Update in Radiology

Central nervous system infections: new diagnostic tools

N. Sgarbi
Clinical Department of Radiology, Hospital de Clínicas, Montevideo, Uruguay

Abstract
Central nervous system infections are an emerging health problem with poor prognosis if treatment is not adequate.
Thus, establishing a correct diagnosis is necessary to quickly start the appropriate treatment. This is a real challenge for
the radiologist, as it frequently requires a multidisciplinary approach.
The relatively low performance of diagnostic imaging is well known. While computed tomography (CT) is limited only
to an initial structural assessment, magnetic resonance imaging (MRI) is the method of choice, despite its low specificity.
However, this has been substantially modified in recent years with the introduction into daily practice of new magnetic
resonance sequences that allow precise structural and functional analysis, and provide essential additional information
for final diagnosis.
Now, due to the techniques of diffusion, perfusion, and spectroscopy, among others, a more detailed analysis can be
made in conjunction with clinical and laboratory studies that significantly improve the sensitivity and specificity of MRI
in this complex patient group.
The patterns of the most common infectious diseases of the nervous system are reviewed in this manuscript, highlight-
ing the contributions of functional sequences in these complex patients.
© 2014 Sociedad Argentina de Radiología. Published by Elsevier Spain, S.L.U. All rights reserved.

Keywords: Brain abscess; Encephalitis; Diffusion; MRI perfusion; Spectroscopy

Introduction fectious diseases. The incorporation of these techniques into

Central nervous system infections (CNSI) are an emerging
health problem given the sustained increase in the number
of cases in recent years. Even if this increase is determined by
the high number of patients with human immunodeficiency
virus (HIV) infection, there are two other important factors:
on the one hand, the significant migration flows worldwide
and, on the other hand, the rise in the number of infectious
agents that are resistant to standard therapy.
The factor that most affects prognosis for survival in patients
with CNSI is the identification of the causative agent. Such
identification allows a rapid institution of a specific therapeu-
tic strategy.
Among the tools available, magnetic resonance imaging
(MRI) has undoubtedly changed the paradigm in the study
of neurological diseases since its implementation. However,
despite its high sensitivity for a final diagnosis, this method
has some limitations at the time of establishing the agent
involved. In this respect, MRI diffusion, perfusion and spec-
troscopy techniques have demonstrated effectiveness in the
detection of various neurological conditions, particularly in-
routine evaluation of the central nervous system (CNS) has
substantially changed the overall performance of this meth-
od, and they have become essential tools.
The aim of this review it to highlight the role of new MRI
techniques in the evaluation of patients with CNSI, discuss-
ing their scope and limitations and illustrating their role by
presenting some cases from our site.
Significance: epidemiology
Nervous system involvement may occur due to infections
caused by multiple agents (such as bacteria, viruses, fungi
and parasites) and each of these agents has typical clinical
and imaging features.
Potentially involved pathogens vary based on the compart-
ment or area involved, geographic location and immune sta-
tus, vaccination and age of the patient, among other factors1.
In 2004, the World Health Organization (WHO) reported
about 350,000 deaths from meningoencephalitis (ME), with
an approximate incidence of 700,000 cases worldwide2.

