Yale University
Yale University
EliScholar – A Digital Platform for Scholarly Publishing at Yale
EliScholar – A Digital Platform for Scholarly Publishing at Yale
Yale Medicine Thesis Digital Library
School of Medicine
January 2020
Do Associated Mri Findinds Improve The Detection Of Elusive
Do Associated Mri Findinds Improve The Detection Of Elusive
Encephaloceles?
Encephaloceles?
Lovemore Makusha
Follow this and additional works at: https://elischolar.library.yale.edu/ymtdl
Recommended Citation
Recommended Citation
Makusha, Lovemore, “Do Associated Mri Findinds Improve The Detection Of Elusive Encephaloceles?”
(2020). Yale Medicine Thesis Digital Library. 3930.
https://elischolar.library.yale.edu/ymtdl/3930
This Open Access Thesis is brought to you for free and open access by the School of Medicine at EliScholar – A
Digital Platform for Scholarly Publishing at Yale. It has been accepted for inclusion in Yale Medicine Thesis Digital
Library by an authorized administrator of EliScholar – A Digital Platform for Scholarly Publishing at Yale. For more
information, please contact elischolar@yale.edu.
DO ASSOCIATED MRI FINDINDS IMPROVE THE DETECTION OF ELUSIVE
ENCEPHALOCELES?
by
Lovemore Makusha
A thesis submitted to the faculty of
Yale University School of Medicine
in partial fulfillment of the requirements for the degree of
Doctor in Medicine
Department of Radiology and Biomedical Imaging
Yale University
February 2020
Copyright © Student’s Lovemore Makusha 2020
All Rights Reserved
Yale University School of Medicine
STATEMENT OF THESIS APPROVAL
The thesis of
Lovemore Makusha
has been approved by the following supervisory committee members:
, Chair
Date Approved
, Member
Date Approved
, Member
Date Approved
iii
ABSTRACT
Objective: Encephaloceles are difficult to detect on imaging, leading to missed diagnosis for
many years. Herein, we aim to describe encephalocele MR characteristics to enhance detection
and interpretation of an abnormality that underlies intractable temporal lobe epilepsy of
approximately 10% of patients. We postulate that some MR features that are easier to visualize
than encephaloceles, such as CSF clefts or cortical distortions, along with MR signs of increased
ICP such as empty sella and Meckel’s cave dilation, should raise neuroradiologists’ suspicion of
potential encephaloceles, and hence improve their detection.
Subjects and Methods: We performed a retrospective study on consecutive patients between
June 2017 to September 2019 who underwent MRI including T2-weighted imaging and high-
resolution CT scans. Demographics, clinical features, radiologic findings, and encephalocele
location data were collected for all patients. Two neuroradiologists reviewed all cases with
particular emphasis on morphological features of encephaloceles and MR signs of increased ICP.
Stratified analysis was used to control for confounding by age, gender, and body mass index.
Results: We included initial 93 patients in our study. Encephaloceles were found in 50 of these
patients, with left temporal lobe and bilateral encephaloceles being the most common at 18%,
and 23%, respectively. MR image characteristics of IIH were found in approximately 25% of
patients. Thirteen of 15 patients found with empty sella or partially empty sella or Meckel’s cave
dilation were obese (BMI > 30 kg/m2) compared to patients with normal BMI (Pcorr = 0.0028).
Conclusion: We describe the various MR morphological features of encephaloceles and
correlate those findings to improved detection of encephaloceles.
4
Dedication
To my mother,
Who taught me diligence and resiliency, and to do unto others as you would have them do
unto you
5
Verily, verily, I say unto you, except a corn of wheat fall into the ground and die, it abideth
alone: but if it dies, it bringeth forth much fruit.
