

p Value

Functional assessment TFC score

11.0 ⫾ 2.0

⫺0.9 ⫾ 0.9

**

Independence scale*

93.3 ⫾ 12.0

⫺6.1 ⫾ 7.0

***

UHDRS aptitude†

25.8 ⫾ 11.4

1.4 ⫾ 7.5

Psychiatric assessment 6.8 ⫾ 6.7

3.0 ⫾ 8.3

11.6 ⫾ 7.7

⫺7.2 ⫾ 28.8

17.3 ⫾ 11.9

13.2 ⫾ 14.1

** *

UHDRS psychiatric† MADRS Motor assessment UHDRS motor† RT, ms

448 ⫾ 191

2.1 ⫾ 182

Accuracy RT

45.1 ⫾ 4.3

⫺2.0 ⫾ 10.6

* p ⬍ 0.05. ** p ⬍ 0.01. *** p ⬍ 0.001. † Number of patients ⫽ 10. TFC ⫽ Total Functional Capacity; UHDRS ⫽ Unified Huntington’s Disease Rating Scale; MADRS ⫽ Montgomery and Asberg Depression Rating Scale; RT ⫽ reaction time.

whereas the PM47 and the MMSE were within the normal range. Attention and executive tasks all were impaired, with a marked slowness in timed tasks. The interference score of the Stroop task was within normal limits, whereas scores for word reading, color naming, and colored-word naming were below normal. One-, two-, and three-digit and figure cancellation performance were worse than those of normal control subjects. Completion of the TMT A and the TMT B was slow. Verbal fluency was poor, with the exception of articulatory rate (controls: 5.6 ⫾ 0.5); language skills were unimpaired, as assessed by picture naming (controls: 19.4 ⫾ 0.8) and by the Token Test. Immediate and short-term memory tasks were within the lowest limits of normal performance. Patients showed impairments in long-term memory for both the total free recall of the FCSRT and the RAVLT. Visuospatial performance was altered at entry on the Mental Rotation task but normal on the JLOT. Decline of average values over time. Performance on attention and executive functions showed significant decline (the three-digit cancellation task, the TMT A–time measurement, verbal fluency, and the Stroop test, except the interference score) or showed a trend toward decline (the TMT A–points; p ⫽ 0.09). In contrast, there was a significant improvement on performance in the WCST. Performance on global cognitive efficiency, language, visuospatial abilities, and memory were stable during the full follow-up period. The Token test and visual span showed a significant decline over time; MDRS decline failed to reach significance (p ⫽ 0.07). Correlation analyses. No correlation was observed between the performance on the tasks that showed a significant decline and the variables that defined the profile of the patient cohort at entry (CAG repeats, TFC, MADRS, MDRS, and RT). Moreover, no correlation between UHDRS motor score and slopes of declining cognitive tasks were obtained.

Table 3 Mean scores at entry and annual slopes of evolution (⫾SD) over the entire follow-up period for the 22 patients

Tests

Normal range limits

Score at entry

Annual slope

27.9 ⫾ 2.8

⫺0.1 ⫾ 1.4

p Value

Global cognitive efficiency MMSE

ⱖ24

PM47

ⱖ26

30.5 ⫾ 5.4

⫺0.8 ⫾ 5.9

MDRS

ⱖ136

131.2 ⫾ 11.4

⫺0.7 ⫾ 9.9

Bonnardel†

ⱖ20

29.5 ⫾ 5.2

0.3 ⫾ 2.2

Attention and executive functions Stroop Word Colour Colour/word 1-Digit cancellation

ⱖ88

73.2 ⫾ 17.5

⫺5.0 ⫾ 9.6

*

ⱖ65

49.3 ⫾ 15.5

⫺3.3 ⫾ 5.8

*

ⱖ35

28.2 ⫾ 10.4

⫺0.3 ⫾ 17.7

ⱖ10

9.1 ⫾ 1.8

⫺0.1 ⫾ 1.1

2-Digit cancellation

ⱖ19

16.4 ⫾ 4.7

⫺0.4 ⫾ 2.7

3-Digit cancellation

ⱖ28

19.5 ⫾ 6.8

⫺2.4 ⫾ 5.0

1-Figure cancellation

ⱖ64

34.5 ⫾ 15.5

0.1 ⫾ 12.5

2-Figure cancellation

ⱖ64

37.5 ⫾ 15.5

⫺1.0 ⫾ 12.0

3-Figure cancellation

ⱖ52

25.7 ⫾ 13.1

⫺2.4 ⫾ 6.0

TMT A (time)