12 Rev. Argent. Radiol. 2015;79(1): 12-31


Despite multiple diagnostic advances, in most patients the
causative agent of infection is still unknown3. In the case of
sporadic encephalitis, herpes simplex virus (HSV) is the most
common 1,3, while in brain abscesses the most common
causative microorganisms are Streptococcus and Staphylo-
coccus species, accounting for approximately 35% and 20%
of cases, respectively4. Cerebral involvement is very impor-
tant in most parasitoses, with cestodes (and within this group
the agent causing cysticercosis) being the most frequent5.
In our setting (Uruguay), it is difficult to know the exact fig-
ures of incidence and significance. Even if it is considered
a rare condition, the most frequent infectious disease until
2002, according to the Ministry of Public Health, was ME,
with a rate of 3.27 to 7 cases per 100,000 inhabitants in the
country and an approximate mortality rate of 10-30%, de-
pending on the agent involved and the patient’s age.
As regards brain abscesses (Bas), mortality has decreased
worldwide from 40 to 10%, with an increase in full recovery
from 33 to 70%4.
In Uruguay, parasitic infections are rare, mainly in immu-
nocompetent patients, unlike other neighboring countries,
where neurocysticercosis is relatively common.
In immunocompromised patients, mainly in HIV-infected
patients, parasites are the most common causative agents,
with toxoplasmosis being the most common infection in this
population, followed by JC virus infection1,5.
Pathology: the basis of imaging
findings
Predisposing factors or conditions are found in over 80%
of patients with BA, mainly contiguous or distant foci from
which dissemination originate4.
Radiology and pathology are inseparable at the time of under-
standing the phenomena causing nervous system infections.
They are the basis for a large number of classifications: some
based on the causative agent and others on topography of the
infection (intra/extra-axial). However, one of the most impor-
tant classifications divides infections into focal, multifocal or
diffuse depending on the type of involvement. Each of these
groups is associated with specific agents and processes.
The response of the nervous system is highly restricted de-
spite the interaction of multiple cell populations6. There is a
limited inflammation with blood-brain barrier (BBB) abnor-
malities and cerebrospinal fluid (CSF) involvement. Thus, a
predominantly vasogenic edema develops, though with vari-
able degrees of cytotoxicity and necrosis in different forms,
hemorrhagic sites and changes in BBB permeability6. All these
changes translate into typical imaging findings and constitute
the basis for an adequate interpretation.
N. Sgarbi
Infections and their alterations (and, therefore, findings) de-
pend on different factors. The leading factors include: the
agent involved, the patient’s immune status and the route
through which the infectious agent reaches the system.
Agents may be differentiated into pathogens proper (includ-
ing bacteria and most viruses) and opportunistic pathogens,
which are those requiring an immunosuppression status to
cause harm (e.g., fungi).
The routes of infection are diverse. The most common route
is hematogenous, but direct spread from contiguous sites can
also occur, e.g. in trauma, or there may be cases of retro-
grade perineural spread.
We should bear in mind the huge variety of associated patho-
logic processes, which are not directly caused by the infec-
tious agent but are triggered by this agent: for example,
vascular abnormalities (such as vasculitis or vein thrombosis),
circulatory disorders and reabsorption of the CSF, and direct
or autoimmune demyelinating phenomena.
Magnetic Resonance Imaging: study
protocol and advanced techniques
MRI is the method of choice in the evaluation of patients
with CNSI, but it is essential to assess images together with
the patient’s age, socioeconomic and immune status, as well
as much clinical data as available7. This technique helps to
establish the presence of an infection, to rule out differen-
tial diagnoses, analyze the mechanism of infection and pro-
vide guidance about the etiologic agent, in addition to being
helpful in follow-up and treatment monitoring. However, it
is also essential to have a CSF study available and to know
whether or not there is suspicion of an extra-neural focus of
infection, as this contributes to an early diagnosis. We should
bear in mind that an early diagnosis and initiation of a spe-
cific therapy are the most important factors having the great-
est impact on mortality from infectious diseases.
Any study protocol should include basic sequences in all 3
planes: T1- and T2-weighted sequences, CSF suppression se-
quences (FLAIR) and T1-weighted sequences after the admin-
istration of gadolinium-based paramagnetic contrast medium.
In recent years, usefulness of functional techniques has also
been confirmed. These techniques provide information about
the baseline process and pathophysiologic abnormalities
(previously reviewed); therefore the protocol should include
diffusion-weighted imaging (DWI) and an apparent diffusion
coefficient (ADC) map, both essential for the study of this
pathology. The use of perfusion-weighted imaging (PWI) and
spectroscopy, as well as of other modalities, is also relevant.
Rev. Argent. Radiol. 2015;79(1): 12-31 13
Central Nervous System Infections: New Diagnostic Tools.
Post-contrast FLAIR sequence
Post-contrast T2 FLAIR sequence is used with an inversion
pulse that nulls the signal intensity of CSF. This increases its
sensitivity to detect lesions and edema. Post-contrast FLAIR
imaging constitutes an essential part of the basic study pro-
tocol for brain conditions when potential infection is being
evaluated, mainly because of the ability of this sequence to
recognize the edema associated with most bacterial and viral
infections of the brain parenchyma.
In recent years, the importance of using this sequence with fat
saturation pulses and after contrast administration has been
highlighted: while the former increases contrast between
grey and white matter, with a consequent improvement in
resolution, post-contrast images have higher sensitivity for
detecting meningeal abnormalities, both fluid collections
(mainly subdural empyema) and leptomingeal involvement
seen in meningitis of various etiologies8-10 (fig. 1).
Some authors suggest that post-gadolinium FLAIR images
have identical sensitivity but a higher specificity compared to
T1-weighted images for the diagnosis of infectious leptomen-
ingitis10. Moreover, other authors, namely Splendiani et al11,
have shown the usefulness of MRI with post-contrast FLAIR
sequences in the early diagnosis of infectious meningitis.
Nevertheless, in spite of these studies and the fact that our
experience is very similar in a significant number of patients,
we should emphasize that conventional contrast-enhanced
T1-weighted images cannot be substituted in this population.
Magnetic susceptibility-weighted
imaging (SWI/SWAN)
Magnetic susceptibility-weighted imaging has a high sensi-
tivity for detecting the presence of hemoglobin breakdown
products. Gradient echo (GRE T2*) is the most commonly
used sequence, although the new susceptibility-weighted
images, such as SWI / SWAN (depending on the trademark),
have a greater ability to detect this artifact.
A multilayer appearance of the wall of nonspecific bacterial
brain abscesses has been reported, which is caused by an in-
tense susceptibility artifact in the part of the wall correspond-
ing to the capsule, coincident with the area of strongest en-
hancement12,13.
The “dual rim sign” of products from phagocytosis determin-
ing the presence of this artifact is very useful in the differen-
tial diagnosis of lesions with ring enhancement, mainly in the
case of high-grade tumors with necrosis, such as glioblas-
toma multiforme14 (fig. 2).
In addition, in some infections (for example herpetic en-
cephalitis) micro- hemorrhages may be noted, specially in
14 Rev. Argent. Radiol. 2015;79(1): 12-31
the cortical area, which can be easily visualized with these
sequences15,16 (fig. 3).
The presence of hemorrhage in some lesions has long been
considered rare, making diagnosis difficult. However, the use
of these sequences has demonstrated that the presence of
hemorrhage is much more common, as it occurs in cases of
toxoplasmosis in HIV patients.
The presence of hemoglobin breakdown products in BAs
due to toxoplasmosis is common and should not invalidate
the suspected diagnosis or delay the initiation of a specific
therapy17.
Diffusion (DWI)
The diffusion technique allows an analysis of the motion of
water molecules in brain tissue and provides unique informa-
tion in the study of infections18.
The presence of intense restricted diffusion has been reported
in the central part of nonspecific bacterial BAs, which has been
related to the poor cellularity and viscosity of pus18,19 (fig. 2).
The presence of restricted diffusion within a ring-enhancing le-
sion allows diagnosis of bacterial BA (with near 100% specific-
ity) and differentiates this entity mainly from necrotic or cystic
tumors18,20,21 (fig. 4). This finding has even been reported to
be pathognomonic for BA, in spite of the fact that it may be
found in other types of lesions, such as some metastases with
high viscosity or glioblastomas with hemorrhage22.
In the early stages of bacterial infection, cerebritis and ear-
ly capsule stage, diffusion restriction is not so marked, and
therefore there are difficulties at the time of making differen-
tial diagnosis15,18.
In practice, it is very complex to distinguish among the vari-
ous types of BAs using only conventional sequences. By us-
ing DWI in combination with a detailed assessment of the
morphology of lesions, clinical data and the patient’s immune
status, a more accurate diagnosis can be reached.
In fungal or tubercular brain abscesses diffusion restriction
is not as homogeneous as in bacterial brain abscesses, with
areas of restricted diffusion and areas of no restriction; there-
fore DWI is an essential tool in diagnostic assessment23.
Fungal BAs show restriction of diffusion in the wall with val-
ues similar to those of bacterial BAs, but also in the projec-
tions, which are usually directed centrally, with low signal on
DWI23,24.
The phenomenon of restricted diffusion related to the pres-
ence of pus occurs independently of the compartment where
pus is found. Therefore, this phenomenon may be seen both
in extradural or subdural suppurative collections (empyemas)
or within the ventricular system, as in ventriculitis25,26 (fig. 5).
Finally, a very interesting point is the use of DWI in the fol-
 N. Sgarbi

a b

c d

e f

Figure 1. Patient with documented tuberculous meningitis. The figure shows the utility and findings of FLAIR imaging after
contrast administration. (a, b and c). Axial slices show strong enhancement and occupation of the Sylvian cistern on the right
side by granulomatous material deposition (typical of this pathology). (d, e and f) Same patient: contrast-enhanced sagittal and
axial T1-weighted SE images show findings clearly identified in the previous sequence.

Rev. Argent. Radiol. 2015;79(1): 12-31 15

Central Nervous System Infections: New Diagnostic Tools.

a b

c d

Figure 2. Nonspecific bacterial brain abscess, confirmed after surgical drainage. Axial MRI selected views show the usefulness
of (a) magnetic susceptibility-weighted imaging, (b) contrast-enhanced T1-weighted SE, (c) diffusion-weighted imaging and
(d) ADC map. Magnetic susceptibility-weighted imaging shows the typical appearance of the bacterial capsule with a target-
shaped or dual wall pattern, due to the orderly deposition of hemoglobin breakdown products. The lesion shows a typical fine,
smooth, complete nodule-free ring enhancement with a necrotic center with intense restricted diffusion.

low-up of patients with BA treated with antibiotics and no
surgical drainage. This is one of the most important tools for
assessing success of the selected therapy, as there is a de-
crease in central signal intensity if antibiotic therapy has the
expected effect, and total disappearance if completely surgi-
cally drained27 (fig. 6).
In viral infections, DWI also plays an essential role16,18. Her-
petic encephalitis by herpes simplex virus type 1 is the most
common infection in this group, with a high mortality1,18.
The classical finding is cortical restricted diffusion with a gy-
ral pattern from the initial stages of disease (48 hours after
the onset of symptoms), which allows establishing diagno-
sis18,28,29 (fig. 3).
In all viral infections, it is common to find high signals on DWI,
with variable degrees of diffusion restriction, in different ana-
tomical regions depending on the causative agent. This allows
establishing diagnostic suspicion with adequate levels of sen-
sitivity18,30 (Fig. 7). DWI has even been reported to have high
sensitivity for detecting areas of viral activity in the early stages
of encephalitis in relation to the onset of symptoms. This is
shown by areas of cortical restricted diffusion not identified
on other sequences (clinical-radiological dissociation) (fig. 8).
Finally, the role of DWI in the diagnosis of progressive multi-
focal leukoencephalopathy (PML) in HIV/AIDS patients should
be highlighted. In cases of severe immunosuppression it is
essential to establish a diagnosis based on the study of C-

16 Rev. Argent. Radiol. 2015;79(1): 12-31

 N. Sgarbi

a b

c d

e f

Figure 3 Herpetic encephalitis (VHS 1) in a 2-year-old male patient with diagnosis confirmed by cerebrospinal fluid study. (a)
FLAIR imaging, (b) diffusion-weighted imaging, (c) T2-weighted FSE image, (d) GRE T2*, (e) T1-weighted SE image and (f)
contrast-enhanced imaging. Findings are typical of this entity: intense cortical restricted diffusion with a gyral pattern, similar
to the distribution of the edema caused by this entity and to the enhancement pattern. (d) The gradient echo image shows the
deep cortical hemorrhagic component.

Rev. Argent. Radiol. 2015;79(1): 12-31 17

Central Nervous System Infections: New Diagnostic Tools.

a b

c d

e f

Figure 4 Nonspecific bacterial brain abscesses. Patients with bacterial cerebellar BA (a and d), subcortical brain abscess (b and
e) and multiple bilateral superficial and deep BAs (e and f). The typical behavior of these lesions is illustrated by the classical
ring enhancement in (a, b and c) contrast-enhanced T1-weighted images and (d, e and f) the intense restricted diffusion in the
necrotic center in diffuse-weighted images, typical of these lesions.

18 Rev. Argent. Radiol. 2015;79(1): 12-31

 N. Sgarbi

a b

c d

e f

Figure 5. Bacterial infections with involvement of different compartments. This figure illustrates the usefulness of diffusion-weighted
imaging for the assessment of bacterial involvement of the various compartments affected. (a) contrast-enhanced T1-weighted SE image
and (d) diffusion-weighted image show an open BA partially ruptured into the lateral ventricle with pus (which accounts for restricted
diffusion within) (b and e) Extradural empyema in right anterior frontal region, secondary to infection in the sinonasal cavity, with (b) clas-
sical enhancement on contrast-enhanced T1-weighted image and (e) restricted diffusion. Finally, we present a subdural empyema with
the same behavior as described above: (c) contrast-enhanced T1-weighted SE image and (f) diffusion-weighted image.

Rev. Argent. Radiol. 2015;79(1): 12-31 19

Central Nervous System Infections: New Diagnostic Tools.

a b

c d

e f

Figure 6. Bacterial brain abscess treated with antibiotics: follow-up. Patient treated with triple antibiotic (ATB) regimen because
of the absence of neurological focus and the size of the abscess. The diagnostic scan shows the typical appearance of the le-
sion: (a) edema on FLAIR imaging, (b) ring enhancement on contrast-enhanced T1-weighted imaging and (c) intense restricted
diffusion. The follow-up MRI performed 15 days later shows (d) reduced size of the lesion and edema, (e) an area of persistent
enhancement and (f) total absence of restricted diffusion, which confirms an excellent response to treatment.

20 Rev. Argent. Radiol. 2015;79(1): 12-31

 N. Sgarbi

a b

c d

e f

Figure 7. Arbovirus encephalitis: multimode MRI. (a) T2-weighted and (b) FLAIR images show edema on the left hemisphere,
consistent with intense restricted diffusion with a gyral pattern and (c) cortical edema on diffusion-weighted image (which
is common in viral infections). (d) Contrast-enhanced T1-weighted imaging shows cortical enhancement while (e) perfusion-
weighted imaging shows that the pathologic area has, despite enhancement, a decrease in regional cerebral blood volume,
delimiting the affected area as an area of hypoperfusion. (f) Spectroscopy shows the pseudotumoral metabolic pattern, with a
decrease in N-acetylaspartate and creatinine, an increase in choline levels and a small lactate peak (arrow).

Rev. Argent. Radiol. 2015;79(1): 12-31 21

Central Nervous System Infections: New Diagnostic Tools.

a b

c d

e f

Figure 8 Encephalitis confirmed by cerebrospinal fluid study. This figure illustrates the special usefulness of diffusion-weighted
imaging in the early diagnosis of viral infections. (a, b and c) FLAIR image interpreted as normal 3 days after the onset of symp-
toms. (d, e and f) Diffusion-weighted image from the same scan shows extensive areas of cortical restricted diffusion at the level
of the right hemisphere, with associated insular involvement, and no significant edema or mass effect.

22 Rev. Argent. Radiol. 2015;79(1): 12-31

 N. Sgarbi

a b

c d

e f

Figure 9 Multimodal MRI in an HIV-positive patient with progressive multifocal leukoencephalopathy Extensive pathological area on the
left anterior frontal region, with mass effect and edema on (a) FLAIR imaging and (b) classical pattern on diffusion-weighted imaging
with high signal halo and hypointense center. (c) No enhancement on contrast-enhanced imaging. (d) Spectroscopy shows a pseudo-
tumoral pattern represented by decreased N-acetylaspartate and increased choline levels, and a very significant presence of lactates
(arrow) visible on (e) the metabolic map as an inverted double peak at 1.3 and (f) which shows low cerebral blood volume on perfusion
weighted-imaging. The presence of the JC virus, the causative agent of this condition, was confirmed by the cerebrospinal fluid study.

Rev. Argent. Radiol. 2015;79(1): 12-31 23

Central Nervous System Infections: New Diagnostic Tools.
reactive protein (CRP) in CSF and imaging findings, with the
purpose of changing immediate prognosis31.
PML is a subacute demyelinating disease caused by reactiva-
tion of the JC virus. Because of the progressive nature and
poor prognosis of this disease, treatment and evaluation of
response are essential31,32. Recently, multiple studies have
used DWI to detect areas of activity in patients with PML and
monitor their response to treatment18,33. In these cases, le-
sions on DWI are characterized by a low signal in the center
of lesions and a high signal in the periphery, which is indica-
tive of activity. This is the area that is susceptible of response
to antiretroviral therapy34 (fig. 9).
Moreover, the ADC map has also been reported to be helpful
in differentiating between rapid responders and non-respond-
ers, with poor progress and the possibility of developing an
immune reconstitution inflammatory syndrome (IRIS)35.
Perfusion
PWI techniques allow a functional and hemodynamic as-
sessment of cerebral circulation, including capillary bed sta-
tus, indemnity or not of the BBB and presence or absence
of angiogenesis 36. Through a dynamic injection of contrast
medium, different functional maps can be generated (e.g.,
cerebral blood flow [CBF]) or cerebral blood volume [CBV]).
Even if the role of PWI has been well established for the
evaluation of stroke or nervous system tumors, it is a tool
of special interest in patients with infection 37. Because of
the constitution of the brain abscess capsule, rich in collagen
and with decreased vascularity, it is easy to distinguish these
lesions from tumors that may have a similar appearance on
conventional imaging 38,39. Thus, on PWI CBV values are
low, with no alterations of permeability in the area of maxi-
mum enhancement (i.e., in the capsule of the lesion)39 (fig.
10). For CBV values found in the necrotic center of tumors
and in BAs there are no significant differences; therefore, it is
necessary to make a correct assessment of the various areas
of the lesion40.
PWI is an effective tool for differentiating between brain
abscesses and necrotic tumors, which --as previously men-
tioned-- may show restricted diffusion in the center of the
lesion. Chan et al have shown that relative CBV values (rCBV)
on the walls of various types of cystic or necrotic tumors are
higher than those of normal white matter, relative to those
of the BA wall41.
The perfusion technique may be used in all CNSIs, since as a
general rule, these are lesions with low CBV values, i.e., they
demonstrate hypoperfusion (figs. 7 and 9).
Different authors have confirmed that PWI is very useful in
patients with HIV/AIDS to differentiate between toxoplasmo-
24 Rev. Argent. Radiol. 2015;79(1): 12-31
sis and primary cerebral lymphoma, as the latter shows high
rCBV values as compared to the decrease observed in infec-
tious lesions42 (fig. 11).
Spectroscopy
Spectroscopy is a technique that allows noninvasive assess-
ment of the metabolic behavior of normal and pathologic tis-
sues. Even if its interpretation may be complex, it has a good
diagnostic performance in a wide range of CNS disorders43.
The major contribution of spectroscopy has been in tumors,
but the usefulness of this modality in infectious lesions is un-
deniable44.
For ring-enhancing lesions, spectroscopy provides very im-
portant data to differentiate between infectious lesions and
other entities, and it even provides information about the in-
fective agent probably involved44-46.
The presence of succinate, acetate, alanine, leucine and va-
line in some cases is considered specific for pyogenic BA,
even in the absence of restricted diffusion on DWI44,47. Some
indexes, such as the acetate / succinate ratio even allow dis-
tinguishing between bacterial and parasitic lesions48 (fig. 12).
If amino acids are present, a BA can be differentiated from a
tumor with adequate sensitivity and specificity levels48.
In the case of viral encephalitis, there is no clear evidence
about the overall performance of this method.
In patients with acute HSV, in the initial stages there is a de-
crease in the levels of N-acteylaspartate (NAA) and an increase
in choline (Cho) with a decrease in the NAA/Cr (creatine) ratio
and variable peaks of lactate, usually high49. However, this
pattern is not very specific and, as it can be seen in multiple
viral agents, it may be interpreted as tumoral (pseudo-tumor-
al pattern), compromising the specificity of the method (figs.
7, 9 and 13).
Some authors have suggested the usefulness of spectroscopy
for the evaluation of infectious lesions in cases of HIV/AIDS,
but its interpretation is complex and it has some difficul-
ties50 (fig. 14). In HIV patients, CNS tuberculosis is one of the
emerging pathologies and, among the forms of involvement
of this condition, meningitis is the most common, while tu-
berculoma constitutes a real diagnostic challenge.
Tuberculomas are lesions with a caseating center, which
results in the typical low signal on T2-weighted imaging51.
These lesions show no restricted diffusion and have low
CBV, as it occurs in previously discussed cases with specific
agents50,51. Metabolic spectra usually show a very significant
lipid/lactates peak. Even if this is a nonspecific finding, when
assessed in conjunction with the information available, it al-
lows suspicion of the final diagnosis51 (fig. 15).
An essential concept that we should bear in mind is that met-
 N. Sgarbi

a b

c d

e f

Figure 10 Bacterial brain abscesses: appearance on perfusion-weighted imaging. MRI in selected sequences, in the 2 patients
with bacterial BA confirmed after surgical drainage shows (a and d) the classical restricted diffusion and (b and e) an ADC map
with (c and fe) low cerebral blood volumes on perfusion weighted images.

Rev. Argent. Radiol. 2015;79(1): 12-31 25

Central Nervous System Infections: New Diagnostic Tools.

a b

c d

e f

Figure 11 Multimodal MRI in HIV-positive patient with cerebral toxoplasmosis. A large-volume lesion in the right basal ganglia
with necrosis, edema and mass effect is seen on (a) T1-weighted images and (C) FLAIR images. Central areas of hemorrhage
on (b) T2-weighted images and (d) a strong heterogeneous enhancement. (e) No restricted diffusion in the center of the lesion,
which differentiates it from bacterial brain abscess and lymphoma, with projections directed inwardly in the form of septa with
high signal (common in atypical agents). (f) Perfusion-weighted imaging typically shows decreased cerebral blood volume (area
of hypoperfusion).

26 Rev. Argent. Radiol. 2015;79(1): 12-31

 N. Sgarbi

a b

c d

Figure 12 Spectroscopy in bacterial brain abscesses: metabolic spectra. Two cases of BA: (a and b) right temporal BA secondary
to otitis media, with poor progress, and (c and d) left temporal BA following high-grade tumor surgery. The first case shows
the typical metabolic spectrum in these lesions: amino acids (AA), acetate (Ac), succinate (Suc) and lactates (arrow). The second
case, of significant diagnostic difficulty, shows (c) the restricted diffusion phenomenon and (d) the most common pattern with
high levels of lipids/lactates in the central area of the lesion.

abolic spectra should be interpreted in conjunction with the
whole set of images and sequences obtained, in order to al-
low adequate diagnosis and potential differential diagnoses.
Recommendations
and final concepts
CNS infections constitute a real health problem, with high
mortality rates in cases of late diagnosis.
The interpretation of images is a challenge, but among diag-
nostic methods, MRI is the method of choice because it has
demonstrated a positive impact in this group of patients: not
only does it help to determine an approximate diagnosis, to
rule out differential diagnoses and to analyze both the agent
most likely involved as well as the etiopathogenic mechanism,
but it also allows a strict follow-up of the selected therapeutic
management.
For an improved performance of this method, images should
be interpreted in conjunction with clinical data, the course
of disease and the patient’s immune status, as well as any
relevant epidemiological information (known contact with
agents, recent trips, etc.) and laboratory tests, mainly blood
and CSF tests.
An early diagnosis reduces mortality, as it allows the rapid in-
stitution of specific treatment. To achieve this, it is imperative

Rev. Argent. Radiol. 2015;79(1): 12-31 27

Central Nervous System Infections: New Diagnostic Tools.

a b

c d

e f

Figure 13 Spectroscopy in a patient with documented arbovirus encephalitis and extensive left frontal pathological area, (b)
with edema not significantly reflected in diffusion, and (f) areas of heterogeneous enhancement. (d and e) Metabolic spectrum
shows a pseudo-tumoral pattern with very high levels of lactates (arrow). (a) FLAIR imaging and (c) non-contrast enhanced T1-
weighted SE imaging complement the study.

28 Rev. Argent. Radiol. 2015;79(1): 12-31


a b
c d
to perform an adequate MRI, which should include conven-
tional, FLAIR, T2- and T1-weigthed imaging, and contrast-
enhanced images. The use of new imaging modalities, main-
ly diffusion- and perfusion-weighted imaging, is essential.
Thus, a final diagnosis can be established with near 100%
sensitivity and specificity.
Conflict of interests
The author declares a possible conflict of interest as a mem-
ber of the International Advisory Board of Revista Argentina
de Radiología.
References
1. Riddell J 4th, Shuman EK. Epidemiology of central nervous system infec-
tion. Neuroimag Clin N Am. 2012;22:543---56.
2. Mathers C, Boerma T, Ma Fat D. The global burden of disease: 2004 up-
date. Switzerland: World Health Organization; 2008.
N. Sgarbi
Figure 14 Perfusion and spectroscopy in HIV-positive patients
with cerebral toxoplasmosis. (a) Right frontal focal lesion with
a ring- and target-shaped enhancement pattern, with an ec-
centric nodule on contrast-enhanced T1-weighted imaging.
Functional imaging shows the following typical findings: (b)
decreased cerebral blood volume and (c) spectrum of me-
tabolites with normal concentrations of N-acetylaspartate
and creatine, increased levels of choline and a very significant
peak of lactates (arrow). This pattern is clearly different from
that of lymphoma, the most common nervous system tumor
in this group of patients.
3. Solomon T, French N. Neurological infections: something for everyone.
Lancet Neurol. 2013;12:20---2.
4. Brouwer MC, Coutinho JM, van de Beek D. Clinical characteristics and
outcome of brain abscess: systematic review and meta-analysis. Neurol-
ogy. 2014;82:806---13.
5. Finsterer J, Auer H. Parasitoses of the human central nervous system. J
Helminthol. 2013;87:257---70.
6. McKeever PE. Pathologic basis of central nervous system infections. Neu-
roimag Clin N Am. 2012;22:773---90.
7. Marchbank ND, Howlett DC, Sallomi DF, Hughes DV. Magnetic resonance
imaging is preferred in diagnosing suspected cerebral infections. BMJ.
2000;320:187---8.
8. Kremer S, Abu Eid M, Bierry G, Bogorin A, Koob M, Dietemann JL, et
al. Accuracy of delayed post-contrast FLAIR MR imaging for the diag-
nosis of leptomeningeal infectious or tumoral diseases. J Neuroradiol.
2006;33:285---91.
9. Kamran S, Bener AB, Alper D, Bakshi R. Role of fluid-attenuated inver-
sion recovery in the diagnosis of meningitis: comparison with contrast-
enhanced magnetic resonance imaging. J Comput Assist Tomogr.
2004;28:68---72.
10. Parmar H, Sitoh YY, Anand P, Chua V, Hui F. Contrast-enhanced flair imag-
ing in the evaluation of infectious leptomeningeal diseases. Eur J Radiol.
2006;58:89---95.
11. Splendiani A, Puglielli E, De Amicis R, Necozione S, Masciocchi C, Gallucci
M. Contrast-enhanced FLAIR in the early diagnosis of infectious meningi-
tis. Neuroradiology. 2005;47:591---8.
Rev. Argent. Radiol. 2015;79(1): 12-31 29
Central Nervous System Infections: New Diagnostic Tools.

a b

c d

e f

Figure 15. Multimodal MRI in an HIV-positive patient with brain tuberculoma diagnosed by pathological examination of the sur-
gical specimen. (a) Left superficial parietal focal lesion with associated edema on FLAIR imaging, (b) low signal on T2-weighted
imaging (typical of these lesions) and (c) ring enhancement on contrast-enhanced T1-weighted imaging. (d) On diffusion-
weighted imaging, no restriction in the center of the lesion. (d) perfusion-weighted imaging and (f) spectroscopy show the
typical characteristics of infectious lesions, with an elevated lipids/lactates peak (arrow).

30 Rev. Argent. Radiol. 2015;79(1): 12-31


12. Chang HC, Chuang TC, Chung HW, Lin HS, Lai PH,Weng MJ, et al. Multi-
layer appearance on contrast-enhanced susceptibilityweigthed images on
patients with brain abscesses: possible origins and effects of postprocess-
ing. J Magn Reson Imaging. 2012;36:1353---61.
13. Lai PH, Chang HC, Chuang TC, Chung HW, Li JY, Weng MJ, et al. Sus-
ceptibility-weighted imaging in patients with pyogenic brain abscesses
at 1.5T: characteristics of the abscess capsule. AJNR Am J Neuroradiol.
2012;33:910---4.
14. Toh CH, Wei KC, Chang CN, Hsu PW, Wong HF, Ng SH, et al. Differentiation
of pyogenic brain abscesses from necrotic glioblastomas with use of suscepti-
bility-weighted imaging. AJNR Am J Neuroradiol. 2012;33:1534---8.
15. Rath TJ, Hughes M, Arabi M, Shah GV. Imaging of cerebritis, encephalitis
and brain abscess. Neuroimag Clin N Am. 2012;22:585---607.
16. Saini J, Gupta RK, Jain KK. Intracranial infections: key neuroimaging find-
ings. SeminRoentgenol. 2014;49:86---98.
17. Bhagavati S, Choi J. Frequent hemorrhagic lesions in cerebral toxoplasmo-
sis in AIDS patients. J Neuroimaging. 2009;19:169---73.
18. Gasparetto EL, Cabral RF, da Cruz LC, DominguesF R.C. Jr. Diffusion imag-
ing in brain infections. Neuroimag Clin N Am. 2011;21:89---113.
19. Mishra AM, Gupta RK, Saksena S, Prasad KN, Pandey CM, Rathore D, et
al. Biological correlates of diffusivity in brain abscess. Magn Reson Med.
2005;54:878---85.
20. Desprechins B, Stadnik T, Koerts G, Shabana W, Breucq C, Osteaux M. Use
of diffusion-weighted MR imaging in differential diagnosis between intra-
cerebral necrotic tumors and cerebral abscesses. AJNR Am J Neuroradiol.
1999;20: 1252---7.
21. Mishra AM, Gupta RK, Jaggi RS, Reddy JS, Jha DK, Husain N, et al. Rol of
diffusion-weighted imaging and in vivo proton magnetic resonance spec-
troscopy in the differential diagnosis of ring-enhancing intracranial cystic
mass lesions. J Comput Assist Tomogr. 2004;28:540---7.
22. Hartmann M, Jansen O, Heiland S, Sommer C, Münkel K, Sartor K. Re-
stricted diffusion within ring enhancement is not pathognomonic for
brain abscess. AJNR Am J Neuroradiol. 2001;22:1738---42.
23. Luthra G, Parihar A, Nath K, Jaiswal S, Prasad KN, Husain N, et al. Com-
parative evaluation of fungal, tubercular and pyogenic brain abscesses
with conventional and diffusion MR imaging and proton MR spectroscopy.
AJNR Am J Neuroradiol. 2007;28:1332---8.
24. Mathur M, Johnson CE, Sze G. Fungal infections of the central nervous
system. NeuroimagClin N Am. 2012;22:609---32.
25. Mohan S, Jain KK, Arabi M, Shah GV. Imaging of meningitis and ventricu-
litis. Neuroimag Clin N Am. 2012;22:557---83.
26. Pezzullo JA, Tung GA, Mudigonda S, Rogg JM. Diffusion-weighted MR
imaging of pyogenic ventriculitis. AJR Am J Roentgenol. 2003;180:71---5.
27. Cartes-Zumelzu FW, Stavrou I, Castillo M, Eisenhuber E, Knosp E, Thurn-
her MM. Diffusion-weighted imaging in the assessment of brain abscesses
therapy. AJNR Am J Neuroradiol. 2004;25:1310---7.
28. Kuker W, Nagele T, Schmidt F, Heckl S, Herrlinger U. Diffusionweighted
MRI in herpes simplex encephalitis: a report of three cases. Neuroradiol-
ogy. 2004;46:122---5.
29. Tien RD, Felsberg GJ, Osumi AK. Herpes virus infections of the CNS: MR
findings. AJR Am J Roentgenol. 1993;161:167---76.
30. Handique SK. Viral infections of the central nervous system. Neuroimag
