вход по аккаунту


Effects of anterior temporal lobectomy on language function A controlled study.

код для вставкиСкачать
Effects of Anterior Temporal Lobectomy
on Ianguage Function: A Controlled Study
Bruce P. Hermann, PhD,*t and Allen R. Wyler, MD"S;3
This study represents the first prospective controlled investigation of preoperative versus postoperative (6 months)
language function in patients who underwent partial resection of the dominant (n = 15) or nondominant (n = 14)
anterior temporal lobe for treatment of medically refractory epilepsy. Language dominance was confirmed by intracarotid sodium amytal test. Thirteen of the 15 patients undergoing anterior temporal lobectomy of the dominant
hemisphere were operated on under local anesthesia in order to map language and memory functions intraoperatively.
Using a standardized languageiaphasia battery, we found a significant trend of worse preoperative language function
in patients with dominant hemisphere temporal lobe foci in comparison to patients with nondominant foci. Following
anterior temporal lobectomy, neither group showed any significant losses in language function, whereas the dominant
hemisphere temporal lobe group showed significant improvement in receptive language comprehension and associative verbal fluency.
Hermann BP, Wyler AR. Effects of anterior temporal lobectomy on language function:
a controlled study. Ann Neurol 1988;23:585-588
Anterior temporal lobectomy (ATL) is a procedure of
documented therapeutic value for the treatment of intractable seizures of temporal lobe origin (1-31. When
ATL. is planned for the dominant cerebral hemisphere,
concerns may be raised regarding potential adverse effects on language function. Unfortunately, only a few
articles address the effects of dominant hemisphere
ATL on language ability. These articles have generally
reported investigations of different and selected aspects of language function and, most importantly, patients have been evaluated only in the postoperative
period 14-61. Therefore, even though some investigators have reported impairment in selected aspects of
language function (e.g, visual naming ability) following
ATL of the dominant hemisphere 14-51, the possibility exists that language deficits identified postoperatively may have existed preoperatively and were not a
consequence of surgical intervention (51. Finally, the
reported findings have shown variability-some investigators have found no postoperative language or naming deficits [GI.
Despite the long history of ATL, no controlled prospective study has compared preoperative versus postoperative language function in resected dominant and
nondominant temporal lobe patients using a standardized aphasidlanguage battery. This report presents the
results of such a study.
From the "Epilepsy Center, Baptist Memorial Hospital, Memphis,
the Departments of ?Psychiatry and $Neurosurgery, University of
Tennessee-Memphis7 and the 5semmes-M~hey Clinic9 Memphis, TN.
Subjects for this investigation consisted of 29 patients who
underwent partial resection of the dominant (n = 15) or
nondominant (n = 14) anterior temporal lobe. Cerebral
dominance for language was determined by the intracarotid
sodium amytal procedure 171.The demographic and seizurerelated characteristics of the subjects are shown in Table 1.
None of the differences was statistically significant. Preoperatively, all patients underwent continuous closed-circuit
TV electroencephalographic monitoring with either scalp,
sphenoidal, andor subdural strip electrodes [8} to confirm
consistent unilateral temporal lobe origin of the seizures.
Subjects also underwent extensive neuropsychological assessment; the Multilingual Aphasia Examination (MAE) was
used to assess language function {9}. MAE is a specialized
language battery designed to evaluate the presence, severity,
and qualitative aspects of language disorder. The following
subtests were administered as part of this investigation and
the descriptions that follow are taken from the MAE manual.
Visual Naming: The stimulus material consists of 10 pictures
calling for 30 naming responses. The primary purpose of
the test is to assess the capacity to apply semantically correct verbal labels to visually presented stimuli, independent of defects in recognition or speech articulation.
Sentence Repetition: This test consists of 14 sentences of progressively increasing length, ranging from 3 to 18 words.
There are examples of each of 7 different grammatical
Received Sep 1, 1987, and in revised form N o v 13. Accepted for
publication Dec 10, 1987.
Address correspondence to Dr Hermann, Epilepsy Center, Baptist
Memorial Hospital, 899 Madison Ave, Memphis, TN 38146.
Copyright 0 1988 by the American Neurological Association 585
Table 1. Subject Characteristia
Age (yr)
Education (yr)
Age at onset (yr)
Duration (yr)
Verbal IQ
Performance IQ
Amount of lateral
resection (cm)'
Seizure free
ATL of
(n = 15)
ATL of
(n = 14)
31.3 (7.2)a
12.4 (1.8)
36.5 (13.2)
11.9 (1.8)
13.5 (9.9)
17.7 (8.9)
87.3 (7.5)
94.5 (11.5)
4.7 (0.4)
18.4 (14.4)
18.1 (13.8)
91.4 (9.7)
92.9 (12.1)
5.1 (0.6)
"Values presented are means with SD in parentheses.