ⱕ54

72.4 ⫾ 42.3

12.9 ⫾ 25.0

TMT B (time)

ⱕ135

171.2 ⫾ 70.7

5.3 ⫾ 42.2

Categorical fluency

⬎21.9

22.1 ⫾ 10.0

*

⫺2.7 ⫾ 3.9

** *

ⱖ12.1

13.6 ⫾ 8.6

⫺1.2 ⫾ 2.0

Digit symbol (WAIS)†

ⱖ37

46.6 ⫾ 17.6

⫺0.8 ⫾ 18.0

WSCT criterions†

ⱖ1.82

2.9 ⫾ 0.3

0.6 ⫾ 1.0

Letter fluency (means P, M)

*

*

Language Articulation rate

ⱖ4.8

6.8 ⫾ 2.0

3.7 ⫾ 12.2

Picture naming

ⱖ18.1

19.0 ⫾ 2.0

0.1 ⫾ 1.1

Token test

ⱖ33

35.2 ⫾ 1.0

⫺0.8 ⫾ 1.1

Forward digit span

ⱖ4.3

5.4 ⫾ 1.3

0.1 ⫾ 1.5

Backward digit span

ⱖ2.74

3.6 ⫾ 1.3

0.3 ⫾ 1.2

**

Memory

Corsi blocks

ⱖ4.8

4.9 ⫾ 1.1

⫺0.3 ⫾ 1.2

Visual span

ⱖ8

6.8 ⫾ 1.6

⫺0.6 ⫾ 1.1

ⱖ26

25.4 ⫾ 7.9

1.3 ⫾ 2.7

ⱖ37

43.5 ⫾ 6.8

2.9 ⫾ 7.7

% Intrusions

2.9 ⫾ 3.7

⫺0.2 ⫾ 1.7

% Perseverations

2.8 ⫾ 3.1

2.4 ⫾ 5.6

Grober & Buschke FTR TRT

RAVLT Total recall

⬎40

Learning

35.8 ⫾ 13.6

1.2 ⫾ 3.6

47.8 ⫾ 18.1

1.6 ⫾ 4.8

% Intrusions

0.9 ⫾ 1.1

0.5 ⫾ 1.2

% Perseverations

3.7 ⫾ 3.2

0.1 ⫾ 2.2

18.0 ⫾ 9.1

1.6 ⫾ 4.8

Logical stories†

ⱖ14.2

*

Visuospatial abilities Mental rotation

ⱖ16

12.0 ⫾ 3.4

0.1 ⫾ 2.9

JLOT

ⱖ21

23.6 ⫾ 4.3

⫺0.4 ⫾ 6.9

Normal range limits are provided as an indication for interpretation of the data. In some cases, they were extracted from published norms; in others, they were calculated to obtain a range of difference of 1.65 standard deviations to the mean or extrapolated from normalized published curves of performance. For the sake of simplicity, values in men and women or for various educational levels and various age ranges have been blended. * p ⬍ 0.05. ** p ⬍ 0.01. † Number of patients ⫽ 10. MMSE ⫽ Mini-Mental State Examination; PM47 ⫽ Raven’s Coloured Progressive Matrix; MDRS ⫽ Mattis Dementia Rating Scale; TMT ⫽ Trail Making Test; FTR ⫽ Free total recall; TRT ⫽ Total recall total; RAVLT ⫽ Rey Auditory Verbal Learning Task; JLOT ⫽ Judgment of Line Orientation Task. April (2 of 2) 2001

