Radiological Studies on Hippocampal Development ...356455/FULLTEXT01.pdf · Limbic lobe/ system, short overview of the anatomy Limbic lobe, a supernumerary lobe, [49] an arbi-trary

Radiological studies on

hippocampal development.

Morphological variants and their

relationship to epilepsy.

Dragan Bajic

�� !"��#�� $��%&��'"%"��"()%*�+�� +��+�,��-.��+�/� ��01�#�� 2�� 1

��3��1�'"%"1�4� ��2�� 5��1�/��6�� 4��7��1�8��+�� 7��+�� 1�*"��1��1�823$�(!&9(%9*":9'%*:9;1

�� +�� +� � � +�� 1�� +��1�+��+�� 1�8+�� +��1�#�� -85801

#��+�� +��<��+�858��9�� +� �� +��1

/��-/40��+�'"%�� %*"�9�� 1�858��+� ��%(�=��>��+��-'"��+�9�� &��01�;"=�+��'"%�� 858�-?"��+�9�� ?��9�� %:��01�#�� ++��+�� -�@"1"'01�'*=�+� �� 8581�#��+��<��+��1�#�� 8581�??=�+��4�� *!=�+��>� �� 8581�858��+�� +��1

A�� +� %:"� �� >� 1� #�� ';9'?��B��'?�� '*9'&��B��!'�� '(9;:��B��:?�� 1�858��+� ��*"=��'*=�� %?=��1�#��+��<�� ++�� @�'*��B�� C�'*��B�� +�� -�@�"1""%01�.��'*��B�� +��<�� +�858�� 1

/48�� +� :;� +�� 1�#��/48��+�� -�B0�%!9%&�� :"�� B�%(9;* � #�� 9� �� ;*� +��B�%!9;'1�#�� +� �� B�;'1� �#�� B�'%�� +��B�;;�� 1�8�':D:;�+��-?%=0�� ';�+�� +��1

��/48��3�� 5��7��

�� !��!��!"��#��#�$��%&'()*+)�#��%��

E��3��'"%"

823$�(!&9(%9*":9'%*:9;��)�)��)��) ��9%;%((:�-��)DD��1��1��D��F��G��)�)��)��) ��9%;%((:0

TO MY FAMILYNe mogu vratiti vrijeme koje nisam proveo sa vama ali ste zauvijek u mojim mislima i mom srcu.

LIST OF PAPERSThis thesis is based on the following studies, which will be referred to in the text by their Roman numerals.

I. Bajic D, Wang C, Kumlien E, Mattsson P, Lundberg S, Eeg-Olofsson O, Raininko R.: Incomplete inversion of the hippocampus - a common developmental anomaly. Eur Radiol. 2008 Jan;18(1):138-42.

II. Bajic D, Kumlien E, Mattsson P, Lundberg S, Wang C, Raininko R.: Incomplete hippocampal inversion - is there a relation to epilepsy? Eur Radiol. 2009 Oct;19(10):2544-50.

III. Bajic D, Ewald U, Raininko R.: Hippocampal development at gestation weeks 23 to 36. An ultrasound study on

preterm neonates. Neuroradiology. 2010 Jun;52(6):489-94.

IV. Bajic D, Canto Moreira N, Wikström J, Raininko R.: Asymmetric hippocampal development is common. A fetal magnetic resonance study.

Variation and asymmetry of the fetal hippocampus. Manuscript submitted.

TABLE OF CONTENTSLIST OF PAPERS ..................................................................................................................................... 5TABLE OF CONTENTS .......................................................................................................................... 7ABBREVIATIONS ................................................................................................................................... 9INTRODUCTION ....................................................................................................................................11

Milestones in brain research ............................................................................................................................. 11History of imaging with CT and MRI ........................................................................................................... 11History of limbic system research ................................................................................................................ 12History of the hippocampus research........................................................................................................... 12

Limbic lobe/ system, short overview of the anatomy ....................................................................................... 13Hippocampal embryology ................................................................................................................................. 14Hippocampal anatomy and functions ............................................................................................................... 16Hippocampal pathology .................................................................................................................................... 18Imaging in hippocampus assessment ................................................................................................................ 19

US ................................................................................................................................................................. 19CT ................................................................................................................................................................. 19MRI ............................................................................................................................................................... 19PET ............................................................................................................................................................... 19SPECT ...........................................................................................................................................................20MRS ..............................................................................................................................................................20Diffusion M RI ...............................................................................................................................................20

Introduction to present studies ..........................................................................................................................20AIMS OF THE STUDY ......................................................................................................................... 21MATERIALS AND METHODS ........................................................................................................... 23

Subjects .............................................................................................................................................................23Children and adults (paper I and II) ............................................................................................................23Premature neonates (paper III) ...................................................................................................................23Fetuses (paper IV) ........................................................................................................................................23

MR methodology in children and adults (paper I and II) .................................................................................24Prestudy –validation of the MR technique ...................................................................................................24Studies of the children and adults (paper I and II) ......................................................................................24

Ultrasound methodology on premature neonates (Paper III) ...........................................................................26MR methodology in fetuses (paper IV) ............................................................................................................ 27Statistical analysis for all studies ......................................................................................................................28

RESULTS ............................................................................................................................................... 29�� .................................................. 29Occurrence of the IHI in the non-epileptic and epileptic populations (Paper II) .............................................30

Control g roup ...............................................................................................................................................30Epilepsy patient groups ................................................................................................................................30

Laterality and distribution of the total and partial forms of the IHI .................................................................30Control g roup ...............................................................................................................................................30Epilepsy patient groups ................................................................................................................................30�� ..................................................................... 32

Ultrasound study on preterm neonates (Paper III) ........................................................................................... 32MRI study on fetuses (Paper IV) ...................................................................................................................... 32

DISCUSSION .......................................................................................................................................... 35Studies on the hippocampal development ......................................................................................................... 35Studies focusing on the morphological variants of the hippocampus and their relation to some epilepsy syndromes ........................................................................................................ 39

CONCLUSIONS ..................................................................................................................................... 41ACKNOWLEDGEMENTS .................................................................................................................... 45REFERENCES ....................................................................................................................................... 47

ABBREVIATIONSCT Computerized tomographyFDG FluorodeoxyglucoseGA Gestation ageGW Gestation weekHIMAL Hippocampal malrotationIHI Incomplete hippocampal inversionLTLE Lateral temporal lobe epilepsyMRI Magnetic resonance imagingMRS Magnetic resonance spectroscopyMTLE Mesial temporal lobe epilepsyMTS Mesial temporal sclerosisNMR Nuclear magnetic resonance PET Positron emission tomography SGA Small for gestational ageSPECT Single photon emission computed tomographyTLE Temporal lobe epilepsyUS UltrasoundGRE Gradient echo

11

INTRODUCTION

Milestones in brain research5000 year B.C. Trepanations performed for thera-peutic purpose. 4000 year B.C.��!��ancient Sumeria. 3000 year B.C. The word “Brain,” is used for the �� "��#��$��Egyptians however believed that the brain was not ��"��!�!��"�""��was discarded while the heart was preserved. 2500 year B.C. The Edwin Smith Surgical Pa-�� anatomy of the brain was discovered. The papyrus documented twenty-six cases of brain injury with treatment recommendations [8].450 year B.C. The ancient Greek physician Al-cmaeon performed anatomical dissection on ani-mals and concluded that the brain was the central organ of intelligence instead of the heart. Hippocrates (460-379, B.C.) also stated that the brain was the site of intelligence. Aristotle (384-322, B.C.) distinguished between long-term and short-term memory.Galen (130-200, A.C.) accepted the Hippocratic approach and as physician to the gladiators was conscious of the consequences of brain and spinal injuries.In 1543 Andreas Vesalius published one of the �� %�� &��%�$� '�� !�-scribed how the nerves and the brain work. He dissected both humans and animals and noticed that the brains of many animals and all mammals have the same ventricles as humans. He claimed that the ventricles could not be the site for emo-tion and memory because animals possessed no soul [81].In 1664� �� "�� !� ��book “Cerebri Anatome” was written by Thomas Willis.

In 1811 the Scottish surgeon Charles Bell was the ��"�%��!��"��!��-sory nerves and claimed that they were anatomi-cally separated in the spinal roots [67]. In 1817 James Parkinson published “An essay on the Shaking Palsy”. He described a degenerative disease of the nervous system that even now is known as Parkinson’s disease [60].It is thought that modern research of the brain started with Paul Broca who discovered a speech center in 1862 [39].In 1873 John Hughlings Jackson (1835-1911) pre-��!�� !�� *��[49]. In 1906 Santiago Ramon y Cajal and Camille Gol-gi won the Nobel prize for research of structure and function of nerve cells [27].In 1929�'��+��<��&-��<��=��in the human brain. He also developed and dem-onstrated the EEG, the equipment for registration of these potentials [79].

History of imaging with CT and MRIIn 1882 Nikola Tesla discovered the rotating mag-��!>��!�"��!��<��$�In 1895� @��"� Q��!� X�� %� �� X-ray image.In 1937 Isidor I. Rabi (Nobel prize for Physics in 1944��"�!��"��presence by absorbing or emitting radio waves �� &��!� �� "��!$In 1946 Edward Mills Purcell published his dis-covery of nuclear magnetic resonance in liquids and in solids and independently Felix Bloch pro-posed the Bloch equations which determine the time evolution of nuclear magnetization. In 1952

12

Radiological studies on hippocampal development

they shared the Nobel Prize for “their develop-ment of new ways and methods for nuclear mag-netic precision measurements”.In 1971 Raymond Damadian discovered that the hydrogen signal in cancerous tissue is different from that of healthy tissue, because tumors con-tain more water.In 1972�\�!��'��!��!�^��Q��"�%�invented CT. In 1979 they were awarded the No-bel Prize for Medicine.In 1973��_��>��!��!��`� �image (Nobel Prize for Medicine in 2003).Between 1974 and 1976��Q��-ners were installed.In 1977��"�� "�!�$

History of limbic system researchIn 1878��+��!��"��"��”le grand lobe limbique” mainly in the context of comparative anatomical studies [55].In 1937 James Papez described his anatomical model of emotion, the Papez circuit. He trans-��"�!�+��{��"��!��"��lobe to the functional system -limbic system. His system included hippocampal formation (subicu-lum), fornix, mammillary bodies, mammillotha-lamic tract, anterior thalamic nucleus, genu of the internal capsule, cingulate gyrus, cingulum, para-hippocampal gyrus, entorhinal cortex, perforant ��>� ��!� �� !� �� >�again the hippocampus [59].In 1948 Yakovlev added new structures associ-ated with emotions (orbitofrontal and insular tem-poral lobe cortex, the amygdala and dorsomedial thalamic nucleus - Yakovlev s circuit) [83].1952 Paul D. MacLean included additional struc-tures in a more dispersed “limbic system,” The term was formally introduced by MacLean. He assumed that the main function of the limbic sys-tem was the formation and expression of emotions ��>��!��&��!��|}}~$

History of the hippocampus researchCa 1564 Giulio Cesare Aranzi coined the name hippocampus from the Greek word for seahorse ��>� �� >� ��>� %�"�� monster). 1886 Camillo Golgi illustrated the unique and complicated hippocampal structure [41].1911 Santiago Ramon y Cajal published “Histolo-gie de Systeme Nerveux” with a drawing of the hippocampus and an explanation of possible im-pulse direction [19].1934 Rafael Lorente de No explained many hip-pocampal cell types and their axonal and dendritic patterns. His terminology is in general use today [47].1956, 1958 Theodor Blackstad, who was called “the father of the modern hippocampus histolo-gy”, presented both internal and external neuronal connections of the hippocampus [12,13].a) Olfactory function of the hippocampus1890 John Hughlings Jackson and Charles Beevor reported a patient with olfactory sensations dur-ing the seizure [38]. 1947 Alf Brodal supported the theory about the hippocampus’s role in olfaction and reported a number of arguments against this theory [14,15].b) Role of the hippocampus in the emotionsIn 1937 Papez explained the role of the limbic sys-tem in the emotions through the construction of the special system called the “Papez circuit”. In that circuit, the cortical and subcortical structures were connected from the hippocampus to the neo-cortex and back to the hippocampus. This concept ��"��!��"��responsible for emotions, and according Papez, the hippocampus’s role was to collect sensory in-formation which in later phases provided the basis for the emotions [59].In 1982 Brodal rejected Papez s theory and con-cluded that his theory “is of historical interest only” p. 672 [15].

