Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print July 24, 2003, 10.1148/radiol.2283020466
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2283020466v1
228/3/683    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Galluzzi, P.
Right arrow Articles by Venturi, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Galluzzi, P.
Right arrow Articles by Venturi, C.
(Radiology 2003;228:683-690.)
© RSNA, 2003

Pediatric Imaging

Retinoblastoma: Abnormal Gadolinium Enhancement of Anterior Segment of Eyes at MR Imaging with Clinical and Histopathologic Correlation1

Paolo Galluzzi, MD, Alfonso Cerase, MD, Theodora Hadjistilianou, MD, Sonia De Francesco, MD, Paolo Toti, MD, Ignazio Maria Vallone, MD, Giuseppe Filosomi, MD, Lucia Monti, MD, Sandra Bracco, MD, Paola Gennari, MD, Chiara Ginanneschi, MD and Carlo Venturi, MD

1 From the Unit of Diagnostic and Therapeutic Neuroradiology, Department of Neurosciences, Azienda Ospedaliera Universitaria Senese, and Interdepartmental Center of Nuclear Magnetic Resonance (P. Galluzzi, A.C., I.M.V., G.F., L.M., S.B, P. Gennari, C.V.), and the Institutes of Ophthalmology (T.H., S.D.F.) and Pathology (P.T., C.G.), University of Siena, Policlinico "Le Scotte," Viale Mario Bracci, 1-53100 Siena, Italy. Received April 18, 2002; revision requested July 8; final revision received November 28; accepted January 20, 2003. Address correspondence to P. Galluzzi (e-mail: neurorad@ao-siena.toscana.it).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
PURPOSE: To evaluate abnormal gadolinium enhancement of the anterior segment of eyes harboring retinoblastoma at magnetic resonance (MR) imaging and correlate this finding with clinical and histopathologic information.

MATERIALS AND METHODS: Three neuroradiologists examined 46 eyes with 34 retinoblastomas in 25 children on gadolinium-enhanced T1-weighted orbital MR images obtained shortly after contrast material injection for evidence of abnormally high signal intensity in the anterior segment. Twenty-two of the 34 affected eyes were enucleated, and six of these 22 eyes were treated with preenucleation adjuvant therapy. Thus, correlation of the clinical, MR imaging, and histopathologic findings in 16 eyes was performed. Statistical analysis was performed with nonparametric methods (Fisher exact test). P < .05 indicated a statistically significant difference.

RESULTS: Fourteen of the 34 affected eyes showed abnormal gadolinium enhancement of the anterior segment. Regarding the 16 eyes evaluated for statistical analysis, a significant correlation (P = .011) between abnormal gadolinium enhancement of the anterior segment and histopathologically documented optic nerve infiltration was noted. A trend toward an association between abnormal enhancement and elevated intraocular pressure (P = .215), tumor growth beyond the equator at MR imaging (P = .125), and histopathologically proved iris neoangiogenesis (P = .182) also was noted. Histopathologic evidence of optic nerve and/or choroid infiltration correlated significantly (P = .001; sensitivity, 100% [nine of nine eyes]; specificity, 86% [six of seven eyes]) with abnormal enhancement.

CONCLUSION: Abnormal gadolinium enhancement of the anterior segments of eyes harboring retinoblastoma seems to indicate more aggressive tumor behavior.

© RSNA, 2003

Index terms: Eye, MR, 224.121411, 224.121413, 224.121415, 224.121416, 224.12143 • Eye, neoplasms, 224.372 • Retina, neoplasms, 224.372


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
During the past century, the prognosis of patients with retinoblastoma has continuously improved owing to early diagnoses and advances in treatment strategy, which have led to the delay or prevention of metastases (1,2). In developed countries, the 5-year survival rate for children with retinoblastoma confined to the eye is nearly 95%, but cancer spread beyond the eye still results in a dismal prognosis, with a 100% mortality rate (2). Reported risk factors for metastases include invasion of the optic nerve, ocular coats, and orbit (3–9). Because fine-needle aspiration biopsy is associated with a substantial risk of procedure-related tumor seeding (2,10), retinoblastoma is one of the few human-inhabiting cancers for which the treatment is generally performed without histopathologic confirmation. Thus, noninvasively rendered diagnostic accuracy is crucial.

