Patients with stroke confined to basal ganglia have diminished response

to rehabilitation efforts Ichiro Miyai, MD, PhD; Alan D. Blau, PhD; Michael J. Reding, MD; and Bruce T. Volpe, MD

Article abstract-Prediction of the functional outcome for patients with stroke has depended on the severity of impair- ment, location of brain injury, age, and general medical condition. This study compared admission and discharge func- tional outcome (Functional Independence Measure, FIM) and deficit severity (Fugl-Meyer, F-M) scores in a retrospective study of patients with similar neurologic impairments: homonymous hemianopia, hemisensory loss, and hemiparesis. CT-verified stroke location was the independent variable: cortical (n = ll), basal ganglia and internal capsule (normal cortex and thalamus, n = 131, or combined (cortical, basal ganglia, and internal capsule, n = 22). By 3 months on average after stroke, all groups demonstrated significantly improved motor function as measured by F-M scores. Patients with cortical lesions had the least CT-imaged damage and the best outcome. Patients with combined lesions and more extensive brain injury had significantly higher FIM scores ( p < 0.05) than patients with injury restricted to the basal ganglid internal capsule. Patients with basal ganglidinternal capsule injury were more likely to have hypotonia, flaccid paralysis, and persistently impaired balance and ambulation performance. While all patients had a comparable rehabilitation experience, these results suggest that patients with stroke confined to the basal ganglia and internal capsule benefited less from therapy. Isolated basal ganglia stroke may cause persistent corticothalamic-basal ganglia interactions that are dysfunctional and impede recovery. NEUROLOGY 1997;48:95-101

In several studies rehabilitative intervention has im- proved the functional outcome of patients with

but until there is a precise definition of the severity of the impairment , lesion location, and quantitative definition of functional outcome, the clinical mechanisms for improvement will remain unclear. To begin to understand these mechanisms for functional improvement, this s tudy quantified neurologic impairment and functional outcome in pa- tients with lesions confined to the cortex, or to the basal ganglia and internal capsule, or to both loca- tions: cortex, and basal ganglidinternal capsule.

Methods. We retrospectively reviewed the rehabilitation outcome of consecutively admitted patients presenting with hemiparesis, hemisensory loss, and hemianopia. All patients admitted to the Burke Rehabilitation Hospital from December 1993 to July 1995 were included. Patients who had had prior stroke, who were not functionally inde- pendent prior to admission, or who had any complications requiring transfer to an acute-care hospital were excluded. These inclusion and exclusion criteria permitted a compar- ison of the outcome of stroke rehabilitation among patients with comparable neurologic deficit^^.^ and with the same quantity and quality of rehabilitation experience. Location of the brain lesions was classified as cortical, basal ganglidinternal capsule, or combined based on CT infor- mation obtained on average 8 2 3 days after the onset. The cortical lesion (CX) involved cortical and subcortical white

matter, but not the basal ganglia, corona radiata, or inter- nal capsule. The basal ganglia lesion (BG) involved the caudate, putamen, and corona radiata with or without in- ternal capsule damage. The combined lesion (COM) in- volved cortex, corona radiata, and the basal ganglia with or without internal capsule damage. The thalamus was also examined in each scan, and was not damaged in any patient. In order to estimate the volume of injury, each lesion on CT was transposed to the right hemisphere of the standardized horizontal brain template^.^ These images were then superimposed using NIH Image Version 1.54. to produce lesion density maps for each group of patients.

Reliable and valid scales were used to measure neuro- logic impairment and functional outcome; namely the Fugl-Meyer scale (F-M),'O ambulation endurance," and the Functional Independence Measure (FIM).12 Each patient was evaluated a t the time of admission and discharge. All patients had multidisciplinary rehabilitation in the same stroke unit. Self-care subscore of FIM was the sum of scores for eating, grooming, bathing, dressing upper body, dressing lower body, toileting, bladder management, and bowel management (max = 56). Ambulation subscore of FIM was the sum of scores for transfers: to bed, to toilet, to tub; and walking on level surfaces and on stairs (max = 35). Communication subscore was the sum of scores for comprehension and expression (max = 14). Cognition sub- score of FIM was the sum of scores for communication, social interaction, problem solving, and memory (max = 35). Total FIM score then was the sum of self-care, ambu- lation, and cognition subscores (max = 126). Total FM

From the Department of Neurology, Cornell University Medical College, The Burke Rehabilitation Center, White Plains, Ny. Received May 17, 1996. Accepted in final form June 25, 1996. Address correspondence and reprint requests to Dr. Bruce T. Volpe, Department of Neurology, Cornell University Medical College, The Burke Rehabilitation Center, 785 Mamaroneck Ave, White Plains, NY, 10605.

