Abstract
Objective: To define the long-term outcome in Gilles de la Tourette syndrome (GTS) using objective rating measures.
Background: Previous historical studies suggest spontaneous improvement of tic symptoms after adolescence, but objective longitudinal data are limited.
Methods: The authors reviewed all videotapes in their database (1978 through 1991) of children with GTS (ages 8 to 14) who were seen in their tertiary care movement disorder center and underwent a standardized 5-minute filming protocol (n = 56). Through multiple contact methods, they successfully located 36 of these patients, who are now adults (age >20 years), and recruited 31 (28 men and 3 women) to volunteer for a second videotape and in-person assessment. A blinded rater evaluated the 62 tapes and rated five tic domains: body areas involved, motor and phonic tic frequency, and motor and phonic tic severity. Using standardized GTS videotape rating scale and Wilcoxon signed-rank tests with Bonferroni correction for multiple comparisons, the authors compared the two videotapes for each tic domain as well as the composite tic disability score.
Results: Ninety percent of adult patients still had tics. Adult patients who considered themselves tic-free were often inaccurate in their self-assessment: 50% had objective evidence of tics. Mean objective tic disability diminished in comparison to childhood (mean composite tic disability score childhood 9.58 vs adulthood 7.52, p = 0.014). All domains improved by adulthood, and significant improvements occurred in motor tic severity (p = 0.008). The improvements in tic disability did not relate to medication use, as only 13% of adults received medications for tics, compared with 81% of children.
Conclusions: In GTS syndrome, tics objectively improve over time but most adults have persistent tics.
Previous studies1-10⇓⇓⇓⇓⇓⇓⇓⇓⇓ suggest that Gilles de la Tourette syndrome (GTS) improves in late adolescence or early adulthood, but objective longitudinal data supporting this conclusion are limited.11 Because the major treatment of GTS involves neuroleptic drugs that can have irreversible side effects, a knowledge of the natural history of this condition may influence both short-term and long-term therapeutic decisions. Several GTS questionnaire and chart review studies on adult outcome suggest improvement with age, but these investigations lack standardized examinations and have varied in patient inclusion criteria, follow-up periods, follow-up ages, and neuroleptic medication use.4,5⇓ The development of clinical assessment measures for GTS over the past decade has increasingly focused on the use of objective data to complement patient reports. Objective data include direct observation of patients and evaluation of videotape-recorded tics. The latter methodology is advantageous as it allows for data assessment by observers without direct patient involvement. It further minimizes the confounding effect of conscious or unconscious tic suppression that is seen during direct patient examinations.12 The current study used parallel, objective videotape-based assessments of motor and vocal tics in a prospective, longitudinal design in a tertiary care GTS population. Using identical videotape assessments from childhood and adulthood in the same patients, we hypothesized that adults would show objective improvement in tics compared to childhood function and that these objective improvements would correlate with perceptions of tic improvement by patient interview.
Methods.
Patient recruitment.
Starting in 1978, all patients with an initial neurologic consultation for childhood tics were videotaped with the Rush video GTS protocol12 at Rush-Presbyterian-St. Luke’s Movement Disorders Center, a tertiary care service. In this protocol, simultaneous audio and visual recordings were made of the full frontal body views and the head and shoulders in close-up as the patient sat in a chair for 5 to 12 minutes. In 1996, we identified all children in this pool (age 8 to 14 years at time of initial videotape) who met Diagnostic and Statistical Manual of Mental Disorders–III–R (DSM-III-R) criteria for GTS.13 Telephone, mail, and investigative services were used to locate these patients, who were now adults (age >20). We attempted to recruit them for a second standardized videotape filming and an in-person interview and examination. Patients were eligible for participation regardless of their use of neuroleptic medications or other drugs prescribed for tic control or behavioral disorder at the time of either videotape session. Furthermore, patients were included irrespective of their use of medications such as methylphenidate or amphetamine products.
Patient evaluation.
After obtaining informed consent, recruited patients were examined in the Department of Neurological Sciences at Rush-Presbyterian-St. Luke’s Medical Center, and the Rush videotape protocol, which followed the same format as the childhood videotape, was completed. During the same visit, the patients completed the Yale Global Tic Severity Scale (YGTSS), which included a subjective (historical subscale) patient/parent perception of current tic disability and an examiner’s assessment.14 Additionally, the patient provided a retrospective YGTSS historical subscale score of tic disability for the time period when the childhood videotape was completed. Current medications and doses were documented by patient interview, and the medications at the time of the childhood videotape were obtained from medical records.
