Journal of Clinical Neuroscience 68 (2019) 111–116

Contents lists available at ScienceDirect

Journal of Clinical Neuroscience

journal homepage: www.elsevier.com/locate/jocn

Clinical study

The clinical characteristics, implicated pathogens and therapeutic

outcomes of culture-proven septic cavernous sinus thrombosis
Che-Wei Hsu a, Wan-Chen Tsai a, Chia-Yi Lien a, Jun-Jun Lee a,b, Wen-Neng Chang a,⇑
a
Department of Neurology, Chang Gung Memorial Hospital-Kaohsiung Medical Center, Chang Gung University College of Medicine, Kaohsiung, Taiwan
b
Department of Information Management, National Sun Yat-sen University, Kaohsiung, Taiwan

a r t i c l e i n f o a b s t r a c t

Article history: In this magnetic resonance imaging-based study, we investigated the clinical features, neuroimaging
Received 16 April 2019 features and therapeutic outcomes of 14 adults (eight men and six women; mean age 60.4 years; range
Accepted 6 July 2019 37–77 years) with septic cavernous sinus thrombosis (CST). Of the underlying conditions, 10 had diabetes
mellitus and 13 had concomitant sphenoid sinusitis. Headache (n = 13) and ophthalmoplegia (n = 13)
were the most common clinical presentations, followed by fever (n = 9) and other neuro-vascular signs
Keywords: and symptoms. The duration from the onset of symptoms to diagnosis ranged from 1 to 61 days, and
Cavernous sinus thrombosis
more than 64% (9/14) of the septic CST patients were diagnosed >7 days after symptom onset.
Septic
Adult
Expansion of the cavernous sinus was the most common neuroimaging feature, followed by convexity
Sinusitis of the lateral wall of the cavernous sinus (5) and filling defect of the cavernous sinus (4).
Diabetes mellitus Staphylococcal species (spp.) was the most commonly implicated pathogen, followed by Aspergillus
Onset-to-diagnosis spp. Despite treatment, 7% (1/14) of the patients died in the hospital and 67% (8/12) of the survivors
MRI had neurological deficits. The duration of onset-to-diagnosis and the presence of hemiparesis were
significant prognostic factors. These results provide a preliminary view of this uncommon infectious
syndrome. Further large-scale studies are needed to better delineate septic CST in adults.
Ó 2019 Elsevier Ltd. All rights reserved.

1. Introduction 2. Patients and methods

Despite advances in diagnostic methods and the development We retrospectively reviewed the clinical and laboratory data
of new antimicrobial agents, septic cavernous sinus thrombosis and initial MRI features of adult patients (>18 years of age) newly
(CST) remains a rare but serious infectious disease of the central diagnosed with septic CST admitted to Kaohsiung Chang Gung
nervous system (CNS) with high mortality and morbidity rates Memorial Hospital (CGMH) during a 10-year study period (January
[1,2]. Early diagnosis and early appropriate antimicrobial agent 2008–December 2017). Kaohsiung CGMH is a 2686-bed acute-care
treatment are the most important therapeutic strategies for septic teaching hospital providing both primary and tertiary care, and it is
CST [3], however the initial clinical manifestation is non-specific in the largest medical center in southern Taiwan. Over the 10-year
the early stage of the disease which may delay the diagnosis [2,4]. study period, 14 adult septic CST patients were identified using
Cranial magnetic resonance imaging (MRI) has been shown to be MRI-based studies, and their clinical and laboratory features and
beneficial in the diagnosis of septic CST [3,5], and may allow for therapeutic outcomes were enrolled for analysis. This study was
an early diagnosis and treatment. To date, only small-scale studies approved by the Ethics Committee of Kaohsiung Chang Gung
of adults with septic CST have been published in the literature Memorial Hospital (IRB No: 201801571B0).
[6–8], of which some did not report culture or MRI findings. In this study, the diagnosis of septic CST was based on: (1) typi-
Therefore, in this study, we analyzed the clinical characteristics, cal clinical features including fever, headache and cavernous sinus
implicated pathogens and therapeutic outcomes of 14 adults with syndrome including chemosis, ptosis, ophthalmoplegia, and tri-
culture-proven septic CST whose diagnosis was based on MRI geminal hypesthesia; (2) presence of direct and/or indirect signs
findings. in MRI; and (3) positive culture of pathogen(s) from surgical
wounds and/or blood and/or cerebrospinal fluid (CSF) [1,9,10].
⇑ Corresponding author at: Department of Neurology, Chang Gung Memorial Examples of MRI findings of septic CST from the 14 included
Hospital-Kaohsiung, 123, Ta Pei Road, Niao Sung, Kaohsiung, Taiwan. patients are shown in Figs. 1 and 2. The direct signs included filling
E-mail address: cwenneng@ms19.hinet.net (W.-N. Chang). defects within the cavernous sinus (Fig. 1A), convexity of the lateral

https://doi.org/10.1016/j.jocn.2019.07.022
0967-5868/Ó 2019 Elsevier Ltd. All rights reserved.
112 C.-W. Hsu et al. / Journal of Clinical Neuroscience 68 (2019) 111–116

Fig. 1. Direct signs of septic cavernous sinus thrombosis in magnetic resonance imaging. A: Gd-enhanced T1WI showing a filling defect within the right cavernous sinus
(arrow). B: Gdenhanced T1WI showing lateral wall convexity of the left cavernous sinus (arrow). C: Gadolinium (Gd)-enhanced T1-weighted image (T1WI) showing
expansion of the right cavernous sinus (arrow).