Copyright 0 1997 by the American Academy of Neurology 95

Figure 1 . This photomontage depicts lesion density maps (A, D. G) of stroke patients with motor, sensory, and visual deficits, and examples of CT from patients with the largest (B, E, H) and smallest (C, F, I!, lesions within each group. Patients had a cortical lesion (CX, A-C) that in- volved cortical and subcortical white matter, but not the basal ganglia, corona radiata, or internal capsule; a basal ganglia lesion (BG', D-F) that involved the caudate lznd puta- men and corona radiata u t h or without internal capsule damage; or a combined lesion (COM, C-I) that involved cortex, corona radiata, and the basal ganglia with or without internal capsule damage. Each le- sion on CT was transposed to the right hemisphere of the standardized horizontal brain templates.' These images were then superimposed us- ing NIH Image Version 1.54 to pro- duce lesion density maps lor each group of patients. (A, CX; D, BG; G, COM).

score (max = 202) was the sum of scores for upper extrem- ity (max = 661, lower extremity (max = 34), balance (max = 14), joint pain (max = 441, and passive motion (max = 44). Sensation subscore of F-M was omitted be- cause approximately half of the patients could not be eval- uated due to aphasia.

Statistical analysis relied on an ANOVA in which lesion group, functional outcome scores, and age were factors.13

Results. Forty-six of 330 patients had hemiparesis, hemisensory loss, and hemianopia. All patients had dam- age in the cortical or subcortical region that was consistent with the neurologic deficit, namely, the sensorimotor cor- tex and the subcortical sensory and motor pathways and the optic radiations (for the BG group, posterior to the internal capsule approximately 3.1 cm dorsal to the can- thomeatal line, for the CX and COM group, temporal and parietal lobes). Further motor evaluation demonstrated that flaccidity (all the articular segments of the affected upper and lower extremities) was present in 0 of 11 pa- tients with CX, 5 of 13 patients (38%) with RG, and 4 of 22 patients (18%) with COM. CT analysis revealed that 11 patients had CX lesions, 13 patients had BG lesions, and 22 patients had COM lesions. Figure 1 demonstrates the lesion density map for each group and examples of the largest and smallest injury within each group. The brain injury was always confined to the territory of the middle cerebral artery, but the thalamus was normal in all scans.

Demographic features of patients in each group are 96 NEUROLOGY 48 January 1997

shown in table 1. Patients in the three groups demon- strated comparable age, sex distribution, interval post- stroke, side of stroke, type of stroke (infarction or hemor- rhage), Mini-Mental State Examination score and compli- cations.

Table 2 and figure 2 display the dynamic (changes in FIM for all groups. The first analysis attempted to test whether age had an effect on the change in FIM scores. A three-factor mixed ANOVA was performed with group (CX versus BG versus COM) and age (above and below 65) as between-subject factors. Time (admission versus dis- charge) was the within-subject factor. The dependent vari- able was the total FIM score. There was no main effect of age [F(1,40) = 0.014, ns], nor was there any significant age interaction with other variables. Since age had no effect on the change in FIM scores over time, it was dropped from further analysis.

The second analysis focused on the change of FIM scores for each group over time. There were significant main effects for group [F(2,40) = 6.107, p < 0.0051, and for time [F(1,40) = 75.785, p < 0.00011. All group:; improved significantly from admission to discharge in FIM scores [CX, F(1,lO) = 23.939, p < 0.001; BG: F(1,12) = 15.431, p < 0.002; COM: F(1,21) = 64.874, p < 0.00111. Impor- tantly, there was a significant interaction between group and time [F(2,40) = 4.996, p < 0.021. Post hoc analysis using the least significant difference test (LSD) revealed that the CX group had a significantly higher admission total FIM score (76 2 5) than either the BG group (54 2 7,

Table 1 Demographic features in stroke patients with motor, sensory, and visual deficits

Age (years)

Sex (M/F)

Days after stroke

Side (WL)

Type (I/H) MMSE

Complications

Hypertension

Ischemic heart disease

Diabetes mellitus

Atrial fibrillation

Congestive heart failure

Hypercholesterolemia

COPD

Depression

cx BG COM p Value

(n = 11)