The diagnosis of retinoblastoma is usually made with ophthalmoscopy supplemented by ultrasonography (US) (11,12). Computed tomography (CT) and/or magnetic resonance (MR) imaging is required for detection of tumor spread beyond the eye—that is, intracranial disease—and is performed for diagnostic purposes only in selected instances—for example, when opaque ocular refractive media or retinal detachment is present or for the differentiation of lesions simulating retinoblastoma (13–23). The clinical management of retinoblastoma depends on the size, location, extent, and laterality of the tumor and the patient’s systemic status. The treatment strategy has changed substantially during the past few years. Enucleation remains the principal therapy to improve survival. There are fewer indications for external beam radiation, and there has been continuing development of more conservative therapies, with combined-modality therapy often being used to reduce treatment-associated morbidity (1,24,25).

At our institution, enucleation is performed for the following indications: secondary glaucoma, phthisis bulbi, cancer recurrence despite conservative treatment, unilateral disease with massive retinal involvement without hope of even minimal residual visual acuity, massive vitreous seeding, and/or anterior segment invasion. Enucleation has also been proposed for patients with elevated intraocular pressure or rubeosis iridis at ophthalmoscopy (7,8). The term rubeosis iridis is used in routine ophthalmoscopy practice to refer to iris neoangiogenesis—that is, a layer of newly formed blood vessels along with a variable amount of fibroblasts on the anterior surface of the iris at histopathologic examination (26–29). Iris neoangiogenesis is thought to be caused by either ischemia in the posterior segment of the eye or neovascular glaucoma. It should be noted that any finding that may indicate an advanced retinoblastoma, and consequently the need for enucleation, is important.

Impairment of the blood-aqueous barrier has been described in association with various diseases, including retinoblastoma, and may be identified by using the fluorescein test, iris fluorescein angiography, fluorophotometry, biochemical analysis of the aqueous humor, or noninvasive quantitative determination of aqueous flare with a laser-cell meter (26–35). Assessment of the integrity of the blood-aqueous barrier with gadolinium-enhanced MR imaging in both animals and humans has been described (36–41). The purpose of our study was to evaluate abnormal gadolinium enhancement of the anterior segment of eyes harboring retinoblastoma at MR imaging and to correlate this finding with clinical and histopathologic information.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Patients and Eyes
We reviewed the medical charts of 29 children (12 girls, 17 boys) who had undergone 1.5-T MR imaging of the orbits at the Unit of Diagnostic and Therapeutic Neuroradiology, Policlinico "Le Scotte" of Siena. These patients were from a series of 270 children with retinoblastoma who were followed up at the Center for Intraocular Tumors at our institution. At our institution, the review board does not require its approval or informed consent for retrospective reviews of medical records.

Patient clinical data were collected from inpatient and outpatient medical records and included patient sex and age, affected eye(s), intraocular pressure, and results of ophthalmoscopy supplemented by US, CT and/or MR imaging, surgery, and histopathologic analysis, if performed. An intraocular pressure of higher than 22 mm Hg was considered abnormal (8). The diagnosis of retinoblastoma was based on generally accepted criteria, including creamy-white vascularized mass(es), calcification(s), retinal detachment, vitreous seeding, dilated feeding vessels at ophthalmoscopy, intratumoral calcifications on US or CT scans, hypointensity of the mass compared with the signal intensity of the vitreous body on T2-weighted MR images, and gadolinium enhancement of the tumor on T1-weighted MR images.

