1. health and fitness
2. disease

Neurosurg Rev (2000) 232:175–204
© Springer-Verlag 2000
REVIEW
E. Reihsaus · H. Waldbaur · W. Seeling
Spinal epidural abscess: a meta-analysis of 915 patients
Received: 18 July 2000 / Accepted: 14 September 2000
Abstract Spinal epidural abscess (SEA) was first described in the medical literature in 1761 and represents
a severe, generally pyogenic infection of the epidural
space requiring emergent neurosurgical intervention to
avoid permanent neurologic deficits. Spinal epidural
abscess comprises 0.2 to 2 cases per 10,000 hospital
admissions. This review intends to offer detailed evaluation and a comprehensive meta-analysis of the international literature on SEA between 1954 and 1997, especially of patients who developed it following anesthetic
procedures in the spinal canal. In this period, 915 cases
of SEA were published. This review is the most comprehensive literature analysis on SEA to date. Most
cases of SEA occur in patients aged 30 to 60 years, but
the youngest patient was only 10 days old and the oldest was 87. The ratio of men to women was 1:0.56. The
most common risk factor was diabetes mellitus, followed by trauma, intravenous drug abuse, and alcoholism. Epidural anesthesia or analgesia had been performed in 5.5% of the patients with SEA. Skin abscesses and furuncles were the most common source of infection. Of the patients, 71% had back pain as the initial symptom and 66% had fever. The second stage of
radicular irritation is followed by the third stage, with
beginning neurological deficit including muscle weakness and sphincter incontinence as well as sensory defiInaugural dissertation, University of Ulm 2000, supported by the
University Clinic of Anesthesiology and the Neurosurgical Clinic
of the German Military Hospital of Ulm
E. Reihsaus
Ludwigsburg-Bietigheim General Hospital,
Department of Anesthesiology, Riedstrasse 12,
74321 Bietigheim-Bissingen, Germany
H. Waldbaur
Neurosurgical Clinic, German Military Hospital,
Oberer Eselsberg 40, 89081 Ulm, Germany
W. Seeling
University Clinic of Anesthesiology,
Section of Pain Therapy, University of Ulm,
Steinhövelstrasse 9, 89075 Ulm, Germany
cits. Paralysis (the fourth stage) affected only 34% of
the patients. The average leukocyte count was
15,700/µl (range 1,500–42,000/µl), and the average
erythrocyte sedimentation rate was 77 mm in the first
hour (range 2–50 mm). Spinal epidural abscess is primarily a bacterial infection, and the gram-positive Staphylococcus aureus is its most common causative
agent. This is true also for patients who develop SEA
following spinal anesthetics. Magnetic resonance imaging (MRI) displays the greatest diagnostic accuracy and
is the method of first choice in the diagnostic process.
Myelography, commonly used previously to diagnose
SEA, is no longer recommended. Lumbar puncture to
determine cerebrospinal fluid protein concentrations is
not needed for diagnosis and entails the risk of spreading bacteria into the subarachnoid space with consequent meningitis; therefore, it should not be performed.
The therapeutic method of choice is laminectomy combined with antibiotics. Conservative treatment alone is
justifiable only for specific indications. Laminotomy is
a therapeutic alternative for children. The mortality of
SEA dropped from 34% in the period of 1954–1960 to
15% in 1991–1997. At the beginning of the twentieth
century, almost all patients with SEA died. Parallel to
improvements in the mortality rate, today more patients
experience complete recovery from SEA. The prognosis of patients who develop SEA following epidural anesthesia or analgesia is not better than that of patients
with noniatrogenic SEA, and the mortality rate is also
comparable. The essential problem of SEA lies in the
necessity of early diagnosis, because only timely treatment is able to avoid or reduce permanent neurologic
deficits.
“The problem with spinal epidural abscesses is not
treatment, but early diagnosis – before massive neurological symptoms occur” (Strohecker and Grobovschek
1986).
Keywords Epidural abscess · Meta-analysis ·
Epidemiology · Outcome
176
Introduction
Spinal epidural abscess (SEA) represents a neurosurgical
emergency requiring immediate action. It usually presents
as a suppurative process localized between the spinal dura
mater and the vertebral periosteum within the spinal epidural space (Fig. 1).
According to Strohecker and Grobovschek [361], “the
problem with spinal epidural abscesses is not treatment,
but early diagnosis – before massive neurological symptoms occur.” Almost all reviews and case series on spinal
epidural abscess emphasize this point [8, 16, 31, 67, 80,
82, 98, 154, 158, 164, 179, 189, 195, 201, 213, 214, 230,
232, 235, 262, 264, 283, 297, 300, 306, 309, 314, 316,
330, 350, 366, 378, 401]. Immediate therapy is necessary
to prevent compression of the spinal cord and cauda equina, with the corresponding neurological lesions (Fig. 2).
Early surgical decompression and abscess drainage supported by broad antibiotic coverage can lead to complete
recovery.
Intramedullary [20, 238] and subdural spinal abscesses [54, 111, 255, 275, 279, 367] occur only rarely. In
contrast, recent studies seem to document an increasing
incidence of purulent infections in the spinal epidural
space [67, 73, 154, 195, 314]. In fact, the authors of a
study on the MRI characteristics of cervical SEA from
Fig. 1 Cross-section through the fourth cervical vertebra. Meninges and epidural space. (With permission from [209] p 108)
1994 commented: “In our experience, this disorder is
more common than previously reported” [112]. Although
the prevalence may be increasing, the incidence of 0.2 to
two cases per 10,000 hospital admissions [16, 154], or
less than one case per million residents [164], is still
quite low. It may present with back pain as the initial
symptom (see “Symptoms and laboratory findings”);
however, back pain also occurs due to a large number of
other disorders. In Great Britain, the yearly incidence of
back pain is estimated to be 28,000 cases per million residents. This helps explain the problem of SEA as described by Strohecker and Grobovschek [361] and others. “Early diagnosis” [361] is only possible if SEA is
considered in the differential diagnosis of back pain,
which in turn is dependent on familiarity with the manifestations of SEA.
The advanced age of the general population may be
responsible for the greater prevalence of SEA, since aging is associated with an increase in risk factors (see
“Pathogenesis and risk factors”). Abuse of intravenous
drugs, especially in the United States, also contributes to
the higher prevalence of SEA, since bacterial contamination of syringes, needles, and other equipment can lead
to SEA by hematogenous spread. More immunocompromised individuals in the last two decades have also contributed to the growing incidence of SEA.
The use of regional spinal and epidural anesthesia –
even in patients with comorbidity factors such as diabetes mellitus – may be one reason for the observed prevalence of SEA according to Du Pen et al. [87], Ngan Kee
177
Fig. 2 Comparison between spinal epidural (left) and subdural
(right) abscess. (With permission from [164] p 105)
et al. [256], and Pegues et al. [278], who predict a continued increase in cases of SEA in the future. However,
not all clinicians share this view [56, 78, 119, 167, 257,
360, 392].
This review intends to offer a comprehensive overview and critical analysis of the literature on SEA in the
form of a meta-analysis. Case reports and case series
from 1954 to 1997 were statistically evaluated with respect to epidemiology, comorbidity factors, symptomatology, causative infectious agents, diagnostic and therapeutic procedures, and outcome of treatment. Several excellent review articles on the neurologic complications
following central nerve block including epidural anesthesia have been published [25, 274, 290, 302, 376, 377]. In
addition, the literature review of Michel et al. [241] from
1997 offers an up-to-date synopsis of the relationship between epidural anesthesia and SEA. These case reports
and case series were also included in the present analysis. Additionally, all reports of SEA not due to nerve
block anesthesia were also included. These cases represent the majority of the publications on the topic.
A total of 915 cases of SEA published between 1964
and 1997 were identified for the present work, which is
the most comprehensive literature review on SEA to
date. By 1969, somewhat more than 300 cases of SEA
had been published [253]. A dissertation from 1962
[333] in the Cologne University Department of Neurosurgery evaluates results obtained with three patients as
well as only 100 other cases from the literature, with
special attention to reports of SEA following vertebral
osteomyelitis. In the present work, the numerous published cases of SEA following purulent or tuberculous
vertebral osteomyelitis are not included [7, 10, 19, 29,
53, 62, 75, 88, 109, 130, 132, 135, 139, 144, 162, 165,
169, 171, 190, 205, 207, 219, 228, 236, 272, 281, 282,
285, 307, 340, 344, 363, 364, 371, 387]. These cases
were excluded from the present analysis because SEA
following purulent or tuberculous vertebral osteomyelitis
is clearly different from SEA of other etiology with respect to therapy and prognosis, with vertebral rather than
epidural infection being the primary clinical problem.
Also, patients with tuberculous or osteomyelitic SEA
don’t represent the majority of patients in the western
hemisphere.
The history of spinal epidural abscess
The first case of SEA in the medical literature was published in 1761 in Venice [252] by the famous Italian
anatomist Giovanni Battista Morgagni, who described a
40-year-old man with severe pain and paralysis of the
lower extremities in his work “De sedibus et causis
morborum per anatomem indagatis” (Locations and
causes of diseases detected by the art of anatomy). The
results of the autopsy demonstrated that the disorder was
SEA. Sixty years later, in 1820, another lethal case of
SEA was presented by Bergamaschi [28], an Italian physician living in Paris.
Albers, a member of the Bonn medical faculty, published one of the first German-language case reports of
SEA in 1833 [6] under the title “Die Entzündung der
harten Haut des Rückenmarks, Perimeningitis medullae
spinalis” (Inflammation of the dura mater of the spinal
cord, perimeningitis medullae spinalis). He described
two cases that would today be diagnosed as SEA. It is
astounding that the report published in 1833 offers an
exact description of the symptomatology that hardly differs from descriptions in modern publications:
“Both of the cases described here displayed the same
predominant symptoms: (1) Severe pain in the entire
lower extremities and the lower part of the trunk and abdomen. (2) Convulsions. (3) Trembling. (4) Difficulties
with urination and defecation. (5) The sensation of a
band around the body.”
178
The “convulsions” and “trembling” described by Albers
[6] are likely to represent the radiculopathic stage of SEA
(see “Clinical and laboratory findings”).
In 1877, Lewitzky [210] of Warsaw and, in 1879,
Spencer [353] of Bristol published other cases of SEA
with lethal clinical courses.
Until the 1930s, there was a lack of consistency in the
terminology used to describe SEA, which was described
as peripachymeningitis spinalis [210], pachymeningitis
(spinalis) externa (purulenta) [72, 113, 150, 243, 329,
331, 353], perimeningitis [159, 175], extradural spinal
suppuration [268], (acute) purulent peripachymeningitis
[145, 251], perimeningitis purulenta [153], extradural
abscess [355], epimeningitis spinalis [42], and extrathecal abscess [368]. As late as 1928, Bensheim [27], a physician at the Internal Medicine Department of the Heidelberg University, then under the direction of professor
Freiherr von Weizsäcker, presented a case of SEA using
the terminology peripachymeningitis spinalis externa purulenta. In 1931, Pollak [286] used the term perimeningitis, although the autopsy report presented in his article
clearly documented that the leptomeninges were
“smooth and free of exudates.” Mixter [247] was the first
to use the concept of epidural (intra)spinal abscess.
Irrespective of the terminology preferred by the different authors, it is remarkable how consistent their descriptions of the disease manifestations are. The connection between distant sources of infection and the occurrence of SEA (see “Pathogenesis and risk factors”) was
already recognized in the early days of the new science
of bacteriology [247, 268, 329].
Furuncles were not the only cause of SEA to be identified. Nonne [258] viewed “otogenic suppuration” as
causal for the development of SEA, and in 1921 Hinz
[153] described the first recognized connection between
the development of SEA and the postpartal period.
Despite the progress in knowledge of the etiology and
symptoms of this disorder, the lack of treatment possibilities led to a more or less nihilistic attitude towards the
therapy until the beginning of the twentieth century. Accordingly, the medical literature on SEA in the nineteenth century consists almost completely of autopsy reports [6, 210, 353]. Unfortunately, until the 1930s most
patients with SEA died [27, 42, 113, 145, 150 153, 243,
247, 251, 268, 286, 329, 355]. In 1926, Nonne [258]
wrote: “The prognosis of spinal cord abscesses is extremely poor.” However, he was mistaken to claim that
no case had been cured up to that date. The American
authors Taylor and Kennedy [368] reported one successfully treated case of SEA, and Dandy [72] described
three other therapeutic successes in 1926. The method of
treatment was operative laminectomy (see “Treatment”).
The first operative intervention, laminectomy, was reported in 1892 by the French physician Delorme (quoted
in Ref. 331). The laminectomy was performed from the
seventh to the 11th thoracic vertebrae and revealed a
“spongy, not especially adherent mass on the dura.” The
patient did not survive the intervention. The first surgeon
to report a successful laminectomy was Barth [21] in
1901. The 21-year-old patient had developed post-traumatic SEA following a knife wound around the
eighth/ninth thoracic vertebrae. The knife wound had
punctured the dura and led to an epidural abscess that
was drained operatively. However, there were postoperative neurological deficits.
However, it was in the 1920s that laminectomy was
first intensively discussed and utilized as a therapeutic
option for SEA. Taylor and Kennedy [368] and Dandy
[72] in the USA were the first to use this method successfully. Since then, laminectomy has been standard
treatment (see “Treatment”). Following the above mentioned articles, there were numerous communications on
the operative treatment aspects of SEA in the medical literature from the 1930s to the 1950s [8, 15, 26, 34, 47,
48, 51, 64, 68, 69, 83, 90, 120, 131, 136, 151, 197, 218,
246, 248, 284, 289, 295, 296, 301, 310, 349, 356, 396].
Initial enthusiasm about the possibility of successfully
treating this disease, which had so long been associated
with an extremely poor prognosis, is reflected in the titles of the articles published in this period:
– 1932: Metastatic epidural abscess of the spinal cord.
Recovery after operation [69].
– 1932: Epidural abscess complicated by staphylococcal meningitis. Report of a case with complete recovery following operation [310].
– 1934: Metastatic spinal epidural abscess. Report of a
case with recovery following operation [349].
– 1940: Acute metastatic spinal epidural abscess. Report of two cases with recovery following laminectomy [301].
– 1941: Emergency laminectomy for acute epidural abscess of the spinal canal. Report of four cases with recovery in three [90].
– 1944: Acute epidural abscess of the spinal canal.
Complete recovery following emergency laminectomy and penicillin [83].
However, case studies continued to be published in
which patients were described who died despite treatment with laminectomy [9, 26, 131, 136, 289, 313], especially in the early years of use of this method. Nonetheless, the prognosis of the disease had been significantly improved. Starting from the early 1930s, early surgical intervention has been emphasized as a prerequisite
for successful treatment of SEA [51, 349, 396].
Grant [129] analyzed 88 cases of SEA published before 1937 and calculated a mortality rate of 48%. In contrast, the same author described a mortality rate about
half as much, or 26%, for cases occurring after 1937. In
addition to laminectomy, the use of antibiotics and sulfonamides for the treatment of patients with SEA after
1941 is likely to be responsible for the drop in mortality.
Donathan [83] from Knoxville, USA described one case
of SEA in 1944 and Heusner [151] from Boston described 12 who received medical treatment in addition to
laminectomy. Sulfadiazine and penicillin were the antibiotics used in that period. According to Donathan [83],
179
penicillin is “of prime importance and should be started
immediately.”
Whereas Donathan [83] had described a 12-year-old
girl with SEA in 1935, Gasul and Jaffe [120] described
cases of SEA in children aged 3, 8, and 12 years. In all
cases, laminectomy was performed. The 3-year-old boy
was the only patient who died in the postoperative period. This publication is the first to deal specifically with
the problem of SEA in children, but the discussions
about SEA in pediatric age groups do not differ from
those on SEA in adults.
Usually, gram-positive cocci and especially Staphylococcus aureus represent the etiologic agent in cases of
SEA (see “Causative organisms”). Therefore, the first
description of SEA due to Actinomyces sp., irregularly
formed, non-spore-forming gram-positive rods, is especially interesting from the point of view of medical history. The case was published in 1940 by Krumdieck and
Stevenson [197] in New York. The 54-year-old patient
was first treated by radiation for suspected Hodgkin’s
disease but developed paraplegia and fever during the
course of his hospital stay. The diagnosis of actinomycotic SEA was made by autopsy.
The first documented case of iatrogenic SEA due to
lumbar puncture was described in 1945 by the American
authors Rangell and Glassman [295]. Viscous pus was
drained from between the third and fourth lumbar vertebrae following laminectomy in this 28-year-old patient.
The patient recovered following postoperative treatment
with sulfadiazine but had residual neurological deficits.
Epidemiology
Spinal epidural abscess occurs primarily in individuals
over 30 years of age. However, published case series
with more than ten patients report differing average ages
for affected patients. The lowest age was reported by
Dei-Anang et al. [80], who found a mean age of only 34
years in a group of 15 patients. Maslen et al. [231] reported a mean of 67 years of age in a group of 28 patients with SEA, which represents the highest average
age in all published case series [16, 31, 73, 74, 80, 81,
98, 141, 154, 172, 177, 189, 201, 212, 213, 231, 235,
253, 283, 297, 300, 306, 316, 370, 391, 401]. However,
Dei-Anang et al. [80] included in their patient group four
children aged 2 to 6 as well as five patients from 19 to
38. The range of ages reported by Dei-Anang et al. was
2 to 74 years, while the patient group investigated by
Maslen et al. [231] had an age range of 45 to 87 years.
Ravicovitch and Spallone [297] of the Botkin Hospital in
Moscow reported a significantly lower average age (30.1
years) for their patient group than Dei-Anang et al. None
of the patients reported by Ravicovitch and Spallone was
younger than 15 years.
In this review, articles published between 1954 and
1997 concerning a total of 915 patients with SEA were
statistically analyzed [1, 3–5, 11–14, 16, 17, 24, 31–33,
35–40, 43–46, 50, 55, 57–61, 63, 65, 66, 70, 71, 73, 74,
77, 79–81, 84–86, 89, 92–94, 96–100, 103–108, 110,
115–117, 121–124, 126–128, 133, 134, 137, 138,
141–143, 146–149, 152, 154, 156, 158, 160, 161, 166,
168, 170, 172, 176–178, 180–182, 184–189, 191, 193,
194, 198–202, 204, 206, 208, 211–217, 220, 222–227,
229, 231–233, 235, 237, 240–242, 244, 245, 250, 253,
254, 256, 259, 260, 264–266, 269–271, 273, 276–278,
280, 283, 293, 294, 297–300, 303–306, 308, 311,
315–318, 321, 324, 326–328, 330, 332, 335, 336, 339,
341–343, 345–348, 351, 352, 354, 357–359, 362, 365,
370, 372, 373, 379–385, 388, 390, 391, 395, 397–399,
401, 403]. Some publications did not include detailed information on patient age [16, 31, 55, 73, 74, 80, 141,
158, 172, 176, 206, 212, 213, 232, 253, 297, 300, 316,
401]. Additionally, some only indicate the average age
and the range of ages, without indicating the exact age of
individual patients [16, 31, 73, 74, 80, 141, 158, 172,
177, 212, 213, 232, 253, 297, 300, 316, 401]. Therefore,
it was possible to calculate the age distribution for 452
patients with SEA (Fig. 3). In contrast to the common
conception that SEA mainly affects individuals older
than 50 [195, 314], there was a broader distribution of
age groups among patients affected by SEA, especially
among males (Fig. 3). The age distribution does not support the view that there is a peak exclusively in the sixth
and seventh decades of life [230].
Of a total of 301 men with SEA, 204 (68%) were between 31 and 70 years old. An age predilection for men
between 51 and 70 was not observed. From a total of 151
women with SEA, 105 (70%) were between 31 and 70
years, again without an obvious predilection for any given decade. The youngest patient with SEA was only 10
days old [133] and the oldest was 87 [231, 235].
Ruiz et al. [314] deny the existence of a gender preference for SEA in their review article published in 1995,
whereas Youmans [402] wrote in a textbook contribution
in 1985 that men develop SEA more frequently. Krauss
and McCormick [195] do not pronounce a definite opinion on this issue. In the present work, essentially all publications on SEA appearing after 1954 were included,
and the evaluation of the correspondingly large number
of cases allows statistically sound analysis.