13

Introduction

c) Role of the hippocampal function in atten-tion controlIn 1938 Richard Jung and Alois Kornmuller dis-covered a “theta” wave on EEG in the rabbit hip-pocampus [42].In 1954 Green and Arduini reported that “theta” waves were connected with enhanced attention [30].2000 Bush et al, revealed that the cingulate gyrus and the hippocampus were responsible for general attention control [18].d) Role of the hippocampus in the memoryIn 1900��!"��+��"�!� ��"�"��!��!��of the hippocampus found at autopsy [9].In 1957 William Scoville and Brenda Milner de-scribed the unexpected outcome of surgery with severe anterograde and partial retrograde amnesia after destruction of the hippocampus in an epi-lepsy patient. The patient was unable to form new episodic memories and could not remember any events that occurred just prior to his surgery, but retained memories for things that happened years earlier [70].In 1963-1972 Robert Isaacson, Robert Douglas and Daniel Kimble suggested “response inhibi-tion hypothesis”, a widely held theory of hyperac-tivity. Interpretation of this theory includes work-ing memory impairment [38].In 1972 Endel Tulving distinguished between memory for episodes and memory for semantic items [80].In 1970-1973 O´Keefe and Nadel independently developed hippocampus cognitive map theory [58].

Limbic lobe/ system, short overview of the anatomyLimbic lobe, a supernumerary lobe, [49] an arbi-trary cortical area [40], is situated in the cerebral hemisphere partly in the medial temporal lobe and partly surrounding the corpus callosum. From an anatomical point of view, its structures are part of the frontal, parietal and temporal lobe. There are cortical structures: the parahippocampal, cingu-late and subcallosal gyrus and the hippocampus, in the limbic lobe. Papez transformed the anatom-��!��"��system the limbic system where some subcortical structures were included. R. Jinkins divided the limbic system into cortical structures (cornu am-monis, gyrus dentatus, subiculum, entorhinal cor-tex, subcallosal and supracallosal gyrus, medial/lateral longitudinal stria of Lancisii, cingulated gyrus, parahippocampal gyrus and paratermi-nal gyrus) and subcortical structures (amygdala, habenular nuclei, septal nuclei, hypothalamus, anterior thalamic nuclei, epithalamus, mesen-cephalic tegmentum, mammillary bodies, and fornices) [40]. Other authors divided the limbic system structures into two arches, outer and in-ner, separated by the sulcus of the corpus cal-losum anteriorly and hippocampal sulcus poste-riorly [45,46]. The main structures of the limbic system being interrelated by a complex system of ��$� �� "�� !�� <��the fornix across corpus callosum to the hypo-��"��!<!��!��""�� $� �� ""�� &�� >� �� "��>��!��""��to the mammillary body and after that, via the mammillothalamic tract to the anterior thalamic ��>��!��<��"��-bres to the anterior cingulated gyrus. From this ��>�� &��-riorly and to the hippocampus via the entorhinal cortex posteriorly [59]. When Papez demonstrated his circuit, he believed that it was responsible for the emotions. More recent research presented its role in the memory [51] and postulated that Perez’s theory “is of historical interest only” [15].

14


Hippocampal embryologyThe hippocampal development has not been stud-ied by a great number of researchers. All these studies were referred to a small number of abort-ed fetuses of different gestational ages and micro-scopically analysed [2,35,45,46,65]. Some of them performed an MRI examination of the speciments [45,46,65]. In recent times, there are MRI stud-ies of the fetal brain analysing the development of some of the hippocampal structure [29,66]. The development of the hippocampus starts before GW 10, where are dentate gyrus and cornu am-monis situated in the posteromedial wall of the lateral ventricle [35] (Fig. 1).

At GW 10 a broad shallow hippocampal sulcus appears on the medial aspect of the temporal lobe, during thickening of the dentate gyrus.

At GW 11 the shallow hippocampal sulcus is lo-calized between the dentate gyrus and the cornu ammonis.

At GW 12 the hippocampal sulcus is still shal-low but pointed toward the cornu ammonis. At the tip of the sulcus the diffuse cellular zone appears. Later during the development this zone moves deeper into the temporal lobe and the hippocam-pal sulcus follows it.

At GW 13-14 the dentate gyrus increases in thickness and makes the hippocampal sulcus more !��!��!�!��>��!��!��cornu ammonis. The diffuse cellular zone be-�"��"��!��!��!��|�}~$�

At GW 15-16 the sulcus hippocampalis is best developed in the temporal region, and inversion of the hippocampal formation into the temporal lobe has begun. The gyrus dentatus is the largest part of the developing hippocampus and its rotation in-ward is the fastest. The cornu ammonis follows this rotation. The diffuse zone becomes linear

because of the pressure from the gyrus dentatus and cornu ammonis and moves over into a com-pression zone. The parahippocampal gyrus and subiculum project more medially than the gyrus dentatus and cornu ammonis [35,45].

By the GW 18 the hippocampus is infolded into the temporal lobe. The form of the hippocampal formation is rounded with the long axis orientated vertically (Fig. 2a). The parahippocampal gyrus and subiculum are larger, the temporal horn is smaller and the germinal matrix reduced in the size compared with the previous period. As the dentate gyrus enlarges and rotates inward, the two sides of the hippocampal sulcus narrows and �� "�!�$�@�� inward rotation of the gyrus dentatus, the diffuse zone gradually becomes aligned with the hip-��"��!�"��"��!��!��-nally transformed into the compression zone. In the deeper portions of the compression zone the cells progressively degenerate and disappear, as the compression increases, to leave a linear zone that becomes less and less cellular. Even before the diffuse zone is in alignment with the hippocam-pal sulcus, fusion has begun as the diffuse zone is transformed into the compression zone. [35,45].

Up to GW 21 the hippocampal sulcus and sur-rounding structures are similar to those found in adults (Fig. 2b). The hippocampal sulcus fuses but sometimes can remain with pia mater and blood <��!�!��$�^��!�-fuse zone narrows progressively it comes com-pletely in line both with the fusion zone and the �� >� ��!� �� "��zone, deeply. At the same time, the compression zone increases in extent, because the gyrus denta-tus is larger [2,3,68].

15

Introduction

Week 9 Week 10-11 Week 14 Week 21

1

4HCS

32

Figure 1. Schematic diagram of the fetal development of the hippocampal formation. ��>��Q��""��>��">� ��"��>�'Q��'��Q�"��$

Figure 2. ��'��"��"��\@��!��"��"��$��}��vertical longitudinal axis. �� "��"��!�<��!��<�!��"��!��*��!��&�>��weighted IR in the coronal plane).

16


Hippocampal anatomy and functionsHippocampal formation includes both gray mat-ter (dentate gyrus, cornu ammonis and subiculum) ��!��"��<��!��"��$��!��dentate gyrus consists of three layers. The cornu ammonis is divided into the 4 segments called CA1 - CA4 (Fig. 3) and contains six layers on mi-croscopy [25].

On coronal MRI images, the normally devel-oped adult hippocampus has an ovoid shape with the horizontal diameter longer than the vertical (Fig. 4 a-c). The sulcus collateralis is situated un-der the hippocampal formation. Between the hip-

pocampal body and the collateral sulcus there is a white matter area “collateral white matter”. The �"��!��"�!��hippocampal body. The temporal horn is almost ��!��"��"��$�

Sometimes, the hippocampal form is not ovoid but more rounded (Fig. 5a) or pyramidal (Fig. 5b) with its vertical diameter equal or longer than the horizontal. In such cases, the collateral sulcus is vertical, the collateral white mater dislocated lat-erally and the lateral part of the temporal horn �� "��$� �� <�!� <�� !��!� ��

Figure 3. MR image of the normal adult hippocampus with horizontal longitudinal axis. �'\��'��"��\��>��">�Q^��Q��^""��>�\��\��$

17

Introduction

some patients with severe developmental anoma-lies [4,69] and in epilepsy patients [6,7,10,61] but sometimes in the normal population also [10,16,61]. There are a few studies regarding this condition in healthy population or patients without develop-mental anomalies and without epilepsy [48].

There were three main theories about hippocam-��$��>� �� "��role in olfaction, was rejected as false, even though there are still some supporters of this theory.

O’Keefe and Nadel developed the second theory about the hippocampus s role in “behavioural in-hibition”. The inhibition theory is currently not abandoned but doesn’t have wide support [11].

The third theory, now generally accepted, reveals the role of the hippocampus in the memory but the

mechanism of this function is not completely ex-plained yet. The hippocampus has an important role in episodic or autobiographical memory, the formation of new memories of experienced events [72] but it is a part of the medial temporal lobe memory system which is responsible for general declarative memory. The other memory func-tion of the hippocampus relates to the memory of space. There is now, almost universal agreement that spatial coding plays an important role in hip-pocampal function. If the hippocampus is dam-aged, humans may not remember where they have been and getting lost is one of the most common symptoms of amnesia.

Figure 4. MR image of the normal adult hippocampus ��^&��!�� "��>� ��Q��!�� "��>� ��!�� "��$

Figure 5. Non-ovoid forms of the hippocampus. �� !�!��"��weighted TSE coronal image. ��"!��"��weighted IR coronal image. Both images perpendicular to the long axis of the hippocampus.

18


Hippocampal pathologyAging in healthy people leads to a gradual decline of some types of memories including substantial loss of neurons in the hippocampus and shrinkage of the hippocampus on MRI. Recent studies us-ing more precise techniques has challenged those beliefs, suggesting instead that aging has a limit-ing effect on medial-temporal lobe cytoarchitec-ture ith [64,75,84]. All elderly people do not reveal hippocampal shrinkage, but those who do tend to perform less well on some memory tasks [71].

Alzheimer’s disease is strongly connected with hippocampal disruption [32] and has a severe im-pact on many types of cognition.

Stress has a negative effect on the hippocam-pus which contains high levels of glucocorticoid receptors. Stress-related steroids affect the hip-pocampus by reducing the excitability of some hippocampal neurons, by inhibiting the genesis of new neurons in the dentate gyrus, and by caus-ing atrophy of dendrites in pyramidal cells of the CA3 region [53]. These effects are evident in post-traumatic stress disorder, schizophrenia and se-vere depression [20] and in Cushing’s syndrome, a disorder caused by high levels of cortisol in the bloodstream [73].

Schizophrenia suffering patients can display abnormalities of the cerebral cortex. Hippocam-pal changes such as reduction in the size of the hippocampus have also been described, as well as changes in synaptic organization and connectivity [33]. It is unclear whether hippocampal alterations play any role in causing the psychotic symptoms, which are the most important feature of schizo-phrenia, but disturbances in long term memory are frequently observed in people with schizo-phrenia [76].

Epilepsy is a brain disorder characterized by an enduring predisposition to generate epileptic sei-zures and by the neurobiologic, cognitive, psycho-logical and social consequences of this condition. An epileptic seizure is a transient occurrence of signs and/or symptoms due to abnormal exces-sive or synchronous neuronal activity in the brain |��~$��!�� !�� International League Against Epilepsy (ILAE)

��*��!��!��"�!��and divided into two groups: 1: Generalized epileptic seizures considered to

originate at some point within, and rapidly en-gage bilaterally networks.

2: Focal seizures considered to originate within networks limited to one hemisphere, which may be discreetly localized or more widely dis-tributed.

There are a number of different disorders hav-ing increased predisposition for seizures [26]. The ��!��!�� *��syndromes are characterized by age at onset or ��"��##\�� !��$��!��are grouped as: genetic, structural/metabolic or ��%��$� �� the word “cryptogenic” was used instead of “epi-lepsies of unknown cause” [26].

Temporal lobe epilepsy (TLE) is a form of the focal epilepsy where seizures originate in either medial or lateral temporal areas. Because of strong interconnections the seizure often extends and in-volves both areas including neighbouring areas on the same side of the brain as well as contralat-eral homologues brain areas. There are two main types of the TLE: mesial temporal lobe epilepsy (MTLE), arising in the hippocampus, parahip-pocampal gyrus and amygdala and, lateral tem-poral lobe epilepsy (LTLE), arising in the lateral neocortex. The abnormality most associated with mesial TLE is hippocampal sclerosis (scarring) characterized by neuronal cell loss and reactive gliosis in the hippocampal area. The hippocampus is one of the most electrically excitable parts of the brain and one of very few brain regions where new neurons are produced throughout life (neuro-genesis) [48].