64 2 4

318

24 2 3

912

11/0

19 * 3

7 (64%)

4 (36%)

4 (36%)

3 (27%)

2 (18%)

3 (27%)

0 (0%)

2 (18%)

(n = 13)

73 2 3

617

31 f 5

914

10/3

13 2 3

7 (54%)

3 (23%)

3 (23%))

4 (31%)

3 (23%)

1(8%)

0 (0%)

3 (23%)

(n = 22)

64 2 2

12/10

25 * 5

11/11

2012

11 t 2

15 (68%)

8 (36%)

10 (45%)

6 (27%)

6 (27%)

2 (9%)

2 (9%)

6 (27%)

n.s.*

n.s.t

n.s.;k

n.s.1

n.s.t

n.s.:b

n.s.t

n.s.t

n.s:t

n.s.t

n.s.t

n s t

n.s.t

n s t

Data are mean 2 SEM.

t Chi-square test. * ANOVA.

CX = cortical lesion; BG = basal ganglia lesion; COM = combined lesion of CX and BG; M = male; F = female; R = right; L = left; H = hemorrhage; MMSE = Mini-Mental State Examination; COPD = chronic obstructive pulmonary disease.

p < 0.01) or COM (52 2 3, p < 0.005). The BG and COM groups had comparable FIM scores on admission. Further, the COM group discharge FIM score (81 2 4) was no dif- ferent from that of the CX group (96 2 5 ) and was signifi- cantly higher than the BG group (65 ? 6, p < 0.05). Post hoc analysis of the subscores of the FIM demonstrated that the differences in total FIM depended on self-care and ambulation subscores ( p < 0.02, COM versus BG), but not on the communication or cognition subscore (see table 2).

Next, an ANOVA was performed with ambulation en- durance as the dependent variable. Results demonstrated main effects for group [F(2,40) = 8.250, p < 0.0021 and time [F(1,40) = 8 7 . 1 5 4 , ~ < 0.00011 and a significant inter- action between group and time [F(2,40) = 9.509, p < 0.00011. These results are consistent with the observation for total FIM score. Post hoc analysis showed that ambula- tion endurance at admission was significantly better in the CX group (140 f 38 ft) than BG (32 2 16 ft, p < 0.02) or COM (46 2 23 ft, p < 0.02). The BG and COM groups had comparable admission ambulation endurance. Once again, on discharge the CX group was significantly better than the other two, but the COM group improved to a higher functional level than BG [discharge endurance scores (see table 2) CX (577 2 111 ft), BG (119 ? 27 ft), COM (320 2 45 ft)].

To analyze the neurologic deficit scores, a similar ANOVA with total F-M score as the dependent variable was applied. Results demonstrate main effects for group [F(2,34) = 1 9 . 8 0 2 , ~ < 0.00011, and time [F(1,34) = 7.391, p < 0.021. Also there was no interaction between group and time [F(2,34) = 0.978, ns]. Admission total F-M score

did not differ between the BG (103 rfr 5 ) and the COM groups (100 t 4), however, it was significantly better in the CX group (149 t 10, p < 0.0001). Similarly, discharge total F-M score did not differ between the BG (109 t 8) and the COM (110 t 6) groups. It was significantly better in CX (160 2 9, p < 0.0001, see table 2). Thus, neurologic deficit improved comparably for all groups. These results suggest that the remarkable functional improvement in the COM group compared with the BG group cannot be dependent on change in neurologic deficit measured by F-M score.

Subscore analysis of F-M revealed that the only sub- score with significant main effects for group [F(2,41) = 8.663, p < 0.0011 and time [F(1,41) = 69.837, p < 0.0011 and a significant interaction [F(2,41) = 3.584, p < 0.051 was the balance score of F-M. Post hoc analysis revealed that the CX group had a significantly higher admission F-M balance score (8 2 1) than either the COM (5 2 1, p < 0.0005) or BG groups (4 2 1, p < 0.0005) without differ- ence between COM and BG. However at the time of dis- charge, the COM group F-M balance score (8, 2 1) was comparable to the CX group (10 2 l ) , and was significantly higher than the BG group (6 5 1, p < 0.05, see table 2).