All large case series observed a preference for the
male gender in the development of SEA [31, 73, 74, 80,
81, 98, 141, 154, 172, 177, 189, 201, 212, 213, 231, 235,
253, 283, 297, 300, 316, 370, 391, 401]. Only one article
on a series of 39 patients with SEA [16] reports a
male:female ratio of 0.5:1. The 915 cases from the literature in the present work included 520 men and 289 women. For 106 patients, no gender information was available in the original publications [55, 158, 172, 297, 401].
Therefore, the male:female ratio is 1:0.56, which corresponds to a definite prevalence for males. The reasons
for this cannot be clearly derived from the literature, but
it may be assumed that different risk factors are partially
responsible (see “Pathogenesis and risk factors”). Alcohol abuse, use of intravenous drugs, and trauma affect
men more often than women and have been observed to
be associated with SEA.
180
Fig. 3 Age distribution of the
patients (n=452) [1, 3–5,
11–14, 17, 24, 32, 33, 35–40,
43–46, 50, 57–61, 63, 65, 66,
70, 71, 77, 79, 81, 84–86, 89,
92–94, 96–100, 103–108, 110,
115–117, 121–124, 126–128,
133, 134, 137, 138, 142, 143,
146–149, 152, 154, 156, 160,
161, 166, 168, 170, 177, 178,
180–182, 184–189, 191, 193,
194, 198–202, 204, 206, 208,
211, 214–217, 220, 222–227,
229, 231, 233, 235, 237,
240–242, 244, 245, 250, 254,
256, 259, 260, 264–266, 269,
273, 276–278, 280, 283, 293,
294, 298, 299, 303–306, 308,
311, 315, 317, 318, 321, 324,
326–328, 330, 332, 335, 336,
339, 342–343, 345–348, 351,
352, 354, 357–359, 362, 365,
370, 372, 374, 379–385, 388,
390, 391, 395, 397–399, 403]
Pathogenesis and risk factors
The pathogenetic mechanisms underlying spinal cord
dysfunction in SEA can only be conjectured from the
histopathological findings upon autopsy. Animal models
allow the consequences of spinal cord compression due
to this extradural, essentially pyogenic infection to be investigated more exactly. However, to date only the group
of the American neurosurgeon John Feldenzer [101, 102]
at the University of Michigan has presented results of
animal experiments. Staphylococcus aureus was injected
into the posterior thoracolumbar portion of the epidural
space of rabbits which were then examined using radiological and histopathological methods [101] as well as
microangiography [102]. The authors visualized the spinal blood vessels and investigated how they reacted to
external pressure due to the developing epidural abscess.
The anterior and posterior spinal vessels were not compressed, even in animals that developed paraplegia
[102]. Histopathological findings of thrombosis or vasculitis associated with damage to the vascular supply of
the spinal cord were not present [101]. These results
would seem to indicate that mechanical compression is
the most important pathogenic factor for the neurologic
symptoms seen in SEA; however, this opinion is not
shared by other authors [16, 48, 151, 316] who favor the
notion of vascular damage due to SEA with secondary
hypoxia being the main pathogenic factor for SEA.
However, these are older publications [48, 151], and histopathological findings of only three [16] or four [316]
deceased patients are presented. According to Hlavin et
al. [154], the combination of spinal cord compression
and vascular damage with resultant hypoxia represents
the pathogenic basis of SEA.
In cases not associated with spinal instrumentation, the
colonization of the spinal epidural space by micro-organisms may occur hematogenously or by contiguous spread
from neighboring organ structures [8, 82, 164, 195, 309,
378]. Infection due to hematogenous spread may originate
from all types of extraspinal infection that lead to a persistent or temporary bacteremia. All invasive procedures can
lead to the introduction of bacteria into the blood stream
with resultant SEA. In addition, many patients have risk
factors that may favor the development of SEA.
For the present review, risk factors and infection
sources could be identified for 854 patients in the literature (Table 1). In many cases, multiple risk factors and
potential infection sources were present in individual patients (e.g., alcohol abuse together with skin abscesses,
furuncles, and paronychia). Therefore, a total of 1005 individual risk factors and infection sources were identified in 854 patients described in the literature.
Diabetes mellitus was the most commonly observed
risk factor. This disorder affected 128 (15%) of the 854
181
Table 1 Risk factors and sources of infection (n=1095) in 854 patients with spinal epidural abscess
Comorbidity factor and/or source of infection
n
Comorbidity factor and/or source of infection
Diabetes mellitus
128
Disorders of different organ systems or body regions:
Degenerative spinal disorders
Chronic renal insufficiency
Colitis ulcerosa or Crohn’s disease
Systemic lupus erythematosus
Dermal sinus
Herpes zoster neuralgia
Decubitus ulcer
Pregnancy
Delivery
Malignancy
Intravenous drug use
Alcohol abuse
Infections:
Skin abscess, furuncle, paronychia
Vertebral osteomyelitis/discitis
Pulmonary/mediastinal infections
Sepsis
Urinary tract infection
Paraspinal abscess
Pharyngitis
Wound infection
Endocarditis
Sinusitis or upper respiratory tract infection
HIV infection
Dental abscess
Retropharyngeal abscess
Soft tissue infection (erysipelas)
Psoas abscess
Fournier’s disease
Hepatitis
Prostate abscess
Septic arthritis
Typhus
Vaginal infection
Trauma:
Extraspinal trauma
Spinal trauma
75
41
377
128
59
41
39
23
14
10
10
10
8
9
7
5
5
4
1
1
1
1
1
1
85
41
44
Invasive procedures:
Epidural anesthesia
Extraspinal operations
Spinal operations
Vascular access
Corticosteroid therapy
Paravertebral injections
Spinal anesthesia
Hemodialysis
Injections in the epidural space
Discography
Coronary angioplasty
Renal transplantation
Dental extraction
Blockade of the stellate ganglion
Chemonucleolysis
Intrauterine spiral
Liver transplantation
Lumbar puncture
n
83
49
18
6
3
3
2
2
5
5
18
188
42
42
25
17
16
12
9
7
5
2
2
2
2
1
1
1
1
1
References 1, 4, 5, 11–13, 16, 24, 31, 33, 35–40, 43–46, 50, 57, 61, 63, 66, 70, 71, 73, 74, 77, 79–81, 85, 86, 89, 92–94, 98, 100, 104,
106–108, 110, 117, 121, 122, 127, 128, 133, 134, 137, 138, 141–143, 146, 149, 154, 156, 160, 166, 168, 170, 172, 176, 177, 182,
185–189, 191, 194, 199–202, 204, 206, 208, 211–215, 217, 220, 222–226, 232, 233, 235, 237, 240, 241, 244, 245, 250, 253, 254, 256,
259, 260, 264–266, 269, 270, 273, 276, 278, 294, 297, 298, 300, 305–306, 311, 316–318, 321, 324, 326–328, 330, 335, 336, 341, 342,
345–348, 351, 352, 354, 357, 359, 362, 365, 370, 372, 383–385, 388, 390, 391, 395, 397–399, 401, 403.
patients with SEA. In the large case series in the literature, diabetes is repeatedly cited as an important risk factor for SEA [74, 154, 177, 213, 231, 262, 300, 306]. The
percentage of patients with diabetes in these series ranges from 18.0% [231] to 53.7% [177]. The high percentage of diabetics (53.7%) in the study of Khanna et al.
[177] is likely to be related to peculiarities of the patient
population under study. The study was performed at the
Henry Ford Hospital in Detroit, a city with a large population of Afro-Americans, who are more commonly affected by diabetes than Caucasians. Maslen et al. [231]
performed a meta-analysis on 254 patients with SEA and
calculated that 16% of all patients also had diabetes.
This result is comparable with that of the present work,
in which 15% of 854 patients were affected by diabetes.
The association of SEA and diabetes mellitus may be
explained by the reduced immunocompetence of diabetics [46]. Inadequately treated blood glucose concentrations are associated with reduced cellular immunity, i.e.,
with reduced chemotaxis, phagocytosis, and bactericidal
activity of neutrophilic granulocytes [46].
Intravenous drug abuse represents another important
risk factor for SEA. Seventy-five (9%) of the 854 pa-
tients with identifiable comorbidity factors were intravenous drug users (Table 1). The case series in the literature report percentages between 17.5% [154] and 31.7%
[177]. However, these studies were performed exclusively in the USA [154, 177, 213, 231, 306]. Khanna et al.
[177] recently reported a percentage of 31.7% for intravenous drug abuse. Again, intravenous drug use is more
extensive in the mainly Afro-American population in
and around Detroit, where the study was performed, than
in the general American population. In the meta-analysis
by Maslen et al. [231] mentioned above, 7% of all SEA
patients were intravenous drug users, which is comparable with the percentage of 9% of 854 patients reported in
the present work. The reasons for the increased risk of
this patient group for the development of SEA include
not only bacterial contamination of the equipment [397]
but also the well-known dysfunction of the cellular and
humoral immune system among chronic heroin abusers
[49].
In addition to drug abuse, alcohol abuse represents
another, albeit less important risk factor for the development of SEA. Published case series report rates of chronic alcohol abuse between 10% [154] and 33% [74]. In
182
the present work, 41 (5%) of 854 patients described in
the international medical literature were alcoholics.
Among the 254 patients from different publications analyzed by Maslen et al. [231], 4% were alcoholics.
In many cases, alcoholics consume diets deficient in
protein, which is thought to compromise the immune
system [46]. It remains speculative whether other factors
such as inadequate personal hygiene or other sociological characteristics among alcoholics may also be associated with the observed increased coincidence of SEA
and chronic alcohol abuse.
On the other hand, there is a definite connection between sources of infection near to and distant from the
vertebral canal. Skin abscesses, furuncles, and paronychia are often found to be SEA’s source of infection. Hematogenous spread into the epidural space is the important pathogenic factor. Patients with SEA who report a
previous infectious condition of the skin made up a higher proportion of all patients in older case series than in
those from the 1990s. In 1954, Hulme and Dott [158] described cutaneous and subcutaneous infections in 33% of
their patients and Yang [401] found 44% in his case series from 1982. However, the latter study was performed
in Tientsin, China. In more recent American studies, the
incidence of cutaneous infections was calculated to
be 7% [231], 12.2% [177], 15.4% [154], and 20% [300];
the present meta-analysis found an incidence of cutaneous infections of 15% (128 of 854 patients) (Table 1).
Maslen et al. [231] observed a rate of 25% for dermal
and soft-tissue infections among the 254 literature cases
they analyzed. Sources of infection near the vertebral canal include paraspinal abscess (14 of 854 cases), retropharyngeal abscess (five of 854), and psoas abscess (four
of 854) (Table 1). In these cases, the colonization of the
spinal epidural space occurred by contiguous spread [46,
61, 71, 133, 168, 188, 294, 321, 372].
Taken together, infectious processes were identified
in 377 (44%) of the 854 cases identified in the literature
and thus represent the most common comorbidity factor.
In most cases, the infection led to seeding of the epidural
space by hematogenous spread.
Trauma represents another important group of risk
factors (Table 1). In the patient group presented in this
work, 85 of 854 patients (10%) had suffered extraspinal
or spinal trauma preceding development of SEA. In the
larger case series of the literature, trauma was reported in
25% [158] to 34.7% [306] of patients with SEA. Trauma
can favor hematogenous spread of micro-organisms by
penetration of anatomic barriers, but spinal trauma is especially important, since it may create a site of entry for
micro-organisms into the epidural space [16, 151, 158,
172]. Spinal hematomas associated with severe blunt
trauma to the back represent an important pathogenetic
factor that can favor the development of SEA and could
also explain the potential development of SEA following
epidural blood patch [221] (see below).
In the present meta-analysis, 49 (6%) of 854 patients
with SEA were identified who also had degenerative vertebral disease as a comorbidity factor (Table 1). Degen-
erative changes of the intervertebral disks could act as
sites of reduced resistance against hematogenous spread,
similarly to hematomas due to spinal trauma, and thereby favor the development of SEA.
Patients with chronic renal insufficiency and especially those who require dialysis often show reduced immunocompetence. Eighteen (2%) of the 854 patients from
the literature discussed in the present work had chronic
renal insufficiency (Table 1) [154, 187, 212, 213, 264,
306, 343, 351].
Of 854 patients with SEA, six had Crohn’s disease or
colitis ulcerosa (Table 1) [4, 110, 146, 254, 300, 383].
Both immunosuppressive treatment and the tendency to
formation of fistulas represent factors that could favor
the development of SEA. One patient with Crohn’s disease developed SEA due to a fistula that led into the
lumbosacral canal [110], and another 42-year-old with
Crohn’s disease developed SEA due to an enteroepidural
fistula following proctocolectomy and creation of an
ilioanal anastomosis [254].
Cancer patients may also display reduced immunocompetence. Eighteen (2%) of the 854 patients with SEA
also had cancer [12, 37, 73, 80, 147, 154, 177, 201, 211,
300, 304, 326, 395]. Some of these patients were also receiving chemotherapy [395]. Because of the simultaneous presence of several different comorbidity factors
in older patients, it is not always possible to determine
the exact contribution of cancer to the development of
SEA in these cases.
Pregnancy and delivery represent other risk factors.
Ten of 854 patients in the literature developed SEA during pregnancy or the postpartal period [16, 70, 122, 160,
176, 182, 185, 224]. Postpartal impairment of immune
defense may be related to the development of SEA in
these patients [224].
Invasive procedures led to SEA in 188 (22%) of the
854 patients in the literature (Table 1). Spinal and extraspinal operations, which may create a contiguous port of
entry for micro-organisms into the epidural space or lead
to hematogenous seeding, and especially anesthesiological procedures may favor the development of SEA. Although spinal anesthesia [76, 174] and lumbar puncture
in children with bacteremia [369] can lead to the development of bacterial meningitis, most authors agree that the
risk of infection following nerve block anesthesia near
the spinal cord should not be overestimated [78, 119, 157,
257, 290, 302, 337, 360, 376, 377]. Spinal epidural abscess is generally thought to be a “very rare, severe complication” [302] of central nerve blocks. The fact that 42
(5%) of the 854 cases of SEA were associated with epidural anesthesia must be seen in relation to the quite common use of this form of regional anesthesia. Among
505,000 epidural anesthetics performed between 1982
and 1986 for obstetric indications, only a single case of
SEA was observed [337]. Perioperative bacteremia is regarded as a relative but not absolute contraindication for
regional epidural or spinal anesthesia [25].
Michel et al. [241] identified 40 patients with SEA following epidural anesthesia or analgesia in articles pub-
183
lished between 1947 and 1996; they also quoted the work
of Du Pen et al. [87], who had described 15 other cases of
infections in the epidural space following epidural anesthetic procedures. Kindler et al. [179] published an evaluation of the literature from 1974 to 1996 with 42 patients
with SEA following epidural anesthesia or analgesia.
There was a remarkably high percentage of 33% following thoracic catheterization compared to 45% following
lumbar catheterization, even though thoracic catheters are
used much less often than lumbar catheters. The authors
speculate that this is related to the generally more difficult insertion of thoracic catheters and their longer duration of use. The information in the literature on the incidence of SEA following anesthesiological procedures is
not reliable because of the rarity of this complication. The
incidence of SEA due to short-duration catheterizations is
about 1:100,000 [25], whereas that of SEA with tunneled
peridural catheters is about 4% [87].
Du Pen et al. [87] report on 350 patients with a total
of 32,354 catheter days and calculate a risk of 1 per 1702
catheter days. In all, there is a low incidence of infectious complications following anesthesiological procedures in the epidural and spinal spaces [25]. In this metaanalysis of 854 patients, there were 42 cases of SEA following catheter epidural anesthesia, nine following spinal anesthesia, and five following “single shot” injections into the epidural space (Table 1) [24, 36, 39, 45,
57, 70, 79, 86, 104, 106, 117, 122, 128, 142, 152, 186,
213, 215, 225, 233, 241, 256, 259, 260, 265, 278, 304,
318, 335, 342, 351, 352, 362, 365, 382, 385].
Williams et al. [394] report on a case of epidural hematoma following repeated steroid injections that had to
be surgically decompressed. Hemorrhages into the epidural space following epidural anesthesia were demonstrated postmortem [400] and could represent a site of
reduced resistance to colonization by micro-organisms,
thus favoring the development of SEA. Occasionally the
formation of fistulas has also been observed after catheterization of the epidural space [334, 386].
Vascular access may also represent a port of entry for
infectious agents. Seventeen (2%) of the 854 patients
with SEA in the literature contracted the infection in this
way (Table 1) [12, 74, 154, 187, 201, 214, 264, 343].
The other invasive procedures performed on patients
who then developed SEA are listed in Table 1.
Of the 738 patients in this meta-analysis for whom
the localization of SEA could be identified from descriptions in the literature, 19% (140) had cervical SEA and
7% (53) had cervicothoracic SEA (Table 2). Although
Lasker [204] indicates a lower frequency of 12% for cervical SEA, Friedman reported a frequency of 32% only 7
years later. In the majority of the cases, cervical epidural
abscesses originated as a consequence of vertebral body
osteomyelitis or discitis affecting vertebrae C4–C7 in an
anterior position [314].
The most common location for SEA is the thoracic
epidural space. Of the 738 patients, 35% (255) had thoracic epidural abscesses and 30% (223) had lumbar or
lumbosacral abscesses (Table 2). This localization corre-
Table 2 Craniocaudal localization of spinal epidural abscess in
738 patients
Localization
Number (%*)
Cervical
Cervicothoracic
Thoracic
Thoracolumbar
Lumbar
Lumbosacral
Sacral
Cervicothoracolumbar
Thoracolumbosacral
Total
140 (19)
53 (7)
255 (35)
54 (7)
133 (18)
90 (12)
3 (–)
8 (1)
2 (–)
738 (100)
*Rounded off.
References 1, 3–5, 11–14, 16, 17, 31–33, 35–40, 43, 44, 50, 55,
57–61, 63, 65, 66, 70, 74, 77, 79, 81, 84, 85, 89, 92–94, 96,
98–100, 103–108, 110, 115–117, 121, 122, 124, 127, 128, 133,
134, 137, 138, 141–143, 146–149, 152, 154, 156, 160, 161, 166,
168, 170, 172, 177–178, 180–182, 185–188, 191, 193, 194,
198–201, 204, 208, 211, 212, 214–217, 222–227, 229, 232, 233,
235, 237, 240, 241, 244, 245, 250, 254, 256, 259, 264–266,
269–271, 273, 276, 277, 280, 283, 293, 297, 299, 300, 303, 305,
306, 308, 311, 315, 316, 318, 321, 324, 326–328, 332, 336, 339,
341–343, 345–348, 351, 352, 354, 357–359, 362, 365, 372, 374,
379, 381–384, 390, 391, 395, 397–399, 401, 403.
sponds to the results of most large case series in the literature [16, 138, 177, 235, 253]. The preferential thoracic
or lumbar localization of SEA is likely related to the
greater extension of the epidural space in the thoracolumbar segment of the vertebral column and to the welldeveloped extradural venous plexus in this region [366].
Clinical and laboratory findings
In 1948, Corradini et al. [68] concisely defined the
meaning of the symptoms of SEA: “Furthermore, we believe that the syndrome is so characteristic that its recognition depends only upon acquaintance with the symptomatology.” Even today, this statement is still correct.