NEUROLOGY 56

1055

Retest effect. Evolution of performance between the first and second assessment (⌬1) differed from those observed between the second and third assessment (⌬2) for TMT B (points): 1.4 ⫾ 3.9 versus ⫺2.3 ⫾ 4.5, p ⫽ 0.02; and categorical fluency: ⫺1.5 ⫾ 6.4 versus ⫺6.6 ⫾ 5.5; p ⫽ 0.02). Memory tasks were very sensitive to retest effect: total free recall of the FCSRT (4.4 ⫾ 3.9 versus ⫺3.6 ⫾ 5.2, p ⬍ 0.001) and backward digit span (0.2 ⫾ 0.7 versus ⫺0.4 ⫾ 0.8, p ⫽ 0.04). Retest effects were not observed for either language or visuospatial performance. Because the learning of the tasks was likely to be more significant between the first and the second assessment than between the second and the third, we excluded the results of the first assessment and reanalyzed the annual evolution for the tasks in which performance suggested a retest effect (for patients who were tested more than three times, n ⫽ 8). In these restricted analyses, a decline was observed for the TMT B (points) (mean slope: ⫺2.7 ⫾ 3.0, p ⫽ 0.05) and for the Color/Word condition of the Stroop test (mean slope: ⫺2.8 ⫾ 3.1, p ⫽ 0.04).

Discussion. This study was designed to evaluate the progression of cognitive functions in patients in the early stages of HD. It followed a relatively restricted cohort of 22 patients over a long period and showed a significant decline in certain cognitive functions, in particular in attention and executive functions. Although cognitive decline in HD patients is certainly not an original finding, it has been difficult to obtain conclusive longitudinal quantitative data up to now. The results of the present study point to difficulties encountered in the evaluation of cognitive performance in HD patients and suggest ways in which some of these may be overcome. Intrinsic features of HD make cognitive decline difficult to assess. Among the cognitive functions analyzed, only attention and executive functions showed statistically significant decline over time. They further proved efficient in revealing the disease progression as indicated by the UHDRS motor, the independence scale,42 and the TFC13 scores. This pattern of evolution and the key assessment role of the Stroop task and verbal fluency are consistent with results obtained in previous neuropsychological longitudinal studies in HD patients,9,11,12 However, these findings differ from what can be inferred from cross-sectional studies, which found, besides attention and executive functions, an early significant decline in memory, and a later one for language and visuospatial skills2,3,5,30,43,44 (see Brandt8 for a review). Intrinsic features of the disease are certainly responsible for much of the difficulty encountered in the evaluation of cognitive functions. Evolution of HD extends for many years (approximately 15–20) after clinical onset.45 A long intertest interval therefore may be necessary to record deterioration on many tasks. In addition, comparison of standard deviations between values at entry and slopes of evolution suggests that interindividual variability over time may be one of the hallmarks of HD. Whereas the cohort of patients was relatively homogeneous at entry, as indicated by the TFC score, individual 1056 NEUROLOGY 56

April (2 of 2) 2001

slopes of evolution were, in contrast, extremely different, leading to very large standard deviations in the group analysis, as already reported.42 Intraindividual variability is also an issue with HD patients: mood or behavioral status may influence motivation to perform tasks, and stress, fatigue, or unknown personal events may transiently increase movement disorders and articulatory dysfunction.46 This, in turn, may alter performance on cognitive tasks that require paper and pencil or verbal answers. To minimize the role of these confounding factors in the elaboration of future follow-up protocols in HD, homogeneity of the cohort at entry should be targeted to reduce interindividual variability. Because of their relative predictability compared with cognitive patterns, the TFC13 and the Independence Scale42 scores may be used as criteria to define the homogeneity of the cohort. Within-individual variability can be reduced by conducting the multiple assessments under similar environmental conditions. This includes keeping the same examiner for all the assessments, conducting the tests as much as possible at the same time of day, in the same context (outpatient or inpatient), and at the same time of year.47,48 Biases in the longitudinal testing procedure. In addition to the confounding features of the disease itself, the tests and the procedure of testing may introduce biases. The high sensitivity to decline of tests devoted to attention-set shifting and executive functions in the current study is not a trivial issue: at entry, patients showed impairments not only in those functions but also in the other cognitive functions. However, that the other functions did not show decline over time is puzzling. If one accepts the generally held clinical view that patients do deteriorate over time in all functions, though not in a homogenous fashion, then some tests did not correctly quantify an existing decline. Interestingly, tests exploring attention and executive functions are timed, whereas tests assessing most of the other functions are not. As a result, the former may be more sensitive to slowness of execution in relation to bradyarthria and bradykinesia or to a general slowness of thinking (so-called bradyphrenia) associated with subcortical dementia,8,49-51 although the design of the current study does not allow the separation of these factors. The wider use of timed tests for assessing functions other than attention and executive functions may sensitize currently uninformative tests and in parallel permit a better dissociation between specific and general processes in patient’s performance. Furthermore, the retest effect, defined as an improvement in performance after repeated presentation of a test, is a general problem for longitudinal neuropsychological studies.52-56 Effects of implicit, or explicit learning, as well as of anxiety reduction are always an issue when the same task is performed twice. Recently, Small et al.57 showed in healthy elderly individuals that the retest effect was not main-

