Behavioural Brain Research 250 (2013) 174–191

Contents lists available at SciVerse ScienceDirect

Behavioural Brain Research

j ournal homepage: www.elsevier.com/locate/bbr

Research report

Assessing gait impairment after permanent middle cerebral artery

ଝ

occlusion in rats using an automated computer-aided control system

a a b a,∗ b,∗ a

SiDi Li , Zhe Shi , Heng Zhang , XinMin Liu , ShanGuang Chen , Jian Jin ,

a c d

Yi Wang , William Jia , HaiQing Li

a

Research Center for Pharmacology & Toxicology, Institute of Medicinal Plant Development (IMPLAD), Chinese Academy of Medical Sciences and Peking

Union Medical College, Beijing 100193, China

b

National Key Laboratory of Human Factors Engineering, Astronaut Centre of China, Beijing 100193, China

c

Department of Surgery and Brain Research Centre, University of British Columbia, F233-2211 Wesbrook Mall, Vancouver, British Columbia, Canada

d

Beijing Xin Hai Hua Yi Technology Co., Ltd., Beijing 100193, China

h i g h l i g h t s

•

pMCAO rats were firstly used to assess long-term gait deficits by a computer-assisted method.

•

Accurate parameters can illuminate impairment and compensation in rats during walk.

•

Average body rotation and propulsion index are first used in cerebral ischemia rats.

•

Correlations between the final infarct size and earlier gait deficits were significant.

•

Gait analysis is a promising tool to investigate mechanisms of stroke and evaluate potential therapies in rats.

a r t i c l e i n f o a b s t r a c t

Article history: Systematic gait analyses have been widely used in clinical settings as a reliable means of evaluating

Received 12 February 2013

stroke severity and the efficacy of rehabilitation on people. However, the extent of gait changes post-

Received in revised form 20 April 2013

stroke in experimental quadrupeds remains to be explored. To date, gait studies in cerebral ischemia

Accepted 25 April 2013

have been limited to the mild ischemia-reperfusion model. However, studies on pathophysiology and

Available online 1 May 2013

therapy of experimental stroke suggest that permanent middle cerebral artery occlusion (pMCAO) is

more similar to naturally occurring cerebral ischemia in humans. This is the first preclinical study to

Keywords:

demonstrate that pMCAO rats can be used to assess long-term functional deficits related to gait by a

Ischemic stroke

computer-assisted method. Our gait analysis results demonstrate obvious gait deficits in the acute phase

Gait analysis

of the disease. During recovery, gait function gradually improved, but deficits were still detectable 42

Permanent middle cerebral artery occlusion

Rat days post-pMCAO. Objective and accurate photogrammetric parameters were used to illuminate laws of

impairment and compensation in rats at different stages of cerebral ischemia in injured and uninjured

limbs during walking. Compared to previous gait studies involving transient (t) MCAO rats, gait changes

observed in pMCAO rats were more similar to changes following naturally occurring cerebral ischemia in

humans. Importantly, the average body rotation and propulsion index, not previously used, are specific

parameters for accurately assessing gait function during the acute phase of post-pMCAO. Furthermore,

the gait test results revealed significant correlations between the final infarction volume and earlier

behavioral outcomes. In conclusion, the gait analysis is a promising tool for assessing cerebral ischemia

severity, and that it may provide a new means of investigating mechanisms of cerebral ischemia and

evaluating potential therapies. © 2013 The Authors. Published by Elsevier B.V. All rights reserved.

ଝ

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which permits non-

commercial use, distribution, and reproduction in any medium, provided the original author and source are credited.

∗

Corresponding authors. Tel.: +86 010 6281 2595/9108; fax: +86 010 6281 2595/9108.

E-mail addresses: [email protected] (S. Li), [email protected], [email protected] (X. Liu).

0166-4328/$ – see front matter © 2013 The Authors. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbr.2013.04.044

S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 175

1. Introduction 2. Materials and methods

2.1. Experimental animals and environmental conditions

Ischemic stroke is one of the most frequent causes of death and

leading cause of adult disability worldwide [1,2]. Even patients who

Male SD rats (HFK, Beijing, China), 10 weeks old, weighing 300–320 g at the

have survived from stroke, 90% of them suffer permanent neuro- beginning of the experiments, were housed with lights on from 7:00 h to 19:00 h.

logical deficits. It is essential for stroke survivors to have a lengthy Surgery was performed after all the rats were allowed to acclimate for 1 week. Rats

were housed in groups of four per cage with ad libitum access to food and water.

program of rehabilitation, followed by life-long clinical support

The animals were divided into two groups for use in gait test at all the observing

with rehabilitation therapy in order to improve their life quality

time points. These groups consisted of pMCAO (n = 8) and sham (n = 8) rats respec-

◦

[3,4]. tively. Room temperature and humidity were maintained at 23 ± 1 C and 65 ± 5%,

However, treatment options are very limited at present and respectively.

All experiments were performed in accordance with protocols approved by the

many promising agents used in stroke models have failed to

Committee for the Care and Use of Laboratory Animals of IMPLAD, CAMS & PUMC,

live up to expectations in clinical studies [5]. This may be

China and were based on the “Principles of Laboratory Animal Care” (NIH publi-

partly because many preclinical studies have employed a narrow

cation no. 86-23, revised in 1996) and PR China legislation for the care and use of

window after the ischemic episode for evaluating functional out- laboratory animals. Sufficient measures were taken to minimize animal suffering

comes in screening therapeutic candidates. These short evaluation during experiments, such as moving or gripping experimental animals with gentle

manipulation, changing litter in cages frequently for animals with adequate food

periods following injury have significantly limited our under-

and water, and so on.

standing due to a lack of information on the delayed effects of

treatment, as well as to a short-lived and reversible neuropro-

2.2. Permanent MCAO surgical procedure

tection, generating so-called false positive results. In addition,

functional deficits associated with impaired sensorimotor ability Permanent focal cerebral ischemia was induced by occluding the middle cere-

bral artery using the previously described intraluminal suture method [20] with

after human strokes often develop over the course of months or

minor modifications. Briefly, rats were anesthetized with 10% chloral hydrate

years [6,7].

diluted in physiological saline (3.5 ml/kg, IP). And a monofilament coated with

Therefore, there is a need to develop a reliable, non-subjective poly-l-lysine (0.1 wt/vol% in deionized water, Sigma, USA), was used to occlude the

means of measuring outcomes that can detect long-term sensor- origin of the middle cerebral artery (MCA). It was inserted into the internal carotid

artery lumen until it met mild resistance, approximately 2.2 cm beyond the common

imotor deficits following middle cerebral artery occlusion (MCAO).

carotid artery bifurcation. The suture was secured with a ligature and maintained in

Although tests for the locomotion-like rope walk [8], grid walk

place until sacrifice. The same surgical procedure was conducted in sham-operated

[9], ladder rung behavior [10] and foot print analysis [11] have

rats in which the MCA was not occluded. During the surgical procedure, the body

◦

unraveled distinct neurological impairments after stroke, their sen- temperature was maintained at 37 C using a homeothermic blanket.

sitivities are too low to distinguish subtle motor impairments. As

2.3. Determination of neurological symptoms

we know, gait impairment often occurs as a result of an ischemic

stroke. Poststroke gait is characterized by temporal asymmetry,

Motor and behavioral changes were assessed one hour after surgery using a five-

reduced walking velocity, and reduced stride length [12]. Impaired

point scale [21] as follows: 0, no neurological deficit (normal); 1, failure to extend

gait function not only reduces ambulation but can also lead to right forepaw (mild); 2, decreased resistance to lateral push (mild to moderate); 3,

imbalance and falls, especially in elderly patients experiencing a circling or walking to the right (moderate); and 4, loss of walking or righting reflex

(severe).

stroke [13]. Analyzing neurological symptoms such as gait impair-

ment may improve evaluation of the neurological outcome, both

2.4. Gait analysis

in humans and animals, and may also aid in the assessment of the

effectiveness of new treatment concepts. To date, several groups Gait assessment was carried out in rats on the 1st, 7th, 21st and 42nd day after

have investigated gait impairments in different stroke models. In pMCAO by the automated computer-assisted method (Xin Hai Hua Yi Instrument

Co., Beijing China). Data were collected and analyzed with Gait Analysis Lab soft-

studies of transient ischemia (tMCAO) with rats or mice [14–17],

ware version 5.0. The equipment was located under natural light in a silent room. In

gait parameters were found to be similarly altered, indicating that

brief, the system consists of an elevated 1.2 m-long glass plate which is illuminated

gait analysis is a feasible method for use with animal stroke mod-

with a fluorescent light coming from the side and the fluorescent light is internally

els. However, permanent ischemia (pMCAO) without reperfusion reflected in the glass allowing the paws to reflect light as they come into contact with

the glass floor. A ceiling on top of the walkway creates a red background to produce

is more similar to naturally occurring cerebral ischemia in humans

the contour of the animal. A high-speed camera (100 frames) underneath the glass

and is thus of greater clinical relevance. One recommendation of the

plate captures the images which are subsequently analyzed by the connected com-

Stroke Therapy Academic Industry Roundtable (STAIR) called for

puter program (Fig. 1A). The video acquisition system is sealed with a PVC sheet to

replicating findings in multiple laboratories using pMCAO models, ensure a uniform dark environment to insure controlled lighting in the experiment.

also because pMCAO models could better simulate typical human Prior to the first testing day, the animals were trained to traverse a glass walkway

toward their home cage. On subsequent training days, three complete runs across

stroke injury without reperfusion on pathophysiology and therapy

the walkway were recorded by a camera positioned below (Fig. 1B). If an animal

of experimental stroke. Moreover, the pMCAO model also pro-

failed to complete a run within 5 s, walked backwards, or reared during the run, the

vides a useful means for testing therapeutic approaches aimed

animal was given an additional re-run. Pixels below a light intensity of 20 units on

at repairing the injured tissue in late stages of cerebral ischemia a 0–255 arbitrary scale were filtered out. Prints can be inspected individually and in

[18,19]. combinations, and timing diagrams for paw placements are available (Fig. 1C and D).

Gait parameters were classified into four main categories as summarized in Table 1.

The goal of the present work was to describe gait changes in

rats after permanent middle cerebral artery occlusion (pMCAO)

2.5. Histology

and to evaluate results of behavioral tests used to detect long-

term gait deficits for up to 42 days after ischemic injury. Several The aim of the histological evaluation was to estimate the total infarct volume.

