-
Journal of Neurology, Neurosurgery, and Psychiatry
1986;49:635-639
Management of subdural intracranial empyemasshould not always
require surgeryD LEYS, A DESTEE, H PETIT, P WAROT
From the Department of Neurology, University of Lille, Lille,
France
SUMMARY Seven patients with subdural empyema were initially
treated by antibiotics withoutsurgery. Six have recovered without
sequelae. One required delayed surgery and has recovered
withepilepsy. The authors emphasise the use of CT for the diagnosis
and follow-up of subdural empy-ema, the principles and modalities
of non-surgical treatment, and the good results, especially for
latemorbidity.
From the medical literature, it is evident that, evenwhen
antibiotics became available, most authors haveagreed there is need
for surgery in all intracranialsubdural empyemas.' However, this
treatment hasnot prevented serious mortality'4 and sequelaes.'
sSince the use ofCT for the diagnosis of various centralnervous
system suppurations,6 7 some authors havetreated brain abscesses, 6
8 -10 extradural/intracranialabscesses" and even spinal epidural
abscesses7 with-out surgery. We report here the results obtained
with7 patients with subdural empyema treated withoutsurgery.
Case reports
The first three patients have been the subject of a
previousreport in a review about clinical and radiological findings
insubdural empyemas. The second patient's history is reportedin
detail, and the six others are summarised in the table.
Eachpatient's CT scan, before and after treatment, is shown in
thefigs 1-5.
Patient 2This 19-year-old woman had a 1 week history of fever
andbifrontal headache and received each day amoxicilline (1
gorally) for 4 days. She was admitted to the neurologicaldepartment
on 13 April 1982 with fever (38°5), headache andvomiting. She was
lethargic, with a left hemiplegia and apalsy of both external
recti. Her neck was stiff. Generalizedseizures occurred. CSF
contained 900 white cells/mm3(100% polymorphonuclear), protein
0-6mg/l and glucose0-6 g/l. ESR was 120 mm/h and WBC count was
15000(80% polymorphonuclear). CT (fig 2a) revealed
aninterhemispheric area of low density with an enhanced thin
Address for reprint requests: Pr P Warot, MD, Department
ofNeurology A, Hospital B, 2 avenue Oscar Lambret, F59037
LilleCedex, France.
Received 12 April 1985 and in revised form 18 September
1985.Accepted 21 September 1985.
margin after contrast and compression of cerebral andventricular
structures. No causative organism was isolatedfrom CSF or blood
cultures. Skull radiographs showedopacification of the right
maxillary and frontal sinuses.The patient was treated by ampicillin
(12gIV), sisomicine(150 mg IM) and trimethoprim-sulfamethoxazole
(320mg-1600mg IV) for 6 weeks, then by oral amoxicillin 6gdaily for
4 months. Clonazepam (3 mg IV) was added duringthe first 48 hours
and mannitol during the first 5 days. Whenthe treatment was
stopped, she had no neurological deficitand CT scan (fig 2b) showed
no abnormalities. Thirtymonths later, she had had no seizure, and
did not receiveanticonvulsant medication.
