1.2.1

Population

The inclusion criterion was hemiplegia after a first ischemic or hemorrhagic unifocal stroke, located on one of the brain’s hemispheres. The exclusion criteria were: recurrent stroke episode, multifocal stroke or not located on one hemisphere, post-stroke delay over 1 year, the existence of a retracted rigid foot, severe spastic claw toe (due to capsular or tendinous retraction) or passive spastic claw toe (during dorsiflexion of the ankle, secondary to a retraction of the long flexors with no spasticity).

Among 81 patients hospitalized for rehabilitation after stroke between September 2000 and September 2001, 39 met the inclusion criterion.

1.2.2

Evaluation

To do the evaluation we looked for claw toes at various times during the patients’ hospital stay: upon admission, during check-ups at 1-month intervals during the entire hospital stay and/or up to 1-year post-stroke at the most.

The patient’s evaluation was done upon admission into our Physical and Rehabilitation (PM&R) department; the date of onset and localization of the stroke were noted.

To unveil this claw toe condition, the ankle was positioned at 0, first while sitting down, brushing the foot on the floor, then standing up into a loading position and finally during gait.

We assessed the spasticity on the triceps surae, patients sitting in a chair, using the modified Ashworth scale .

Leg impairment was evaluated and analyzed with the Held-Pierrot Deseilligny index score, used in France, derived from the Daniels and Worthingham’s muscle testing while trying to eliminate movement synkinesis. Muscle testing was done on the muscles of the leg (and foot) with the patient sitting in a chair; we conducted a testing for the triceps surae (for which we evaluated the spasticity) with the patient’s knee in extension.

Functional gait autonomy was assessed using the Functional ambulation classification (FAC) ; postural balance was assessed using the Postural assessment scale for stroke patients (PASS) . Global functional autonomy was evaluated using the Barthel Index .

The scores from the Barthel Index and PASS scale were divided in order to obtain respectable-sized populations: the Barthel Index, with a maximum score of 100, was broken down into 10 categories and the PASS scale with its 12 items quoted from 0 to three (for a maximum score of 36), was divided into 12 categories.

During their hospital stay, all patients benefited from twice-daily physiotherapy care.

1.2.3

Statistical analysis

The statistical analysis was done with the SAS software, version 5.0 (SAS Institute, Inc., Cary, NC) using the χ ² independence test and Student’s t test (comparing averages) with p < 0.05 considered significant.

1.4.1

Methodology

The aim of this study was to analyze some of the severity criteria for hemiplegia and the functional autonomy scores in stroke patients with claw toes, each evaluation could be considered independent from the other. This study was strictly limited to the first year post-stroke.

Recruitment bias was another of our study’s limit; our population was made of patients hospitalized in a PM&R department without any homogeneity in post-stroke delay and hospital stay duration and not a random population of stroke patients with hemiplegia. Furthermore, three patients were recruited at a later stage (9–10 months post-stroke).

1.4.2

Analysis of the results

This study shows the prevalence and early onset (under 3 months post-stroke) of claw toes in a population of hemiplegic patient hospitalized in the PM&R department of a hospital after a first episode of unifocal stroke, located on one of the brain hemispheres. The onset of claw toes is correlated to equinus and/or varus foot; however, no significant correlation was found with spasticity severity, muscle testing of the tibialis anterior and the various autonomy scores (PASS, FAC, Barthel Index).

The pathology of claw toes has been known for many years; in 1914, Dejerine described it as associated to equinus foot and reported that this condition was less common in hemiplegic adults than children . Alajouanine et al. , in over two years of neurological consultations, reported only 12 cases of claw toes among hundreds of patients. In their series, only seven patients had claw toes, considered as minor sequelae of their stroke and were consulting for this reason. This proves that claw toes, as an isolated condition is not a common reason for consulting even though its incidence in our population is important. Thus, claw toes affected 46% of the patients included in our study. This percentage is largely higher than the 2% found by Verdié et al. in a population of stroke patients with hemiplegia seen at 1-year post stroke. This difference is probably linked to the recruitment method: patients hospitalized in a PM&R department had severe strokes. This difference should be compared to the one described by Mazaux and Debelleix for the observation of equinovarus foot: the incidence of equinovarus foot (50%) in a population of patients hospitalized in a PM&R department is largely higher (20%) than the one observed in a regular population of stroke patients.

Claw toes are rarely described in a population with severe motor impairments, partly because looking for this pathology implies for patients to walk barefoot, with a great risk of falls. Furthermore, in these patients, this condition is not inducing the main gait difficulties and thus is rarely mentioned.

Claw toes occurring early on in the recovery stage, before the third month post-stroke were in favor of a spastic component whereas its lingering during gait was in favor of a dystonic component .

This shows that the distinction between spastic and dystonic claw toe is still under discussion. The only validated points are the lesion’s topology: corticospinal for spastic claw toe and noncorticospinal for dystonic claw toe and the fact that dystonia appears at a distance from the stroke (even after a few years) in patients who recovered quite well from their motor impairment.

For some authors , “claw toe appearing when standing up in loading and getting worse during gait will be qualified as spastic. Conversely, a more versatile claw toe, appearing after a few steps, more likely when walking backwards and without hypertonia of the toes’ flexors will be mostly qualified as dystonic”.

For others , spastic claw toes are intermittent, characterized by the flexion of all toes and occur in the final phase of putting the foot down on the floor when the toes are in extension; the condition disappears during the swing phase. Dystonic claw toes affect all toes including the great toe and linger throughout all gait phases. However, the condition disappears when resting. We differentiate two types of toe dystonia: the type with the isolated extension of the great toe and the one associating the great toe and flexion of all other toes.