After completion of the last behavioral test time-point, rats were sacrificed and

parameters that have not been employed in previous studies

transcardially perfused with 4% paraformaldehyde in PBS. Brains were removed

were added to observe their usefulness in assessing gait in cere- ◦

from skulls and immersed in paraformaldehyde at 4–8 C. Each brain was then cut

bral ischemia. Furthermore, correlation analyses were carried

into six coronal 2 mm-thick slices. Histology sections (approximately 5 ␮m thick)

out between gait parameters and the cerebral infarction vol- were obtained for evaluation by light microscopy and scanned into the computer.

ume in order to evaluate associations between gait and cerebral In each section, an outline was traced around the contralateral hemisphere and the

2

ipsilateral hemisphere using Image J (NIH, USA). The infarct area (mm ) was cal-

impairment in ischemic stroke, and to identify ischemic models

culated by two independent and experimentally blind assistants by subtracting the

and behavioral tests which could be reliably used for screening

area of the ipsilateral hemisphere from the contralateral hemisphere. The infarct

in pharmacological studies and in bench to bedside transla-

volume was calculated as the mean cavity area of two adjacent sections multiplied

tion. by the distance between them. The total volume was the sum of the volumes for

176 S. Li et al. / Behavioural Brain Research 250 (2013) 174–191

Fig. 1. Gait analysis system. (A) A photograph of gait analysis system in this work. (B) Example of image of rat in the corridor. (C) Footprints diagram. (D) Timing diagram.

Color coding: bright yellow, right fore (Rf); dark purple, right hind (Rh); bright green, left fore (Lf); dark blue, left hind (Lh). (For interpretation of the references to color in

this figure legend, the reader is referred to the web version of the article.)

each section. The ratio of the cerebral infarction volume is the volume of cere- significant (*p < 0.05; **p < 0.01; ***p < 0.001). Behavioral data were first analyzed

bral infarction in the ipsilateral hemisphere in relation to the total cerebral volume. using repeated measures analysis of variance (ANOVA) with days as the within-

Brain infarct volumes were expressed as the percentage of the volume of the injured subject variable and different treatment groups as the between-subject variable

hemisphere. (sham vs. pMCAO). Mauchley’s test was used to evaluate the sphericity of the within-

subject effects, and when necessary, Greenhouse–Geisser was applied to adjust

2.6. Statistical analysis the degrees of freedom. When significant effects were detected, post hoc multi-

ple pairwise comparisons were made using the LSD comparisons test after ANOVA.

All analyses were performed using SPSS version 16.0 (Chicago, IL, USA). Data are Furthermore, a Pearson correlation coefficient was used to evaluate the relationship

± presented as means S.E.M. A p-value of less than 0.05 was considered statistically between infarction volumes and behavioral outcomes.

Table 1

Terminology and definition of the main gait parameter categories relevant to this study.

Parameter Definition

Based on the body position

Walk speed The speed obtained by summing the total distances that all the feet have moved over the entire video and dividing by the sum

of the stance times for all the feet.

Body rotation average The average orientation measured in degrees with 0 orientation being directly in front of the animal, representing the

capacity of controlling the walking direction in certain degree.

Body rotation standard deviation The standard deviation of the body rotation measuring the amount of sway of the animal in terms of body rotation from its

average orientation, and representing walking stability of the animal to some extent.

Lateral movement average This average lateral movement yields the average position of the center of the mass of the animal along its minor body axis

(waist axis) with respect to the image.

Lateral movement standard deviation The standard deviation of the lateral movement measuring the amount of variability of the center of mass along the minor

body axis (waist axis).

Based on relative positions between paws

Track width The distance between the midpoint of the trajectory of the left foot stance and the midpoint of the trajectory of the right foot

stance.

Print position Distance from the former forepaw position to the consecutive hindpaw position

Stride length The real-world distance the animal has traveled (assuming a stationary tread with the animal walking forward) between two

successive initiations of stances.

Temporal parameters

Stride time The time elapsed between two successive initiations of stances: the sum of the stance time and swing time.

Stance time The time elapsed while the foot is in contact with the floor in its stance phase.

Swing time The time elapsed while the foot is in the air, its swing phase.

Brake time The time elapsed between the start of a stance and the instance the foot reaches the normal stance position.

Propulsion time The time elapsed between the instance the foot reaches the normal stance position and the time when it leaves the floor

surface.

Propulsion index The ratio of stance time and propulsion time

Duty cycle (%) The percentage of stride time spent in the stance phase.

Support time (%) The relative duration of one, two (diagonal, lateral, or girdle pairs), three, or four paws in contact with the floor, expressed as

percentages.

Swing speed The ratio of the stride length to the swing time.

Homologous coupling The fraction of the stride of a reference foot, when the given foot on the same half (front half or rear half) starts its stride.

Homolateral coupling The fraction of the stride of a reference foot, when the given foot on the same side (left side or right side) starts its stride.

Diagonal coupling The fraction of the stride of a reference foot, when the given foot diagonally opposite to the reference foot starts its stride.

Spatial parameters based on individual paws

Print intensity Pressure of paw in contact with the floor.

Average print area The size of the foot print over the entire stance.

Max print area The maximum area of a paw (in pixels) that comes into contact with the glass plate.

S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 177

3. Results p = 0.0049); no significant differences were seen in the standard

deviation of the body rotation (Fig. 2D: effect of surgery F1,15 = 3.14,

3.1. Functional deficits p = 0.0981). Post hoc analysis of the data showed that on the 1st

day after pMCAO, the average body rotation of the pMCAO rats

After recovering from the anesthetic, all occluded rats circled or was not significantly different from that of the sham group, while

veered to the affected side, earning them a score of 3 on the acute the body rotation standard deviation was increased (F1,15 = 5.80,

neurological assessment one hour after pMCAO. Signs of stress p = 0.0304). On the 7th day post-surgery, the average body rota-

were evident for the first few days after the occlusion by the large tion was increased significantly (F1,15 = 22.81, p = 0.0003), but the

amounts of red harderian gland secretions around their eyes and standard deviation showed no obvious changes. Significant changes

nostrils. After 10–14 days, rats no longer drooled or circled, and a in the body rotation standard deviation and in the average body

gross inspection of the animals in their home cages failed to detect rotation appeared uniquely at acute and subacute stages, respec-

any apparent behavioral deficits. tively, with no significant changes noted at other stages of the

disease.

3.2. Gait impairment following pMCAO Propulsion time and propulsion index: In this study, pMCAO

induced a significant increase in the propulsion time of the left

A summary of the gait parameters measured with the gait anal- forepaw (Fig. 3A: effect of surgery F1,15 = 26.68, p = 0.0001), left

ysis system in SD rats subjected to pMCAO is provided in Table 1. hindpaw (Fig. 3B: effect of surgery F1,15 = 41.55, p < 0.0001), right

Table 2 provides an overview of the gait parameters used, in this forepaw (Fig. 3C: effect of surgery F1,15 = 28.76, p = 0.0001), and right

study, as well as changes that occurred from day 1 to day 42 hindpaw (Fig. 3D: effect of surgery F1,15 = 34.93, p < 0.0001). A grad-

after pMCAO. Gait assessment of sensorimotor function in post- ual improvement in the propulsion time deficit was detected in

pMCAO rats revealed long-lasting deficits for several weeks in the all four paws, and Fig. 3 shows the left forepaw (Fig. 3A: effect

behavioral parameters related to each paw. In particular, significant of time F3,45 = 11.84, p = 0.0007), left hindpaw (Fig. 3B: effect of

differences in most of the evaluated gait parameters were obviously time F3,45 = 26.13, p < 0.0001), right forepaw (Fig. 3C: effect of time

found between sham and pMCAO groups during the 1st week. F3,45 = 8.57, p = 0.0001), and right hindpaw (Fig. 3D: effect of time

The parameters showing the most significant differences are F3,45 = 20.74, p < 0.0001). Post hoc analysis of the data showed

described in further detail below, and these included walk speed, significant changes in the propulsion time of all paws at each obser-

stride time, body rotation average, body rotation standard devi- vation point.

ation, duty cycle, double support time, three support time, Animals subjected to pMCAO showed an increased propulsion

propulsion time, propulsion index and homologous coupling index of only Rf during locomotion (Fig. 4C: effect of surgery

(Figs. 2–9). F1,15 = 7.17, p = 0.018). However, post hoc analysis of the data

Walk speed and stride time: In this study, as with the find- showed that the propulsion index of the affected limbs (Rf,

ings reported in stroke people, there was a significant difference in F1,15 = 7.60, p = 0.0155; Rh, F1,15 = 6.32, p = 0.0248; respectively) and

both walk speed (Fig. 2A: effect of surgery F1,15 = 55.23, p < 0.0001) the uninjured hindlimb (F1,15 = 7.54, p = 0.0157) was increased

and stride time (Fig. 2B: effect of surgery F1,15 = 36.32, p < 0.0001) on the 1st day after pMCAO, and that of the uninjured fore-

between sham and pMCAO rats with the latter showing obvious limb was decreased (F1,15 = 10.78, p = 0.0054), which might be

deficits in gait parameters. With the recovery time prolonged, due to deficits in swing initiation and forward propulsion in

the walk speed continued to increase (Fig. 2A: effect of time walking due to weakness of the affected limbs, a hyperfunc-

F3,45 = 13.65, p < 0.0001), and the stride time was continually short- tion of swing initiation and a compensating forward propulsion

ened (Fig. 2B: effect of time F3,45 = 11.99, p < 0.0001) indicating that of the uninjured forelimb. On the 7th day after pMCAO, the

the functional walking deficits of pMCAO rats had continued to propulsion index of the two forelimbs (Lf, F1,15 = 5.07, p = 0.0409;

improve. Rf, F1,15 = 15.09, p = 0.0016; respectively) and affected hindlimb

Furthermore, post hoc analysis of the data showed that walk (F1,15 = 6.22, p = 0.0258) was significantly increased, but that of

speed was decreased, while the stride time was increased, on the the uninjured hindlimb had a tendency to decrease. On the 21st

1st (F1,15 = 7.8161.06, p < 0.0001; F1,15 = 43.64, p < 0.0001; respec- and 42nd day post-surgery, the propulsion index of the forelimbs

tively), 7th (F1,15 = 13.79, p = 0.0023; F1,15 = 18.20, p = 0.0008; showed an upward tendency, while that of the hindlimbs showed

respectively), 21st (F1,15 = 61.12, p < 0.0001; F1,15 = 26.00, a downward tendency. Especially noteworthy, on the 42nd day

p = 0.0002; respectively), and 42nd day (F1,15 = 13.60, p = 0.0024; after pMCAO, the propulsion index of the uninjured hindlimb was

F1,15 = 16.76, p = 0.0011; respectively) after pMCAO. In contrast significantly reduced (F1,15 = 5.52, p = 0.034). This showed that the

with the sham group, sustained impairments were detected in compensatory effect of the uninjured forelimb disappeared on the

the walk speed and the stride time for the duration of the pMCAO 7th day, and that the effect of the hindlimb subsequently increased

study, and these findings were statistically significant up to six over time.

weeks. Duty cycle (%): Animals subjected to pMCAO displayed an

Body rotation average and standard deviation: The average increase in the duty cycle (%) of each paw, and Fig. 5 shows the duty

body rotation is the average orientation measured in degrees with cycles (%) of Lf (Fig. 5A: effect of surgery F1,15 = 55.57, p < 0.0001), Lh

0 orientation being the direction immediately in front of the ani- (Fig. 5B: effect of surgery F1,15 = 16.8, p = 0.0011), Rf (Fig. 5C: effect

mal. It essentially represents the overall orientation of the animal. of surgery F1,15 = 27.82, p = 0.0001), and Rh (Fig. 5D: effect of surgery