Discharge disposition (home versus nursing home) did not differ among the three groups. Home/nursing home was 11/0 in CX, 10/3 in BG, and 19/3 in COM. Length of stay (LOS) was 42 2 5 days in the CX group, 54 ? 5 days in BG, and 65 ? 5 days in COM. The COM group had significantly longer LOS than CX ( p < 0.01, LSD), how- ever, there was no difference between BG and COM nor between CX and BG.

January 1997 NEUROLOGY 48 97

Table 2 FIM, ambulation endurance, and F-M in stroke patients with motor, sensory, and visual deficits

Post hoc tests

CX vs BG CX vs COM COM vs BG CX(n = 11) BG (n = 13) COM (n = 22)

FIM Total

Admission

Discharge

FIM Self-care

Admission

Discharge

FIM Ambulation

Admission

Discharge

FIM Communication

Admission

Discharge

FIM Cognition

Admission

Discharge

Ambulation (ft)

Admission

Discharge

F-M Total

Admission

Discharge

F-M UE+Le

Admission

Discharge

F-M UE

Admission

Discharge

F-M LE Admission

Discharge

F-M Balance

Admission

Discharge

F-M Sense"

Admission

Discharge

F-M Pain

Admission

Discharge

F-M ROM

Admission

Discharge

76 + 5

96 t 5

54 t 7

65 + 6

52 t 3

81 t 4

<0.01 <0.005

<0.001 n s .

n.s.

0.05

33 t 2

42 -+ 3

25 t 3

28 t 3

23 t 2

37 t 2

<0.05 (0.01

<0.005 n.s.

n.s.

(0.02

17 t 2

24 + 2

10 t 1

15 + 1

11 t 1

20 t 1

=0.0001 < 0.000 1

<0.001 n.s.

11,s.

(0.02

10 t 1

11 2 1

8 2 1

LO + 1

7 5 1

9 ? 1

n.s. n.s.

n.s. n.s.

n.s.

n s .

26 + 3

30 t 1

18 t 3

22 ? 3

18 2 2

24 f 2

n.s. CO.05

n.s. n s .

n.s.

n.s.

140 t 38

577 t 111

32 -+ 16

119 2 27

46 f 23

320 + 45

C0.02 (0.02

<0.0005 <0.02

n s .

=0.005

149 + 10

160 + 9

103 ? 5

109 t 8

100 t 4

110 f 6

~ 0 . 0 0 0 1 ~ 0 . 0 0 0 1

<0.0005 <0.0001

n.s.

n s .

56 t 9

67 % 7

20 -+ 4

30 t 3

17 + 3

29 t 4

~ 0 . 0 0 0 1 <0.0001

~ 0 . 0 0 0 1 <0.0005

n.s.

n.s.

34 ? 6

40 + 6

10 f 3

16 t 5

8 2 2

13 & 3

<0.0001 ~ 0 . 0 0 0 1

=0.001 <0.0001

n.s.

ns.

21 t 3

27 + 2

10 + 2

14 t 3

10 t 2

16 t 2

<0.005 =0.0005

<0.005 10.005

n.s.

n s .

8 + 1

10 ? 1

4 2 1

6 2 1

5 + 1

8 + 1

<0.0005 <0.0005

(0.005 n s .

n s .

<0.05

13 -+ 3

16 f 3

17 2 4

16 -+ 4

8 2 2

15 ? 2

n.s. n s .

n.s. n.s. n s .

11,s.

42 -+ 1

41 t 1

40 t 2

36 + 1

38 2 1

34 f 2

n.s. <0.05

n.s. co.01

n.s.

n s .

42 ? 1

41 t 1

41 + 1

38 t 1

39 t 1

37 + 1

n s . n.s.

n.s. 11,s.

n s .

11,s.

Data are mean f SEM.

.h For F-M Sense, n = 9 in CX, n = 6 in BG, n = 8 i n COM.

UE = upper extremity; LE = lower extremity; ROM = range of motion; FIM = Functional Independence Measure; F-M = Fugl-Meyer score; CX = cortical lesion; BG = basal ganglia lesion; COM = combined lesion.

98 NEUROLOGY 48 January 1997

110-

100 - 90 - 80 - 70 -

$ 6 0 -

50 - 40 - 30 - 20 - 10 -

T - cx - BG - COM

" I I

Figure 2. Dynamic changes of FIM total score for each group. CX group had a significantly higher admission to- tal FIM score than either BG group (p < 0.01) or COM (p < 0.005). COM group discharge FIM score was no dif- ferent from that of the CX group and significantly higher than the BG group (p < 0.05). See Results and table 2 for details.