tained after the presentation of a second task. In our study, the retest effect predominated between the first and the second test, presumably because at this time the evolution of the disease was not severe enough to counterbalance learning and familiarization effects. An improvement in performance over the entire follow-up period, however, appeared for the WCST, suggesting that this task is inappropriate for longitudinal follow-up in HD. To reduce the influence of the retest effect, we propose to discard the results of the first assessment from the analysis of the results. The two first assessments should be performed at short delay, and the second should be used as baseline. We also recommend use of parallel forms for those tasks that involve explicit learning, such as the FCSRT, in which retest effects may be maintained despite the exclusion of the data of the first assessment. More directly identifiable, and therefore less problematic for the evaluation of results, were floor and ceiling effects. Certain tasks appeared to be very easy for the patients (e.g., one-figure and one-digit cancellations and language production tasks), as exemplified by average group values very close to maximal (normal) values (“ceiling” effect). In contrast, on other tests, such as three-figure and three-digit cancellation, or the TMT B, subjects performed extremely poorly at entry (“floor effect”). As a result of this low baseline, performance on such tasks did not deteriorate as for other tasks. Because the ease with which patients perform the tasks depends on the stage of the disease, it is important to adapt the choice of the tasks to the presumed abilities of the cohort during the full follow-up period. Finally, despite retest effects and interindividual and intraindividual variability, the standardization of testing procedure and the use of a long follow-up provided an opportunity to observe a significant decline in cognitive function in this restricted cohort of early HD patients. Acknowledgment The authors thank K. White, C. Manning, and A. Doble for review of the manuscript, the centers for genetic testing for characterization of patients, and V. Ribeil and D. Delbos for skillful secretarial help. Discussions in meetings of various networks and groups, including the Créteil/Orsay Huntington Research group for Experimental therapy and Assessment (COHREA), NECTAR, NESTHD, CAPIT-HD, and HSG, have been particularly helpful during the completion of this study.

References 1. Sprenglemeyer R, Lange H, Hömberg V. The pattern of attentional deficits in Huntington’s disease. Brain 1995;118:145– 152. 2. Butters N, Wolfe J, Granholm E, et al. An assessment of verbal recall, recognition and fluency abilities in patients with Huntington’s disease. Cortex 1986;22:11–32. 3. Huber SJ, Paulsen GW. Memory impairment associated with the progression of Huntington’s disease. Cortex 1987;23:275– 283. 4. Andrew DL, Sahakian BJ, Hodges J, et al. Executive and mnemonic functions in early Huntington’s disease. Brain 1996;119:1633–1645.