'Two of the left-handed patients (both male) were right-hemispheredominant for language and underwent a right temporal lobectomy.
The remaining left-handed patient was a female who was lefthemisphere-dominant and underwent a left temporal lohectomy.
'p = 0.07.
ATL = anterior temporal lohectomy; IQ = intelligence quotient.
Controlled Oral WordAssociation: This is an oral fluency test in
which the subject is required to make verbal associations
with a letter of the alphabet by saying all the words that he
or she can think of beginning with that letter. Three letters
of progressively increasing associative difficulty are presented successively as stimuli, and the subject is given 1
minute per letter in which to respond.
OralSpelling:The subject is asked to spell 11 different words
Token Test: This 28-item test of oral language comprehension
is an abbreviation and modification of the Token Test of
DeRenzi and Vignolo { 10). Twenty small and large circles
and squares in 5 colors are employed to assess the patient's
ability to comprehend and carry out simple commands.
Aural Comprehension of Words and Phrases: This is a multiple
choice test of aural comprehension calling for 18 responses. Six pages, each with 4 possible choices, assess
comprehension of 11 single words and 7 short phrases.
Reading Comprehension of Words and Phrases: This test is the
reading counterpart of the multiple choice aural comprehension test. Stimulus materials consist of 18 single words
or short phrases in %th inch print on 5 X 8 cards calling
for pointing responses to the same stimulus materials as
the above test.
Raw scores for each of the above subtests were corrected
for age and education where appropriate, and the corrected
scores were converted to standardized t scores (2 = 50, SD
= 10) to normalize the distribution for statistical purposes.
586 Annals of Neurology Vol 23 N o 6 June 1988
All subjects subsequently underwent craniotomy with partial resection of the anterior temporal lobe. Thirteen of the
15 patients undergoing ATL of the dominant hemisphere
were operated on under local anesthesia so that intraoperative mapping of speech and memory functions could be conducted. A tailored resection was carried out based on the
results of the electrocorticography and speech-mapping procedures. The remaining 2 patients received general anesthesia because it was believed they would not tolerate the procedure well under local anesthesia. The surgical procedure was
fairly standardized. There was generally about a 4.5-cm resection of lateral cortex from the dominant temporal lobe
and a 5.0-cm resection from the lateral nondominant temporal lobe. As shown in Table 1, the actual mean amount of
resected lateral dominant temporal cortex was 4.7 cm, with
5.1 cm of nondominant cortex resected. Electrocorticography modified these lateral margins minimally and was used
primarily to determine whether larger medial resections
were necessary. In all cases, the amygdala and 4.5 to 5.0 cm
of the hippocampus were resected with no more than 5.0 cm
of mesial resection on any patient.
Six months postoperatively the subjects underwent complete neuropsychological reevaluation, and the MAE was
again administered. Alternate forms were available for 4 of
the 7 subtests (Sentence Repetition, Controlled Oral Word
Association, Oral Spelling, Token Test).
No differences between the two groups were noted in
clinical seizure outcome. N i n e of the dominant and 10
of the nondominant hemisphere ATL patients were
seizure free (total absence of clinical seizures as well
as auras), and the remainder of each group was
significantly improved (> 75% reduction in seizure
Preoperatively there were no significant differences
between the dominant and nondominant temporal
lobe groups on the individual subtests of the MAE
(Table 2). However, the dominant temporal lobe
group performed in a more impaired fashion o n 6 of
the 7 subtests, a consistent and significant (p < 0,001)
The patients were reevaluated 6 months postoperatively and difference scores (postoperative score minus
preoperative score) for the dominant and nondominant hemisphere ATL groups were analyzed by unpaired t tests (Table 3). No significant differences were
found between the two groups on any of the MAE
subscales, nor was there an overall tendency for the
dominant hemisphere ATL subjects to show greater
loss across the MAE subtests.
Paired t tests were used to compare within-group
(dominant, nondominant) preoperative versus postoperative performance on the MAE subtests (see Table
3). Following surgery, the dominant hemisphere ATL
group showed significantly improved performance on
the T o k e n Test ( p = 0.035) and the Controlled Oral
Word Association Test ( p = 0.021). No other pre-
Table 2. Preoperative PeYfrmunce on the
Multilingual Aphasia Examination”
Visual Naming
Sentence Repetition
Controlled Oral Word
Oral Spelling
Token Test
Aural Comprehension
Reading Comprehension
ATL of
(n = 15)
ATL of
(n = 14)
43.5 (11.6)
39.3 (9.6)
37.6 (8.4)
45.7 (13.9)
46.4 (12.7)
41.4 (11.9)
“Scaled scores reported are t scores (X = 50, SD
presented are means with SD in parentheses.