Fig. 2. Indirect signs of septic cavernous sinus thrombosis in magnetic resonance imaging. A & B: Gd-enhanced T1-weighted image (T1WI) showing dilation of the right
superior ophthalmic vein (arrow) on coronal (A) and transverse (B) views. C: T1WI showing exophthalmos of the right eye (star). D & E: T2-weighted images showing partial
thrombosis in the left internal jugular vein (IJV) (arrowhead in D) in comparison with a patent right IJV (star in E).

wall of the cavernous sinus (Fig. 1B) and expansion of the cavernous most common initial clinical presentations, found in 13 and 10
sinus (Fig. 1C). The indirect signs included superior ophthalmic vein patients, respectively. The other clinical presentations were fever
(SOV) dilation (Fig. 2A and B), exophthalmos (Fig. 2C) and thrombosis (9), ptosis (9), chemosis (8), eyelid swelling (5), trigeminal hypes-
of other venous sinuses or cerebral veins (Fig. 2D). thesia (5), hemiparesis (5), altered consciousness (4), exophthal-
In this study, we used the modified Rankin scale (mRS) to assess mos (4), eye pain (4), decreased visual activity (3), facial palsy
the outcomes at discharge, and divided the 14 patients into two (3), dysarthria/dysphagia (3), seizure (2), and Horner’s syndrome
subgroups for comparisons: those with a good outcome (mRS 5 2), (1). Except for Case 11, the other 13 patients had unilateral clinical
and those with a poor outcome (mRS = 3). The underlying condi- features as the initial presentation, and six (Cases 2, 6–9, 14) of
tions, symptoms and signs, MRI findings and complications were these 13 patients evolved to bilateral involvement during the hos-
compared between the two subgroups of septic CST patients. All pitalization course. Among the 14 patients, only five (Cases 3, 6,
statistical calculations were performed using the Statistical Pack- 11–13) came to our hospital for first aid management, and the
age for the Social Sciences (SPSS) version 22.0. Continuous vari- other nine were transferred from other hospitals to our hospital
ables were expressed as mean (SD), and categorical data were for further diagnosis and treatment. The interval from the clinical
presented as number. Continuous variables with and without nor- presentation to the diagnosis of septic CST (onset-to-diagnosis)
mal distribution were analyzed using the Student’s t-test and ranged from 1 to 61 days (mean 18.7 days).
Mann-Whitney test, respectively. Categorical data were analyzed The MRI findings of the 14 patients are listed in Table 2. The
using the chi-square test and Fisher’s exact test because of the most common findings were cavernous sinus expansion (10), fol-
small sample size of the cohort in this study. Statistical significance lowed by convexity of the cavernous sinus lateral wall (5), filling
was defined as a p-value < 0.05. defect within the cavernous sinus (4), SOV dilation (1), exophthal-
mos (1), and internal jugular vein thrombosis (1). Concomitant
sphenoid sinusitis, ethmoid sinusitis, maxillary sinusitis, and fron-
3. Results tal sinusitis were found in 13, 11, 10 and 3 of the patients, respec-
tively. Inflammation of the clivus (5), brain abscesses (3), internal
The 14 septic CST patients enrolled in this study included eight carotid artery stenosis and/or occlusion (3), cerebral infarction
men and six women, aged 37–77 years (mean 60.4 years), and (2), and internal jugular vein thrombosis (1) were the other neu-
their clinical data are listed in Table 1. Diabetes mellitus (DM) roimaging findings of these 14 patients.
was the most common underlying condition (in 10 patients), fol- Except for Case 11, the other 13 patients received surgical inter-
lowed by chronic renal insufficiency (2), congestive heart failure ventions for the concomitant sinusitis and/or brain abscesses, and
(1), chronic obstructive pulmonary disease (1), liver cirrhosis (1), the implicated pathogens were isolated from pus cultures and
alcoholism (1), heroin abuse (1), nasopharyngeal carcinoma (1), pathologic specimens in 12 (except Case 10) and 7 (Cases 1, 5, 6,
and rheumatoid arthritis with prednisolone and hydroxychloro- 7, 9, 10, 12) patients, respectively. CSF cultures were performed
quine treatment (1). Headache and ophthalmoplegia were the in nine patients (Cases 2, 3, 5, 6, 7, 9, 12, 13, 14), but none had a
 C.-W. Hsu et al. / Journal of Clinical Neuroscience 68 (2019) 111–116 113

Table 1
Demographic and clinical features of the 14 patients with septic cavernous sinus thrombosis.