Also in 1948, Heusner [151] published his classic description of the various stages of SEA that continues to
be referred to in the literature up to the present. Following the initial stage of severe back pain associated with
fever and local tenderness in the area of the spinal
column, the second stage is dominated by signs of spinal
irritation. These manifestations include Lasègue’s,
Kernig’s, and Lhermitte’s signs, Brudzinski’s reflex, and
neck stiffness [241, 309]. Additionally, back pain can radiate into the arms or legs, depending on the craniocaudal location of the abscess. In Heusner’s third stage
[151], initial neurologic deficits are observed such as
weakness of the voluntary musculature or fecal or urinary incontinence. Sensory deficits may also occur. In
the fourth stage, muscle weakness progresses to paralysis. The duration of the individual stages varies. Importantly, the transition to the terminal stage with paralysis
can occur very quickly [158]. In addition to neurological
deficits [151], patients with SEA may also develop fever
184
Table 3 Symptoms and signs of spinal epidural abscess in 871 patients
Clinical findings
n patients (%)
Fever
571 (66)
Initial pain symptom:
Back pain
Neck pain
Headache
Local tenderness
Irritability
Spinal irritation
620 (71)
29 (3)
25 (3)
151 (17)
8 (1)
174 (20)
Beginning neurologic deficit:
Muscle weakness
Incontinence
Sensory deficit
225 (26)
210 (24)
115 (13)
Advanced neurologic deficit:
Paraparesis/paraplegia
Tetraparesis/tetraplegia
268 (31)
29 (3)
References 1, 3–5, 11–14, 16, 17, 24, 31–33, 35–40, 43–46, 50,
55, 57–61, 63, 65, 66, 71, 74, 77, 79–81, 84, 85, 89, 92–94, 96, 97,
99, 100, 103–108, 110, 115–117, 121–124, 126–128, 133, 134,
137, 138, 141–143, 146–149, 152, 154, 156, 160, 161, 166, 168,
170, 172, 176–178, 180–182, 185, 187–189, 191, 193, 194,
198–202, 204, 206, 208, 211, 213–217, 220, 222–227, 229,
231–233, 235, 237, 240–242, 244, 245, 250, 253, 254, 256, 259,
260, 264–266, 269–271, 273, 276, 277, 280, 293, 294, 297–300,
303–306, 308, 311, 315–318, 321, 324, 326–328, 330, 332,
335–336, 339, 341–343, 345–348, 351, 352, 354, 357–359, 362,
365, 370, 372, 374, 379–385, 388, 390, 391, 397–399, 401, 403.
in the initial stage due to the inflammation associated
with the abscess [82, 164, 230, 309, 378]. This is accompanied by an accelerated erythrocyte sedimentation rate
(ESR) and leukocytosis.
The combination of severe pain near the spine with
fever is characteristic [68] and should always be regarded as a warning sign of possible SEA. The early recognition of these symptoms before the occurrence of neurologic deficits is exactly the “problem of spinal epidural
abscesses” described by Strohecker and Grobovchek
[361].
For 871 of the 915 patients with SEA identified in the
literature, information about the symptoms of the disease
were available from case reports or case series (Table 3).
Back pain was the most common symptom, occurring
in 71% of the patients with SEA. Fever was present in
66%. Accordingly, two thirds of the SEA patients could
be diagnosed in an early stage if this disorder is considered in the differential diagnosis, and neurologic deficits
could be avoided with timely therapeutic intervention.
Most of the large case series in the literature with at
least 18 patients describe a somewhat higher incidence of
back pain with SEA. All 18 patients of Yang [401], all 29
of Del Curling et al. [81], and all 39 of Baker et al. [16]
displayed spine-related back pain due to SEA. Eighty
percent of the SEA patients described by Nussbaum et al.
[262], 89% of those of Khanna et al. [177], 89% of those
of Liem et al. [213], and 92% of those of Maslen et
al. [231] also reported back pain. Darouiche et al. [74]
found 72% of their patient group to have back pain, and
Rigamonti et al. [306] found 74%; these are the only two
large case studies with rates of back pain comparable
to that calculated from the present meta-analysis. The
case studies describe groups of patients treated by one
physician or group of physicians, while the present metaanalysis summarizes published data from numerous authors, so it is possible that back pain was not noted in
some articles. Even if one author performs a retrospective
chart analysis, clinical data are not always complete. For
instance, Hancock [141] reports that the clinical data on
the symptoms experienced by 49 SEA patients in his
analysis were sufficiently documented in only 24 cases.
Fever developed in 66% of the patients with SEA in
the present meta-analysis (Table 3). Even though increased body temperature is an essential symptom of this
severe inflammation of the epidural space, published
case series document a quite variable incidence of fever
in these patients. During the course of disease, 29%
of the patients of Liem et al. [213], 39.1% of those of
Rigamonti et al. [306], 45% of those of Nussbaum et al.
[262], 52% of those of Maslen et al. [231], 61% of those
of Khanna et al. [177], and 62% of those of Del Curling
et al. [81] as well as all patients described by Baker et al.
[16] developed increased body temperature. This broad
range may be related to differing definitions of fever in
the different studies. In the present meta-analysis, the adjective “feverish” in the case reports was regarded as
sufficient to include a patient in the group with fever.
In the initial stage of back pain, 17% of the 871 patients from the international medical literature had local
tenderness of the spine and its surroundings (Table 3).
This finding is not always consistently documented in
the literature. Most case series do not analyze the incidence of local tenderness of the spine using statistical
methods. The only exceptions to this are the articles of
Baker et al. [16] and Maslen et al. [231]. Whereas the
former document the occurrence of local tenderness in
all examined patients with SEA, Maslen et al. [231]
found this manifestation to be present in only 31% of
their patients and 24% of the 207 patients they analyzed
from the international literature.
Irritability was seen in eight of 871 patients (Table 3)
and is a typical symptom of SEA in small children [3,
32, 96, 125, 228, 271, 315, 380]. Aicardi and Lepintre
[3] describe a 3-month-old infant with SEA and conclude that conspicuous crying and obvious pain on being
touched may be the only signs of SEA in this age group.
Spinal irritability and signs of initial or progressive
neurological deficits occurred in only a fifth to a third of
the 871 patients reported in the international literature
from 1954 to 1997 (Table 3). Perhaps many patients
were diagnosed in early stages. Comparable indications
on the incidence of signs of spinal irritation are generally
not found in the large case series. Radiculopathic complaints were present in 12.2% of the patients described
by Khanna et al. [177], as the presenting symptom in
19% of those described by Maslen et al. [231], and in a
total of 47% of patients described by Darouiche et al.
185
Fig. 4 Leukocytes/µl in 218
patients with SEA [3, 5, 11, 13,
14, 17, 31, 32, 40, 44, 46, 50,
55, 58, 60, 61, 63, 77, 79, 84,
93, 94, 96, 97, 100, 103, 105,
108, 110, 117, 122, 123,
126–128, 133, 134, 143,
147–149, 152, 160, 161, 166,
168, 172, 176, 178, 180, 181,
185–189, 191, 193–194, 198,
202, 204, 208, 211, 214, 215,
225, 227, 229, 232, 233, 240,
242, 244, 245, 250, 256, 259,
264, 266, 269, 271, 273, 276,
293, 299, 303, 304, 306, 308,
315, 317, 321, 324, 326, 328,
332, 335, 336, 339, 341, 342,
348, 352, 354, 357, 362, 372,
374, 379–381, 383–385, 390,
399, 403]
[74]. In this meta-analysis of 871 patients, 20% showed
signs of spinal irritation (Table 3) including radicular
complaints.
Weakness of the voluntary musculature as well as urinary or fecal incontinence were seen in 26% and 24% of
the cases included in this meta-analysis, respectively
(Table 3). There were sensory deficits in 13%.
In case series of the literature, 27% to 35% of the patients had weakness of the voluntary musculature [74,
231]. Sphincter dysfunction was present in 30.4% to
38% [74, 213, 262, 306]. The incidence of sensory deficits is reported to be between 23% and 43.4% [74, 213,
231, 262, 306].
The results of this meta-analysis concerning the percentual prevalence of signs of the second and third stages of SEA are somewhat lower than those of individual
case series. This could be due to poorer documentation
of these manifestations on the part of some or many of
the case reports that were included in this meta-analysis,
since individual case reports often emphasize one specific aspect of SEA such as an unusual concurrent disease,
a rare infectious agent, or special radiological tech-
niques. Systematic descriptions of signs and symptoms
similar to those of large case series is generally not a feature of individual case reports. However, these case reports were included in the evaluation for the present
work.
In 268 (31%) of the 871 patients from the literature,
paraparesis or paraplegia developed, and 29 (3%) developed tetraparesis or tetraplegia (Table 3). Published case
series seldom differentiate between paraparesis/paraplegia and tetraparesis/tetraplegia or even between paresis
and paralysis. Instead, the findings are generally subsumed under the heading of paralysis. Of the patients of
Darouiche et al., 21% [74] developed paralysis, and 32%
of the patients of Maslen et al. [231] had it upon admission to the hospital. Twenty-nine (3%) of the 871 patients from the literature developed tetraparesis or tetraplegia. Most of these cases were due to cervical or cervicothoracic SEA [38, 44, 59, 99, 100, 116, 117, 121, 143,
199, 223, 311, 326, 345, 359, 382, 395].
The manifestations of the different stages of SEA are
accompanied by various laboratory findings that reflect
the severe inflammation, including leukocytosis and an
186
Fig. 5 Erythrocyte sedimentation rate (ESR) in 117 patients
with SEA [5, 11, 14, 17, 31, 40,
44, 46, 50, 60, 63, 84, 93, 99,
103, 105, 108, 110, 117, 121,
123, 127, 128, 142, 149, 160,
161, 180, 184, 186, 187, 194,
198, 208, 211, 214–216, 220,
229, 232, 237, 240, 253, 259,
265, 293, 299, 306, 308, 315,
317, 321, 324, 326, 332, 336,
342, 347, 374, 380, 383, 384,
399, 403]
accelerated erythrocyte sedimentation rate (ESR). For
the 218 patients from the international literature whose
leukocyte counts were described in case reports, the average was 15,700 leukocytes/µl and the count ranged
from 1,500 [189] to 42,000 [14] leukocytes/µl (Fig. 4).
The patient with the lowest leukocyte count (1,500]
was a 42-year-old alcohol- and drug-dependent man with
a positive tuberculin test [189]. Whether this patient was
HIV-positive cannot be determined from the article,
which describes a patient group in New York treated
from 1984 to 1987. The patient with the highest leukocyte count (42,000) was a 10-year-old African child from
Brazzaville [14]. Hlavin et al. [54] report an average leukocyte count of 13,400/µl, with a range of 4,000 to
33,000/µl. The leukocyte count of the seven patients described by Mattle et al. [232] was between 10,000/µl and
24,600/µl with an average of 14,100/µl. In the present
work, 171 (78%) of the 218 patients from the literature
had a leukocyte count above 10,000/µl. For 68% of the
patients of Maslen et al. [231] and 76% of the patients of
Del Curling et al. [81], a leukocyte count was shown
above 10,000/µl. In the 88 patients identified in the literature by Maslen et al. [231], there was leukocytosis (i.e.,
above 10,000/µl) in 77%.
The ESR was indicated for only 117 of the 915 patients with SEA identified in the literature (Fig. 5). The
average ESR was 77 mm in the first hour, with a range
of 2 to 150 mm. One hundred ten (94%) of the patients
showed an ESR above 20 mm (Fig. 5). A high ESR was
consistently found in most patients described in the literature. All 40 patients examined by Hlavin et al. [154]
showed an ESR above 30 mm in the first hour, and all 28
patients described by Maslen et al. [231] had more than
25 mm.
In summary, the evaluation of 915 patients with SEA
described in the literature allows the conclusion that the
combination of back pain and abnormal inflammation
parameters (fever, leukocytosis, high ESR) is characteristic and should always make the clinician include SEA
in the differential diagnosis.
However, many patients are misdiagnosed initially.
Table 4 summarizes the diagnoses for which patients
187
Table 4 Initial diagnosis (n=159) in patients with spinal epidural
abscess
Diagnosis
n
Cerebral nervous system disorders:
Meningitis
Cerebral ischemia
25
2
Vertebral/spinal disorders:
Intervertebral disk prolapse
Vertebral osteomyelitis
Lumbar ischialgia
Spinal tumor or metastases
Poliomyelitis
Paraparesis
Guillain-Barré syndrome
Idiopathic back pain
Discitis
Polyneuritis
Transverse myelitis
Arachnoiditis
Syndrome of the anterior spinal artery
Brown-Séquard’s syndrome
Herpes zoster
Infectious vertebral spondylitis
Partial cauda equina syndrome
Intraspinal space-occupying lesion
Landry’s paralysis
Spinal hematoma
Spinal tuberculosis
Idiopathic spinal lesion
Vertebral compression fracture
Cervical myeloradiculopathy
Cervical spondylosis
30
9
9
8
4
4
3
3
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Infectious diseases except spinal, vertebral, and cerebral infections:
Urinary tract infection
9
Sepsis
8
Endocarditis
3
Paravertebral abscess
2
Dentogenic abscess
1
Pelvic inflammatory disease
1
Pneumonia
1
Prostatitis
1
Retropharyngeal abscess
1
Septic arthritis
1
Virus infection syndrome
1
Other disorders:
Hysteria
Urolithiasis
Rheumatoid arthritis
Appendicitis
Burkitt’s lymphoma
Diabetic coma
Mastoiditis
Myocardial infarction
Sacroileitis
4
2
2
1
1
1
1
1
1
References 4, 13, 16, 33, 35, 36, 38, 50, 57, 58, 66, 70, 74, 80, 92,
96, 100, 110, 116, 117, 137, 143, 148, 154, 156, 170, 176, 178,
180, 184, 186, 187, 193, 204, 206, 215, 227, 232, 237, 244, 245,
250, 253, 256, 264, 265, 271, 293, 318, 326, 336, 342, 346, 348,
362, 374, 379–381, 384, 388, 403.
with SEA in the literature were initially treated. This information was available for only 159 of 915 patients.
The most common mistaken diagnosis was meningitis,
followed by intervertebral disk prolapse.
Causative organisms
Spinal epidural abscess is primarily a bacterial infection.
The few reported cases of fungal infections were seen
mainly in immunocompromised patients, and individual
cases of SEA due to parasites have been reported from
certain geographic regions.
At the beginning of the twentieth century, Staphylococcus was described as the principle etiologic agent of
SEA [159, 247, 268], and this has remained true up to
the present day. Staphylococcus aureus was present in
551 (73%) of 753 patients from the international literature for whom the causative agent of SEA was identified
(Table 5). Skin abscesses and furuncles represent common risk factors for SEA (see “Pathogenesis and risk
factors”), and are mainly due to Staphylococcus. Staphylococcus was isolated in 80% of the 42 SEA patients of
Ravicovitch and Spallone [297], and infections of the
skin were again the primary source of infection. In other
large case series, the incidence of staphylococcal infection was 59.2% [231], 61% [177], 62% [74, 154, 213],
64% [300], 67.5% [262], 69.6% [306], 71% [235], and
93% [316]. The Canadian group of Russell et al. [316]
published a case series in 1979 describing 30 patients
with SEA, 93% of whom had staphylococcal infection.
The study included patients treated in Toronto starting
in 1946. All other case series reporting an incidence of
Staphylococcus as the causative agent from 59.2% to
69.6% were published in the mid-1990s and describe patients treated in 1979 or later [262, 306]. With the exception of an Australian [235] and a Canadian [300] study,
all studies were performed in the USA. The smaller proportion of Staphylococcus aureus among patients with
SEA in the more recent publications may reflect a partial
change of etiologic agents [74]. However, the ubiquity of
Staphylococcus aureus continues to make it the most important etiologic agent of SEA.
Gram-negative bacteria may be more common in
cases of SEA in drug users, according to the experience
of Kaufman et al. [172]. The use of intravenous drugs
has become much more common in industrial nations in
the last several decades, especially in the USA. Since intravenous drug use represents a comorbidity factor for
the development of SEA, this would result in an increased incidence of gram-negative bacteria in this patient group with a corresponding reduction in the incidence of Staphylococcus aureus. Koppel et al. [189] of
the Metropolitan Hospital in New York report on a group
of 18 drug-dependent patients with SEA. In contrast to
the results of Kaufman et al. [172], an increased incidence of gram-negative bacteria was not observed in
these patients. Hlavin et al. [154] reach a similar conclusion. The lower proportion of Staphylococcus aureus in
the above mentioned publications from the 1990s on
SEA patients from the USA could reflect a worldwide
shift in bacteria in this age of antibiotics and isn’t necessarily due to a predilection of certain subgroups of patients with SEA for other infectious agents.
188
Table 5 Infectious agents in
830 patients with SEA
Taxonomic classification of
bacteria was done according to
Kayser et al. [173].
References 1, 3–5, 11–14, 16,
17, 24, 31, 32, 35, 37–40, 43,
44, 50, 55, 58, 59, 61, 63, 66,
70, 74, 77, 79–81, 84, 85, 89,
93, 94, 96–100, 103–105, 107,
108, 115–117, 121–124,
126–129, 133, 134, 138,
141–143, 146–149, 152, 154,
156, 161, 166, 168, 170, 172,
176–178, 180–182, 185–189,
191, 193, 194, 202, 204, 208,
211–217, 222–227, 229,
231–233, 235, 240, 242, 244,
245, 253, 254, 256, 259, 260,
264–266, 269–271, 276–278,
280, 293, 294, 298–300,
303–306, 308, 311, 315–318,
321, 324, 326–328, 330, 332,
335, 336, 339, 341, 345–347,
351, 352, 354, 358, 359, 362,
365, 374, 379–384, 390, 391,
397–399, 401.
Infectious agent
Number
Bacteria
753
Gram-positive cocci. Micrococcaceae:
Staphylococcus aureus
Coagulase-neg. staphylococci:
Staphylococcus epidermidis
Staphylococcus mitis
Staphylococcus sp.
551
35
9
1
25
Gram-positive bacteria, catalase-negative. Streptococcaceae:
Streptococcus pneumoniae
Streptococcus viridans
Streptococcus pyogenes
Streptococcus agalactiae
Streptococcus milleri
Streptococcus sanguis
Streptococcus bovis
Streptococcus constellatus
Streptococcus sp.
Peptostreptococcus sp.
Enterococcus sp.
12
11
7
4
4
2
1
1
14
1
1
Facultatively anaerobic, gram-negative rods. Enterobacteriaceae:
Escherichia coli
Proteus mirabilis
Proteus sp.
Enterobacter cloacae
Enterobacter sp.
Enterobacter typhi
Salmonella typhimurium
Salmonella sp.
Klebsiella sp.
Citrobacter diversus
Serratia liquefaciens
21
3
1
2
2
1
1
2
2
1
1
Facultatively anaerobic, gram-negative rods. Pasteurellaceae: haemophilus paraphrophilus
1
Anaerobic, gram-negative, straight, curved, and helical rods. bacteriodaceae:
Bacteroides melanogenicus,
Bacteroides bovis
Bacteroides necrophorum
Bacteroides urealyticus
2
1
1
1
Aerobic, gram-negative, flagellated rods. Pseudomonaceae:
pseudomonas aeruginosa, pseudomonas spec.
14, 1
Gram-negative cocci and rods. Neisseriaceae: neisseria sp., acinetobacter
1, 1
Genera/species without family classification: brucella melitensis, brucella spec.
1, 1
Spore-forming gram-positive rods: Bacillaceae. Clostridium perfringens
1
Irregularly formed, non-spore-forming rods: actinomyces bovis, actinomyces sp.
1, 2
Regularly formed, non-spore-forming, gram-positive rods: propionibacterium sp.
2
Mycobacteria: Mycobacterium tuberculosis
9
Pleomorphic, gram-positive rods: Nocardia asteroides
2
Mixed bacterial infections
27
Fungi:
Aspergillus fumigatus
Aspergillus sp.
Sporotrichium schenkii
Torulopsis glabrata
13
9
1
1
2
Parasites:
Dracunculus midinensis, echinococcus granulosus
No growth or agent not specified
3
2, 1
61
189
Among gram-negative bacteria, 21 (3%) of the 753
patients from the literature with bacterial SEA had infections due to Escherichia coli and 14 (2%) had infections
due to Pseudomonas aeruginosa (Table 5). Other gramnegative bacteria were identified in individual cases.
Spinal epidural abscess due to Escherichia coli and
Pseudomonas aeruginosa was described in case series
that did not undertake differentiated analysis of infection
sources or risk factors [16, 74, 81, 98, 154, 172, 177,
189, 213, 235, 300, 391]. There are only five published
case reports that allow such an analysis; SEA due to
Escherichia coli was identified in two diabetic patients,
one of whom was also alcoholic and underwent several
endoscopic investigations because of pancreatitis [148,
327] and the other three case reports described Pseudomonas aeruginosa as the infectious agent [24, 278, 351].
These three patients all underwent spinal procedures. In
one case, an epidural catheter was left in place for 5 days
[278] and in another for 6 weeks [351]. The third patient
received five single shots for a nonoperative analgesic
indication [24].