If an animal has an average body rotation of +5 degrees, it has a ten- F1,15 = 7.52, p = 0.0159). Stance time was increased at each observa-

dency to walk with the body angled 5 degrees clockwise on average. tion point in the model group (data not shown). Post hoc analysis

Standard deviation of the body rotation is the amount of sway of showed that after pMCAO, the duty cycle (%) of each limb was

the animal in terms of the rotation of the body from its average increased significantly at the early stages of pMCAO, on day 1 (Lf,

orientation. In the example above, if the standard deviation was F1,15 = 21.98, p = 0.0003; Lh, F1,15 = 13.31, p = 0.0026; Rf, F1,15 = 9.45,

3, then the animal had a tendency to sway, for example, from +8 p = 0.0083; Rh, F1,15 = 13.76, p = 0.0023; respectively) and day 7 (Lf,

degrees to +2 degrees. In this study, a larger body rotation average F1,15 = 13.07, p = 0.0028; Lh, F1,15 = 12.71, p = 0.0031; Rf, F1,15 = 9.74,

was noted in pMCAO animals during locomotion (Fig. 2C: effect p = 0.0075; Rh, F1,15 = 4.91, p = 0.0437; respectively). In the mid-

of surgery F1,15 = 10.51, p = 0.0059) as was a gradual improvement term of the disease, on the 21st day post surgery, the duty cycle

in this average during recovery (Fig. 2C: effect of time F3,45 = 4.96, (%) was increased in all limbs except the affected hindlimb (Lf,

178 S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 * * * * * ** * ** ** ** ** ** ** ** ** ** * ** ** ** ** *** *** *** *** *** *** *** p 0.0023 0.0003 0.3162 0.4298 0.1703 0.0141 0.9842 0.7326 0.0006 0.0490 0.0919 0.0026 0.0022 0.0009 0.0029 0.0014 0.0003 0.0027 0.0001 0.0888 0.0605 0.0014 0.1329 0.0018 0.0016 0.0017 0.0003 0.0409 0.8665 0.0016 0.0258 0.0028 0.0031 0.0075 0.0437 0.0136 0.0584

127.786 20.382 0.0215 13.203 12.402 19.628

7.392 0.3490

4.526 3.369

± ± ± ± ±

2.161 0.470 2.457 0.701 0.084 0.093 0.096 0.041 0.027 0.025 0.019 0.069 0.062 0.073 0.066 0.091 0.102 0.028 0.052 0.034 0.035 0.065 0.055 0.096 0.103 3.427 0.111 0.0008 0.108 0.027 0.045 0.024 0.049

±

± ±

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

3.762 − pMCAO 397.665 3.666 1.970 1.761 22.757 24.422 4.114 5.050 146.602 150.980 0.253 0.296 0.292 0.306 0.189 0.152 0.163 0.148 0.080 0.080 0.106 0.078 0.173 0.216 0.186 0.227 0.665 0.738 0.651 0.756 0.574 0.656 0.634 0.662 0.248 0.251

21.337 16.307 132.184 15.055 160.943 35.589 163.184

6.710 1.310

4.802 2.192

± ± ± ± ±

0.473 0.695 2.681 2.884 0.0220.032 0.450 0.028 0.031 0.017 0.014 0.026 0.023 0.055 0.046 0.064 0.028 0.030 0.042 0.032 0.063 0.024 0.007 0.007 0.017 0.012 0.015 0.012 0.013 0.077 0.035 0.018

±

±

± ±

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.603 0.342 7d Sham 639.028 − 1.659 − 1.519 19.385 20.438 4.087 4.514 192.302 165.412 200.713 134.657 0.279 0.140 0.166 0.144 0.162 0.127 0.106 0.126 0.108 0.061 0.043 0.060 0.052 0.079 0.122 0.084 0.111 0.570 0.745 0.577 0.683 0.518 0.598 0.549 0.612 0.350 0.339 *** *** *** *** *** *** *** ***

* * * * * * * * ** ** ** ** ** ** ** *** *** ***

p <0.0001 0.5897 0.0304 0.2412 0.0127 0.0782 0.8239 0.7120 0.0569 0.121 <0.0001 <0.0001 <0.0001 <0.0001 0.0369 0.2788 0.1957 0.0702 <0.0001 0.0834 0.0257 0.3549 0.0001 <0.0001 0.0003 <0.0001 0.0054 0.0157 0.0155 0.0248 0.0003 0.0026 0.0083 0.0045 0.0014

49.457 20.92816.113 0.0110 8.232 16.231 0.0016

7.976 2.028

± ± ± ± ±

0.741 2.053 2.163 0.664 0.0540.045 <0.0001 0.056 0.018 0.056 0.015 0.022 0.026 0.027 0.050 0.062 0.034 0.045 0.081 0.056 0.045 0.067 0.0640.086 0.0023 1.248 2.629 0.3632 0.059 0.040 0.013 0.016 0.012 0.058 0.128 0.109

± ±

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

pMCAO 258.998 0.033 2.324 2.508 19.058 31.575 4.727 130.831 135.338 0.315 0.360 0.310 0.349 0.154 0.110 0.165 0.124 0.158 0.054 0.118 0.049 0.157 0.306 0.299 0.499 0.846 0.649 0.852 0.669 0.762 0.709 0.156 0.140

18.047 142.886 10.342 151.072 25.050 17.068 75.817

1.612

1.553 7.709

± ± ± ± ±

0.373 2.449 2.286 5.914 0.724 0.043 0.475 0.034 0.008 0.013 0.013 0.028 0.035 0.027 0.078 0.037 0.045 0.020 0.032 0.0340.068 0.741 3.944 2.080 0.039 0.031 0.034 0.013 0.025 0.007 0.027 0.013 0.016 0.01520.023 0.184 0.043

±

± ±

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.365 1d Sham 509.076 − 1.618 0.505 2.083 16.481 24.124 4.983 5.495 171.496 150.171 177.629 149.037 0.314 0.180 0.210 0.181 0.203 0.138 0.103 0.137 0.111 0.081 0.043 0.085 0.043 0.099 0.167 0.096 0.160 0.556 0.796 0.539 0.793 0.563 0.671 0.565 0.646 0.288 0.304

p 0.5913 0.4656 0.1107 0.2299 0.1848 0.4401 0.3503 0.2374 0.9998 0.9845 0.9125 0.9459 0.4282 0.5881 0.7646 0.7383 0.2102 0.5889 0.7175 0.6148 0.4692 0.4591 0.1988 0.2041 0.1703 0.3244 0.8570 0.4316 0.8009 0.7224

116.342 23.909 29.480 0.8012 28.364 0.3542 27.469

8.106 1.821

± ± ± ± ±

0.181 1.9860.434 0.1786 1.568 2.354 0.2857 1.127 0.4220 0.0430.033 0.040 0.9532 0.037 0.039 0.014 0.011 0.036 0.038 0.048 0.057 0.061 0.070 0.038 0.048 0.0480.052 0.069 0.7316 0.0070.011 0.9603 0.020 0.0109 0.026 0.0090.016 0.0643 0.013 0.5410 0.054

± ±

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

Model 638.885 1.415 1.792 13.800 27.032 4.619 159.919 160.020 0.147 0.167 0.147 0.168 0.134 0.131 0.106 0.069 0.039 0.066 0.078 0.128 0.127 0.535 0.763 0.564 0.748 0.516 0.607 0.519 0.355 0.352

32.60531.072 26.148 200.995 196.773 125.533 29.627

1.340 0.219

5.812 4.724

± ± ± ± ±

0.313 0.341 2.100 2.366 4.146 4.138 2.852 0.0350.029 0.033 0.275 0.030 0.031 0.013 0.013 0.024 0.028 0.067 0.068 0.071 0.045 0.048 0.044 0.071 0.072 0.0080.011 0.105 0.014 0.015 0.019 0.0080.016 0.041 0.0150.030 0.081 0.036 0.612

±

± ±

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.294 1d gait.

− − Sham 605.633 1.511 1.459 16.503 22.405 5.355 5.456 186.353 146.525 193.197 142.390 0.147 0.167 0.148 0.169 0.128 0.105 0.128 0.108 0.066 0.047 0.072 0.049 0.081 0.121 0.076 0.115 0.550 0.721 0.521 0.697 0.537 0.611 0.538 0.605 0.347 0.365 of

recovery

deviation

deviation

functional (%)

averagestandard 1.130 on

time

average standard

time index

(%)

statistics

support

movement movement

time

time

length time 0.274 time

width

speed

rotation rotation

position cycle

2

Lf Lf Lf Lf Lf Lf Lf Lh Rf Rh Lh Rf Rh Lh Rf Rh Rh Lh Rf Rh Lh Rh Lh Rf Rh Lh-Rf Left Lf-Rh Forepaw Hindpaw Right Lh Rf Rf Walk Body Body Print Stance Swing Brake Propulsion Duty Double Lateral Lateral Track Stride Stride Propulsion Table Parameter

S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 179 * * ** ** ** ** ** ** *** *** ** *** *** * * * * ** ** *** *** p 0.0024 0.0609 0.6500 0.9195 0.7563 0.9987 0.1395 0.2236 0.8627 0.6654 0.9017 0.1994 0.9064 0.0011 0.0007 0.0024 0.0006 0.0047 0.0080 0.0003 0.0507 0.0011 0.0091 0.0113 0.0219 0.1592 p 0.0072 0.0252 0.0482 0.0534 0.7659 <0.0001 0.0004 0.0007 0.0362 0.3699 0.0898 0.0068 0.0146

99.001 100.513 136.243 133.387

± ± ± ±

0.036 0.031 0.067 0.029 0.061 0.064 0.038 0.034 0.055

± ± ± ± ± ± ± ± ±

22.183 16.735 110.851 28.822 21.027

4.568

5.711 7.634

± ± ± ± ±

0.577 2.620 0.695 0.467 3.107 0.063 0.047 0.042 0.045 0.028 0.030 0.025 0.037 0.015 0.011 0.040 0.022 0.023

±

pMCAO 0.114 0.129 0.090 0.177 0.037 377.796 346.864 389.002 360.820 0.586 0.661 0.629 0.661

± ±

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

3.211

pMCAO 457.560 1.310 1.446 − 1.480 18.600 18.417 4.520 5.899 199.758 175.173 196.024 150.582 0.366 0.207 0.236 0.214 0.227 0.161 0.140 0.151 0.139 0.083 0.081 0.090 0.061

64.141 108.355 78.295 177.332

± ± ± ±

0.032 0.037 0.039 0.015 0.021 0.033 0.022 0.032 0.033

± ± ± ± ± ± ± ± ±

7d Sham 0.059 0.095 0.041 0.121 0.032 695.247 591.276 730.979 487.525 0.571 0.615 0.556 0.619

23.353 25.919 30.351 114.205 23.049

9.735

7.199 5.459

± ± ± ± ±

0.324 0.366 3.385 2.600 1.461 0.029 0.020 0.030 0.029 0.016 0.018 0.023 0.013 0.013 0.006 0.018 0.018 0.009