No Age/Sex Underlying condition Onset-to- Clinical features Lesion

diagnosis side
1 56/F DM, sinusitis 20 days# H/A*, ophthalmoplegia*, chemosis, eyelid swelling, eye pain* L’t
2 77/M DM, CHF, sinusitis 48 days# H/A*, ophthalmoplegia, fever, ptosis, chemosis, hemiparesis*, altered consciousness, L’t ? Bil.
exophthalmos, facial palsy, dysarthria
3 55/F DM, CRI, sinusitis 4 days Ophthalmoplegia*, fever*, trigeminal hypesthesia*, eye pain*, facial palsy*, seizure* R’t
4 52/M DM, sinusitis 11 days# H/A*, ophthalmoplegia*, fever*, ptosis*, eyelid swelling* R’t
5 61/M CRI, alcoholism, sinusitis 7 days# H/A*, ophthalmoplegia, fever*, trigeminal hypesthesia*, Horner’s syndrome* L’t
6 61/M RA with prednisolone & 1 day H/A*, ophthalmoplegia*, fever*, ptosis, chemosis*, eyelid swelling*, trigeminal R’t ? Bil.
hydroxychloroquine treatment, hypesthesia*, hemiparesis, altered consciousness, exophthalmos*
sinusitis
7 59/M DM, COPD, sinusitis 44 days# H/A*, ophthalmoplegia, fever*, ptosis, chemosis, hemiparesis*, altered R’t ? Bil.
consciousness*, seizure*
8 81/F DM, sinusitis 14 days# H/A*, ophthalmoplegia*, ptosis*, exophthalmos, dysphagia R’t ? Bil.
9 77/F DM, sinusitis 61 days# H/A*, ophthalmoplegia*, ptosis*, chemosis*, hemiparesis*, altered consciousness*, R’t ? Bil.
eye pain*, decreased visual activity*
10 63/F DM, sinusitis 33 days# H/A*, ophthalmoplegia*, fever*, ptosis*, hemiparesis*, dysarthria* L’t
11 53/F NPC 5 days H/A*, ophthalmoplegia*, fever*, chemosis*, eyelid swelling* Bil.
12 37/M DM, liver cirrhosis, heroin abuse, 1 day H/A*, ophthalmoplegia*, fever*, ptosis*, chemosis*, trigeminal hypesthesia*, L’t
sinusitis exophthalmos*, eye pain*, decreased visual activity *, facial palsy*
13 55/M Sinusitis 6 days H/A*, ophthalmoplegia, chemosis, eyelid swelling, trigeminal paresthesia L’t
14 58/M DM, sinusitis 7 days# H/A*, ophthalmoplegia*, ptosis*, decreased visual activity* R’t ? Bil.

–Abbreviations: No = case number; F = female; M = male; DM = diabetes mellitus; CHF = congestive heart failure; CRI = chronic renal insufficiency; RA = rheumatoid arthritis;
COPD = chronic obstructive pulmonary disease; NPC = nasopharyngeal carcinoma; H/A = headache; L’t = left; R’t = right; Bil = bilateral.
#
suggests that the patient visited a local hospital or specialists other than a neurologist when the initial symptoms occurred.
*
means presenting symptoms at admission, and those without marks means symptoms that appeared later during hospitalization.

Table 2
Magnetic resonance imaging findings of the 14 patients with septic cavernous sinus thrombosis.

Case number 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Direct signs
Expansion of the cavernous sinus + + + + + + + + + +
Convexity of the lateral wall of the cavernous sinus + + + + +
Filling defect in the cavernous sinus + + + +
Indirect signs
Superior ophthalmic vein dilation +
Exophthalmos +
Internal jugular vein thrombosis +
Concomitant sphenoid sinusitis + + + + + + + + + + + + +
Ethmoid sinus involved + + + + + + + + + + +
Maxillary sinus involved + + + + + + + + + +
Frontal sinus involved + + +
CNS complications
Clivus inflammation + + + + +
Brain abscess + + +
Stenosis and/or occlusion of internal carotid artery + + +
Cerebral infarction + +

CNS: central nervous system.

