Wallerian degeneration of the pyramidal tract does not affect stroke rehabilitation outcome

Recent studies^1-6 suggest that Wallerian degeneration (WD) of the pyramidal tract detected by MRI may be related to poor functional outcome in stroke. Although these studies have produced a new perspective on clinical neurorehabilitation, they have presented only single-time point analysis and have focused on impairment or disability instead of functional outcome. In this report we sought to specify the onset of WD, the relation of the rehabilitation experience to the stroke event, and the effect on functional outcome. We also focused the evaluation on patients with isolated subcortical infarcts and pure motor hemiplegia because sensory and visual deficit also affect functional outcome.⁷

Methods. A total of 77 consecutive patients with pure motor hemiparesis after ischemic infarcts in either the posterior limb of the internal capsule or the corona radiata were evaluated prospectively. MRI (1.0-T superconductive; Shimadzu, Kyoto, Japan. MAGNEX Epios 10; 8.5-mm slice thickness) included T2-weighted (repetition time [TR], 3,630 msec; echo time [TE], 110 msec) axial spin-echo images and T1-weighted (TR, 500 msec; TE, 15 msec) axial spin-echo images. Positive delineation of WD occurred when a high-intensity area was detected along the pyramidal tract in three consecutive slices below the level of lesion either in the corona radiata or the internal capsule in the T2-weighted image (figure). Because WD had been demonstrated by this class of MRI criteria to occur at least 3 months after an acute stroke,^8,9 all patients had an MRI evaluation at least 3 months after the acute event.

• Download figure
• Open in new tab
• Download powerpoint

Figure. A patient with MRI evidence of Wallerian degeneration of the pyramidal tract. Note the high-intensity area along the pyramidal tract below the level of lesion in the right corona radiata. See text for details.

All patients were transferred to Bobath Memorial Hospital after conservative therapy at acute hospitals. None had complications severe enough to discontinue rehabilitation. All patients participated in a neurodevelopmental technique (NDT)¹⁰ rehabilitation program that included two 45-minute sessions of physical therapy, two 45-minute sessions of occupational therapy, and one 45-minute session of speech therapy as needed, 5 days a week. The NDT approach views segmental and suprasegmental spinal cord movement patterns as abnormal movements that should always be inhibited and never used to elicit movement.^10,11

On admission and discharge, disability was measured utilizing the Functional Independence Measure¹² (FIM), and neurologic impairment was evaluated utilizing motor subscore of Stroke Impairment Assessment Set¹³ (SIAS). We also analyzed subscores for activities of daily living (ADL), mobility, and cognition subscores of FIM.¹⁴ Motor subscores of SIAS (0 to 25) consist of two tests for upper extremity (0 to 10) and three tests for lower extremity (0 to 15).¹³

To estimate the severity of injury, each lesion on MRI was transposed to the standardized horizontal brain templates,¹⁵ and volume of lesion was calculated using the National Institutes of Health (NIH, Baltimore, MD) Image Version 1.60. Statistical analysis for outcome measures was performed using Wilcoxon's ranked sum test and analysis of variance (ANOVA). Demographic data were analyzed using unpaired t-tests and chi-square tests.

Results. There were 51 patients with WD (WD+) and 26 patients without WD (WD-). Age, sex, duration after the onset of stroke, side of infarction, Mini-Mental State Examination score, and volume of infarction were comparable for each group (table 1). However, length of stay (LOS) in the rehabilitation environment was significantly longer (p < 0.05) in WD+ (130 days) than in WD- (105 days; see table 1). There were no significant differences between the groups in the mean admission total, ADL, mobility, and cognition scores of FIM, and gains in these scores at the time of discharge (table 2). Improvement in total FIM score was due mainly to changes of ADL and mobility subscores in both WD+ and WD-. Similarly, there were no significant differences in the mean SIAS scores for upper and lower extremities or gains of these scores at the time of discharge (table 3).