Clin N Am. 2011;21:777---94.
31. Sahraian MA, Radue EW, Eshaghi A, Besliu S, Minagar A. Progressive mul-
tifocal leukoencephalopathy: a review of the neuroimaging features and
differential diagnosis. Eur J Neurol. 2012;19:1060---9.
32. Barata Tavares J, Geraldo AF, Neto L, Reimão S, Campos JG. Diffusion-
N. Sgarbi
weighted MR imaging characterization of progressive multifocal leucoen-
cephalopathy (PML) lesions. Acta Med Port. 2012;25Suppl1:34---7.
33. Mader I, Herrlinger U, Klose U, Schmidt F, Küker W. Progressive multifo-
cal leukoencephalopathy: analysis of lesion development with diffusion-
weighted MRI. Neuroradiology. 2003;45:717---21.
34. Shah R, Bag AK, Chapman PR, Curé JK. Imaging manifestations of pro-
gressive multifocal leukoencephalopathy. Clin Radiol. 2010;65:431---9.
35. Buckle C, Castillo M. Use of diffusion-weighted imaging to evaluate the
initial response of progressive multifocal leukoencephalopathy to highly
active antiretroviral therapy: early experience. AJNR Am J Neuroradiol.
2010;31:1031---5.
36. Cha S. Perfusion MR imaging: basic principles and clinical applications.
Magn Reson Imaging Clin N Am. 2003;11:403---13.
37. Aksoy FG, Lev MH. Dynamic contrast-enhanced brain perfusion im-
aging: technique and clinical applications. Semin Ultrasound CT MR.
2000;21:462---77.
38. Nguyen JB, Black BR, Leimkuehler MM, Halder V, Nguyen JV, Ahktar N. In-
tracranial pyogenic abscess: imaging diagnosis utilizing recent advances in
computed tomography and magnetic resonance imaging. Crit Rev Com-
put Tomogr. 2004;45:181---224.
39. Grand S, Lefournier V, Krainik A, Bessou P, Tropres I, Chabardes S, et
al. Imagerie de perfusion: principes et applications cliniques. J Radiol.
2007;88:444---71.
40. Chiang IC, HsieH TJ, Chiu ML, Liu GC, Kuo YT, Lin WC. Distinction between
pyogenic brain abscess and necrotic brain tumor using 3-tesla MR spectros-
copy, diffusion and perfusion imaging. Br J Radiol. 2009;82:813---20.
41. Chan JH, Tsui EY, Chau LF, Chow KY, Chan MS, Yuen MK, et al. Discrimi-
nation of an infected brain tumor from a cerebral abscess by combined
MR perfusion and diffusion imaging. Comput Med Imag Graph. 2002;26:
19---23.
42. Ernst TM, Chang L, Witt MD, Aronow HA, Cornford ME, Walot I, et al.
Cerebral toxoplasmosis and lymphoma in AIDS: perfusion MR imaging
experience in 13 patients. Radiology. 1998;208:663---9.
43. Tran T, Ross B, Lin A. Magnetic resonance spectroscopy in neurological
diagnosis. Neurol Clin. 2009;27:21---60.
44. Oz G, Alger JR, Barker PB, Bartha R, Bizzi A, Boesch C, et al. Clinical
proton MR spectroscopy in central nervous system disorders. Radiology.
2014;270:658---79.
45. Gupta RK. Magnetic resonance spectroscopy in intracranial infection. En:
Gillard JH, Waldman AD, Barker PB, editores. ClinicalMR neuroimaging.
2nd ed. London, England: Cambridge University Press; 2010. p. 426---54.
46. Chang KH, Song IC, Kim SH, Han MH, Kim HD, Seong SO, et al. In vivo
single-voxel proton MR spectroscopy in intracranial cystic masses. AJNR
Am J Neuroradiol. 1998;19:401---5.
47. Pal D, Bhattacharyya A, Husain M, et al. In vivo proton MR spectroscopy
evaluation of pyogenic brain abscesses: a report of 194 cases. Am J Neu-
roradiol AJNR. 2010;31:360---6.
48. Gupta RK, Jobanputra KJ, Yadav A. MR spectroscopy in brain infections.
Neuroimag Clin N Am. 2013;23:475---98.
49. Gupta RK, Lufkin RB. Viral infections. En: Gupta RK, Lufkin RB, editores.
MR imaging and spectroscopy of central nervous system infection. New
York: Kluver Academic/Plenum Publishers; 2001. p. 147---75.
50. Simone IL, Federico F, Tortorella C, et al. Localised 1H-MR spectroscopy
for metabolic characterisation of diffuse and focal brain lesions in patients
infected with HIV. J NeurolNeurosurg Psychiatry. 1998;64:516---23.
51. Gupta RK, Kumar S. Central nervous system tuberculosis. Neuroimag Clin
N Am. 2011;21:795---814.
Rev. Argent. Radiol. 2015;79(1): 12-31 31
Central Nervous System Infections: New Diagnostic Tools.

32 Rev. Argent. Radiol. 2015;79(1): 12-31