2.1

Patients

Nineteen patients admitted to the neurological rehabilitation medicine unit of Hôpital Henry-Gabrielle, Hospices Civils de Lyon and presenting with left SN secondary to a right-hemisphere stroke were included in the study from September 2001 until September 2005 ( Fig. 1 ). All patients were right-handed according to the criteria of the Edinburgh questionnaire .

Study design. Mild neglect: BIT score > 55; Severe neglect: BIT score ≤ 55.

The inclusion criteria were:

• •
  

  age ranging from 18 to 90 years;

  
  
• •
  

  a single stroke confirmed by a tomodensitometry examination or by brain MRI;

  
  
• •
  

  left SN confirmed by several neuropsychological tests (line bisection test , balloon test , copy of a drawing, dictation and reading of a text);

  
  
• •
  

  a time lapse of at least one month following the ischemic event.

The exclusion criteria were:

• •
  

  existence of multiple brain lesions;

  
  
• •
  

  temporo-spatial disorientation;

  
  
• •
  

  psychiatric disorders;

  
  
• •
  

  an associated, non-stabilized pathology.

The characteristics of the patients in the two groups are detailed in Table 1 .

2.2

Study description

This was a double-blind monocentric controlled randomized clinical trial involving 2 groups of patients presenting with left SN: a group undergoing PA rehabilitation, and a control group. The primary outcome measure of the study was the functional improvement in daily life activities following rehabilitation as assessed by the Functional Independence Measure (FIM) . This score was previously used in a non-randomized clinical trial evaluating the effectiveness of a trunk orthosis in SN rehabilitation in two groups of patients . The authors demonstrated, at 6 months, a statistically significant difference between the mean scores of the 2 groups (mean difference = 24; n = 11). In this study, the FIM end-point had a standard deviation of 10 units. Based on this variability measure, and to have at least a 90% chance of showing a 25-point difference between mean responses in the two treatments with a risk of type 1 error not exceeding 5%, 9 subjects had to be included in each group, (1–β = 0.90; β = 0.10; α = 0.05 and Δ = 25; n = σ ² × M/Δ ² = 100 × 54/625 = 8.6, i.e. at least 9 subjects).

Block randomization was carried out with the distinction of two levels of deficit severity used according to the initial seriousness of SN as assessed through the inclusion tests. Severe SN: was defined as deficit shown in all the inclusion tests, Schenkenberg score greater than 50° of deviation and BIT score ≤ 55. Moderate SN: was defined as deficit shown in some (from 1 to 4) of the inclusion tests, Schenkenberg score ranging from 11 to 50° of deviation and BIT score > 55.

The study was double-blinded: The examiners carrying out the evaluation (GR, SL, EM), did not know whether a given patient had undergone PA. They were distinct from the examiners performing the task of exposure to prismatic or neutral glasses (YR, SJC, LP). The double-blind procedure was facilitated by the fact that the SN patients were not aware of the disturbance induced by prism deviation and did not present the vegetative reactions expected during the appearance of motor errors when the prisms were worn for the first time (cf. infra and ). Consequently, they could be assigned without their knowledge to the “prism” and “control” groups. This also entailed that examiners performing the assessment did not receive information from the patients all of which that might have compromised the double-blind trial.

Block randomization and drawing by lots of the patients into the “prism” or “control” groups was carried out by Denis Pelisson, director of the ImpAct team in the Lyon Neuroscience Research Centre. The randomization was produced at 2 levels, firstly by selection of patients for the “prism’ or “control” group; secondly, by selection of patients according to the severity of the initial SN assessed by the BIT score, the objective being that the ratio of patients with severe neglect and patients with moderate neglect is comparable in the two groups.

All patients gave their consent to participate in the study. The experimental procedure was approved by the Comité Consultatif de Protection des Personnes dans la Recherche Biomédicale Lyon B on 5 June 2001 le (Dossier 2001-040 B). The Hospices Civils de Lyon sponsored this study, which was registered on 2 August 2001 under the number 2001/0294. The study was financed by Inserm, the Hospices Civils de Lyon and the university Claude-Bernard Lyon 1.