Discussion
The use of CTfor diagnosis andfollow-upAs in brain abscesses,6 a
non-surgical treatment ofsubdural empyemas is possible only ifCT
can be per-formed. It reveals small empyemas which could not
bediagnosed otherwise, as in the second case reported byRosazza:'3
this was a patient with purulent meningitisand without any focal
deficit, in whom CT showed asmall subdural empyema. CT also allows
easy andatraumatic follow-up.2 14
The classical treatment of subdural empyemasFor most authors,
surgery is always required in allsubdural empyemas: they often
prefer a largecraniotomy' 3 15 16 to burr holes, so as to
providepurulent material and allow an irrigation of the sub-dural
space with antibiotics.'6 For these authors, thesurgical treatment
must be performed in emergency,but Pimontel-Appel'4 prefers to wait
for an im-provement of the neurological state, 24 to 48 hoursafter
the onset of antibiotic therapy. In spite of thepossibility of
improvement with surgical treatment,the importance of antibiotics
cannot be neglected.The first successfully treated cases occurred
only
635
by copyright. on July 4, 2021 by guest. P
rotectedhttp://jnnp.bm
j.com/
J Neurol N
eurosurg Psychiatry: first published as 10.1136/jnnp.49.6.635 on
1 June 1986. D
ownloaded from
http://jnnp.bmj.com/
-
Table Cases of non-surgically treated subdural empyemas
Authors Sex and Clinical signs CSF ESRmm/h WBC/mm3 CT
scanage(yr) Cells P G
Rasazza13 M 14 fever 33 1200 720 17400 IH SDEheadacheleft
hemiplegia
Rasazza13 M 11 fever 316 1500 721 ? 17300 IH SDEmeningism
Rousseaux20 M 24 fever 255 ? ? IH SDEmeningismleft
hemiplegia
Case 1 M 29 fever 830 800 500 26 11600 IH SDE802284 focal
seizures left PA
generalised status SDEwith coma right (fig 1)hemiplegia and
aphasia
Case 2 F 19 fever 900 600 600 120 15000 IH SDE820869 generalised
seizures (fig 2)
left hemiplegiaVI palsy meningism
Case 3 M 22 fever 50 550 700 85 15100 FR SDE821121 focal
seizures (fig 3)
comaright hemiplegia
Case 4 M 40 fever 5 400 800 60 13000 FR SDE830286 generalised
seizures (fig 4)
right hemiplegiameningism
Case 5 M 56 fever 50 400 500 100 22000 whole convexity SDE831075
generalized status (fig 5)
with coma meningismright hemiplegia
Case 6 M 15 fever 80 300 500 15 6000 FR SDE840544 meningism (fig
6)
Case 7 F 24 fever 100 1100 740 80 12000 IH SDE841050 focal and
generalised TE SDE
seizures with coma (fig 7)meningism aphasia
M, male; F, female; P, proteins (mg/i); G, glucose (mg/i); SDE,
subdural empyema; IH, interhemispheric; PA, parietal; FR, frontal;
TE, temporal;T R, total recovery; Ampi, ampicillin; TMP-SMX,
trimethoprime-sulfamethoxazole; Siso, sisomicine; Amox,
amoxicilline; Metro, metronidazole;PRIS, pristinamycine.
after the introduction of penicillin.'7 More recently,ampicillin
and especially chloramphenicol have beenpreferred, because of their
good diffusion into thecentral nervous system'3 1' and their
effectiveness onanaerobic organisms, which are frequently
isolatedfrom subdural empyemas.18
rrWhy have we tried a non-surgical treatment?Surgery has usually
been performed as an emergencybecause of two objections to a
non-surgical treatment:firstly, antibiotics do not penetrate into
loculated in-tracranial suppurations, and secondly, it is
necessaryto know the causative organism and its sensibility to
Fig I (Patient 1) (a) On admission: interhemispheric and left
parietal subduralempyema. (b) After a 30 day course of antibiotics:
disappearance of the most part ofthe empyema, but increase of the
posterior part. (c) A year later: no residual empyema.