±

± ±

*** *** ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

** ** ** ** *** *** *** *** ***

3.602 42d Sham 665.078 0.144 1.338 − 1.413 18.594 22.793 6.589 6.129 204.755 176.451 214.484 152.145 0.265 0.143 0.162 0.140 0.163 0.125 0.106 0.125 0.100 0.065 0.039 0.063 0.046

p 0.0003 0.0016 0.0078 0.0002 0.7094 <0.0001 0.0002 <0.0001 0.0004 0.8227 0.0047 0.0005 0.0093

51.949 68.043 60.098 65.234

± ± ± ±

0.045 0.029 0.038 0.053 0.150 0.063 0.043 0.078 0.032 *** *** *** ***

* * * * ** *** *** *** ** ** ± ± ± ± ± ± ± ± ±

pMCAO 0.185 0.199 0.121 0.272 0.034 308.670 283.379 347.721 293.793 0.564 0.724 0.664 0.752 p <0.0001 0.5362 0.1383 0.6298 0.2178 0.3722 0.1797 0.3446 0.8084 0.0178 0.4778 0.7805 0.3193 0.0002 0.0001 0.0002 <0.0001 0.0010 0.0134 0.0021 0.0865 <0.0001 <0.0001 0.0152 0.0010 0.0129

88.646 101.771 102.890 88.358

± ± ± ±

0.024 0.036 0.057 0.032 0.046 0.041 0.061 0.030 0.027

± ± ± ± ± ± ± ± ±

65.323 11.497 30.003 11.309 28.001

5.931

2.743 5.936

± ± ± ± ±

0.640 1.668 3.568 0.438 0.040 0.041 0.052 0.025 0.026 0.016 0.785 0.056 0.047 0.018 0.018 0.036 0.026 0.039

±

1d Sham 0.075 0.147 0.070 0.155 0.027 554.432 484.890 574.810 485.439 0.577 0.620 0.580 0.662

± ±

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.741

pMCAO 383.939 0.780 2.032 − 1.664 18.516 24.334 3.531 4.687 170.248 141.674 180.492 136.951 0.390 0.232 0.260 0.237 0.255 0.159 0.140 0.154 0.147 0.102 0.091 0.099 0.088

p 0.1183 0.7514 0.6904 0.7906 0.8318 0.7074 0.5204 0.9305 0.2786 0.6674 0.4818 0.9132 0.2659

136.882 122.292 145.513 206.810

81.172 35.440 30.384 25.469 26.234

4.269 ± ± ± ±

9.142 6.263

0.039 0.037 0.050 0.063 0.036 0.030 0.032 0.038 0.032

± ± ± ± ±

0.438 2.063 0.459 2.675 0.035 0.0256 0.031 0.025 0.037 0.014 1.749 0.016 0.015 0.008 0.008 0.008 0.013 0.023

±

± ± ± ± ± ± ± ± ± ± ±

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

2.014 Model 0.025 0.112 0.056 0.098 0.023 721.324 603.419 727.101 596.925 0.534 0.563 0.549 0.558 21d Sham 671.919 0.350 1.688 − 1.305 20.745 29.776 4.448 5.077 196.706 151.282 184.642 151.500 0.271 0.141 0.159 0.137 0.163 0.130 0.109 0.134 0.107 0.062 0.050 0.059 0.047

160.491 173.767 123.670 153.274

± ± ± ±

0.061 0.038 0.047 0.053 0.068 0.048 0.045 0.038 0.032

± ± ± ± ± ± ± ± ±

1d − Sham 0.069 0.105 0.049 0.104 0.026 692.755 550.592 719.020 527.607 0.545 0.547 0.545 0.585 deviation

deviation

(%)

average standard (%)

) time average standard

time coupling

movement movement time Lh

Rh speed time continued length time Lf support time Rf

width

speed (

rotation rotation

position

support

→

→

→ 2

→

Lf Lf Lf Lf Lf Lh-Lf-Rh Rf-Lf-Rh Rh-Lh-Rf Lh Rf Rh Lh Lh Rf Rh Lh Rf Lh Rf Rh Lf-Lh-Rf Left Forepaw Rf Hindpaw Right Lh Rf Rh Rh Rh Lf Three Homologous Walk Body Stance Swing Brake Body Lateral Lateral Track Print Stride Stride Four Swing Table

180 S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 * * * ** * * ** *** *** ** p 0.1093 0.0229 0.0998 0.0340 0.3502 0.3269 0.1192 0.0126 0.9707 0.0677 0.1647 0.0127 0.5437 0.056 0.2089 0.8871 0.0006 0.0302 0.0033 0.0010 0.455 0.7535 0.0525 0.4541

108.367 138.469 45.759 133.615

± ± ± ±

0.0310.044 0.0250.043 0.054 0.129 0.063 0.088 0.0014 0.035 0.0002 0.0370.036 0.033 0.057 0.4561 0.045 0.051 0.041 0.038 0.058 0.042 0.029 0.036 0.043 0.028

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

pMCAO 0.154 0.166 0.592 0.643 0.584 0.727 0.565 0.588 0.615 0.302 0.312 0.086 0.101 0.076 0.130 0.014 559.357 497.019 552.673 415.566 0.563 0.613 0.588 0.615

112.210

85.220 141.639 137.822

±

± ± ±

0.027 0.033 0.053 0.040 0.064 0.038 0.042 0.0510.044 0.070 0.615 0.043 0.066 0.037 0.063 0.0120.009 0.124 0.124 0.072 0.041 0.059 0.032 0.067 0.017

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

42d Sham 0.077 0.123 0.077 0.117 0.545 0.756 0.553 0.761 0.533 0.598 0.531 0.614v0.040 0.366 0.360 0.027 0.087 0.041 0.093 0.016 774.391 662.996 797.097 592.8467 0.546 0.623 0.547 0.595 ***

* * * * * * ** ** * ** ** ** * ** ** *** ***

p 0.0005 0.0193 0.4457 0.2733 0.5642 0.3265 0.0026 <0.0001 0.3601 0.0271 0.0209 0.0039 0.331 0.0144 0.0176 0.5465 0.0013 0.0193 0.3059 0.0031 0.0270 0.0017 0.0048

66.081 123.910 128.809 116.345

± ± ± ±

0.030 0.0300.044 0.045 0.0998 0.118 0.0004 0.0520.067 0.093 0.074 0.043 0.0285 0.055 0.056 0.0660.038 0.6334 0.032 0.0024 0.040 0.037 0.045 0.044 0.032 0.037

0.085 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

pMCAO 0.169 0.167 0.576 0.658 0.657 0.590 0.605 0.625 0.278 0.259 0.102 0.094 0.159 0.028 464.006 361.552 435.583 381.515 0.597 0.606 0.665 hindpaw.

left

Lh:

and

100.937 115.760 93.743 181.135

0.023 0.139

± ± ± ±

0.033 0.059 0.089 0.033 0.0320.053 0.070 0.060 0.643 0.026 0.0770.040 0.047 0.632 0.021 0.0730.071 0.025 0.583 0.039 0.0450.034 0.036 0.026 0.116 0.022 0.130

±

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

hindpaw;

21d Sham 0.080 0.110 0.0783 0.116 0.555 0.711 0.560 0.706 0.522 0.591 0.506 0.596 0.379 0.347 0.029 0.093 0.039 0.097 0.019 684.510 490.945 720.579 562.583 0.543 0.531 0.594 right

Rh:

forepaw;

(%)

(%) left

(%)

Lf: ) time

time

time coupling

time index (%)

support

Lh 0.615 forepaw; Rh speed continued Lf

support Rf

(

cycle

support 0.05. 0.01. 0.001.

→

→ →

2

→ < < <

Lf Lf Lf Lf Lf-Lh-Rf Lh Rh Rf Rh Lh-Lf-Rh Lh Rf Rh Lf-Rh Rf Rf Lh Rh Lh Lh-Rf Rf-Lf-Rh Rh-Lh-Rf Rh Lf Lh Rf p p p right

* Propulsion Propulsion Homologous Duty Double Three Four Swing ** *** Table Rf:

S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 181

Fig. 2. Effect of pMCAO on SD rats. (A) Deficits in walk speed persisted for 42 days. (B) Stride time increased after pMCAO. (C) Deficits were also observed in the body rotation

average. (D) Body rotation standard deviation was observed up to 42 days. (Error bar: S.E.M. *p < 0.05, **p < 0.01, ***p < 0.001 LSD post hoc. The X-axis indicates days.) pMCAO

n = 8, sham n = 8.

F1,15 = 13.29, p = 0.0026; Lh, F1,15 = 5.96, p = 0.0285; Rf, F1,15 = 39.95, asymmetry F1,15 = 8.08, p = 0.013), and Rh-Lh-Rf three support time

p < 0.0001; respectively). At the later stage, 42 days after pMCAO, (%) was increased as seen on post hoc analysis of the 1st day

the duty cycle (%) of almost all the paws had recovered to a normal (F1,15 = 11.71, p = 0.0041) results. A significant difference was also

level. seen between Rf-Lf-Lh and Rf-Lf-Rh (Fig. 7D: effect of asymme-

Double and three support time (%): In this gait test, three try F1,15 = 4.65, p = 0.049). Post hoc analysis of the data showed

support times (%) were increased after pMCAO as follows: Lf- that Rf-Lf-Lh three support time (%) was increased on the 1st

Lh-Rf (Fig. 6A: effect of surgery F1,15 = 44.05, p < 0.0001), Lh-Lf-Rh day (F1,15 = 9.55, p = 0.008) post-surgery. This demonstrated severe

(Fig. 6B: effect of surgery F1,15 = 5.86, p = 0.0296), Rf-Lf-Rh (Fig. 6C: deficits in the walking symmetry of the pMCAO rats, especially at

effect of surgery F1,15 = 15.01, p = 0.0017), and Rh-Lh-Rf (Fig. 6D: the acute stage of the disease, probably leading to changes in the

F1,15 = 16.35, p = 0.0012). All three support times (%) were signifi- walking direction during this phase.

cantly improved over time post-surgery: Lf-Lh-Rf (Fig. 6A: effect Animals subjected to pMCAO showed a shorter double support

of time F3,45 = 10.79, p < 0.0001), Lh-Lf-Rh (Fig. 6B: effect of time time (%) during locomotion, and Fig. 8 shows that of Lf-Rh (Fig. 8A:

F3,45 = 20.03, p < 0.0001), Rf-Lf-Rh (Fig. 6C: effect of time F3,45 = 3.82, effect of surgery F1,15 = 15.5, p = 0.0015) and Lh-Rf (Fig. 8B: effect

p = 0.0166), and Rh-Lh-Rf (Fig. 6D: effect of time F3,45 = 13.71, of surgery F1,15 = 20.57, p = 0.0005). A gradual improvement in the

p = 0.0003). And importantly, it was interesting to note that there double support time (%) was seen over time, and Fig. 8 shows that

was no significant difference between Rh-Lh-Lf and Rh-Lh-Rf three of Lf-Rh (Fig. 8A: effect of time F3,45 = 10.81, p < 0.0001) and Lh-Rf

support times (%), as was also seen with those of Rf-Lf-Lh and Rf- (Fig. 8B: effect of time F3,45 = 5.03, p = 0.0175). There was a ten-

Lf-Rh in the sham group, due to a symmetric distribution of gravity dency toward a longer interval on four paws in the pMCAO group,

on both sides of the body when normal rats engage in walking compared with the sham group, with no statistical significance

(Table 3). However, in the pMCAO group, a significant difference (data not shown). Thus, the decreased double support time (%),

was observed between Rh-Lh-Lf and Rh-Lh-Rf (Fig. 7B: effect of and the increased time on three and four paws showed that the

182 S. Li et al. / Behavioural Brain Research 250 (2013) 174–191

Fig. 3. pMCAO caused an increase in propulsion time in SD rats. (A) Left forelimb, Lf. (B) Left hindlimb, Lh. (C) Right forelimb, Rf. (D) Right hindlimb, Rh. (Error bar: S.E.M.