Discussion. Patients with comparable neurologic deficit (as measured by neurologic examination and F-M deficit scores) 3 months after stroke have func- tional outcome dependent on CT localization. Age was not a factor in our study, as has been r ep~r t ed . '~ - '~ Patients with only cortical lesions had the least damage and the best outcome. However, the comparison of the combined cortex, basal gan- glia, and internal capsule (COM) group with the basal ganglia and internal capsule (BG) group dem- onstrates improved functional outcome for the COM group. These findings show that larger lesions do not necessarily predict worse functional outcome. A de- tailed examination of the differences among these groups demonstrates that the BG group had persis- tently impaired balance, ambulation, and persistellt hypotonia. There is precedent for these findings. Dromerick and RedingZO suggested that discharge Barthel Index scores were superior in a group with combined damage compared with a group with in- jury confined to the basal ganglia. The nonsignifi- cant differences they observed may have been due to the use of the Barthel Scale instead of more sensitive measures of functional outcome and neurologic defi- cit, namely the FIM and the F-M scales."ST2 Another precedent is found in studies that have demon- strated poor functional outcome in patients with damage in the posterior limb of the internal capsule and either the basal ganglia or the t h a l a m ~ s . ~ ~ - ~ ~

A closer analysis of the FIM and F-M subscales supports the contention that persistent imbalance and ambulation impairment contributed strongly to

the worse outcome of the BG group. Put differently, the improvements in balance and ambulation ap- peared to account for the greater improvements in functional outcome in the COM group. The BG and COM groups performed comparably on admission with respect to F-M evaluation of the upper extrem- ity, lower extremity, sensory, range of motion, and pain subscores. Disequilibrium in elderly people may be associated with diffuse subcortical white matter lesions and with poor functional outcome.26 Other investigators have also reported that balance as measured by F-M scores and ambulation perfor- mance on the FIM correlate better with functional performance and activities of daily living than the standard F-M scores of raw motor p o ~ e r . ~ ~ . ~ ~ Along these lines, investigators have reported that damage to the basal ganglia and internal capsule, particu- larly the lentiform nucleus (putamen and globus pal- l i d ~ ~ ) ~ ~ or the striatum,so was associated with hypo- tonia and muscle flaccidity. Pantano et al.29 reported further that functional outcome was poorest in pa- tients with these lesions. Similarly, our data demon- strate a higher proportion of patients with hypotonia and flaccidity in the BG group (38%) compared with the COM group (18%) and CTX group (0%). Taken together these data suggest that when the basal gan- glia and internal capsule are damaged decreased muscle tone is likely, and may be associated with persistent imbalance and poor ambulation. Yet corti- cal damage alone more often causes increased mus- cle tone, and this factor may have contributed to the best functional outcome in the CX group.

The unusual improved outcome in the COM group with more damage than the BG group prompts a detailed inspection of the anatomic injury (see figure 1). The CX group had normal internal capsular anat- omy. This lack of damage to the internal capsule almost certainly contributes to the best FIM and F-M measures, and the best functional outcome in the CX group. However, patients with COM and BG lesions had comparable damage to the anterior and to the posterior limb of the internal capsule. In sup- port of this structural analysis is the functional mea- sure, namely, the F-M demonstrated comparable neurologic deficit at admission and discharge for these two groups. Differential damage to the internal capsule cannot account for the difference of func- tional outcome between the combined and subcorti- cal group. In fact, the only difference on CT between the COM and BG groups was the larger region of damage in the overlying cortex in the COM group. Since both the BG and COM groups had similar basal ganglia damage, the difference in functional outcome among the groups might be attributed to altered basal ganglia modulation of the cortex and thalamus.31r3Z Our clinical data show that residual cortex impedes recovery, and suggests that cortex disconnected from the basal ganglia may cause per- sistent hemispheric dysfunction.