Among the 915 patients with SEA analyzed in the
present meta-analysis, 26 were identified who had received epidural anesthesia by means of an indwelling
catheter and in whom the etiologic agent was determined
[35, 36, 39, 70, 79, 104, 152, 233, 241, 259, 260, 265,
278, 304, 318, 335, 342, 351, 352, 362, 365]. There
were 18 patients with Staphylococcus aureus, four with
Staphylococcus epidermidis, two with Pseudomonas aeruginosa, and one each with coagulase-negative Staphylococci and Pyocyaneus species. Among the 915 patients, only nine received spinal or peridural anesthesia
by single-shot injections [24, 45, 57, 94, 128, 186, 256,
317, 382]. In one case, trigger point infiltration had been
performed in the paraspinal cervical region at five different trigger points [94]. In eight patients, Staphylococcus
aureus was identified and in one a Pseudomonas species.
Therefore, Staphylococci were the most common etiologic agents in the 31 patients with epidural anesthesia or
analgesia for whom an infectious agent was identified,
being found in 26 of them (81%). Holt et al. [155] found
mainly coagulase-negative Staphylococci in the 78 patients they investigated by cultures of epidural catheter
tips, and in only 35% did they find Staphylococcus
aureus. They report on two patients with SEA; in one
Staphylococcus aureus was identified and in the other
Pseudomonas aeruginosa.
Infection due to epidural catheters can occur via the
catheter lumen or the insertion canal [119]. However,
bacterial colonization of the catheter tip does not necessarily result in SEA, especially since catheter tips are
colonized by bacteria in about 25% of cases, according
to Ungemach et al. [375]. In contrast, Bauer et al. [23]
identified bacterial contamination in only 5.7% of epidural catheters, in one case with an aerobic organism
(Staphylococcus epidermidis) and in three others with
anaerobic bacteria (Propionibacterium species). The authors investigated women who received epidural anesthesia during labor and do not specify the average time
the catheters remained in place. The average time in the
study of Ungemach et al. [375] was 32 hours; this study
classifies results according to specialty (orthopedics, surgery, urology), number of subsequent injections, and site
of insertion. If the catheter was inserted at L2/L3 or
above, the rate of contamination was only half that if the
catheter was inserted at L3/L4 or below. However, bacterial colonization of the catheter tip does not correlate
with bacterial colonization of the epidural space [18].
According to Sato et al. [325], bacterial colonization
of the skin represents a potential source of infection.
They found that a 10% polyvidone iodine solution is inferior to 0.5% chlorhexidine solution with respect to its
bactericidal effect. However, even after application of
these disinfectants, bacteria remain on the skin. In the
study of Abouleish et al. [2], the application of polyvidone-iodine solution for 1 minute was sufficient to prevent infection of the epidural space. Zenz et al. [405]
used suture fixation to secure and maintain indwelling
catheters for morphine analgesia of carcinoma patients;
bathing these patients in a solution containing polyvidone-iodine may help greatly as infection prophylaxis.
According to a Japanese study, positive bacterial cultures
from the skin of the back are more common in summer
than in winter [320], which may be due to increased
sweating in summer. Therefore, the authors recommend
more frequent disinfection of the skin in summer before
the insertion of epidural catheters. Continuous infusions
and avoidance of frequent manipulations of the epidural
catheter in order to give intermittent bolus injections,
and strict aseptic technique for catheter insertion can
also reduce the risk of infection [362]. It is also interesting that local anesthetics, morphine, and related opioids
may possess a certain antimicrobial effect that varies according to types of local anesthetic and infectious agent,
temperature, and pH of the solution [79].
Two comprehensive and recent studies that investigate
the risk of infection for children following epidural anesthesia conclude that the risk is very small for short-term
(2- or 3-day) catheterization performed with strictly sterile
technique [192, 360]. In nine children aged 10 years or
less who had developed spontaneous SEA, seven cases of
infection with Staphylococcus aureus [14, 17, 133, 271,
336, 380, 384] and one case of infection with Streptococcus pneumoniae [96] were identified. In another child
with SEA, unspecified Staphylococci were identified [32].
Among a total of 830 patients with SEA from the international literature in whom the infectious agent was
identified, there were 13 cases of fungal infections, nine
of which were due to Aspergillus fumigatus (Table 5)
[85, 126, 149, 177, 339, 341, 381]. Three of these patients had acute myeloid leukemia [85], one was HIVpositive, and three others were receiving corticosteroids
[149, 341, 381].
In three patients, parasites were identified as the causative organism for SEA (Table 5). Two Nigerian patients
had a 70-cm long worm that was operatively removed
from the epidural space. In both cases, the worm was
identified as Dracunculus medinensis, which is endemic
190
Table 6 Radiologic and isotopic examinations (n=1142) in
patients with spinal epidural
abscess
Examination technique
1954–1980
1981–1990
1991–1997
n
n
n
%
%
%
Spinal radiograph
Myelography
Computed tomography
Myelography and computed tomography
MRI
Gadolinium-enhanced MRI
Isotopic examinations*
53
104
–
–
–
–
–
34
66
–
–
–
–
–
142
104
33
89
38
–
50
31
23
7
20
8
–
11
90
112
56
96
90
57
28
17
21
11
18
17
11
5
Total
157
100
456
100
529
100
*Includes radioisotope bone scan, 111indium scan, 67gallium, and 99mtechnetium scan.
References 1, 3–5, 11–14, 16, 24, 32, 35–40, 43, 46, 50, 57, 61, 63, 65, 66, 71, 74, 77, 79, 84, 85, 89,
92, 94, 96, 97, 99, 100, 103, 108, 110, 115, 117, 121, 124, 126, 128, 133, 134, 137, 138, 142, 143,
146, 149, 152, 154, 156, 161, 166, 168, 170, 172, 176–178, 180–182, 184–189, 191, 193, 194,
198–202, 204, 206, 208, 211, 213–217, 220, 222–227, 229, 232, 233, 235, 240–242, 245, 250, 254,
256, 259, 260, 264–266, 269–271, 273, 276, 277, 280, 293, 294, 299, 300, 303–306, 308, 311,
315–318, 321, 326–328, 330, 332, 335, 336, 339, 341–343, 345, 346, 348, 351, 352, 354, 357–359,
362, 365, 372, 374, 379–385, 388, 390, 391, 395, 397–399, 401, 403.
in Africa [178], causing dracunculiasis (Medina or Guinea
worm infection). Another case involved a cyst of Echinococcus granulosus located in the epidural space at Th11 in
a 44-year-old man from New York [172]. The cyst was
also operatively removed.
Diagnosis
Radiological investigations have been the mainstay of
diagnosis for patients with SEA. However, the dictum of
Schlossberg and Shulman from 1997 [330] on the central
aspect of the diagnosis of SEA still remains valid: “The
most important step in diagnosing spinal epidural abscess is consideration of the entity.”
Before the era of computed tomography (CT) and
magnetic resonance imaging (MRI), the only imaging
modalities of use in diagnosis were conventional radiographs and myelography, and these are the only investigations mentioned in the literature up to 1980 as diagnostic methods for patients with SEA (Table 6).
However, conventional radiography of the vertebral
column is not necessarily reliable for SEA [8, 82, 164,
230, 234, 267, 314, 350, 378]. Only after arrosion or
sclerotic changes of bony structures, which may occur
after SEA has been present for some time, do conventional radiographic investigations of the spine offer valuable diagnostic information. However, radiologically visible changes are to be expected only after several weeks
of chronic SEA [8, 378]. Only nine (23%) of 39 patients
with SEA evaluated by Baker et al. [17] and 11 (37%) of
the 30 evaluated by Russell et al. [316] demonstrated
spinal radiograph findings compatible with an inflammatory spinal process. Such a low sensitivity is not sufficient for diagnostic purposes.
Myelography has long been considered the method of
choice for the diagnosis of SEA. It reveals evidence of SEA
reliably because there is contrast medium blockage above
or below the abscess [8, 82, 164, 230, 350, 378, 402].
In all three periods evaluated for this meta-analysis of
the international literature, myelography was the most
commonly used diagnostic method (Table 6). However,
MRI is beginning to replace myelography as the method of
choice; for instance, MRI alone was used in the diagnosis
of a patient with SEA published by this center [241]. Although myelography provides reliable results, it is an invasive procedure and may lead to the spread of micro-organisms into the subarachnoidal space [164, 300, 350, 402].
The introduction of CT into routine clinical use at the
beginning of the 1980s replaced myelography as the diagnostic method of first choice for patients with SEA
[41, 52, 181, 211, 270, 354, 391]. Computed tomographic imaging enables axial tomographs of the spine with
high resolution and reproducibility. In contrast to myelography, CT is noninvasive. However, delineation of the
spinal cord from the epidural space can be difficult with
CT [350]. Of nine patients with SEA, the diagnosis by
CT could not be made in six; repeat CT could only demonstrate SEA in one of these six patients [73]. Therefore,
the authors judged the value of CT in the diagnosis of
SEA to be limited. On the other hand, it may enable early diagnosis [52], which is not usually possible with conventional radiographs or sonography. In addition, CT is
advantageous for planning the operative procedure and
performing percutaneous needle biopsies of the epidural
space or of bony structures involved in osteomyelitis in
order to isolate the causative micro-organism [46, 52].
The combination of CT with myelography (myeloCT) is also possible. Among the patients represented in
this meta-analysis, myelo-CT was used more often than
CT alone (Table 6). It enables high diagnostic accuracy
and exact judgments concerning the paraspinal space
[154, 164, 350]. In one of the patients reported by O’Sullivan et al., only the use of myelo-CT enabled a definitive diagnosis, which had not been possible with CT
alone [270]. The diagnostic value of CT in patients with
SEA can also be increased by the application of intravenous contrast medium [354, 391].
191
Fig. 6 CSF protein concentration in 117 patients with spinal epidural abscess. Thirteen other patients demonstrated protein concentrations of 1500 mg/dl to 5100 mg/dl [3, 14, 17, 24, 31, 35, 36,
38, 50, 59, 60, 66, 77, 80, 84, 89, 99, 100, 103, 104, 108, 117, 121,
134, 152, 166, 172, 178, 180, 187–189, 198, 204, 206, 208,
214–216, 232, 244, 253, 266, 271, 293, 299, 315, 317, 318, 328,
330, 332, 336, 341, 342, 345, 348, 379, 381, 383, 384, 390]
In contrast to CT, MRI is able to form multiplanar
tomographic images with high contrast among soft-tissue
structures and without bone artifacts. In addition, imaging
of the spinal canal can be varied with respect to the intensities of its components by varying the excitation parameters [249, 287]. The lack of radiation represents another
advantage of MRI over CT. In this meta-analysis, the frequency of use of MRI had increased more than that of
any other diagnostic tool when comparing the periods
1981–1990 and 1991–1997 (Table 6). Hlavin et al. [154]
report a sensitivity of 91% in MRI for the definitive diagnosis of SEA. This is comparable with the 92% sensitivity with myelo-CT, which these authors evaluated in parallel. Beyond that, MRI allows spinal tumors, hematomas, transverse myelitis, spinal cord infarction, or intervertebral disk prolapse to be differentiated from SEA
[154, 203]. In one of the two patients described by
Schmutzhard et al. [332], CT demonstrated an isodense
space-occupying lesion in the spinal canal but did not allow the delineation of the lesion as an abscess, scar, or
granulomatous process. Subsequently, MRI was used to
identify the process as SEA.
The literature on SEA since 1991 has contained reports of the use of gadolinium as a contrast medium for
MRI [196, 200, 225, 261, 319, 322, 370]. In the present
meta-analysis, MRI examinations with gadolinium were
performed in 57 (11%) of the 529 radiological or nuclear-medicine examinations reported in the SEA literature
from 1991 to 1997. This rate is comparable to that of CT
(Table 6). The use of gadolinium in MRI allows better
delineation of SEA from contiguous structures [319].
Nuclear-medicine imaging is being used less and less often in the diagnosis of SEA (Table 6).
In the youngest reported patient with SEA, a 10-dayold newborn, MRI and sonography were employed to
make the diagnosis [133]. Epidurography has not yet
been used for the diagnosis of SEA [338, 373].
Figure 6 displays the distribution of CSF protein
concentrations in patients with SEA. The average CSF
protein concentration in the 130 patients for whom in-
192
formation was available in the original publications
was 538 mg/dl (normal range: 15–45 mg/dl) (references: see legend to Fig. 6). The reported concentrations
ranged from 17 mg/dl [383] to 5100 mg/dl [80]. Most
authors regard lumbar puncture as an unnecessary examination that does not contribute to exact diagnosis in
cases of suspected SEA. Because of the danger of inducing bacterial meningitis by spreading the bacteria
into the subarachnoidal space, many authors recommend that lumbar puncture should be avoided [8, 82,
154, 195, 300].
In a review from 1996 about spinal infections, Martin
and Yuan [230] categorically reject lumbar puncture:
“Lumbar puncture to obtain the cerebrospinal fluid
(CSF) for analysis is mentioned to be condemned. This
procedure risks spreading the infection to the intrathecal
compartment”.
In summary, MRI, especially in combination with gadolinium, now represents the method of first choice for
the diagnosis of SEA. It has made other diagnostic procedures essentially superfluous [99, 225, 370].
Treatment
In 1986, Strohecker and Grobovschek [361] summarized
the principles of treatment for SEA: “Although conservative treatment modalities have occasionally been reported to be successful ... , in our opinion, an operation
is the method of choice, according to the surgical motto
‘ubi pus, ibi evacua’.” This is also the conclusion of the
overwhelming majority of the articles in the literature [8,
46, 67, 82, 164, 195, 230, 241, 309, 378, 390, 402]. Surgical intervention is usually performed with dorsal access as a laminectomy or removal of the spinous process
and vertebral arch (lamina) including the posterior longitudinal ligament in order to drain the abscess [140, 239].
The surgical intervention can be combined with suctionirrigation drainage [118] and is generally performed with
pre- and postoperative administration of antibiotics.
Information on treatment details was available for
639 of the 915 patients with SEA from the international
literature (Table 7). In 567 patients (89%), surgical and
conservative treatment were combined as described
above. The most commonly used operative procedure
was laminectomy (339 patients). For 197 patients, the
surgical procedure was not specified in the corresponding publications, but it may be assumed that laminectomy was used in most cases. Intraoperative sonography is
recommended by many authors to judge the extent of the
abscess and to plan the laminectomy [101, 222, 288,
291]. Intraoperative sonography is especially useful in
the surveillance of surgical decompression of anteriorly
located extensions of the abscess [103, 222, 228].
Anterior decompression is necessary for SEA in the
anterior segment of the epidural space; this operation is
usually combined with anterior corpectomy [100, 124,
133, 177, 213]. This technique was used in only 14 of a
total of 639 patients (Table 7). Surgery is performed
Table 7 Therapeutic approach (n=639) in patients with spinal epidural abscess
Treatment
Solely conservative treatment
Combined surgical and conservative treatment
Laminectomy/hemilaminectomy
Anterior decompression
Percutaneous abscess drainage
Laminotomy
Spondylodesis
Unspecified operation
N patients
72
567
339
14
9
7
1
197
References 1, 3–5, 11–14, 16, 17, 24, 32, 33, 35–40, 43–46, 50,
55, 57–61, 63, 65, 66, 70, 71, 74, 77, 79, 81, 84, 85, 89, 92–94,
96–100, 103–108, 110, 115–117, 121–124, 126, 128, 133, 134,
137, 138, 142, 143, 147–149, 154, 156, 160, 161, 166, 168, 170,
172, 176–178, 180–182, 185–188, 191, 193, 194, 198, 199, 202,
204, 206, 208, 214–217, 220, 222–224, 226, 227, 229–233, 235,
237, 240, 241, 244–245, 250, 253, 256, 259, 260, 265, 266, 270,
271, 273, 276, 277, 293, 294, 297–299, 303–305, 308, 311,
315–318, 324, 327, 328, 332, 335, 336, 339, 341–343, 345, 346,
348, 352, 354, 357–359, 362, 365, 370, 372, 374, 379–385, 388,
397, 398, 401, 403.
from a ventral approach in the cervical region and from
either an anterior transthoracic or dorsolateral approach
for thoracic abscesses; the anterior transthoracic approach is associated with the risk of infection of the
pleural cavity or mediastinum [64].
Percutaneous abscess drainage is a further option that
was used in nine of the 639 patients [71, 235, 306, 336,
365, 384]. One of them was a 9-month-old child with
SEA [384]. The authors utilized fluoroscopically guided
percutaneous drainage because the abscess was easily delineated by MRI in the dorsolumbar portion of the epidural space. In addition, the authors did not want to expose
the child to increased risk of long-term complications following lumbar laminectomy. At the age of 15 months, the
patient displayed no neurological deficits. Walter et al.
[384] view nonlumbar or anterior localization as a relative contraindication to percutaneous drainage. Schweich
and Hurt [336] successfully performed percutaneous abscess drainage on 3-year-old and 11-year-old children,
both of whom recovered completely.
Laminotomy was performed on seven of the 639 patients from the literature for whom details of the therapeutic approach were available [108, 170, 193, 380].
Five of the seven patients were children aged between 2
months and 11 years [108, 380]. In contrast to laminectomy, laminotomy represents an attempt to close the posterior covering of the spinal canal following the surgical
intervention. Following temporary removal of the spinous processes and vertebral laminae, the spinous processes are reconnected to the roots of the vertebral arch
of the corresponding vertebrae, once abscess drainage
has been performed (Fig. 7) [239, 292].
Restoration of the anatomical integrity of the spine is
especially advantageous for children operated upon for
SEA. Kyphosis, anterior subluxation, and spinal instability are potential complications of multisegment laminec-
193
Fig. 7 Laminotomy: the continuity of the vertebral arch is repaired with osteoplastic technique. (With permission from [292]
p 558)
tomy in the pediatric age group [108, 158, 292]. Even
though two reports on laminotomy in adults have been
published [170, 193], laminectomy is the procedure of
first choice for the treatment of SEA in adults and should
be performed emergently following diagnosis [46, 164,
241, 309, 361, 378].
Only 72 (11%) of 639 patients with SEA analyzed in
this work received exclusively antibiotic treatment without operative drainage of the abscess (Table 7). In addition to several case reports, three case series concerning
the conservative treatment of SEA have been published,
although only three to six patients were analyzed in each
report [142, 211, 226]. The authors share the opinion that
conservative therapy of SEA is only indicated when patients do not yet have severe neurologic symptoms.
Computed tomography or MRI must be available at all
times in order to perform emergent radiological examinations in case the patient’s clinical condition worsens
suddenly [142]. Close follow-up is absolutely necessary.
Early identification of the causative infectious agent is
an absolute prerequisite for sole conservative treatment
[142, 226, 263] to allow specific antibiotic treatment.
For patients who have been completely paralyzed for 3
days or more, Leys et al. [211] recommend just conservative treatment, because surgical decompression is not
likely to produce significant benefits such as recovery of
neurological function after this duration of complete paralysis. However, this opinion is not shared by all clini-
cians. On the other hand, treatment with antibiotics alone
of patients with other severe medical problems may be
indicated [142, 211]. Those with large abscesses extending from the cervical to the lumbar segments may also be
candidates for conservative treatment [211]. However,
such patients have also been successfully treated surgically. For instance, Donowitz et al. [84] successfully
treated a 13-year-old girl with cervicothoracolumbar
SEA by laminectomy. The patient recovered completely.
The admittedly cautious recommendation of Hanigan et
al. [142] to treat children conservatively if an operation
would require multisegment laminectomy is debatable.
Not only the above mentioned case [84] but also the option of laminotomy represent arguments against this recommendation. In summary, purely conservative treatment is an option for selected patients but not indicated
for the majority of cases.
The antibiotics used for purely conservative or combined conservative and surgical treatment should fulfill
the criteria of Leys et al. [211]: (1) efficacy against
Staphylococcus aureus, the most common cause of SEA,
(2) low toxicity to enable treatment over several weeks,
and (3) the ability to penetrate bony tissues, as also is
necessary in treating spondylodiscitis.
Over the period of 1954 to 1997, a large number of
different antibiotics were used (Table 8) which reflect
the preferences of each center and the development of
antibiotics in the last four decades.
