Traumatic brain injury (TBI) is a common problem in the United States, with an annual hospitalization rate approaching 200 per 100,000 people.1 The inpatient neurological and neurobehavioral evaluation of these individuals usually includes at least an elementary neurological examination, a general mental status examination, and some type of bedside cognitive assessment.2 Our Behavioral Neurology & Neuropsychiatry consultation service, which serves an acute inpatient brain injury rehabilitation unit, also routinely includes assessment for subtle neurological signs, including several primitive reflexes and paratonia. In this context, multiple primitive reflexes and paratonia are elicited commonly, particularly among patients with marked posttraumatic cognitive impairment and functional disability.

These observations prompted a retrospective study intended to address the following hypotheses: the number of subtle neurological signs predicts (a) raw and Z-transformed Mini-Mental State Examination (MMSE)3 scores; (b) raw and Z-transformed Frontal Assessment Battery (FAB)4 scores; (c) total Functional Independence Measure (FIM)5 score at time of Behavioral Neurology & Neuropsychiatry consultation; (d) total FIM score at time of rehabilitation discharge; and (e) rehabilitation length of stay.

This retrospective study was approved by the Colorado Multiple Institutional Review Board and the HealthONE Institutional Review Board.

Participants

Medical records of 83 patients consecutively admitted to an acute inpatient neurorehabilitation unit following traumatic brain injury were reviewed. Inclusion criteria for this study were: clinical diagnosis of TBI by American Congress of Rehabilitation Medicine definition6; nonpenetrating TBI; age 20—89 years; English as the primary language; not aphasic; Behavioral Neurology & Neuropsychiatry consultation performed, including neurological and cognitive examinations required to address the study hypotheses; and medical records containing all relevant assessment and rehabilitation outcome measures. Among participants meeting these inclusion criteria, medical records data extraction included: subject age, gender and education; mechanism of injury; presence or absence of intracranial abnormality on computed tomography (CT) or MRI of the brain; whether or not the patient was in the period of posttraumatic amnesia as assessed by the Galveston Orientation and Amnesia Test (GOAT)7 upon rehabilitation admission and also upon Behavioral Neurology & Neuropsychiatry consultation; neurological examination findings; MMSE, and FAB scores; FIM scores at time of rehabilitation admission, proximate (±3 days) to Behavioral Neurology & Neuropsychiatry consultation, and at rehabilitation hospital discharge; and acute hospital and rehabilitation hospital lengths of stay.

Subtle Neurological Signs Score

A subtle neurological signs score was constructed based on the neurological examination data recorded during the Behavioral Neurology & Neuropsychiatry consultations. The elements contributing to the subtle neurological signs score included paratonia (mitgehen and/or gegenhalten) and primitive reflexes. Mitgehen and gegenhalten reflect impairments of volitional motor inhibition during assessment of resistance to passive manipulation (hence the designation of "paratonia"). Mitgehen (from German, "go along" or "go with") refers to a patient’s inability to inhibit active movement of the extremity under examination despite instructions to the patient to remain passive during the examiner’s movement of that extremity. Gegenhalten (from German, "hold against" or "go against") describes purposeless resistance by the patient against movement of the extremity by the examiner ("pull-counter-pull" phenomenon). The primitive reflexes included in the subtle neurological signs score: glabellar response, snout response, grasp response (left and right), palmomental (left and right), suck reflex, and rooting response. The examination for these findings and interpretation of primitive reflexes were performed according to the methods described in Arciniegas and Beresford.8 One point was allocated for each abnormal finding, and missing data points were scored as 0. The subtle neurological signs score was equal to [# of primitive reflexes] + [mitgehen] + [gegenhalten], with a range of 0—10 points.

Cognitive Assessment Measures

The MMSE was administered as described by Folstein et al.3 and the FAB was administered as described by Dubois et al.4 To correct for the effects of age and education on MMSE and FAB scores, Z-transformations were performed using normative data for these measures.9,10

Functional Independence Measure Assessments

The FIM assessments were performed by physical therapy, occupational therapy, speech language pathology, and nursing staff assigned to the acute inpatient rehabilitation unit using the method described in the Guide for Uniform Data Set for Medical Rehabilitation, Version 5.1.5

Statistical Analyses

Statistica 8.0 (StatSoft, Inc., Tulsa, Okla.) was used for all descriptive data analyses as well as for regression analyses to test the study hypotheses. Simple regression modeling was used to investigate the proportion of variance in MMSE and FAB scores (raw and Z-scores), FIM scores, and rehabilitation length of stay accounted for by subtle neurological signs score.

Fifty-one participants (13 women) met study inclusion criteria; these patients are described in Table 1. Among these patients, causes of TBI included motor vehicle accidents (50.9%), falls (31.4%), sports/recreational (11.8%), and assaults (5.9%). Neuroimaging (CT and/or MRI) demonstrated intracranial abnormalities consistent with TBI among 47 patients (92.2%). At the time of rehabilitation admission, 23 patients (45.1%) were in posttraumatic amnesia, and 14 (27.5%) were still in posttraumatic amnesia at the time of Behavioral Neurology & Neuropsychiatry consultation (usually 5—6 days after rehabilitation admission).