*p < 0.05, **p < 0.01, ***p < 0.001 LSD post hoc. The X-axis indicates days.) pMCAO n = 8, sham n = 8.

walking stability of the pMCAO rats had decreased and that three We found that pMCAO led to delayed placement of the right

and four paws were needed to improve the walking stability. How- forepaw during the left forepaw stance (Fig. 9A: effect of surgery

ever, changes in the double and three support times (%) gradually F1,15 = 34.66, p = 0.0001). When the left hindpaw was instance, there

became smaller in pMCAO rats over time, which suggested that was a significant difference between sham and pMCAO groups dur-

walking stability had significantly improved. ing right hindpaw placement (Fig. 9B: effect of surgery F1,15 = 20.25,

Homologous coupling: In this study, homologous coupling is a p = 0.0005). Post hoc analysis showed that the left forepaw anchor

coordination-related parameter that illustrates the temporal rela- with the right forepaw target had a deficit on the 1st (F1,15 = 20.26,

tionship between the placement of two forepaws or hindpaws. p = 0.0005), 7th (F1,15 = 10.03, p = 0.0068), and 21st day (F1,15 = 15.05,

Table 3

Comparison of three support time (%).

Effect

Group Asymmetry Recovery time Interaction

F value p value F value p value F value p value

***

Rh-Lh-Rf vs Rh-Lh-Lf Sham 0.75 0.4013 7.57 0.0004 0.26 0.8529

*

Rf-Lf-Rh vs Rf-Lf-Lh Sham 0 0.9865 3.94 0.0146 0.66 0.5842

* ***

Rh-Lh-Rf vs Rh-Lh-Lf Model 8.08 0.013 34.64 <.0001 1.09 0.3637

* ***

Rf-Lf-Rh vs Rf-Lf-Lh Model 4.65 0.049 13.23 <.0001 2.36 0.0854

Rf: right forepaw; Lf: left forepaw; Rh: right hindpaw; and Lh: left hindpaw.

*

p < 0.05.

***

p < 0.001.

S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 183

Fig. 4. Effect of pMCAO on the propulsion index in SD rats. (A) Left forelimb, Lf. (B) Left hindlimb, Lh. (C) Right forelimb, Rf. (D) Right hindlimb, Rh. (Error bar: S.E.M. *p < 0.05,

**p < 0.01 LSD post hoc. The X-axis indicates days.) pMCAO n = 8, sham n = 8.

p = 0.0017). In addition, the left hindpaw anchor with the right Lh, r = 0.46; Rf, r = 0.76; Rh, r = 0.69; respectively) were significantly

hindpaw target also showed a deficit on the 1st (F1,15 = 9.07, correlated with walk speed in the sham group, while in the

p = 0.0093), 7th (F1,15 = 7.76, p = 0.0146), and 21st day (F1,15 = 11.18, pMCAO rats, only Lh propulsion time (r = −0.91), Rh propulsion

p = 0.0048). These results suggested that pMCAO could have led to time (r = −0.72) and Lh stride length (r = 0.83) were correlated with

the obvious deficits in homologous coupling, which recovered at walk speed. Swing time only showed a tendency to decrease when

later stages. the walk speed was increased. There were no correlations between

walk speed and other temporal or spatial parameters such as track

3.3. Correlations between the walk speed and the other gait width, relative paw placement, and body rotation average (data not

parameters shown).

In sham-operated rats, correlations (r > 0.5) with walk speed

Velocity is a very important determinant that can potentially were significant with respect to both hindpaws and forepaws with

influence several parameters of gait function [22]. However, the certain temporal or spatial parameters. However in pMCAO rats,

relationship between walk speed and other gait parameters in the significant correlations (r > 0.5) were found with only a few param-

pMCAO animal model has not previously been reported, there- eters and these did not pertain to all paws, which might have been

fore we conducted Pearson’s correlation analyses during the acute due to asymmetric changes affecting the limbs in movement among

phase (1st day after pMCAO) according to gait studies of cerebral these animals. Furthermore, there were no correlations between

ischemia rats [16] (Table 4). Consistent with a previous finding walk speed and many other temporal or spatial parameters such

[23], results of this study showed that stride time and swing speed as relative paw placement and the body rotation average. This sug-

were interdependent with walk speed in both sham and pMCAO gests that the changes in gait parameters mentioned above may

rats. Propulsion time (Lf, r = −0.95; Lh, r = −0.97; Rf, r = −0.85; Rh, not necessarily be due to a lower walk speed in the pMCAO rats,

r = −0.94; respectively), brake time (Lf, r = −0.97; Lh, r = −0.76; Rf, although we did conjecture that a slower speed might be due to

r = −0.86; Rh, r = −0.93; respectively), and stride length (Lf, r = 0.5; the combined effect of these changed gait parameters.

184 S. Li et al. / Behavioural Brain Research 250 (2013) 174–191

Fig. 5. Functional deficits in the duty cycle. (A) Left forelimb, Lf. (B) Left hindlimb, Lh. (C) Right forelimb, Rf. (D) Right hindlimb, Rh. (Error bar: S.E.M. *p < 0.05, **p < 0.01,

***p < 0.001 LSD post hoc. The X-axis indicates days.) pMCAO n = 8, sham n = 8.

3.4. Correlations between the volume of cerebral infarction and time was correlated with the final cerebral infarction volume. The

gait parameters results are presented in Table 5.

Measurement of the cerebral infarction volume using sequen-

tial coronal sections revealed no infarcts in any of the sham-treated 4. Discussion

rats but significant cerebral infarction in pMCAO-treated rats

±

(0.264 0.084%) (Fig. 9C). 4.1. Gait impairment in pMCAO rats

Statistical analyses were carried out on a number of obviously

correlated parameters at some observation point. There was no Most studies of neuroprotection following MCAO have

significant correlation between any of the gait parameters on the restricted their assessments to simple reflex and sensory-motor

1st day after pMCAO and the cerebral infarction volume on the function during the early phase of the infarction, or at best, to

42nd day post surgery. On the 7th day after pMCAO, 16 parameters behavioral tests within about one month of the infarct. This is

correlated with the cerebral infarction volume and these were as because rats often show quick recovery times following MCAO and

follows: stride length, swing speed, stance time, duty cycle (%), dou- long-term deficits are hard to detect. However, in human stroke,

ble support time (%), three support time (%), homologous coupling, functional deficits associated with impaired sensorimotor ability

propulsion time, propulsion index, and brake time. On the 21st day often develop over the course of months or years. Gait analysis in

after pMCAO, 13 parameters correlated (r > 0.5) with the final cere- this study demonstrated significant long-lasting deficits for sev-

bral infarction volume and these were as follows: stride time, walk eral weeks and recovery was slow in several areas including walk

speed, homologous coupling, brake time, propulsion time, stance speed, stride time, duty cycle (%), support time (%), propulsion time,

time, and duty cycle (%). On the 42nd day after pMCAO, only swing and interlimb coordination of girdle pairs. Changes in these gait

S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 185

Fig. 6. Deficits in three support time (%) were observed in SD animals subjected to pMCAO. (A) Lf-Lh-Rf three support time (%) impaired post-op. (B) A deficit was also

observed in Lh-Lf-Rh three support time (%). (C) Deficits in Rf-Lf-Rh three support time (%). (D) Increased Rh-Lh-Rf three support time (%) during locomotion. (Error bar:

S.E.M. *p < 0.05, **p < 0.01, ***p < 0.001 LSD post hoc. The X-axis indicates days.) pMCAO n = 8, sham n = 8.

parameters make them suitable for behavioral evaluation of cere- body axis and the horizon, the standard deviation of the angle, the

bral ischemia in animals. In addition, some gait indices changed average distance along the waist axis, and the standard deviation

significantly at early stages after pMCAO but recovered rapidly, of the distance traversed were all observed in the walking process,

making these parameters suitable for evaluating stroke models and thereby insuring that the changing posture in pMCAO rats was

the degree of efficacy of treatment during the acute phase. These assessed in an objective and accurate manner. In the acute phase

parameters included the body rotation average, three support time of cerebral ischemia, the ability to control direction in the pMCAO

(%), and the propulsion index. rats was not completely lost until the decompensation stage, but

A previous report has suggested that body weight might affect walking stability was severely impaired. The stable formation

the intensity of paw contact, the contact area, and the print length of the lesion had obviously affected the directional control of

[24]. In this study, sham-operated rats were significantly heavier walking, but obvious changes of walking stability had disappeared.

than pMCAO rats (F1,15 = 17.74, p = 0.0009). Therefore, parameters Significant changes in the body rotation standard deviation and

directly related to the body weight were left out (e.g. the intensity the average body rotation appeared uniquely at acute and suba-

of paw contact and paw size). cute stages, respectively. This showed that these two parameters

In previous assessments using pMCAO models, many may be useful for sensitive evaluation of the degree of cerebral

researchers found that rats circled or inclined to the affected ischemia at acute and subacute stages. In addition, many gait

side, and this phenomenon was often used for evaluating the parameters were obviously correlated with the walk speed in

degree of cerebral ischemia after MCAO [25–29]. However, the sham or pMCAO rats, while the body rotation standard deviation

scoring of the behavior depended on the subjective judgment of the and average body rotation showed no significant correlation with

observer, and the phenomenon itself disappeared quickly. Hence, this gait feature at any time point in either group. This suggests

the objectivity, accuracy and sensitivity of the scoring method that the two parameters may be used to assess the walking process

were unreliable. In this study, the average angle between the without having to take walk speed into consideration. However,

186 S. Li et al. / Behavioural Brain Research 250 (2013) 174–191

Fig. 7. Three support time (%) observed in sham and model groups. (A) No difference in Rh-Lh-Lf and Rh-Lh-Rf three support time (%) in the sham group. (B) Effect of pMCAO

on Rh-Lh-Lf and Rh-Lh-Rf three support time (%). (C) Rf-Lf-Lh and Rf-Lf-Rh three support time (%) in rats of the sham group. (D) Differences showed between Rf-Lf-Lh and