2.3

Study parameters

The spatial frames of reference and precision of pointing at a visual target (without visual feedback) were measured from patient inclusion data through six months so as to monitor their evolution while the patients were recovering. Two measurements took place, before and after each exposure session, in view of quantifying the after-effects of prism adaptation.

The primary outcome measure for therapeutic efficacy was the functional score achieved according to the Functional Independence Measure (FIM) . The secondary outcome measure was the total score in the Behavioural Inattention Test (BIT) , which constitutes a good indicator of SN severity. Measurement with regard to the primary end-point was carried out 4 times: in pre-tests and in post-tests at 1 (M1), 3 (M3) and 6 (M6) months after the initial PA session. The BIT evaluations were performed twice: in pre-test, and then in post-test at 6 months. No intermediate evaluation took place during monitoring, in order to avoid the confounding effect of learning through repeated testing.

2.4

Prism adaptation and sensory-motor parameters

In the “prism” group, PA was carried out by wearing a pair of glasses producing a 10° rightward optical deviation of the visual field (OptiquePeter.com). The prismatic lenses were composed of two superimposed, curved, point-to-point lenses fitted with a “glacier” frame containing lateral leather protectors designed to avoid access to non-shifted vision. The prisms covered a total visual field of 105% in which each monocular field represented 75°, while the central visual binocular field represented 45°.

During prism exposure, the patient had to execute 80 rapid pointing movements towards visual targets located 10 degrees to the left or to the right of the middle of her/his body, the targets being made to pseudo-randomly alternate. In spite of repeated instructions to carry out rapid movements, the movements produced in brain-damaged patients generally remain too slow as to allow visual retroaction, and the errors committed by our patients did not necessarily reflect the amplitude of optical deviation or phase of adaptation ( Fig. 2 ). However, their degree of rapidity remained compatible with the development of actual sensory-motor adaptation by reducing the strategic components of compensation . The pointing movements were performed with a pause of 30 s after each series of 20, thereby favoring an increased number of errors at the start of the following series. During exposure, the patient did not see the initial position of her/his hand, which entered the visual field only once the movement was approximately 30 to 50% complete , in such a way as to favour proprioceptive-visual coding of the movement . All in all, prism exposure lasted from 6 to 10 mins (video tutorial: http://www.chu-lyon.fr/web/4531 ). While the “control” group patients carried out this visuo-motor task under the same conditions, they were wearing a pair of placebo glasses fitted out with neutral lenses of the same weight consisting in two 5° prismatic lenses set-up so as not to produce any optical deviation ; (OptiquePeter.com). Each patient carried out the exposure task (with prismatic glasses or neutral lenses) 4 times: at D0 (Expo1), at D + 7 (Expo2), at D + 14 (Expo3) and at D + 21 (Expo4). All of the exposure sessions took place under the same conditions and with the same operators.