Paratonia and Primitive Reflexes

Paratonia was present in 30 patients (58.8%); all of these demonstrated mitgehen, and eight patients (15.7%) also demonstrated gegenhalten. Fifty patients (98%) demonstrated at least one primitive reflex; two or more primitive reflexes were present in 45 patients (88.2%), and three or more were present in 34 patients (66.7%). Glabelar response was the most common primitive reflex (82.4%), followed by snout response (74.5%), palmomental response (left 49% and right 47.1%), grasp response (left 41.2% and right 41.2%), suck reflex (19.6%), and rooting response (3.9%).

Regression Modeling of Cognitive and Functional Status Using Subtle Neurological Signs

Subtle neurological signs score predicts raw FAB score (adjusted R²=0.28, β=−0.54, p<0.00005), raw MMSE score (adjusted R²=0.25, β=−0.51, p<0.0002), FAB Z-score (adjusted R²=0.24, β=−0.51, p<0.0002), and MMSE Z-score (adjusted R²=0.15, β=−0.41, p<0.003). Subtle neurological signs score also predicts total FIM score at the time of Behavioral Neurology & Neuropsychiatry consultation (adjusted R²=0.12, β=−0.37, p<0.01) and rehabilitation discharge (adjusted R²=0.13, β=−0.38, p<0.006). Finally, subtle neurological signs score predicts rehabilitation length of stay (adjusted R²=0.24, β=0.51, p<0.0002).

Paratonia and primitive reflexes predict cognitive performance, functional independence, and inpatient rehabilitation length of stay in the subacute period following TBI. Assessment for these subtle neurological signs among inpatients with recent TBI therefore may contribute useful information regarding their need for further neurobehavioral rehabilitative evaluation and treatments.

After correcting for age and education, subtle neurological signs score accounted for a greater proportion of the variance in performance on the FAB than on the MMSE. This observation, that subtle neurological signs score is more strongly associated with performance on a measure of executive function (FAB) than one of general cognition (MMSE), is concordant with the common attribution of paratonia and primitive reflexes to frontal dysfunction,11 although the anatomic localization of these findings remains a matter of controversy.11,12 Since the presence of paratonia and primitive reflexes predicts cognitive impairment in this population, eliciting such findings on the elementary neurological examination should prompt the clinician to perform a more detailed bedside cognitive examination, such as that included in our Behavioral Neurology & Neuropsychiatry consultations.

Subtle neurological signs also predicted functional status as assessed by the FIM, similar to the association between primitive reflexes and functional status in neurodegenerative dementias.13,14 When planning rehabilitation treatment, expectations regarding duration of inpatient treatment may also be anticipated by the number of subtle neurological signs given that 24% of the variance in rehabilitation length of stay was accounted for by subtle neurological signs score.

Describing these findings and their implications with respect to cognitive performance, functional status, and rehabilitation outcomes may also be of use in the education of family and other care providers during the acute rehabilitation period. In our experience, some families have difficulty accepting as objective the cognitive and functional assessments performed in the acute rehabilitation setting. Neurological examination findings are, for many of these families, more easily accepted as objective and as evidence of injury-related neurological dysfunction. Complementing discussions of posttraumatic cognitive, neurobehavioral and functional impairments with the presentation of neurological findings such as paratonia and primitive reflexes may allow, in such circumstances, families to engage in these discussions more fully and to appreciate more readily the serious nature of their loved one’s clinical condition.

The principal limitation of the present study is its retrospective design and predication on data collected originally for clinical rather than research purposes. Additionally, the preliminary nature of the subtle neurological signs metric employed in this study is of concern: this metric has not yet been validated, and its reliability has not been established. While acknowledging these limitations, it is noteworthy that statistically significant relationships were identified between the subtle neurological signs score and the cognitive and functional measures in a data set derived from everyday practice. This observation suggests that the generalizability of these findings may be substantial despite the retrospective nature of this study. Other limitations include the lack of a data set of size sufficient to permit item-level analyses, such as prediction of cognitive or functional status by individual primitive reflex or paratonia findings, or to establish a cutoff subtle neurological signs score that is predictive of cognitive or functional impairment.

Despite these limitations, the present findings suggest that assessment of persons with recent TBI for paratonia and primitive reflexes may yield information with important diagnostic and treatment implications. Given the need to develop further clinical metrics that identify persons with recent TBI in need of further diagnostic and functional interventions,15 we recommend the development of prospective studies of this subject, including validation of the subtle neurological signs metric employed here, determination of its reliability, and further investigation of its clinical usefulness.

This work was presented as a poster at the 19th Annual Meeting of the American Neuropsychiatric Association, Savannah, Ga., on March 3, 2008. This work was also published in abstract form as: Wortzel HS, Frey KL, Anderson CA, Arciniegas DB: Subtle neurological signs predict the severity of subacute posttraumatic cognitive and functional impairments in traumatic brain injury (Abstract # P2), in Abstracts Presented at the 19th Annual Meeting of the American Neuropsychiatric Association. J Neuropsych Clin Neurosci 2008; 20:232—234 (available online at http://neuro.psychiatryonline.org/content/vol20/issue2/index.dtl#ABSTRACTS)

This work was supported by HealthONE Spalding Rehabilitation Hospital (KLF, DBA) and the VISN-19 MIRECC (HSW, CAA, DBA). The authors are grateful to Kenneth L. Tyler, M.D., for his assistance during the preparation of this manuscript.