Content uploaded by Selim R Benbadis
Author content
All content in this area was uploaded by Selim R Benbadis on Nov 25, 2022
Content may be subject to copyright.
ORIGINAL ARTICLE
EEG Findings in the Persistent Vegetative State
Vidya P. Kulkarni, Kaiwen Lin, and Selim R. Benbadis
Summary: The definition of the persistent vegetative state (PVS) is
relatively straightforward, but its diagnosis can be challenging. We
reviewed a series of EEG performed in patients with PVS to assess
the diagnostic value of EEG. We reviewed records of all hospital
patients with a diagnosis of persistent vegetative PVS. EEG findings
included normal, continuous generalized slowing, intermittent gen-
eralized slowing, background slowing, background suppression,
alpha, generalized periodic pattern, PLEDS, and triphasic waves.
EEG findings had no association with etiology and varied from one
pattern to another in the same patients’ EEGs obtained at different
times (see table). We conclude that EEG findings in PVS are
heterogeneous and too variable to be of diagnostic value.
Key words: Vegetative state, Minimally conscious state, Coma,
EEG.
(J Clin Neurophysiol 2007;24: 433–437)
The persistent vegetative state (PVS) is defined as a “clin-
ical condition of complete unawareness of self and envi-
ronment, accompanied by sleep-wake cycles with either com-
plete or partial preservation of hypothalamic and brain stem
autonomic functions” (Bernat, 2006; Luders and Noachtar,
2000; Practice Parameters, 1995). PVS represents a challenge
for neurologists because the patient may seem aware to the
family, as illustrated by the March 2005 court battles of the
Terri Schiavo case. This case had a significant impact on
social awareness, the encouragement of advance directives
and even legislation (Bernat, 2006). Because the definition of
PVS is clinical and disagreements can have serious conse-
quences, objective data to support or refute the diagnosis
would be useful. Although EEGs are typically performed for
evaluation of altered mental status, little data are available on
EEG findings in PVS. Thus, the goal of this study was to
review a series of EEG performed in patients with PVS.
METHODS
We reviewed records of all patients with a diagnostic
code of persistent vegetative state (ICD-9 code 780.03) ad-
mitted to Tampa General Hospital between 2001 and 2005. A
total of 69 patients’ charts were reviewed. Only those patients
who carried a confirmed diagnosis of persistent vegetative
state by a board-certified Neurologist or Neurosurgeon at the
time of the EEG recording were included in the study. The
chart review was performed with special attention to the EEG
findings and the etiology. A standardized EEG classification
system is used at our center (Luders and Noachtar, 2000), and
this was used to characterize records.
RESULTS
Results are summarized in Table 1. A total of 31 EEGs
were obtained on 16 patients. The number of EEGs per
patient ranged from 1 to 3. The etiologies of PVS comprised
the following:
YStroke in 6 of 16 (37%): 4 ischemic and 2 hemorrhagic
(ICH)
YAnoxia/hypoxia in 4 of 16 (25%)
YMultiorgan failure in 4 of 16 (25%)
YTraumatic injury in 1 of 16, and
YSubarachnoid hemorrhage in 1 of 16
EEG findings (classifications) by individual patients are
shown in Table 1. The various patterns were often combined,
so that the total number of patterns is greater than 31:
YEEG was normal in 1 of 31 (3.2%).
YContinuous generalized slowing (continuous slow, gen-
eralized): This consists of unreactive polymorphic delta
for ⬎80% of the record, and was found in 10 of 31.
YIntermittent generalized slowing (intermittent slow, gen-
eralized): This consists of polymorphic delta for ⬍80%
of the record and reactive and was found in 8 of 31.
YBackground slowing (posterior dominant reactive back-
ground but ⬍8 Hz) in 4 of 31.
YBackground suppression (amplitude ⬍10
V and unre-
active) in 4 of 31.
YAlpha coma (continuous unreactive alpha frequency
activity) in 4 of 31.
YGeneralized periodic pattern in 1.
YPLEDS in 1 (combined with generalized burst-suppres-
sion).
YTriphasic waves in 1 (combined with generalized slowing).
EEG findings had no association with etiology, and
varied from one pattern to another in the same patients’ EEGs
obtained at different times (Table 1).
University of South Florida and Tampa General Hospital, Tampa, Florida.
Address correspondence and reprint requests to Dr. Selim R. Benbadis, 4,
Columbia Drive, Suite 730, Tampa, FL 33606; e-mail: sbenbadi@health.
usf.edu.