TLE may be a progressive condition where sei-zures become more resistant to medications over time. The association between temporal lobe epi-lepsy and hippocampal sclerosis has been recog-nised for over a century, but despite many decades of basic and clinical research it is still not possible

19

Introduction

to assign an arrow of causality. One explanation is that hippocampal sclerosis may represent a pheno-typically similar manifestation of a heterogeneous group of pathologies with diverse pathogenesis, derived from a complex interplay of numerous factors (including genetic, developmental and en-vironmental components) Hopefully prospective neuroimaging studies which can be used to follow its evolution. Although hippocampal sclerosis is ��"��""��!��_#>��"-mon pathological conditions in TLE, are malfor-mation of cortical development, low grade glio-mas and vascular/ traumatic lesions.

LTLE can also be affected by strokes, cortical malformations, multiple sclerosis plaques, etc., and be the cause of seizures from this area. Exact-ly how any of the previously mentioned abnormal-ities actually cause groups of neurons to generate seizure activity is complex and not fully known. Because there can be different lesions, there can also be different mechanisms of generating a sei-zure. So the temporal lobe epilepsies could be caused by many etiologic factors.

Rolandic epilepsy is the most common focal epi-lepsy syndrome in the pediatric age group. This “benign childhood epilepsy with centrotemporal spikes” [26] has the excellent prognosis regarding the seizures, which usually resolve at puberty, and EEG, which will be normalized.

Imaging in hippocampus assessmentIn hippocampus imaging there are 2 groups of methods: 1: Morphological, including Ultrasound (US),

Computerized tomography (CT) and Magnetic resonance imaging (MRI).

2: Functional, including Positron emission tomography (PET), Single photon emissions computed tomography (SPECT), Magnetic resonance Spectroscopy (MRS) and functional MRI (fMRI).

USCranial US is a useful method in the assessment of the neonates but there is no data concerning the use of US in hippocampal imaging.

CTCT is not good enough to show anatomical details, but some pathological conditions can be detected: tumors, atrophic or ischemic changes.

MRI� �� "��visualization of the hippocampal structure due to the great contrast between different tissues. MR images have to be of high resolution because the hippocampus is a fairly small structure. The hip-pocampi are cylindrical structures and the MR images has to be acquired perpendicular to the hippocampal long axis (i.e. in an oblique coronal plane) to respect spatial orientation [22]. MRI ex-aminations of the histological specimens can be ��<��"��!�strength and exposition time theoretically can be unlimited, which is not the case with clinical ex-aminations.

PETPET is a nuclear medicine technique which pro-duces images of functional brain anatomy using gamma rays emitted indirectly by a positron-emit-ting radionuclide (tracer). It can measure changes in glucose metabolism, and neurotransmitter re-ceptor distribution, which are associated with �_#$��!��&��\��!�=�"�*��are the most commonly used tracers in epilepsy.

20


SPECTSPECT shows blood perfusion in the brain and, in contrast to other methods, patients could be sub-mitted both to ictal and interictal examinations. SPECT is the only method which can reveal in-��!��!�=��!��*��$��#Q��have a quick uptake in the brain and reach peak effect within 2 minutes after injection. This initial tracer uptake remains unchanged after 2 hours. Practically if the radiotracer is injected during the seizure, the SPECT examination can be acquired after 2 hours without changing blood perfusion pattern.

MRSMRS is a special technique used for characteri-zation of the biochemistry of the tissues. MRS is based on the unique radio signals of the non-�� "�� "�� and distinct from the frequency of water protons. It can illustrate decrease of functioning neurons in epileptogenic tissue .

Diffusion MRIDiffusion MRI is a technique that can be used to access the integrity of cerebral tissue. Diffusion abnormalities such as increased postictal diffusion are an indicator of seizure localization in TLE.

Introduction to present studiesThe fully inverted hippocampus has an oval con-�� "��>�� <��-sion is not always simultaneous or symmetrical. If the inversion is not completed, the hippocampus retains the round or pyramidal shape. Incomplete inversion has been described in patients with epi-lepsy [6,7,10,61] or severe midline malformations [4,69]. Two studies [10,61] also included healthy controls with abnormal hippocampal shape in 4–10% of the subjects. Bronen and Cheung [16] studied hippocampal formation in 29 volunteers.

They reported that 12 of the 58 hippocampi had a shape deviating from the normal oval shape. The criteria of IHI have been poorly evaluated; sym-metry, laterality, form and relationship to other deviating features in the temporal lobes are not known and the occurrence of IHI and its relation to epilepsy is unclear.

When the change of the general shape and the direction of the longitudinal axis of the hip-pocampus occur to reach the adult form is not well known. Hevner and Kinney [34] reported that the cornu ammonis regions had reached a mature, fully rotated position after 18 weeks of gestation. Kier et al. [45,46] found that the infolded rela-tionship of the dentate gyrus and cornu ammonis was unchanged in a 24-week-old specimen when compared to an 18-week-old specimen. Accord-ing to the study of Arnold and Trojanowski [2], the hippocampus only increases in volume after 20th–21st GW with the increasing of the size of pyramidal cells in the subiculum and cornu am-monis but the hippocampal shape remains un-changed. However, the fusion of the walls of the ��"�� \@� ��!��"��|�}~$�The earlier studies have described the microscopi-cal, not macroscopical development and the mate-rials have been very small. Also the shape of the brain can be changed after removal from the skull.

In different studies, this condition has been designated as Incomplete hippocampal inversion (IHI) [4], hippocampal malrotation (HIMAL) [6] ��"��"��"�!��|��~$�We use the term “incomplete hippocampal inver-sion”. The reason for this choice of terminology is that the hippocampus was, most likely, never completely rotated in this condition, therefore it would not be described as “malrotated”.

21

Aims of the study

AIMS OF THE STUDY

The goals of the present studies are:To document the frequency, laterality, and symme-try of IHI in subjects without epilepsy or obvious developmental anomalies of the brain (paper I).

To evaluate if there are relations between the hip-pocampal form and type of epileptic syndromes by comparing the occurrence of IHI in epileptic populations and non-epileptic populations with neither obvious developmental anomalies of the brain nor any pathological conditions affecting the temporal lobes (paper II).

To assess hippocampal development by analysing hippocampal form using cranial US in neonates born preterm (paper III).

To review the development of the hippocampal re-gion in the fetuses using MRI (paper IV).

22


23

Materials and methods

MATERIALS AND METHODS

Premature neonates (paper III)Cranial US images of about 400 premature neo-nates born at Uppsala university hospital and at regional hospitals gravitating to the Uppsala uni-versity hospital in 2006 – 2008 were reviewed. Premature neonates with intracranial haemor-rhages grade II-IV, ventriculomegaly, massive periventricular leukomalacia (PVL) or intracra-nial expansive processes and those cases in which the examination was technically faulty were ex-��!�!$��"��!��the cranial US, and infants with small germinal matrix haemorrhage (grade I) or small PVL out-side the temporal lobes were included.

Finally selected population consisted of 160 ne-onates, 86 of them were boys and 74 girls, with gestational ages (GA) of 22 to 34 weeks (mean 27 weeks) at delivery. Mean birth weight was 1021 g (range 434 g - 2695 g).

Fetuses (paper IV)The brain images from post mortem MRI of 12 aborted fetuses and from 306 clinical fetal in vivo MRI were evaluated.

The fetuses that underwent post mortem exami-nations had been aborted for extracranial pathol-ogy. The fetuses in which central nervous system (CNS) pathology was revealed in autopsy or MRI were excluded.

The clinical fetal MR examinations were per-formed in Uppsala, Sweden and in Porto, Portu-gal. The indications were suspected body mal-formations, central nervous system abnormalities suspected on previous ultrasound or abnormal screening laboratory tests. Subjects in which re-peated ultrasound or clinical studies revealed brain pathology were excluded from the study. ��!�<��!�� &��!�!$��"��""��!��was a widening of the lateral ventricles.

Subjects

Children and adults (paper I and II)Three hundred patients, subjected in MRI exami-nation, were drawn from the regional epilepsy register. 78 subjects with conditions that could affect the temporal lobes such as obvious in-tracranial developmental anomalies, intracranial masses, hydrocephalus or mesial temporal sclero-sis together with 21 subjects with incomplete data were excluded from the study. 150 subjects were used as controls: 34 were healthy volunteers, 91 patients selected from the hospital archives and 25 patients enrolled in the study. In patients included ��>�� !��the brain were not expected. These patients were suffering from common headaches without neuro-logical symptoms, from cranial nerve symptoms, or from predominantly spinal symptoms. They had no obvious developmental anomalies and did not have epilepsy.

Examinations of the epilepsy patients and con-trol subjects were mixed by an administrator and then analysed by experienced radiologists blinded to clinical data.

When 50 men and 50 women among the con-trols were analysed the data was evaluated to re-veal frequency and type of the IHI in this group. Subjects were aged 9 months to 73 years, with a mean age of 29 years (75% were between 10 and 54 years). At this time point, the control subjects included 20 healthy volunteers, 17 patients en-rolled in the study, as well as 63 patients that were selected from the hospital archives.

From the 300 epilepsy patients, examined with � �>��&��!�!$��"��-rial consisted of 201 epilepsy patients (100 female, 101 male), aged 2 months to 74 years, mean age 26 years (75% were 6–50 years old); and 150 controls (75 female, 75 male), aged 2 months to 75 years, mean age 31 years (75% were 11–54 years old).

24


Fetuses with the atria wider than 10 mm were excluded [1,24,31,82]. Technically poor examina-tions and all examinations in which the hippocam-pal regions were not well evaluable bilaterally or the coronal slices were asymmetrically positioned were excluded (180 fetuses, all examined in utero).

Three post mortem examinations and 60 in utero� �&�"�� !� �� analysis.

MR methodology in children and adults (paper I and II)

Prestudy –validation of the MR techniqueBefore we began the study we created criteria for the studies and validated MR techniques. We test-ed for the type of MR examination needed to eval-uate the structures of the hippocampal region. The images of 30 individuals from the hospital archive were evaluated. All of them were examined with two different MR techniques, “a routine protocol” and “an epilepsy protocol”, using thin slices per-pendicular to the long axis of the hippocampus. Both protocols are presented in Table 1. All the epilepsy protocol examinations were made at 1.5 T, and the routine protocol examinations at 1.5 T in 22 patients and at 0.5 T in 8 patients.

Two experienced radiologists analysed the im-ages independently and separately for the two pro-tocols, using the previously established variables used for analysis of the hippocampal region (Table 2).

One of the radiologists repeated the evaluations after 6 months. One radiologist found incomplete inversion in 16 of the 60 hippocampi, the other in 17 of 60. Intraobserver agreement for variables 1–7 was 97% for epilepsy protocol, and 98% for routine protocol. Interobserver agreement was 97% for epilepsy protocol and 95% for routine

��$��>��"��>��!��!��!��"��protocol examinations, and we decided to exclude this variable from further analyses. We concluded ��>��!��>��!�be used for evaluation of the hippocampal regions.

Studies of the children and adults (paper I and II)All the examinations both with the routine proto-col and the epilepsy protocol, were analysed by one radiologist according to the variables 1–7 in Table 2. If all the coronal slices through the hippocam-pus (the head could not be evaluated) showed a deviating shape, we categorized the IHI as total. If ��!��"��!��!��others deviating, the IHI was of partial type. The images were analysed without knowledge of clini-cal data. The MR examinations were performed using two techniques (Table 1). A total of 142 epi-lepsy patients were examined according to an epi-lepsy protocol. Fifty-nine patients were examined using a routine protocol as they were successfully treated without a further evaluation according to ��$��the control group were examined in accordance with the epilepsy protocol and 91 in accordance with the routine protocol.

Clinical records of the patients, including EEG, �� %�!� �� !�� of epilepsy and location of the epileptic focus. #�� !� ��!�� "��!��"��and related disorders by the International League Against Epilepsy (ILAE) from 1989.

For these studies the research plan was accepted by the National Ethics Committee. All the volun-teers and the patients included in the prospective study gave their informed consent.