–
John 12:24
vi
Table of Contents
ABSTRACT ……………………………………………………………………………………………………………………………………….. III
ACKNOWLEDGMENTS ……………………………………………………………………………………………………………………….. IX
CHAPTER 1 ……………………………………………………………………………………………………………………………………… 10
INTRODUCTION ………………………………………………………………………………………………………………………………. 10
CHAPTER 2 ……………………………………………………………………………………………………………………………………… 18
MATERIALS AND METHODS ………………………………………………………………………………………………………………. 18
PATIENTS STUDY GROUP AND METHODS ……………………………………………………………………………………………………….. 18
MRI
……………………………………………………………………………………………………………………………………………….. 19
COMPUTED TOMOGRAPHY (CT) …………………………………………………………………………………………………………………… 19
IMAGE ANALYSIS …………………………………………………………………………………………………………………………………….. 20
CLINICAL EVALUATION ………………………………………………………………………………………………………………………………. 22
SURGERY
………………………………………………………………………………………………………………………………………………. 22
STATISTICAL ANALYSIS ………………………………………………………………………………………………………………………………. 22
CHAPTER 3 ……………………………………………………………………………………………………………………………………… 23
RESULTS
…………………………………………………………………………………………………………………………………………. 23
CLINICAL FEATURES, RADIOLOGIC FINDINGS, AND ENCEPHALOCELE LOCALIZATION ………………………………………………………….. 23
TABLE 2 CONTINUED. THE CHARACTERISTICS OF THE FINAL PATIENTS INCLUDED IN THE ANALYSIS: ( ……………. 26
CHAPTER 5 ……………………………………………………………………………………………………………………………………… 36
DISCUSSION ……………………………………………………………………………………………………………………………………. 36
CHAPTER 6 ……………………………………………………………………………………………………………………………………… 44
CONCLUSION ………………………………………………………………………………………………………………………………….. 44
APPENDIX 1 ……………………………………………………………………………………………………………………………………. 45
REFERENCES
……………………………………………………………………………………………………………………………………. 47
vii
vii
LIST OF TABLES
TABLE 1: CLASSIFICATION OF ENCEPHALOCELES: …………………………………………………………………………………………………….. 12
TABLE 2. THE CHARACTERISTICS OF THE FINAL PATIENTS INCLUDED IN THE ANALYSIS:
…………………………………………………………… 25
TABLE 3: SUMMARY OF THE LOCATION OF THE ENCEPHALOCELES. ………………………………………………………………………………… 27
TABLE 4: MRI AND CT IMAGE FINDINGS: …………………………………………………………………………………………………………….. 40
viii
viii
LIST OF FIGURES
Figures
FIGURE 1: SOME LOCATIONS OF ENCEPHALOCELES: …………………………………………………………………………………………………. 14
FIGURE 2:FLOW CHART OF THE STUDY: ……………………………………………………………………………………………………………….. 23
FIGURE 3: BMI OF 51 PATIENTS WITH ENCEPHALOCELES AND MENINGOENCEPHALOCELES: …………………………………………………… 29
FIGURE 4: T1W1 SAGITTAL POST-CONTRAST MRI DEMONSTRATING EMPTY SELLA: ……………………………………………………………. 30
FIGURE 5: MRI SCAN. ………………………………………………………………………………………………………………………………….. 31
FIGURE 6: AXIAL CT SCAN.
……………………………………………………………………………………………………………………………… 32
FIGURE 7: MRI SAGITTAL 3-DIMENSIONAL (3D) T2 SPACE IMAGE. …………………………………………………………………………….. 33
FIGURE 8: MAGNETIC RESONANCE IMAGING. ……………………………………………………………………………………………………….. 34
FIGURE 9: MRI SCAN SHOWING AN EXAMPLE OF AN OCCIPITAL ENCEPHALOCELE. ………………………………………………………………. 35
ix
ix
ACKNOWLEDGMENTS
I would like to thank…
Dr. Richard Bronen
For his guidance and patients throughout the year, and for his insight and support in this
thesis project. It was a privilege to work with you and I am extremely grateful for everything you
taught me.
Dr. Sean Lisse and Dr. Anthony Abou Karam
For all your guidance, help, mentorship throughout this project. I simply will not have
been able to complete it without your unwavering support.
Yale University Department of Radiology and Biomedical Imaging
For giving me the opportunity to do a thesis project, which has been a rewarding
academic experience at Yale, and also for providing me with the resources that I needed to
complete this project.
My Family
For all the support, love and motivation that I could ever want and for tolerating my
absence from y’all over the last 10 years. I also want to thank my surrogate families—the
Loughrans, Riemanns, and Rashins for all your love and unwavering support.