Rf-Lf-Rh three support time (%) after surgery. (Error bar: S.E.M. **p < 0.01 LSD post hoc. The X-axis indicates days.) pMCAO n = 8, sham n = 8.

these two parameters were not employed in a previous study of was mainly due to an increase in the propulsion phase duration

animal gait analysis, which may indicate that their use could be of about 10% that progressed over the disease process, and further

complementary in automated gait analysis evaluating systems. demonstrated for the first time that these abnormalities arise in the

The strength of the extensor muscles of the affected knee is one acceleration. Krouchev [35] showed that the impairments observed

of the most important factors in walking on humans [30,31]. In in the locomotor performance of SOD1 mice point to the contribu-

hemiplegic patients, weakened surae triceps muscle, an unstable tion of muscles involved at the end of the stance phase and the

knee joint, poor ankle joint movement, a muscle spasm of the lower beginning of the swing phase, such as hamstrings, gastrocnemius

extremity and imbalance all eventually lead to walking abnormal- and flexor digitorum longus (FDL) muscles. In comparison, post-

ities and a decline in the ability to push away from the glass plate, stroke hemiparetic patients showed deficits in swing initiation and

which in gait analysis hemiplegic people should walk on for data forward propulsion [36]. We have shown that the propulsion times

acquisition [32]. Propulsion time and the propulsion index are used of all limbs in pMCAO rats were significantly increased almost

to describe the time spent in lifting a paw from the glass plate, a throughout the whole study. However, the propulsion time and

process in which the area of footprint gradually becomes smaller. walk speed were correlated in all paws of the sham group and

These features could reflect the gait parameters of swing initia- the hindpaws of pMCAO rats. The significantly increased propul-

tion and forward propulsion in moving the body, which could also sion index had no correlation with walk speed, suggesting that an

reflect to some extent the force condition of the limbs while walk- increased propulsion index might be related to the reduced mus-

ing. In a treadmill gait analysis using 8-week-old mice, Wooley et al. cle strength in the pMCAO animals. Deficits in swing initiation and

[33] noticed an increase in the stance duration of SOD1 transgenic forward propulsion in walking might be due to weakness of the

mice compared to wild type mice. A few years later, Mancuso [34] affected limbs, and a hyperfunction of swing initiation might be due

reported an early increase in the hindlimb stance duration that to a compensating forward propulsion of the uninjured forelimb.

S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 187

Fig. 8. Double support time (%) after pMCAO in SD rats. (A) Decreased Lf-Rh double

support time (%). (B) A deficit was observed in Lh-Rf double support time (%). (Error

bar: S.E.M. *p < 0.05, **p < 0.01 LSD post hoc. The X-axis indicates days.) pMCAO n = 8,

sham n = 8.

In addition, both hindlimbs, but mainly the uninjured hindlimb,

participated in compensatory effort. Several previous studies have

reported significant correlations between gait velocity and lower

limb muscle strength in stroke people [37,38]. This is further sup-

Fig. 9. Homologous coupling of paws after pMCAO in SD rats. (A) The right forepaw

ported by our current results in which changes in muscle strength

was delayed during swing while the left forepaw was in stance. (B) Delayed place-

may have been an important cause of the reduced walk speed, ment of the right hindpaw, Rh, during swing. (C) A representative histological image

from each group. (Error bar: S.E.M. *p < 0.05, **p < 0.01 LSD post hoc. The X-axis

the increased propulsion time, and the altered propulsion index

indicates days.) pMCAO n = 8, sham n = 8.

in pMCAO rats.

In a study of hemiplegic patients, the stance time of the affected

lower limb and the uninjured limb was extended compared to nor-

mal individuals. The stance time of the uninjured lower limb was increased duty cycle (%), which was reduced over time. On the 21st

longer than that of the affected limb. And the duty cycle (%) of day after pMCAO, no differences were observed in the duty cycle

the affected lower limb was lower, while the uninjured one was of the affected hindlimb between the sham and model groups. This

higher, since the decreased weight-bearing capacity of the affected lack of difference might have been due to support by the three less

limb was compensated by the increased contribution of the unin- severely affected limbs as the rats would have avoided the use of

jured lower limb in hemiplegic patients [39]. The results suggested the most affected limb as far as possible in order to improve walking

that at the early stages, pMCAO may have led to a significantly stability.

188 S. Li et al. / Behavioural Brain Research 250 (2013) 174–191

Table 4

Walk speed and other measures of gait.

Sham pMCAO

p value cor p value cor

*** ***

Stride time 2.96E−05 −0.96 Stride time 1.44E−04 −0.90

Stance time Stance time

*** ***

Lf 1.69E−07 −0.99 Lf 2.80E−04 −0.89

− *** *** − Lh 7.18E 06 0.98 Lh 3.44E−05 −0.93

− ***

Rf 8.93E 06 −0.97 Rf 1.81E−01 −0.44

*** ***

Rh 1.18E−05 −0.97 Rh 2.85E−04 −0.89

Propulsion time Propulsion time

***

Lf 7.48E−05 −0.95 Lf 1.05E−01 −0.51

*** ***

Lh 8.02E−06 −0.97 Lh 1.27E−04 −0.91

** − − Rf 4.03E 03 0.85 Rf 2.14E−01 −0.41

*** *

− −

Rh 1.41E 04 0.94 Rh 1.19E−02 −0.72

Brake time Brake time

***

Lf 1.11E−05 −0.97 Lf 1.59E−01 −0.46

*

Lh 1.76E−02 −0.76 Lh 6.45E−02 −0.57

**

Rf 3.00E−03 −0.86 Rf 5.95E−01 −0.18

***

Rh 2.70E−04 −0.93 Rh 2.66E−01 −0.37

Swing speed Swing speed

** ***

Lf 4.06E−03 0.85 Lf 2.47E−04 0.89

− * *** − Lh 1.08E 02 0.79 Lh 3.82E 06 0.96

*** *

Rf 3.04E−05 0.96 Rf 2.57E−02 0.66

** ***

Rh 1.69E−03 0.88 Rh 2.14E−04 0.89

Swing time Swing time

Lf 6.74E−01 −0.16 Lf 2.26E−01 −0.40

Lh 7.71E−02 −0.62 Lh 8.49E−01 0.07

Rf 9.38E−02 −0.59 Rf 4.55E−01 −0.25

−

Rh 1.49E 01 −0.52 Rh 9.52E−01 0.02

Stride length Stride length

− *

Lf 3.34E 02 0.50 Lf 8.37E−02 0.54

* *

Lh 4.56E−02 0.46 Lh 1.77E−03 0.83

*

Rf 1.80E−02 0.76 Rf 8.15E−02 0.55

* − −

Rh 3.88E 02 0.69 Rh 5.26E 02 0.60

Rf: right forepaw; Lf: left forepaw; Rh: right hindpaw; and Lh: left hindpaw.

*

p < 0.05.

**

p < 0.01.

***

p < 0.001.

A study on hemiplegic people showed that single support time Inter-limb coordination is a key characteristic of locomotion and

(%) was decreased, while double support time (%) and double stance therefore the parameters governing coordination are of particular

time was increased. In addition, the gait stability was improved by interest. In quadrupeds, walking on a flat surface is mainly coordi-

the increased double stance time [39]. In both sham and ischemic nated by neuronal networks within the spinal cord that generate a

rats, the longest simultaneous usage of paws during glass plate basic motor pattern. These networks are collectively referred to as

contact was with two or three paws, and rats rarely supported central pattern generators (CPGs) [40,41]. The generated pattern is

themselves with only one paw or four paws at any given time while characterized by a rhythmic alternation between flexor and exten-

standing [16]. Similarly, in our study, three and four limb support sor muscles of the limbs and is likely only initiated and terminated

times (%) of pMCAO rats were increased in an effort to improve by higher motor centers. Diagonal, girdle and ipsilateral phase lags

walking stability to compensate for the decreased double support are parameters that can be used to measure very subtle differ-

time (%). Importantly, it was interesting to uncover a symmetrical ences in inter-limb coordination [42,43]. A very recent report [17]

phenomenon in three support time (%) in the sham group between showed changes induced by cerebral ischemia-reperfusion injury

Rf-Lf-Rh and Rf-Lf-Lh, Rh-Lh-Rf and Rh-Lh-Lf, a phenomenon which in phase dispersions in diagonal pairs as well as in the back girdle

was not observed in the model group. We concluded that the rea- pair. Wang et al. also found that the affected hindpaws were placed

son was as follows: The three support time (%) was increased when later relative to the contralateral hindpaws [16]. Then it is espe-

the three limbs were engaged in support at the same time in order cially important to take into consideration that phase lags depend

to reduce the utilization of the dysfunctional affected limb, as the on speed, and changing from walking to trotting will change the

incongruous movement showed. It suggested that the functional diagonal lateral phase lag. Accordingly, comparison of the diagonal

deficits of the affected hindlimb were relatively more serious. Then and lateral phase lags among quadrupeds is only valid when speeds

the affected forelimb had to be more supportive in order to limit are similar. However, with symmetrical gaits such as the walk and

drooping on the affected side. A similar phenomenon was reported trot, the girdle phase lag is 50% and does not depend on velocity

in a study of impaired gait in a mouse model that underwent [42]. In this study, the walk speed was significantly decreased in

60 min ischemia-reperfusion [17]. This may represent a compen- pMCAO rats, therefore the coupling observed was mainly homolo-

satory mechanism in which the forepaws support the function of gous. In addition, we found that pMCAO led to obvious deficits in

the impaired hindpaws. The largest stroke-induced deficits in the homologous coupling, which recovered at later stages.

hindpaws, especially in the contralateral hindpaw, may reflect neu- In conclusion, a selection of photogrammetric parameters were

roanatomical conditions. Furthermore, the results suggested that used to objectively and accurately illuminate laws of impair-

comparison of gait between injured and uninjured limbs in exper- ment and compensation in walking at different stages of cerebral

imental animals may provide a new way to explore asymmetric ischemia in injured and uninjured rat limbs. Our results show that

gait deficits in cerebral ischemia or certain other diseases with an this system of very sensitive measures may be a promising way of

underlying pathogenesis. evaluating the performance of affected paws after a stroke.

S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 189

Table 5

The volumes of cerebral infarction and indicators of gait impairment.