Comparison of childhood with adulthood videotapes.
From each patient’s original childhood and adulthood tapes, a technician blind to the project hypothesis compiled a randomly selected 5-minute videotape segment for rating purposes. The choice of a 5-minute segment was based on the methods of Chappell et al.,15 who demonstrated that videotape data over this period provided highly reliable and stable measurements of tic frequency assessed over 30 minutes of objective observation and 1 week of global function. The 5-minute childhood and adulthood edited video segments were randomly ordered on the final rating tape, using computer-generated randomization tables.
Each videotape segment was rated with the Rush Videotape Rating Scale12 by an experienced movement disorder neurologist (E.D.L.) at a separate institution who was blind to the hypothesis of the project and who had no personal interaction with the patients. In spite of the hours required to review all videotapes for data consistency, we elected to choose one rater to review all the data, rather than select a panel with each member receiving only part of the sample. The scale assessed five tic domains: anatomic distribution as determined by number of involved body areas, frequency of motor and phonic tics, and severity of motor and phonic tics. For motor tic distribution, the number of affected areas (eyes, nose, mouth, neck, shoulders, arms, hands, trunk, pelvis, legs, and feet) was counted. The frequency score consisted of the total number of discrete motor or phonic tics during each 5-minute video segment. Tic severity was assessed by rating the most severe motor and phonic tic.12 This scale has demonstrated inter-rater reliability, temporal stability, and validation with other GTS rating scales.12 To derive an objective global tic disability score and to determine the relative contribution of each of the five domains of tic disability, we used the modified Rush Videotape Rating Scale for data analysis.16 In this modification, the data for each of the five domains (body distribution, frequency, and severity) were transformed to a 0 to 4 rating with a resultant composite global disability score of 0 to 20.
Data analysis.
Because patients with tics can show a wide range of tic counts, we summarized data both as medians and ranges as well as means and standard deviations and used Wilcoxon signed-rank tests to compare the videotapes for each domain and the composite score. Spearman correlation coefficients were computed to measure the correlation between objectively derived videotape-based scores and subjective changes in the YGTSS historical ratings and the duration between videotape assessments. McNemar test was used to compare the percentages of patients having social or educational dysfunction as children with percentages of individuals having adult social and vocational difficulties. Statistical significance was determined by whether p < 0.05, except that Bonferroni correction for multiple comparisons was used to determine which tic disability domain changed significantly between childhood and adulthood. In these analyses, a nominal alpha = 0.05/5 = 0.01 was used for comparisons.
Results.
Patients.
Fifty-six GTS childhood videotapes meeting entry criteria were identified. Of the possible 56 patients, 36 adults were located and 31 (28 men and 3 women) were successfully recruited for study participation. The mean (SD) age at disease onset, defined as the age the parent or patient first observed tics, was 7.2 (±2.4) years, and the average disease duration at the time of the adulthood videotape was 16.2 (±4.0) years. The average age at the time of the childhood videotape was 12.2 (±2.2) years, and the adulthood videotape was 24.2 (±3.5) years. There were no differences in age, sex, or severity of childhood tics between the unlocated and unrecruited patients and the study patients (all p > 0.05).
Objective Rush GTS Scale rating.
During childhood, all patients included in the study fulfilled DSM-III-R diagnostic criteria for GTS, recognized the presence of tics, and demonstrated multiple motor tics on videotape. On the childhood examination, the median body area score from the modified Rush Videotape Rating Scale16 was 2 (range 0, 4). The three most frequently involved areas were neck (80%), mouth (77%), and eyes (71%), whereas the least frequently involved were the feet (9%) and pelvis (6%). The median motor tic frequency score was 2 (range 0, 4), and the median tic severity score was 4 (range 0, 4). Twenty-one of 31 children (68%) had no phonic tics on the video segment, and therefore, the median frequency and severity scores were 0 (range 0, 4 for both). The mean global tic disability score derived from these five measures was 9.58 (median 10, range 3, 18, SD 3.5) (table).
