[Topography of the hypophysis and its neighboring structures]

The first part of the article informs about four points: the skeletal structures of the middle cranial fossa, the basic divisions of the gland and their connections with the hypothalamus, its blood supply, and its development. The topographic part deals with the incorporation of the pituitary gland in the sella turcica, its relationship to the meninges, the subarachnoid cavity and other neighbouring structures, especially the sphenoidal sinus, the posterior ethmoidal cells as well as the chiasma opticum. With respect to the latter, the position of the fibres in the optic nerve, chiasma and optic tract and the consequences of lesions in characteristic areas are described. Finally, the variant structures of the cavernous sinus and the topography of its contents are discussed. Detailed data can be found about the course of the oculomotor nerves and the possible locations of their lesions.

The superior wall of the cavernous sinus: a microanatomical study

The superior wall of the cavernous sinus was studied in 30 specimens obtained from 15 cadaver heads fixed in formalin. Trapezoidal in shape, the superior wall of cavernous sinus is limited laterally by the anterior petroclinoid ligament, medially by the dura of the diaphragma sellae, anteriorly by the endosteal dura of the carotid canal, and posteriorly by the posterior petroclinoid ligament. An interclinoid ligament bisects the wall, dividing it into two triangles: the carotid trigone anteromedially and the oculomotor trigone posterolaterally. Similar to the lateral wall of the cavernous sinus, the superior wall is formed by two layers: a smooth superficial dural layer and a thin, less defined deep layer. In the area of the carotid trigone, both layers separate to wrap the anterior clinoid process. The removal of this process will reveal a "clinoid space" medial to which the internal carotid artery can be identified. This clinoid segment of the artery, still extracavernous, is surrounded by two fibrous rings: a distal ring formed by fibers of the superficial dural layer and a proximal ring related to the deep dural layer. Below the proximal ring, the internal carotid artery becomes intracavernous; above the distal ring, the artery is continuous with its supraclinoid segment. The complex dural anatomy of the superior wall, its fibrous rings, and the clinoid space in relation to a superior surgical approach to the cavernous sinus are discussed.

Rectangular field of view for dynamic MR imaging of sellar and parasellar regions

The effectiveness and quality of a rectangular field of view (R-FOV) were compared with those of a square FOV (S-FOV) in dynamic MR study of sellar regions. Cranial nerves in the cavernous sinus were identified more clearly with R-FOV. The image quality grading of R-FOV was significantly superior to that of S-FOV.

Natural cervical inclination and craniofacial structure

The purpose of this study was to search for the statistical associations between postural and morphologic variables of the head. Interpretation of the facial structure was made by using both intracranial and the extracranial reference lines. The sample comprised natural head posture (NHP) cephalograms of 106 dental students, aged 19 to 29 years. Results showed that, when the facial structure was evaluated by using a NHP analysis based on extracranial reference lines, it was associated with the inclination of the cervical column to the true horizontal. In addition, in the natural position of the head, inclination of the NSL reference was found to be associated with the vertical localization of sella turcica (r = -724, p < or = 0.001), rather than the "extension" or "flexion" of the head. It was concluded that associations between posture and structure of the head are merely caused by the functional factors related to "forward cervical posture" and "vertical cervical posture".

Imaging of the sella: anatomy and pathology

The hypophysis cerebri, or pituitary gland, is a complex neuroendocrine organ involved in the control of a variety of homeostatic mechanisms. Subtleties in the internal anatomy of this gland are now becoming appreciated, such as the topographic arrangements of pituitary cell types and specializations in its regional microcirculation. A large number of potential lesions may affect the pituitary, including tumors (adenomas, Rathke cleft cysts, and craniopharyngiomas), inflammatory processes (adenohypophysitis, infections), and vascular lesions (apoplexy, infarction). Diseases isolated to the posterior lobe and stalk (eg, pituitary dwarfism, choristoma, diabetes insipidus) are also well recognized. Modern radiologic techniques like computed tomography (CT) and magnetic resonance (MR) imaging have provided new insights into the morphologic changes of the gland that occur both in health and in a variety of diseases. The picture of the gland that has emerged from this research is a dynamic one: The pituitary undergoes dramatic changes in size and shape throughout life that must be recognized when assessing it for pathologic change.