Clinical and experimental evidence suggest that direct damage to the basal ganglia-nigral network or

January 1997 NEUROLOGY 48 99

to the corticothalamic network initiates additional secondary damage to structures within the network, not damaged by the initial injury, that develops weeks after the ictus. This diaschisis injury may in- clude substantia nigra degeneration following mas- sive basal ganglia ~ t r o k e , ~ ~ - ~ ~ thalamic degeneration after cortical stroke in the territory of the MCA,40-47 and atrophy of ipsilateral putamen and thalamus following temporal l o b e c t ~ m y . ~ ~ ~ ~ ~ Further investiga- tion is necessary to determine whether these alter- ations within the corticothalamo-basal ganglia net- work are related to functional prognosis observed in our study.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

References Feigenson JS, Gitlow HS, Greenberg SD. The disability ori- ented rehabilitation unit-a major factor influencing stroke outcome. Stroke 1979;10:5-7. Jorgenson HS, Nakayama H, Raaschou HO, Larsen K, Hubbe P, Olsen TS. The effect of a stroke unit: reductions in mortal- ity, discharge rate to nursing home, length of hospital stay, and cost. A community-based study. Stroke 1995;26:1178- 1182. Kaste M, Palomaki H, Sarna S. Where and how should elderly stroke patients be treated. A randomized trial. Stroke 1995;

Reding MJ, McDowell FH. Focused stroke rehabilitation pro- grams improve outcome. Arch Neurol 1989;46:700-701. Smith DS, Goldenberg E, Ashburn A, e t al. Remedial therapy after stroke: a randomized controlled trial. BMJ 1981;282:

Strand T, Asplund K, Eriksson S, Hagg E, Lithner F, Wester PO. A non-intensive stroke unit reduces functional disability and the need for long-term hospitalization. Stroke 1985;16:

Reding MJ, Potes E. Rehabilitation outcome following initial unilateral hemispheric stroke: life table analysis approach. Stroke 1988;19:1354-1358. Litchman CD, Dietz M, Reding MJ. Clinical examination is more predictive of ambulation recovery than CT lesion loca- tion following subcortical ischemic stroke. J Neuro Rehab

Matsui T, Hirano A. An atlas of the human brain for comput- erized tomography. Tokyo: Igaku-shoin, 1978. Fugl-Meyer AR, Jaasko L, Leyman I, Olsson S, Steglind S. The post-stroke hemiplegic patient. A method for evaluation of physical performance. Scand J Rehabil Med 1975;7:13-31. Reding MJ. A model stroke classification scheme and its use in outcome research. Stroke 1990;21(suppl 2):35-37. Keith RA, Granger CV, Hamilton BB, Sherwin FS. The func- tional independence measure: a new tool for rehabilitation. In: Eisenberg MG, Grzesiak RC, eds. Advances in clinical rehabil- itation. Vol 2. New York: Springer, 1987:6-18. Rosner B. Fundamentals of biostatistics. 3rd ed. California: Duxbury Press, 1990. Adler MK, Brown CC, Acton P. Stroke rehabilitation-Is age a determinant? J Am Geriatr SOC 1980;28:499-503. Borucki S, Volpe BT, Reding MJ. The effect of age on mainte- nance of functional gains following stroke rehabilitation. J Neuro Rehab 1992;6:1-5. Dombovy ML, Basford JR, Whisnant JP, Bergstralh EJ. Dis- ability and use of rehabilitation services following stroke in Rochester, Minnesota, 1975-1979. Stroke 1987;18:830-836. Reding MJ. Age and its effect on rehabilitation outcome fol- lowing stroke. In: Frengley JD, Murray P, Wykle ML, eds. Practicing rehabilitation with geriatric clients. New York: Springer, 1990:135-153. Wade DT, Hewer RL, Wood VA. Stroke: influence of patient's sex and side of weakness on outcome. Arch Phys Med Rehabil

Wylie CM. Age and the rehabilitative care of stroke. J Am Geriatr SOC 1968;16:428-435. Dromerick AW, Reding MJ. Functional outcome for patients

26:249-253.

5 17-520.

29 -34.

1994;8:55-61.

1984;65:513-516.

with hemiparesis, hemihypesthesia, and hemianopia. Does le- sion location matter? Stroke 1995;26:2023-2026.

21. Wade DT, Collin CE, The Barthel ADL index: a standard mea- sure of physical disability? Int Disabil Stud 1988;10:61-63.

22. Keith RA. Status of measurement in stroke rehabilitation out- comes. Stroke 1990;21(Suppl 2):30-31.

23. Fries W, Danek A, Scheidtmann K, Hamburger C. Motor re- covery following capsular stroke. Role of descending pathways from multiple motor areas. Brain 1993;116:369-382.

24. Miyai I, Kajiura I, Suzuki T, Kang J , Nozaki S. Mechanism of functional recovery of hemiplegia in cerebral infarction. Wal- lerian degeneration and basal ganglia damage affect func- tional prognosis (Japanese). Neurol Med 1995;42:39-47.