Four children did not receive intravenous gadolinium-based contrast material and thus were excluded from the study. Thus, the final study population consisted of 11 girls and 14 boys (mean age, 11.2 months; median age, 9 months; age range at admission, 1–28 months). Retinoblastoma was unilateral in 12 patients (left-sided in four and right-sided in eight patients) and bilateral in the remaining 13. At the time the first MR imaging examinations were performed, four children (patients 1, 5, 14, and 15) with bilateral retinoblastoma had already undergone enucleation on one side. Thus, a total of 46 eyes of 25 children with 34 retinoblastomas were examined with MR imaging (Table 1). The nonenhanced CT scans of the orbits of 15 of the 25 children were available.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Patient Demographics and Clinical and MR Findings at Admission

 
MR Imaging
At admission, all 25 patients underwent 1.5-T MR imaging (Gyroscan ACS-NT 15; Philips Medical Systems, Best, the Netherlands) of the brain and orbits while in a state of general anesthesia induced with 100 mg of chloral hydrate per kilogram of body weight. A quadrature head coil and a surface coil were used. All orbital MR examinations were performed by using a surface coil and included the acquisition of serial transverse 3-mm-thick nonenhanced T1-weighted spin-echo (500/20 [repetition time msec/echo time msec], 256 x 256 matrix) and T2-weighted turbo spin-echo (4,000/120, turbo factor of 11, 512 x 512 matrix) MR images. Transverse 3-mm-thick fast-field-echo T2*-weighted (373/23, 15° flip angle, 256 x 256 matrix) and fast fluid-attenuated inversion-recovery (5,000/100, 2,000-msec inversion time, turbo factor of 18, 256 x 256 matrix) MR images were obtained in 17 and 16 patients, respectively.

The 25 children were administered gadoterate meglumine (0.1 mmol/kg) (Dotarem; Guerbet, Roissy, France) through a peripheral vein. Shortly after gadolinium-based contrast material administration, transverse T1-weighted spin-echo orbital MR images (500/20, 256 x 256 matrix) of at least 3 mm in thickness were obtained with (13 patients) or without (12 patients) fat suppression.

Treatment and Follow-up
Twenty-two of the 34 eyes of the 25 patients with retinoblastomas were enucleated. Sixteen of these 22 eyes (in 16 patients) were enucleated within 7 days after MR imaging, without preenucleation treatment. Twelve of these 16 eyes (in 16 patients) were dark. The five patients (patients 5, 13, 17, 18, and 19) with the remaining six retinoblastomas underwent preenucleation treatment. Patient 5 (with bilateral retinoblastoma and right enucleation performed at another institution before MR imaging at our institution) underwent numerous preenucleation treatments, including several cycles of radiation therapy, cryotherapy, and laser therapy in the residual left eye, but no further follow-up MR imaging. Patients 13 and 19 (both with bilateral retinoblastoma) underwent early enucleation of one eye, followed by laser applications, cryotherapy, radiation therapy, or chemotherapy in the contralateral affected eyes; follow-up MR imaging was performed four times in patient 13 and twice in patient 19. Patients 17 and 18 (both with bilateral retinoblastoma) underwent chemotherapy after MR imaging and before enucleation on one side (patient 17) and both (patient 18) sides. Both of these patients underwent one follow-up MR imaging examination. All 25 patients underwent clinical follow-up for at least 8 months.

Histopathologic Evaluation
The 22 enucleated eyes were fixed in saline-buffered formalin, sampled, and embedded in paraffin, and microtome sections were stained with hematoxylin-eosin. Tumor differentiation and growth pattern (ie, endophytic, exophytic, mixed endophytic-exophytic, or diffusely infiltrating); amount of intratumoral necrosis and calcification; tumor infiltration of the anterior segment, optic nerve, choroid, and/or scleral extension; and absence or presence of iris neoangiogenesis (6,20) were evaluated by two pathologists (P.T., C.G.) who were unaware of the MR imaging results.

Image Evaluation
Three neuroradiologists (A.C., P. Galluzzi, I.M.V.) reviewed the nonenhanced T1- and T2-weighted orbital MR images and the gadolinium-enhanced T1-weighted orbital MR images obtained in the 25 children for tumor signal intensity and extent (ie, beyond or behind bulb equator) and the gadolinium-enhanced T1-weighted images specifically for evidence of abnormally high signal intensity in the anterior segment of the eyes—that is, abnormal gadolinium enhancement just anterior to the lens in the section showing the maximum size of the lens. The three neuroradiologists worked together in consensus.