Initial treatment with a combination of antibiotics is indicated while waiting for definitive identification of the
causative micro-organism. In 1991, Ingham et al. [164]
recommended the combination of flucloxacillin, ampicillin, gentamicin, and metronidazole. Mampalam et al. [226]
194
Table 8 Antibiotics used for treatment of patients with spinal epidural abscess according to period
Period
Antibiotics
1954–1980
Achromycin, amikacin, azlocillin, cefazolin, ceftriaxone, cephalexin, cephalothin, chloramphenicol, chlormycetin,
cloxacillin, dicloxacillin, erythromycin, gentamicin, kanamycin, methicillin, nafcillin, oxacillin, penicillin, polybactrim,
polymyxin B, procaine-penicillin, streptomycin, sulfadiazine, sulfisoxazole, tetracycline
1981–1990
Amikacin, ampicillin, amoxicillin-clavulanic acid, benzylpenicillin, cefataxime, cefotiam, cefuroxime, cephalexin,
chloramphenicol, clindamycin, cloxacillin, dicloxacillin, erythromycin, flucloxacillin, fosfomycin, fusidinic acid,
gentamicin, lincomycin, metronidazole, minocycline, nafcillin, natilmicin, oxacillin, penicillin G, pristinamycin,
rifampicin, streptomycin, tetracycline, trimethoprim-sulfmethoxazole, tobramycin, vancomycin
1991–1997
Amikacin, amoxicillin, amoxicillin-clavulanic acid, ampicillin, arbekacin, benzyl penicillin, cefalexin, cefazolin,
cefmenoxim, cefotaxime, cefpirom, ceftazidime, ceftriaxone, cefuroxime, cephalothin, cephmetazole, cephradine,
chloramphenicol, ciprofloxacin, clindamycin, cloxacillin, dicloxacillin, doxycycline, erythromycin, flomexef-sodium,
flucloxacillin, fosfomycin, fusidinic acid, gentamicin, imipenem, metronidazole, mezlocillin, minocycline, nafcillin,
natilmycin, ofloxacin, oxacillin, penicillin G, piperacillin, rifampicin, streptomycin, sulfbactam, tobramycin,
tosufloxacin, trimethoprim-sulfmethoxazole, vancomycin
References 1, 3, 5, 12–14, 17, 32, 35–40, 44, 59–61, 65, 66, 77,
79, 84, 85, 93, 96–100, 105, 108, 110, 117, 121–123, 126–128,
133, 134, 143, 146–149, 152, 154, 160, 161, 166, 170, 178, 180,
182, 185–187, 193, 198, 201, 204, 211, 214–216, 224, 227–229,
233, 240, 241, 244, 245, 250, 254, 259, 260, 264–266, 270, 271,
273, 276, 277, 280, 293, 294, 299, 303, 304, 308, 315, 324, 326,
328, 332, 335, 336, 339, 341, 342, 345, 352, 354, 357, 358, 362,
365, 372, 374, 381, 383–385, 388, 390, 397–399, 401, 403.
recommend the combination of a third-generation cephalosporin with antibiotics with activity against Staphylococcus
such as vancomycin or nafcillin. In the case reported by
our department, in which Staphylococcus epidermidis was
isolated, vancomycin and ofloxacin were administered
[241].
The duration of antibiotic administration is up to 12
weeks but usually 4 to 6 weeks. Antibiotics should be
initially given intravenously but may be continued with
oral formulations [164, 195, 309, 323, 378, 402].
A standard antibiotic treatment regimen cannot be derived from published results. The choice of antibiotics is
based on the preferences of the clinician and the institution, resistance testing, and the country of treatment.
– 5%: two deaths in 43 patients; year of publication:
1992 [74]
– 7%: two deaths in 29 patients; year of publication:
1990 [81]
– 13%: three deaths in 23 patients; year of publication:
1994 [306]
– 13%: four deaths in 30 patients; year of publication:
1979 [316]
– 14%: five deaths in 35 patients; year of publication:
1987 [73]
– 14%: three deaths in 21 patients; year of publication:
1994 [213]
– 18%: seven deaths in 39 patients; year of publication:
1975 [16]
– 20%: eight deaths in 41 patients; year of publication:
1996 [177]
– 20%: five deaths in 25 patients; year of publication:
1992 [300]
– 22%: four deaths in 18 patients; year of publication:
1982 [401]
– 23%: nine deaths in 39 patients; year of publication:
1990 [154]
– 36%: nine deaths in 25 patients; year of publication:
1954 [158]
Outcome
In 1926, Dandy [72] published one of the first comprehensive reviews on SEA. Of the 32 patients summarized
in his report, 26 died, corresponding to a mortality rate
of 81%. In the series published by Nussbaum et al. [262]
in 1992, just two of 40 patients died (5%). Laminectomy
and antibiotic therapy have greatly improved the prognosis of this severe pyogenic infection.
For the 915 cases of SEA evaluated in the present review, information as to treatment results was available
for 599 patients. Death occurred in 94 (16%) of these patients. If results are classified according to periods, a
continual reduction in mortality from 34% to 16% can be
seen from 1954 to 1980 (Table 9). Since 1980, the mortality rate has remained constant and was 15% from
1991 to 1997 (Table 9). The large case series from 1954
to 1996 with at least 18 patients with SEA report mortality rates from 0% to 36%:
– 0%: no deaths in 21 patients; year of publication:
1991 [235]
For all patients, the average mortality of the three periods of 1971–1980, 1981–1990, and 1991–1997 was consistently between 13% and 16% (Table 9). Despite modern diagnostic methods (see “Diagnosis”), there has been
no significant change in the mortality of SEA patients in
the last 25 years. This may be related to the rarity of the
disorder, which is apparently still taken into consideration all too infrequently [195].
In parallel to the overall reduction in mortality since
1954, there has been an increase in the number of patients who completely recover from SEA (Table 9).
However, the proportion has not changed significantly
since 1971 and lies between 41% and 47% (Table 9). A
195
Table 9 Outcome for 599 patients with SEA
Outcome
1954–1960
1961–1970
1971–1980
1981–1990
1991–1997
n
n
n
n
n
%
%
%
%
%
Complete recovery
Neurological deficits (without paresis)
Paresis/paralysis
Death
8
6
5
10
28
21
17
34
8
7
5
6
31
27
19
23
63
27
24
22
46
20
18
16
89
57
19
24
47
30
10
13
90
63
34
32
41
29
16
15
Total
29
100
26
100
136
100
189
100
219
100
References 1, 3–5, 11–13, 16, 17, 32, 33, 35–40, 44, 45, 50,
57–61, 65, 66, 71, 77, 79–81, 84, 92–94, 96–100, 103–108, 110,
116, 117, 121–124, 126–128, 133–134, 137, 138, 142, 143, 146,
149, 152, 154, 156, 158, 160, 161, 166, 170, 172, 176–178,
180–182, 185–188, 191, 193, 194, 198–202, 204, 206, 208, 211,
Table 10 Outcome for 589
patients with spontaneous SEA
or following anesthesiologic
procedures in the epidural
space (epiduralanesthesia/
analgesia – single shot, catheter
insertion, and epidural steroid
application)
214–217, 222–227, 232, 233, 235, 237, 240–242, 244, 245,
253, 256, 259, 260, 264, 266, 269, 271, 276, 277, 280, 283,
294, 298–300, 303–306, 308, 311, 315–318, 321, 326–328,
335, 336, 339, 341–343, 345, 346, 348, 352, 354, 357–359,
365, 370, 372, 374, 381–385, 390, 395, 397–399, 403.
250,
293,
332,
362,
Outcome
n (%) of patients
with spontaneous SEA
n (%) of patients with SEA following
anesthetic procedures in the epidural space
Complete recovery
Neurologic deficits
except paresis/paralysis
Paresis/paralysis
Death
240 (43)
146 (26)
11 (38)
6 (21)
84 (15)
90 (16)
8 (27)
4 (14)
Total
560 (100)
29 (100)
References 1, 3–5, 11–13, 16, 17, 32, 33, 35–40, 44, 45, 50, 57–61, 65, 66, 71, 77, 79–81, 84, 92, 93,
96–100, 103–108, 110, 116, 117, 121–124, 126–128, 133, 134, 137, 138, 142, 143, 146, 149, 152,
154, 156, 158, 160, 161, 166, 170, 172, 176–178, 180–182, 185–188, 191, 193, 194, 198–202, 204,
206, 208, 211, 214–217, 222–227, 232, 233, 235, 237, 240–242, 244, 245, 250, 253, 256, 259, 260,
264, 266, 269, 271, 276, 277, 280, 283, 293, 294, 298–300, 303–306, 308, 311, 315–318, 321,
326–328, 332, 335, 336, 339, 342, 343, 345, 346, 348, 352, 354, 357–359, 362, 365, 370, 372, 374,
381–385, 390, 395, 397–399, 403.
total of 160 of 599 patients (27%) retained permanent
neurologic deficits (except paresis and paralysis) in the
period between 1954 and 1997, and 87 had either paresis
or paralysis (15%). There has been no obvious change in
these proportions over the entire period (Table 9). Tacconi et al. [366] published a case series on ten patients with
SEA in 1996 and emphasized the reduction in mortality
coupled with unchanged problems of morbidity. However, the present meta-analysis demonstrates a clear increase in the proportion of patients with complete recovery (see above), with a parallel decrease in mortality. It
is therefore conceivable that a shift has occurred between
patient groups with different outcomes. Patients who
might have died in earlier decades may today survive
with residual paralysis, those that might have developed
paralysis might today have other, milder neurologic deficits, and those who earlier would have recovered with
neurologic deficits may today experience complete recovery. This speculation seems probable but of course
cannot be proven. Proof would entail controlled, prospective case studies in which the complete therapeutic
repertoire is not administered to one group in order to
compare it with another group with all therapeutic options. This is not possible because the severity of the disorder makes it ethically impossible to withhold therapeutic options to the patients afflicted. Moreover, its rarity
would not allow sufficient numbers of cases to be evaluated.
Among the 559 patients of the international literature
for whom information on the outcome was available,
there were 29 with SEA following epidural anesthesia or
analgesia (Table 10) [33, 35, 36, 45, 57, 79, 98, 104, 106,
122, 128, 142, 152, 186, 225, 233, 241, 259, 304, 317,
318, 335, 352, 362, 365, 382, 385, 386]. Patients in
whom instrumental intervention of the spinal canal had
been performed (i.e., mostly catheterization) had a mortality rate of 14%. The mortality of patients with spontaneous SEA was slightly higher at 16% (Table 10). Complete recovery was experienced by 43% of patients with
spontaneous SEA and 38% of patients with SEA following anesthesiological interventions in the epidural space.
The percentage of patients with mild neurological deficits
was 26%, clearly higher than the 21% with such deficits
due to SEA following anesthesiologic interventions in the
epidural space. Paresis or paralysis developed in 27% of
SEA patients following anesthesiologic interventions,
which was significantly higher than in patients with spontaneous SEA (15%). All in all, there has been no improvement in outcome for patients with SEA following
anesthetic or analgesic procedures in the epidural space.
A possible explanation for this observation could be that
signs of the infection in the epidural space are confused
196
with consequences of the operation or the disorder for
which the operation was performed. This could in turn
lead to delays in the diagnosis of SEA. On the other hand,
the index of suspicion for a pyogenic infection of the epidural space should be higher in these patients, especially
if back pain and fever develop. Kindler et al. [179] found
a similar outcome in their retrospective study of 42 cases
of SEA following insertion of peridural catheters in patients treated between 1974 and 1996; 45% of patients recovered completely and 48% had residual neurological
deficits (including paresis or paralysis).
The case published by this department was unusual in
that the infection developed under antibiotic therapy
[241]. Since the patient developed only a mild neurological deficit, surgical decompression in combination with
further antibiotic therapy led to complete recovery. It
may be assumed that the index of suspicion for SEA is
higher with patients who have received nerve block procedures near the spine or other procedures in the spinal
canal than for patients developing spontaneous SEA.
This confirms once again the statement of Strohecker
and Grobovchek [361]: “The problem with spinal epidural abscesses is not treatment, but early diagnosis – before massive neurological symptoms occur.”
It is often emphasized in textbooks and review articles that early diagnosis and timely therapeutic intervention in the form of surgical decompression lead to better
outcomes for patients with SEA [8, 67, 82, 156, 164,
230, 231, 309, 378, 402]. More severe preoperative neurologic deficits are generally associated with a worse
prognosis, which is confirmed by all the large case series
of the literature [16, 73, 81, 154, 213, 235, 300, 306,
316, 401].
Additionally, the duration of neurologic abnormalities
has an influence on the outcome. Heusner [151] noted in
1948 that SEA patients without paralysis preoperatively
or whose paralysis had developed less than 36 hours before the operation had better prognoses with respect to
survival and recovery of function. In contrast, no patients with paralysis developing 48 hours or more before
surgical decompression showed recovery of neurologic
function. All of the deaths reported in the work of
Heusner [151] occurred in this patient group. These results from more than three decades ago were confirmed
by Yang [401] in a case series on 18 patients with SEA.
Russell et al. [316] also reported a lack of recovery of
neurologic function in five of a total of 30 patients.
These five patients had demonstrated paraplegia before
surgery. Information on the duration of paraplegia is not
available in that publication. The duration of 36 hours,
after which, according to Heusner [151], patients with
paralysis due to SEA generally do not recover neurologic
function, was recently confirmed by Rigamonti et al.
[306]. The extent of spinal cord compression is thought
to represent the decisive pathophysiological factor for
outcome and prognosis [177]. A rehabilitation program
is desirable for SEA patients [389].
In summary, analysis of 915 patients from the international literature on SEA allows the following conclu-
sion: a high index of suspicion for the early symptoms
and signs of SEA (spinal back pain, local tenderness,
meningitic signs, and parameters of acute inflammation), emergent diagnostic measures (MRI), and treatment (laminectomy, antibiotics) before onset of neurologic deficits can prevent these deficits from becoming
permanent.
References
1. Abdel-Magid RA, Kotb HIM (1990) Epidural abscess after
spinal anesthesia: favorable outcome. Neurosurgery 27:310–
311
2. Abouleish E, Orig T, Amortegui AJ (1980) Bacteriologic comparison between epidural and caudal techniques. Anesthesiology 53:511–514
3. Aicardi J, Lepintre J (1967) Spinal epidural abscess in a 1month-old child. Am J Dis Child 114:665–667
4. Aitken RJ, Wright JP, Bok A, Elliot MS (1986) Crohn’s desease precipitating a spinal extradural abscess and paraplegia.
Br J Surg 73:1004–1005
5. Akai S, Ohno T, Kamura S, Kadoya S, Kadoya H, Matsuzaki 0
(1990) Epidural abscess following septic arthritis in a rheumatoid patient. Spine 15:603–605
6. Albers E (1833) Die Entzündung der harten Haut des Rückenmarks, perimeningitis medullae spinalis. J Chir Augenheilk
19:347–369
7. Albright L, Toczek S, Brenner VJ, Ommaya AK (1974) Osteomyelitis and epidural abscess caused by Arachnia propionica.
Case report. J Neurosurg 40:115–119
8. Allen Jr MB, Beveridge WD (1985) Spinal epidural and subdural abscesses. In: Wilkins RH, Rengachary SS (eds): Neurosurgery. McGraw-Hill, New York, pp 1972–1975
9. Allen SS, Kahn EA (1932) Acute pyogenic infection of the
spinal epidural space. J Am Med Ass 98:875–878
10. Altrocchi PH (1963) Acute spinal epidural abscess vs. acute
transverse myelopathy. Arch Neurol 9:17–25
11. Angtuaco EJC, McConnell JR, Chadduck WM, Flanigan S
(1987) MR imaging of spinal epidural sepsis. Am J Roentgenol
149:1249–1253
12. Ashpole RD, Jenkins A (1995) Cervical extradural abscess
complicating discitis and associated disc prolapse, secondary
to a long line infection. J Neurol Neurosurg Psychiatr 58:510–
511
13. Atkinson L, Deambrosis W, Sheehy J, Teoh G (1978) A spinal
epidural abscess complicating an intrauterine contraceptive
device. Aust N Z J Obstetr Gynecol 18:272–273
14. Bader B, Labrune P, Zerah M, Lebras P, Wood C, Devictor 0,
Huault G (1986) Traitement médical d’une épidurite à staphylocoques. Arch Fr Pédiatr 43:335–336
15. Bailey M, Love JG (1941) Epidural abscess of the spinal cord:
report of two cases. Proc Mayo Clin 16:24–28
16. Baker AS, Ojemann RG, Swartz MN, Richardson EP (1975)
Spinal epidural abscess. New Engl J Med 293:463–468
17. Baker CJ (1971) Primary spinal epidural abscess. Am J Dis
Child 121:337–339
18. Barreto RS (1962) Bacteriological culture of indwelling epidural catheters. Anesthesiology 23:643–646
19. Bartal AD, Schiffer J, Heilbronn YD, Yahel M (1972) Anterior
interbody fusion for cervical osteomyelitis. Reversal of quadriplegia after evacuation of epidural spinal abscess. J Neurol
Neurosurg Psychiatr 35:133–136
20. Bartels RHMA, Gonera EG, van der Spek JAN, Thijssen
HOM, Mullaart RA, Gabreëls FJM (1995) Intramedullary spinal cord abscess. A case report. Spine 20:1199–1204