25


Table 1. Technique of coronal MR slices

Abbreviations: T1-w = T1 weighted T2-w = T2 weightedSE = spin echo TSE = turbo spin echoIR = inversion recovery FLAIR = fluid attenuated inversion recovery

Epilepsy protocol Routine protocol

Field strength 1.5 T 1.5 T0.5 T (in the minority)

Slice thickness 2.5 mm 5 mm

Gap 0.5 mm 0.5 mm

Coronal slice adjusting Perpendicular to the hippocampus long axis

In the plane of the posterior pontinecontour

Sequences types T2-w TSE, T1-w IR, FLAIR T2-w TSET1-w SE (not in all)

Variable Normally In incomplete hippocampal

1. Hippocampal shape Ovoid Round or pyramidal

2. Localization of HC Lateral part of the choroidal fissure Medial part of the choroidal fissure

3. Craniocaudal diameter of the HC

Shorter than the transversal diameter

Equal or longer than the transversal diameter

4. Tip of the temporal horn Lateral part filled by the HC Lateral part empty and looks large

5. Orientation of the collateral sulcus

Horizontal or with an angle <70 degrees to HC More vertical

6. Position of the collateral white matter* Under the HC Lateral to the HC

7. Long axis of the choroidal fissure

Horizontal, parallel to HC long axis Skew or vertical

8. Position of the fimbria Cranial to HC Medial to HC

Table 2. Variables used for analysis of hippocampal region

Abbreviations: HC = hippocampus* Collateral white matter is the white matter between the collateral sulcus grey matter and

hippocampus just medial to the collateral eminence [13]

26


Ultrasound methodology on premature neonates (Paper III)All examinations were performed in the Neona-tal intensive care unit with an Acuson Sequoia US scanner using a high frequency transducer 10 MHz. US examination was a part of the routine neurological assessment of the premature neo-nates. An anterior fontanel approach with stand-ard six coronal or paracoronal slices in addition to �<��!$��experienced paediatric radiologists performed examinations and the hippocampal forms on the images was subsequently analysed by one ex-perienced paediatric radiologist. Coronal slices of the routine cranial US performed at 1st – 7th postnatal days, preferably on day 1 were analysed. Gestational ages upon examination are shown in Table 3.

One of the parents of each child gave written permission to use the images in the study.

��"��!��!��!��!��slice III or IV at the level of the third ventricle (Fig. 6). The hippocampal form was evaluated and the ratio between the horizontal and vertical di-ameters of the hippocampal body was calculated. The ratios were divided in three categories: �� '��"��$�� $}��"��$�3: If the ratio was between 1 and 1.5, additional

criteria [10,11] such as the enlargement of the lateral part of the temporal horn and the orien-tation of the collateral sulcus were used to de-cide if hippocampal inversion was completed or not.

Table 3. Gestation ages at delivery and at ultrasound examination

Number of neonatesGestations age(Weeks) At delivery At examination

22 3 0

23 13 8

24 22 16

25 24 19

26 24 22

27 9 20

28 13 9

29 17 20

30 14 8

31 8 9

32 7 8

33 2 5

34 2 2

35 0 9

36 0 3

Total 158 158

27


MR methodology in fetuses (paper IV)All examinations were performed with Philips Intera MR imagers (Best, The Netherlands) op-erating at 1.5 T. The post mortem examinations were performed using a birdcage knee-foot coil. T2-weighted 2D fast spin echo (FSE) sequences were performed in the sagittal, coronal and trans-verse planes, with slice thickness of 2 mm and in-plane resolution of 0.50x0.53 mm (transverse plane) and 0.59x0.62 mm (coronal and sagittal planes). A T1-weighted magnetization-prepared 3D gradient echo (GRE) sequence was performed in the sagittal plane, with slice thickness of 1 mm and in-plane resolution 0.59x0.91 mm. From the acquired T1-weighted images, reformats were constructed in the axial and coronal planes. The in utero studies were performed using abdominal ��!��=�&��$��"��included T2-weighted single-shot FSE sequences

Figure 6. Ultrasound image of the both hippocampal regions. An ovoid hippocampus on the right side (white arrow) and non-ovoid on the left side (empty arrow).

in the sagittal, axial and coronal planes. The slice thickness was 3-4 mm, and the matrix between 256x256 and 364x256. Fields of view ranged from 20x20 to 30x30 cm. No sedation was given to the mothers.

Hippocampal regions were evaluated in one or more coronal slices through the area. Progress of the hippocampal inversion was analysed and a comparison between the right and left side was made in every case. In cases where the hippocam-pal sulcus was not yet closed, the angle between the upper and lower lip of the hippocampal sul-cus was evaluated. If the hippocampal sulcus was closed the shape of the infolded hippocampus was ��!��<�!��!>��"!��ovoid with a vertical long axis) or ovoid with a horizontal long axis. The existence and orienta-tion of the collateral sulcus were also recorded.

28


One experienced radiologist evaluated all the 318 examinations and made exclusions according to the criteria presented in the material descrip-��$� �� &�"�� !� �� analysis (63 examinations, 126 hippocampal re-gions) were then also evaluated by another expe-rienced radiologist independently. Both radiolo-gists made analyses blinded for gestational ages. Inter-observer agreement in evaluation of the hip-pocampal structure was 90% and in evaluation of the collateral sulcus 97%. In cases when there was disagreement, the radiologists made reevaluations and consensus was reached.

The study protocol was accepted by the local Ethical Committees in Sweden and in Portugal.

In Sweden, all mothers gave written informed consent before the examinations. In Portugal, the Ethical Committee approved that the clinical studies could be used in the project without asking mothers for consent.

Statistical analysis for all studiesThe statistical packages JMP (version 5.1) and StatsDirect 2.5.7 were used. Differences in pro-��!��%��!�� method and by Fisher’s (generalised) exact test when expected numbers were small. P values less ��$�}�� !��!� ��¡��}¢��-�!�� <�� !$� ^�� tailed.

29

Results

RESULTS

Morphological MR analysis in the �� (Paper I)A round or pyramidal hippocampal shape was ��!��¢��}¢��!��-terval (95% CI) 11.3–26.7%). In all these cases, the variables 2–7 (Table 1) were also deviating. There were no differences between the volunteers and the patients. The frequency was the same in men and women (Table 4).

All the unilateral malformations were found on the left side. Total IHI was reported in 9 of 13 left-sided IHI and the remaining 4 were partial. Among 6 bilateral IHI four were total bilaterally, one total on the left side, partial on the right and

��!��!��left. A vertical collateral sulcus was also found in 17/81 subjects (in 10/40 men and in 7/41 women) with normal oval hippocampal shape. The orien-tation could differ from the anterior to the poste-rior part. Defects in the corpus callosum or other developmental brain anomalies were not revealed in any of the subjects.

Two aspects of the IHI were analysed in paper two: occurrence and laterality and distribution of the total and partial IHI forms.

Table 4. Number of subjects with incomplete hippocampal inversion

MenN=50

WomenN=50

TotalN=100

Subjects with incomplete hippocampal inversion 10 9 19

Unilateral incomplete inversion (all on the left side) 9 4 13

Total 7 2 9Partial 2 2 4

Bilateral incomplete hippocampal inversion 1 5 6

Total bilaterally 1 3 4

Total on the left 1 1

Total on the right 1 1

30


Occurrence of the IHI in the non-epileptic and epileptic populations (Paper II)

Control groupIHI including a round or pyramidal hippocampal shape combined with presence of all other devi-ating variables previously named was found in 28/150 (19%) subjects (Table 5).

Epilepsy patient groups��'��-ly larger in patients with epilepsy than in controls (P<0.02) (Table 5). The proportion of the patients �� '�� $��>� ��!�� "��-�� _#�� $��>� �"��!� ��the controls. There were no differences in the age of onset or in therapy refractoriness between the groups having or not having IHI. In this group, Rolandic epilepsy differed, with a higher propor-��'��$��$��'��with generalised epilepsy in comparison to con-trols (P<0.01). Notably, the distribution of IHI among the subtypes of generalised epilepsies (2.1, 2.2 and 2.4) was not homogeneous (Fisher’s gen-��!� �&�� >� ��$��$� �� !�� high proportion of IHI among patients with cryp-togenic generalised epilepsies (code 2.2) and spe-��!��"��!��$��$�`��!�-pathic generalised epilepsy had an IHI. There was ��!��'��in patients with epilepsy of undetermined status ��"��!�>� ��$�>��£�$��!��¤��&��>��$�}}�$

Laterality and distribution of the total and partial forms of the IHI

Control groupOf the persons with IHI, 71% had an unilateral IHI and 29% had a bilateral IHI. All the unilat-eral IHIs (20 subjects) were found on the left side: 13/20 unilateral IHIs were of total type (65%) and 7/20 were partial (35%). Bilateral IHI, found in 8/150 of all control subjects, was total in 5/8 (63%) subjects and partial at least on one side in 3/8 (37%) (Table 6).

Epilepsy patient groupsAll epilepsy patientsThe distribution of the laterality of IHI did not dif-fer between controls and epilepsy patients (Fish-��¤��!��&��>��$}��"��-lepsy patient subgroups (codes 1.1, 1.2, 1.3, 2.1, 2.2 ��!��$�>��¤��!��&�� >��$�¥�$�In subgroups 2.2 (generalised epilepsy syndromes - cryptogenic) and 3 (undetermined), there was a ��>��bilateral IHI compared with both controls and oth-er epilepsy patients. The distribution of total and partial IHI did not differ in controls and epilepsy patients with the exception of cryptogenic gener-alised epilepsy, in which all bilateral IHIs (7/21) were total.Generalised cryptogenic epilepsyTwenty-one patients had generalised cryptogenic epilepsy syndrome, 12 of them had IHI. Of these, 57% (7/12) were bilateral. The frequency of bilat-�� '�� $�¥�� "��!� �� !!� �� $� ^�� -tients in this subgroup who had bilateral IHI had the total form (7/7). The frequency of total IHI was highest in this subgroup compared to both controls and other epilepsy subgroups. Cognitive defects were common both in patients with IHI (12/12) and without IHI (7/9). Therapy refractori-ness was similar in patients with IHI (9/12) and without IHI (6/9) (Table 6).

31

Results

Group

Controls

Epilepsy patients (all)

1 Focal epilepsy syndromes

1.1 Idiopathic

Rolandic

Other

1.2 Symptomatic

Temporal

Frontal

Other*

* = parietal lobe epilepsies, occipital lobe epilepsies, chronic progressive epilepsia partialis continua of childhood, syndromes characterized by seizures with specific modes of precipitation [23]

1.3 Cryptogenic

2 Generalised epilepsy syndrome

2.1 Idiopathic

2.2 Cryptogenic

2.4 Specific syndromes

3 Undetermined (all types)

Table 5. Occurrence of IHI in epilepsy syndromes

No. of subjects Subjects with IHI N (%) P value *

150 28 (19)

201 60 (30) 0.017

163 44 (27) 0.08

31 13 (42) 0.004

25 11 (44) 0.004

6 2 (33) 0.33

127 27 (21) 0.59

57 14 (25) 0.34

7 2 (29) 0.62

63 11 (18) 0.83

5 4 (80) 0.0062

33 13 (39) 0.009

10 0 0.212

21 12 (57) 0.0001

2 1 (50) 0.346

5 3 (60) 0.0549

Table 6. Distribution of the laterality of IHI in epilepsy syndromes

GroupSubjects with

IHI NSubjects with unilateral IHI – left N (%)

Subjects with bilateral IHI N (%)

Subjects with unilateral IHI – right N (%)

Controls 28 20 (71) 8 (29) 0

Epilepsy patients (all) 60 40 (67) 16 (27) 4 (7)

1 Focal epilepsy syndromes 44 33 (75) 7 (16) 4 (9)

1.1 Idiopathic 13 10 (77) 2 (15) 1 (8)

Rolandic 11 9 (82) 1 (9) 1 (9)

Other 2 1 (50) 1 (50) 0

1.2 Symptomatic 27 20 (74) 4 (15) 3 (11)

Temporal 14 11 (79) 1 (7) 2 (14)

Frontal 2 1 (50) 1 (50) 0

Other**

* = parietal lobe epilepsies, occipital lobe epilepsies, chronic progressive epilepsia partialis continua of childhood, syndromes characterized by seizures with specific modes of precipitation [23]

11 8 (78) 2 (18) 1 (9)

1.3 Cryptogenic 4 3 (75) 1 (25) 0

2 Generalised epilepsy syndrome 13 6 (46) 7 (54) 0

2.1 Idiopathic 0 0 0 0

2.2 Cryptogenic 12 5 (42) 7 (58) 0

2.4 Specific syndromes 1 1 (100) 0 0

3 Undetermined (all types) 3 1 (33) 2 (67) 0

32


Laterality of IHI in relation to EEG �� Laterality of the IHI was compared with focus laterality on EEG. Among 14 TLE patients with IHI, 11 had isolated left-sided, 2 had isolated right-sided and 1 had bilateral IHI. On EEG, the ��'��!��!��!��>�and both patients with IHI on the right side had an EEG focus on the left side. Among the remain-ing 11 patients with IHI on the left side, 8 had an EEG focus on the left side, 2 had a bilateral focus and data for 1 patient were missing. Among 43 TLE patients without IHI, 19 had an EEG focus on the left side, 9 on the right side, 2 bilaterally, and for the remaining 13, data were not complete. Occurrence of left-sided EEG focus was higher than right-sided in both groups: in 62% of patients having IHI and in 63% of patients without IHI.