636 Leys, Destee, Petit, Warot
by copyright. on July 4, 2021 by guest. P
rotectedhttp://jnnp.bm
j.com/
J Neurol N
eurosurg Psychiatry: first published as 10.1136/jnnp.49.6.635 on
1 June 1986. D
ownloaded from
http://jnnp.bmj.com/
-
Management of subdural intracranial empyemas should not always
require surgery
Point of entry Organism Treatment Clinical outcome
Antibiotics Associated medications
sinusitis unknown Ampi Dexamethazone TRChloramphenicol(3 weeks
IV5 weeks orally)
sinusitis staphylococcus Ampi IV sinusitis drainage TR6
weeks
sinusitis unknown Ampi IV Tetracosactide TR6 weeks sinusitis
drainagesinusitis streptococcus Ampi IV Mannitol seizures
TMP-SMX 6 weeks ClonazepamSiso IM J craniotomy
(26th day)
sinusitis unknown Ampi Mannitol TRTMP-SMX 6 weeks ClonazepamSiso
f
Amox 4 monthssinusitis unknown Ampi Mannitol TR
TMP-SMX 8 weeks ClonazepamSiso JAmox 5 months
sinusitis unknown Ampi Mannitol TRTMP-SMX 4 weeks
ClonazepamMetro J sinusitisAmox 3 months drainage
otitis unknown Ampi Mannitol TRTMP-SMX 4 months ClonazepamSiso J
surgical treatment
otitissinusitis unknown Ampi } 5 weeks TR
Amox 4 monthspost-traumatic unknown Ampi \4 weeks Mannitol
TRsinusitis TMP-SMX > Clonaxepam surgical
PRIS 3 months treatment ofpost-traumatic sinuslesion
antibiotics. To our knowledge, antibiotic has neverbeen found in
the pus of subdural empyemas, as it hasin brain abscesses;"9
nevertheless, in our cases 2 to 7,antibiotics were sufficient to
improve the patients'state and to normalise the CT scan. In two of
Rosa-zza's cases13 and in Rousseaux' case,20 antibioticshad also
been able to cure such lesions. Our firstpatient was surgically
treated one. month after the
Fig 2 (Patient 2) (a) On adnission: interhemisphericsubdural
empyema. (b) During the 6th month: CT scan isnormal.
onset of the antibiotherapy: his neurological condi-tion had
improved, but surgery was decided becausethe size of the most
posterior part of the empyema hadgradually increased; this patient
was our firstmedically treated patient but now, with Rosazza
andRousseaux' experiences,1320 and from our next sixpatients, we
think that it would have been possible totreat him without surgery.
In Kaufman' 21 and
0rT)_IL ieLN&:Fig 3 (Patient 3) (a) On admission: frontal
subduralempyema. (b) Eight months later: no residual empyema.
637
by copyright. on July 4, 2021 by guest. P
rotectedhttp://jnnp.bm
j.com/
J Neurol N
eurosurg Psychiatry: first published as 10.1136/jnnp.49.6.635 on
1 June 1986. D
ownloaded from
http://jnnp.bmj.com/
-
Leys, Destee, Petit, Warot
a.... .....
Fig 4 (Patient 4) (a) On admission: interhemisphericsubdural
empyema. (b) Four months later. no residualempyema.
Holtzman' 22 cases, the neurological condition deteri-orated in
spite of antibiotics. We are not sure whetherthe dosage was
sufficient, but, in our cases, althoughthe patients often showed a
little deterioration duringthe first 24 or 48 hours, we always
continued the sametreatment. So, we think that antibiotics are
probablyable to penetrate into subdural empyemas. The pushas been
free of organisms in our patient 1 and re-ported by Borzone et
al.23 This penetration is perhapsmade possible by an unusual
development of men-ingeal arteries, as in our third case, which
broughtlarge quantities of antibiotics in the margin of theempyema.
24 25
It is not always necessary to know the causativeorganism from
the empyema itself. In our seven cases,the causative organism was
found only in the firstpatient, from blood cultures. Surgery is not
indicatedfor identification of the organism as in brain abs-cesses;
this is possible in less than 50% of the operatedcases, and in 30%
of the non-surgically treated ones,from blood or CSF cultures, or
from the point ofentry; moreover, the organism is, in most cases,
sensi-tive to large spectrum antibiotics used intravenouslywith
high doses.
Modalities of the medical treatmentWe have used intravenous
antibiotic therapy for 4 to
Fig 6 (Patient 6) (a) On admission: right frontalsubdural
empyema. (b) Four months later: no residualempyema.