21. Barth H (1901) Operative Behandlung der eitrigen Meningitis.
Dtsch Med Wochenschr I:136–138
22. Batson 0V (1940) The function of the vertebral veins and their
role in the spread of metastases. Ann Surg 112:138–149
197
23. Bauer M, Knippenberger H, Kubli F, Fournier D, Radacic A,
Mehringer R (1979) Die bakterielle Kontamination von EpiduralKathetern mit aeroben und anaeroben Keimen in der Geburtshilfe. Geburtsh Frauenheilk 39:1054–1056
24. Beaudoin MG, Klein L (1984) Epidural abscess following
multiple spinal anaesthetics. Anaesth Intens Care 12:163–164
25. Beland B, Prien T, van Aken H (1997) Rückenmarknahe
Regionalanästhesien bei Bakteriämie. Anaesthesist 46:536–
547
26. Bellerose A, Amyot R (1932) Metastatic epidural abscess of
the spinal marrow. Can Med Ass J 27:629–631
27. Bensheim H (1928) Über Peripachymeningitis spinalis externa
purulenta. Ztschr Ges Neurol Psychiatr 110:777–782)
28. Bergamaschi G (1820) Sulla mielitide stenica e sul tetano, loro
identia. Thesis. G. Torri, Parvia
29. Bergman I, Wald ER, Meyer JD, Painter MJ (1983) Epidural
abscess and vertebral osteomyelitis following serial lumbar
punctures. Pediatrics 72:476–480
30. Bertino RE, Porter BA, Stimac GK, Tepper SJ (1988) Imaging
spinal osteomyelitis and epidural abscess with short TI inversion recovery (STIR). Am J Neuroradiol 9:563–564
31. Bischof W, Nittner K (1965) Zur Pathogenese, Klinik und
Behandlung des spinalen Epiduralabszesses. Zentralbl Neuchir
26:193–210
32. Bock SA, Sickler D, Chhabra OP (1976) Spinal epidural abscess in a five-week-old infant. A problem in diagnosis. Clin
Pediatr 15:286–287
33. Boden 0 (1954) Letale Komplikation einer extraduralen Spinalanaesthesie. Anaesthesist 3:127–128
34. Boharas S, Koskoff YD (1941) The early diagnosis of acute
spinal epidural abscess. J Am Med Ass 117:1085–1088
35. Bollensen E, Menck S, Buzanoski J, Prange HW (1993) Iatrogenic epidural spinal abscess. Clin Invest 71:780–786
36. Bollensen E, Prange HW (1991) Epiduraler spinaler Abszeß
als letale Komplikation einer Periduralanaesthesie. Reg Anaesth 14:101–103
37. Bonomo RA, Strauss M, Blinkhorn R, Salata RA (1996) Torulopsis (Candida) glabrata: a new pathogen found in spinal
epidural abscess. Clin lnfect Dis 22:588–589
38. Boongird P, Vimolchalao M, Khantanaphar S, Bunyaratavej S
(1974) Acute spinal epidural abscess: a report of two cases,
one with autopsy findings. J Med Ass Thailand 57:564–570
39. Borum SE, McLeskey CH, Williamson JB, Harris FS, Knight
AB (1995) Spinal epidural abscess. Anesthesiology 82:1523–
1526
40. Bouchez B, Arnott G, Delfosse JM (1985) Acute spinal epidural abscess. J Neurol 231:343–344
41. Brant-Zawadzki M, Burke VD, Jeffrey RB (1983) CT in the
evaluation of spine infection. Spine 8:358–364
42. Braun H (1922) Über Epimeningitis spinalis. Zentralbl Chir
35:1374–1376
43. Brian Jr JE, Westerman GR, Chadduck WM (1984) Septic
complications of chemonucleolysis. Neurosurgery 15:730–734
44. Brisseau JM, Derriennic M, Fritz A, Leveiller D, Courtieu AL,
Barrier JH (1988) Septicaemia due to Actinobacillus actinomycetemcomitans with endocarditis and spinal epidural abscess. J Infection 17:131–134
45. Bromage PR (1993) Spinal extradural abscess: pursuit of vigilance. Br J Anaesth 70:471–473
46. Broner FA, Garland DE, Zigler JE (1996) Spinal infections in
the immunocompromised host. Orthop CIin North Am 27:
37–46
47. Browder J, Meyers R (1937) Infections of the spinal epidural
space: an aspect of vertebral osteomyelitis. Am J Surg 37:4–26
48. Browder J, Meyers R (1941) Pyogenic infections of the spinal
epidural space. A consideration of the anatomic and physiologic pathology. Surgery 10:296–308
49. Brown SM, Stimmel B, Taub RN, Kochwa S, Rosenfield RE
(1974) lmmunologic dysfunction in heroin addicts. Arch Intern Med 134:1001–1006
50. Bullock R (1981) Unusual presentation of pyogenic spinal epidural abscess. A case report. South Afr Med J 59:723–724
51. Bunch GH, Madden E (1933) Acute extradural abscess with
compression of the cord. Am J Surg 20:763–771
52. Burke DR, Brant-Zawadzki M (1985) CT of pyogenic spine
infection. Neuroradiology 27:131–137
53. Buruma 0JS, Craane H, Kunst MW (1979) Vertebral osteomyelitis and epidural abscess due to mucormycosis. A case report. Clin Neurol Neurosurg 81:39–44
54. Butler EG, Dohrmann PJ, Stark RJ (1988) Spinal subdural abscess. CIin Exp Neurol 25:67–70
55. Byers K, Axelrod P, Michael S, Rosen S (1995) lnfections
complicating tunneled intraspinal catheter systems used to
treat chronic pain. Clin lnfect Dis 21:403–408
56. Carson D, Wildsmith JAW (1995) The risk of extradural abscess. Br J Anaesth 75:520–521
57. Chan S-T, Leung S (1989) Spinal epidural abscess following
steroid injection for sciatica. Spine 14:106–108
58. Chang C-Z, Huang T-Y, Howng S-L (1997) Spinal epidural
abscess – a case report. Kaoshiung J Med Sci 13:457–461
59. Chappel R, Verhelst JA, Nagler JM, Dom L, Appel B, Herregods P (1986) Epidural abscess causing tetraparesis: case report. Paraplegia 24:364–369
60. Chee YC, Poh SC (1983) Aspergillus epidural abscess in a patient with obstructive airway disease. Postgrad Med J 59:43–
45
61. Chen JY, Chen W-J, Huang T-J, Shih C-H (1992) Spinal epidural abscess complicating cervical spine fracture with hypopharyngeal perforation. A case report. Spine 17:971–974
62. Cheng T-C, Tsai T-C, Lin G-J (1995) Successful medical treatment for staphylococcal vertebral osteomyelitis complicated
by spinal epidural abscess, psoas abscess and meningitis: a
case report. Kaohsiung J Med Sci 113:295–299
63. Clark R, Carlisle JT, Valainis GT (1989) Streptococcus pneumoniae endocarditis presenting as an epidural abscess. Rev Infect Dis 11:338–340
64. Cohen I (1938) Epidural spinal infections. Ann Surg 108:
992–1000
65. Colle I, Peeters P, Le Roy I, Diltoer M, D’Haens J (1996) Epidural abscess: case report and review of the literature. Acta
CIin Belg 51:412–416
66. Cooper K (1984) Spinal epidural abscess: a case report. Del
Med J 56:343–348
67. Corboy JR, Price RW (1993) Myelitis and toxic, inflammatory, and infectious disorders. Curr Opin Neurol Neurosurg
6:564–570
68. Corradini EW, Turney MF, Browder EJ (1948) Spinal epidural
infection. N Y State J Med 48:2367–2370
69. Craig WMK, Doyle JB (1932) Metastatic epidural abscess of
the spinal cord. Recovery after operation. Ann Surg 95:58–
66
70. Crawford JS (1975) Pathology in the extradural space. Br J
Anaesth 47:412–415
71. Cwikiel W (1991) Percutaneous drainage of abscess in psoas
compartment and epudural space. Case report and review of
the literature. Acta Radiol 32:159–161
72. Dandy WE (1926) Abscesses and inflammatory tumors in the
spinal epidural space (so-called pachymeningitis externa).
Arch Surg 13:477–494
73. Danner RL, Hartman BJ (1987) Update of spinal epidural abscess: 35 cases and review of the literature. Rev lnfect Dis
9:265–274
74. Darouiche RO, Hamill RJ, Greenberg SB, Weathers SW,
Musher DM (1992) Bacterial spinal epidural abscess. Review
of 43 cases and literature survey. Medicine 71:369–385
75. David CV, Balasubramaniam P (1981) Acute osteomyelitis
of the spine with paraplegia. Aust N Z J Surg 51:544–545
76. Davidson IM (1947) Pseudomonas pyocyanea meningitis following spinal analgesia. Lancet I:653–654
77. Davidson RI, Zajac JJ (1978) Spinal epidural empyema with
Brown-Séquard syndrome. Surg Neurol 9:195–197
78. Davies JM, Thistlewood JM, Rolbin SH, Douglas MJ (1988)
Infections and the parturient: anaesthetic considerations. Can
J Anaesth 35:270–277
198
79. Dawson P, Rosenfeld JV, Murphy MA, Hellyar AG (1991)
Epidural abscess associated with postoperative epidural analgesia. Anaesth Intens Care 19:569–572
80. Dei-Anang K, Hase U, Schürmann K (1990) Epidural spinal
abscesses. Neurosurg Rev 13:285–288
81. Del Curling O, Gower DJ, McWhorter JM (1990) Changing
concepts in spinal epidural abscess: a report of 29 cases.
Neurosurgery 27:185–192
82. DiNubile MJ (1990) Spinal epidural abscess. In: Schlossberg
D (ed) Infections of the nervous system. Springer, Berlin
Heidelberg New York Tokyo, pp 171–178
83. Donathan ER (1944) Acute epidural abscess of the spinal canal. Complete recovery following emergency laminectomy
and penicillin. J Am Med Ass 126:956–957
84. Donowitz LG, Cole WQ, Lohr JA (1983) Acute spinal epidural abscess presenting as hip pain. Pediatr lnfect Dis 2:44–45
85. Dubbeld P, van Oostenbrugge RJ, Twijnstra A, Schouten HC
(1996) Spinal epidural abscess due to Aspergillus infection
of the vertebrae: report of 3 cases. Neth J Med 48:18–23
86. Ducroq JL, Tchaoussoff J, Zarour J, Le Gars D, Dorde T,
Boulard M, Ossart M (1983) Complication iatrogène infectieuse
de l’analgésie péridurale en continu. Agressologie 24:179–180
87. Du Pen SL, Peterson DG, Williams A, Bogosian AJ (1990)
Infection during chronic epidural catheterization: diagnosis
and treatment. Ansthesiology 73:905–909
88. Durity F, Thompson GB (1968) Localized cervical extradural
abscess. Case report. J Neurosurg 28:387–390
89. Dús V (1960) Spinal peripachymeningitis (epidural abscess).
J Neurosurg 17:973–983
90. Echols DH (1941) Emergency laminectomy for acute epidural abscess of the spinal canal. Report of four cases with recovery in three. Surgery 10:287–295
91. Editorial (1988) Frequency of neurosurgical disorders in the
UK. Br J Neurosurg 2:281–283
92. Egashira T, Nose T, Maki Y (1984) A case of spinal epidural
abscess in a renal transplant recipient. No Shinkei Geka 12:
739–744
93. Elian D, Hassin D, Tomer A, Bank H, Eisenstein Z (1984)
Spinal epidural abscess: an unusual complication of bacterial
endocarditis. Infection 12:258–259
94. Elias M (1994) Cervical epidural abscess following trigger
point injection. J Pain Sympton Manage 9:71–72
95. Emmanuel ER (1994) Post-sacral extradural catheter abscess
in a child. Br J Anaesth 73:548–549
96. Enberg RN, Kaplan RJ (1974) Spinal epidural abscess in
children. Early diagnosis and immediate surgical drainage is
essential to forestall paralysis. CIin Pediatr 13:247–253
97. Epstein S, Holden M, Feldshuh J, Singer JM (1963) Unusual
case of spinal cord compression: nocardiosis. N Y State J
Med 63:3422–3427
98. Ericsson M, Algers G, Schliamser SE (1990) Spinal epidural
abscesses in adults: review and report of iatrogenic cases.
Scand J lnfect Dis 22:249–257
99. Erntell M, Holtas S, Norlin K, Dahlquist E, Nilsson-Ehle I
(1988) Magnetic resonance imaging in the diagnosis of spinal epidural abscess. Scand J Infect Dis 20:323–327
100. Esposito DP, Gulick TA, Sullivan HG, Allen MB Jr (1984)
Acute anterior spinal epidural abscess. South Med J
77:1171–1172
101. Feldenzer JA, McKeever PE, Schaberg DR, Campbell JA,
Hoff JT (1987) Experimental spinal epidural abscess: a pathophysiological model in the rabbit. Neurosurgery 20:859–867
102. Feldenzer JA, McKeever PE, Schaberg DR, Campbell JA,
Hoff JT (1988) The pathogenesis of spinal epidural abscess:
microangiographic studies in an experimental model. J Neurosurg 69:110–114
103. Feldenzer JA, Waters DC, Knake JE, Hoff JT (1986) Anterior cervical epidural abscess: the use of intraoperative spinal
sonography. Surg Neurol 25:105–108
104. Ferguson JF, Kirsch WM (1974) Epidural empyema following thoracic extradural block. Case report. J Neurosurg 41:
762–764
105. Ferree BA, Stambough JL, Greiner AL (1989) Spinal epidural abscess. A case report and literature review. Orthop Rev
18:75–80
106. Fine PG, Hare BD, Zahniser JC (1988) Epidural abscess following epidural catherization in a chronic pain patient: a diagnostic dilemma. Anesthesiology 69:422–424
107. Firsching R, Frowein RA, Nittner K (1985) Acute spinal epidural empyema. Observations from seven cases. Acta Neurochir 74:68–71
108. Fischer EG, Greene CS, Winston KR (1981) Spinal epidural
abscess in children. Neurosurgery 9:257–260
109. Forsythe M, Rothman RH (1978) New concepts in the
diagnosis and treatment of infections of the cervical spine.
Orthop Clin North Am 9:1039–1051
110. Frank B, Dörr F, Penkert G, Vogel E, Tidow G (1991) Epiduraler spinaler Abszeß mit Kaudasymptomatik als Komplikation
eines Morbus Crohn. Dtsch Med Wochenschr 116:1313–1316
111. Fraser RAR, Ratzan K, Wolpert SM, Weinstein L (1973) Spinal subdural empyema. Arch Neurol 28:235–238
112. Friedman DP, Hills JR (1994) Cervical epidural spinal infection: MR imaging characteristics. Am J Roentgenol 163:699–
704
113. Fuchs L (1920) Kasuistischer Beitrag zur Pathologie der
Rückenmarkshäute (Pachymeningitis externa). Dtsch Ztschr
Nervenheilkde 66:231–242
114. Fuente Aguado J, Arzuaga Torre JA, Yusta Izquierdo A,
Garcı́a Andrade L, Martínez López de Letona L (1992)
Absceso epidural espinal. Experienca de ocho años (Spinal
epidural abscess. 8 years’ experience). Rev Clı́n Esp 191:
264–266
115. Fujimoto N, Nabatame H, Nakamura K, Konishi T, Kamiya
Y (1991) Spinal epidural abscess as the cause of torticollis –
diagnosis by magnetic resonance imaging. Clin Neurol
31:49–53
116. Fujiwara S, Morioka T, lshibashi H, Takaki T, Fukui M
(1994) Acute purulent discitis with epidural abscess of the
cervical spine in an adult. Case report. Neurol Med Chir
34:382–384
117. Fukui T, Ichikawa H, Kawate N, Nozawa T, Sugita K (1992)
Acute spinal epidural abscess and spinal leptomeningitis: report of 2 cases with comparative neuroradiological and autopsy study. Eur Neurol 32:328–333
118. Garrido E, Rosenwasser RH (1983) Experience with the suction-irrigation technique in the management of spinal epidural infection. Neurosurgery 12:678–679
119. Gastmeier K, Gastmeier P (1991) Zur lnfektionsgefahr bei
Patienten mit Langzeitperiduralkathetern. Anaesthesiol Reanimat 16:267–270
120. Gasul BM, Jaffe RH (1935) Acute epidural spinal abscess – a
clinical entity. Arch Pediatr 52:361–390
121. Gelber BR, Pierson EW, Birkmann LW (1981) Spinal epidural abscess. Nebraska Med J 66:10–14
122. Gelfand MS, Bakhtian BJ, Simmons BP (1991) Spinal sepsis
due to Streptococcus milleri: two cases and review. Rev Infect Dis 13:559–563
123. Ghosh K, Duncan R, Kennedy PGE (1988) Acute spinal epidural abscess caused by Streptococcus milleri. J lnfection
16:303–312
124. Gilliland K (1989) Epidural abscesses of the spine: case
comparisons. J Neurosci Nurs 21:185–189
125. Glazer PA, Hu SS (1996) Pediatric spinal infections. Orthop
Clin North Am 27:111–123
126. Go BM, Ziring DJ, Kountz DS (1993) Spinal epidural abscess due to Aspergillus sp in a patient with acquired immunodeficiency syndrome. South Med J 86:957–960
127. Godeau B, Brun-Buisson C, Brugières P, Roucoules J,
Schaeffer A (1993) Complete resolution of spinal epidural
abscess with short medical treatment alone. Eur J Med
2:510–511
128. Goucke CR, Graziotti P (1990) Extradural abscess following
local anaesthetic and steroid injection for chronic low back
pain. Br J Anaesth 65:427–429
199
129. Grant FC (1945) Epidural spinal abscess. J Am Med Ass
128:509–512
130. Griffiths HED, Jones DM (1971) Pyogenic infection of the
spine. A review of twenty-eight cases. J Bone Joint Surg 53
B:383–391
131. Grossman MO, Kesert BH, Voris HC (1944) Chronic spinal
epidural granuloma. Report of two cases. Ann Surg 119:897–
902
131. Gruß P, Friedrich B, Mertens HG, Bockhorn J (1976) Purulent osteomyelitis of the cervical spine with epidural abscess.
Operative treatment by means of dorsal and ventral approach. Clin Neurol Neurosurg 79:57–61
133. Gudinchet F, Chapuis L, Berger D (1991) Diagnosis of anterior cervical spinal epidural abscess by US and MRI in a
newborn. Pediatr Radiol 21:515–517
134. Guerrero IC, Slap GB, MacGregor RR, Lawner P, Ruggeri S,
Gennarelli T (1978) Anaerobic spinal epidural abscess. Case
report. J Neurosurg 48:465 469
135. Gutmanis PA, Rorabeck CH (1979) Pyogenic infections of
the adult spine. Can J Surg 22:162–166
136. Guttmann E, Singer L (1931) Der Epiduralabszeß oder die
Pachymeningitis spinalis externa purulenta. Arch Klin Chir
166:183–191
137. Hainline B, Tuszynski MH, Posner JB (1992) Ataxia in epidural spinal cord compression. Neurology 42:2193–2195
138. Hakin RN, Burt AA, Cook JB (1979) Acute spinal epidural
abscess. Paraplegia 17:330–336
139. Hamada J, Sato K, Seto H, Ushio Y (1991) Epidural tuberculoma of the spine: case report. Neurosurgery 28:161–163
140. Hamer J, Dosch C (1978) Neurochirurgische Operationen.
Springer, Berlin Heidelberg New York
141. Hancock DO (1973) A study of 49 patients with acute spinal
extradural abscess. Paraplegia 10:285–288
142. Hanigan WC, Asner NG, Elwood PW (1990) Magnetic resonance imaging and the nonoperative treatment of spinal epidural abscess. Surg Neurol 34:408–413
143. Harris LF, Haws FP, Triplett JN Jr, Maccubbin DA (1987)
Subdural empyema and epidural abscess: recent experience
in a community hospital. South Med J 80:1254–1258
144. Harston PK (1992) Spinal epidural abscess as a complication
of duodenolumbar fistula. A case report. Spine 17:593–596
145. Hassin GB (1925) Circumscribed suppurative (nontuberculous) peripachymeningitis. Histopathologic study of a case.
Arch Neurol Psychiatr 20:110–129
146. Hefter H, Piontek M, Aulich A (1991) Bacterial meningitis
and dorsal spinal epidural abscess caused by Crohn’s disease.
Neurology 41:606–608
147. Henderson JM, Coonrod JD (1990) Spinal epidural abscess:
an unusual complication of a duodenal ulcer. J Clin Gastroenterol 12:672–674
148. Hendlisz A, Rypens F, Gay F, Vangossum A, Cremer M
(1995) Spinal epidural abscess after endoscopic treatment for
chronic pancreatitis. Pancreas 10:204–215
149. Hendrix WC, Arruda LK, Platts-Mills TAE, Haworth CS,
Jabour R, Ward GW Jr (1992) Aspergillus epidural abscess
and cord compression in a patient with aspergilloma and empyema. Survival and response to high dose systemic amphotericin therapy. Am Rev Resp Dis 145:1483–1486
150. Henneberg NN (1921) Pachymeningitis externa purulenta nach
Nackenabsceß. Zentralbl Ges Neurol Psychiatr 25:96–97
151. Heusner AP (1948) Nontuberculous spinal epidural infections. New EngI J Med 239:845–854
152. Higuchi T, Imagawa A, Murahashi M, Hara H, Wakayama Y
(1996) Spinal epidural abscess associated with epidural anesthesia: gadolinium-enhanced magnetic resonance imaging
and its usefulness in diagnosis and treatment. Intern Med
35:902–904
153. Hinz F (1921) Ueber einen Fall von Perimeningitis purulenta. Dtsch Med Wochenschr 47:1229
154. Hlavin ML, Kaminsky HJ, Ross JS, Ganz E (1990) Spinal
epidural abscess: a ten-year perspective. Neurosurgery 27:
177–184
155. Holt HM, Andersen SS, Andersen 0, Gahrn-Hansen B,
Siboni K (1995) Infections following epidural catheterization. J Hosp lnfect 30:253–260
156. Hoppe B (1996) Spinal epidural abscess: the nurse’s role in
early detection and intervention. Heart Lung 25:463–466
157. Horlocker TT, McGregor DG, Matsushige DK, Schroeder
DR, Besse JA, Perioperative Outcome Group (1997) A retrospective review of 4767 consecutive spinal anesthetics: central nervous system complications. Anesth Analg 84:578–
584
158. Hulme A, Dott NM (1954) Spinal epidural abscess. Br Med J
I:64–68
159. Hunt J (1904) Acute infectious osteomyelitis of the spine and
acute suppurative perimeningitis. Med Rec 65:641–650
160. Hunter JC, Ryan MD, Taylor TKF, Pennington JC (1977)
Spinal epidural abscess in pregnancy. Aust N Z J Surg 47:
672–674
161. Hutchinson FD (1955) Non-tuberculous spinal epidural abscess. Can Med Ass J 72:208–210
162. Hutton PW (1956) Acute osteomyelitis of cervical spine with
epidural abscess. Br Med J I:153–154
163. Iannettoni MD, Whyte RI, Orringer MB (1995) Catastrophic
complications of the cervical esophagogastric anastomosis. J
Thorac Cardiovasc Surg 110:1493–1501
164. Ingham HR, Sisson PR, Mendelow AD, Kalbag RM (1991)
Pyogenic neurosurgical infections. Edward Arnold, London
165. Jabban R, Pierce JF (1977) Spinal cord compression due to
pseudomonas in a heroin addict. Case report. Neurology 27:
1034–1037
166. Jacobsen FS, Sullivan B (1994) Spinal epidural abscesses in
children. Orthopedics 17:1131–1138
167. Jakobsen KB, Christensen M-K, Carlsson PS (1995) Extradural anaesthesia for repeated surgical treatment in the presence of infection. Br J Anaesth 75:536–540
168. Jamison MH, Stanworth P, Maclennan I (1984) An usual rectal fistula: extradural abscess discharging per rectum. Br J
Surg 71:651–652
169. Johnston JDH, Ashbell TS, Rosomoff HL (1962) Isolated intraspinal extradural tuberculosis. New Engl J Med 266:703–
705
170. Junila J, Niinimäki T, Tervonen 0 (1997) Epidural abscess
after lumbar discography. A case report. Spine 22:2191–
2193
171. Kannangara DW, Tanaka T, Thadepalli H (1981) Spinal epidural abscess due to Actinomyces israelii. Neurology 31:202–204
172. Kaufman DM, Kaplan JG, Litman N (1980) lnfectious agents
in spinal epidural abscesses. Neurology 30:844–850
173. Kayser FH, Bienz KA, Eckert J, Zinkernagel RM (1998) Medizinische Mikrobioiogie. 9. Aufl. Georg Thieme, Stuttgart
174. Keefer CS (1950) Bacterial meningitis following lumbar
puncture and spinal anesthesia. Am Pract 1:679–682
175. Keienburg F (1924) Über akute eitrige Perimeningitis. Med
Klin 19:640–642
176. Keon-Cohen BT (1968) Epidural abscess simulating disc hemia. J Bone Joint Surg 50B:128–130
177. Khanna RK, Malik GM, Rock JP, Rosenblum ML (1996)
Spinal epidural abscess: evaluation of factors influencing
outcome. Neurosurgery 39:958–964
178. Khwaja MS, Dossetor JFB, Lawrie JH (1975) Extradural
guinea-worm abscess. J Neurosurg 43:627–630
179. Kindler CH, Seeberger MD, Staender SE (1998) Epidural abscess complicating epidural anesthesia and analgesia. An analysis of the literature. Acta Anaesthesiol Scand 42:614–620