5. Mohr E, Browers P, Claus JJ, et al. Visuospatial cognition in Huntington’s disease. Mov Disord 1991;6:127–132. 6. Fisher JM, Kennedy JL, Caine ED, et al. Dementia in Huntington’s disease: a cross-sectional analysis of intellectual decline. In: Mayeux R, Rosen WG, eds. The dementias. New York: Raven Press, 1983. 7. Volkman J, Heffter H, Lange HW, et al. Impairment of temporal organization of speech in basal ganglia diseases. Brain Lang 1992;43:386 –399. 8. Brandt J. Cognitive impairments in Huntington’s disease: insights into the neuropsychology of striatum. In: Boller F, Grafman J, eds. Handbook of neuropsychology. Amsterdam: Elsevier, 1991:241–264. 9. Hodges JR, Salmon DP, Butters N. Differential impairment of semantic and episodic memory in Alzheimer’s and Huntington’s diseases: a controlled prospective study. J Neurol Neurosurg Psychiatry 1990;53:1089 –1095. 10. Buschke H. Selective reminding for analysis of memory and learning. J Verb Learn Verb Behav 1973;12:543–550. 11. Bamford KA, Caine ED, Kido DK, et al. A prospective evaluation of cognitive decline in early Huntington’s disease: functional and radiographic correlates. Neurology 1995;45:1867– 1873. 12. Kremer B, Clark CM, Almqvist EW, et al. Influence of lamotrigine on progression of early Huntington disease: a randomized clinical trial. Neurology 1999;53:1000 –1011. 13. Shoulson I. Huntington disease: functional capacities in patients treated with neuroleptic and antidepressant drugs. Neurology 1981;31:1333–1335. 14. Shoulson I, Fahn S. Huntington disease: clinical care and evaluation. Neurology 1979;29:1–3. 15. Britton JW, Uitti RJ, Robinson RG, et al. Hereditary lateonset chorea without significant dementia: genetic evidence for substantial phenotypic variation in Huntington’s disease. Neurology 1995;47:443– 447. 16. Roos RAC, Hermans J, Vegter-Vander Vlis M, et al. Duration of illness in Huntington’s disease is not related to age at onset. J Neurol Neurosurg Psychiatry 1993;56:98 –100. 17. Quinn Q, Brown R, Craufurd D, et al. Core Assessment Program for Intracerebral Transplantation in Huntington’s Disease (CAPIT-HD). Mov Disord 1996;11:143–150. 18. Folstein MF, Folstein SE, McHugh PR. “Mini Mental State”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189 –198. 19. Raven JC, Court JH, Raven J. Manual for the Raven’s progressive matrices and vocabulary scales. Los Angeles: Western Psychological Services, 1978. 20. Mattis S. Mental status examination for organic mental syndrome in the elderly patient. In: Bellak L, Karasu TB, eds. Geriatric psychiatry. New York: Grune & Stratton, 1976. 21. Beauregard D. Le test des automatismes verbaux. Issy-lesMoulineaux: Editions Scientifiques et Psychotechniques, 1971. 22. Golden CJ. Stroop color and word test. Chicago: Stoelting, 1978. 23. Reitan RM. Validity of the TMT as an indication of organic brain damage. Percept Motor Skills 1958;8:271–276. 24. Spinnler H, Tognoni G. Standardizzazione e tarature italiana di test neuropsicologici. Ital J Neurol Sci 1987;(suppl 8). 25. Wechsler D. Wechsler adult intelligence scale–revised manual. New York: Psychological Corporation, 1981. 26. Benton AL, Hamsher K, Varney NR, Spreen O. Contributions to neuropsychological assessment. Oxford: Oxford University Press, 1983. 27. Nelson HE. A modified card sorting test sensitive to frontal defects. Cortex 1976;12:313–324. 28. Cardebat D, Doyon B, Puel M, et al. Evocation lexicale formelle et sémantique des sujets normaux: performances et dynamiques de production en fonction du sexe, de l’âge et du niveau d’étude. Acta Neurol Belg 1990;90:207–217. 29. De Renzi E, Faglioni P. Normative data and screening power of a shortened version of the Token Test. Cortex 1978;14:41– 49. 30. Podoll K, Caspary P, Lange HW, et al. Language functions in Huntington’s disease. Brain 1988;111:1475–1503. 31. Ludlow CL, Connor NP, Bassich CJ. Speech timing in Parkinson’s and Huntington’s disease. Brain Lang 1987;32:195–214. April (2 of 2) 2001