Table Raw data and modified Rush Videotape Rating Scale scores
At the adulthood examination, 90% of patients still had tics. The median body area score was 2 (range 0, 4), and the three most frequently involved areas were the mouth (84%), eyes (62%), and neck (36%), whereas the least frequently involved were the pelvis (6%) and feet (3%). The median motor tic frequency score was 1 (range 0, 4), with a median motor severity score of 3 (range 0, 4). As in the childhood tapes, most adult subjects had no phonic tics during the filming (19/31, 61%) and therefore the median phonic tic frequency and severity scores were 0 (range 0, 4 for both). The mean global tic disability score derived from these five measures was 7.52 (median 8, range 0, 19, SD 2.0). This value was improved compared to the childhood composite score (p = 0.014) (see the table).
In addition to the temporal improvement in global tic impairment score, all five domains of tic impairment improved by adulthood with significant improvement in motor tic severity (p = 0.021). Ten patients, however, had worse objective global tic disability scores as adults. A significant correlation existed between the severity of objective global tic disability at childhood and the severity of objective global tic impairment in adulthood (Spearman correlation = 0.42, p = 0.017). No correlation existed between the change in objective global tic disability and the length of time between videotape assessments.
Subjective patient ratings.
Adults considered themselves tic-free. As a group, however, they were inaccurate in this self-assessment and 50% had objective evidence of tics. At the time of the childhood videotape, the median YGTSS historical subscale rating of tic disability was 14 (range 6, 20). In comparison, there was a significant reduction in tic disability at the time of the adulthood videotape (median 6 [range 0 to 20], p < 0.0001). The difference between childhood and adulthood videotape tic disability scores did not correlate, however, with the difference in childhood and adulthood YGTSS tic disability ratings (Spearman correlation = 0.289, p = 0.1164). A significant correlation did exist between the adult subjective YGTSS tic disability and the objective tic assessment (Spearman correlation = 0.43, p = 0.017). In contrast, there was no correlation between the retrospective childhood subjective YGTSS tic disability and the corresponding objective tic assessment (Spearman correlation = 0.29, p = 0.11).
Effect of medication use.
The improvements did not relate to neuroleptic use, as only 13% of adults received medications for tics, compared with 81% of children. No significant difference existed between childhood and adulthood mean chlorpromazine equivalent neuroleptic doses. At the time of the childhood videotape, 61% received clonidine and 16% received stimulant medications (methylphenidate, pemoline, amphetamine derivatives). At the adulthood assessment, none received clonidine or stimulants.
Social and educational assessments.
Fifty-two percent of children experienced social or educational dysfunction. Thirty-nine percent required special education classroom placement, 10% were held back at least 1 year in school, and 29% experienced disciplinary problems in school or community. In contrast, 32% of adults had significant social or educational dysfunction. All had completed high school and 52% had finished at least 2 years of college. At the time of the interview, however, only 71% were either currently employed or pursuing higher education full-time. This adult subgroup’s composite tic severity ratings from the videotape assessments did not cluster above or below the median (60% below the median, 40% above the median). Of the 16 patients who had social or educational dysfunction as children, 50% had dysfunction as adults (χ2 = 8.424, p = 0.004). Thirteen percent of the patients who did not have social or education dysfunction as children developed these problems as adults.
Discussion.
Our study documented persistence of tics into adulthood and is the first study to use objectively derived data in a large patient sample to assess longitudinal tic outcome in GTS. Because we successfully recruited 86% of all located patients and no significant difference in childhood tic severity existed between the recruited and nonrecruited patients, we are confident that our sample is not heavily biased toward less disabled or more motivated individuals. We found a moderate disparity between adult patients’ perceptions of the presence of tics and the objective tic documentation. Whereas all adults who reported tics showed objective evidence of tics on videotape, half of the patients who considered themselves tic-free where incorrect and showed multiple tics on videotape. Because all prior longitudinal estimates involved subjectively derived data dependent on patient impressions, we consider the accuracy of those reports in regards to tic resolution as suspect.2-4⇓⇓ We recognize that quiescent tics can be unmasked by stress and it is possible the study visit was sufficiently stressful to precipitate tics that did not ordinarily occur in daily life among these subjects. Because the stress, however, was not overtly apparent and the subjects were familiar with the office setting, video protocol, and staff, we do not consider the study participation to be an event that provoked stress that was outside the likely realm of adult life.