Normal imaging anatomy of the suprasellar cistern and floor of the third ventricle

This article discusses the gross, functional, and imaging anatomy of the suprasellar cistern and its contents as well as the structures that form its borders, among the most important being the floor of the third ventricle and associated structures. Specific structures discussed include the optic chiasm and optic tracts, the anterior third ventricle, the tuber cinereum, the pituitary stalk (infundibulum), the choroidal fissure, the lamina terminalis, the gyrus rectus, the uncus, and the cerebral peduncles.

Comparison of full- and half-dose gadolinium-DTPA: MR imaging of the normal sella

PURPOSE

This study compares sellar intensities in 17 patients without sellar pathology using half (0.05 mmol/kg, nine patients) and full-dose (0.1 mmol/kg, eight patients) gadopentetate dimeglumine (Gd-DTPA).

METHODS

Sellar MR studies of eight patients who received full and nine patients who received half-dose Gd-DTPA were compared, retrospectively. Sequences included pre-, immediate, and delayed postcontrast T1-weighted coronal images (1.5 T). Intensity measurements were obtained by two observers using a uniform region of interest.

RESULTS

Comparison of normalized intensities revealed no significant difference between intensities obtained from immediate half- and full-dose techniques for any of the tissues examined (Student's t test, P < .90). Delayed scans likewise demonstrated no significant intensity differences between full- and half-dose studies.

CONCLUSION

Our findings suggest that a 50% reduction in dosage of Gd-DTPA for sellar MR at 1.5 T results in no significant diminution in intensity of enhancement of the pituitary gland or adjacent tissues.

Sellar susceptibility artifacts: theory and implications

PURPOSE

To investigate the prevalence and physical basis of a specific form of MR susceptibility artifact that may be seen in the pituitary gland near the junction of sellar floor and sphenoidal septum.

MATERIALS AND METHODS

Coronal, T1-weighted MR images of the pituitary glands in 50 subjects without clinical evidence of pituitary or sphenoidal sinus disease were reviewed to determine the prevalence of a focal susceptibility artifact near the sellar floor. A plexiglass phantom was constructed to duplicate this artifact in vitro, the appearance of which was studied by varying the direction and intensity of the readout gradient.

RESULTS

In the clinical studies, a focal artifact larger than 1 mm2 was observed in MR studies of seven (14%) of 50 subjects and was sufficiently large to mask or mimic pathology in all cases. The location of this artifact was always within the pituitary gland but closely related to the junction of the sphenoidal septum and sellar floor. The artifact was successfully reproduced in the phantom, and its magnitude was shown to be linearly related to the strength and direction of the readout gradient. An explanation for the focal nature and shape of this artifact is presented based on consideration of the boundary conditions of the Maxwell equations of electromagnetism.

CONCLUSION

A focal susceptibility artifact may be seen on MR images of the pituitary gland closely related to the junction between the sellar floor and sphenoidal septum that may mimic or obscure a microadenoma.

Relationship between the development of diaphragma sellae and the morphology of the sella turcica and its content

The impaired formation of the diaphragma sellae may lead to the development of the empty-sella syndrome. This structure, when fully formed, is a protective barrier against the pulsating action that the cerebrospinal fluid exerts on the sellar content. There are anatomical features which support this belief, but they also suggest that the development of the diaphragma sellae is a factor which determines the morphology of the sella turcica and its contents. Those human specimens which do not have diaphragma sellae or in which it is only partially developed, are characterized by a smaller hypophysis, always located at the inferior and/or posterior half of the sella, with a larger sellar volume and frequently greater fragility of its bony walls. These findings, although rare (5% of the cases), are indirect signs of the important role which the diaphragma sellae plays in the sellar region.