Time point Indicators of gait impairment p value cor

related to the volume of

cerebral infarction

1d None

*

7d Stride length Lh 3.18E−02 −0.75

*

Swing speed Lh 3.11E−02 −0.75

*

Swing speed Rh 1.78E−02 −0.80

*

Stance time Lf 1.88E−02 0.79

*

Stance time Rf 3.99E−02 0.73

*

Duty cycle Rf 1.38E−02 0.81

− *

Duty cycle Rh 4.95E 02 0.71

**

Double support time Lf-Rh 2.48E−03 −0.90

*

Three support time Rf-Lf-Rh 2.39E−02 0.77

*

Homologous coupling Lf→Rf 1.34E−02 0.82

*

Homologous coupling Lh→Rh 3.02E−02 0.76

*

Propulsion time Rf 1.97E−02 0.79

*

Propulsion time Lf 2.40E−02 0.77

*

Propulsion time Rh 3.77E−02 0.74

* − Propulsion index Rf 2.36E 02 0.78

*

Brake time Lf 4.36E−02 0.72

*

21d Stride time 2.77E−02 0.76

*

Walk speed 3.19E−02 −0.75

**

Homologous coupling Lh→Rh 8.85E−03 0.84

**

Brake time Lf 4.26E−03 0.88

*

Brake time Rf 1.73E−02 0.80

** − Propulsion time Lh 5.03E 03 0.87

*

Propulsion time Rh 4.58E−02 0.72

− *

Propulsion time Rf 4.62E 02 0.71

**

Stance time Lh 5.33E−03 0.87

*

Stance time Rf 1.37E−02 0.81

* − Stance time Lf 2.40E 02 0.77

*

Stance time Rh 3.38E−02 0.75

**

Duty cycle Lh 8.14E−03 0.85

*

42d Swing time Lh 3.23E−02 −0.75

Rf: right forepaw; Lf: left forepaw; Rh: right hindpaw; and Lh: left hindpaw.

*

p < 0.05.

**

p < 0.01.

4.2. Correlations between the volume of cerebral infarction and Functional deficits in walking may be closely related with sum of

indicators of walking skill reversible and irreversible infarct volume. However, the cerebral

infarct volume of pMCAO rats after 42 days of recovery may repre-

Functional scoring has become increasingly important in test- sent the volume of completely necrotic cerebral tissue. Therefore,

ing neuroprotective drugs, although little consensus exists in the the fact that there were no correlations at acute stages may be due

literature about the different scoring systems for neurological to differences in the ischemic and infarcted areas at acute and later

assessment in animal models of stroke. Infarct size and functional stages, and this may in turn depend on the establishment of collat-

deficits have been reported to be influenced by various factors, such eral circulation. In addition, in a recent study [14], results showed

as model, strain, vendor, age, and cerebral artery occlusion time that there was no direct correlation in tMCAO rats between the

[44–51]. Some authors have even reported that unsuccessful trial extent of an infarct and the behavioral outcome after three months

results in the application of certain tests were due to the rat strain of recovery. However, significant correlations between the final

or experimental design used, while others have questioned the reli- infarction volume and earlier behavioral outcomes were appar-

ability of some tests [52,53]. Therefore, in studying animal models, ent in the composite neuroscore and rearing test results, but not

subjective, qualitative, and inaccurate behavioral approaches may beyond one month of recovery. Similarly, there was no correlation

probably result in that it is extremely difficult to uncover associ- between the infarct volume and gait parameters in pMCAO rats

ations between functional deficits in behavior and the degree of on the 42nd day. This is likely because the gait function had been

ischemia. Gait analysis, however, has been developed as a consis- restored to a relatively stable state.

tent, non-subjective measure of outcome that can detect long-term Moreover, several researchers have found that ischemic damage

sensorimotor deficits following MCAO. Hence, gait analysis may be in one cerebral hemisphere can cause an adaptive response in the

a more stable and reliable behavior method to uncover associations intact hemisphere whereby cross-midline sensorimotor functions

between functional deficits in behavior and the degree of ischemia. are restored over time [56–58]. Disturbances in gait parameters

Our gait analysis results showed that the cerebral infarction vol- can be restored over time but this may not result from repair of

ume of pMCAO rats was correlated with functional deficits of gait the cerebral infarct, but rather from compensation by the intact

on the 7th and 21st day after pMCAO. This was in good agreement hemisphere. In our study, gait function of pMCAO rats was recov-

with previous animal studies [54] in which a correlation of gait dis- ered to a certain degree, 7 and 21 days post surgery, and some

turbances with the infarct size was also found on days 7 and 10 in gait parameters were correlated with the infarct size. This sug-

animals subjected to pMCAO. gests that there might be some association between the degree of

During the acute phase of pMCAO, both reversible and recovery of gait function and cerebral ischemia. The more serious a

irreversible changes [55] are included in cerebral ischemia. Recov- cerebral ischemia, the less complete would be the recovery of gait

ery of the reversible changes may be either partial or even complete. function.

190 S. Li et al. / Behavioural Brain Research 250 (2013) 174–191

4.3. Comparing gait impairment in patients and experimental mechanisms of focal cerebral ischemic disease and potential ther-

stroke models apies.

There are a variety of commercially available systems for mea-

Acknowledgements

suring gait and posture changes in humans, including GAITRite [59],

three-dimensional gait analysis and a plantar pressure distribution

This work was supported by the Ministry of Science and Tech-

measure system [60]. Certain systems allow physicians to ana-

nology of China (Fund Nos.: 2011DFA32730; 2009ZX09502-014;

lyze not only footprint changes, but also joint angle changes from

and 2011CB711000).

the pelvis to ankles as well as plantar pressure distribution during

walking [61]. After a stroke, most patients asked to walk at a com-

References

fortable pace will walk at speeds that range from approximately

0.2 m/s to 0.8 m/s [62–65]. Walking velocities of these patients are

[1] World Health Organization (WHO), editor. The global burden of disease –

significantly lower than the comfortable walking speed exhibited

update 2004. Geneva: WHO Library Cataloguing-in-Publication Data; 2008.

by age-matched individuals (1.3–1.4 m/s). Although between 52% [2] Lloyd-Jones D, Adams RJ, Brown TM, Carnethon M, Dai S, De Simone G, et al.

Heart disease and stroke statistics – 2010 update: a report from the American

and 85% of patients can regain the capacity to walk, their gait

Heart Association. Circulation 2010;121:46–215.

usually remains different from that of healthy subjects [66,67].

[3] Jorgensen HS, Nakayama H, Raaschou HO, Olsen TS. Stroke neurologic and

And similar to the results of our experimental stroke study, one functional recovery the Copenhagen stroke study. Physical Medicine and Reha-

bilitation Clinics of North America 1999;10:887–906.

group [12] noted that velocity, cadence and stride length in stroke

[4] Mayo NE, Wood-Dauphinee S, Ahmed S, Gordon C, Higgins J, McEwen S,

patients tended to decrease while stance was increased. However,

et al. Disablement following stroke. Disability and Rehabilitation 1999;21:

in previous tMCAO gait studies in rats [14,16], walk speed, cadence 258–68.

and stance time showed no significant changes, which also indi- [5] Endres M, Engelhardt B, Koistinaho J, Lindvall O, Meairs S, Mohr JP, et al.

Improving outcome after stroke: overcoming the translational roadblock. Cere-

cates that pMCAO is more similar to naturally occurring cerebral

brovascular Diseases 2008;25:268–78.

ischemia in humans than is tMCAO, and thus is of greater clinical

[6] Merino JG, Hachinski V. Stroke-related dementia. Current Atherosclerosis

relevance. The present study is the first to examine gait in pMCAO 2002;Rep 4:285–90.

[7] Sachdev PS, Brodaty H, Valenzuela MJ, Lorentz LM, Koschera A. Progression of

rats using an automated computer-assisted method. For a more

cognitive impairment in stroke patients. Neurology 2004;63:1618–23.

comprehensive evaluation of gait deficits, we added average body

[8] Carlini EA, Teresa M, Silva A, Cesare LC, Endo RM. Effects of chronic

rotation, propulsion time and the propulsion index, none of which administration of beta-(3,4-dimethoxyphenyl)-ethylamine and beta-(3,4,5-

trimethoxyphenyl)-ethylamine on the climbing rope performance of rats.

had previously been used in gait studies of cerebral ischemia. How-

Medicina et Pharmacologia Experimentalis International Journal of Experimen-

ever, the distal MCAO model used in the current study limits the

tal Medicine 1967;17:534–42.

infarct to the sensorimotor cortex, whereas in hemiparetic human [9] Kunkel-Bagden E, Dai HN, Bregman BS. Methods to assess the development and

recovery of locomotor function after spinal cord injury in rats. Experimental

strokes, the injury often involves the internal capsule, resulting in

Neurology 1993;119:153–64.

damage to the thalamocortical or extrapyramidal tracts in addition

[10] Metz GA, Whishaw IQ. Cortical and subcortical lesions impair skilled walk-

to the corticospinal tract. Thus, in the future, researchers should ing in the ladder rung walking test: a new task to evaluate fore- and

hindlimb stepping, placing, and co-ordination. Journal of Neuroscience Meth-

employ a variety of cerebral ischemia models in order to analyze

ods 2002;115:169–79.

gait from other angles in a meticulous and comprehensive manner.

[11] de Medinaceli L, Freed WJ, Wyatt RJ. An index of the functional condition of rat

In addition, parameters related to joint angle and distribution of sciatic nerve based on measurements made from walking tracks. Experimental

foot pressure should also be utilized. Gait observations should also Neurology 1982;77:634–43.

[12] Mah CD, Hulliger M, O’Callaghan IS, Lee RG. Quantitative kinematics of gait

be carried out using different kinds of cerebral ischemia models,

patterns during the recovery period after stroke. Journal of Stroke and Cere-

such as two-vessel occlusion, four-vessel occlusion, and throm-

brovascular Diseases 1999;8:312–29.

botic emboli. Accordingly, more gait parameters and other cerebral [13] Scherder E, Eggermont L, Sergeant J, Boersma F. Physical activity and cognition

in Alzheimer’s disease: relationship to vascular risk factors, executive functions

ischemia models will be studied in our laboratory in order to

and gait. Reviews in the Neurosciences 2007;18:149–58.

provide a broader understanding of the theoretical basis underlying

[14] Encarnacion A, Horie N, Keren-Gill H, Bliss TM, Steinberg GK, Shamloo M. Long-

the pathological mechanisms of cerebral ischemia and to prepare term behavioral assessment of function in an experimental model for ischemic

stroke. Journal of Neuroscience Methods 2011;196:247–57.

the way for better tools for evaluating the therapeutic effectiveness

[15] Vandeputte C, Taymans JM, Casteels C, Coun F, Ni Y, Van Laere K, et al. Auto-

of potential drugs.

mated quantitative gait analysis in animal models of movement disorders. BMC

Neuroscience 2010;11:92.

[16] Wang Y, Bontempi B, Hong SM, Mehta K, Weinstein PR, Abrams GM, et al.

A comprehensive analysis of gait impairment after experimental stroke and

5. Conclusion

the therapeutic effect of environmental enrichment in rats. Journal of Cerebral

Blood Flow and Metabolism 2008;28:1936–50.