Statistical Analysis
Because of the preenucleation treatment and the possible resulting changes, the six eyes with retinoblastoma in patients 5, 13, 17, 18, and 19 were excluded from statistical analysis. Thus, 16 eyes that were enucleated within 7 days after MR imaging were assessable for correlation of the MR imaging results with the clinical and histopathologic findings. Absence or presence of abnormal gadolinium enhancement of the anterior segment of the affected eyes was correlated with rubeosis iridis at ophthalmoscopy; intraocular pressure; tumor extent at MR imaging; and tumor differentiation, necrosis, calcification, iris neoangiogenesis, and infiltration of the anterior segment, optic nerve, choroid, and/or sclera at histopathologic analysis. Statistical analysis was performed with nonparametric methods (Fisher exact test). P < .05 indicated a statistically significant difference.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 
Clinical Findings
In all 34 eyes with retinoblastoma, ophthalmoscopy showed a single or multiple creamy-white vascularized masses with dilated feeding vessels with or without retinal detachment. Diffuse vitreous seeding was appreciable in four (12%) eyes, and calcifications were appreciable in seven (21%). Intraocular pressure was elevated in 10 (29%) eyes. Rubeosis iridis was seen in seven (21%) eyes. Pseudohypopyon resulting from whitish flares of tumor cells in the anterior chamber was appreciable in two (6%) eyes.

MR Imaging Findings
The MR imaging findings at admission are summarized in Table 1. All 34 retinoblastomas were slightly hyperintense on the nonenhanced T1-weighted MR images, were hypointense compared with the vitreous body on the nonenhanced T2-weighted MR images, and showed mild to marked enhancement on the gadolinium-enhanced T1-weighted MR images. The tumor extended beyond the bulb equator in 23 eyes and behind the bulb equator in 11. Abnormal gadolinium enhancement of the anterior segment was observed in 14 (41%) of the 34 eyes with retinoblastoma: six eyes in children with unilateral retinoblastoma (Figs 1, 2) and eight eyes in children with bilateral retinoblastoma (Fig 3). In the patients with unilateral retinoblastoma, no abnormal gadolinium enhancement was detected in the anterior segment of the normal other eye.



View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1a. Patient 6. Left eye retinoblastoma (asterisk in a) extending beyond bulb equator. (a) Transverse and (b) sagittal gadolinium-enhanced T1-weighted fat-suppressed spin-echo MR images (500/20, 3-mm section thickness, 256 x 256 matrix) obtained shortly after contrast material injection show area of high signal intensity (arrows) in anterior segment of left eye. Normal contralateral eye does not show abnormal signal intensity. (c) Histopathologic specimen from enucleated left eye shows numerous neovessels (arrows) in anterior part of iris. (Hematoxylin-eosin stain; original magnification, x250.)

 


View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1b. Patient 6. Left eye retinoblastoma (asterisk in a) extending beyond bulb equator. (a) Transverse and (b) sagittal gadolinium-enhanced T1-weighted fat-suppressed spin-echo MR images (500/20, 3-mm section thickness, 256 x 256 matrix) obtained shortly after contrast material injection show area of high signal intensity (arrows) in anterior segment of left eye. Normal contralateral eye does not show abnormal signal intensity. (c) Histopathologic specimen from enucleated left eye shows numerous neovessels (arrows) in anterior part of iris. (Hematoxylin-eosin stain; original magnification, x250.)

 


View larger version (144K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1c. Patient 6. Left eye retinoblastoma (asterisk in a) extending beyond bulb equator. (a) Transverse and (b) sagittal gadolinium-enhanced T1-weighted fat-suppressed spin-echo MR images (500/20, 3-mm section thickness, 256 x 256 matrix) obtained shortly after contrast material injection show area of high signal intensity (arrows) in anterior segment of left eye. Normal contralateral eye does not show abnormal signal intensity. (c) Histopathologic specimen from enucleated left eye shows numerous neovessels (arrows) in anterior part of iris. (Hematoxylin-eosin stain; original magnification, x250.)