positive result. Blood cultures were performed for all of the tum, one strain of Prevotella spp., one strain of Propionibacterium
patients except for Case 1, and implicated pathogens were identi- spp. and one strain of Veillonella dispar.
fied in five patients (Cases 4, 5, 7, 8, 11). Nine of the patients had Table 3 also shows the antimicrobial agents used for the treat-
polymicrobial infections, and the other five patients (Cases 2, 10, ment of these 14 patients and the therapeutic outcomes. The most
11, 13, 14), had monomicrobial infections. The implicated patho- commonly used empiric antimicrobial regimen was third- or
gens in these 14 patients are listed in Table 3, and they included fourth-generation cephalosporin ± vancomycin (11/14), and
nine strains of staphylococcal spp. (methicillin-resistant Staphylo- antibiotic combination was later adjusted according to the culture
coccus (S.) aureus (MRSA) (2), S. epidermidis (4), S. lugdunensis (1), results and susceptibility testing. Among the eight patients with
and coagulase-negative staphylococcus CoNS (2)), eight strains of fungal infections (Cases 1, 2, 5, 6, 7, 9, 10, 12), seven received addi-
fungi (Aspergillus spp. (6), unknown fungus (1) and mucormycosis tional anti-fungal agents. Case 2 did not receive antifungal treat-
(1)), four strains of streptococcal spp. (Streptococcus intermedius ment because the culture results were not available, and he died
(1), Streptococcus constellatus (1) and viridians streptococci (2)), soon due to a large cerebral infarction. As shown in Table 3, Cases
four strains of Pseudomonas aeruginosa, two strains of Citrobacter 4 and 8 were discharged after a full recovery, and the other 11
freundii, two strains of Acinetobacter baumannii, one strain of Kleb- patients were discharged with ocular or neurologic sequelae
siella pneumoniae, one strain of Enterobacter faecalis, one strain of including ophthalmoplegia (6), decreased visual activity (4), hemi-
Peptostreptococcus anaerobius, one strain of Fusobacterium nuclea- paresis (4), ptosis (3), trigeminal hypesthesia (3), altered state of
114 C.-W. Hsu et al. / Journal of Clinical Neuroscience 68 (2019) 111–116

Table 3
Pathogens, treatment and outcomes of the 14 patients with septic cavernous sinus thrombosis.

No Pathogens (sample) Antibiotic regimen Surgery mRS at discharge One-year follow-up sequelae (mRS)
1 MRSA, Aspergillus spp., Citrobacter freundii fluconazole (7 days) Yes 1 (0)
(sinus) ? micafungin (5 days)
? voriconazole (29 days)
2 Aspergillus spp. (sinus) ceftriaxone (2 days) Yes 6 –
? meropenem + vancomycin (3 days)
? ceftriaxone + vancomycin +
metronidazole (9 days)
? meropenem + vancomycin (13 days)
3 MRSA, Pseudomonas. aeruginosa, piperacillin-tazobactam (4 days) Yes 1 Ophthalmoplegia, trigeminal
Enterobacter faecalis, Veillonella dispar, ? meropenem (12 day paresthesia, facial palsy, hearing
Acinetobacter baumannii (sinus) ? ampicillin-sulbactam (19 days) impairment (1)
? flomoxef (2 days)
? meropenem (2 days)
? teicoplanin (25 days)
? amoxicillin-clavulanate (6 days)
4 Streptococcus constellatus, ceftriaxone (28 days) Yes 0 (0)
Peptostreptococcus anaerobius, CoNS
(sinus)
5 Pseudomonas aeruginosa, Aspergillus spp. ceftriaxone + amphotericin B (15 days) Yes 1 Ophthalmoplegia (1)
(sinus) ? voriconazole (3 days)
6 Pseudomonas aeruginosa, Aspergillus spp., ceftriaxone + vancomycin + metronidazole + Yes 2 Hemiparesis (2)
viridians streptococci, CoNS (sinus) voriconazole (5 days)
? cefepime + vancomycin +
voriconazole (31 days)
7 Staphylococcus epidermidis, Fusobacterium ceftriaxone + vancomycin (9 days) Yes 4 Hemiparesis, altered consciousness
nucleatum, Prevotella spp., unknown ? ceftazidime + fluconazole (16 days) (5)
fungus, Citrobacter freundii, Acinetobacter ? teicoplanin + fluconazole (7 days)
baumannii (sinus) ? cefepime + fluconazole (27 days)
? piperacillin-tazobactam +
voriconazole (7 days)
? tigecycline + voriconazole (18 days)
? piperacillin-tazobactam +
voriconazole (7 days)
? ceftriaxone + vancomycin +
voriconazole (6 days)
? meropenem (26 days)
8 Viridans streptococci, Staphylococcus ceftriaxone + Metronidazole (3 days) Yes 0 (0)
lugdunensis, Pseudomonas aeruginosa ? penicillin G (26 days)
(sinus)
9 Aspergillus fumigatus, Propionibacterium amoxicillin-clavulanate (10 days) Yes 4 Hemiparesis, decreased visual
spp. (sinus) ? liposomal amphotericin B (7 days) activity, altered consciousness (5)
? voriconazole (13 days)
10 Aspergillus spp. (brain abscess) cefepime + metronidazole + linezolid + Yes 4 Hemiparesis, dysphagia (5)
voriconazole (3 days)
? meropenem + voriconazole (21 days)
11 Streptococcus intermedius (blood) ceftriaxone (13 days) No 1 Death* (6)
12 Staphylococcus epidermidis, Klebsiella ceftriaxone + amphotericin B (17 days) Yes 1 Ophthalmoplegia, decreased visual
pneumoniae, Mucormycosis (sinus) ? amphotericin B (24 days) activity, trigeminal hypesthesia,
facial palsy (1)
13 Staphylococcus epidermidis (sinus) ceftriaxone + metronidazole (21 days) Yes 1 (0)
14 Staphylococcus epidermidis (sinus) ceftriaxone + vancomycin (22 days) Yes 3 Ophthalmoplegia, decreased visual
? meropenem + vancomycin (4 days) activity, hearing impairment (4)
? ceftriaxone (25 days)