25. Rascol A, Clanet M, Manelfe C, Guiraud B, Bonafe A. Pure motor hemiplegia: CT study of 30 cases. Stroke 1982;13:11- 17.

26. Baloh RW, Yue Q , Socotch TM, Jacobson KM. White matter lesions and disequilibrium in older people. I. Case-control comparison. Arch Neurol 1995;52:970-974.

27. Wood-Dauphinee SL, Williams JI , Shapiro SH. Examining outcome measures in a clinical study of stroke. ;Stroke 1990;

28. Bohannon RW, Leary KM. Standing balance and function over the course of acute rehabilitation. Arch Phys Med Reha- bil 1995;76:994-996.

29. Pantano P, Formisano R, Ricci M, et al. Prolonged muscular flaccidity after stroke. Morphological and functional brain al- terations. Brain 1995;118:1329-1338.

30. Aring CD. Flaccid hemiplegia in man. Arch Neurol Psychiatry

31. Albin RL, Young AB, Penney JB. The functional anatomy of basal ganglia disorders. Trends Neurosci 1989;12:366-375.

32. Alexander GE, Crutcher MD. Functional architecture of basal ganglia circuits: neural substrates of parallel processing. Trends Neurosci 1990;13:266-271.

33. Forno LS. Reaction of the substantia nigra to massive basal ganglia infarction. Acta Neuropathol (Berl) 1983;62:96-102.

34. Nakane M, Teraoka A, Asato R, Tamura A. Degeneration of the ipsilateral substantia nigra following cerebral infarction in the striatum. Stroke 1992;23:328-332.

35. Tamura A, Kirino T, Sano K, Takagi K, Oka H. Atrophy of the ipsilateral substantia nigra following middle cerebral artery occlusion in the rat. Brain Res 1990;510:154-157.

36. Saji M, Volpe BT. Delayed histological damage and neuron death in the substantia nigra reticulata following transient forebrain ischemia depends on the extent of initial striatal injury. Neurosci Lett 1993;155:47-50.

37. Saji M, Cohen M, Blau AD, Wessel TC, Volpe BT. Transient forebrain ischemia induces delayed injury in the substantia nigra reticulata: degeneration of GABA neurons, compensa- tory expression of GAD mRNA. Brain Res 1994;643:234-244.

38. Saji M, Blau AD, Volpe BT. Rescue of neurons in the substan- tia nigra reticulata from transneuronal degeneration by abla- tion of the subthalamic nucleus. Exp Neurol 19196;141:120- 129.

39. Volpe BT, Blau AD, Wessel TC, Saji M. Delayed. histopatho- logical neuronal damage in the substantia nigra compacta (nucleus A9) after transient forebrain ischemia. Keurobiol Dis

40. Fujie W, Kirino T, Tomukai N, Iwasawa T, Taniura A. Pro- gressive shrinkage of the thalamus following middle cerebral artery occlusion in rats. Stroke 1990;21:1485-1488.

41. Iizuka H, Sakatani K, Young W. Neural damage in the rat thalamus after cortical infarcts. Stroke 1990;21:7'90-794.

42. Kataoka K, Hayakawa T, Yamada K, Mushiroi T, Kuroda R, Mogami H. Neuronal network disturbance after focal ischemia in rats. Stroke 1989;20:1226-1235.

43. Rae ASL. Bilateral infarction of carcarine cortex with lateral geniculate degeneration. Conf Neurol 1961;21:225-232.

44. Ross DT, Ebner FF. Thalamic retrograde degeneration follow- ing cortical injury: an excitotoxic process? Neuroscience 1990;

45. Russell GV. Histological alterations in the thalamic nuclei of man following cortical lesions. Tex Rep Biol Med 1958;16:483- 492.

46. Strefling AM, Urich H. Cross cerebellar atrophy: an old prob- lem revisited. Acta Neuropathol (Berl) 1982;57:197-202.

21:73 1-739.

1940;43:302-317.

1995;2:119-127.

35~525-550.

100 NEUROLOGY 48 January 1997

47.

48.

Tamura A, Tahira Y, Nagashima H, et al. Thalamic atrophy following cerebral infarction in the territory of the middle cerebral artery. Stroke 1991;22:615-618. Parashos IA, Oxley SL, Boyko OB, Krishnan KRR. In viva

changes in two temporal lobectomy patients with affective disorder. J Neuropsychiatry Clin Neurosci 1993;5:337-341.

