Effect of hand splints on stereotypic hand behavior of girls with Rett Syndrome: a replication study.

The purposes of this study were to replicate a recent report of the positive effects of band splinting on the stereotypic hand movement of children with Rett syndrome and to evaluate the generality of these results to a different setting. Two 5-year-old girls diagnosed with early Stage-III Rett syndrome were introduced to hand splints in accordance with the multiple-baseline design used in the Naganuma and Billingsley study Splint wear ranged from 30 to 50 days for the two subjects Data were analyzed as a percentage of time and as actual time in minutes. Unlike the previous study, in which a decrease in hand-wringing behavior was noted, neither subject in our study demonstrated a decrease in stereotypic band behavior or a subsequent increase in independent feeding skills when wearing the splints. There was also no evidence of increased hand wringing following withdrawal of the splints The differences in ages of the subjects and different functional levels (stages) may have been contributing factors to the conflicting results and should be considered in managing this group of children. [Tuten H, Mtedanerj Effect of band splints on stereotypic band behavior of girls with Rett syndrome: A replication study. Phys Ther 69:1099-1103, 1989] Key Words: Hand; oribotics/splints/casts, upper extremity; Stereotyped behavior. Rett syndrome is a progressive sexlinked disorder characterized by a sudden loss of motor and cognitive abilities in young girls between the ages of 7 and 18 months.1-9 Other characteristic symptoms include deceleration of head growth, loss of facial expression, increased seizure tendencies, autistic behaviors, and gait apraxia. The most apparent characteristic of Rett syndrome is the loss of purposeful hand movements and the concurrent appearance of the stereotypic hand-wringing, mouthing, and biting behaviors.1-9 A variety of management strategies have been proposed in an attempt to decrease the stereotypic hand behavior of children with Rett syndrome. Previous authors were unable to modify the stereotypic behavior by using a stimulating joyful environment, drugs, or psychological means. Esptein et al were successful in reducing inappropriate hand and foot movements in two children with schizophrenia by using overcorrection.13 Once the intervention was stopped, the inappropriate hand and foot movements either reverted to the original state or became exacerbated. In 1988, Naganuma and Billingsley reported reduction of stereotypic hand behavior and a subsequent increase in purposeful hand movements for eating and playing with the use of hand splints., Because of the variability of reported characteristics and management effectiveness for this syndrome, we felt that it was important to replicate the most recent management study for validation and additional insight. Additionally, we were interested in assessing whether hand wringing would increase during the orthosis withdrawal phase as seen in the Naganuma and Billingsley study. Our supposition was that stereotypic hand behavior would decrease with the implementation of hand splints. Method Subjects Two 5-year-old girls with a diagnosis of Rett syndrome identified at 2 years of age and enrolled in a preschool program for children with developmental disabilities served as the subjects for this replication study. Both children attended school six hours per day, five days per week. Both subjects participated in a physical therapy program two or three times per month while at school. Subject A" typically demonstrated stereotypic hand-to-mouth and rocking behaviors. Subject B" demonstrated stereotypic hand-to-hand behavior patterns, with the dorsal surface of one hand always meeting the palmar surface of the other hand at about chin level. Subject B would attempt to grasp a cracker when placed in front of her using a rhythmical raking-mass-grasp pattern while maintaining a modified version of the stereotypic hand behavior. Neither subject demonstrated any functional hand movements; neither subject had ever used hand splints. Written informed parental consent was obtained for each of the subjects prior to the study. Procedure The hand splints used in this study were fabricated according to the specifications outlined in the Naganuma and Billingsley study (Fig. 1), with wearing times increased according to tolerance. Both subjects adapted to full wearing time within two days, and die splints were subsequently worn Monday through Friday from 8:30 AM to 2 PM. A small piece of foam rubber was added to the outside of the palmar surface of the right splint to prevent Subject B from irritating the dorsal surface of the left hand while performing stereotypic hand behavior. The measurement technique used was identical to that of the original study. Each child was videotaped for a five-minute segment of free-time sitting at a table listening to music through headphones and for a five-minute segment where a small cracker was placed on the table in front of the child finger feeding). Data were collected in a similar manner, except free-play observations were videotaped five days per week instead of three times per week as in the original study. A college graduate with no physical therapy background participated as the observer for our data collection. He was blind to the purpose of the study. Following a one hour training session, this observer's assessments were compared with those of a physical therapist (HT) employed at a local preschool center. Using the first hour of recorded observations, interrater reliability (r) for observing stereotypic hand movement during both free play and finger feeding was .98. Reliability was monitored once a week throughout the study and did not vary significantly from the initial results. The multiple-baseline across-subjects design was replicated from the original study; however, additional data were collected during the withdrawal phase to assess changes in stereotypic behavior. Hand splints were applied in a sequential fashion once a steady baseline level was achieved. On Day 1 1, Subject A began wearing the hand splints. Baseline data were collected as before for Subject B. Subject A tolerated the splints all day on Day 12 and subsequently without complications. The splints were introduced to Subject B on Day 31. She was able to tolerate the splints full-time on Day 32 and subsequently without complications. The withdrawal phase for both subjects began on Day 60 and continued for eight days. Neither child wore the splints from Day 36 to Day 44 while they were on summer vacation. Otherwise, wearing compliance was 100%. Results Unlike the previous study, we did not observe a consistent decrease in stereotypic hand behavior following intervention with the hand splints (Figs. 2-5). Figure 3 illustrates the percentage of time spent performing stereotypic hand movements during the feeding portion of the study for both subjects. In the original study, the results were displayed on a similar graph. Figure 2 represents these same data reported in actual minutes instead of percentages of time involved in stereotypic hand behavior. During our feeding observations, Subject A maintained stereotypic hand behavior 93.2% of the time. Following introduction of the hand splints, her percentage of time spent performing stereotypic hand behavior remained consistent at 93.5%. After removal of the hand splints, her stereotypic hand behavior decreased slightly to 88.3%. Subject B demonstrated a similar pattern during feeding observations. Subject B's mean duration of stereotypic hand behavior was 70.0% during the baseline phase, 77.7% during the intervention phase, and 80.3% during the withdrawal phase. During freetime observations, Subject A's mean duration of stereotypic hand behavior was 78.2% during the baseline phase, 84.6% during the intervention phase, and 80.7% during the withdrawal phase. Similar results were found for Subject B during the free-time portion of the study. As in the feeding portion of the study, Subject B showed more variability than the subjects in the original study, with a baseline frequency of 67.4%, an intervention frequency of 83.1%, and a withdrawal frequency of 82.7%. Like Subject A, Subject B demonstrated no decrease in hand wringing following intervention with the splints, nor did she demonstrate a subsequent increase of stereotypic hand behavior following withdrawal, as noted in the original study. Discussion The Naganuma and Billingsley study claimed a decrease of stereotypic hand behavior in all three of their subjects after application of hand splints. Our replication of this study showed no change in the occurrence of stereotypic hand behavior for two children. Our study replicated the original study, except for the following modifications: 1. There were two subjects in our

study as compared with three subjects

in the previous study. 2. Data collection during the withdrawal

phase was increased compared

with the original study. 3. Both feeding and free-time portions

of our study were videotaped

on a daily basis, whereas the original

study videotaped free-time

behavior three times per week. 4. An observer blind to the purpose

of the study was used in our study

to eliminate any bias during data

collection, whereas no blinded

observer participated in the original

study. 5. The girls in our study were 5 years

of age, whereas the girls in the

original study were 13 and 16

years of age. 6. All three subjects in the original

study showed some degree of

functional hand skills, ranging from

holding an object if placed into the

hand to independent finger feeding

prior to the use of hand splints,

in contrast to both subjects in our

study. According to Hagberg and WittEngerstrom's staging system of skill levels in Rett syndrome, the three subjects from the original study most likely represent Stage IV, the late motor deterioration stage. This stage is characterized by significant weakness; muscle wasting; spasticity; deformities, particularly of the feet and spine; a concurrent improvement in emotional contact; and less active, more easily controlled seizure activity. Neither of the subjects in our study showed any purposeful hand movement prior to the use of hand splints. The girls in our study demonstrated rocking behaviors and intermittent respiratory dysfunction during wakefulness, including breath holding and hyperventilation, and one subject was prone to seizure activity. According to the staging system of Hagberg and Witt-Engerstrom, these two subjects represent early Stage III, also known as the "pseudo stationary stage." This stage is characterized by low-level hand use and speech ability.9 Perhaps the benefit of hand splinting cannot be generalized to patients with Rett syndrome as a whole. Possibly these benefits can be generalized to specific stages, or perhaps each case of Rett syndrome must be individually analyzed. It does appear that the existence of some degree of purposeful hand movement is a prerequisite to the beneficial use of hand splints. In the previous study, the stereotypic hand behavior was displayed as a percentage of time. Smaller changes tend to be more dramatic when displayed as a percentage of time rather than actual time in minutes. When we translated percentages of time (Figs. 3, 5) to actual time in minutes (Figs. 2, 4), the changes did not appear as significant. For example, a change of 25% represents a reduction of stereotypic hand behavior of 1.2 minutes. A 50% reduction represents 2.5 minutes. If the initial figure is extrapolated over a 12-hour period, a 25% decrease would mean a decrease of 2.9 hours in stereotypic hand behavior. Whether this decrease is clinically significant is a question that we must consider. Conclusion The purpose of this study was to determine the generality of the results of a previous study in which the effect of hand splints on stereotypic hand behavior for patients with Rett syndrome was investigated. The stereotypic hand behavior of the girls in our study did not decrease as in the previous study following the introduction of the hand splints. The differences in ages of the subjects for the two studies and the different functional levels (stages) may have been contributing factors and should be considered in managing this group of children. Hopefully, this replication study will provide increased information by which management decisions for children with Rett syndrome can subsequently be made.