My Friends
Thank you so much for always being there for me, and for listening or pretending to
listen to me talk about my work.
10
CHAPTER 1
INTRODUCTION
Epilepsy afflicts 0.5-1.0% of the population. Medically refractory or uncontrolled
epilepsy occurs in 30% of these patients, and results in significant socioeconomic disability for
the individual and society. Treatment for medically refractory epilepsy includes identifying the
seizure cause through a combination of physiologic and imaging methods, and then removing the
epileptogenic focus through surgical resection. Epilepsy can be caused by a wide variety of
things, including injuries, inflammations of the brain or the lining of the brain, strokes or tumors.
In this context, temporal lobe encephalocele (TE), a herniation of the brain parenchyma
through a defect of the adjacent dura and bone space, can be a cause of intractable medically
refractory temporal lobe epilepsy in up to 10% of the patients.1-3 Encephaloceles are increasingly
recognized as a surgically treatable cause of medically refractory epilepsy. The proposed
mechanisms of their epileptogenicity is likely due to mechanical irritation and secondary changes
such as inflammation and gliosis.4 Unfortunately, TEs are difficult to identify on imaging, an
average time to identification having been cited at 5 years, prolonging patients’ suffering.5 In
contrast, predisposing factors of encephalocele development such as arachnoid granulation,
empty sella changes, and skull base osseous defects are often easier to identify.
TEs are skull base defects arising from idiopathic, iatrogenic, traumatic, congenital,
inflammatory, or neoplastic etiologies. While spontaneous encephaloceles are presumed to be
rare, their true prevalence is likely higher than currently recognized.2,5,6 This is because these
11
lesions can be missed on standard imaging modalities on initial medical evaluation. Because not
all defects are symptomatic, coupled with the risk of surgery, most cases can be managed
conservatively. The majority of encephaloceles are managed conservatively because they are
asymptomatic, and the risks of surgery are not justified. However, the encephaloceles may not
stay asymptomatic and indeed progress resulting in herniation of intracranial contents through
the bony dehiscence leading to CSF leaks and seizures if the dural tear exposes the subarachnoid
space.4,7 TEs are increasingly being visualized during neuroimage assessment of patients with
refractory temporal lobe epilepsy, and their identification could indicate an intracranial potential
seizure focus.7-9 Upon TE identification that is localized to a right or left temporal lobe based on
seizure semiology, EEG monitoring and potentially FDG PET imaging, focused resection of the
epileptogenic tissue associated with herniation, and repair of the temporal floor defect can
provide definitive treatment. This is because knowing the general region where a potential
epileptogenic focus could be located increases the chances of localizing TEs. There are different
encephaloceles classifications as shown in Table 1, with nonuniformity of terminology and
confusion in image delineation for reporting. This complicates the accurate diagnosis and further
management of patients.
12
Table 1: Classification of encephaloceles:
There are various ways of classifying encephaloceles with significant nonuniformity of
terminology and confusion in delineation of imaging for accurate reporting. Adapted from
Manjubashini et al.10
TEs mainly occur in two patterns: lateral encephaloceles which often present with
cerebrospinal fluid (CSF) otorrhea and on rare occasions—epilepsy. Medial encephaloceles often
presents with epilepsy and rarely, with CSF leaks. They occur in any supratentorial regions of
the cranium, although the anterior region of the middle cranial fossa is the most frequently
affected area. Besides the anteroinferior temporal lobe, other common anatomic locations
include fronto-ethmoidal, spheno-orbital, spheno-maxillary, nasopharyngeal, and aural (Figure
1).1,11 In anteroinferior TE, the parenchyma protrudes through the middle cranial fossa defect and
13
are increasingly seen as a cause of medically refractory but surgically treatable epilepsy. Skull-
base CT imaging, coupled with high resolution MRI including specific sequences, are useful
imaging modalities for detecting these subtle lesions. Furthermore, heavily T2-weighted MRI
sequences that emphasize bright CSF are useful in identifying encephaloceles. In the literature,
three large retrospective studies of 230 patients with medically refractory epilepsy and TE who
underwent surgical treatment,5,11,12 showed only four patients continued to have seizures after
surgical resection; importantly, all subjects showed significantly decreased seizure occurrence
with no complications. These patients underwent anterior temporal lobectomy with
amygdalohippocampectomy (ATL-AH) while some had focal surgical therapy with
encephalocele resection with or without local temporal pole resection. For seizures originating in
the right temporal lobe, there is a lower threshold for performing a larger resection such as an
ATL-AH in conjunction with resection of encephalocele. However, seizures originating in the
dominant temporal lobe, standard temporal lobectomy poses memory and language risks which
has led other others to opt for more tailored encephalocele resection with limited associated
cortex.