Ultrasound study on preterm neonates (Paper III)��"��$}��"�$��-tribution of the hippocampal forms (complete versus incomplete inversed) at different ages is ��¥$��<��!�in IHI frequencies between GW 24 and 25. The patients are divided into three age groups accord-ing to the IHI frequency. The frequency of the IHI was the highest in the group I of infants examined at GWs 23-24 12 out of 24 (50%). In the group II GWs 25-28 frequency was 18 out of 72 (25%). In the group III GWs 29-36 frequency was 9 out of ��¢��!��frequency of the IHI between group I and group II ��$��!��!��$��>��!��$�¥}�$�

Also frequency of the bilateral IHI was higher in the youngest group: it was found in 7 of 24 children (29%) and in 11 of the other 136 children ��¢�$��!��$��>� ��9.08). Frequency of isolated left-sided IHI was 4 of 24 (17%) in 23-24 GW group, and 14 of 136 (10%) in the other premature newborns. This difference �� $� ��!� ��!�!� �'��rare: one in every group. In the total material, the difference between the frequencies of the isolat-

ed right sided IHI and isolated left sided IHI was ��$��>� ��$��$�

MRI study on fetuses (Paper IV)The three post mortem examinations were per-formed at GW 17-18 and the 60 in vivo examina-��\@��$�^��!��!��hippocampal sulcus bilaterally. The hippocampal sulcus was completely or widely open, bilater-ally or unilaterally, in 39/51 fetuses examined at GW 17-32 (Fig. 8). In 21/35 fetuses with a bilat-eral open hippocampal sulcus, the angle between the upper and lower lip, was less than 90 degrees ��!��$� ��<��>� �� than 90 degrees on the right side but more than 90 degrees on the left side. The hippocampal sul-cus was closed at GW 21 at the earliest (right side) and always bilaterally closed from GW 33 on-wards. Asymmetric development was observed in 11 fetuses in which both hippocampal sulci were open but the angles between the upper and lower lips were different on the right and left sides. The more closed hippocampal sulcus was on the right side in 8 fetuses and on the left side in 3 fetuses.

A closed hippocampal sulcus was found in 28 fetuses (Fig. 8). One fetus at GW 21 had a closed sulcus on the left side only. Three fetuses, at GW 27-30, had a closed right-sided sulcus but the left-sided sulcus was still open. From GW 33 onwards, all hippocampal sulci were closed.

The youngest fetus with an open hippocampal sulcus on one side and a closed sulcus on the oth-er side showed a non-ovoid shape of the inverted hippocampus but the three older fetuses, showed an ovoid shape with a horizontal long axis. The non-ovoid hippocampal shapes were found bilat-erally at GW 24-29 or unilaterally with the ovoid shape on the contralateral side at GW 31-36. From GW 33 onwards when all hippocampal sulci were closed, the hippocampal shape was non-ovoid on the left side and ovoid on the right side in 4/12 fetuses (33%).

Asymmetric hippocampal development was found in 26 (41%) fetuses when the non-ovoid hip-

33

Results

GW 23

GW 24

GW 25

GW 26

GW 27

GW 28

GW 29

GW 30

GW 31

GW 32

GW 33

GW 34-36

Bilat. IHI Right IHILeft IHI No IHI

3 5

7

1

1

1

4

2

2

2

2

2 18

61

1

1

1

4 15

1613

2 17

4

0 5 10 15 20 25

1

6

7

6

13

5

Figure 7. Occurrence of IHI in premature newborn according to gestation age at the examination

��"�� !� �� !�<��-oped than the ovoid shape. The development was faster on the right in 22 (85%) fetuses and on the left only in four (15%).

Development of the collateral sulcus is shown in Figure 9. A shallow collateral sulcus was de-tected at the earliest in one fetus at GW 17 but another fetus showed no collateral sulci at GW 29. The deep collateral sulcus was seen at the earliest at GW 26 on the right and at GW 27 on the left. At GW 31 or later, all 17 fetuses had developed a

deep collateral sulcus bilaterally. Fourteen of the 24 fetuses with a bilateral deep collateral sulcus had an uni- or bilateral vertical orientation (58%). The collateral sulcus was vertical in 5/26 fetuses (19%) on the right side and in 13/25 fetuses (52%) on the left side. This difference was statistically ��$��$�

Nine fetuses with a non-ovoid hippocampal shape had a deep collateral sulcus. In six of them, the nearby collateral sulcus was vertical.

34


R LR L

R LR L R L R L

R L

R L R LR L

R LR L R L R L

R L R L R L R L R L R L

17

1

4

32

567

8

18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

Gestation week

Number of fetuses

R LR L

R LR L R L R L

R L

R L R LR L

R LR L R L R L

R L R L R L R L R L R L

17

1

4

32

567

8

18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

Gestation week

Number of fetuses

Figure 8. Distribution of the hippocampal shape at different gestational ages. Each square pair describes a fetus. ��!�>�_��!�$ ��"��>��<�!��<��!��"��> ��<�!��<��!��"��>��&��*��"��<��$

Figure 9. Collateral sulcus at different gestational ages. Each square pair describes a fetus. ��!�>�_��!�$� Collateral sulcus: ��<��>��>��<��>��*��$

35

Discussion

DISCUSSION

Studies on the hippocampal developmentOur fetal MRI study was the second prenatal MR ��!�� !� �� "�� !� �� in which sulcal development in the hippocampal region and the shape of the inverted hippocampus has been evaluated and our ultrasound study of ��"��!��<��-ating hippocampal form using ultrasound.

In the fetal MRI study, we found that there are wide individual temporal variations in the deepen-ing and closure of the hippocampal sulcus. After the closure of the hippocampal sulcus, the invert-ed hippocampus could be of a non-ovoid shape (Fig. 10) or of an ovoid shape with a near hori-zontal long axis (Fig. 11). The last-named shape was presumed to present complete hippocampal inversion.

In all fetuses the hippocampal sulcus was iden-�� &��!� �� !�� the previously published data [28,45,46,65] (Ta-ble 7). According to the studies on the formalin-�&�!��>��"��!��<��!��adult appearance at GW 21 [2,3,35,45,46]. We could found a closed hippocampal sulcus only in one fetus at that age. Time variation for the clo-sure is very wide and the majority of the foetus-es had open hippocampal sulci. We found open hippocampal sulci up to GW 32. In cases with �� !� ��"�� >� �� !� ��shape of the inverted hippocampal formation us-ing the same criteria as in earlier studies on adults, children and premature neonates. The majority of the population have ovoid hippocampus with a

Figure 10. In utero MR study. T2-weighted single-shot FSE coronal images of fetus at 29 GW. A non-ovoid hippocampus with a vertical long axis bilater-ally with closed hippocampal sulcus. The arrows indicate the hippocampal region and the stars the collateral sulcus.

Figure 11. In utero MR study. T2-weighted single-shot FSE coronal images of fetus at 29 GW. An ovoid hip-pocampus with a closed hippocampal sulcus and a horizontal long axis bilaterally. The white arrows indicate the hippocampal region, the black closed hippocampal sulcus and the star the collateral sulcus.

36


horizontal long axis and that shape we categorize as ovoid. All the other shapes were categorized as non-ovoid. A non-ovoid hippocampus was found at GW 21 at the earliest and an ovoid hippocampus at GW 27. Bilateral non-ovoid hippocampi were not observed after GW 29 and bilateral ovoid not before GW 29.

In our ultrasound study of preterm neonates, �� !�� !� �� higher frequency of non-ovoid, incomplete in-versed, hippocampi before GW 25 and the lower frequency from GW 25 onwards. Of the neonates examined up to GW 24, about one half had In-complete hippocampal inversion (IHI). The IHI frequency in preterm neonates both at GWs 25–28 (24%) and at GWs 29–36 (14%) approached the IHI frequency (19%) in our MRI studies on chil-dren and adults.

The fetal MRI study did not reveal a threshold of the frequency of the non-ovoid shape at GW 25 as the ultrasound study had done. This discrep-

ancy may have been caused by the small numbers ��!��!�� "��!��$� �� !�study only the hippocampal shape was evaluated, the hippocampal sulci were not evaluated. In the fetal MRI study the hippocampal shape was only evaluated if the hippocampal sulcus was closed. The number of subjects was higher in the US study. In this study IHI frequency was 14-20% compared with 19% in a MR study of children and adults. These frequencies are best compara-ble to the frequency of the uni- or bilateral non-ovoid hippocampi in the fetuses in which both hippocampal sulci were closed (from GW 33 on-wards). That frequency was 33%. The differences between the frequencies are not statistically sig-��$�^��!��-tal MRI study of the material of 63 fetuses and in US study, in the material of 158 infants) the hip-pocampal inversion seems to continue longer than presumed on the basis of the earlier studies (only

Table 7. Detectability of some anatomical structures in the hippocampal region in different ages

Author Bajic et al. Bajic et al. Humphrey Kier et al. Chi et al. Garel et al.

Study method

Fetal MRI US of thepremature neonates

Autopsy material

MRI of the specimens

Photos of the brain (autopsy)

Fetal MRI(gyration

study)

Number of subjects

63 158 10 6 507 173

Age in GW

17 - 35 23 - 36 10 - 30 13 - 24 10 - 44 22 - 38

Collateral sulcus

17 24 23 26

Open HC sulcus

17 - 32 10 13 - 14 22 - 23

Non ovoid HC shape

21 23 15.5 - 18.5 18 - 20

Ovoid HC shape

27 23 20.5 21

37

Discussion

one hippocampus had ovoid shape before GW 29 and about one half had IHI.

In our fetal MRI study even the development of the collateral sulcus was analysed. The collat-eral sulcus showed different types of orientation. It can be vertical or horizontal which is the more common type in adults. In 58% of the fetuses with a deep bilateral sulcus, the orientation was ver-tical, uni- or bilaterally. The vertical orientation ��"��""�� !�$�Even the collateral sulcus developed earlier on the ��!�$��!��!��early as at GW 17 but was not visible in one fe-tus at GW 29. The orientation of the sulcus, hori-zontal or vertical, was not related to the gestation age. However, the vertical orientation seems to be slightly more common in fetuses than in children and adults. In children and adults, 36% had a ver-tically oriented collateral sulcus uni- or bilaterally. In fetuses with a deep bilateral sulcus, 58% showed that orientation. The difference reached statistical ��$��}�$�� !��-dren and adults, a non-ovoid hippocampal shape was always associated with a vertically oriented collateral sulcus. In 3/9 fetuses with a non-ovoid hippocampus and deep collateral sulcus, the near-by sulcus did not have a vertical orientation. This might be a sign that the hippocampus had not yet reached the ovoid shape but the inversion process was continuing.

In our fetal MRI study the number of fetuses was higher than in the earlier hippocampal studies ��"��&�!��!��<��show the wide time variation in the development of the hippocampal region. The timetable of the development of some anatomical structures in the hippocampal region presented in different studies, including ours was shown in Table 7.

