Elsevier

The Lancet

Volume 351, Issue 9118, 13 June 1998, Pages 1763-1767
The Lancet

Articles
Prediction of recovery from post-traumatic vegetative state with cerebral magnetic-resonance imaging

https://doi.org/10.1016/S0140-6736(97)10301-4Get rights and content

Summary

Background

The early post-traumatic vegetative state (VS) is compatible with recovery. Various clinical and laboratory tests have failed to predict recovery so we assessed the value of cerebral magnetic-resonance imaging (MRI) in prediction of recovery.

Methods

80 adult patients in post-traumatic VS had cerebral MRI between 6 weeks and 8 weeks after injury. MRIs were reviewed by three neuroradiologists for the number, sizes, and location of brain lesions. Three neurologists assessed the patients at the time of MRI and at 2 months, 3 months, 6 months, 9 months, and 12 months after injury using the Glasgow Outcome Scale.

Findings

At 12 months, 38 patients had recovered while 42 patients remained in the VS. The demographic characteristics and causes and severity of injury were similar in patients in persistent VS (PVS) and those who recovered (NPVS). An average of 6·1 different brain areas were injured in patients in PVS compared with 4·6 areas in patients who had NPVS. Patients in PVS revealed a significantly higher frequency of corpus callosum, corona radiata, and dorsolateral brainstem injuries than did patients who recovered. Logistic regression analysis showed that corpus callosum and dorsolateral brainstem injuries were predictive of non-recovery. The adjusted odds ratios for non-recovery of patients with a corpus callosum lesion and dorsolateral brainstem injury were 213·8 (95% Cl 14·2–3213·3), and 6·9 (1·1–42·9), respectively. In contrast, clinical characteristics, such as initial score on the Glasgow Coma Scale, age, and pupillary abnormalities failed to predict recovery.

Interpretation

Cerebral MRI findings in the subacute stage after head injury can predict the outcome of the post-traumatic VS. Corpus callosum and dorsolateral brainstem lesions are highly significant in predicting non-recovery.

Introduction

There is no more devastating or morally challenging condition in modern medicine than the persistent vegetative state (PVS).1 The term vegetative state (VS) is used to describe the condition of patients with severe brain damage, in whom vegetative functions (sleep-wake cycles, autonomic control, and breathing) persist, but awareness (including all cognitive function and emotion) is abolished.1, 2, 3 The estimated number of adult patients in PVS in the USA ranges from 10 000 to 25 000,1 and in 1996 it was estimated that between US$1 billion and US$7 billion may be spent annually in providing their medical care.4

The diagnosis of a PVS can have a major influence on decision making concerning the level of care or services provided and may prompt an application to be made to the courts for a directive on withdrawal of tube feeding. Unfortunately, misdiagnosis of VS is not uncommon.5 A report by Andrews and colleagues in 19966 provided evidence that up to 43% of patients were wrongly diagnosed as being in a VS. In addition, recovery from VS is not unlikely; half of the patients in post-traumatic VS may recover within 1 year of the injury.7 Therefore, the first step in VS management requires correct clinical diagnosis. Moreover, ancillary diagnostic tests, in conjunction with a clinical assessment may provide important information for confirming a diagnosis of VS, and may also be helpful in predicting the potential recovery.

Various neurodiagnostic tests have been assessed in a quest for improved prediction of recovery from VS. These have largely centered on evoked potentials, the electroencephalogram, and cerebral computed tomography (CT). However, these tests have failed to predict the potential for recovery.

Magnetic resonance imaging (MRI) has been shown to be considerably more sensitive than cerebral CT for detection of traumatic and ischaemic cerebral lesions.8, 9 However, MRI has not been used to characterise the pattern of brain lesions in post-traumatic VS. Moreover, the value of cerebral MRI for prediction of recovery from a VS has not been investigated.

This study was done to define the MRI signs of cerebral injury in patients in post-traumatic VS. We also examined whether lesions in certain brain areas can predict that there will be non-recovery from a post-traumatic VS.