Prism adaptation phases. A. Pre-test: the subject is comfortably seated in front of a table. A chinrest maintains the trunk in an upright position and prevents the subject from seeing her/his hand as the movement begins; in order to program the movement from proprioceptive to visual coordinates . Pre-tests are carried out so as to provide a baseline for later measurement of compensatory after-effects. In the SN patient, OLP remains correct, but straight-ahead pointing (MSA and VSA) is deviated to the right. B. Active exposure: the subject wears a pair of “point-by-point” prismatic glasses deviating vision on the right side by 10 degrees. Rapid pointing movement in the direction of a visual target is shifted to the side of the optical deviation (red arrow) and towards the virtual target. The motor system can then take into account the spatial error consecutive to prism deviation, regardless of whether the subject shows phenomenological awareness of the error , and finally compensate for the optical deviation. The three examples of error reduction curves correspond to a patient managing to point rapidly enough to generate initially important errors (gray), a patient pointing slowly enough to produce moderate errors (black), and a patient whose pointing variability is increased (broken line). Progressive correction of pointing deviation is observed, with transient increase following breaks. During active exposure to prism, the subject compensates for the optical deviation of vision, but the actual presence of adaptation can only be evidenced by the presence of after-effects . C. After-effects: following removal of the prismatic glasses, when the subject is asked to once again rapidly point towards a target, the movement is shifted in the direction opposed to the optical deviation (leftward: green arrow). This compensatory after-effect is known as total shift, or spatial realignment, and it can be divided into visual and proprioceptive components, both of which are quantified by straight-ahead measurements . The amplitude of total adaptation is quantified in terms of the difference between OLP measurements in post-tests and pre-tests (OLP _post –OLP _pre ). It reveals a compensatory leftward shift when right-shifting prisms are used. The visual adaptation components (VSA _post –VSA _pre ) illustrate a compensatory shift from the side of the optical deviation, which means rightward. The proprioceptive component (MSA _post –MSA _pre ) shows a leftward compensatory shift. OLP: open-loop pointing (i.e. without visual feedback); MSA: manual straight-ahead; VSA: visual straight-ahead

The patients’ perceptual awareness of the optical deviation and its effects on movement trajectories were systematically studied using a phenomenological questionnaire (cf. Appendix A ). This open questionnaire consists in 20 questions divided into 3 main parts and progressing from open-ended formulations to highly specific questions on prism deviation. The first part includes 5 open questions put forward after ten preliminary pointing movements carried out prior to putting on glasses. The questions progress from (Q1: How is the exercise going?) and (Q3: Have you observed anything in particular?) to (Q5: Is it easy to aim toward the target?). The second and main part is administered after 5 movements carried out with glasses on, that is to say during the early exposure period, which usually generates maximal pointing errors . This consisted of 12 questions, the first of which are a reprise of the 5 preceding open questions (Q6 to Q10), while the following queries become increasingly explicit (Q11 to Q17) (Q17: In some patients these glasses can render it difficult to aim with the hand. How about you?). The third and final part of the questionnaire (Q18 to Q20) is given after 20 trials and at the end of active exposure to the prisms.

The after-effects of PA were evaluated by means of repeated MSA measurements in the dark ( n = 10), by VSA in the dark ( n = 10) and by OLP in the direction of a visual target ( n = 10). The VSA and MSA measurements present a double interest in the framework of our study as they are classically used independently to evaluate an egocentric reference and by subtraction of the measurements obtained before and after prism exposure (VS = visual shift and PS = proprioceptive shift) in view of quantifying the after-effects of PA . OLP is used to measure through the same subtraction operation (TS = total shift) the total after-effects of PA . The three sensory-motor parameters were evaluated in pre-test and in post-test at 1 (M1), 3 (M3) and 6 (M6) months in order to monitor the development of each patient’s frames of reference, and also before and after each prism exposure session (Expo1, Expo2, Expo3 and Expo4) in order to quantify the adaptation. To avoid any contamination of the sensory-motor parameters by cognitive parameters involved in compensation for prism deviation occurring independently of adaptation , we carefully ensured that the pre-test and post-test evaluations were not organized under the same conditions as prism exposure. The target used for OLP and the precision/rapidity instructions given differed from those employed during exposure.

MSA measurement was carried out by asking patients to point with their right hand in darkness in the “straight-ahead’ position in the direction of an imaginary line dividing their body into two equivalent halves. The subject was required to employ her/his arm without any speed or amplitude constraint, and if necessary, was reminded that she/he was not expected to proceed rapidly or stretch out her/his arm to the greatest possible extent. Pointing was measured using a contractor attached to a thimble threaded into the index finger, atop a table covered with isoresistant carbon paper on which two 65 × 1 cm electrodes were applied, thereby delimiting a section at an angle of 50° and a depth of 70 cm centered at the starting position near the torso. A direct 5 V current was generated between the electrodes. When the finger touched the surface of the table, tension between the thimble contact point and the reference electrode was measured as in a potentiometer. Tension measurement enabled us to calculate the angular position in relation to the objective sagittal axis, and this position could then be converted into degrees and conventionally signed (negative on the left, positive on the right) ( Fig. 3 A ). Measurement precision was estimated at ± 0.5 degrees.