In our fetal MRI study, the development of the hippocampus, including the hippocampal sulcus and the inverted hippocampal formation, was �� ""�� ¢�� !� �� majority of the asymmetric cases, the right side ��}¢��!�!�<��!��$�^��US study in cases of asymmetric shape of the hip-pocampus the left sided IHI was more common

which indicates the faster hippocampal develop-ment on the right side. Asymmetrical hippocam-pal development was frequent and evenly distrib-uted in different age groups, which were from 17 GW to 36 GW. In a prenatal MR study, where hippocampal infolding angles were measured at GW 20-37, no asymmetry was observed [66]. However, another group that used the same meth-odology in infants and children found a left-right asymmetry and concluded that it may result from slower development of the left hippocampus [57]. Chi and co-workers evaluated photographs of 507 ��"��&�!��%��age [21]. 207 brains were also sectioned. They de-scribed the development of many sulci separately but did not mention the hippocampal sulcus in the report. They found that the right cerebral hemi-sphere, in general, shows gyral complexity earlier than the left. Even neurophysiological develop-ment is slower in the left temporal lobe [62].

It is not easy to depict small hippocampal struc-tures neither on the fetal MRI nor on the US. Im-age quality and resolution were the best in the post mortem fetal MRI images (Fig. 12) in which there are no movement artefacts and the examination time is not restricted by the risk of movement. The hippocampal region was well assessable in all 12 post mortem examinations performed at GW 17-22, included at the start of the study, but only three could be accepted in the study owing to brain ab-normalities in the other cases. In the examinations in utero, the image quality was much lower (Fig. 13) and 59% of those examinations were rejected for technical reasons. The youngest living fetus with a resolution good enough was examined at GW 19. Many examinations were rejected as small details could not be assessable, which is a weak-ness of the study. A general weakness of both our fetal MRI study and fetal MRI studies of the for-"��&�!� ��"�� "�� "��the normal foetuses. An additional weakness of �� !��"��&�!��"��>�were possible changes of the anatomical shape and proportions as a result of the removal of the brains from the skull, placement in the vessels for ��"�� &�� !� ��

38


for cutting or imaging. In our ultrasound study as well as the fetal MRI study the hippocampus was examined in a natural position and shape.

Imaging techniques progress continuously and examination times will be shorter. Images with a very high resolution can be obtained even now at ��!��|¥¥~��limiting factor, in particular, in fetuses which are more vulnerable than adults. Repeated examina-tions would be the best way to demonstrate devel-opmental progress but fetal examinations without clinical indications are not ethically acceptable.

The hippocampal shape was well seen in neo-nates also with US, especially with high frequency transducers (10 MHz) but the occurrence of IHI at GWs 29-36 was a little lower than in children and adults in our MR studies. The difference was not ��"��=�� "��of the partial IHIs might be missed in US exami-nations because brain morphology is not depicted as well with US as with MRI.

Hippocampal development during the late ges-tational period was analysed using cranial ultra-sound on preterm neonates. The hippocampal development in preterm neonates may be af-fected by pre- and postnatal stress. This problem exists in all studies on fetal hippocampal devel-opment. Developmental milestones have been based on histological and MR studies on forma-��&�!� ��"�� %�� "� ��!� ��[2,3,35,45,46,65]. The etiologies of abortions were not reported. There are no developmental criteria for normal hippocampi in utero because even pre-natal MR studies were performed for medical in-dications. We have documented the hippocampal shape in preterm newborns and tried to minimize all those factors possibly affecting hippocampal development. Children and adolescents born pre-term have been reported to have smaller brain vol-umes than those born full-term. The volume re-duction has been proportionally great particularly in the hippocampi [36,56,63]. In contrast, in more

Figure 12. Post mortem MR study of a fetus aborted at 17 GW. The arrows indicates widely open symmetri-cal hippocampal sulci. T2-weighted TSE coronal image.

Figure 13. In utero MR study. T2-weighted single-shot FSE coronal images of fetus at 25 GW. The arrows indicate the hippocampal sulci. The hippocampal sulcus is larger on the left side.

39

Discussion

recent samples without major perinatal complica-tions, the hippocampi have shown a marginal, if any, volume reduction [44,50,52,78]. Intrauterine growth restriction affects the hippocampal vol-ume [50]. A generally small brain volume and cognitive dysfunction consistent with hippocam-pal injury has been found in adolescents born at full-term but SGA [52]. Our study excluded neo-nates SGA. However, there might have been some unknown factors affecting prenatal hippocampal development because there is always a cause for a preterm birth. The time intervals between the birth and the US examination were in our study, most likely, too short to allow demonstrable post-natal developmental deviations.

Studies focusing on the morphological variants of the hippocampus and their relation to some epilepsy syndromesIn our MRI study, including 100 subjects without obvious developmental anomalies and without ��>� �� !� �� &�� !��-tion of the IHI as a developmental hippocampal anomaly where the infolding of the hippocampus in the temporal lobe was incomplete and the hip-pocampal form in the MRI coronal slices found to be round or pyramidal. We found that there is a frequent (19%, 95% CI 11.3–26.7%) hippocam-pal form variation in the common population, and that this form variation could be total, if the hip-pocampal shape irregularity is present in all slices on the coronal MR images of the hippocampal re-gion; or partial if the shape irregularities are pre-sent only on some slices. IHI can be bilateral or unilateral most often on the left side.

There are a few studies regarding IHI in sei-zure free subjects. One of them is Gamss study which was published after our papers I and II [28]. They found rounded hippocampal form and verti-cal collateral sulcus in 7/497 (1.4%) patients. The frequency was lower than in the nonepileptic con-trols in the epilepsy studies of Peltier (2/50, 4%) [61] or Bernasconi (5/50, 10%) [10]. Bronen and Cheung [16,17] reported that 12/58 hippocampi (20.8%) had a shape deviating from the normal

in healthy volunteers. In our study, 25/200 hip-pocampus (12.5%) were incompletely inverted. The frequency of IHI was higher in our subjects without epilepsy than many other authors have reported in epilepsy patients without corpus cal-losum anomalies. The great variations in fre-�� "�� !��!� �� !�� !�� incomplete hippocampal inversion, i.e. including or excluding of partial incomplete inversions, and also on genetic and environmental factors. All incomplete inversions of the hippocampus in our study were bilateral or situated on the left side. In our fetal MRI and US studies, the majority of IHI was also bilateral or on the left side. In the litera-ture, the laterality has not always been mentioned but in the Gamss study, all of the seven deviating hippocampal forms were on the left side [28].

We found that IHI is more common in epilep-sy patients than in the control subjects group (19 resp. 30%). The frequency of IHI in our epilepsy patients was higher than in many other studies on epilepsy patients without major brain anomalies. Barsi et al. [6] described IHI in 6% of 527 patients with suspected epilepsy, Peltier et al. [23] in 14% of 97 epilepsy patients. Baulac et al. [7] found ab-normal hippocampal form in 1.7% (19/1,100) of patients with partial epilepsy. Thirteen of those 19 patients had isolated hippocampal abnormalities (2 bilateral, 11 unilateral). In TLE, the frequency of IHI ranged from 2.2% (5/222 patients) [10,61] to 43% (13/30 patients) but in the last-named study, 15/30 patients had previously known unilateral hippocampal atrophy. In our study, patients with MTS were excluded because of shrunken hip-pocampi are and the original form may not there-fore be assessable.

The frequency of IHI was not statistically signif-icantly higher in focal epilepsies than in controls ��¥�<�$��¢>��$��$�`��>��only 25% in TLE. Rolandic epilepsy differs from the other types of focal epilepsies by having a ��'��¢>� ��$�� $� �� !�<��"��-tal deviation of the hippocampus in Rolandic epi-lepsy, although common, may be of milder type than in the other epilepsy syndromes. The group

40


with the most frequent occurrence of IHI (57%) was that of patients suffering from generalized cryptogenic epilepsy. Compared to the controls, the occurrence was almost three times higher and reached the same occurrence of IHI as reported in patients with severe developmental anomalies [4,69]. The IHI frequency was high also in focal cryptogenic epilepsy (4/5 patients), but this sub-group was very small.

Isolated left-sided IHI is far more common than bilateral IHI. This distribution pattern was found both in controls and in epilepsy patients, except for those with generalized cryptogenic epileptic syn-dromes where 58% of IHIs were bilateral. Isolated right-sided IHI seems to be a very rare entity, and ��!!��!� �� "�� and in only four epilepsy patients. Laterality has not always been mentioned in articles describing hippocampal morphological changes including IHI. In literature, isolated rightsided IHI has only been reported in one patient with partial epilepsy [7]. In cases with persistent left-sided IHI, cere-bral development may have been disturbed when the right hippocampus had completed inversion process but the left side had not yet reached that developmental stage.

In TLE patients, there was no statistically signif-icant difference in frequency, laterality or distri-bution of total and partial forms of IHI compared with controls. The laterality of the EEG focus did ��'�$�@��!��!��-sality between the focus in TLE and IHI.

Cryptogenic epilepsy refers to a condition for which the nature of the underlying cause is un-

%��>� ��!�� _^#� �� [23]. We found a high frequency of IHI in this het-erogeneous group of patients. In addition, many ��!��'�$��that such a high frequency of morphological brain changes has been detected in epilepsies previously ��!��$��has yet to be elucidated.

Studies including patients with severe develop-mental brain anomalies present much higher fre-quencies of IHI. Sato and coworkers [69] found in-complete inversion in 64% (28/44) and Baker and Barkovich [4] in 82% of the patients diagnosed with congenital brain malformations such as cor-pus callosum agenesis, lissencephaly, or holopros-encephaly.

�� '�� such is still unknown. A recent study indicates that IHI does not have a detrimental effect on memory functioning but is associated with dys-function of other areas [74]. That also supports the hypothesis that persistent IHI may be a sign of a disturbance in brain development with clini-cal consequences elsewhere in the brain, and new, both clinical and imaging, follow-up studies could be very informative.

IHI cannot be the etiology of epilepsy as such but could be a sign of general failure in brain de-velopment, which might be responsible, both for incomplete inversion of the hippocampus and fo-cal or multifocal pathological changes in the cer-ebral cortex.

41

Conclusions

CONCLUSIONSIHI is a common morphologic variation. Leftsided IHI is more common than bilateral, and isolated right-sided incomplete hippocampal inversion seems to be very rare.

The frequency of IHI is higher in the epilepsy population in comparison to the non-epileptic controls, mainly due to a high proportion of IHI in generalised cryptogenic epilepsy syndromes. IHI is not more frequent in TLE than in nonepileptic controls and no causality between TLE and IHI seems to exist. IHI, a morphological variant of the hippocampus, is not an aetiological factor in epi-lepsy but can be a sign of disturbed cerebral de-velopment that may affect other parts of the brain leading to epilepsy.

Hippocampal inversion was not completed up to GW 24 in one half of the preterm neonates. From the GW 25 onwards, the frequency of IHI is simi-lar to that in the general adult population, at ap-proximately the 20% level. Hippocampal inver-sion seems to continue longer than presumed on the basis of the earlier studies in limited samples. The higher IHI frequency on the left than right side, demonstrated in earlier studies in children and adults, exists even in preterm neonates.

In our fetal MRI study, there were a higher number of fetuses than in the earlier hippocampal studies ��"��&�!��!��<��show the wide time variation in the development of the hippocampal region and also that the right side often develops faster.

@��!��"��"��"��malrotation. In particular, we thought that incom-plete hippocampal inversion is a better descrip-tion of this condition because the hippocampus is not “malrotated” but rather probably the inversion was never completed.

In our limited case material we cannot prove that the non-ovoid shape precedes the ovoid shape but our data does not contradict that hypothesis.

42


SUMMARY IN SWEDISHDetta avhandlingsarbete avbildar utvecklingen av hippocampal regionerna i perioden från 17 - 36 gestationsveckor (GV) med hjälp av prenatal MR-undersökning av foster, samt i perioden mellan 23 och 35 GV med hjälp av kranialt ultraljud på pre-matura nyfödda barn. Dessutom har vi genom att analysera MR hjärna av 300 epilepsi patienter och 150 epilepsifria subjekt, evaluerat förekomst av icke färdigutvecklade hippocampala former och deras relation till olika epilepsisyndromen.

I delarbete I har vi analyserat förekomsten av ”In-complete hippocampal inversion” (IHI) som är en hippocampal formvariant, bland den ”vanliga” populationen, dvs. bland 100 friska frivilliga och patienter utan hjärnanomalier, tumörer, epilepsi, hydrocephalus och vilket tillstånd som helst som kunde påverka temporala loben. Vi har studerat MR hjärna i det coronara planet och funnit att 19% av de undersökta hade IHI, bilateralt eller på vänster sida. Samtliga med IHI hade också verti-kal kollateral sulcus på IHI sida. Detta visar att IHI inte alls är en ovanlig hippocampusform.