NEUROLOGY 56

1057

32. Deloche G, Hannequin D, Dordain M, et al. Picture confrontation oral naming: performance differences between aphasics and normals. Brain Lang 1996;53:105–120. 33. Milner C. Interhemispheric differences in the localisation of psychological processes in man. Br Med Bull 1971;27:272–277. 34. Wilson JTL. Visual short-term memory. In: Stachowiack FJ, ed. Development in the assessment and rehabilitation of brain-damaged patients. Krager-Verlag, 1993. 35. Kaplan E, Fein D, Morris. WAIS-R as a neuropsychological instrument. San Antonio, TX: The Psychological Corporation, 1991. 36. Grober E, Buschke H. Genuine memory effects in dementia. Dev Psychol 1987;3:13–36. 37. Rey A. L’examen clinique en psychologie. Paris: Presses Universitaires de France, 1964. 38. Ratcliff G. Spatial thought, mental rotation and the right cerebral hemisphere. Neuropsychologia 1979;17::49 –54. 39. Benton AL, Hanney HJK, Varney NR. Visual perception of line direction in patients with unilateral brain disease. Neurology 1975;25:907–910. 40. Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry 1979;134:382–389. 41. Pasquier F. Neuropsychological features and cognitive assessments in fronto-temporal dementia. In: Pasquier F, Leber F, Scheltens PH, eds. Fronto-temporal dementia. Dordrecht: ICG Publications, 1996:49 – 69. 42. Huntington Study Group. Unified Huntington’s Disease Rating Scale: reliability and consistency. Mov Disord 1996;11: 136 –142. 43. Butters N, Sex D, Montgomery K, et al. Comparison of the neuropsychological deficits associated with early and advanced Huntington’s disease. Arch Neurol 1978;35:585– 589. 44. Moses JA, Golden CJ, Berger PA, et al. Neuropsychological deficits in early, middle, and late stage Huntington’s disease as measured by the Luria-Nebraska neuropsychological battery. Int J Neurosci 1981;14:95–100. 45. Rosenstein LD. Differential diagnosis of the major progressive dementias and depression in middle late and late adulthood; a summary of the literature of the early 1990s. Neuropsychol Rev 1998;8:109 –176.

1058 NEUROLOGY 56

April (2 of 2) 2001

46. Bachoud-Lévi AC, Bourdet C, Brugières P, et al. Safety assessment of intrastriatal neural allografts in 5 patients with Huntington’s disease. Exp Neurol 2000;161:194 –202. 47. Porterfield T, Cook M, Deary IJ, et al. Neuropsychological function and diurnal variation in depression. J Clin Exp Neuropsychol 1997;19:906 –913. 48. Moffoot APR, O’Caroll RE, Bennie J, et al. Diurnal variation of mood and neuropsychological function in major depression with melancholia. J Affect Disord 1994;32:257–269. 49. Albert ML, Feldman RG, Willis AL. The. “subcortical dementia” of progressive supranuclear palsy. J Neurol Neurosurg Psychiatry 1974;37:121–130. 50. Owen AM, James M, Leigh PN, et al. Fronto-striatal cognitive deficits at different stages of Parkinson’s disease. Brain 1992; 115:1727–1751. 51. Lees AJ. The concept of bradyphrenia. Rev Neurol 1994;50: 823– 826. 52. Mitrushina M, Satz P. Effect of a repeated administration of a neuropsychological battery in the elderly. J Clin Psychol 1991; 47:790 – 801. 53. Kendrick DC, Post F. Differences in cognitive status between healthy, psychiatrilly ill, and diffusely brain damaged elderly subjects. Br J Psychiatry 1967;113:75– 81. 54. Katzman R, Brown T, Thal LJ, et al. Comparison of rate of annual change of mental status score in four independent studies of patients in Alzheimer’s disease. Ann Neurol 1988; 24:384 –389. 55. Galasko D, Abramson I, Corey-Bloom J, et al. Repeated exposure to the Mini-Mental State Examination and the Information-Memory-Concentration Test results in a practice effect in Alzheimer’s disease. Neurology 1993;43:1559 – 1563. 56. Campodonico JR, Codori AM, Brandt J. Neuropsychological stability over two years in asymptomatic carriers of the Huntington’s disease mutation. J Neurol Neurosurg Psychiatry 1996;61:621– 624. 57. Small SA, Stern Y, Tang M, et al. Selective decline in memory function among healthy elderly. Neurology 1999;52:1392– 1398.