180. King M (1994) Spinal epidural abscess: an elusive diagnosis.
South Med J 87:288–289
181. Kirzner H, Oh YK, Lee SH (1988) Intraspinal air: a CT finding of epidural abscess. Am J Roentgenol 151:1217–1218
182. Kitching AJ, Rice ASC (1993) Extradural abscess in the
postpartum period. Br J Anaesth 70:703–704
183. Klein M, Ahn C-S, Drum DE, Tow DE (1994) Gallium-67
scintigraphy as an aid in the detection of spinal epidural abscess. Clin Nucl Med 19:761–762
200
184. Klinger M, Druschky KF, Mokrusch T (1987) Spinaler epiduraler Abszeß als Ursache eines progredienten Querschnittsyndroms. In: Bock WJ, Schirmer M (eds) Differentialdiagnosen in der Neurochirurgie. Urban and Schwarzenberg,
Munich, pp 170–172
185. Klygis LM, Reisberg BE (1991) Spinal epidural abscess
caused by group G streptococci. Am J Med 91:89–90
186. Knight JW, Cordingley JJ, Palazzo MGA (1997) Epidural abscess following epidural steroid and local anaesthetic injection. Anaesthesia 52:576–585
187. Kolmos HJ (1979) Spinal epidural abscess in patients on
maintenance haemodialysis (a presentation of two cases). Int
Urol Nephrol 11:249–253
188. Koopmann CF Jr, Miller RW, Couthard SW (1984) Retropharyngeal abscess associated with progressive quadriplegia,
an epidural abscess, renal failure, and jaundice. Otolaryngol
Head Neck Surg 92:114–118
189. Koppel BS, Tuchman AJ, Mangiardi JR, Daras M, Weitzner I
(1988) Epidural spinal infections in intravenous drug abusers. Arch Neurol 45:1331–1337
190. Korbin WM (1958) Spinal epidural abscess. Review of the
literature and report of case. Bull Los Angeles Neurol Soc
23:21–26
191. Korovessis P, Sidiropoulos P, Piperos G, Karagiannis A
(1993) Spinal epidural abscess complicated closed vertebral
fracture. A case report and review of literature. Spine 18:
671–674
192. Kost-Byerly S, Tobin JR, Greenberg RS, Billett C, Zahurak
M, Yaster M (1998) Bacterial colonization and infection rate
of continuous epidural catheters in children. Anesth Analg
86:712–716
193. Kotilainen E, Sonninen P, Kotilainen P (1996) Spinal epidural
abscess: an unusual cause of sciatica. Eur Spine J 5:201–203
194. Kotlarek F, Schütz E (1981) Spinaler Epiduralabszeß bei
einem 11 Jahre alten Jungen. Monatsschr Kinderheilkde 129:
48–50
195. Krauss WE, McCormick PC (1992) Infections of the dural
spaces. Neurosurg Clin North Am 3:421–433
196. Kricun R, Shoemaker EI, Chovanes GI, Stephens HW (1992)
Epidural abscess of the cervical spine: MR findings in five
cases. Am J Roentgenol 158:1145–1149
197. Krumdieck N, Stevenson L (1940) Spinal epidural abscess
associated with actinomycosis. Arch Pathol 30:1223–1226
198. Kuiters RRF, Douma G, Hekster REM (1987) Spinal epidural abscess. Report of two cases. Clin Neurol Neurosurg 89:
255–260
199. Kuwata T, Nishiguchi T, Kinoshita Y, Kubo K, Komai N
(1990) Cervical spinal epidural abscess: case report. No
Shinkei Geka 18:285–288
200. Lang IM, Hughes DG, Jenkins JPR, St Clair Forbes W,
McKenna F (1995) MR imaging appearances of cervical epidural abscess. Clin Radiol 50:466–471
201. Lange M. Tiecks F, Schielke E, Yousry T, Haberl R, Oeckler R
(1993) Diagnosis and results of different treatment regimes in
patients with spinal abscesses. Acta Neurochir 125:105–114
202. Larkin EB, Scott SD (1994) Metastatic paraspinal abscess
and paraplegia secondary to dental extraction. Br Dent J
177:340–342
203. Larsson E-M, Holtås S, Cronqvist S (1988) Emergency magnetic resonance examination of patients with spinal cord
symptoms. Acta Radiol 29:69–75
204. Lasker BR, Harter DH (1987) Cervical epidural abscess.
Neurology 37:1747–1753
205. Latronico N, Tansini A, Gualandi GF, Bonetti M, Scalzini A,
Candiani A, Chiesa A (1993) Successful nonoperative treatment of tuberculous spinal epidural abscess: the role of magnetic resonance imaging. Eur Neurol 33:177–180
206. Law WB (1969) Acute spinal epidural abscess. Aust N Z J
Surg 38:354–357
207. Lee HJ, Bach JR, White RE (1992) Spinal epidural abscess
complicating vertebral osteomyelitis: an insidious cause of
deteriorating spinal cord function. J Am Paraplegia Soc 15:
19–21
208. Leela Naidu PS, Dinakar I, Sitharma W, Rao SY, Sathyavathi
S, Fakruddin L (1979) Spinal epidural abscess of Salmonella
etiology – case report. Indian J Pathol Microbiol 22:367–370
209. Leonhardt H, Tillmann B, Töndury G, Zilles K (1987) Anatomie des Menschen. Lehrbuch und Atlas. Bd. 3: Nervensystem, Sinnesorgane. Georg Thieme, Stuttgart
210. Lewitzky P (1877) Ein Fall von Peripachymeningitis spinalis. Berlin Klin Wochenschr 14:227–230
211. Leys D, Lesoin F, Viaud C, Pasquier F, Rousseaux M, Jomin
M, Petit H (1985) Decreased morbidity from acute bacterial
spinal epidural abscesses using computed tomography and nonsurgical treatment in selected patients. Ann Neurol 17:350–355
212. Leys D, Lesoin F, Destee A, Jomin M, Christiaens JL, Warot
P (1984) Les épidurites aigues à germes banales. Vingt-trois
observations. Presse Méd 13:597–599
213. Liem LK, Rigamonti D, Wolf AL, Robinson WL, Edwards
CC, DiPatri A (1994) Thoracic epidural abscess. J Spinal Dis
7:449–454
214. Lindner A, Warmuth-Metz M, Becker G, Toyka KV (1997)
Iatrogenic spinal epidural abscesses: early diagnosis essential
for good outcome. Eur J Med Res 2:201–205
215. Liu K-N, Wu J-M, Fairholm D (1981) Acute spinal epidural
abscess. Report of 4 cases. Taiwan I Hsueh Hui Tsa Chi
(J Formosan Med Ass) 80:971–976
216. Loar CR, Connelly TR (1976) Acute spinal epidural abscess.
West Virg Med J 72:56–57
217. Loarie DJ, Fairley HB (1978) Epidural abscess following
spinal anesthesia. Anesth Analg 57:351–353
218. MacDonald NS (1928) A case of intraspinal extradural abscess. J Am Med Ass 90:1114–1115
219. Madsen JR, Rosenberg AE (1992) Case records of the Massachusetts General Hospital: weekly clinico-pathological exercises. New Engl J Med 16:1070–1076
220. Mahendru V, Bacon DR, Lema MJ (1994) Multiple epidural
abscesses and spinal anesthesia in a diabetic patient. Case
Report. Reg Anesth 19:66–68
221. Maier C (1997) Aufklärung bei epiduralem Blutpatch. Anaesthesist 46:255–258
222. Mak KH, Au KK, Fung KY, Chan YW (1996) Spinal epidural
abscess: a report of nine cases and the use of intra-operative
ultrasonography. Aust N Z Surg 66:287–290
223. Makiuchi T, Kondo T, Yamakawa K, Shinoura N, Yatsushiro
K, Ichi S, Yoshioka M (1993) Stellate ganglion blocks as the
suspected route of infection in a case of cervical epidural abscess. No Shinkei Geka 21:805–808
224. Male CG, Martin R (1973) Puerperal spinal epidural abscess.
Lancet I:608–609
225. Mamourian AC, Dickman CA, Drayer BP, Sonntag VKH
(1993) Spinal epidural abscess: three cases following spinal
epidural injection demonstrated with magnetic resonance imaging. Anesthesiology 78:204–207
226. Mampalam TJ, Rosegay H, Andrews BT, Rosenblum ML,
Pitts LH (1989) Nonoperative treatment of spinal epidural infections. J Neuosurg 71:208–210
227. Mansharami RK, Singh SD, Ranawat CS, Jain AC, Thakar
SV (1965) Spinal epidural abscess. Case reports with review
of literature. Ind J Med Sci 19:136–141
228. Marks WA, Bodensteiner JB (1987) Anterior cervicaI epidural abscess with pneumococcus in an infant. J Child Neurol
2:25–29
229. Martin MJ, Lee PYC (1995) Streptoccoccus mitis causing
epidural abscess. Postgrad Med J 71:251
230. Martin RJ, Yuan HA (1996) Neurosurgical care of spinal epidural, subdural, and intramedullary abscesses and arachnoiditis. Orthop Clin North Am 27:125–136
231. Maslen DR, Jones SR, Crislip MA, Bracis R, Dworkin RJ,
Flemming JE (1933) Spinal epidural abscess. Optimizing patient care. Arch Intern Med 153:1713–1721
232. Mattle H, Jaspert A, Forsting M, Sieb JP, Hanny P, Ebeling V
(1986) Der akute spinale Epiduralabszeß. Dtsch Med
Wochenschr 111:1642–1646
233. McDonogh AJ, Cranney BS (1984) Delayed presentation of
an epidural abscess. Anaesth Intens Care 12:364–365
201
234. McGahan JR, Dublin AB (1985) Evaluation of spinal infections
by plain radiographs, computed tomography, intrathecal metrizamide, and CT-guided biopsy. Diagn Imag Clin Med 54:11–20
235. McGee-Collett M, Johnston IA (1991) Spinal epidural abscess: presentation and treatment. A report of 21 cases. Med
J Aust 155:14–17
236. McKnight P, Friedman J (1992) Torticollis due to cervical
epidural abscess and osteomyelitis. Neurology 42:696–697
237. Meinecke F-W (1964) Epiduraler Absceß am Rückenmark
als Unfallfolge. Monatsschr Unfallheilkde 67:403–405
238. Menezes AH, Graf CJ, Perret GE (1977) Spinal cord abscess:
a review. Surg Neurol 8:461–467
239. Merrem G, Goldhahn W-E (1981) Neurochirurgische Operationen. 2. Aufl. Springer, Berlin Heidelberg New York
240. Messer HD, Lenchner GS, Brust JCM, Resor S (1977) Lumbar spinal abscess managed conservatively. J Neurosurg
46:825–829
241. Michel H, Steffen P, Weichel T, Seeling W (1997) Epiduritis
nach Langzeitschmerztherapie mit einem Epiduralkatheter –
eine Literaturübersicht mit einem aktuellen Fallbericht. Anaesthesiol Reanimat 22:69–79
242. Miller WH, Hesch JA (1962) Nontuberculous spinal epidural
abscess. Report of a case in a 5-week-old infant. Am J Dis
Child 104:269–275
243. MilIs CK, Spiller WG (1902) Case of external spinal pachymeningitis, implicating the entire ventral surface of the spinal
dura. Brain 25:318–325
244. Mincks JR, Pulaski EJ (1956) Acute lumbar epidural abscess
in a thirty months old child; complete recovery following
surgery and antibiotic therapy. Antibiot Med Clin Ther 3:
202–206
245. Mitohell DIG, Crandon IW (1994) Spinal epidural abscess.
Diagnosis and management. West Ind Med J 43:104–106
246. Mitchell WRD (1938) Acute epidural abscess. Br Med J
I:1149–1151
247. Mixter WJ (1916) A case of epidural intraspinal abscess of
pyogenic origin. Boston Med Surg J 175:864–865
248. Mixter WJ, Smithwick RH (1932) Acute intraspinal epidural
abscess. New Engl J Med 207:126–131
249. Modic MT, Masaryk T, Paushter D (1986) Magnetic resonance
imaging of the spine. Radiol Clin North Am 24:229–245
250. Mooney RP, Hockberger RS (1987) Spinal epidural abscess:
a rapidly progressive disease. Ann Emerg Med 16:1168–1170
251. Morawitz P (1919) Über akute eitrige Perimeningitis (Peripachymeningitis). Ein charakteristisches Krankheitsbild bei
Staphylokokkenerkrankungen. Dtsch Arch Klin Med 128:
294–316
252. Morgagni G (1967) De sedibus et causis morborum (Venedig
1761). Ausgewählt, übertragen, eingelassen und mit Erkenntnissen versehen von Markwart Michler. Huber, Bern
253. Müller A (1969) Der spinale epidurale Abszeß. Schweiz
Arch Neurol Neurochir Psychiatr 104:247–273
254. Murr MM, Metcalf AM (1993) Spinal epidural abscess complicating an ileal J-pouch-anal anastomosis. Report of a case.
Dis Colon Rectum 36:293–294
255. Negrin J Jr, Clark RA Jr (1952) Pyogenic subdural abscess of
the spinal meninges. Report of two cases. J Neurosurg 9:
95–100
256. Ngan Kee WD, Jones MR, Thomas P, Worth RJ (1992) Extradural abscess complicating extradural anaesthesia for caesarean section. Br J Anaesth 69:647–652
257. Nickels JH, Poulos JG, Chaouki K (1989) Risks of infection
from short-term epidural catheter use. Reg Anesth 14:88–89
258. Nonne M (1926) Otogener Rückenmarksabsceß. Z HalsNasen- und Ohrenheilkde 13:574–579
259. Nordström O, Sandin R (1993) Delayed presentation of
an extradural abscess in a patient with alcohol abuse. Br J
Anaesth 70:368–369
260. North JB, Brophy BP (1979) Epidural abscess: a hazard of
spinal epidural anaesthesia. Aust N Z J Surg 49:484–485
261. Numaguchi Y, Rigamonti D, Rothman MI, Sato S, Mihara F,
Sadato N (1993) Spinal epidural abscess: evaluation with gadolinium-enhanced MR imaging. Radiographics 13:545–559
262. Nussbaum ES, Rigamonti D, Standiford H, Numaguchi Y,
Wolf AL, Robinson WL (1992) Spinal epidural abscess: a report of 40 cases and review. Surg Neurol 38:225–231
263. Obana WG, Rosenblum ML (1992) Nonoperative treatment
of neurosurgical infections. Neurosurg Clin North Am 3:
359–373
264. Obrador GT, Levenson DJ (1996) Spinal epidural abscess in
hemodialysis patients; report of three cases and review of the
literature. Am J Kidney Dis 27:75–83
265. Oepen G, Thron A, Thoden U (1978) Pyocyaneusspondylitis
mit Epiduralabszeß und Kompression. Nervenarzt 49:609–612
266. Ohtsuka K, Tsunoda M, Nakagawa Y, Tashiro K (1979) Spinal epidural abscess – report of two cases. No Shinkei Geka
7:499–504
267. Olcott EW, Dillon WP (1993) Plain film clues to the diagnosis of spinal epidural neoplasm and infection. Neuroradiology 35:288–292
268. Oppenheim EA (1910) Ueber einen Fall von extraduraler
Spinaleiterung. Berlin Klin Wochenschr 47:1412–1414
269. Oriscello RG, Fineman S, Vigario JC, Robertello ME (1994)
Epidural abscess as a complication of coronary angioplasty: a
rare but dangerous complication. Cath Cardiovasc Diagn
33:36–38
270. O’Sullivan R, McKenzie A, Hennessy 0 (1988) Value of CT
scanning in assessing location and extent of epidural and
paraspinal inflammatory conditions. Aust Radiol 32:203–206
271. Palmer JJ, KeIly WA (1972) Epidural abscess in a 3-weekold infant: a case report. Pediatrics 50:817–820
272. Pareyson D, Savoiardo M, D’Incerti L, Sghirlanzoni A
(1995) Spinal epidural abscess complicating tuberculous
spondylitis. Ital J Neurol Sci 16:321–325
273. Park CW, Shin YS, Shin MJ, Koh SH, Chang KU, Ahn YB,
Chang YS, Bang BK (1997) Pyoderma gangraenosum and
spinal epidural abscess after subcutaneous administration of
recombinant human erythropoeitin. Nephrol Dial Transplant
12:1506–1508
274. Parnass SM, Schmidt KJ (1990) Adverse effects of spinal
and epidural anaesthesia. Drug Safety 5:179–194
275. Patronas NJ, Marx WJ, Duda EE (1979) Radiographic presentation of spinal abscess in the subdural space. Am J
Roentgenol 132:138–139
276. Pawlak A, Learner J, Kurtz A, Blackburn GW (1984) Spinal
epidural abscess caused by streptococcus MG intermedius
with complicating leptomeningitis: report of case. J Am
Osteopath Ass 83:589–592
277. Paz IF, Alvarez FJ, Roda IM, Frutos R, IsIa A (1994) Spinal
epidural abscess caused by Brucella: case report. J Neurosurg
Sci 38:245–249
278. Pegues DA, Carr DB, Hopkins CC (1994) Infectious complications associated with temporary epidural catheters. Clin
lnfect Dis 19:970–972
279. Penkert G, Fliedner E (1982) Spinales subdurales Empyem
nach Stromverletzung. Unfallheilkunde 85:473–477
280. Pérez-Calvo J, Matamala C, Sanjoaquin I, RodriguezBenavente A, Ruiz-Laiglesia F, Bueno-Gomez J (1994)
Epidural abscess due to acute Brucella melitensis infection.