Abbreviations: No = case number; MRSA = methicillin-resistant Staphylococcus aureus; spp. = species; CoNS = coagulase-negative Staphylococcus species; mRS = modified
Rankin scale.
*
Case 11 died from a deep neck infection and massive bleeding within 3 months after discharge.

consciousness (2), facial palsy (2), hearing impairment (2), exoph- 5, 12, and 14) still presented with variable degrees of
thalmos (1), and dysarthria (1). The mRS scores of these 14 patients ophthalmoplegia.
at discharge were 0 in two patients, 1 in six patients, 2 in one
patient, 3 in one patient, 4 in three patients, and 6 in one patient. 4. Discussion
The prognostic factors of the 14 patients are listed in Table 4, and
the comparative results showed that a longer onset-to-diagnosis The cavernous sinuses drain blood from the abundant venous
interval and the presence of hemiparesis were significant prognos- systems of the eye, nose, and brain into the pterygoid venous
tic factors. Case 11 died from a deep neck infection and massive plexus and internal jugular vein [1–3,11,12]; therefore, infections
bleeding within 3 months after discharge. At one year of follow- in these related structures may also contribute to the development
up in the other 12 survivors, the mRS scores were 0 in four of septic CST [1,12,13]. After the wide use of antibiotics, septic CST
patients, 1 in three patients, 2 in one patient, 4 in one patient, has become an uncommon although still serious infectious syn-
and 5 in three patients. The neurologic sequelae of Cases 3, 5–7, drome of the CNS, and septic CST remains a therapeutic challenge
9, 10, 12 and 14 are listed in Table 3, and four patients (Cases 3, [1,12,14–16].
 C.-W. Hsu et al. / Journal of Clinical Neuroscience 68 (2019) 111–116 115