10). Values
anterior temporal lobectomy.
Table 3. Preoperative versus Postoperative Difference
Scores on the Multilingual Aphasia Examination”
Visual Naming
Sentence Repetition
Controlled Oral Word
Oral Spelling
Token Test
Aural Comprehension
Reading Comprehension
ATL of
(n = 15)
-0.1 (8.0)
3.3 (9.9)
3.7 (5.5)
0.3 (5.8)
6.8 (11.3)
- 1.5 (4.5)
-0.8 (5.5)
ATL of
(n = 14)
3.1 (8.7)
3.8 (12.2)
1.3 (8.2)
3.4 (8.5)
- 1.3 (4.1)
1.9 (7.9)
Wegative scores indicate poorer postoperative performance,
whereas positive scores indicate improved performance relative to
preoperative performance. Values presented are means with standard deviations in parentheses.
A l l = anterior temporal lobectomy.
operative versus postoperative comparisons reached
statistical significance for either group.
In addition to the above analyses that inquired into
group performances, we conducted several subsequent analyses to determine whether there were any
significant deviations in the performances of individuals who underwent dominant hemisphere ATL. To
conduct these analyses, first we operationally defined
an alteration in performance of -+ 1 standard deviation
or greater as a change in performance of probable clinical significance. Each subject’s preoperative versus
postoperative performance on the language battery
was then classified into one of three categories: (1)
showing more significant improvements than losses;
(2) showing more significant losses than improvements; and (3) exhibiting an equal number of gains and
losses. Primarv interest was of course in the outcome
of the dominant hemisphere ATL patients, with the
nondominant hemisphere ATL. patients serving as control subjects.
There was no significant difference between dominant and nondominant hemisphere ATL patients in
the individual patterns of language outcome (x2 =
0.46, df = 2). Additionally, there were no significant
differences between the dominant and nondominant
hemisphere ATL patients in the overall number of
significant losses on the MAE subtests (x2 = 4.02, df
= 2) or in the total number of subtests falling in the
impaired range postoperatively.
Among the dominant hemisphere ATL patients,
language outcome was independent of surgical outcome (improved versus seizure free) (x2 = 1.64, df =
2, ns) or whether the patient underwent more or less
than the mean amount of lateral resection (x2 = 3.65,
df = 2 , ns).
In summary, there were individual differences in the
pattern of language outcome using the classification
scheme outlined above. However, these patterns were
unrelated to whether surgery was carried out on the
dominant or nondominant temporal lobe and were
also unrelated to a host of other variables.
Four aspects of this investigation are noteworthy. First,
the results of the preoperative assessment of language
function revealed a significant trend indicating poorer
overall language performance in patients with complex
partial seizures of dominant temporal lobe origin relative to those of nondominant temporal lobe origin.
Although the dominant hemisphere patients generally
performed less adequately, none of the differences
on the individual MAE subtests reached statistical
significance. These findings could be interpreted as
suggesting that there is mild generalized impairment of
language function in dominant temporal lobe patients
relative to nondorninant patients. This mild generalized impairment would not be unexpected in the case
of epileptiform dysfunction of the dominant temporal
Second, partial resection of the dominant anterior
temporal lobe did not result in any significant losses in
language ability. Tailored resection of the dominant
temporal lobe, based on findings derived from intraoperative electrocorticography and speech mapping,
resulted in improved seizure control with preservation
of preoperative levels of language ability, just as was
found to be the case for ATL of the nondominant
Third, following ATL of the dominant temporal
lobe those patients showed significantly improved performance on two MAE subtests: the Token Test (a
measure of receptive language comprehension) and
Hermann and Wyler: Language Function and Temporal Lobectomy 587
the Controlled Oral Word Association Test (a measure
of associative verbal fluency). Alternate forms of both
tests were used so the postoperative gains could not be
attributed to practice effects, especially when such
gains were not shown by the nondominant hemisphere
ATL patients. The cortical areas traditionally hypothesized to be important for successful task performance include the dominant posterior (superior)
temporal lobe (Token Test), and the dominant frontal
lobe (Controlled Oral Word Association Test).