6 weeks and oral antibiotics until the CT scan wasnormal in all
cases except the first in which the patienthimself stopped the
treatment in the sixth week. Wethink it is possible to stop
earlier, as in Rosazza'cases,'3 but care is required to ensure
sterilisation ofthese lesions. Clonazepam was used when
generalisedseizures occurred. Corticosteroids have been
avoidedduring the acute phase as they prevent antibioticsfrom
penetrating into the abscesses.8 To prevent oe-dema, 10% hypertonic
mannitol was used during thefirst few days. Of course, surgical
treatment mighthave been necessary for patients who were
rapidlydeteriorating neurologically with medical treatment.However,
in our second and seventh cases, a littledeterioration did not lead
to surgery. In four cases,antibiotics alone were sufficient to
treat the initialinfection of paranasal sinuses; in one case,
delayedsurgery prevented relapse and in two cases, earlysurgery was
necessary to treat the paranasal sinusitis.
Results
With classical treatment, associating emergencysurgery and
antibiotics, the mortality washigh' 23 26 - 28 and sequelae (focal
deficits or epilepticseizures) were frequent.' 5 In our cases, only
one pa-tient had sequelae, (generalised seizures) and he had
_..1:.
....
Fig 5 (Patient S) (a) On admission: subdural empyemaof the whole
convexity. (b) One year later: CT scan isnormal.
638
by copyright. on July 4, 2021 by guest. P
rotectedhttp://jnnp.bm
j.com/
J Neurol N
eurosurg Psychiatry: first published as 10.1136/jnnp.49.6.635 on
1 June 1986. D
ownloaded from
http://jnnp.bmj.com/
-
Management of subdural intracranial empyemas should not always
require surgery
been operated upon. After 6 to 30 months, the othersix have no
focal deficit or seizure. The three otherpatients previously
reported13 20 in the literature, alsohave had no sequelae.
For these 10 cases, summarised in the table, themorbidity and
mortality obtained by medical treat-ment alone seem better than
those by surgical treat-ment,28 as also shown in brain abscesses.'0
Manystudies have shown that the most important prognos-tic factor
in intracranial infection is the level of con-sciousness when the
treatment is commenced. Threepatients (cases 2, 4, 6) were not in
coma and they maytherefore have been expected to have a better
prog-nosis, no matter how they were treated. Nevertheless,the four
others were in coma, and three had totalrecovery, and one little
sequelae. In the literature, withsurgery, the mortality and
morbidity seem higher.28A long period of intravenous treatment may
be a
financial disadvantage as compared with perhaps amore rapid
response to surgical drainage, leading toearlier discharge and
cheaper overall treatment; nev-ertheless, shorter treatments are
possible, as in ourfirst patient, and it would be possible to
dischargethese patients earlier in the future, when our experi-ence
will be greater. Moreover, less sequelae is also afinancial
advantage.
The authors thank Mr Francois Leung for the re-vision of the
English manuscript.
References
1Bannister G, Williams B, Smith S. Treatment of subduralempyema.
J Neurosurg 1981;55:82-8.
2Hadj-Djilana M, Calliauw L. A contribution to the
RapidDiagnosis of Subdural Empyema. Acta Neurochir(Wein)
1982;61:187-99.
3Stern J, Bernstein CA, Whelan MA, Neu HC. Pasteurellamultocida
subdural empyema: Case Report. J Neuro-surg 1981;54:550-2.
'Kaufman DM, Miller MH, Steigbigel NH. Subduralempyema: analysis
of 17 recent cases and review of theliterature. Medicine
1975;54:485-98.
sCowie R, Williams B. Late seizures and morbidity aftersubdural
empyema. J Neurosurg 1983;58:569-73.
6Petit H, Rousseaux M, Lesoin F, Destee A, Clarisse J,Warot P.
Primaut6 du traitement medical des abcescer6braux (19 cas). Rev
Neurol (Paris) 1983;139:575-81.
7Leys D, Lesoin F, Viaud C, et al. Decreased morbidityfrom acute
bacterial spinal epidural abscesses usingcomputed tomography and
non surgical treatment inselected patients. Ann Neurol
1985;17:350-5.
'George B, Roux F, Pillon M, Thurel C, George C. Rele-vance of
antibiotics in the treatment of brain abscesses.Report ofa case
with eight simultaneous brain abscessestreated and cured medically.
Acta Neurochir (Wein)
1979;47:285-91.Petit H, Dest&e A, Leys D, Warot P.
Volumineux abces
listerien du tronc cerebral. Effet favorable
del'antibiotherapie. Rev Neurol (Paris) 1983;139:149-54.
°Rousseaux M, Lesoin F, Destee A, Jomin M, Petit H.Long term
sequelae of hemispheric abscesses as func-tion ofthe treatment.
Acta Neurochir (Wein). (in press).
l Leys D, Destee A, Warot P. Empyeme extradural en
fosseposterieure. Traitement medical exclusif. Presse Med1983;12:
1549.
12 Leys D, Dest&e A, Combelles G, Rousseaux M, Warot P.Les
empyemes sous-duraux intracraniens. Trois obser-vations. Sem Hop
(Paris) 1983;59:3347-50.
13Rosazza A, De Triboulet N, Deonna TH. Non surgicaltreatment
ofinterhemispheric subdural empyemas. HelvPaed Acta
1979;34:577-81.
14Pimontel-Appel B, Bochner A, Ghevens J, Klaes R, BrossJ,
Ebinger G. Subdural empyema: improvement of theemergency diagnosis
by use of computed tomography(CT). Two personal cases study. J
Belge Radiol1980;63:109-17.
s Le Beau J, Creissard P, Harispe L, Recondo A.
Surgicaltreatment of brain abscess and subdural empyema. JNeurosurg
1973;38:198-203.
16Smith HP, Hendrick EB. Subdural empyema and epiduralabscess in
children. J Neurosurg 1983;58:392-7.
17Schiller F, Cairns H, Russel DS. The treatment of puru-lent
pachymeningitis and subdural suppuration withspecial reference to
penicillin. J Neurol NeurosurgPsychiatry 1948;11:143-82.
'8Yoshikawa TT, Chow AW, Guze LB. Role of anaerobicbacteria in
subdural empyema. Report offour cases andreview of 327 cases from
the English literature. Am JMed 1975;58:99-104.
Black P, Graybill JR, Charache P. Penetration of brainabscess by
systemically administered antibiotics. JNeurosurg
1973;38:705-9.
20Rousseaux M, Lesoin F, Jomin M. Traitement medicald'un empyeme
sous-dural parasagittal. Presse Med1984;13:2153.
21 Kaufman DM, Litman N, Miller MH. Sinusitis inducedsubdural
empyema. Neurology (Cleveland) 1983;33:123-32.
22 Holtzman RNN, Tepperberg J, Schwartz 0. Parasagittalsubdural
empyema: a case report with computerizedtomographic scan
documentation. Mount Sinai J Med1980;47:62-7.
23Borzone M, Capuzzo T, Rivano C, Tortori-Donati P.Subdural
empyema: fourteen cases surgically treated.Surg Neurol
1980;13:449-52.
24Handa J, Hanakita J, Koyama T, Handa H. Inter-hemispheric
subdural empyema with an enlarged ten-torial artery and vein.
Neuroradiology 1976;9:167-70.
2SKim KS, Weinberg PE, Magidson M. Angiographic fea-tures of
subdural empyema. Radiology 1976;118:621-5.
26Bhandari YS, Sarkari NBS. Subdural empyema. A reviewof 37
cases. J Neurosurg 1970;32:35-9.
27Weinman D, Samarasinghe HHR. Subdural empyema.Aust NZ J Surg
1972;41:324-30.
28Williams B. Subdural empyema. In: Krayenbuhl H, ed.Advances
and Technical Standards in Neurosurgery, Vol9. New York:
Springer-Verlag, 1983:133-70.
639
by copyright. on July 4, 2021 by guest. P
rotectedhttp://jnnp.bm
j.com/
J Neurol N
eurosurg Psychiatry: first published as 10.1136/jnnp.49.6.635 on
1 June 1986. D
ownloaded from
http://jnnp.bmj.com/