In spite of the persistence of tics, our objective data corroborate prior subjective studies reporting improvement of tic disability over time.3,4⇓ Furthermore, our objective data delineate tic motor severity as the primary area of tic improvement. In a questionnaire study, 73% of respondents felt that their tics had diminished considerably by late adolescence.4 The analysis considered overall tic dysfunction and did not individually investigate the temporal changes in motor and phonic tic frequency and severity. Lucas followed 20 patients for 2 to 15 years into adulthood and found 14 had experienced improvement in their tic disorder.6 Similarly, in Zausmer’s study, 75% of patients followed longitudinally rated themselves as improved, although the specific domains of improvement were not investigated.8
Objectively derived videotape data offer several distinct advantages to the study of GTS. First, because tics are fleeting and occur simultaneously in different body regions, they are difficult to assess completely with a single observation. Videotapes capture the tic repertoire and can be replayed to permit the meticulous assessment of complex clinical material. Second, because tics are often suppressed in the presence of the observer, the videotape permits the recording of data in a relaxed setting without direct clinical scrutiny. In our prior study, the number of tics that occurred in the presence of an observer was only 27% of the number seen on videotape, when no observer was present during filming.12 Third, videotape protocols have been validated, incorporated into standardized scales like the Unified Tic Rating Scale, and are currently used in multicenter clinical trials of GTS.17
A concern regarding the use of videotapes in tic assessment is the short time frame of filming and the questionable correlation of videotape findings with overall clinical function. In recognition of this concern and based on the study by Chappell et al.,15 we used 5-minute videotapes in this study. This group examined the temporal stability of tic counts taken from different segments of a 30-minute videotape to estimate the minimum length of time needed for a reliable index of overall tic activity during the full videotape. The authors found that a 5-minute videotape segment provided a valid index of the frequency of both motor and vocal tics present during the entire videotape. These 5-minute video segments were also highly correlated with overall tic severity and frequency as assessed by the YGTSS and the Tourette’s Syndrome Global Scale.18 Based on these several considerations, we view videotape data as a reliable and valid means of assessing tic dysfunction with the advantages of objectivity and permanent documentation.
Whereas this study monitored tics at rest under structured filming conditions in the doctor’s office, a greater repertoire can be seen when the same methods are applied at home.19 Comparative videotapes during childhood and adulthood could likewise be used to assess the impact of tics on various activities like speaking, writing, conversing, or reading. Our childhood protocol did not include any additional activities besides sitting at rest, and therefore we cannot address these issues in our sample.
In our patients all tic domains improved over time. The statistically significant improvement in the composite tic disability score was driven primarily by improvement in motor tic severity. Because phonic tics during childhood were infrequent and mild in our sample, the ability to detect improvements in these domains was hampered by a “floor effect.” The advantage of a composite tic severity score based on equally weighted domains permits an overall assessment of tic severity in addition to a clear delineation of the specific impact of change in anatomic distribution, motor tic frequency and severity, and phonic tic frequency and severity.
As adults, most patients were either employed or attending school full-time. In spite of substantial childhood educational or social problems, most adults were well integrated in their community and economically independent. A subgroup of over one-quarter of the sample, however, was disabled as adults with problems that included alcohol abuse, unemployment, or criminal activity. The fact that this subgroup had composite tic severity scores that were distributed across the spectrum of tic ratings suggests that tic severity was not the primary factor related to adult social dysfunction. However, early life dysfunction closely correlated with adult dysfunction, suggesting that early and assertive intervention in childhood may impact on the prevention of dysfunction in adulthood.
Future studies will focus on longer follow-up and longitudinal tic assessments across life epochs that include middle age and senescence. Because all patients in the series were enthusiastic about the follow-up examinations, they were willing to maintain contact with our group so that future assessments are feasible. We are simultaneously interested in extending our videotape data acquisition methods to maximize available information on tics. As such, with enhanced and more readily available electronic technology, we are investigating the difference in tic frequency and severity as obtained from videotapes recorded in a clinical environment compared with the same standardized methods recorded in the patient’s natural home environment. These methods have the potential to be extended to clinical trials involving new pharmacologic treatments for GTS.
Acknowledgments
Supported by a grant from the Tourette’s Syndrome Association. Federal Grant NIH NS01863 and the Paul Beeson Physician Faculty Scholars supported E.D.L. in aging research award.
Footnotes
-
Presented in part at the 50th annual meeting of the American Academy of Neurology; Minneapolis, MN.
- Received September 16, 1998.
- Accepted June 16, 2003.
References
Disputes & Debates: Rapid online correspondence
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within the last 8 weeks.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.