Table 4 sentations of septic CST are all related to involvement of the corre-
The prognostic factors of the 14 adults with septic cavernous sinus thrombosis. sponding neuro-vascular structures [12–15,18]. In the current
Good Poor p study, apart from headache and fever, all of the other clinical fea-
outcome outcome tures (Table 1) were neuro-vascular related, of which eye signs
(N = 9) (N = 5) were the most common. Because bilateral cavernous sinuses can
Age at onset (years) 56.78 ± 11.5 66.8 ± 9.5 0.124 communicate with each other [12,14,19] and there is no valve
Sex 5 3 1.000 within the cavernous sinus so that blood can flow in either direc-
Underlying condition
Diabetes mellitus 5 5 0.221
tion depending on the pressure gradient [1,11,13,20], most
Chronic renal insufficiency 2 0 0.505 patients with septic CST eventually have bilateral cavernous sinus
Congestive heart failure 0 1 0.357 involvement, as found in 50% (7/14) of our adult septic CST
Chronic obstructive pulmonary 1 0 1.000 patients.
disease
The initial suspicion and diagnosis of septic CST are based on
Liver cirrhosis 1 0 1.000
Alcoholism 1 0 1.000 the clinical presentations [2,11,12,14,20,30], however the initial
Heroin abuse 1 0 1.000 symptoms and signs are usually nonspecific or subtle in the early
Nasopharyngeal carcinoma 1 0 1.000 stage [4,12] which may cause a delay in diagnosis. As shown in
Rheumatoid arthritis with 1 0 1.000 Table 1, 64% (9/14) of our patients were diagnosed >1 week after
prednisolone &
symptom onset, and all nine of these patients were transferred
hydroxychloroquine treatment
Onset-to-diagnosis (days) 7.67 ± 6.28 38.6 ± 20.31 0.007* from other hospitals. Among these nine transferred septic CST
Clinical manifestations patients, four had the diagnosis 1 month after symptom onset. This
Headache 8 5 1.000 delay in diagnosis has also been reported in other studies of septic
Fever 6 3 1.000
CST [4,13,15,21,22], with an onset-to-diagnosis duration ranging
Ptosis 4 5 0.086
Chemosis 5 3 1.000 from 2 weeks to 5 weeks. In the detection and diagnosis of septic
Eyelid swelling 5 0 0.086 CST, MRI is much better than cranial computed tomography (CT)
Trigeminal hypesthesia 5 0 0.086 studies [1,3,10,12,19], because MRI can detect all stages of throm-
Hemiparesis 1 4 0.023* bus formation in septic CST [4,23], especially structural changes in
Altered state of consciousness 1 3 0.095
the intra-cavernous internal carotid artery [12] as well as other
Exophthalmos 3 1 1.000
Eye pain 3 1 1.000 CNS complications such as cerebral infarction, brain abscess, and
Decreased visual activity 1 2 0.505 subdural empyema. [1,2,11,19,24,25]. In this study, the most com-
Facial palsy 2 1 1.000 mon MRI feature was cavernous sinus expansion (Table 2). In a
Dysarthria/dysphagia 1 2 0.505
previous study of neuroimaging findings of septic CST using cranial
Seizure 1 1 1.000
Horner’s syndrome 1 0 1.000
CT studies [7], filling defects were the most common finding. This
MRI findings difference in neuroimaging findings may be due to the different
Filling defect in the cavernous sinus 3 1 1.000 study tools used and deserves further large-scale studies.
Expansion of the cavernous sinus 6 4 1.000 In this study, 64.3% (9/14) of the patients had mixed infections,
Convexity of lateral wall of the 4 1 0.580
and the other 35.7% (5/14) had monomicrobial infections. In total,
cavernous sinus
Superior ophthalmic vein dilatation 1 0 1.000 20 implicated pathogens were isolated from clinical specimens,
Exophthalmos 0 1 0.357 including 92% (12/13) from pus cultures, 53% (7/13) from patho-
Internal jugular vein thrombosis 1 0 1.000 logic specimens, and 38% (5/13) from blood cultures. Staphylococ-
Complications
cal spp. was the most commonly identified pathogen, accounting
Brain abscess 1 2 0.505
Stenosis and/or occlusion of internal 2 1 1.000
for 64% (9/14) of the patients. The next most commonly implicated
carotid artery pathogen was fungi, especially Aspergillus, which was found in 57%
Cerebral infarction 0 2 0.110 (8/14) of the patients. The other implicated pathogens included
*
Indicates statistical significance (p < 0.05).
streptococcal spp., Gram-negative bacilli, and anaerobes. The high
incidence of staphylococcal spp. as the implicated pathogen in our
adult patients is consistent with other reported cases of septic CST
Of the 14 patients with septic CST in the current study, except of any age groups [2,11,14,15,26]. This complexity of implicated
for Case 13, all of the others had underlying conditions, of which pathogens in septic CST is similar to our previous studies of other
DM (10/14; 71.4%) and concomitant sinusitis (13/14; 93%) were CNS infections [27–30] and reveals the epidemiologic trend of
the most common (Table 2). Sphenoid sinusitis was the most com- adult bacterial meningitis, including changes in infectious state,
mon type of sinusitis (13/13; 100%), followed by ethmoid sinusitis implicated pathogens, emergence of uncommon pathogens and
(11/13; 86.6%), maxillary sinusitis (10/13; 76.9%) and frontal superinfection during the therapeutic course. Acinetobacter bau-
sinusitis (3/13; 23.1%). These finding of sinusitis, especially sphe- mannii, Citrobacter freundii, and Propionibacterium species as found
noid sinusitis, are consistent with the findings of other studies on in the present study have not been reported in other studies of sep-
septic CST [15,16] in which the presence of sinusitis, and especially tic CST in adults except when they were found to be the causative
sphenoid sinusitis, was an important preceding event of this pathogens of sinusitis-related complications [31,32].
uncommon infectious syndrome. In Taiwan, more than 10% of Because of the rarity of this infectious syndrome, there are no
adults have DM [17], and therefore the presence of DM and con- definite guidelines for the treatment of adult septic CST. Neverthe-
comitant sinusitis, and especially sphenoid sinusitis, are important less, early diagnosis and early use of appropriate empiric antimi-
underlying conditions for the development of septic CST. As shown crobial agents are essential therapeutic strategies for CNS
in Tables 1 and 2, the presence of other underlying conditions as infectious diseases [3]. The most commonly suggested empirical
well as sinusitis other than sphenoid sinusitis was also important antibiotics used for septic CST include broad-spectrum intravenous
for the development of septic CST. antibiotics with a good blood-brain-barrier penetration, of which
The cavernous sinuses are trabeculated structures close to the the empiric combination of third-generation cephalosporin and
sphenoid sinus and pituitary gland situated in the sella turcica high-dose vancomycin are usually suggested [11,14]. In the pre-
[1,3,10,12], and several important neuro-vascular structures are sent study, both Gram-positive and Gram-negative bacteria and
contained in the cavernous sinuses [18]. Therefore, the clinical pre- fungi were important implicated pathogens, and among the
116 C.-W. Hsu et al. / Journal of Clinical Neuroscience 68 (2019) 111–116