Regarding equinovarus foot, it is also very difficult to make the distinction between spasticity, dystonia or a combination of both (spastic and dystonic) .

We can elaborate that the mechanism of onset for claw toes or equinovarus foot is similar since claw toes and equinovarus foot are strongly correlated whereas neither one nor the other are significantly correlated to spasticity of the triceps surae.

In our series, claw toes did not affect two groups of patients. The first group included patients in the early post-stroke stage, with very low scores at the Barthel Index and PASS scale; these patients, with severe motor impairments, were not yet affected by spasticity. The second group was made of patients who achieved good functional recovery with a less severe hemiplegia and an acceptable autonomous gait with high Barthel Index score.

The frontier between spasticity and dystonia remains poorly defined in stroke patients with hemiplegia. Nowadays, we mostly use the term of muscle excitability, in a context of upper motor neuron syndrome, by the disappearance of the inhibitory control of the reticular formation, associated to axonal nerve sprouting phenomena or neuronal plasticity .

This muscle excitability is responsible for three elements:

• •
  

  spasticity defined by exaggerated spinal reflex that can progress towards biomechanical changes in the muscle, responsible for elastic hypertonia;

  
  
• •
  

  co-contractions of agonist and antagonist muscles;

  
  
• •
  

  “spastic” dystonia related to the activation of the descending motor pathways, independently of all reflexes like for example, involuntary spasms of the elbow, wrist and fingers during gait .

The Barthel Index and the PASS scale results must be carefully analyzed. In fact, in our population, two subgroups presented very little incidence of claw toes: the first subgroup were patients with severe motor impairments or in a very early post-stroke stage (very low Barthel Index and PASS scale score) in whom spasticity did not set in yet. The other subgroup was made of patients that recovered well or with a less severe hemiplegia with good scores.

The incidence of claw toes was high in patients with severe motor impairments and noticeable spasticity. It is likely that in that case the condition observed had a both spastic and dystonic components. The only arguments that could have validated the spastic or dystonic characteristic would have been the persistence, or lack of, this condition during the swing phase of the gait and walking backwards.

However, the onset of claw toes concomitant to spasticity of the triceps surae in three patients is in favor of a spastic involvement.

In light of the observed results, a mixed clinical score could be suggested it would combine “severity of the leg impairment” associated to a presence or lack of equinus and/or varus foot, muscle testing of the tibialis anterior and evaluation of the spasticity of the triceps surae. It could be used as a risk factor index for the incidence of claw toes in stroke patients. This index would need to be validated on a larger population.

2.2.1

Population

Le critère d’inclusion était l’existence d’une hémiplégie secondaire à un premier AVC ischémique ou hémorragique, unifocal, de localisation hémisphérique.

Les critères d’exclusion étaient : une récidive d’AVC, un AVC multifocal ou non hémisphérique, un délai post-AVC supérieur à un an, l’existence d’un pied spastique rétracté, de griffe des orteils irréductible (par rétraction capsulaire ou tendineuse) ou de griffe des orteils passive (lors de la dorsiflexion de cheville, secondaire à une rétraction des longs fléchisseurs, en l’absence de spasticité).

Parmi 81 patients hospitalisés entre septembre 2000 et septembre 2001 pour rééducation d’un hémiplégie, 39 répondaient aux critères d’inclusion et d’exclusion.

2.2.2

Évaluation

La recherche d’une griffe des orteils était réalisée à l’admission et pouvait être renouvelée lors des bilans à intervalles d’au minimum un mois pendant toute la durée d’hospitalisation et/ou pour un délai post-AVC d’un an maximum.

L’évaluation du patient était réalisée à l’entrée dans le service ; la date de survenue et la localisation de l’AVC étaient notées.

L’existence d’une griffe des orteils était recherchée cheville à 0, d’abord en position assise, par frottement du pied sur le sol, puis lors du passage en orthostatisme et enfin à la marche.

La spasticité était évaluée sur le triceps surae à l’aide de l’échelle d’Ashworth modifiée , patient assis.

Le déficit du segment jambier était évalué de manière analytique par la cotation de Held-Pierrot Deseilligny, utilisée en France, dérivée de la méthode de Daniels et Worthingham en essayant d’éliminer les syncinésies. Le testing musculaire était réalisé en position assise sur les muscles du segment jambier (et du pied) ; le testing du triceps (dont la spasticité était évaluée) était réalisé genou en extension.

L’autonomie de déambulation était évaluée au moyen de l’échelle Functional ambulation classification (FAC) , l’équilibre par l’échelle Postural assessment scale for stroke patients (PASS) . L’autonomie fonctionnelle globale était évaluée par l’index de Barthel .

Les scores de l’index de Barthel et de l’échelle PASS ont été divisés de façon à obtenir des populations de taille acceptable : l’index de Barthel, de score maximum 100, en dix classes et l’échelle PASS, comportant 12 items cotés de 0 à trois, soit un score maximum de 36, en 12 classes.

Pendant leur hospitalisation, tous les patients bénéficiaient d’une prise en charge kinésithérapique biquotidienne.

2.2.3

Analyse statistique

L’analyse statistique réalisée avec le logiciel SAS, version 5.0 (SAS Institute, Inc., Cary, NC) utilisait le test χ ² d’indépendance et le test t de Student (comparaison de moyennes) avec une valeur de p < 0,05 considérée comme significative .