Because there was a significant difference of LOS between WD+ and WD-, we tested whether time that elapsed between stroke and the commencement of the rehabilitation experience affects LOS. Patients were divided into two groups-one referred to rehabilitation ≤90 days after stroke and the other referred to rehabilitation >90 days after stroke-because the literature demonstrate that WD appears to occur at least 3 months after an acute stroke^8,9 by the MRI criteria we used. A total of 31 patients were referred ≤90 days after stroke (mean, 57 ± 4 days; WD+, n = 19; WD-, n = 12), and 46 patients were referred >90 days after stroke (mean, 186 ± 21 days; WD+, n = 32; WD-, n = 14). Chi-square analysis demonstrated that incidence of WD was not significantly different between groups. An analysis for LOS was performed using two-factor ANOVA with WD as the intersubject factor, and days after stroke set to ≤90 or >90 days from stroke to rehabilitation as the intrasubject factor. There was a significant main effect for WD (F[1,73] = 7.139, p < 0.01). However, there was no effect for the timing of the rehabilitation experience, nor was there any interaction between WD and the rehabilitation starting point. The results suggest that WD+ patients had a longer LOS in the rehabilitation setting, which was independent of timing of that experience. Finally, an ANOVA of the admission/discharge FIM and SIAS scores and their subscores demonstrated no effects for either set of scores, nor any significant interaction between the factors. These results suggest that the presence of WD and the timing of the rehabilitation with respect to the stroke had no effect on the ultimate functional outcome.

Discussion. Several studies^1-6 suggest that WD of the pyramidal tract detected by MRI is related to poor functional outcome in stroke. Most studies, however, employed only one-point evaluation of either disability or impairment. Sawlani et al.⁶ reported that WD was related to a poor Barthel index score at 1 and 4 months in 18 stroke patients.

We investigated the relation between WD and rehabilitation outcome, including both impairment and disability, in a prospective study. We studied patients with motor deficit without sensory and visual deficits to analyze the pure motor aspect of functional recovery due to damage of the subcortical motor pathway.

It is not known whether patients without MRI evidence of WD actually have WD in vivo. Increased signal intensity on T2-weighted images as seen in patients with WD might be due in part to increased water content secondary to glial proliferation.⁸ Incidence of WD of the pyramidal tract on MRI by capsular infarction has been reported to be as high as 78.6%.⁹ The incidence of WD in our study was comparable (51 of 77, 66%), and suggests the incidence of WD may change with larger lesions. It is likely that patients with WD+ have more damage in the pyramidal tract than do patients with WD-. Nevertheless, WD+ had comparable rehabilitation outcome compared with WD- both in terms of disability and impairment in pure motor hemiparesis. Patients with WD of the pyramidal tract had a longer LOS and therefore a longer rehabilitation effort than those without WD, independent of the start of rehabilitation after stroke. These data suggest that more damage of the pyramidal tract might cause a delay in functional improvement, as reported previously.^5,6 However, pathways other than the residual pyramidal tract ipsilateral to the lesions might substitute the damage.

Recovery from hemiplegia due to subcortical lesions disrupting the pyramidal tract might occur by ipsilateral and contralateral substituting pathways that bypass the site of lesions. Fries et al.¹⁶ demonstrated almost complete recovery of finger movements in patients with MRI evidence of WD. Transcortical electric response of the affected cerebral hemisphere produced bilateral motor response, suggesting polysynaptic corticoreticulospinal connection bypassing the lesions. They consider the reorganization of parallel-acting multiple motor areas (i.e., primary motor cortex, premotor cortex, and supplementary motor area) ipsilateral to the lesion side as the central mechanism in motor recovery.¹⁷ Conversely, there have been several reports that suggest reorganization of the cerebral hemisphere contralateral to the lesion is associated with the mechanism of functional recovery. Jones et al.¹⁸ demonstrated impairment of ipsilateral sensorimotor function due to unilateral cerebral infarction. Furthermore, Fisher¹⁹ reported two patients with pure motor hemiplegia who developed another pure motor stroke on the opposite side, causing exacerbation of preexisting hemiplegia as well as new hemiplegia. However, ipsilaterally evoked motor responses to transcranial magnetic stimulation in the nonaffected hemisphere after stroke were not correlated to clinical improvement of hemiparesis.^20-22 Recent PET studies^23-26 revealed that recovery of motor function is associated with complex changes of regional cerebral blood flow (rCBF), including both the damaged and the undamaged hemisphere, and that the pattern of rCBF varies between patients. Findings of functional MRI studies are in good agreement with these results of PET studies.^27,28

Although the pathways that substitute for damage to the subcortical motor tract were not determined in our study, it is likely that WD of the pyramidal tract does not affect final rehabilitation outcome and that residual pyramidal tract contributes little to functional recovery, especially in chronic stroke.

Acknowledgment

The authors thank Dr. Bruce T. Volpe for reviewing the manuscript and Ms. Naomi Hoshina for assistance with data collection.