bacterial pathogens, anaerobic bacteria were also isolated from [2] Desa V, Green R. Cavernous sinus thrombosis: current therapy. J Oral Maxillof
Surg: Off J Am Assoc Oral Maxillof Surg 2012;70:2085–91.
clinical specimens. Therefore, we suggest that broad-spectrum
[3] Chen CM, Fan JY, Chen WL, et al. Lethal sphenoid fungal sinusitis in an
coverage for Gram-positive, Gram-negative, and anaerobic organ- immunocompromised elderly patient. B-ent 2009;5:173–7.
isms should include combinations of empiric antibiotics for adults [4] Ito E, Saito K, Nagatani T, et al. Cavernous sinus thrombophlebitis caused by
with suspected septic CST. If fungi are found, further adjustments porphyromonas gingivalis with abscess formation extending to the orbital
cavity. Case report. Neurologia medico-chirurgica 2009;49:370–3.
in the antimicrobial regimen may include antifungal agents. [5] Chick RS, Glisson JK, Pierce S. Bilateral cavernous sinus thrombosis following
There is currently no consensus on the duration of antimicrobial community-acquired methicillin-resistant staphylococcus aureus infection: a
agent therapy in the treatment of adults with septic CST [1,2,25], case report and review of the literature. J Miss State Med Assoc 2010;51:317–20.
[6] Bhatia K, Jones NS. Septic cavernous sinus thrombosis secondary to sinusitis:
however at least 3–4 weeks of treatment with parenteral antimi- are anticoagulants indicated? A review of the literature. J Laryngol Otol
crobial agents is reasonable [1–3,12,19,24]. If not contraindicated, 2002;116:667–76.
surgical interventions should be performed for the primary infec- [7] Chen HW, Su CP, Su DH, et al. Septic cavernous sinus thrombosis: an unusual
and fatal disease. J Formosan Med Assoc = Taiwan yi zhi 2006;105:203–9.
tious focus or for intracranial complications as soon as possible [8] Lize F, Verillaud B, Vironneau P, et al. Septic cavernous sinus thrombosis
[2,14], because this can decrease infectious burden and obtain cul- secondary to acute bacterial sinusitis: a retrospective study of seven cases. Am
tures to guide further modifications of the therapeutic strategy J Rhinol Allergy 2015;29:e7–12.
[9] DiNubile MJ. Septic thrombosis of the cavernous sinuses. Arch Neurol
[11]. In the current study, 13 of the 14 patients (except for Case 1988;45:567–72.
11) received surgical interventions, and the causative pathogens [10] Komatsu H, Matsumoto F, Kasai M, et al. Cavernous sinus thrombosis caused
were subsequently identified in 12 of these 13 patients (except by contralateral sphenoid sinusitis: a case report. Head Face Med 2013;9:9.
[11] Barahimi B, Murchison AP, Bilyk JR. Forget me not. Surv Ophthalmol
for Case 10) through these surgical procedures. Controversy exists
2010;55:467–80.
with regards to the use of anticoagulants for septic cavernous sinus [12] Kiddee W, Preechawai P, Hirunpat S. Bilateral septic cavernous sinus
thrombosis, and no prospective trials have been performed on the thrombosis following the masticator and parapharyngeal space infection
use of anticoagulants for septic CST [2,14,23,24,33]. Therefore, only from the odontogenic origin: a case report. J Med Assoc Thailand = Chotmaihet
thangphaet 2010;93:1107–11.
limited evidence suggests that anticoagulant drugs are probably [13] Pendharkar HS, Gupta AK, Bodhey N, et al. Diffusion restriction in thrombosed
safe and may be beneficial for people with septic CST [6,34]. In this superior ophthalmic veins: two cases of diverse etiology and literature review.
study, none of the 14 adults with septic CST received anticoagulant J Radiol Case Rep 2011;5:8–16.
[14] Songu M, Can N, Onal K, et al. Staphylococcus aureus cavernous sinus
therapy. Nevertheless, six patients (Cases 2, 6, 7, 8, 9, 14) had an thrombosis mimicking complicated fungal sinusitis. Ear Nose Throat J
evolving clinical course during the therapeutic course, and eight 2012;91:E26–30.
patients (Cases 3, 5, 6, 7, 9, 10, 12, 14) had thrombosis-related [15] Horowitz A, Spendel D, Kraut R, et al. Cavernous sinus thrombosis as a result of
a fungal infection: a case report. J Oral Maxillof Surg: Off J Am Assoc Oral
neuro-vascular sequelae. Therefore, further studies are warranted Maxillof Surg 2013;71:1899.e1891–e1895.
to investigate the need for anticoagulation therapy in patients with [16] van der Poel NA, Mourits MP, de Win MML, et al. Prognosis of septic cavernous
septic CST in order to prevent evolving thrombosis and reduce neu- sinus thrombosis remarkably improved: a case series of 12 patients and
literature review. Eur Arch Oto-Rhino-Laryngol 2018;275:2387–95.
rologic sequelae. [17] <https://www.idf.org/our-network/regions-members/western-pacific/