14
Figure 1: Some locations of encephaloceles:
Color-coded multidetector CT image with volume rendering shows the usual locations of arachnoid
granulations adjacent to pneumatized structures of the skull base, such as the midline sphenoid sinus
(orange area); the lateral wall of the sphenoid sinus (teal area); the tegmen tympani (purple area); and
along the floor of the median fossa, from the tegmen tympani to the lateral surface of the sella turcica
(yellow area). Arachnoid granulations have a characteristic lobulated surface and should not be mistaken
for a bone defect (arrows). Sites of inherent structural weakness in the skull base, such as the perforations
in the cribriform plate (pink area) and the sellar diaphragmatic (blue area) may play a role in the
development of encephaloceles. Adapted from Alonso et al.6
15
Further limiting our understanding of encephaloceles and other osteodural defects is the
variability in nomenclature such as such as meningocele, meningoencephalocele, meningeal or
arachnoid hernia, arachnoid diverticulum or arachnoid cyst. These distinctions in naming reflect
variable contents of herniation and occasional inaccuracy because of the limited ability to
visualize the lesions by imaging and during transcranial surgery.13 Endoscopic skull base
approach can render direct visualization of defects and the contents in them, thus allowing for
more definitive definitions to be made; however, this is only rarely done after initial delays in
diagnosis by imaging. The difficulties to visualize the content of the sac-like protrusions at
imaging and during transcranial surgery and an imprecise nomenclature limits our understanding
of why and when these lesions develop.13 If a lesion contains meninges or both meninges and
brain matter, or if brain herniates into a prominent arachnoid granulation, or whether the dura is
intact or not, may not be confirmed on microscopic examination.14 Furthermore, varying
histopathological descriptions (e.g., focal cortical dysplasia (FCD) associated with temporal lobe
meningoencephaloceles) further limit our understanding of the pathogenesis of these lesions.15,16
Therefore, even after the surgical resection of supposed encephaloceles, chances are that the
surgically resected foci and location may remain difficult to accurately describe.
There is increased knowledge of the link between encephaloceles, obesity and increased
intracranial pressure (ICP); however, the incidence of this link is unknown.6,17 Further, the
mechanism by which obesity contributes to TEs has not been completely elucidated but is
proposed to be due to increased intra-abdominal pressure leading to increased intrathoracic
pressure. Subsequently, this leads to increased cardiac filling pressure impeding brain venous
16
return.17,18 The increase in pressure enhances the propensity of ectopically positioned arachnoid
granulations along the middle cranial fossa dura to induce erosions and middle cranial fossa
dehiscence causing encephaloceles and meningoceles.18 It has been hypothesized that chronically
increased ICP may also contribute by causing localized thinning of the bone, with eventual
herniation of part of meninges with/without brain tissue. While previous studies have suggested
an association of obesity with spontaneous CSF leaks and TEs, the majority of these studies have
been observational without comparison groups.19,20 Literature review notes that encephaloceles
commonly occur in middle aged women who are obese, and have clinical symptoms and
radiologic signs of elevated ICP similar to those of benign or idiopathic intracranial hypertension
(IIH).6 However, a substantial number of patients with TEs are not reported on MR imaging due
to their difficulty in detection; hence, careful review with high-resolution T2 sequences may
improve results. Even on CT imaging, encephalocele are indistinguishable from cholesteatoma,
cholesterol granuloma, granulation tissues, or middle ear fluid, appearing as non-enhancing
lucency making diagnosis challenging.5,12,18,21,22
IIH is a common clinical problem that may lead patients to present to a primary care
physician, neurologists, neurosurgeon, or even an ENT surgeon. Empty sella, cystic lesions in
the petrous part of the temporal bone, Meckel’s cave dilation, and CSF clefts, tethering and
distortions are associated findings in cases of IIH.14,21,23 The role of empty sella as an indicator of