Arch Intern Med 154:1410–1411
281. Petersen IM, Awad I, Ahmad M, Bay J, McHenry MC (1983)
Nocardia osteomyelitis and epidural abscess in the nonimmunosuppressed host. Cleve Clin Quart 50:453–459
282. Petty BG, Burrow CR, Robinson RA, Bulkley GB (1985)
Haemophilus aphrophilus meningitis followed by vertebral
osteomyelitis and suppurative psoas abscess. Am J Med
78:159–162
283. Phillips GE, Jefferson A (1979) Acute spinal epidural abscess.
Observations from fourteen cases. Postgrad Med J 55:712–715
284. Pincoffs MC (1949) Epidural spinal abscess with paraplegia.
A report of three cases. Int Clin 3:49–69
285. Pirofski L-A, Casadevall A (1990) Mixed staphylococcal and
cryptococcal epidural abscess in a patient with AIDS. Rev
Infect Dis 12: 964–965
286. Pollak E (1931) Zur Frage der Perimeningitis. Arb Neurol
Inst Univ Wien 33:297–314
202
287. Post MJD, Bowen BC, Sze G (1991) Magnetic resonance imaging of spinal infection. Rheum Dis Clin North Am 17:773–794
288. Post MJD, Quencer RM, Montalvo BM, Katz BH, Eismont
FJ, Green BA (1988) Spinal infection: evaluation with MR
imaging and intraoperative US. Radiology 169:765–771
289. Puckett HL, Harris EM (1933) Acute epidural abscess of the
spinal cord. US Nav Med Bull 31:29–301
290. Pujol S, Torrielli R (1996) Accidents neurologiques après
anesthésie péridurale en milieu obstétrical. Cah Anesthesiol
44:341–345
291. Quencer RM, Montalvo BM, Green BA, Eismont FJ (1984) Intraoperative spinal sonography of soft-tissue masses of the spinal cord and spinal canal. Am J Nucl Med Radiol 5:507–515
292. Raimondi AJ, Gutierrez FA, Di Rocco C (1976) Laminotomy
and total reconstruction of the posterior spinal arch for spinal
canal surgery in childhood. J Neurosurg 45:555–560
293. Ramani A, Raja A, Kundaje GN, Subramania NKR, Shubha
K (1986) Acute spinal epidural abscess. J Ass Phys Ind
34:522–523
294. Rana PVS, Raghunath D, Parakkal U, Biswas S, Prakash S
(1985) Spinal epidural abscess due to Salmonella Group C
monophasic 1.5. J Neurosurg 62:942–943
295. Rangell L, Glassman F (1945) Acute spinal epidural abscess
as a complication of lumbar puncture. J Nerv Ment Dis
102:8–18
296. Rankin RM, Flothow PG (1946) Pyogenic infection of the spinal epidural space. West J Surg Obstetr Gynecol 54:320–323
297. Ravicovitch MA, Spallone A (1982) Spinal epidural abscesses. Surgical and parasurgical management. Eur Neurol 21:
347–354
298. Rea GL, McGregor JM, Miller CA, Miner ME (1992) Surgical treatment of the spontaneous spinal epidural abscess.
Surg Neurol 37:274–279
299. Reddy DR, Murthy DK, Waghray VN, Krishna RV (1972) A
case of chronic spinal epidural abscess. J Ass Phys Ind
20:803–804
300. Redekop GJ, Del Maestro RF (1992) Diagnosis and management of spinal epidural abscess. Can J Neurol Sci
19:180–187
301. Reeves DL (1940) Acute metastatic spinal epidural abscess.
Report of two cases with recovery following laminectomy.
Arch Surg 41:994–1003
302. Renck H (1995) Neurological complications of central nerve
blocks. Acta Anaesthesiol Scand 39:859–868
303. Renjhen RC, Nangia VK, Joshi V, Khare NP (1977) Spinal
epidural abscess. A case report. Neurol India 25:193–194
304. Reynolds PC, Hahn MB (1991) Early diagnosis of a spinal
epidural abscess. Reg Anesth 16:57–58
305. Richmond BK, Schmidt JH III (1994) Seventeen level laminectomy for extensive spinal epidural abscess: case report
and review. West Virg Med J 90:468–471
306. Rigamonti D, Liem L, Wolf AL, Fiancada MS, Numaguchi
Y, Hsu FP, Nussbaum ES (1994) Epidural abscess in the cervical spine. Mt Sinai J Med 61:357–362
307. Roberts WA (1988) Pyogenic vertebral osteomyelitis of a
lumbar facet joint with associated epidural abscess. Spine 13:
948–952
308. Rockney R, Ryan R, Knuckey N (1989) Spinal epidural abscess. An infectious emergency. Case report and review. Clin
Pediatr 28:332–334
309. Roos KL (1988) Epidural abscess: a review. Indiana Med
81:1015–1019
310. Rosamond E (1932) Epidural abscess complicated by staphylococcus meningitis. Report of a case with complete recovery
following operation. J Pediatr 1:230–232
311. Rossitch E Jr, Maggio MI, Andersson EE (1991) Spinal epidural abscess as a complication of Fournier’s gangrene. J
Neurosurg Sci 35:225–227
312. Rubin G, Michowiz SD, Ashakenasi A, Tadmor R, Rappaport ZH (1993) Spinal epidural abscess in the pediatric age
group: case report and review of the literature. Pediatr Infect
Dis J 12:1007–1011
313. Rucks WW, Nicholson BH (1932) Acute epidural abscess of
the spinal canal of probable metastatic origin. Am J Dis
Child 43:569–662
314. Ruiz A, Post MJD, Ganz WI (1995) Inflammatory and infectious processes of the cervical spine. Neuroimag Clin North
Am 5:401–425
315. Rushworth RG, Martin PB (1958) Acute spinal epidural abscess: a case in an infant with recovery. Arch Dis Child 33:
261–264
316. Russell NA, Vaughan R, Morley TP (1979) Spinal epidural
infection. Can J Neurol Sci 6:325–328
317. Rustin MHA, Flynn MD, Commes EN (1983) Acute sacral
epidural abscess following local anaesthetic injection. Postgrad Med J 59:399–400
318. Saady A (1976) Epidural abscess complicating thoracic epidural analgesia. Anesthesiology 44:244–246
319. Sadato N, Numaguchi Y, Rigamonti D, Kodama T, Nussbaum E, Sato S, Rothman M (1994) Spinal epidural abscess
with gadolinium-enhanced MRI: serial follow-up studies and
clinical correlations. Neuroradiology 36:44–48
320. Sakuragi T, Higa K, Dan K, Okubo M (1990) Skin disinfectants for nerve blocks and their long-lasting antimicrobial effects. Masui 39:328–334
321. Sanchez J, Jiménez-Escrig A, Saldaña C, Santoni C, de Andres R, Varrela de Seijas E (1992) Cervical epidural abscess:
approaches to diagnosis. J Neurosurg Sci 36:121–125
322. Sandhu FS, Dillon WP (1992) Spinal epidural abscess: evaluation with contrast-enhanced MR imaging. Am J Roentgenol 158:405–411
323. Sapico FL (1996) Microbiology and antimicrobial therapy of
spinal infections. Orthop Clin North Am 27:9–13
324. Sarrell WG, LaFia DJ (1954) Acute lumbar epidural abscess.
Report of a case. New Engl J Med 250:318–320
325. Sato S, Sakuragi T, Dan K (1996) Human skin flora as a potential source of epidural abscess. Anesthesiology 85:1276–
1282
326. Sawada M, Iwamura M, Hirata T, Sakai N (1996) Cervical
discitis associated with spinal epidural abscess caused by
methicillin-resistent Staphylococcus aureus. Case report.
Neurol Med Chir 36:40–44
327. Sawamura Y, Kinomoto H, Mabuchi S, Iwasaki Y, Itoh T,
Abe H, Tsuru M (1983) A case of spinal epidural abscess extending over twenty-one vertebral levels from cervical
through lumbar spines. No Shinkei Geka 11:299–303
328. Scerpella EG, Wu S, Oefinger PE (1994) Case report of spinal epidural abscess caused by Haemophilus paraphrophilus.
J Clin Microbiol 32:563–564
329. Schick K (1909) Pachymeningitis spinalis externa purulenta
als Metastase nach Diplokokkenbronchitis. Wien Klin
Wochenschr 22:1185–1188
330. Schlossberg D, Shulman JA (1977) Spinal epidural abscess.
South Med J 70:669–673
331. Schmalz A (1925) Über akute Pachymeningitis spinalis externa. Virchow Arch Pathol Anat 257:521–560
332. Schmutzhard E, Aichner F, Dierckx RA, Gerstenbrand F,
Willeit J (1986) New perspectives in acute spinal epidural
abscess illustrated by two cases. Acta Neurochir 80:105–108
333. Schnell U (1962) Zur Pathogenese und Klinik der spinalen
Epiduralabscesse. Klinische Studie unter besonderer Berücksichtigung der Wirbelosteomyelitis. Dissertation, University of Cologne
334. Schregel W, Hartmann K, Schmitz C, Baumgärtner D, Cunitz
G (1992) Entzündliche Fistel nach Periduralkatheter. Anaesthesist 41:346–347
335. Schröter J, Djamba D, Hoffmann V, Bach A, Motsch J (1997)
Epidural abscess after combined spinal-epidural block. Can J
Anaesth 44:300–304
336. Schweich PJ, Hurt TL (1992) Spinal epidural abscess in children: two illustrative cases. Pediatr Emerg Care 8:84–87
337. Scott DB, Hibbard BM (1990) Serious non-fatal complications associated with extradural block in obstetric practice.
Br J Anaesth 64:537–541
203
338. Seeling W, Tomczak R, Merk J, Mrakovcic N (1995) Vergleich von konventionellen und CT-Epidurographien nach
Kontrastmittelinjektion über thorakale Epiduralkatheter.
Anaesthesist 44:24–36
339. Seres JL, Ono H, Benner EJ (1972) Aspergillosis presenting
as a spinal cord compression. Case report. J Neurosurg 36:
221–224
340. Seward DNL (1976) Some problems in diagnosis and management of spinal tuberculosis. Med J Aust 1:822–826
341. Sheth NK, Varkey B, Wagner DK (1985) Spinal cord Aspergillus invasion – complication of an aspergilloma. Am J Clin
Pathol 84: 763–769
342. Shintani S, Tanaka H, Irifune A, Mitoh Y, Udono H, Kaneda
A, Shiigai T (1993) Iatrogenic acute spinal epidural abscess
with septic mengitis: MR findings. Clin Neurol Neurosurg
94:253–255
343. Shirakawa H, Nakagawa H, Itoh H, Hara H, Asano G (1988)
Spinal epidural abscess in a hemodialyzed patient. Case report. Neurol Med Chir 28:828–832
344. Shope TR, Garrett AL, Waecker NJ Jr (1994) Mycobacterium bovis spinal epidural abscess in a 6-year-old boy with
leukemia. Pediatrics 93:835–837
345. Siao TC P, McCabe P, Yagnik P (1989) Nocardial spinal epidural abscess. Neurology 39:996
346. Siao TC P, Yagnik P (1988) Spinal epidural abscess. J Emerg
Med 6:391–396
347. Sirang H (1977) Chronischer, epiduraler intraspinaler Abszeß
nach Lumbalpunktion. Neurochirurgia 20:173–177
348. Slade WR, Lonano F (1990) Acute spinal epidural abscess.
J Natl Med Assoc 82:713–716
349. Slaughter RF, Fremont-Smith F, Munro D (1934) Metastatic
spinal epidural abscess. Report of a case with recovery following operation. J Am Med Assoc 102:1468–1470
350. Smith AS, Blaser SI (1991) Infectious and inflammatory processes of the spine. Radiol Clin North Am 29:809–827
351. Sollmann W-P, Gaab MR, Panning B (1987) Lumbales epidurales Hämatom und spinaler Abszeß nach Periduralanaesthesie. Reg Anaesth 10:121–124
352. Sowter MC, Burgess NA, Woodsford PV, Lewis MH (1992)
Delayed presentation of an extradural abscess complicating
thoracic extradural analgesia. Br J Anaesth 68:103–105
353. Spencer WH (1879) Case of idiopathic inflammation of the
spinal dura mater (Pachymeningitis spinalis externa). Lancet
I:836–837
354. Spiegelmann R, Findler G, Faibel M, Ram Z, Shacked I,
Sahar A (1991) Postoperative spinal epidural empyema.
Clinical and computed tomography features. Spine 16:1146–
1179
355. Spiller WG, Wright VWM (1921) Extradural abscess of the
midthoracic region of the spinal canal secondary to a boil in
the neck. Arch Neurol Psychiatr 5:107–108
356. Stammers FAR (1938) Spinal epidural suppuration, with special reference to osteomyelitis of the vertebrae. Br J Surg
26:366–374
357. Stephanides GC, Gibson RM (1984) Paraplegia caused by
spinal epidural abscess. Postgrad Med J 64:603–605
358. Stewart IC, Fort MJ, Heading RC, Mendelow AD, Harris P
(1981) Acute spinal epidural abscess – a cause of meningism.
Scott Med J 26:348–349
359. Still JM, Abramson R, Law EJ (1995) Development of
an epidural abscess following staphylococcal septicemia in
an acutely burned patient: case report. J Trauma 38:958–
959
360. Strafford MA, Wilder RT, Berde CB (1995) The risk of infection from epidural analgesia in children: a review of 1620
cases. Anesth Analg 80:234–238
361. Strohecker J, Grobovschek M (1986) Der spinale Epiduralabszeß: eine interdisziplinäre Notfallsituation. Zentralbl Neurochir 47:120–124
362. Strong WE (1991) Epidural abscess associated with epidural
catherization: a rare event? Report of two cases with markedly delayed presentation. Anesthesiology 74:943–946
363. Struyf N, Moens E, Convens C, Verhelst J, Appel B, Mahler
C (1994) Tuberculous spinal epidural abscess: case report
and MR imaging. Eur J Radiol 18:36–37
364. Sugimoto M, Tamaki N, Nagashima T, Kokunai T, Tomita H,
Sakagami Y (1993) Cervical spinal epidural abscess caused
by methicillin-resistant staphylococcus aureus (MRSA).
Spine 20:10–21
365. Tabo E, Ehkuma Y, Kimura S, Nagaro T, Arai T (1994) Successful percutaneous drainage of epidural abscess with epidural needle and catheter. Anesthesiology 80:1393–1395
366. Tacconi L, Johnston FG, Symon L (1986) Spinal epidural abscess – review of 10 cases. Acta Neurochir 138:520–523
367. Takenaka K, Kobayash H, Niikawa S, Hattori T, Ohkuma A,
Nokura H, Sakai N, Yamada H, Sasaoka I (1989) Spinal subdural abscess – report of a case and a review of the literature
of 43 cases. No To Shinkei 41: 331–336
368. Taylor AS, Kennedy F (1923) A case of extrathecal abscess
of the spinal cord. Arch Neurol Psychiatr 9:652–653
369. Teele DW, Dashefsky B, Rakusan T, Klein JO (1981) Meningitis after lumbal puncture in children with bacteremia. New
Engl J Med 305:1079–1081
370. Teman AJ (1992) Spinal epidural abscess. Early detection
with gadolineum magnetic resonance imaging. Arch Neurol
49:743–746
371. Terada Y, Matsunobe S, Nemoto T, Tsuda T, Tonomura S
(1992) Contents of chest wall cold abscess flowing into the
epidural space in miliary tuberculosis. Chest 101:590–591
372. Tolnay M, Bregenzer T, Heilbronner R, Stock KW, Dalquen
P (1996) Zervikale prävertebrale Phlegmone mit spinalem
Epiduralabszess als Komplikation des Diabetes mellitus.
Schweiz Rundschau Med 85:197–202
373. Tomczak R, Seeling W, Rieber A, Sokiranski R, Rilinger N,
Brambs H-J (1996) Die Epidurographie: Vergleich mit CT-,
Spiral-CT- und MR-Epidurographie. Fortschr Röntgenstr
165:123–129
374. Tyson GW, Grant A, Strachan WE (1979) Spinal epidural abscess presenting as acute abdomen in a child. Br J Surg
66:3–4
375. Ungemach J, Jaminet M, Striebel J-P, Tolksdorf W (1980)
Bakteriologische Studien bei Katheter-Periduralanästhesien.
In: Weis KH, Cunitz G (eds) Anaesthesiologie und Intensivmedizin. Vol 130. Springer, Berlin Heidelberg New York,
pp 362–365
376. Usubiaga JE (1975) Neurological complications following
epidural anesthesia. Int Anaesthesiol Clin 13:1–157
377. Vandermeulen E, Gogarten W, van Aken H (1997) Risiken
und Komplikationsmöglichkeiten der Periduralanästhesie.
Anaesthesist 46 [Suppl 3]:179–186
378. Verner EF, Musher DM (1985) Spinal epidural abscess. Med
Clin North Am 69:375–384
379. Vilke GM, Honingford EA (1996) Cervical spine epidural
abscess in a patient with no predisposing risk factors. Ann
Emerg Med 27:777–780
380. de Villiers JC, de Clüver PF (1978) Spinal epidural abscess
in children. South Afr J Surg 16:149–155
381. Wagner DK, Varkey B, Sheth NK, Da Mert GJ (1985) Epidural abscess, vertebral destruction, and paraplegia caused by
extending infection from an aspergilloma. Am J Med 78:
518–522
382. Waldman SD (1991) Cervical epidural abscess after cervical
epidural nerve block with steroids. Anesth Analg 72:713–718
383. WaIIace JR, Luchi M (1995) Crohn’s disease complicated by
meningitis and spinal epidural abscess. Clin Infect Dis 20:
1428–1429
384. Walter RS, King JC Jr, Manley J, Rigamonti D (1991) Spinal
epidural abscess in infancy: successful percutaneous drainage
in a nine-month-old and review of the literature. Pediatr Infect Dis J 19:860–864
385. Wang JS, Fellows DG, Vakharia S, Rosenbaum AE, Thomas
PS (1996) Epidural abscess – early magnetic resonance imaging detection and conservative therapy. Anesth Analg
82:1069–1071
204
386. Wanscher M, Riishede L, Krogh B (1985) Fistula formation
following epidural catheter. A case report. Acta Anaesthesiol
Scand 29:552–553
387. Watts JW, Mixter WJ (1931) Spinal epidural granuloma.
New Engl J Med 204:1336–1344
388. Wayne DA, Muizelaar JP (1989) Acute lumbosacral epidural
abscess after percutaneous transluminal angioplasty. Am J
Med 87:478
389. Weingarden SI, Swarczinski C (1991) Non-granulomatous
spinal epidural abscess: a rehabilitation perspective. Paraplegia 29:628–631
390. Wheeler D, Keiser P, Rigamonti D, Keay S (1992) Medical
management of spinal epidural abscesses: case report and review. Clin Infect Dis 15:22–27
391. Whelan MA, Schonfeld S, Post JD, Svigals P, Meisler W,
Weingarten K, Kricheff II (1985) Computed tomography of
nontuberculous spinal infection. J Comp Ass Tomogr 9:280–
287
392. Wildsmith JAW (1993) Extradural abscess after dentral neural block. Br J Anaesth 70:387–388
393. Wilensky AO (1929) Osteomyelitis of the vertebrae. Ann
Surg 89:561–570
394. Williams KN, Jackowski A, Evans PJD (1990) Epidural haematoma requiring surgical decompression following repeated
cervical epidural steroid injections for chronic pain. Pain
42:197–199
395. Williams JW, Powell T (1990) Epidural abscess of the cervical spine: case report and literature review. Br J Radiol 63:
576–578
396. Wilson G, Kammer AG (1934) Abscess of the spinal epidural
area. Med Clin North Am 18:287–295
397. van Winter JT, Nielsen SNJ, Jr PL (1991) Epidural abscess
associated with intravenous drug abuse in a pregnant patient.
Mayo Clin Proc 66:1036–1039
398. Wszolek ZK, McCashland TM, Witte RJ, Brandenberg GA,
Steg RE (1996) Spinal epidural abscess in a liver transplant
recipient. Transplant Proc 28:2978–2979
399. Wu L-L, Chen S-T, Tang L-M (1994) Nonsurgical treatment
of spinal epidural abscess: report of a case. Taiwan I Hsueh
Hui Tsa Chi (J Formos Med Ass) 93:253–255
400. Wulf H, Striepling E (1990) Postmortem findings after epidural anaesthesia. Anaesthesia 45:357–361
401. Yang S-Y (1982) Spinal epidural abscess. N Z Med J
95:302–304
402. Youmans JR (1985) Infections of the spine and spinal cord.
In: Youmans JR (ed) Neurological surgery. Part XI: Infection. Saunders, Philadelphia, pp 3270–3304
403. Yu L, Emans JB (1988) Epidural abscess associated with
spondylosis. A case report. J Bone Joint Surg 70A:444–447
404. Zamora-Chávez A, Soto-Mancilla JL, de la O-Cavazos ME,
Berrones Espericueta D, Paredes-Diaz E (1990) Epidural spinal abscess. Diagnosis of a case using nuclear magnetic resonance. Bol Med Hosp Infant Mex 47:589–592
405. Zenz M, Piepenbrock S, Hüsch M, Schappler-Scheele B,
Neuhaus R (1981) Erfahrungen mit längerliegenden Periduralkathetern – peridurale Morphin-Analgesie bei Karzinompatienten. Reg Anaesth 4:26–28