 


View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a. Patient 8. Left eye retinoblastoma (asterisk in a) extending beyond bulb equator. (a) Transverse and (b) coronal gadolinium-enhanced T1-weighted fat-suppressed spin-echo MR images (500/20, 3-mm section thickness, 256 x 256 matrix) obtained shortly after contrast material injection show area of high signal intensity (arrows) in anterior segment of left eye. Normal contralateral eye does not show abnormal signal intensity.

 


View larger version (173K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b. Patient 8. Left eye retinoblastoma (asterisk in a) extending beyond bulb equator. (a) Transverse and (b) coronal gadolinium-enhanced T1-weighted fat-suppressed spin-echo MR images (500/20, 3-mm section thickness, 256 x 256 matrix) obtained shortly after contrast material injection show area of high signal intensity (arrows) in anterior segment of left eye. Normal contralateral eye does not show abnormal signal intensity.

 


View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a. Patient 17. Bilateral retinoblastoma. Transverse gadolinium-enhanced T1-weighted fat-suppressed spin-echo MR images (500/20, 3-mm section thickness, 256 x 256 matrix) obtained shortly after contrast material injection (a) at admission and (b) after chemotherapy. (b) Marked reduction of the retinoblastoma volume and disappearance of the mildly high signal intensity (arrows in a) in the anterior segment of the right eye are seen after treatment.

 


View larger version (132K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b. Patient 17. Bilateral retinoblastoma. Transverse gadolinium-enhanced T1-weighted fat-suppressed spin-echo MR images (500/20, 3-mm section thickness, 256 x 256 matrix) obtained shortly after contrast material injection (a) at admission and (b) after chemotherapy. (b) Marked reduction of the retinoblastoma volume and disappearance of the mildly high signal intensity (arrows in a) in the anterior segment of the right eye are seen after treatment.

 
In patient 13, abnormal gadolinium enhancement of the anterior segment of the left eye was appreciable only at the 3-month follow-up MR examination performed after the first cycle of chemotherapy and laser therapy. Patient 17 had tumor growth beyond the equator bilaterally and abnormal gadolinium enhancement in the anterior segment of the right eye at MR imaging performed at admission. At 12-month follow-up MR imaging performed after treatment that included cryotherapy, chemotherapy, laser therapy, and radiation therapy, this patient’s tumor growth was behind the equator bilaterally and the abnormal gadolinium enhancement of the anterior segment of the right eye had disappeared (Fig 3).

Follow-up Findings
At the time of this writing, 24 (96%) of the 25 children had survived and were disease free, whereas one patient (4%, patient 5) had died owing to hematogenous metastases.

Histopathologic Findings
The diagnosis of retinoblastoma was confirmed in all 22 enucleated eyes. The tumor growth pattern was endophytic in eight (36%) eyes, exophytic in one (4%) eye, and mixed endophytic-exophytic in 13 (59%) eyes; there were no diffusely infiltrating retinoblastomas. Nineteen (86%) tumors were well differentiated, and the remaining three (14%) were poorly differentiated. Intratumoral calcifications were observed in all 22 enucleated retinoblastomas. The amount of necrosis was less than 50% of the tumor size in 13 (59%) eyes and more than 50% of the tumor size in the remaining nine (41%) eyes. Choroidal invasion was observed in six (27%) eyes. Scleral invasion was observed in two (9%) eyes. Optic nerve involvement was observed in eight (36%) eyes. Anterior segment invasion was observed in six (27%) eyes. Iris neoangiogenesis (Fig 1) was seen in 12 (54%) eyes.

Image Evaluation and Statistical Analysis
The histopathologic findings in the 16 enucleated eyes considered for clinical, MR imaging, and histopathologic statistical evaluation are summarized in Table 2. The corresponding correlations between gadolinium enhancement of the anterior segment of the affected eyes and the main clinical, MR imaging, and histopathologic findings are presented in Table 3. There were no statistically significant correlations between the finding of abnormal gadolinium enhancement in the anterior segment of the affected eyes at MR imaging and the ophthalmoscopic findings. There was a trend toward an association between abnormal enhancement and elevated intraocular pressure (P = .215). Optic nerve involvement correlated significantly with abnormal enhancement (P = .011), whereas growth beyond the equator at MR imaging (P = .125) and histopathologically documented iris neoangiogenesis (P = .182) showed a trend toward being associated with abnormal enhancement. There was a significant correlation between histopathologic evidence of optic nerve and/or choroid infiltration and abnormal gadolinium enhancement (P = .001; sensitivity, 100% [nine of nine eyes]; specificity, 86% [six of seven eyes]).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Histopathologic Findings in 16 Enucleated Eyes Considered for Statistical Evaluation