The mortality rate of septic CST was nearly 100% before the members/114-taiwan.html>. Accessed date: 2019/03/31.
introduction of antibiotics, however it has markedly decreased to [18] Le H, Prejean S, Heck M. Septic cerebral venosinus thrombosis secondary to an
odontogenic infection. J Louisiana State Med Soc: Off Organ Louisiana State
approximately 20–30% in the modern era. [1–4,11,15,19,21]. This
Med Soc 2017;169:33–6.
improvement may be due to supportive medical care and newer [19] Munckhof WJ, Krishnan A, Kruger P, et al. Cavernous sinus thrombosis and
imaging techniques, which help with the early identification and meningitis from community-acquired methicillin-resistant Staphylococcus
diagnosis of septic CST [2,25]. In the current study, the therapeutic aureus infection. Int Med J 2008;38:283–7.
[20] Sethi P, Jones ST, Valenzuela AA. Septic cavernous sinus thrombosis with
in-hospital mortality rate of the 14 patients was 7% (1/14). A diffuse spread leading to cerebral ischemia. Orbit (Amsterdam, Netherlands)
shorter onset-to-diagnosis duration was a significant prognostic 2013;32:330–2.
factor, and this finding supports the fact that favorable outcomes [21] Shum JW, Lee AW, Lustman K, et al. Cavernous sinus thrombosis secondary to
fracture of posterior maxillary sinus wall. Case report. The New York State
of patients with septic CST depend on an early diagnosis and early Dental J 2009;75:41–4.
appropriate antimicrobial treatment [1,4,5,21,24]. The other prog- [22] Shams PN, Policeni B, Carter KD, et al. Bilateral septic cavernous sinus
nostic factor found in this study was the presence of hemiparesis, thrombosis, congestive orbitopathy, and ischemic optic neuropathy. Canadian
J Ophthalmol. Journal canadien d’ophtalmologie 2016;51:e75–77.
which may indicate that the disease course was advanced. This [23] Lopez F, Santamarta E, Martinez P, et al. Cavernous sinus thrombosis during
study also showed that 67% (8/12) of the survivors had neurologic pregnancy. Auris Nasus Larynx 2017;44:232–6.
deficits evaluated within 1 year of discharge from the hospital. [24] Leo QJ, Bolger Jr DT. Septic cavernous sinus thrombosis due to Campylobacter
rectus infection. BMJ Case Re 2014;2014.
In conclusion, the diagnosis of septic CST in adults in this MRI- [25] Cheung EJ, Scurry WC, Isaacson JE, et al. Cavernous sinus thrombosis
based study was delayed in most of the cases. Most of the patients secondary to allergic fungal sinusitis. Rhinology 2009;47:105–8.
had underlying medical conditions, especially DM, and concomi- [26] Xia P, Jiao Y. Septic cavernous sinus thrombosis caused by tuberculosis
infection. BMJ Case Rep 2014;2014.
tant sinusitis, and especially sphenoid sinusitis, as preceding
[27] Chang WN, Lu CH, Huang CR, et al. Mixed infection in adult bacterial
events. Of the implicated pathogens, staphylococcal spp. was still meningitis. Infection 2000;28:8–12.
the most common, followed by fungi, and especially Aspergillus [28] Huang CR, Chen SF, Lu CH, et al. Clinical characteristics and therapeutic
spp. Despite treatment, the in-hospital mortality rate was about outcomes of nosocomial super-infection in adult bacterial meningitis. BMC
Infect Dis 2011;11:133.
7.9%, however about 67% of the survivors had neurologic deficits. [29] Lien CY, Huang CR, Tsai WC, et al. Epidemiologic trend of adult bacterial
Of the prognostic factors, onset-to-diagnosis and hemiparesis were meningitis in southern Taiwan (2006–2015). J Clin Neurosci 2017;42:59–65.
significant. [30] Lee JJ, Lien CY, Chien CC, et al. Anaerobic bacterial meningitis in adults. J Clin
Neurosci 2018;50:45–50.
[31] Lim JH, Sardesai MG, Ferreira Jr M, et al. Transorbital neuroendoscopic
Appendix A. Supplementary material management of sinogenic complications involving the frontal sinus, orbit, and
anterior cranial fossa. J Neurol Surg B Skull Base 2012;73:394–400.
[32] Chaiyasate S, Fooanant S, Navacharoen N, et al. The complications of sinusitis
Supplementary data to this article can be found online at in a tertiary care hospital: types, patient characteristics, and outcomes. Int J
https://doi.org/10.1016/j.jocn.2019.07.022. Otolaryngol 2015;2015:709302.
[33] Suntrup S, Kemmling A, Dziewas R, et al. Septic cavernous sinus thrombosis
complicated by occlusion of the internal carotid artery and multiple embolic
References strokes after surgery of an anorectal abscess: a clinical chameleon. The
Neurologist 2012;18:310–2.
[1] Ebright JR, Pace MT, Niazi AF. Septic thrombosis of the cavernous sinuses. Arch [34] Levine SR, Twyman RE, Gilman S. The role of anticoagulation in cavernous
Intern Med 2001;161:2671–6. sinus thrombosis. Neurology 1988;38:517–22.