Changes in the mandibular rest position after removal of remaining teeth and insertion of complete dentures

A 24-month study was conducted to determine changes in the mandibular rest position after removal of the remaining teeth and insertion of complete dentures. A total of 24 patients with at least 10 teeth acting as occlusal stops were included in this study. Serial lateral cephalometric radiographs were made for each patient on four occasions (1) before extraction (2) 7 to 10 days after the initial denture insertion (3) 12 months later, and (4) 24 months later. Two angular measurements and one linear measurement were made for each radiograph and compared. A statistically significant increase in the face rest height with a tendency for posterior mandibular rotation was recorded between stages 1 and 2. This initial posterior rotation was followed by a gradual forward and upward movement, which finally brought the mandibular rest position even below the preextraction levels. The findings confirm the concept of the variability of posture and suggest that the mandibular rest position is an unreliable means of reestablishing the vertical dimension of occlusion that existed before extraction of the natural teeth.

Dimensional variation of craniofacial structures in relation to changing masticatory-functional demands

A reduction in masticatory stress has been an important factor in the evolution of the human skull. Similarly, the recent increase in the occlusal variation has been related to a change in masticatory activity. The present study investigates short-term variation in craniofacial dimensions by examining cephalometrically two Finnish samples, one exposed to a hard and the other to a soft diet. The samples comprised 32 skulls, derived from the 16th and 17th centuries, and 50 living individuals. Out of 18 dimensions measured, 12 showed only non-significant differences between the samples. In the present-day sample, the cranial length and the anterior cranial base were significantly longer, and the upper incisors segment significantly higher. In the skull sample, the posterior facial height, the height of the mandibular ramus, and the antero-posterior width of the pharynx were significantly larger. The results suggest that hard diet, which requires more chewing force and time, promotes vertical growth of the ramus and anterior translocation of the maxilla. The greater posterior face height and greater height of ramus are in accordance with the earlier finding that the mandible shows more anterior growth rotation in an attritive environment. These findings support the hypothesis that the growth of the craniofacial skeleton is regulated by masticatory stress. It is suggested that both the dimensional changes and the lack of dental attrition may have contributed to the higher occlusal variation of modern individuals.

Magnetic susceptibility artifact in spin-echo MR imaging of the pituitary gland

PURPOSE

To evaluate the magnetic susceptibility effect on spin-echo MR images of the pituitary gland.

METHODS

Air-in-water phantom experiments and studies in normal volunteers were performed using various sampling bandwidth (8.3-33.3 kHz) spin-echo sequences with the polarity of the readout gradient normal or reversed.

RESULTS

Attachment of a sphenoid septum to the sellar floor was the major factor in the appearance of the inferior surface of the pituitary gland. Patterns of distortion and/or artifactual signal intensities, related to the presence or absence of the attachment, were accentuated on the images with narrower bandwidth. A "spearhead shape" deformity of the sphenoid sinus was observed both in the air-in-water phantom experiment, and when no sphenoid septum was present. When a sphenoid septum was present, two spearhead shapes side by side were present. High-intensity artifacts were seen where the sellar floor was misplaced into the pituitary gland.

CONCLUSION

Knowledge of the patterns of distortion and high-intensity artifact is important in the diagnosis of pituitary lesions.

The assessment of antero-posterior dental base relationships

The reliability and validity of four methods of assessing skeletal pattern (Reidel's method, Eastman correction, Ballard's method, and 'Wits' analysis) from cephalometric tracings have been examined, and the levels of agreement between them investigated. The possibility of the Y-axis length and SN-Y-axis angle having a bearing on the skeletal pattern was also examined. Strong correlation was found between the four methods considered. Ballard's method correlated very closely with Reidel's method, and like 'Wits' method, is not as unreliable as previously reported. Measurement of the SN-Y-axis angle was not found to be of benefit to the assessment of the anteroposterior dental base relationship. Reidel's method, which is the easiest to apply, is recommended as the method of choice, and routine application of the Eastman correction is not recommended.