 

View this table:
[in this window]
[in a new window]

  		TABLE 3. Correlations of Main Clinical, MR Imaging, and Histopathologic Findings with Gadolinium Enhancement of Anterior Segment of 16 Affected Eyes

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
 
The eyeball can be considered to be composed of two differently sized sphere segments—the anterior and posterior segments—that are separated by the lens. The anterior segment is filled by the aqueous humor and is divided into two chambers: The anterior chamber is located between the posterior edge of the cornea and the anterior aspect of the iris, and the posterior chamber is located between the posterior aspect of the iris and the anterior aspect of the lens, which is encircled by the ciliary body. The posterior segment includes the vitreous chamber between the lens and the pars plana anteriorly and the retina posteriorly (42–44).

The eyeball is protected by the blood-ocular barrier system (45), which is similar to the blood-brain barrier in terms of neuroepithelial origin, microanatomy, and mode of regulating the regional microenvironment (46). The blood-ocular barrier system includes the blood-retinal (47) and blood-aqueous (48) barriers. The blood-retinal barrier borders the posterior edge of the vitreous chamber and is composed of outer (choroidal) and inner (retinal) membranes that completely seal the intercellular space (46). The blood-aqueous barrier borders the posterior edge of the posterior chamber and has a complex structure that includes a posterior epithelial portion that restricts penetration into the posterior chamber and an anterior endothelial portion that restricts leakage into the anterior chamber (44,49).

Aqueous humor is secreted into the posterior chamber by the ciliary body, enters the anterior chamber through the pupil, and leaves the eye by way of the canal of Schlemm. There is no anatomic barrier between the vitreous body and the posterior chamber. At MR imaging, aqueous humor shows low signal intensity on T1-weighted images and high signal intensity on T2-weighted images. The uveal tract shows slightly high signal intensity on T1-weighted MR images probably because of its melanin-containing epithelium (17,40,43).

Shortly after intravenous gadolinium chelate administration, the normal eyes of humans show mild enhancement of the choroid and the ciliary body, very slight enhancement of the iris, and low signal intensity within the anterior segment (17,40). At 0.5 T, gadolinium-enhanced MR images obtained 40–50 minutes after contrast material injection show persistence of the mildly high signal intensity inside the uveal layer but still no high signal intensity in either the aqueous fluid or the vitreous body (40). Results of experimental studies performed with high-field-strength magnets in healthy rabbits (37) and humans (38) have shown that gadolinium-based contrast material enters the anterior chamber directly by way of the iris root but does not enter the posterior chamber, even on delayed-enhancement MR images. It is notable that the increased signal intensity in the anterior chamber was not appreciable until 10 and 24 minutes after contrast material injection in rabbits (37) and humans (38), respectively.

The diffusion of gadolinium-based contrast material into the aqueous fluid depicted on 0.5-T MR images obtained 40–50 minutes after contrast material injection has been reported as a sign of disruption of the blood-aqueous barrier in adult patients with neovascular glaucoma or central retinal venous thrombosis (40). The diffusion of gadolinium-based contrast material into the anterior segments of both eyes of a 3-month-old girl was seen at 1.5-T MR imaging performed 15 days after cranial trauma that resulted in drowning and hypoxic brain damage (41). This diffusion of gadolinium-based contrast material has been attributed to the lack of tight junctions in newly growing iridociliary vessels resulting from hypoxic injury (40,41,46). To our knowledge, MR imaging evidence of abnormal gadolinium enhancement of the anterior segment of the eyes harboring retinoblastoma seen shortly after contrast material injection had not been described before the present study (13–23). To limit the level of general anesthesia induced, we did not perform delayed MR imaging.

The possible mechanisms linking a posterior segment lesion such as retinoblastoma to anterior segment phenomena such as blood-aqueous barrier impairment remain speculative. Impairment of the blood-aqueous barrier occurs in various conditions and diseases, including retinal detachment (which is almost always present in eyes harboring retinoblastoma), benign and malignant uveal tumors, diabetic retinopathy, uveitis, pseudoexfoliation syndrome, intraocular surgery, and retinal venous occlusion (30–35). Infiltration of the optic nerve by retinoblastoma is often accompanied by compression of central retinal vessels (50–52), and a correlation between retinal venous occlusion and blood-aqueous barrier impairment has been demonstrated (30,31). This latter mechanism might explain the statistically significant correlation between optic nerve infiltration and abnormal gadolinium enhancement of the anterior segment at MR imaging in our study.

Furthermore, there was a positive trend toward an association between abnormal gadolinium enhancement and elevated intraocular pressure, tumor extent at MR imaging, and histopathologically documented iris neoangiogenesis. A possible explanation is that proposed to explain the blood-aqueous barrier impairment in uveal tumors, whose growth might result in alterations of the tight junctions between retinal pigmented epithelial cells and, consequently, leakage of proteinaceous fluid into the subretinal space and the sensory retina. Further diffusion or transportation of proteinaceous fluid into the anterior segment by way of the vitreous body and the posterior segment, due to the lack of a dense barrier that prevents diffusion, might lead to increased aqueous protein concentration (32). Additionally, vasogenic factors, growth factors, or substances released by viable or necrotic tumor cells, altered retinal pigmented epithelium cells, or damaged photoreceptors, as well as an intraocular inflammatory response to immunologic phenomena involving the host’s defense against tumor cell antigens, may induce iris neoangiogenesis and blood-aqueous barrier disruption.

Retinoblastoma-associated iris neoangiogenesis is thought to be induced by hypoxia caused by the overgrowth of the tumor blood vessels and the typically associated retinal detachment, which result in the production of vascular endothelial growth factor (29). This phenomenon seems to parallel the reported correlation between tumor size and hypoxia in uveal melanomas (32) and the iris neoangiogenesis associated with proliferative diabetic retinopathy, which is induced by the release of vascular endothelial growth factor from the diabetic retina into the vitreous and aqueous fluids (35). Decreasing aqueous flare values have been reported to correlate well with ophthalmoscopic and US signs of tumor regression in eyes with choroidal melanoma that are treated with brachytherapy (34). In diabetic eyes in which retinal coagulation is performed, the concentration of vascular endothelial growth factor decreases, and this results in suppressed neovascularization (53). These treatment-induced changes may explain the reduced abnormal gadolinium enhancement in the anterior segment of the right eye of patient 17 following several cycles of adjuvant therapies.

Presurgical diagnosis of iris neoangiogenesis is important because this condition is often followed by increased intraocular pressure and neovascular glaucoma, and both of these conditions may lead to early enucleation. The lack of correlation between clinically and histopathologically documented iris neoangiogenesis in our study is consistent with the low sensitivity of ophthalmoscopy in the detection of early or subtle stages of rubeosis iridis (26,35), especially in patients with dark eyes (34,54). Twelve of 16 enucleated eyes were dark in our series. Fluorescein angiography of the iris is the method of choice for clinical detection of rubeosis iridis; however, it is not routinely performed in patients with retinoblastoma. Another possible explanation for the abnormal enhancement might be neoplastic infiltration into the anterior segment of the affected eyes; however, this is an uncommon condition (4,22,23) and is scarcely detectable at histopathologic analysis because free cells may float away during tissue sampling and processing. These considerations might explain the lack of correlation in our study.

Finally, a highly significant correlation between histopathologically documented presence of optic nerve and/or choroid infiltration—both of which have been reported as statistically significant predictors of metastases—and abnormal gadolinium enhancement was observed. The presurgical diagnosis of optic nerve and/or choroid infiltration is difficult. Ophthalmoscopy may reveal involvement of the optic disk, but it cannot be used to further evaluate extension into the optic nerve or choroidal involvement. The specific roles of CT and MR imaging have been sporadically addressed with conflicting results (9,18,50,55); however, the subject of radiologic detection of retinoblastoma extension into the optic nerve and choroid was beyond the scope of our investigation.