Copyright © 2007 by the American Clinical Neurophysiology Society
ISSN: 0736-0258/07/2406-0433
Journal of Clinical Neurophysiology • Volume 24, Number 6, December 2007 433
TABLE 1. EEG Findings
ID Date of Insult Cause Imaging Results EEG (Date) EEG 2 (Date) EEG 3 (Date)
WB 2/13/2004 Multiple strokes CT: bilateral infarcts CSgen (2/19) CSgen (3/4) CSgen, triphasic (3/9)
CB 3/8/2005 ICH MRI: Large ICH BS, ISgen (3/10) BS (4/8) Alpha coma (4/26)
KV 12/8/2004 Anoxic injury MRI: Atrophy Normal (12/11) BS (12/15) ISgen (1/21)
WG 2/12/2005 ICH MRI: Large ICH Alpha coma, ISgen (2/14) ISgen (3/1)
DS Unclear Multi-organ failure CT: Severe atrophy Bsupp, ISgen (10/10) ISgen (4/2)
JT 8/15/2004 Anoxic injury CT: Negative Alpha coma (8/16) Bsupp (8/21)
MC 1/6/2003 TBI CT: severe atrophy CSgen (5/15) CSgen (5/17)
RL 7/24/2003 Multi-organ failure MRI: Atrophy BS, ISgen (7/28) CSgen (8/4)
LK 7/11/2004 Anoxic injury MRI: abn bl diffusion Periodic pattern (7/12) Bsupp (7/13)
PB 2/8/2004 Multiple strokes MRI: bilat MCA strokes Bsupp, PLEDS (2/20)
AL 10/24/2004 SAH CT: large SAH CSgen (10/27/2004)
LW 9/9/2001 Anoxic injury MRI: atrophy Alpha coma (10/29)
CM 11/13/2004 Multiple strokes CT: bilateral MCA infarct ISgen (11/16)
LJ 6/22/2002 Stroke MRI: MCA stroke Csgen (6/29)
FW 2/28/2004 Multi-organ failure CT: L frontal ischemia CSgen (3/1)
JM 4/28/2004 Multi-organ failure CT: extensive atrophy CSgen (5/3)
BS, background slow; Bsupp, background suppression; CSgen, Continuous, slow, generalized; ICH, intracerebral hemorrhage; ISgen, intermittent, slow, generalized; MCA,
middle cerebal artery; SAH, subarachnoid hemorrhage; TBI, traumatic brain injury.
FIGURE 1. Normal EEG. Note the normal and reactive alpha rhythm.
Kulkarni et al. Journal of Clinical Neurophysiology • Volume 24, Number 6, December 2007
Copyright © 2007 by the American Clinical Neurophysiology Society434
FIGURE 2. Background slowing. Note that poorly defined background at about 6 Hz, with some reactivity.
FIGURE 3. Intermittent generalized slowing (transverse montage). Note the brief bursts of polymorphic delta slowing.
Journal of Clinical Neurophysiology • Volume 24, Number 6, December 2007 EEG Findings in the Persistent Vegetative State
Copyright © 2007 by the American Clinical Neurophysiology Society 435
FIGURE 4. Background suppression. Note the very low amplitude (less than 10 microvolts) and lack of reactivity.
FIGURE 5. Alpha coma. Note the diffuse alpha frequency activity, which is continuous and unreactive.
Kulkarni et al. Journal of Clinical Neurophysiology • Volume 24, Number 6, December 2007
Copyright © 2007 by the American Clinical Neurophysiology Society436
DISCUSSION
Patients in PVS commonly undergo an EEG as part of
an evaluation of their altered state of consciousness. How-
ever, few systematic studies have been done regarding EEG
findings among such patients. In this study, we found that
EEG findings from patients in a persistent vegetative state are
variable and nonspecific, and thus of no value in distinguish-
ing between PVS and other states of altered consciousness.
The significance of the variations that occur in EEGs over
time within one patient is unclear.
In keeping with our findings, a study of 8 patients with
PVS found that EEGs showed a range of patterns that were
unchanged from the comatose through the vegetative state
(Hansotia, 1985). Thus, there is marked interindividual EEG
variability, which severely restricts any practical value for
EEG in PVS (Guerit, 2005).
One potential limitation of our study is that there is
some uncertainty regarding the accuracy of the diagnosis of
PVS versus minimally conscious or other similar states (e.g.,
locked-in syndrome). We were careful to include only pa-
tients diagnosed by board-certified neurologists or neurosur-
geons, but errors could conceivably have occurred. Another
limitation is that the variable EEG patterns seen in patients
who had more than one EEG were done mostly within the
first 30 days of the insult, which is before they can be
considered in persistent vegetative state at the time of the
EEG. Thus, the evolution of the EEG might be seen to reflect
an evolving clinical picture before the diagnosis of PVS.
It is possible that other functional imaging modalities
(e.g., FMRI) will prove useful to evaluate patients with PVS.
However, currently proper clinical evaluation comprising of
detailed history and careful neurologic examination cannot be
replaced by “tests,” and PVS remains entirely a clinical
diagnosis.
REFERENCES
Bernat JL. Chronic disorders of consciousness. Lancet 2006;367:1181–1192.
Guerit JM. Neurophysiological patterns of vegetative and minimally con-
scious states. Neuropsychol Rehabil 2005;15:357–371.
Hansotia PL. Persistent vegetative state: Review and report of electro
diagnostic studies in eight cases. Arch Neurol 1985;42:1048–1052.
Luders HO, Noachtar S (eds.). Atlas and Classification of Electroencepha-
lography. Philadelphia: W.B. Saunders Company 2000.
Practice Parameters: Assessment and management of patients in the persis-
tent vegetative state. Neurology 1995;45:1015–1018.
Journal of Clinical Neurophysiology • Volume 24, Number 6, December 2007 EEG Findings in the Persistent Vegetative State
Copyright © 2007 by the American Clinical Neurophysiology Society 437