It may be reasonable to hypothesize that resection
of the epileptogenic focus in the dominant anterior
temporal lobe resulted in a marked reduction in the
amount of abnormal epileptiform activity propagated
via association pathways to posterior temporal and
frontal (as well as other) brain regions, which thereby
allowed these cortical areas to express their neuropsychological function more normally. It has been hypothesized that postoperative reduction in interictal
“neural noise” following ATL is responsible for observed improvements in memory functions mediated
by neuronal systems contralateral to the surgical hemisphere 1111. Our findings may represent an instance in
which select ipsilateral cognitive functions appear to be
improved following ATL. Future prospective investigations of preoperative and postoperative neuropsychological performance in larger samples of dominant
and nondominant hemisphere ATL patients using preoperative and postoperative electroencephalographic
findings will be needed to evaluate this “neural noise”
Fourth, in the context of the group findings reviewed above, the expected individual variability in
postoperative language performance was apparent. Examining individual patterns of language performance,
we did not find any evidence of subtle or systematic
language loss associated with dominant hemisphere
ATL. Further, among patients who underwent ATL of
the dominant temporal lobe, the individual variability
in language outcome was not associated with gender,
surgical outcome, or amount of lateral cortical resection. It should be noted that the latter two variables
were rather truncated in their distribution; that is,
there were no patients who were not at least significantly improved (275% reduction in seizure frequency), and the mean amount of lateral dominant
temporal lobe resection was conservative (2 = 4.7
cm). Perhaps other samples, with a wider distribution
of these and other variables, can better account for
individual variability. Overall, we believe that one interpretation of our findings, in total, is that in the context of the group findings reviewed previously (i.e.,
588 Annals of Neurology
Vol 23 No 6 June 1988
signals), considerable individual variability existed (i.e.,
noise). We were not able to find predictors of this
“psychometric noise” in the analyses that were conducted and could not find evidence of subtle language
loss associated with dominant hemisphere ATL.
T o our knowledge, this study represents the first
prospective investigation of basic language function
following ATL. Given the methodology described in
this investigation, ATL was found to have no adverse
effects on language function in either dominant or
nondominant hemisphere ATL groups and to have
some positive effects for dominant hemisphere ATL
patients. These findings suggest that ATL can be conducted on patients who harbor an epileptogenic focus
within their dominant temporal lobe without significant compromise of language function, thereby
reinforcing clinical impressions that have long been
reported from surgical centers { 123.
We sincerely thank Roger Vander Zwagg, PhD, Health Services
Research, Baptist Memorial Hospital, Memphis, TN, for statistical
consultation and for conducting the analyses of our data. We also
thank Ruby Green for typing and processing this manuscript.
1. Rapport RL, Ojemann GA, Wyler AR, Ward AA Jr. Surgical
management of epilepsy. West J Med 1977;127:185-189
2. Green JR, Scheetz DG. Surgery of epileptogenic lesions of the
temporal lobe. Arch Neurol 1964;10:135-148
3. Rasmussen TB. Surgical treatment of complex partial seizures:
Results, lessons, and problems. Epilepsia (suppl 1) 1983;
4. Heilman KM, Wilder BJ, Malzone WF. Anomic aphasia following anterior temporal lobectomy. Trans Am Neurol Assoc
5. Lifrak MD, Novelly RA. Language deficits in patients with temporal lobectomy for complex partial seizures. In: Porter RJ, et
al, eds. Avances in neurology: XVth epilepsy international symposium. New York: Raven, 1984:469-473
6. Cherlow DG, Serafetinides EA. Speech and memory assessment in psychomotor epileptics. Cortex 1976;12:21-26
7. Blume WT, Grabow JD, Darley FL., et al. Intracarotid amobarbital test of language and memory before temporal lobectomy
for seizure control. Neurology 1973;23:812-819
8. Wyler AR, Ojemann G, Lettich E, Ward AA. Subdural strip
electrodes for localizing epileptogenic foci. J Neurosurg
9. Benton AL, Hamsher K. Multilingual aphasia examination.
New York: Oxford University Press, 1983
10. Let& M. Neuropsychological assessment (2nd ed.). New York:
Oxford University Press, 1983
11. Novelly RA, Augustine EA, Mattson RH, et al. Selective memory improvement and impairment in temporal lobectomy for
epilepsy. Ann Neurol 1984;23:64-67
12. Penfield W, Roberts L. Speech and brain mechanisms. Princeton, NJ: Princeton University Press, 1959
Без категории
Размер файла
453 Кб
effect, language, stud, controller, temporal, function, anterior, lobectomy
Пожаловаться на содержимое документа