49. Shedlack KJ, Lee EK, Radtke RA, et al. Ipsilateral subcortical atrophy associated with temporal lobectomy. Psychiatry Res

quantitation of basal ganglia and thalamic degenerative 1994;54:295-304.

Continuous bromocriptine therapy in menstrual migraine

Andrew G. Herzog, MD, MSc

Article abstract-This investigation assessed the effects of an open, prospective trial of adjunctive continuous bromocrip- tine therapy on the frequency of refractory, disabling menstrual migraine. It compared continuous bromocriptine with previously optimal baseline therapy and cyclic perimenstrual bromocriptine use. The subjects were 24 women with disabling migraines that occurred exclusively or at least 50% of the time within 3 days before or after the onset of menstruation despite treatment. We added bromocriptine 2.5 mg three times a day to their existing regimen and compared menstrual migraine frequency during the first year with the year prior to bromocriptine. Eighteen of the 24 women experienced a 25% or greater decline in migraine frequency. Migraine frequency declined by 72% overall ( p < 0.01). Three women did not tolerate bromocriptine, and three did not benefit. None of the women had a 10% or greater increase in headaches. Continuous bromocriptine therapy was also significantly more effective than intermittent bro- mocriptine use ( p < 0.05). Continuous bromocriptine therapy appears to benefit menstrual migraine. NEUROLOGY 1997;48:101-102

Migraine commonly occurs in women with a cyclic catamenial pattern of exacerbation.' It occurs with menstruation in more than 60% of cases and exclu- sively at that time in 14%.2 Menstrual migraine is often refractory to treatment1Z2 and can cause monthly visits to the emergency room. Various ergot derivatives, such as methysergide and lisuride, show efficacy in migraine prophylaxis without particular benefit for menstrual mig~-aine.~.~ Cyclic, intermit- tent bromocriptine use benefits some features of pre- menstrual syndrome, but has little effect on men- strual Continuous-use bromocriptine in seven women with regularly occurring menstrual mi- graine, however, was associated with only one epi- sode during a total of 12 cycles of observation.s This investigation assessed the effects of an open, pro- spective trial of adjunctive, continuous bromocriptine therapy on the frequency of refractory, disabling menstrual migraine. We compared continuous bro- mocriptine with previously optimal baseline therapy and cyclic perimenstrual bromocriptine use.

Methods. The subjects were 24 women between the ages of 20 and 45 years who had refractory, disabling migraine,9 with or without auras, occurring exclusively, or at least on 50% of occasions, within 3 days before or after the onset of

menstruation. Menstrual migraines were intractable de- spite prophylactic trials of beta blockers, antidepressants, and calcium channel blockers, alone or in combination. They also failed to respond adequately to acute therapy with ergotamine and, in some cases, sumatriptan. Disabil- ity referred to an inability to pursue activities that were ordinarily carried out at work or a t home.

The 24 women were placed on bromocriptine 2.5 mg tid in addition to their previously optimal existing treatment. The number of migraine episodes during the first year was compared with the year prior to bromocriptine using the Wilcoxon matched-pairs signed rank test.

Eight of the women were also tried on cyclic use of bromocriptine prior to continuous use, starting on day 21 of each menstrual cycle until day 3 of the next cycle for four consecutive cycles.

Results. Eighteen of the 24 women (75%) experienced a 25% or greater decline in disabling migraine on continuous bromocriptine therapy. Fifteen of the women (62.5%) had a reduction of over 50%; seven (29%) experienced complete resolution. Three did not tolerate bromocriptine because of light-headedness or nausea. Headache frequency among the 21 who remained on therapy declined by 72%, from a total of 330 during the baseline year to 93 during the year of bromocriptine therapy [ p < 0.01). Three did not benefit

From the Neuroendocrine Unit, Charles A. Dana Research Institute, Beth Israel Hospital, and the Department of Neurology, Harvard Medical School, Boston, MA. Presented in part at the 47th annual meeting of the American Academy of Neurology, Seattle, WA, May 1995. Received March 1, 1996. Accepted in final form June 12, 1996. Address correspondence and reprint requests to Dr. Andrew G . Herzog, Neuroendocrine Unit, Beth Israel Hospital, 330 Brookline Avenue, Boston, MA 02215.

Copyright 0 1997 by the American Academy of Neurology 101