Face height and tooth eruption in adults--a 20-year follow-up investigation

A longitudinal roentgen cephalometric investigation of vertical craniofacial and dentoalveolar changes during 20 years of adulthood, was performed in 15 men and 15 women. Two lateral cephalograms taken at the average ages 25 and 45 years, were available of each subject. Skeletal and dental changes were described by 13 linear and four angular cephalometric measurements. The analysis of the linear variables showed that total anterior face height increased by 1.60 mm on average. Approximately one-fifth of this increase occurred in the upper and four-fifths in the lower face. In the dentoalveolar region, significant increments of all dimensions except overjet and overbite were found, indicating an eruptive movement of the teeth and a vertical development of their investing tissues. The analysis of the angular measurements showed that posterior rotation of the mandible and uprighting of the upper incisors had occurred during the period of investigation.

Vertical facial growth: a longitudinal study from 7 to 14 years of age

Multilevel analysis is used to describe the changes in vertical facial dimensions and their relationships in Dutch girls between 7 and 14 years of age. For the anterior face height (nasion-gnathion) and the posterior face height (sella-gonion) a fourth degree polynomial was fitted. The mandibular plane angle (sella nasion-gonion gnathion) could be described with a second degree polynomial. Low correlations were found between the growth parameters (intercept and age coefficient) of the anterior and posterior face height, which points to independent growth and development of the anterior and posterior face height. The individual length and growth velocity (at age 10.8) of the anterior face height and the mandibular plane angle showed substantial correlations. It was concluded that changes during growth with regard to the mandibular plane angle are more strongly related to the anterior than to the posterior face height.

High-intensity signals within the posterior pituitary fossa: a study with fat-suppression MR techniques

Five different theories have been proposed to explain the high-intensity signals within the posterior pituitary fossa seen on MR: (1) a paramagnetic effect of phospholipids in the posterior lobe, (2) lipid in pituicytes in the posterior lobe of the pituitary, (3) neurosecretory granules in the posterior lobe, (4) fat within the sella but outside the pituitary gland, and (5) fat in bone marrow in the dorsum sellae. Previous reports have contained conflicting evidence on which of these structures is the cause of the high-intensity signals within the posterior sella. The purpose of this study was to examine the high-intensity signals of the normal posterior sella with fat-suppression MR techniques to reevaluate the contribution of fat to those signals. The sellae of 19 normal volunteers and two cadavers were imaged with MR with a commercially available unit and a research fat- water-suppression technique. High-intensity signals in the posterior sella were observed in all 21 subjects on conventional T1-weighted MR images. In two volunteers, the high-intensity signals in the posterior sella were suppressed with fat-suppression techniques; in 17 subjects the signals were suppressed with water-suppression techniques. In two volunteers the results were indeterminate. The high-intensity signals in the posterior sella do not behave like lipid in the majority of cases. Our study supports the conclusion that high-intensity signals in the posterior sella may have more than one source. It appears that most of these sources do not suppress with fat-suppression techniques.

[CT anatomy of the sella turcica and its environment]

If MRI is not considered because of its limited capacity, computed tomography has become an exclusive diagnostic method in recent years in our country for the radiologic visualization of the sella turcica and its vicinity. Interpretation of the findings of this method lies on the knowledge of normal anatomical situations and variations. In direct coronal sections the hypothalamus, pituitary gland, suprachiasmatic cystern, optic chiasm, infundibular recess, sellar flor and the nerves transversing the cavernous sinus are visualised. The axial sections are more suitable for the visualization of the posterior lobe of pituitary gland and dorsum sellae. The Meckel's cavity, infundibulum and the blood vessels around the sella turcica are seen in both sections. The paper presents findings on the optimal CT evaluation of the above anatomical structures and variations as related to their diagnostic value.

[Linear versus circular polarization of head coils: comparison on phantoms and in the clinic]

Two different head coils--one with linear polarization and the other with circular polarization--were compared under the same measurement conditions. Comparison was done on a phantom with water-filled and gadolinium-filled pin-holes, as well as on anatomical MR images of 23 volunteers. In three volunteers the whole brain was examined while, in the remaining 20 volunteers the sella region or cerebellopontine angle region was examined. Criteria for comparison were signal-to-noise ratio, background noise, and detail resolution (phantom), as well as subjective criteria--image sharpness, anatomical, contrast, and recognition of anatomical details--, evaluated on anatomical MR images by four radiologists independently of each other. The results show a significant improvement of signal-to-noise ratio, lower background noise and therefore marked improvement of images sharpness, and moderate improvement in the recognition of anatomical details using the circular polarized head coil; as for as detail resolution and anatomical contrast were concerned, however, no significant difference was seen between the two coils. Major advantages of the circular, polarized head coil in clinical application are shorter measurement times (reduced number of acquisitions), as well as thinner slices without loss of signal-to-noise ratio.