[17] Hetze S, Römer C, Teufelhart C, Meisel A, Engel O. Gait analysis as a method

In this research, some objective and accurate photogrammet-

for assessing neurological outcome in a mouse model of stroke. Journal of

ric parameters were used to illuminate the laws of impairment

Neuroscience Methods 2012;206:7–14.

and compensation in rats at different stages of cerebral ischemia [18] Richard Green A, Odergren T, Ashwood T. Animal models of stroke: do they

in injured and uninjured limbs during walking. We found that have value for discovering neuroprotective agents? Trends in Pharmacological

Sciences 2003;24:402–8.

the average body rotation and propulsion index, which none of

[19] Hossmann K-A. Pathophysiology and therapy of experimental stroke. Cellular

these have previously been used, were specific parameters for

and Molecular Neurobiology 2006;26:1055–81.

accurately assessing gait function during the acute phase after [20] Longa EZ, Weinstein PR, Carlson S, Cummins R. Reversible middle cerebral

artery occlusion without craniectomy in rats. Stroke 1989;20:84–91.

pMCAO. This indicates that the sensitive measures for skilled

[21] Bederson JB, Pitts LH, Tsuji M, Nishimura MC, Davis RL, Bartkowski H. Rats

walk may be a reliable way to evaluate the performance of

middle cerebral artery occlusion evaluation of the model and development of

affected paws after a stroke. In addition, the gait test results a neurologic examination. Stroke 1986;17:472–6.

[22] Gorska T, Majczynski H, Zmyslowski W. Overground locomotion in intact

revealed significant correlations between the final infarction vol-

rats: contact electrode recording. Acta Neurobiologiae Experimentalis (Wars)

ume and earlier behavior outcomes. Furthermore we demonstrated 1998;58:227–37.

that the gait changes observed in pMCAO rats were more sim- [23] Gillis GB, Biewener AA. Hindlimb muscle function in relation to speed and

gait: in vivo patterns of strain and activation in a hip and knee extensor

ilar to changes following naturally occurring cerebral ischemia

of the rat (Rattus norvegicus). Journal of Experimental Biology 2001;204:

in humans. In summary, we propose that the gait analysis has

2717–31.

promise as a potential tool for assessing cerebral ischemia severity, [24] Cendelín J, Voller J, Vozeh F. Ataxic gait analysis in a mouse model of the

olivocerebellar degeneration. Behavioural Brain Research 2010;210:8–15.

and that this method may provide a new means of investigating

S. Li et al. / Behavioural Brain Research 250 (2013) 174–191 191

[25] Lenzlinger PM, Saatman KE, Hoover RC, Cheney JA, Bareyre FM, Raghupathi [46] Grabowski M, Nordborg C, Johansson BB. Sensorimotor performance and rota-

R, et al. Inhibition of vascular endothelial growth factor receptor (VEGFR) tion correlate to lesion size in right but not left hemisphere brain infarcts in

signaling by BSF476921 attenuates regional cerebral edema following trau- the spontaneously hypertensive rat. Brain Research 1991;547:249–57.

matic brain injury in rats. Restorative Neurology and Neuroscience 2004;22: [47] Grabowski M, Sorensen JC, Mattson B, Zimmer J, Johansson BB. Influence of

73–9. an enriched environment and cortical grafting on functional outcome in brain

[26] Bederson JB, Pitts LH, Tsuji M, Nishimura MC, Davis RL. Rat middle cerebral infarcts of adult rats. Experimental Neurology 1995;133:96–102.

artery occlusion: Evaluation of the model and development of a neurological [48] Kuge Y, Minematsu K, Yamaguchi T, Miyake Y. Nylon monofilament for intra-

examination. Stroke 1986;17:472–6. luminal middle cerebral artery occlusion in rats. Stroke 1995;26:1655–7.

[27] Garcia JH, Wagner S, Liu KF, Hu XJ. Neurological deficit and extent of neu- [49] Markgraf CG, Green EJ, Hurwitz BE, Morikawa E, Dietrich WD, McCabe PM, et al.

ronal necrosis attributable to middle cerebral artery occlusion in rats. Stroke Sensorimotor and cognitive consequences of middle cerebral artery occlusion

1995;26:627–35. in rats. Brain Research 1992;575:238–46.

[28] Menzies SA, Hoff JT, Betz AL. Middle cerebral artery occlusion in rats: a neu- [50] Memezawa H, Zhao Q, Smith ML, Siesjo BK. Hyperthermia nullifies the amelio-

rological and pathological evaluation of a reproducible model. Neurosurgery rating effect of dizocilpine maleate (MK-801) in focal cerebral ischemia. Brain

1992;31:100–7. Research 1995;670:48–52.

[29] Petullo D, Masonic K, Lincoln C, Wibberley L, Teliska M, Yao DL. Model develop- [51] Oliff HS, Marek P, Miyazaki B, Weber E. The neuroprotective efficacy of MK-

ment and behavioral assessment of focal cerebral ischemia in rats. Life Sciences 801 in focal cerebral ischemia varies with rat strain and vendor. Brain Research

1999;64:1099–108. 1996;731:208–12.

[30] Lehmann JF. Push off and propulsion of the body in normal and abnormal gait. [52] Van der Staay FJ, Augstein KH, Horvath E. Sensorimotor impairments in Wistar

Clinical Orthopaedics and Related Research 1993;28:97–108. Kyoto rats with cerebral infarction, induced by unilateral occlusion of the mid-

[31] Olney SJ, Carol R. Hemiparetic gait following stroke. Part I: Characteristrics. Gait dle cerebral artery: recovery of function. Brain Research 1996;715:180–8.

& Posture 1996;4:136–48. [53] Zausinger S, Hungerhuber E, Baethmann A, Reulen HJ, Schmid-Elsaesser R.

[32] Nadeau S, Gravel D, Arsenault AB, Bourbonnais D. Plantarflexor weakness as a Neurological impairment in rats after transient middle cerebral artery occlu-

limiting factor of gait speed in stroke subjects and the compensating role of hip sion: a comparative study under various treatment paradigms. Brain Research

flexors. Clinical Biomechanics 1999;14:125–35. 2000;863:94–105.

[33] Wooley CM, Sher RB, Kale A, Frankel WN, Cox GA, Seburn KL. Gait analy- [54] Reglodi D, Tamás A, Lengvári I. Examination of sensorimotor performance

sis detects early changes in transgenic SOD1(G93A) mice. Muscle and Nerve following middle cerebral artery occlusion in rats. Brain Research Bulletin

2005;32:43–50. 2003;15:459–66.

[34] Mancuso R, Oliván S, Osta R, Navarro X. Evolution of gait abnormalities in [55] Minematsu K, Li L, Sotak CH, Davis MA, Fisher M. Reversible focal ischemic

SOD1(G93A) transgenic mice. Brain Research 2011;1406:65–73. injury demonstrated by diffusion-weighted magnetic resonance imaging in

[35] Krouchev N, Kalaska JF, Drew T. Sequential activation of muscle synergies dur- rats. Stroke 1992;23:1304–11.

ing locomotion in the intact cat as revealed by cluster analysis and direct [56] Jones TA, Allred RP, Adkins DL, Hsu JE, O’Bryant A, Maldonado MA. Remodeling

decomposition. Journal of Neurophysiology 2006;96:1991–2010. the brain with behavioral experience after stroke. Stroke 2009;40:136–8.

[36] Hall AL, Peterson CL, Kautz SA, Neptune RR. Relationships between muscle [57] Nudo RJ. Postinfarct cortical plasticity and behavioral recovery. Stroke

contributions to walking subtasks and functional walking status in per- 2007;38:840–5.

sons with post-stroke hemiparesis. Clinical Biomechanics (Bristol, Avon) [58] Woodlee MT, Asseo-Garcia AM, Zhao X, Liu SJ, Jones TA, Schallert T. Testing fore-

2011;26:509–15. limb placing “across the midline” reveals distinct, lesion-dependent patterns

[37] Damiano DL, Abel MF. Functional outcomes of strength training in spastic cere- of recovery in rats. Experimental Neurology 2005;191:310–7.

bral palsy. Archives of Physical Medicine and Rehabilitation 1998;79:119–25. [59] Ng SS, Hui-Chan CW. The timed up & go test: its reliability and association

[38] Nadeau S, Arsenault AB, Bourbonnais D, GraveI D. Analysis of the clinical factors with lower-limb impairments and locomotor capacities in people with chronic

determining natural and maximal gait speeds in adults with a stroke. American stroke. Archives of Physical Medicine and Rehabilitation 2005;86:1641–7.

Journal of Physical Medicine and Rehabilitation 1999;78:123–30. [60] Chen G, Pateern C, Kothari DH, Zajac FE. Gait differences between individuals

[39] Brandstater ME, de Bruin H, Gowland C, Clark BM. Hemiplegic gait: analy- with post-stroke hemiparesis and non-disabled controls at matched speeds.

sis of temporal variables. Archives of Physical Medicine and Rehabilitation Gait and Posture 2005;22:51–6.

1983;64:583–7. [61] Woodford HJ, Price C. Electromyographic biofeedback for the recovery of motor

[40] Cazalets JR, Borde M, Clarac F. Localization and organization of the central pat- function after stroke. Stroke 2007;19:1999–2000.

tern generator for hindlimb locomotion in newborn rat. Journal of Neuroscience [62] Bohannon RW. Walking after stroke: comfortable versus maximum safe speed.

1995;15:4943–51. International Journal of Rehabilitation Research 1992;15:246–8.

[41] Kjaerulff O, Kiehn O. Distribution of networks generating and coordinating [63] Turnbull GI, Charteris J, Wall JC. A comparison of the range of walking speeds

locomotor activity in the neonatal rat spinal cord in vitro: a lesion study. Journal between normal and hemiplegic subjects. Scandinavian Journal of Rehabilita-

of Neuroscience 1996;16:5777–94. tion Medicine 1995;27:175–82.

[42] Hamers FP, Koopmans GC, Joosten EA. CatWalk-assisted gait analysis in the [64] Jonsdottir J, Recalcati M, Rabuffetti M, Casiraghi A, Boccardi S, Ferrarin M. Func-

assessment of spinal cord injury. Journal of Neurotrauma 2006;23:537–48. tional resources to increase gait speed in people with stroke: strategies adopted

[43] Koopmans GC, Brans M, Gomez-Pinilla F, Duis S, Gispen WH, Torres-Aleman compared to healthy controls. Gait and Posture 2009;29:355–9.

I, et al. Circulating insulin-like growth factor I and functional recovery from [65] Jonkers I, Delp S, Patten C. Capacity to increase walking speed is limited by

spinal cord injury under enriched housing conditions. European Journal of impaired hip and ankle power generation in lower functioning persons post-

Neuroscience 2006;23:1035–46. stroke. Gait and Posture 2009;29:129–37.

[44] Belayev L, Alonso OF, Busto R, Zhao W, Ginsberg MD. Middle cerebral artery [66] Bohannon RW. Gait performance of hemiparetic stroke patients selected vari-

occlusion in the rat by intraluminal suture. Stroke 1996;27:1616–23. ables. Archives of Physical Medicine and Rehabilitation 1987;68:777–81.

[45] Dirnagl U, Tanabe J, Pulsinelli W. Pre- and post-treatment with MK-801 but not [67] Eng JJ, Chu KS. Reliability and comparison of weight-bearing ability dur-

pretreatment alone reduces neocortical damage after focal cerebral ischemia ing standing tasks for individuals with chronic stroke. Archives of Physical

in the rat. Brain Research 1990;527:62–8. Medicine and Rehabilitation 2002;83:1138–44.