raised intracranial pressure was supported by the observation of elevated CSF pressure in
individual patients,24 and in a series of 10 patients who underwent lumbar puncture after sealing
of the defect.25 In addition to the presence of an empty sella as a radiologic sign, patients with
encephaloceles are usually of female sex, middle aged, and obese. Elevated ICP and obesity are
associated with development of spontaneous CSF fistulas in the temporal (petrous) bone, lateral
17
recess of the sphenoid sinus, and cribriform plate as well as the occasional association of
temporal lobe seizures and rhinoliquorrhoea; this further points towards a role of elevated
intracranial pressure in the pathogenesis of temporal lobe encephaloceles.5,6,17,26 MRI is useful in
distinguishing between these pathologies, with encephaloceles showing as isointense extending
herniations in continuity with the brain.16,18,27,28
We analyzed clinical, imaging, and surgical findings in a consecutive series of 56 patients
undergoing presurgical work-up for medically refractory epilepsy with radiographic evidence of
encephaloceles. A careful analysis of high-resolution CT and T2-weighted MR images was
performed with emphasis on imaging signs of IIH as potential image surrogates to aid in the
identification of encephaloceles.29,30 Herein, we hypothesize that elevated ICP, coupled to obesity
leads to increased encephaloceles, and in the right microenvironment, refractory temporal lobe
epilepsy.
18
CHAPTER 2
MATERIALS AND METHODS
Patients Study Group and Methods
Our study included 93 consecutive patients with anterior-inferior temporal lobe
encephaloceles detectable on MRI at Yale New Haven Hospital. Patients were identified using
the search words “encephalocele and not encephalocele” from our Montage database.
Consequently, the validity of the search results was confirmed by manual searches and
encephaloceles confirmation in the individual patients’ electronic medical records. We reviewed
the location, presence of granulations and distortions, empty sella and Meckel’s curve dilations,
CT data and other signs of increased intracranial pressure (ICP) such as CSF clefts, cortical
distortions, and tethering.
Patients with a history of trauma, encephaloceles resulting from surgical procedures,
previous paranasal sinus or skull base surgery, and congenital malformation of the skull base
were excluded from the study. Measurements of the size of the osseous defect was performed on
axial and coronal images. The maximum dimension of the defect was recorded. To assess
differences in the sizes among the locations of the ethmoid, midsphenoid, and lateral sphenoid,
we analyzed the data using the unpaired Student t test, with a significance set at P 30 kg/m2 were considered obese and patients with a BMI < 30
kg/m2, non-obese.
Surgery
The multidisciplinary surgical epilepsy team at Yale evaluated the results of all tests and
surgical treatment decisions were made at the conference. Surgical procedures were selected on
the basics of each patient’s symptomatology, neurophysiologic findings, and duration of
epilepsy. Short duration of epilepsy, lack of mesial temporal symptoms, risk of significant verbal
memory deficit, and palliative treatment goal are some reasons that have been cited to support a
more temporal pole resection and sparing of mesial temporal structure.
Statistical Analysis
Stratified analysis was used to control for potential confounding by age, gender, and body
mass index. A one-factorial ANOVA was calculated for the BMI and the morphology of the sella
(normal, partial empty sella, empty sella and Meckel’s cave dilation).
23
CHAPTER 3
RESULTS
Clinical features, radiologic findings, and encephalocele localization
For the study, the demographics, radiologic, and neurophysiologic data for all the patients
involved are analyzed in Table 1. We started with a total of 94 patients from the Montage
database who potentially had encephaloceles on imaging from Yale New Haven Hospital
between 2010 and 2019. Patients were excluded from the initial count for false positives, no true
encephaloceles, post-surgical encephaloceles, and underaged patients defined as those below 2
years of age at the time of imaging (Figure 1). Patients who had TEs after surgical procedures
were also excluded.
Figure 2:Flow chart of the study:
