Abstract
Objective
The variety and extent of impairments occurring after traumatic brain injury vary according to the nature and severity of the lesions. In order to better understand their interactions and long-term outcome, we have studied and compared the cognitive and neurobehavioral profile one year post onset of patients with and without traumatic brain injury in a cohort of motor vehicle accident victims.
Method
The study population is composed of 207 seriously injured persons from the ESPARR cohort. This cohort, which has been followed up in time, consists in 1168 motor vehicle accident victims (aged 16 years or more) with injuries with all degrees of severity. Inclusion criteria were: living in Rhone county, victim of a traffic accident having involved at least one wheel-conducted vehicle and having occurred in Rhone county, alive at the time of arrival in hospital and having presented in one of the different ER facilities of the county. The cohort’s representativeness regarding social and geographic criteria and the specificities of the accidents were ensured by the specific targeting of recruitment. Deficits and impairments were assessed one year after the accident using the Neurobehavioral Rating Scale – Revised and the Trail-Making Test. Within our seriously injured group, based on the Glasgow Score, the presence of neurological deficits, aggravation of neurological condition in the first 72 hours and/or abnormal cerebral imaging, we identified three categories: (i) moderate/severe traumatic brain injury ( n = 48), (ii) mild traumatic brain injury ( n = 89), and (iii) severely injured but without traumatic brain injury ( n = 70).
Results
The most frequently observed symptoms were anxiety, irritability, memory and attention impairments, depressive mood and emotional lability. While depressive mood and irritability were observed with similar frequency in all three groups, memory and attention impairments, anxiety and reduced initiative were more specific to traumatic brain injury whereas executive disorders were associated with moderate/severe traumatic brain injury.
Discussion-Conclusion
The presence and the initial severity of a traumatic brain injury condition the nature and frequency of residual effects after one year. Some impairments such as irritability, which is generally associated with traumatic brain injury, do not appear to be specific to this population, nor does depressive mood. Substantial interactions between cognitive, affective and neurobehavioral disorders have been highlighted.
Résumé
Objectif
La diversité et l’étendue des déficits observés après un traumatisme crânien varient suivant la nature des lésions et leur gravité. Afin de mieux en comprendre les interactions et leurs conséquences à long terme, nous avons étudié et comparé le profil cognitif et neuro-comportemental d’une cohorte de victimes d’accidents de la voie publique avec et sans traumatisme crânien, un an après l’accident.
Méthode
La population d’étude est constituée de 207 blessés graves de la cohorte ESPARR. Cette cohorte, qui a été suivie dans le temps, est composée de 1168 accidentés de la route dans le Rhône âgés de 16 ans et plus et de toutes gravités, inclus selon certains critères : résidant dans le Rhône, victime d’un accident de la circulation impliquant au moins un véhicule à roues, ayant eu lieu dans le Rhône, vivant au moment de son arrivée à l’hôpital, et se présentant dans l’un des différents services d’urgences du Rhône. La représentativité de cette cohorte en termes de critères sociogéographiques et de caractéristiques de l’accident a été assurée par la nature spécifique du recrutement. Les plaintes et déficiences cognitivo-comportementales ont été évaluées un an après l’accident avec la Neurobehavioral Rating Scale-Revised et le Trail Making Test. Nous avons distingué, dans notre population d’étude de blessés graves, trois catégories de patients : les traumatisés crâniens modérés/graves ( n = 48), les traumatisés crâniens légers ( n = 89) et les blessés graves sans traumatisme crânien ( n = 70), basées sur le score de Glasgow, la présence de déficits neurologiques, d’une aggravation de l’état neurologique dans les 72 heures et/ou d’une imagerie cérébrale anormale.
Résultats
Les symptômes les plus fréquemment observés étaient l’anxiété, l’irritabilité, les troubles mnésiques et attentionnels, la dépression et la labilité de l’humeur. L’humeur dépressive et l’irritabilité sont relevées aussi fréquemment dans les trois groupes. Les troubles de mémoire, d’attention, l’anxiété et la diminution des initiatives sont plus spécifiques au traumatisme crânien et les troubles exécutifs sont typiques d’un traumatisme crânien modéré/grave.
Discussion-Conclusion
La présence et la sévérité initiale du TC conditionnent la nature et la fréquence des séquelles à un an d’évolution. Certains troubles comme l’irritabilité, habituellement associée à un traumatisme crânien, n’apparaissent pas comme spécifiques à cette population, tout comme l’humeur dépressive. Ainsi, les interactions entre troubles cognitifs, émotivo-affectifs et neuro-comportementaux sont mises en avant.
1
English version
1.1
Introduction
Traffic accidents remain a major public health problem in France; in 2005, 5543 victims died, and of the 111,683 additional persons suffering injury, more than 40,000 were hospitalized . The consequences of road accidents are wide-ranging and likely to entail numerous sequelae. Prognoses pose difficulties in assessment, tend to be varied and are subject to underestimation , especially among the most mildly or moderately injured.
Since 1995, a Registre des Victimes des Accidents Corporels de la Circulation Routière (registry of motor accident victims) has been used in the French Rhône county; its purpose is not simply to enumerate accidents, but also to provide precise indications on the victims and their injuries . In 2005, the Rhône county registry comprised 8608 persons, of whom 1453 presented at least one brain lesion, whatever its gravity, and of whom 88 had died. Over recent years, a nationwide decrease of road accidents has been replicated on a county-wide scale; in 2008, for example, only 7460 traffic accident victims, 13% fewer than in 2005 were mentioned in the registry; 1326 had suffered some kind of brain lesion, while 57 had lost their lives. Recent studies based on the registry have shown that the number of severely injured persons with sequelae was roughly equivalent to the number of fatalities .
Traumatic brain injury (TBI) is particularly important on account of its frequency (it is ranked second following cervical injuries and trauma) and the severity of its sequelae .
In fact, independently of cause, long-term prognosis is closely associated with cognitive and behavioral impairments. In a recent review of the literature, Dikmen et al. confirmed the importance of cognitive sequelae following moderate-to-severe brain injury, whereas the sequelae are hardly obvious subsequent to mild TBI. Even though many relevant studies have presented a variety of methodological biases (no control group, heterogeneous etiology, lack of representativeness with regard to a reference population), they clearly underscore typical cognitive sequelae including memory or attention disorders as well as slowed handling of information , while from a behavioral standpoint, the most frequently reported abnormalities are irritability , impulsiveness , disinhibition and loss of initiative . In some patients, the psychic consequences are of paramount importance: altered personality and/or thymus disorders (depression, anxiety, emotional numbing) . They are often associated with ansognosia, in which patients are not always fully aware of the seriousness and the repercussions of their dysfunctions . Taken as a whole, these disorders constitute what is sometimes termed the “invisible disability” of brain trauma. Along with the personality alterations, troubles related to behavior and mood represent a major complaint of close friends and family, and this continues to be the case long after the accident . All in all, the multiplicity of variables in highly diversified anatomo-clinical and social situations strongly underscores the indispensability of a rigorous approach. Dikmen et al. , for example, have insisted on the need for control groups tailored to the situations being explored; they have also underlined the usefulness of longitudinal follow-up.
This study is aimed at indicating the precise characteristics specific to brain trauma that have been found among the cognitive and behavioral manifestations observed in the 207 seriously injured persons forming a cohort of Rhône county road accident victims followed from October 2004 until June 2006 and assessed one-year post onset. By comparing three groups of victims, we shall determine the nature of attendant complaints in accordance with the presence or absence of brain trauma; initial injury severity shall likewise be taken into close account.
1.2
Material and methods
1.2.1
The study population
The ESPARR (étude de suivi d’une population d’accidentés de la route dans le Rhône) cohort is a prospective cohort consisting in Rhône county traffic accident casualties. Recruitment was carried out from 1 October 2004 through 30 June 2006 in all the public and private hospitals of Rhône county. Any person fulfilling the inclusion criteria (living in Rhône county, victim of a traffic accident involving at least one wheel-conducted vehicle and having occurred in Rhône county, alive at the time of arrival in hospital and having presented in one of the different ER facilities of the county) was invited to participate in the ESPAAR study. The most severely injured patients were recruited directly in hospital wards throughout the study period. For the most serious cases (intensive care patients), relations with close family and/or friends were handled by the intensive or critical care physician. Recruitment structuring (time frames extended through 24 hours, 7 days a week) ensured the cohort’s representativeness regarding social and geographic criteria as well as the specificities of the accident: season, day of the week, hour, type of accident (work-related, during a journey), means of transportation being used at the time . All in all, 1372 subjects including 1168 adults (at least 16 years of age) agreed to be followed-up in the ESPARR cohort. Medical data were collected from hospital records and the declaration forms of the Registre des Victimes d’accidents de la circulation du Rhône (Rhône county registry of motor accident victims). The injuries were coded by a registry physician with experience in use of the Abbreviated Injury Scale (AIS) , a coding system through which lesions are assigned severity scores ranging from 1 (minor injury) to 6 (lethal injury). The M-AIS (Maximum Abbreviated Injury Score) is equal to the AIS of the most serious injury and indicates the initial level of overall severity; the injury is considered as minor when the M-AIS is 1, moderate when it is 2, and serious when it is at least equal to 3 (M-AIS3 + ).
The Marshall et al. score was calculated for all the subjects having undergone medical imaging in view of classifying the brain injuries according to degree of severity (grade 1: no visible injury to grade 4: midline shift).
Within the cohort, the population having been singled out specifically for the study corresponded to subjects of at least 16 years of age at the time of the accident and considered at that time as seriously injured (M-AIS ≥ 3), whether or not they had incurred traumatic brain injury. Three groups of patients were constituted:
- •
the “moderate/severe TBI” group was defined by one of the following criteria: Glasgow score < 13, neurological deficit, aggravated neurological status during the 72 hours after the accident, abnormal brain imagery;
- •
the “mild TBI” group was defined by a Glasgow score ≥ 13, with neither aggravation over 72 hours nor abnormal neurological signs on clinical examination. In this group, we also included patients with no documented Glasgow score but presenting brain trauma with loss of consciousness for less than an hour, as well as patients without consciousness disorder but with facial trauma, given that by definition, facial trauma is associated with brain trauma ;
- •
The “seriously injured non-TBI” group without TBI was defined by M-AIS ≥ 3 but without head injury.
At the initial stage, immediately subsequent to the accident, each patient (or, if necessary, a close relation) was asked, during an interview at the hospital with an investigator, to answer questions in a questionnaire comprising the following items: sex, age at the time of accident, means of transportation (motorcycle, bicycle…), family and socio-professional status, level of studies.
A year after the accident, each patient was interviewed for a second time and given a neuropsychological test and a questionnaire, which were administered by two specifically trained neuropsychologists. The evaluation comprised:
- •
the Neurobehavioral Rating Scale – revised: this multidimensional 29-item scale has been revised and validated with American and French populations; all of the items have been simultaneously defined in both English and French. As an integrated tool that can be rapidly applied with minimal constraint, it is particularly suited for epidemiological studies. It consists in a semi-structured interview and succinct cognitive tests. The different items are designed to evaluate not only cognitive aspects such as problems pertaining to memory, attention and planning ahead, but also behavioral and thymus-related items involving anxiety, depression, mood lability, etc. Susceptibility to fatigue is likewise taken into account. Disorders or impairments are classified according to 4 categories: (i) none, (ii) mild, (iii) moderate or (iv) severe. Given the number of subjects participating in our study, each NRS item was considered as either present or absent;
- •
the Trail Making Test is composed of two parts. In Part A, the subject is asked to connect as rapidly as possible, in ascending order, a set of 25 randomly distributed dots; in Part B, the subject is once again asked to connect dots, as quickly as possible and in ascending numerical and alphabetical order, through systematic alternation of a number and a letter (1-A-2-B-3-C…). This test provides recognized measurement of mental flexibility or attention-switching capability ;
- •
a questionnaire bringing together the socio-demographic data (sex, age, return home and return to work) collected at the time of inclusion. Post-traumatic stress syndrome (PTSS) is assessed in accordance with the Post-traumatic Stress Disorder Checklist Scale . A threshold score ≥ 44 provides a likely indication of PTSS with significant repercussions in daily life .
The one-year face-to-face interviews took place at patients’ homes (72.5%), during a hospital consultation (18.4%), in a medicine faculty office (7.8%) or at their workplaces (1.4%). The median time interval between the accident and the interviews was 437 days (interquartile interval Q1–Q3 = [396–498]).
1.2.2
Ethical considerations
Informed written consent was secured from each patient, or by a close relation if the patient was unable to express consent.
1.2.3
Analysis
The 3 groups were initially compared on an overall basis, after which each of the two TBI groups was compared to the group of seriously injured patients without head injury, which was considered as a control group. Lastly, the two TBI groups were compared with each other so as to determine whether or not there existed a difference in degree of injury. Chi-square tests (or Fisher tests when the number of subjects was too small) were carried out for the categorical variables, and Student’s t -tests (or analysis of variance tests for general variance) were performed for the quantitative variables. When overall test significance failed to reach the threshold of 5%, inter-group comparisons were not performed.
Modified Poisson regression models were constructed to determine whether or not associations between the different symptoms and the fact of having suffered TBI persisted once adjustments had been made for age, sex and level of studies (Chi-square tests). Resumption or non-resumption of professional activities one year after the accident was integrated to the multivariate model when analysis using a univariate model was significant at 10%. The only symptoms modeled were those present in at least 20% of the population and for which comparison between the moderate/severe TBI and the “control” population presented a significant difference following verification of the overall or omnibus test.
Statistical analyses were carried out through SAS 9.3 software (SAS Institute Inc., Cary, NC, USA). The genmod procedure was employed for construction of modified Poisson regression models.
1.3
Results
Out of the 1168 adults (aged at least 16 years) of the ESPARR cohort, 324 subjects fulfilled the previously defined criteria (M-AIS ≥ 3). Among those persons, 68 patients agreed to reply to the questionnaire but refused the face-to-face interview and failed to take the tests; 20 did not reply for the one-year follow-up but participated in subsequent follow-up; 15 refused to carry on with their participation in the study, while 12 others were lost to follow-up, and 1 subject died. All in all, 116 injured persons did not take part in the neuropsychological tests one year after the accident, and the neuropsychological follow-up was consequently performed for 208 injured persons, which represents a participation rate of 64%.
One subject presenting with mild TBI was excluded from the analyses because he had suffered hypovolemic shock that raised his Glasgow score to 9. As a result, our study population consisted of 207 adult subjects: 48 in the moderate/severe TBI group, 89 in the mild TBI group, and 70 in the severely injured non-TBI group.
With regards to the age and sex criteria and the degree of brain injury severity, the seriously injured subjects who did not participate in the one-year neuropsychological follow-up did not differ from those who did participate.
1.3.1
Description of the sample
The different medical and socio-demographic data of the sample have been detailed in Table 1 .
| Brain trauma | Severe injuries without brain trauma | Chi-squared or Student’s t -test | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Moderate/Severe (A) | Mild (B) | (C) | ||||||||
| n = 48 | n = 89 | n = 70 | ||||||||
| n | % | n | % | n | % | (A) vs (B) vs (C) | (A) vs (B) | (A) vs (C) | (B) vs (C) | |
| Age at time of accident | NS | – | – | – | ||||||
| [16–20[years | 15 | 31.30% | 14 | 15.70% | 7 | 10.00% | ||||
| [20–30[years | 11 | 22.90% | 24 | 27.00% | 17 | 24.30% | ||||
| [30–40[years | 5 | 10.40% | 20 | 22.50% | 9 | 12.90% | ||||
| [40–50[years | 7 | 14.60% | 14 | 15.70% | 16 | 22.90% | ||||
| [50–60[years | 5 | 10.40% | 9 | 10.10% | 11 | 15.70% | ||||
| ≥ 60 years | 5 | 10.40% | 8 | 9.00% | 10 | 14.30% | ||||
| Average age [mean(SD)] | 34.4 | (16.5) | 35.4 | (15.3) | 41.2 | (17.7) | P < 0.05 a | NS | P < 0.05 | P < 0.05 |
| Sex | NS | – | – | – | ||||||
| Male | 39 | 81.30% | 69 | 77.50% | 49 | 70.00% | ||||
| Female | 9 | 18.80% | 20 | 22.50% | 21 | 30.00% | ||||
| Family status | NS | – | – | – | ||||||
| Single | 25 | 52.10% | 41 | 46.10% | 23 | 32.90% | ||||
| In couple | 16 | 33.30% | 38 | 42.70% | 36 | 51.40% | ||||
| Divorced/separated/widow(er) | 17 | 35.40% | 10 | 11.20% | 11 | 15.70% | ||||
| Type of user | P < 0.05 | NS | P < 0.05 | P < 0.01 | ||||||
| Pedestrian/roller skate/scooter | 8 | 16.70% | 11 | 12.40% | 16 | 22.90% | ||||
| 4 wheels | 21 | 43.80% | 37 | 41.60% | 13 | 18.60% | ||||
| 2 motorized wheels | 14 | 29.20% | 34 | 38.20% | 29 | 41.40% | ||||
| Bicycle | 5 | 10.40% | 7 | 7.90% | 12 | 17.10% | ||||
| Study level | NS c | – | – | – | ||||||
| < baccalaureate | 30 | 62.50% | 55 | 61.80% | 45 | 64.30% | (NS b ) | |||
| Baccalaureate | 4 | 8.30% | 14 | 15.70% | 6 | 8.60% | ||||
| > baccalaureate | 11 | 22.90% | 18 | 20.20% | 18 | 25.70% | ||||
| Other | 3 | 6.30% | 2 | 2.20% | 1 | 1.40% | ||||
| Socio-professional categories d | NS | – | – | – | ||||||
| Student | 12 | 25.00% | 12 | 13.50% | 10 | 14.30% | ||||
| Farmer/artisan | 5 | 10.40% | 4 | 4.50% | 7 | 10.00% | ||||
| Executive | 6 | 12.50% | 8 | 9.00% | 6 | 8.60% | ||||
| Intermediate profession | 4 | 8.30% | 6 | 6.70% | 8 | 11.40% | ||||
| Employee | 18 | 37.50% | 40 | 44.90% | 27 | 38.60% | ||||
| Worker | 3 | 6.30% | 19 | 21.30% | 12 | 17.10% | ||||
| Initial M-AIS | P < 0.0001 c | P < 0.0001 | P < 0.0001 | NS c | ||||||
| 3 | 14 | 29.20% | 76 | 85.40% | 64 | 91.40% | ||||
| 4 | 22 | 45.80% | 13 | 14.60% | 4 | 5.70% | ||||
| 5 | 12 | 25.00% | 0 | 0.00% | 2 | 2.90% | ||||
| Initial GCS | P < 0.0001 c | P < 0.0001 | P < 0.0001 | P < 0.0001 | ||||||
| [13–15] | 23 | 47.90% | 72 | 80.90% | 8 | 11.40% | ||||
| [9–12] | 9 | 18.80% | 0 | 0.00% | 0 | 0.00% | ||||
| [3–8] | 16 | 33.30% | 0 | 0.00% | 0 | 0.00% | ||||
| Non réalisé | 0 | 0.00% | 17 | 19.10% | 62 | 88.60% | ||||
| Marshall score | – | – | – | – | ||||||
| No Scanner/MRI (head) | 0 | 0.00% | 33 | 37.10% | 66 | 94.30% | ||||
| Grade I diffuse injury (neither effusion, nor edema) | 4 | 8.30% | 56 | 62.90% | 4 | 5.70% | ||||
| Grade II diffuse injury | 33 | 68.80% | 0 | 0.00% | 0 | 0.00% | ||||
| Grade III diffuse injury | 3 | 6.30% | 0 | 0.00% | 0 | 0.00% | ||||
| Grade IV diffuse injury | 0 | 0.00% | 0 | 0.00% | 0 | 0.00% | ||||
| Massive evacuated injury | 1 | 2.10% | 0 | 0.00% | 0 | 0.00% | ||||
| Massive non-evacuated injury | 7 | 14.60% | 0 | 0.00% | 0 | 0.00% | ||||
| Back to work at 1-year questionnaire | NS | – | – | – | ||||||
| No | 15 | 31.30% | 20 | 22.50% | 16 | 22.90% | (NS b ) | |||
| Yes | 15 | 31.30% | 43 | 48.30% | 31 | 44.30% | ||||
| Not relevant (student/retired/home-based) | 18 | 37.50% | 26 | 29.20% | 23 | 32.90% | ||||
| Presence of PTSD signs at 1 year | NS c | – | – | – | ||||||
| No | 30 | 62.50% | 66 | 74.20% | 57 | 81.40% | ||||
| Yes | 15 | 31.30% | 21 | 23.60% | 10 | 14.30% | ||||
| Undetermined | 3 | 6.30% | 2 | 2.20% | 3 | 4.30% | ||||
a Anova test – variance analysis.
b Test excluding the last category.
d Retirees and persons on parental leave are included in the socio-professional group corresponding to their most recent employment.
As regards age, the mild TBI subjects (36 years old on average) and the moderate/severe TBI subjects (34 years old on average) were significantly younger than the severely injured non-TBI subjects (41 years old on average). On the other hand, age-class distribution showed no difference between the 3 groups.
The patients with brain injury in both the moderate/severe TBI and the mild TBI groups were mainly users of 4-wheel vehicles, while the severely injured non-TBI subjects comprised a majority of users of 2-wheel motor vehicles.
If 48% of the moderate/severe TBI patients had an initial Glasgow score ranging from 13 to 15, it was due either to aggravation of the latter over the first 72 hours after the accident, or to a lesion observed in initial brain imaging (92% of the patients in this group had a lesion and a Marshall score equal to or higher than 2). In the mild TBI group, 3 subjects had a Glasgow score of 13, 13 had a Glasgow score of 14 and 56 subjects had a Glasgow score of 15. In this group, brain imagery was considered as normal at the time it was carried out ( n = 56).
The other variables taken into consideration (sex, family and socio-professional status, level of studies) did not significantly differ from one group to another (comprehensive statistical test).
At one year, no significant difference between the three groups was observed with regards to resumption of professional activities, which had occurred in 31.3% of the moderate/severe TBI group, in 48.3% of the mild TBI group and in 44.3% of the severely injured non-TBI group. Nor was there any significant difference between the three groups with regards to the proportion of persons presenting signs of PTSD, which affected 31.3% of the moderate/severe TBI group, 23.6% of the mild TBI group and 14.3% of the severely injured non-TBI group.
1.3.2
NRS-R at one year – description of the symptoms
In the total population of the three groups ( n = 207), the most frequently observed symptoms (> 15%) were: anxiety (52.2%), irritability (38.6%), memory disorders (38.6%) depressive mood (30.4%), attention disorders (28.0%), mood lability (26.6%) and guilt feelings (16.4%) ( Table 2 ).
| TBI moderate/severe (A) n = 48 | TBI mild (B) n = 89 | Seriously injured (C) n = 70 | Total n = 207 | Chi-squared test | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | n | % | (A) vs (B) vs (C) | (A) vs (B) | (A) vs (C) | (B) vs (C) | |
| Anxiety | 31 | 64.60 | 51 | 57.30 | 26 | 37.10 | 108 | 52.20 | P < 0.01 | NS | P < 0.01 | P < 0.05 |
| Memory disorders | 29 | 60.40 | 39 | 43.80 | 12 | 17.10 | 80 | 38.60 | P < 0.0001 | NS | P < 0.0001 | P < 0.001 |
| Irritability | 21 | 43.80 | 38 | 42.70 | 21 | 30.00 | 80 | 38.60 | NS | |||
| Depressive mood | 17 | 35.40 | 28 | 31.50 | 18 | 25.7 | 63 | 30.40 | NS | |||
| Attention disorders | 29 | 60.40 | 21 | 23.60 | 8 | 11.40 | 58 | 28.00 | P < 0.001 | P < 0.0001 | P < 0.0001 | P < 0.05 |
| Mood lability | 20 | 41.70 | 24 | 27.00 | 11 | 15.70 | 55 | 26.60 | P < 0.01 | NS | P < 0.01 | NS |
| Guilt feelings | 9 | 18.80 | 15 | 16.90 | 10 | 14.30 | 34 | 16.40 | NS | |||
| Diminished initiative and motivation | 13 | 27.10 | 12 | 13.50 | 3 | 4.30 | 28 | 13.50 | P < 0.01 | P < 0.05 | P < 0.001 | P < 0.05 |
| Diminished activity | 14 | 29.20 | 9 | 10.10 | 4 | 5.70 | 27 | 13.00 | P < 0.001 | P < 0.01 | P < 0.001 | NS |
| Withdrawal from society | 11 | 22.90 | 11 | 12.40 | 5 | 7.10 | 27 | 13.00 | P < 0.05 | NS | P < 0.05 | NS |
| Conceptual disorganization | 9 | 18.80 | 8 | 9.00 | 5 | 7.10 | 22 | 10.60 | NS | |||
| Lower cognitive flexibility | 12 | 25.00 | 4 | 4.50 | 3 | 4.30 | 19 | 9.20 | P < 0.0001 | P < 0.001 | P < 0.001 | NS a |
| Difficulties in oral expression | 11 | 22.90 | 5 | 5.60 | 1 | 1.40 | 17 | 8.20 | P < 0.0001 | P < 0.01 | P < 0.0001 a | NS a |
| Susceptibility to mental fatigue | 8 | 16.70 | 3 | 3.40 | 1 | 1.40 | 12 | 5.80 | P < 0.01 a | P < 0.05 a | P < 0.01 a | NS a |
| Decreased planning ability | 8 | 16.70 | 3 | 3.40 | 0 | 0.00 | 11 | 5.30 | P < 0.0001 a | p < 0.05 a | P < 0.0001 a | NS a |
| Disorientation | 8 | 16.70 | 2 | 2.20 | 1 | 1.40 | 11 | 5.30 | P < 0.0001 a | P < 0.01 a | P < 0.01 a | NS a |
| Difficulties in oral comprehension | 6 | 12.50 | 3 | 3.40 | 1 | 1.40 | 10 | 4.80 | P < 0.05 a | NS a | P < 0.05 a | NS a |
| Slower speed and stride | 8 | 16.70 | 2 | 2.20 | 0 | 0.00 | 10 | 4.80 | P < 0.0001 a | P < 0.01 a | P < 0.0001 a | NS a |
| Exaggerated somatic preoccupations | 3 | 6.30 | 3 | 3.40 | 3 | 4.30 | 9 | 4.30 | NS a | |||
| Unusual thought contents | 4 | 8.30 | 3 | 3.40 | 2 | 2.90 | 9 | 4.30 | NS a | |||
| Limited self-criticism | 6 | 12.50 | 0 | 0.00 | 1 | 1.40 | 7 | 3.40 | P < 0.0001 a | P < 0.01 a | P < 0.05 a | NS a |
| Mistrust | 3 | 6.30 | 2 | 2.20 | 2 | 2.90 | 7 | 3.40 | NS a | |||
| Disinhibition | 3 | 6.30 | 1 | 1.10 | 1 | 1.40 | 5 | 2.40 | NS a | |||
| Diminished vigilance | 4 | 8.30 | 0 | 0.00 | 0 | 0.00 | 4 | 1.90 | P < 0.01 a | P < 0.05 a | P < 0.05 a | – |
| Hallucinations | 3 | 6.30 | 0 | 0.00 | 0 | 0.00 | 3 | 1.40 | P < 0.05 a | P < 0.05 a | NS a | – |
| Excitation | 0 | 0.00 | 1 | 1.10 | 2 | 2.90 | 3 | 1.40 | NS a | |||
| Hostility | 1 | 2.10 | 1 | 1.10 | 1 | 1.40 | 3 | 1.40 | NS a | |||
| Hyperactivity, agitation | 0 | 0.00 | 1 | 1.10 | 2 | 2.90 | 3 | 1.40 | NS a | |||
| Articulation disorders | 2 | 4.20 | 0 | 0.00 | 0 | 0.00 | 2 | 1.00 | NS a | |||
Irritability and depressive mood were as frequent in seriously injured non-TBI as in TBI subjects. Whatever the group under consideration, signs of depression and anxiety were associated at one year with post-traumatic stress ( P < 0.01).
Anxiety, memory and attention disorders as well as diminished initiative and motivation were significantly more frequently found in the two TBI groups than in the non-TBI group. As regards attention disorders, they were significantly more prevalent in the moderate/severe TBI group than in the mild TBI group and significantly less prevalent in the severely injured non-TBI group than in the other two. This was also the case for mood lability, even though only between the moderate/severe TBI group and the non-TBI group was the difference significant.
The moderate/severe TBI groups significantly differed from the other two groups with regards to the following items: decreased planning ability, lower cognitive flexibility, diminished affectivity, limited capacity for self-criticism, slower speed and stride, susceptibility to mental fatigue, difficulties in oral expression, disorientation, diminished vigilance and social withdrawal. Only in the moderate/severe TBI group were hallucinations reported ( n = 3).
Given the relatively small number of subjects participating in the study, there was no significant difference between the three groups for the following symptoms: disinhibition, hyperactivity, exaggerated somatic preoccupations or mistrust; this was likewise the case for signs of conceptual disorganization, even though its occurrence was somewhat more frequent in moderate/severe TBI subjects.
1.3.3
NRS-R at one year–modified Poisson regression model
Following adjustment for age, sex and level of studies, the moderate/severe TBI subjects were more at risk than the seriously injured non-TBI subjects of presenting attention and memory disorders, anxiety, mood lability, diminished affectivity, lowered cognitive flexibility, difficulties in oral expression and loss of initiative and motivation ( Table 3 ). In addition to anxiety, the memory and attention disorders were associated with mild brain trauma.
| TBI moderate/severe vs controls | TBI mild vs controls | |||||||
|---|---|---|---|---|---|---|---|---|
| RR | CI95% | P | RR | CI95% | P | |||
| Attention disorders | 1.59 | 1.38 | 1.83 | P < 0.0001 | 1.16 | 1.03 | 1.29 | P < 0.05 |
| Memory disorders | 1.6 | 1.37 | 1.87 | P < 0.0001 | 1.26 | 1.12 | 1.42 | P < 0.001 |
| Anxiety | 1.25 | 1.05 | 1.48 | P < 0.05 | 1.25 | 1.08 | 1.44 | P < 0.05 |
| Difficulties in oral expression | 1.26 | 1.12 | 1.41 | P < 0.0001 | – | – | – | NS |
| Mood lability | 1.2 | 1.03 | 1.39 | P < 0.05 | – | – | – | NS |
| Diminished affectivity | 1.27 | 1.11 | 1.45 | P < 0.001 | – | – | – | NS |
| Social withdrawal | 1.2 | 1.04 | 1.37 | P < 0.05 | – | – | – | NS |
| Lower cognitive flexibility | 1.28 | 1.11 | 1.48 | P < 0.001 | – | – | – | NS |
| Loss of initiative and motivation | 1.29 | 1.1 | 1.51 | P < 0.01 | – | – | – | NS |
1.3.4
Trail Making Test after one year
No significant difference between the groups was observed in the results of the Trail Making Test ( Table 4 ).
| TBI moderate/severe (A) n = 48 | TBI mild (B) n = 89 | Seriously injured (C) n = 70 | Chi-squared test or Anova | ||||
|---|---|---|---|---|---|---|---|
| n | % or SD | n | % or SD | n | % or SD | (A) vs (B) vs (C) | |
| Trail Making Test A | NS b | ||||||
| No error | 46 | 95.80% | 89 | 100.00% | 70 | 100.00% | |
| Existence of error(s) | 2 | 4.20% | 0 | 0.00% | 0 | 0.00% | |
| Trail Making Test B | NS b | ||||||
| No error | 36 | 75.00% | 66 | 74.20% | 57 | 81.40% | (NS a ) |
| Existence of error(s) | 9 | 18.80% | 21 | 23.60% | 13 | 18.60% | |
| Inability to take the test c | 3 | 6.30% | 2 | 2.20% | 0 | 0,0% | |
| Trail Making Test A time (seconds) | NS | ||||||
| Mean | 33.2 | (15.3) | 28.8 | (15.6) | 32.2 | (14.4) | |
| 1st quartile | 22.5 | 21 | 23 | ||||
| Median | 29.5 | 25 | 29 | ||||
| 3rd quartile | 36.5 | 31 | 37 | ||||
| Trail Making Test B time (seconds) | NS | ||||||
| Mean | 82.3 | (42.1) | 75.6 | (51.4) | 83.5 | (57.5) | |
| 1st quartile | 55 | 56 | 58 | ||||
| Median | 71 | 63 | 69 | ||||
| 3rd quartile | 98 | 80 | 88 | ||||
| No response | 3 | 2 | 0 | ||||
| Time difference between TMTB and TMTA (seconds) | NS | ||||||
| Mean | 50.8 | (35) | 47.2 | (38.1) | 51.3 | (46,0) | |
| 1st quartile | 30 | 32 | 31 | ||||
| Median | 42 | 38 | 38 | ||||
| 3rd quartile | 61 | 50 | 52 | ||||
| No response | 3 | 2 | 0 | ||||
a Test excluding subjects unable to take the test.
c 4 subjects could not take TMT B because they did not adequately know the alphabet and 1 subject (TBI moderate/severe) was unable to finish TMT B (“massive failure, nervous irritation”).
1.4
Discussion
Our study represents an original approach to the specificity of cognitive, neurobehavioral and psycho-affective disorders with regards to TBI. Our results provide confirmation of the importance of cognitive and behavioral sequelae after moderate/severe TBI (for a review, see Dikmen et al. ). Some disorders occur in such elevated proportions as to appear specific to this group: impaired cognitive flexibility and ability to plan, diminished affectivity, loss of initiative and motivation, and social withdrawal. While the nature and the range of damage to the executive systems in TBI patients have been extensively documented in the past , our results are also congruent with those of more recent studies that demonstrate large-scale interaction between executive impairments and behavioral, psycho-affective and social disorders .
Other disorders affect not only moderate/severe TBI subjects, but also mild TBI subjects, in whom they are generally less severe. This is the case not only for memory complaints, which have been extensively reported in the literature , but also for anxiety.
On the other hand, some of the disorders observed in our study were not specifically associated with TBI. One example is irritability, an excessive manifestation that is described as frequent and at times specific in the aftermath of brain trauma. In the study carried out by Kim et al. , irritable behavior was reported during the first year of follow-up in 30% of patients having suffered brain trauma, and was also associated with a prolonged period of more pronounced depression. In our study as well, irritability was one of the most frequently observed manifestations, occurring in 40% of the patients over periods largely posterior to the accident, whatever the group under consideration. The expression characterizing this type of behavior may be influenced not only by the alteration of one or several psychological factors (imperiousness, premeditation, perseverance, urgency and quest for sensations…), as has already been shown as concerns impulsiveness , but also by environmental factors such as family structure and financial resources. It should be noted, in conclusion, that irritability is a symptom of which the definition remains unclear; it is consequently difficult to evaluate .
As concerns depressive moods, their frequency in our two TBI groups is comparable to that observed in previous studies . Equivalent frequency is likewise found in our severely injured non-TBI group, and it may mean that depression-related complaints occur relatively independently of the nature of the injury. Conversely, anxiety is more prevalent in the TBI groups and could constitute a sequel that would be more specific to brain injury. Similar results have been reported by Demakis et al. , who showed that a year after an accident entailing brain trauma, depression-related complaints are distinguishable from those connected with anxiety. Other works have even shown that the two manifestations involve distinct areas of the brain .
It is also noteworthy that in the sub-group composed of patients with moderate/severe TBI, the frequency of several neurobehavioral disorders (diminished affectivity, social withdrawal, disinhibition, hyperactivity, excitation) was comparable to previously published findings by McCauley et al. , who made use of the NRS-R in their evaluation of neurocognitive sequelae in a severe TBI population 6 months after the trauma. On the other hand, in the McCauley study, the proportion of patients presenting a loss of initiative and motivation along with irritability was more elevated than in our study, and conversely, several cognitive disorders (planning difficulties, lessened cognitive flexibility, memory disorders, limited self-criticism, conceptual disorganization) were more frequently found in our study. Perhaps, the time elapsed prior to evaluation provides an explanation for this apparent discrepancy; while McCauley et al. performed their follow-up at 6 months, our assessment took place one full year after the accident. Our results correspondingly suggest a tendency for some complaints (executive, attention-related and, more generally, the cognitive sphere) to diminish in frequency over time, whereas others are likely to endure (neurobehavioral disorders).
As regards return to work, our results for the moderate/severe TBI group are in agreement with those reported by Van Velzen et al. and estimated at 40% in their review of the literature . On the other hand, for the mild TBI group our results are pronouncedly inferior to those usually reported, as most of the persons in other studies with mild TBI returned to work on an average of 3 to 6 months after the injury . The difference could be explained by the fact that in our study, mild TBI patients also suffered more serious injury (fracture…). It is likely that to a higher degree than the mild TBI, it was the serious injury that delayed their return to work. That much said, the low percentage of return to work underscores the importance of offering specific forms of treatment and management designed to facilitate professional reintegration . In this respect, 3-to-5-year follow-up of our study population should provide interesting information on the clinical outcomes of our three groups.
Rather astonishingly, no noteworthy differences between the three groups were found in the results of the Trail Making Test. Classical measurement such as comparison of the times needed to complete each part were limited by the fact that some of the more severely impaired patients were unable to take the second part. It would undoubtedly have been equally pertinent to specifically compare uncorrected errors, which are indicative as response time , but the relevant information was not collected. It is also conceivable that the sensitivity of the Trail Making Test fails to detect possible attention-related and dysexecutive disorders at one year. In all likelihood, the existing data reflect a frequently observed discrepancy between impairments that are hard to objectify given the inadequate sensitivity of the tests and the fact that many complaints are voiced from a distance.
When interpreting our results, the limits and potential biases of our study should be taken into consideration; they consist in:
- •
the relatively low number of participants limits analyses and inter-group comparisons;
- •
even when account is taken of aggravation during the first 72 hours, definition of the groups based on the initial data may not sufficiently distinguish the different populations in terms of cognitive and behavioral consequences once substantial time has elapsed since the trauma;
- •
initial cognitive disorders are often initially underestimated by the patient, whose possible anosognosia is readily adopted by many of his close relations, especially during an initial period that may be characterized by great “hope”, while awareness of impairments evolves over the course of time. Some of the symptoms are consequently likely to be under-reported, especially those mentioned in a declarative mode.
The NRS-R was chosen for use in this study on account of its status as a validated and recognized scale suited to an epidemiological approach . It should also be noted that in our study, the scale was applied in a face-to-face interview, which was conducted in most cases in the patients’ homes by two neuropsychologists who had received training in its use and who regularly compared their modes of questionnaire administration, thereby ensuring good homogeneity in test-taking conditions.
From a methodological standpoint, one of the strong points of our study consists in the representativeness of the population involved. The subjects were seriously injured patients forming a representative cohort of traffic accident victims in a clearly delimited geographical zone (Rhône county) for which, since 1995, all road casualties having undergone treatment have been systematically registered. And the fact of having chosen as controls and included in the same cohort a group of non-TBI patients who were nonetheless seriously injured (M-AIS ≥ 3) and had been selected according to the same procedures as the TBI patients and studied over the same time period is of particular interest in analysis of the different symptoms and of their specificity with regards to brain trauma.
1.5
Conclusion
Our study is one of the first to show, under strict epidemiological conditions, that some emotional or affective complaints generally associated with brain trauma such as irritability and depressive mood are in fact more directly related, independently of any brain damage, to an overall traumatic condition. As for cognitive complaints, they are often engendered by head injury, as occurs with regards to control of emotional reactions such as anxiety. So, it is that our results confirm the hypothesis that some disorders are not limited to severe traumatic brain injury but may also appear in cases of mild traumatic brain injury, and even in cases of trauma other than head trauma, as have been suggested by more than one author .
Disclosure of interest
The authors declare that they have no conflicts of interest concerning this article.
Acknowledgments
The authors are grateful to the victims for their cooperation in data collection. The authors would like to thank all those who assisted in carrying out this study: Nadia Baguena, Jean Yves Bar, Amélie Boulanger, Elodie Paquelet, Stuart Nash and Véronique Sotton for collecting the data; Irène Vergnes for organising the databases; Anne-Marie Bigot, Nathalie Demangel and Geneviève Boissier for subject database management; Blandine Gadegbeku, Amina NDiaye and The Association for the Rhône Road Trauma Registry (ARVAC) for their help in collecting and providing medical data; the Scientific Committee (Daniel Floret, François Chapuis, Jean Michel Mazaux, Jean Louis Martin and Jacques Gaucher); all the hospital staff who accepted the interviewers’ presence and referred victims; and the SAMU team who reported their daily emergency interventions. Special thanks to Réjean Prévost for the NRS.
Funding sources : We acknowledge funding from the French Ministry of Equipment, Transport, Housing, Tourism and Sea (Program Predit 3 “New Knowledge in the Field of Road Safety”: N o SU0400066), from the National Agency for Research (Program Predit “Safe, reliable and adapted transport” N o ANR-07-TSFA-007-01) and from the French Ministry of Health (Programs PHRC 2003: PHRC-N03 and PHRC 2005: PHRC-N051). The role of the funding source was in the collection and the analysis of the data.
2
Version française
2.1
Introduction
Les accidents de la route restent un problème de santé publique majeur en France : en 2005, 5543 personnes ont trouvé la mort, 111 683 ont été blessées, parmi lesquelles plus de 40 000 ont été hospitalisées . Les conséquences de ces accidents peuvent être très diverses et entraîner de nombreuses séquelles. Le pronostic est variable, difficile à évaluer et parfois sous-estimé , particulièrement chez les blessés les plus modérément touchés.
Depuis 1995, le registre des victimes des accidents corporels de la circulation routière a été mis en place dans le département du Rhône afin de dénombrer les accidents mais aussi de caractériser les victimes et leurs lésions . En 2005, le registre comptabilisait 8608 victimes dans le département du Rhône. Parmi ces victimes, 1453 présentaient au moins une lésion céphalique, toutes gravités confondues, et 88 sont décédées. Au cours des dernières années, la diminution du nombre d’accidents constatée à l’échelon national a également été observée dans ce département : ainsi en 2008, 7460 victimes d’accidents corporels ont été recensées par le registre (soit une baisse de 13 % en 3 ans), parmi lesquelles 1326 présentaient une lésion céphalique et 57 ont perdu la vie. Les études réalisées à partir du registre ont montré que le nombre de blessés graves avec séquelles était équivalent au nombre de décès .
Le traumatisme crânien (TC) a une place à part en raison de sa fréquence (il se situe au second rang après les lésions du rachis cervical) et de la gravité de ses séquelles .
En effet, indépendamment de la cause, le pronostic à long terme est surtout lié aux atteintes cognitives et comportementales. Dans une revue récente de la littérature, Dikmen et al. ont confirmé l’importance des séquelles cognitives après un traumatisme crânien grave ou modéré alors que celles-ci n’apparaissent pas évidentes après un TC léger. Même si beaucoup d’études comportent des biais méthodologiques (absence de groupe témoin, étiologie variée, manque de représentativité par rapport à une population de référence), elles font apparaître, au niveau cognitif, des séquelles pouvant prendre la forme de difficultés de mémoire, d’attention, ou encore un ralentissement du traitement de l’information alors qu’au niveau comportemental, les troubles les plus fréquemment rapportés sont notamment, l’irritabilité , l’impulsivité , la désinhibition et la perte d’initiative . En outre, les conséquences psychiques peuvent être au premier plan chez certains patients : modifications de la personnalité et/ou troubles thymiques (dépression, anxiété, indifférence affective) . Ces séquelles sont souvent associées à une anosognosie, les patients n’étant pas toujours pleinement conscients de l’importance et de l’implication de leurs déficits . Ces divers troubles forment ce que certains appellent le « handicap invisible » du traumatisé crânien. Les troubles du comportement et de l’humeur comme les modifications de la personnalité constituent d’ailleurs une plainte majeure de l’entourage même longtemps après l’accident . Ainsi, la multiplicité des variables dans des situations anatomo-cliniques et sociales très hétérogènes rend donc indispensable une approche rigoureuse. Dikmen et al. ont notamment insisté sur la nécessité de disposer de groupes témoins adaptés aux situations explorées, ainsi que sur l’utilité d’un suivi longitudinal.
Le travail présenté vise à repérer ce qui est spécifique au traumatisme crânien parmi les manifestations cognitives et comportementales présentes chez 207 blessés graves d’une cohorte d’accidentés de la route suivie d’octobre 2004 à juin 2006, dans le département du Rhône, et revue à un an. En comparant trois groupes de victimes, nous pourrons déterminer la nature des plaintes suivant l’existence ou non d’un traumatisme crânien, et en tenant compte de sa gravité initiale.
2.2
Matériel et méthodes
2.2.1
Population d’étude
La cohorte étude de suivi d’une population d’accidentés de la route dans le Rhône (ESPARR) est une cohorte prospective de victimes d’accidents corporels de la route. Le recrutement a été réalisé entre le 1 er octobre 2004 et le 30 juin 2006, dans tous les hôpitaux publics ou privés du département du Rhône. Toute personne répondant aux critères d’inclusion (personne résidant dans le Rhône, victime d’un accident de la circulation impliquant au moins un véhicule à roues, ayant eu lieu dans le Rhône, vivante au moment de son arrivée à l’hôpital, et se présentant dans l’un des différents services d’urgences du Rhône) a été invitée à participer à l’étude ESPARR. Les blessés les plus graves étaient recrutés directement dans les services d’hospitalisations sur toute la période. Pour les cas les plus graves (patients en réanimation), la relation avec la famille passait par le médecin réanimateur. La structure du recrutement (plages horaires réparties sur 24 heures et tous les jours de la semaine) assure la représentativité de la cohorte en termes de critères sociogéographiques, de caractéristiques de l’accident (saison, jour de la semaine, heure, type d’accident (travail, trajet…) et de types de moyens de transport . Au total, 1372 sujets ont accepté d’être suivis dans la cohorte ESPARR, dont 1168 adultes de 16 ans et plus. Les données médicales ont été recueillies à partir des données enregistrées dans le dossier médical hospitalier et des fiches de déclaration au registre des victimes d’accidents de la circulation du Rhône. Les lésions ont été codées par le médecin du registre, expérimenté dans la passation de l’Abbreviated Injury Scale (AIS) qui permet de coder chaque lésion et de lui attribuer un score de gravité allant de 1 (lésion mineure) à 6 (lésion mortelle). Le maximum des AIS observés (M-AIS) est égal à l’AIS de la lésion la plus grave et indique le niveau initial de gravité globale : un blessé est considéré comme léger si le M-AIS s’élève à 1, modéré pour un M-AIS égal à 2 et grave pour un M-AIS au moins égal à 3 (M-AIS3 + ).
Le score de Marshall et al. a été calculé pour tous les sujets ayant eu une imagerie médicale pour classer les lésions cérébrales selon leur gravité (grade 1 : pas de lésion visible à grade 4 : engagement).
Au sein de cette cohorte, la population isolée spécifiquement pour cette étude correspond aux sujets âgés de 16 ans et plus, lors de l’accident, et considérés comme des blessés graves (M-AIS ≥ 3) au moment de l’accident, qu’ils aient eu ou non un traumatisme crânien. Trois groupes de patients ont été considérés :
- •
groupe « TC modéré/grave » défini par l’un des critères suivants : score de Glasgow < 13, déficit neurologique, aggravation de l’état neurologique dans les 72 heures après l’accident, imagerie cérébrale anormal ;
- •
groupe « TC léger » défini par un score de Glasgow ≥ 13, sans aggravation dans les 72 heures et sans signe neurologique à l’examen clinique. Nous avons également inclus dans ce groupe les patients avec un score de Glasgow non renseigné, mais présentant un traumatisme crânien avec une perte de connaissance inférieure à une heure, ainsi que les patients sans trouble de conscience ayant un traumatisme facial, dans la mesure où celui-ci est associé par définition à un traumatisme crânien ;
- •
groupe « blessé grave » sans TC défini par un M-AIS ≥ 3 mais sans lésion à la tête.
Au stade initial de l’accident, chaque patient (ou son proche, en cas d’impossibilité) a été invité à répondre, au cours d’un entretien à l’hôpital avec l’enquêteur, à un questionnaire comportant, entre autres, les items suivants : sexe, âge lors de l’accident, type d’usager (moto, vélo…), situation familiale, catégorie socioprofessionnelle et niveau d’études.
Un an après l’accident, chaque patient a été revu lors d’un entretien au cours duquel un test neuropsychologique et un questionnaire ont été administrés par deux neuropsychologues, préalablement formés à leur passation. L’évaluation a comporté :
- •
la Neurobehavioral Rating Scale–Revised (NRS-R) : un outil intégré permettant une passation rapide et peu contraignante, particulièrement adapté aux études épidémiologiques. Cette échelle multidimensionnelle, composée de 29 items, a été révisée et validée auprès d’une population américaine et française , les items ayant tous été définis simultanément en langue anglaise et française. Elle associe un entretien semi dirigé et des tests cognitifs succincts. Les différents items évaluent les aspects cognitifs (comme par exemple les troubles de mémoire, de l’attention ou de la planification) mais également comportementaux et thymiques (comme l’anxiété, la dépression ou encore la labilité de l’humeur). La fatigabilité est également prise en compte. La cotation des troubles s’effectue habituellement selon 4 niveaux : (i) absence de trouble, (ii) troubles discrets, (iii) modérés ou (iv) sévères. Compte tenu des effectifs de notre étude, chaque item de la NRS a été côté comme présent ou absent ;
- •
le Trail Making Test, se composant de deux parties : dans la partie A, le sujet doit relier le plus rapidement possible et par ordre croissant des points numérotés de 1 à 25 répartis aléatoirement sur une feuille ; dans la partie B, le sujet doit relier le plus rapidement possible des points en alternant systématiquement un chiffre et une lettre, les chiffres par ordre croissant et les lettres par ordre alphabétique (1-A-2-B-3-C…). Cette épreuve est reconnue pour mesurer les capacités de flexibilité attentionnelle ;
- •
un questionnaire reprenant les données sociodémographiques relevées lors de l’inclusion (sexe, âge, reprise de travail, retour à domicile). Le syndrome de stress post-traumatique (SSPT) est évalué à partir du Post-traumatic Stress Disorder Checklist Scale . Un seuil du score ≥ 44 indique la présence vraisemblable d’un SSPT ayant des répercussions significatives sur la vie quotidienne .
Les entretiens à un an réalisés en tête à tête ont été effectués pour 72,5 % des patients à leur domicile, lors d’une consultation à l’hôpital (18,4 %), dans un bureau à la faculté de médecine (7,8 %) ou sur le lieu de travail (1,4 %). Le délai médian entre l’accident et les entretiens est de 437 jours (intervalle interquartile Q1–Q3 = [396–498]).
2.2.2
Éthique
Le consentement a été recueilli par écrit auprès de chaque patient, ou de son proche en cas d’impossibilité pour lui de répondre.
2.2.3
Analyse
Les 3 groupes ont, tout d’abord, été comparés globalement, puis chacun des deux groupes de TC a été comparé au groupe de blessés graves sans lésion céphalique, considéré comme groupe témoin. Enfin, les deux groupes de TC ont également été comparés entre eux, afin de voir si une différence d’atteinte entre les deux niveaux de gravité du TC existait. Des tests du Chi 2 (ou des tests de Fisher lorsque les effectifs étaient trop petits) ont été réalisés pour les variables catégorielles et des tests de Student (ou des tests d’analyse de variances pour l’analyse générale) pour les variables quantitatives. Lorsque le test global n’est pas significatif au seuil de 5 %, les comparaisons entre groupes n’ont pas été réalisées.
Des modèles en régression de « Poisson modifié » ont été construits pour rechercher si les associations persistent entre les divers symptômes et le fait d’avoir souffert d’un TC quand on ajuste sur l’âge, le sexe, le niveau d’études (test du Chi 2 ). Le fait d’avoir repris ou non le travail un an après l’accident a aussi été intégré au modèle multivarié lorsque l’analyse était significative à 10 % en univarié. Seuls les symptômes pour lesquels la comparaison entre les TC modérés/graves (respectivement TC légers) et la population « témoin » présentait une différence significative (après vérification du test global) et pour lesquels au moins 20 % des sujets souffraient de ce symptôme, ont été modélisés.
Les analyses statistiques ont été réalisées sous le logiciel SAS 9.3 (SAS Institute Inc., Cary, NC, États-Unis). La procédure genmod a été utilisée pour construire les modèles de régression en « Poisson modifié ».
2.3
Résultats
Parmi les 1168 adultes de 16 ans et plus de la cohorte ESPARR, 324 sujets répondaient aux critères précédemment définis (M-AIS ≥ 3). Parmi ceux-ci, 68 sujets ont accepté de répondre au questionnaire mais n’ont pas effectué les tests, par refus de l’entretien en face à face. Vingt n’ont pas répondu au suivi à un an mais ont répondu à un suivi ultérieur de l’étude, 15 ont refusé de poursuivre leur participation à l’étude tandis que 12 ont été perdus de vue et 1 sujet est décédé. Au total, 116 accidentés n’ont donc pas participé aux tests neuropsychologiques un an après l’accident ; ainsi, le suivi neuropsychologique a effectivement été réalisé pour 208 blessés, soit un taux de participation de 64 % des sujets.
Un sujet présentant un TC léger a été exclu des analyses car il avait subi un choc hypovolémique ayant entraîné la cotation du score de Glasgow à 9. Notre population d’étude porte donc sur 207 sujets adultes dont 48 dans le groupe TC modérés/graves, 89 dans le groupe TC légers et 70 dans celui des blessés graves sans TC.
Sur les critères d’âge et de sexe et au regard du niveau de sévérité du traumatisme crânien, les blessés graves n’ayant pas répondu au suivi neuropsychologique à un an ne diffèrent pas de ceux ayant répondu.
2.3.1
Description de l’échantillon
L’ensemble des données médicales et sociodémographiques de cet échantillon est détaillé dans le Tableau 1 .
Related posts:
Enhancing Annalsvalue through quality
Lack of repercussions of sleep apnea syndrome on recovery and attention disorders at the subacute stage after stroke: A study of 45 patients
Detection of progressive idiopathic scoliosis during growth using back surface topography: A prospective study of 100 patients
Quality of life and satisfaction after multilevel surgery in cerebral palsy: Confronting the experience of children and their parents
Communication disorders and executive function impairment after severe traumatic brain injury: An exploratory study using the GALI (a grid for linguistic analysis of free conversational interchange)
The motor function measure to study limitation of activity in children and adults with Charcot-Marie-Tooth disease
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