Surgical approaches to the cavernous sinus: a microsurgical study

The surgical approaches to the cavernous sinus were examined in 50 adult cadaveric cavernous sinuses using magnification of X3 to X40. The following approaches were examined: 1) the superior intradural approach directed through a frontotemporal craniotomy and the roof of the cavernous sinus; 2) the superior intradural approach combined with an extradural approach for removing the anterior clinoid process and unroofing the optic canal and orbit; 3) the superomedial approach directed through a supraorbital craniotomy and subfrontal exposure to the wall of the sinus adjacent to the pituitary gland; 4) the lateral intradural approach directed below the temporal lobe to the lateral wall of the sinus; 5) the lateral extradural approach for exposure of the internal carotid artery in the floor of the middle cranial fossa proximal to the sinus; 6) the combined lateral and inferolateral approach, in which the infratemporal fossa was opened and the full course of the petrous carotid artery and the lateral wall of the sinus were exposed and; 7) the inferomedial approach, in which the medial wall of the sinus was exposed by the transnasal-transsphenoidal route. It was clear that a single approach was not capable of providing access to all parts of the sinus. The intracavernous structures best exposed by each route are reviewed. The osseous relationships in the region were examined in dry skulls. Anatomic variants important in exposing the cavernous sinus are reviewed.

Optic chiasm position on MR images

This article describes the results of a study of the distance from the optic chiasm to the tuberculum sellae as seen on sagittal MR images. Measurements revealed an average chiasm-tuberculum distance of 3.8 mm (2.6 mm for females, 4.3 mm for males) and a sizable group in whom the distance was effectively zero. These results show a fairly close correlation with previous anatomic studies. The closer the chiasm is to the tuberculum, the earlier one may expect clinical manifestations of pituitary disease and the more difficult an intracranial surgical approach to the pituitary will be.

MR imaging of the posterior hypophysis in children

The posterior lobe of the pituitary gland was studied by MR imaging in 30 children without pituitary gland disease and compared with studies from a group of 13 children with central diabetes insipidus, including eight cases of primary diabetes insipidus and five cases of diabetes insipidus secondary to suprasellar tumors (four proved germinomas, one still unknown tumor). Two components in the sella turcica were identified in all 30 children without pituitary gland disease, and the posterior lobe was identified as a high-intensity structure on T1-weighted images. In all 13 patients with diabetes insipidus, the normal hyperintense signal of the posterior hypophysis was absent on T1-weighted images. Three patients with suprasellar tumors presented with a progressively enlarging pituitary stalk on follow-up. Our findings show that absence of the normal hyperintense signal of the posterior lobe is closely related to a loss of function of the neurohypophysis. Size or signal modification of the pituitary stalk should suggest the development of a suprasellar tumor.

Morphology of arachnoid granulations and villi in the region of the human sella turcica--light microscopy and three-dimensional image analysis

The distribution and structure of the arachnoid granulations and villi in the region of the sella turcica in human adult brains were observed under light microscopy. In order to study the interrelationships between the arachnoid projections, hypophysis, cavernous sinus, and sella turcica, we performed a three-dimensional reconstruction and analysis of the region with the use of a digital image-processing method. Arachnoid projections penetrating the dura were villous when they were few in number, but more frequently granular when they were numerous. No relationship was observed between the type of projections and the age or the underlying disease of the subjects. Brains which showed no or few arachnoid granulations or villi had a small and deep hypophyseal fossa with a poorly-developed intercavernous sinus or venous plexus. Three-dimensional image analysis revealed that arachnoid granulations and villi form a maze in the dura or connective tissue between the venous plexuses.