Quantitative evaluation of prism adaptation. A. Measurement of manual straight-ahead demonstrations: the subject carries out a series of 10 pointing movements in the “straight-ahead” direction with her/his right index finger in darkness. B. Measurement of visual straight-ahead: the subject interrupts the lateral movement of a visual target (10 trials) when the target is perceived in “straight-ahead” position. This measurement is performed in total darkness. C. Measurement of pointing without visual feedback: the subject carries out a series of 10 pointing movements with her/his right hand in the direction of a visual target without seeing her/his hand.

VSA measurement was carried out in total darkness. The patient was comfortably seated in front of a table, with her/his head held straight up by a chinrest . A luminescent red diode was moved by the experimenter onto a 2-m horizontal ramp set face to the patient at a distance of 1 m. Speed of traverse ranged from 20 to 30 cm/s. Ten measures were successively carried out by alternating target movement in the left-right and right-left directions. The patient was asked to vocally interrupt (“Stop!”) the movement of the target as soon as she/he perceived it to be in a “straight-ahead” position ( Fig. 3 B). Measurement of the deviation was performed using a calibrated galvanometer and converted into angular deviation with regard to the objective straight-ahead position.

OLP accuracy measurement was carried out under the same conditions of darkness and with the same set of devices ( Fig. 3 C). The luminous visual target was aligned with the patient’s sagittal axis. The instruction given to the patients was to place their right hand at the target drip-line as precisely as possible but without time constraint, the goal being to distance themselves from the pointing conditions employed during exposure in view of obtaining measurements of sensory-motor after-effects that would be less influenced by cognitive factors .

During prism exposure, the terminal errors of each movement were captured by means of the thimble and converted into degrees of angular error with regard to the target.

2.5

Statistical analysis

First of all, the pre-test results of the two groups were compared. Initial variance analysis with repeated measurements were compared the two groups according to the parameters of age, FIM, BIT and mean time lapse after stroke. Sensory-motor parameters were compared by means of a repeated measures (RM) Anova including the “Session” factor since two pre-tests were available. Reliability of the sensory-motor measurements was evaluated using correlations.

Evaluation of the sensory-motor effects proceeded using two main steps. The first tests compared the different after-effects measured in terms of post-pre difference using unilateral Student t -tests against a theoretical value of zero. A three-factor Anova with repeated measurements then compared the two treatments (group factor, inter-subject: prism and control) with two intra-subject factors: pre-post (measured before and after exposure) and session (Expo 1 to 4). The error reduction curves produced during exposure could not be interpreted, due to highly variable and relatively slow movement speed.

In order to study the long-term impact of the 4 PA sessions on the sensory-motor variables, repeated measures analysis of variance was carried out with the “session” (pre, M1, M3, M6 or pre and M6), and “group” (prism, control) factors for each functional parameter under examination.

To conclude, the results enabled us to explore the predictive value of the sensory-motor parameters on the BIT and FIM scores, along with evolution between the pre-tests and M6.

2.1

Patients

Nineteen patients admitted to the neurological rehabilitation medicine unit of Hôpital Henry-Gabrielle, Hospices Civils de Lyon and presenting with left SN secondary to a right-hemisphere stroke were included in the study from September 2001 until September 2005 ( Fig. 1 ). All patients were right-handed according to the criteria of the Edinburgh questionnaire .

Study design. Mild neglect: BIT score > 55; Severe neglect: BIT score ≤ 55.

The inclusion criteria were:

• •
  

  age ranging from 18 to 90 years;

  
  
• •
  

  a single stroke confirmed by a tomodensitometry examination or by brain MRI;

  
  
• •
  

  left SN confirmed by several neuropsychological tests (line bisection test , balloon test , copy of a drawing, dictation and reading of a text);

  
  
• •
  

  a time lapse of at least one month following the ischemic event.

The exclusion criteria were:

• •
  

  existence of multiple brain lesions;

  
  
• •
  

  temporo-spatial disorientation;

  
  
• •
  

  psychiatric disorders;

  
  
• •
  

  an associated, non-stabilized pathology.

The characteristics of the patients in the two groups are detailed in Table 1 .

Table 1

Patient characteristics.

Case	N	Age/sex	LI	MD	SD	LHH	OCD	Delay	Aetiology	Lesion
Prism group
1	+	40/F	100	3	3	P	1	44	Isch	(Frontal), parietal, occipital, temporal, insula, corona radiata, putamen
7	++	40/F	100	3	3	P	2	47	Isch	Frontal, parietal, temporal, insula, corona radiata, putamen
8	+	47/F	100	3	0	P	1	34	Isch	Frontal white matter, (corona radiata), insula, internal capsule, putamen, caudate nucleus
9	+	69/M	100	3	3	P	0	54	Isch	Temporal, (occipital), corona radiata, internal capsule, putamen
10	++	66/M	16.66	3	3	A	2	60	Isch	Frontal, parietal (temporal), corona radiata, putamen
11	+	59/M	100	2	2	P	1	88	Hem	Parietal, occipital
13	++	49/M	100	3	3	P	1	30	Isch	Frontal (temporal, parietal), putamen
15	+	63/F	100	3	3	P	1	42	Isch	Corona radiata, capsule interne, putamen
16	++	71/M	100	3	3	P	1	60	Isch	Frontal, parietal, corona radiata, putamen
Control group
2	+	45/M	83.33	3	3	A	1	35	Isch	Frontal white matter, corona radiata, insula, internal capsule, putamen
3	++	45/M	100	3	3	A	1	92	Isch	Frontal white matter, corona radiata, insula, putamen, caudate nucleus
4	++	57/M	100	3	3	P	1	38	Isch	Frontal, temporal, parietal, insula, putamen
5	+	72/M	100	3	3	A	1	60	Hem	Corona radiata, insula, internal capsule, putamen
6	+	62/F	91.66	3	3	P	1	46	Isch	Parietal, occipital
12	+	79/F	100	2	2	P	1	38	Hem	Parietal, occipital
14	++	51/M	100	3	3	P	2	67	Isch	Temporal, parietal, occipital
17	+	75/F	100	1	1	P	1	34	Isch	Occipital, parietal white matter, putamen, caudate nucleus
18	++	69/F	100	3	2	P	2	60	Isch	Frontal, temporal, parietal, corona radiata, putamen, caudate nucleus

N: neglect (+ = mild neglect; ++ = severe neglect); LI: Laterality Index (Edinburgh Test ); MD: motor deficit (0 = absent; 1 = monoparesis; 2 = incomplete hemiparesis; 3 = complete); SD: somatosensory deficit (0 = absent; 1 = superficial; 2 = incomplete superficial and deep; 3 = complete); LHH: left homonymous hemianopia (A = absent; P = present); OCD: right ocular and cephalic deviation (0 = no deviation; 1 = spontaneously reducible; 2 = reducible under order; 3 = not reducible); Etiology (Isch = ischemia; Hem = hemorrhage); Lesions: brackets indicate partial involvement of cerebral lobe.

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Prism adaptation and unilateral neglect: Beyond an apparent negative result Effets sensori-moteurs et fonctionnels à long terme d’un traitement hebdomadaire par adaptation prismatique dans la négligence : un essai randomisé et contrôlé en double insu Comparison between tibial nerve block with anaesthetics and neurotomy in hemiplegic adults with spastic equinovarus foot Effect of table top slope and height on body posture and muscular activity pattern What are the perspectives for ventilated tetraplegics? A French retrospective study of 108 patients with cervical spinal cord injury Ulnar nerve lesion at the wrist and sport: A report of 8 cases compared with 45 non-sport cases

Stay updated, free articles. Join our Telegram channel

Join