I delarbete II har vi analyserat förekomsten av IHI hos epilepsipatienter. Kontrollgruppen var 100 personer från arbete I plus 50 nya utvalda med samma kriterier. Epilepsigruppen hade i början 300 patienter och efter exkluderingen av patien-ter med hjärnanomalier, tumörer, hydrocephalus och patienter med inkompletta data blev det 201 kvar. 30% av dem hade IHI, och i förhållande till %�� ¦<��!�� %�� %��-nad. Hög frekvens var bland patienter med kryp-

togenisk epilepsi och Rolandisk epilepsi, men inte bland dem med temporallobsepilepsi. Det var ingen korrelation med epilepsifokus vid EEG och IHI. Så trots att förekomsten av IHI var hög i vis-sa grupper, kunde inte IHI vara orsak till epilepsi utan snarare visat att det någonstans i hjärnan har skett någon förändring som kunde vara orsak till epilepsi.

I delarbete III har vi studerat normal hippocam-pusutveckling genom att analysera coronara bil-der på ultraljud hjärna. Efter att samtliga prematu-��"�!��%��§��<��&%��!��!�>�analyserades bilder av 158 nyfödda i gestations-ålder 23 till 35 gestationsveckor (GV). Vi har stu-!��"��"�"�!�"¦��!��när hippocampus är fullt utvecklad. Cirka 50% av prematura har inte fullt utvecklad hippocampus före GV 25. Efter GV 25 var frekvensen av IHI, som vi har tolkat som omogen hippocampal form, lika som i barn och vuxna i arbete I och II.

I manuskriptet under bearbetning som är delarbe-te IV har vi analyserat morfologisk utveckling av hippocampus under fetalperioden, genom analys av 60 fetala MR-undersökningar och 3 postmor-tala MR av aborterade foster. Ingen av dem hade hjärnpatologi. Vi har analyserat hippocampus-området under gestationsveckorna 17-36. Det var stora individuella variationer i utvecklingen av både hippocampus och collateral sulcus. Ofta sker utvecklingen asymmetriskt, och det är höger sida som utvecklas snabbare.

43

Summary

SUMMARY IN SERBIANOva disertacija odslikava razvoj hipokampusa i susjednih anatomskih struktura u periodu od 17-36 gestacione nedjelje koristeci prenatalni pregled fetusa magnetnom rezonancom (MR), kao i u pe-riodu od 23-35 gestacione nedjelje koristeci ultraz-vucni pregled mozga prerano rodjene novorodjen-cadi. Pored toga analizirajuci MR snimke mozga kod 300 pacijenata koji boluju od epilepsije i 150 subjekata koji nemaju epilepsiju diskutovali smo ucestalost nepotpuno razvijenih formi hipokam-pusa i njihov odnos sa pojedinim epileptickim sin-dromima.

U prvoj od cetiri studije u ovoj disertaciji ana-lizirali smo ucestalost nekompletne hipokampalne inverzije (NHI) koja predstavlja jednu varijantu oblika hipokampusa, medju ”normalnom” popu-lacijom, sacinjenom od 100 subjekata, zdravih do-brovoljaca i pacijenata bez anomalija mozga, tu-mora, hidrocefalusa, epilepsije ili bilo kog drugog patoloskog stanja koje moze uticati na temporalni rezanj mozga. Proucavali smo koronarne MR sekvence i nasli da 19% av ispitanika ima zna-kove NHI, obostrano ili na lijevoj strani. Kod svih koji su imali NHI, vertikalni kolateralni sulkus je takodje postojao na istoj strani na kojoj je i NHI.

Druga studija obuhvata analizu ucestalosti NHI kod oboljelih od epilepsije. Kontrolnu grupu su sacinjavali subjekti iz prvog dijela zajedno sa 50 novih izabranih na osnovu istih kriterija kao za prvu studiju. Grupu oboljelih od epilepsije sacin-javalo je u pocetku 300 pacijenata ali poslije isklju-cenja iz studije pacijenata sa anomalijama mozga, tumorima, hidrocefalusom, kao i onih sa nepot-punim podacima u istoriji bolesti, u studiji je os-tao 201 pacijent. 30% od njih je imalo NHI, sto je predstavljalo statisticki znacajnu razliku u odnosu

na kontrolnu grupu. Visoka ucestalost je nadjena kod pacijenata sa kriptogenom epilepsijom kao i kod pacijenata sa rolandickom epilepsijom, ali ne i medju pacijentima sa epilepsijom temporalnog reznja. Korelacija izmedju epileptickog fokusa na EEG i NHI nije nadjena. Iako je ucestalost NHI u pojedinim grupama visoka NHI se ne moze sma-trati direktnim uzrocnikom epilepsije, nego je pri-je indikator da se tokom razvoja negdje u mozgu desila promjena koja je uzrocnik epilepsije.

Treca studija se bavi normalnim razvojem hipokampusa kroz analizu koronarnih presjeka ultrazvuka mozga kod 158 prerano rodjene no-vorodjencadi u starosti od 23 do 35 gestacionih nedelja. Posto su iskljucana novorodjencad sa sig-��%��"� ��"� "�*��>� ��!�� "��oblik hipokampusa sa ciljem da utvrdimo vrijeme kada je hipokampus potpuno razvijen. Oko 50% prerano rodjene novorodjencadi nema potpuno razvijen hipokampus prije 25 gestacione nedelje, ali je poslije tog perioda procenat novorodjencadi sa NHI, hipokampusnom formom koju smatramo nezrelom, isti kao kod kontrolne grupe u studiji I i II.

Cetvrta studija sa manuskriptom u obradi pred-stavlja analizu morfoloskog razvoja hipokam-pusa za vrijeme fetalnog perioda. Izucavali smo hipokampalni region kod 60 prenatalnih MR fetu-sa i 3 postmortalna MR abortiranih fetusa, gesta-cione starosti 17-36 nedelja. Nijedan nije imao pa-toloske promjene na mozgu. Uocili smo znacajne individualne varijacije u razvoju kako hipokam-pusa tako i kolateralnog sulkusa. Cesto razvoj ide asimetricno i pri tome je desna strana ta koja se brze razvija.

44


45

Acknowledgements

ACKNOWLEDGEMENTSThis thesis is not the work of one person, it has been produced by many sup-porting people and I would like to express my sincere gratitude to everyone who helped me to complete this work:

Professor Raili Raininko, my supervisor, for introducing me to research, all positive advice, patience, constant support and kindness.

My associate supervisors Eva Kumlien and Peter Mattsson for their enthusi-��">��!��"��$��%��!�help in clinical analyze of the patients, kind advices and help in statistical evaluation of the studies.

My co-authors Staffan Lundberg, Chen Wang, Professor Orvar Eeg-Olofsson, Professor Uwe Ewald, and Johan Wikström for generous sharing their knowl-edge and support.

My co-author and Portuguese colleague Nuno Canto Moreira for collecting and generous sharing of his Portuguese patients data.

Håkan Pettersson and Nora Velastegui for their technical support, kindness and patience.

Christl Richter-Frohm for fantastic secretary help.

Peter Pech, Torsten Lönnerholm, Marie Raiend and Eva Penno my colleagues on the Department for paediatric radiology for support during this project and friendship.

Ognjen Gasic for assistance during the patients selection.

All my colleagues and personal at the Department of radiology for helping me in various ways.

46


47

References

REFERENCES1. Almog B, Gamzu R, Achiron R, Fainaru O, Za-

lel Y. (2003). Fetal lateral ventricular width: what should be its upper limit? A prospective cohort study and reanalysis of the current and previous data. J Ultrasound Med. 22:39-43.

2. Arnold SE, Trojanowski QJ. (1996). Human fetal hippocampal development: I. Cytoarchitecture, myeloarchitecture and neuronal morphologic features. J Comp Neurol. 367:274-292.

3. Arnold SE, Trojanowski QJ. (1996). Human fetal hippocampal development: II. The neuronal cy-toskeleton. J Comp Neurol. 367:293-307.

4. Baker LL, Barkovich AJ. (1992). The large tem-poral horn: MR analysis in developmental brain anomalies versus hydrocephalus. AJNR Am J Neuroradiol. 13:115-122.

5. Barry GF, Judith A. (1989). Whitworth Diction-ary of Medical Eponyms, new edn in 2002. The Parthenon Publishing Group.

6. Barsi P, Kenez J, Solymosi D, Kulin A, Halasz P, Rasonyi G, Janszky J, Kaloczkai A, Barcs G, Neuwirth M, Paraicz E, Siegler Z, Morvai M, Jerney J, Kassay M, Altman A. (2000). Hippo-campal malrotation with normal corpus callo-sum: a new entity? Neuroradiology. 42:339-345.

7. Baulac M, De Grissac N, Hasboun D, Oppen-heim C, Adam C, Arzimanoglou A, Semah F, Lehericy S, Clemenceau S, Berger B. (1998). Hippocampal developmental changes in patients with partial epilepsy: Magnetic resonance imag-ing and clinical aspects. Ann Neurol. 44:223-233.

8. Bear M. (2006). Neuroscience exploring the Brain, 3rd edn. Lippincott Williams & Wilkins, (Chapter I Neuroscience: Past, present and future).

9. Bechterew W. (1900). Demonstration eines Ge-hirns mit Zerstoring der vorderen und inneren Teile der Hirnrinde bei der Schläfelappen. Neu-rol Zbl. 19:990-991.

10. Bernasconi N, Kinay D, Andermann F, Antel S, Bernasconi A. (2006). Analysis of shape and positioning of the hippocampal formation: an MRI study in patients with partial epilepsy and healthy controls. Brain. 128: 2442-2452

11. Best PJ, White AM. (1999). Placing hippocampal single-unit studies in a historical context. Hippo-campus. 9:346-351.

12. Blackstad TW. (1956). Commissural connections of the hippocampal region in the rat, with special reference to their mode of termination. Journal of Comparative Neurology. 105:417-528.

13. Blackstad T.W. (1958). On the termination of some afferents to the hippocampus and fascia dentata. Acta Anatomica 35:202-214.

14. Brodal A. (1947). The hippocampus and the sense of smell. A review. Brain 70:179–222.

15. Brodal A. (1982). Neurological Anatomy. New York, Oxford Univ. Press.

16. Bronen RA, Cheung G. (1991). MRI of the normal hippocampus. Magn Reson Imaging. 9:497-500.

17. Bronen RA, Cheung G. (1991). MRI of the tem-poral lobe: Normal variations, with special ref-erence toward epilepsy. Magn Reson Imaging. 9:501-507.

18. Bush G, Luu P, Posner MI. (2000). Cognitive and �"��=��&$�Trends Cogn Sci. 4:215-222.

19. Cajal SRY. (1995). Vols I & II Histology of the Nervous System of Man and Vertebrates. Trans-lated (from the French) by Neely and Larry Swanson. New York, NY: Oxford University Press, Oxford.

20. Campbell S, Macqueen G. (2004). The role of the hippocampus in the pathophysiology of major depression. J Psychiatry Neurosci. 29:417-26.

21. Chi JG, Dooling EC, Gilles FH. (1977). Gyral development of the human brain. Ann.neurol. 1:86-93.

22. Clifford RJJ. (1994). MRI based Hippocam-pal volume measurement in epilepsy Epilepsia. 35:S21-S29.

��$� Q�""��Q��!��"��of the International League Against Epilepsy. ��$� �� <��!� Q�� Epilepsies and Epileptic syndromes. Epilepsia 30:389-399.

48


24. Farrell TA, Hertzberg BS, Kliewer MA, Harris L, Paine SS. (1994). Fetal lateral ventricles: reas-sessment of normal values for atrial diameter at US. Radiology 193:409-411.

25. Fatterpekar GM, Naidich TP, Delman BN, Agui-naldo JG, Gultekin SH, Sherwood CC, Hof PR, Drayer BP, Fayad ZA. (2002). Cytoarchitecture of the human cerebral cortex: MR microscopy of excised specimens at 9.4 Tesla. AJNR Am J Neu-roradiol. 23:1313-1321.

