ITEM METADATA RECORD
Title: Trunk impairment after stroke : development of a clinical tool to predict functional outcome and determine recovery patterns
Other Titles: Belang van romprevalidatie na een cerebrovasculair accident : schaalontwikkeling, predictieve waarde en herstelpatroon
Authors: Verheyden, Geert
Issue Date: 27-May-2006
Abstract: This study focussed on three major issues: (1) can we develop a clinical tool to measure trunk performance which takes into account the assessment of selective trunk movements, evaluation of quality of trunk movement and is a valid instrument for measuring rehabilitation outcome? (2) What is the importance of trunk function in relation to motor and functional recovery after stroke? And (3) what is the pattern of recovery of the trunk and is it comparable to the pattern of recovery of arm, leg and functional activity after stroke?

At the start of this doctoral study, there was a clear lack of clinical tools which assess trunk impairment after stroke. In the systematic review presented in chapter 1 we discussed three clinical tools which specifically evaluate trunk performance after stroke, the Trunk Control Test and two Trunk Impairment Scales. Psychometric properties of the three scales were compared. The most extensively appraised tools are the two Trunk Impairment Scales. It was established that the Trunk Impairment Scale developed in this doctoral study had sufficiently robust psychometric characteristics to make it a valuable assessment tool. Furthermore and contrary to the other two scales it has no ceiling effect.

The Trunk Impairment Scale developed at the Department of Rehabilitation Sciences at the Katholieke Universiteit Leuven is presented in chapter 2. The scale consists of three subscales: static and dynamic sitting balance and trunk co-ordination. Static sitting balance assesses if the patient can remain in the seated position at the side of the bed with the feet flat on the floor and if the patient can maintain this position when the legs are crossed passively by the therapist and actively by the patient. Dynamic sitting balance evaluates if the patient can selectively perform a lateral flexion, initiated from the upper and lower part of the trunk. Finally co-ordination examines if the patient can perform a selective rotation of the upper and lower part of the trunk. The score for the static, dynamic and co-ordination subscales range from 0 to 7, 10 and 6 points, respectively. The total score for the Trunk Impairment Scale ranges from 0 to 23 points, a higher score indicating a better trunk performance. Test-retest and interrater reliability were established. Kappa and weighted Kappa values, percentages agreement and intraclass correlation coefficients (ICC) were calculated on a sample of 28 stroke patients. ICC values for test-retest and interrater agreement for the total Trunk Impairment Scale score were 0.96 and 0.99. The 95% limits of agreement for the test-retest and interexaminer measurement error were between -2.90 and 3.68 and -1.84 and 1.84, respectively. Furthermore internal consistency and content, construct and concurrent validity were established. It was concluded that the Trunk Impairment Scale has good clinimetric characteristics. The scale evaluates selective movements of the trunk, taking into account the quality of the trunk movement and is related to actual treatment of the trunk in stroke patients because of its established content validity.

The study presented in chapter 3 answers the question if the items of the scale would be able to discriminate stroke patients from healthy individuals. Forty stroke patients and 40 age- and sex-matched healthy individuals were evaluated with the Trunk Impairment Scale. It was concluded that the Trunk Impairment Scale clearly discriminates between stroke patients and healthy individuals. Significant differences (p<.0001) were found for the total Trunk Impairment Scale and its subscale scores between the two groups. Forty-five percent of the healthy individuals did not score the maximum value on the Trunk Impairment Scale. It was suggested that a score of 20 out of 23 could be considered as a borderline value to discriminate a normal from an impaired trunk performance.

Chapter 4 and 5 present the psychometric properties of the Trunk Impairment Scale for patients with multiple sclerosis and traumatic brain injury. Intraclass correlation coefficients of 0.95 and 0.97 for the Trunk Impairment Scale and 0.92 and 0.93 for the trunk assessment of the Melsbroek Disability Scoring Test established test-retest and interrater reliability for both tests on a sample of 30 patients with MS. Bland Altman plots provided further evidence of consistency of scores over the full range without observer bias. It was concluded that both clinical tools have good clinimetric characteristics and are applicable for measuring trunk performance in persons with multiple sclerosis. Reliability and validity of the Trunk Impairment Scale was also established for patients with traumatic brain injury on a sample of 30 patients. ICC values of 0.88 and 0.95 were reported for test-retest and interrater reliability for the total Trunk Impairment Scale. The 95% limits of agreement for test-retest and interexaminer measurement error were -4,4 and -3,3 respectively. A correlation coefficient between the Trunk Impairment Scale and the Barthel index of 0.59 (p = .0007) established construct validity. It was concluded that acceptable psychometric properties allows the use of the Trunk Impairment Scale in rehabilitation treatment and research for patients with traumatic brain injury.

Chapter 6 presents the cross-sectional study which was performed to compare the Trunk Control Test with the Trunk Impairment Scale and to relate trunk function to measures of gait, balance and functional ability. Fifty-one non-acute and chronic stroke patients were evaluated. Measures of gait, balance and functional ability comprised the Tinetti balance and gait test, Functional Ambulation Category score, 10-m walk test, timed Up and Go test and motor part of the Functional Independence Measure. This study demonstrated that the trunk is still impaired in non-acute and chronic stroke patients. Median score on the Trunk Control Test was 61 out of 100 (61%) and 11 out of 23 (48%) on the Trunk Impairment Scale. Twelve participants (24%) obtained the maximum score for the Trunk Control Test, indicating an important ceiling effect in the examined population. No subject obtained the maximum score on the Trunk Impairment Scale. Multiple regression analysis emphasized the importance of trunk performance in relation to measures of balance, gait and functional ability (Model R² = 0.55 to 0.62).

The predictive validity of the Trunk Impairment Scale was addressed in chapter 7. A total of 102 subjects were evaluated on admission to a rehabilitation centre and at six months after stroke. The study was conducted in three European rehabilitation centres and was part of the ‘Collaborative Evaluation of Rehabilitation of Stroke across Europe’ (CERISE) project. The impact of trunk performance on functional recovery after stroke was further revealed by the results of this study indicating that Trunk Impairment Scale total score (R² = 0.52, p<.0001) and static sitting balance subscale score on admission (R² = 0.50, p<.0001) were significant predictors of Barthel index score at six months after stroke. The total TIS and static sitting balance subscale score were better predictors of Barthel index score at six months than Barthel index score on admission itself.

Having established that trunk performance as measured by the Trunk Impairment Scale was an important determining factor of the outcome of rehabilitation of stroke patients, chapter 8 focused on possible differences between trunk recovery patterns and that of distal parts and of functional ability. Measures of arm, leg, trunk and functional performance were assessed at one week, one month, three months and six months after stroke. Contrary to common belief, it was found that the recovery patterns for trunk, arm, leg and functional activity showed remarkable similarities. Furthermore, the results of the study revealed that patients after stroke improved significantly for all measures between one week an one month and between one month and three months after stroke with the most improvement noted from one week to one month after stroke. No significant difference was found between three and six months after stroke, indicating an important stagnation or ‘plateau phase’ and even a deterioration in motor and functional performance.
Publication status: published
KU Leuven publication type: TH
Appears in Collections:Research Group for Neuromotor Rehabilitation

Files in This Item:
File Description Status SizeFormat
Geert Verheyden_pdf_doctoraat_definitief.pdf Published 616KbAdobe PDFView/Open Request a copy

These files are only available to some KU Leuven Association staff members

 




All items in Lirias are protected by copyright, with all rights reserved.