Section snippets

Patients and methods

The 80 patients who took part in the study were from a pool of adult patients with closed-head injuries, who were consecutively admitted to our trauma and rehabilitation centre between Jan 1, 1988, and March 31, 1996. Patients were enrolled in our study if VS continued to the subacute stage 6–8 weeks after injury. Patient's age, sex, initial pupillary response, initial score on the Glasgow Coma Scale (GCS), and time from injury to MRI were recorded.

Neurological outcome was scored on a consensus

Results

Most of the patients in PVS and NPVS groups were men, and the most common cause of head injury was a motor vehicle accident (table 1). The two groups did not differ significantly in terms of age, sex, pupillary abnormalities, or initial GCS score. The rates of incidence of a hypotensive episose in the prehospital setting (within 24 h of injury), and of a craniotomy for evacuation of an epidural or subdural haematoma were also similar in the PVS and NPVS patients. The frequency of medical

Discussion

Our data suggest that cerebral MRI may assist in early prediction of outcome from a post-traumatic VS. Our findings indicate that lesions of the corpus callosum and the dorsolateral upper brainstem are predictive of a patient not recovering. Clinical features such as initial GCS score, pupillary reactivity, and patient age failed to predict recovery from a VS.

Because of the prognostic and therapeutic uncertainties concerning VS, several professional medical organisations began a comprehensive

References (30)

  • A Zeman

    Persistent vegetative state

    Lancet

    (1997)
  • B Jennet et al.

    Persistent vegetative state after brain damage

    Lancet

    (1972)
  • Medical aspects of the persistent vegetative state (first of two parts)

    N Engl J Med

    (1994)
  • K Payne et al.

    Physicians' attitudes about the care of patients in the persistent vegetative state: a national survey

    Ann Intern Med

    (1996)
  • NL Childs et al.

    Accuracy of diagnosis of persistent vegetative state

    Neurology

    (1993)
  • K Andrews et al.

    Misdiagnosis of the vegetative state: retrospective study in a rehabilitation unit

    BMJ

    (1996)
  • Medical aspects of the persistent vegetative state (second of two parts)

    N Engl J Med

    (1994)
  • LR Gentry

    Imaging of closed head injury

    Radiology

    (1994)
  • NM Ramadan et al.

    Magnetic resonance and clinical cerebrovascular disease. An update

    Stroke

    (1989)
  • B Jennet et al.

    Assessment of outcome after severe brain damage: a practical scale

    Lancet

    (1975)
  • J R Coll Physicians Land

    (1996)
  • R Braakman et al.

    Prognosis of the posttraumatic vegetative state

    Acta Neurochir

    (1988)
  • HS Levin et al.

    Vegetative state after head injury. A traumatic coma data bank report

    Arch Neural

    (1991)
  • I Ahmed

    Use of somatosensory evoked responses in the prediction of outcome from coma

    Clin Electroencephalogr

    (1988)
  • JA Judson et al.

    Early prediction of outcome from cerebral trauma by somatosensory evoked potentials

    Crit Care Med

    (1990)
  • Cited by (197)

    • Neuroimaging in Disorders of Consciousness and Recovery

      2024, Physical Medicine and Rehabilitation Clinics of North America
    • ACR Appropriateness Criteria® Head Trauma: 2021 Update

      2021, Journal of the American College of Radiology
    • Subdivisions of the posteromedial cortex in disorders of consciousness

      2018, NeuroImage: Clinical
      Citation Excerpt :

      Others have also implicated tracts between the PMC and thalamus in the pathophysiologic basis of DOCs (Laureys and Schiff, 2012), and in the recovery from these (Koenig et al., 2014; Norton et al., 2012; Wu et al., 2015). Recovery from a vegetative state is also more likely with less damage to the corpus callosum (Kampfl et al., 1998), through which many of the PMC to thalamus tracts pass. These previous reports on recovery from DOCs fit with our finding of less PMC-thalamic disruption in DOC patients who recovered within 12 months than in those who did not.

    View all citing articles on Scopus
    View full text