26. Fisher RS, Van Emde Boas W, Blume W, Elger C, Genton P, Lee P, Engel J Jr. (2005). Epileptic Sei-*��!�#��!��International League Against Epilepsy (ILAE) and the International Bureau for Epilepsy (IBE). Special Article Epilepsia, 46:470–472.

27. Fishman RS. (2007). The Nobel Prize of 1906. Arch Ophthalmol. 125:690-694.

28. Gamss RP, Slasky SE, Bello JA, Miller TS, Shin-nar S. (2009). Prevalence of hippocampal mal-rotation in a population whiteout seizure. AJNR Am J Neuroradiol. 30:1571-1573.

29. Garel C, Chantrel E, Elmaleh M, Brisse H, Sebag G. (2003). Fetal MRI: normal gestational land-marks for cerebral biometry, gyration and my-elination. Childs Nerv Syst. 19:422-425.

30. Green, JD, Arduini A. (1954). “Hippocampal ac-tivity in arousal”. J Neurophysiol 17:533–57.

��$� \��>� ��<��«>�� «#>��\>�Russell SA, Paley MNJ, Whitby EH. (2010). A prospective study of fetuses with isolated ven-triculomegaly investigated by antenatal sonogra-phy and in utero MR imaging. AJNR Am J Neu-roradiol. 31:106-111.

32. Hampel H, Bürger K, Teipel SJ, Bokde AL, Zetterberg H, Blennow K. (2008). “Core candi-date neurochemical and imaging biomarkers of Alzheimer’s disease”. Alzheimers Dement. 4: 38–48.

33. Harrison PJ. (2004). The hippocampus in schizo-phrenia: a review of the neuropathological evi-dence and its pathophysiological implications. Psychopharmacology (Berl). 174:151-162.

34. Hevner FR, Kinney CH. (1996). Reciprocal en-torhinal-hippocampal connections established by human fetal midgestation. J Comp Neurol 372: 384-394.

35. Humphrey T. (1967). The development of the hu-"��"��«$�^��$��}}��¥�$

36. Isaacs EB, LucasA CWK, Wood SJ, Johnson CL, Marshall C, Vargha-Khadem F, Gabisn DG. (2000). Hippocampal volume and everyday memory in children of very low birth weight. Pe-diatric Res. 47:713–720.

37. Isaacson RL, Kimble DP. (1972). Lesions of the limbic system: Their effects upon hypotheses and frustration. Behavioral Biology. 7:767-793.

38. Jackson JH, Beevor CE. (1889). Case of tumour of the right temporosphenoidal lobe bearing on the localisation of the sense of smell and on the interpretation of a particular variety of epilepsy Brain. 12:346-349.

39. Jay V. (2002). Pierre Paul Broca. Arch Pathol Lab Med 126:250-251.

40. Jinkins R. (2000). Atlas of neuroradiologic Em-bryology, Anatomy and variants p. 122 LWW.

41. Jones EG. (1999). Golgi Cajal and the Neuron Doctrine. Journal of the History of the Neurosci-ences. 8:170-178.

42. Jung R, Kornmuller AE. (1938). Eine methodik der ableitung lokalisierter potential schwankin-gen aus subcorticalen hirnyebieten. Arch Psy-chiat Neruenkr. 109:1-30.

43. Kadar T, Arbel I, Silbermann M, Levy A. (1994). Morphological hippocampal changes during nor-mal aging and their relation to cognitive deterio-ration J Neural Transm Suppl. 44:133-43.

44. Kesler S, Reiss AL, Vohr B, Watson C, Schnei-der KC, Katz KH, Maller-Kesselman J, Silbereis J, Constable RT, Makuch RW, Ment LR. (2008). Brain volume reductions within multiple cog-nitive systems in male preterm children at age twelve. J Pediatr. 152:513–520.

45. Kier EL, Fulbright RK, Bronen RA. (1995). Limbic lobe embryology and anatomy: Dissec-tion and MR of the medial surface of the fetal cerebral hemisphere. AJNR Am J Neuroradiol. 16:1847-1853.

46. Kier EL, Kim JH, Fulbright RK, Bronen RA. (1997). Embryology of the human fetal hippo-campus: MR imaging, anatomy, and histology. AJNR Am J Neuroradiol. 18:525-532.

49

References

47. Kruger L. Woolsey TA. (1990). Rafael Lorente de Nó: 1902-1990 Obituary. The Journal of Com-parative Neurology. 300:1-4.

48. Kuruba R, Hattiangady B, Shetty AK. (2009). Hippocampal neurogenesis and neural stem cells in temporal lobe epilepsy. Epilepsy Behav. 1:65-73.

49. Kuzniecky R, Jackson GD. (1995). Magnetic Resonance in Epilepsy. Raven Press chapter I.

}�$� _�!��%��\^>��_>�@��!��¬>�+��-radori Tolsa C, Sizonenko S, Lazeyras F, Hüppi PS. (2008). Intrauterine growth restriction af-fects the preterm infant’s hippocampus. Pediatric Res. 63:438–443.

51. Mark LP, Daniels DL, Naidich TP, Hendrix LE. (1995) Limbic connections. AJNR Am J Neuro-radiol. 16:1303-1306.

52. Martinussen M, Flanders DW, Fischl B, Busa E, Lø haugen GC, Skranes J, Vangberg TR, Brubakk AM, Haraldseth O, Dale AM. (2009). Segmental brain volumes and cognitive and perceptual cor-relates in 15-year-old adolescents with low birth weight. J Pediatr. 155:848–853.

53. McEwen BS. (1999). Stress and hippocampal plasticity. Annu Rev Neurosci. 22:105-122.

54. Montenegro MA, Kinay D, Cendes F, Bernasconi A, Coan AC, Li LM, Guerreiro MM, Guerreiro CAM, Lopes-Cendes I, Andermann E, Dubeau F, Andermann F. (2006). Patterns of hippocam-pal abnormalities in malformations of cortical development. J Neurol Neurosurg Psychiatry. 77:367-371.

55. Nakano I. (1998). The limbic system: An outline and brief history of its concept. Neuropathology Symposium 18:211-214.

56. Nosarti C, Al-Asady MHS, Frangon S, Stewart AL, Rifkin L, Murray RM. (2002). Adolescents who were born very preterm have decreased brain volumes. Brain 125:1616–1623.

57. Okada Y, Kato T, Iwai K, Iwasaki N, Ohto T, Matsui A. (2003). Evaluation of hippocampal in normal infolding using magnetic resonance im-aging. Neuroreport. 14:1405-1409.

58. O’Keefe J, Nadel L. (1978). The Hippocampus as a Cognitive Map, Oxford: Oxford U.P.

59. Papez JW. (1937). A proposed mechanism of emotion. Arch Neurol psychiat. 38:725-743.

60. Parkinson J. (1817). “An essay on the shaking pal-sy”. The Journal of Neuropsychiatry and Clinical Neurosciences. 2002 (2): 223–236.

61. Peltier B, Hurtevent P, Trehan G, Derambure P, Pruvo J-P, Soto-Ares G. (2005). IRM des malfor-mations de l`hippocampe dans l`epilepsie tempo-rale refractaire. J Radiol. 86:69-75.

62. Petersen I, Eeg-Olofsson O. (1971) The develop-ment of the electroencephalogram children from the age of 1 through 15 years. Non-paroxysmal activity. Neuropediatrics. 2:247–304.

63. Peterson BS, Vohr B, Staib LH, Cannistraci CJ, Dolberg A, Schneider KC, Katz KH, Westerveld M, Sparrow S, Anderson AW, Duncan CC, Makuch RW, Gore JC, Ment LR. (2000). Re-gional brain volume abnormalities and long-term cognitive outcome in preterm infants. JAMA 284:1939–1947.

64. Prull MW, Gabrieli JDE, Bunge SA. (2000). “Ch 2. Age-related changes in memory: A cognitive neuroscience perspective”. in Craik FIM, Salt-house TA. The handbook of aging and cognition. Erlbaum.

65. Rados M, Judas M, Kostovic I. (2006). In vitro MRI of brain development. Eur J Radiology. 57: 187-198.

66. Righini A, Zirpoli S, Parazzini C, Bianchini E, Scifo P, Sala C, Triulzi F. (2006) Hippocampal infolding angle changes during brain develop-ment assessed by prenatal MR imaging. AJNR Am J Neuroradiol. 27:2093-2097.

67. Robinson V. (editor). (1939). ”Bell’s Law, within the description of Bell’s Palsy, including a brief discussion about Charles Bell, 1774-1842”. The Modern Home Physician, A New Encyclopedia of Medical Knowledge. WM. H. Wise & Com-pany (New York), page 92.

68. Sasaki M, Sone M, Ehara S, Tamakawa Y. (1993). Hippocampal sulcus remnant: potential cause of change in signal intensity in the hippocampus. Radiology. 188:743-746.

69. Sato N, Hatakeyama S, Shimizu N, Hikima A, Aoki J, Endo K. (2001). MR evaluation of the hippocampus in patients with congenital malfor-mations of the brain. AJNR Am J Neuroradiol. 22:387-936.

50


70. Scoville WB, Milner B. (1957). Loss of recent memory after bilateral hippocampal lesions. Journal of Neurology, Neurosurgery and Psychi-atry 20:11-21.

71. Small S, Stern Y, Tang M, Mayeux R. (1999). Selective decline in memory function among healthy elderly Neurology. 52: 1392–1396.

72. Smith DM, Mizumori SJY. (2006). Hippocampal Place Cells, Context and Episodic Memory. Hip-pocampus. 16:716-729.

73. Starkman NM, Gebarski SS, Berent S, Schtein-gart DE. (1992). Hippocampal formation volume, memory dysfunction, and cortisol levels in pa-tients with Cushing’s syndrome. Biological Psy-chiatry 32: 756-765

74. Stiers P, Fonteyne A, Wouters H, D’Agostino A, Sunaert S, Lagae L. (2010). Hippocampal malro-tation in pediatric patients with epilepsy associ-ated with complex prefrontal dysfunction. Epi-lepsia 51:546-555.

¥}$� "®�\>�¬��<®��>�@��!�+>�+��!��<®�`$�(1997). Volume and number of neurons of the hu-man hippocampal formation in normal aging and Alzheimer’s disease. The Journal of Compara-tive Neurology. 379: 482–494.

76. Talamini LM, Meeter M, Elvevåg B, Murre JM, Goldberg TE. (2005). Reduced parahippocampal connectivity produces schizophrenia-like mem-��!��"��!��-duced parahippocampal connectivity. Arch Gen Psychiatry. 62:485-493.

77. Theysohn JM, Kraff O, Maderwald S, Schla-mann MU, de Greiff A, Forsting M, Ladd SC, Ladd ME, Gizewski ER. (2009). The human hip-pocampus at 7 T – In vivo MRI. Hippocampus 19:1-7.

¥�$� ��"�� ¬>� @��!� �«>� �� _@>� @��!�SK, Lodygensky GA, Anderson PJ, Egan GF, In-der TE. (2008). Neonate hippocampal volumes: prematurity, perinatal predictors, and 2-year out-come. Ann Neurol. 63:642–651.

79. Tudor M, Tudor L, Tudor KI. (2005). “[Hans Berger (1873-1941) - the history of electroen-cephalography]”, Acta medica Croatica 59 (4): 307–13.

80. Tulving E. (1972). Episodic and semantic mem-ory. In E. Tulving and W. Donaldson (Eds.), Or-ganization of Memory (pp. 381-402). New York: Academic Press.

81. Vallejo-Manzur F. (2003). The resuscitation greats. Andreas Vesalius, the concept of an arti-��$� ��}��¥$

82. Wyldes M, Watkinson M. (2004). Isolated mild fetal ventriculomegaly. Arch Dis Child Fetal Neonatal Ed. 89:F9–F13.

83. Yakovlev PI. (1948). Motility, behavior and the brain: Stereodynamic organization and neural coordinates of behavior. J Nerv Ment Dis. 107: 313-335.

84. Yi Zhang. (2010). Brain MRI and CT morphol-ogy in healthy aging and Alzheimer’s disease. Thesis for doctoral degree, Karolinska Institutet Stockholm.

Documents

Radiological Studies on Hippocampal Development ...356455/FULLTEXT01.pdf · Limbic lobe/ system, short overview of the anatomy Limbic lobe, a supernumerary lobe, [49] an arbi-trary