Given the lack of sufficient long-term follow-up in our study, we cannot hypothesize as to whether abnormal gadolinium enhancement of the anterior segment of eyes harboring retinoblastoma at MR imaging may be a statistically significant predictor of metastases. However, due to the paucity of neuroradiologic indicators, we emphasize the need for further retrospective and prospective studies to evaluate the significance of the MR findings reported herein, possibly in conjunction with a noninvasive quantitative evaluation of aqueous flare.

In conclusion, the MR imaging finding of abnormal gadolinium enhancement seems to be an indicator of more aggressive behavior of retinoblastoma, possibly alerting neuroradiologists and ophthalmologists to suspect choroid and/or optic nerve infiltration. Furthermore, gadolinium-enhanced MR imaging may be useful in the detection of incipient iris neoangiogenesis. The series of patients described herein was small compared with the large populations usually required to help demonstrate the efficacy of a diagnostic marker; however, we hope that the reported results contribute to the better treatment and outcome of children with retinoblastoma.


   ACKNOWLEDGMENTS
 
We thank the following individuals from Azienda Ospedaliera Universitaria Senese, Policlinico "Le Scotte," Siena, Italy: Roberta Benvenuti and Roberta Bellini, from the Unit of Diagnostic and Therapeutic Neuroradiology, for their support in the pediatric neuroimaging procedures; Antonella Buscalferri, MD, from the First Unit of Intensive Care, for her invaluable assistance with pediatric neuroanesthesia; and Guido Garosi, MD, from the Unit of Nephrology, for his help in the statistical analysis. We also thank Roberto Faleri and Ombretta Bugiani, from the Central Library, School of Medicine, University of Siena, Policlinico Le Scotte, Siena, Italy, for providing important references for the preparation of this article.


   FOOTNOTES
 
Author contributions: Guarantors of integrity of entire study, P. Galluzzi, C.V., T.H., P.T.; study concepts, P. Galluzzi, A.C.; study design, P. Galluzzi, A.C., T.H.; literature research, A.C., P. Gennari, I.M.V., G.F.; clinical studies, P. Galluzzi, A.C., L.M., S.B., I.M.V.; data acquisition, P. Galluzzi, I.M.V., A.C., P.T., C.G., S.D.F.; data analysis/interpretation, P. Galluzzi, A.C., P.T., T.H.; statistical analysis, P. Galluzzi, A.C., P.T.; manuscript preparation, P. Galluzzi, A.C., P.T., I.M.V.; manuscript definition of intellectual content, P. Galluzzi, C.V., T.H., P.T.; manuscript editing, P. Galluzzi, A.C., P.T.; manuscript revision/review, P. Galluzzi, A.C., P.T., C.V., T.H.; manuscript final version approval, C.V., P.T., T.H.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 REFERENCES
 

  1. Abramson DH. Retinoblastoma 1990: diagnosis, treatment, and implications. Paediatr Ann 1990; 19:387-395.
  2. Singh AD, Shields CL, Shields JA. Prognostic factors in retinoblastoma. J Pediatr Ophthalmol Strabismus 2000; 37:134-141.[Medline]
  3. Messmer EP, Heinrich T, Hopping W, de Sutter E, Havers W, Sauerwein W. Risk factors for metastases in patients with retinoblastoma. Ophthalmology 1991; 98:136-141.[Medline]
  4. Haik BG, Dunleavy SA, Cooke C, et al. Retinoblastoma with anterior chamber extension. Ophthalmology 1987; 94:367-370.[Medline]
  5. Magramm I, Abramson DH, Ellsworth RM. Optic nerve involvement in retinoblastoma. Ophthalmology 1989; 96:217-222.[Medline]
  6. Tosi P, Cintorino M, Toti P, et al. Histopathological evaluation for the prognosis of retinoblastoma. Ophthalmic Paediatr Genet 1989; 10:173-177.[Medline]
  7. Shields CL, Shields JA, Baez KA, Cater J, De Potter PV. Choroidal invasion of retinoblastoma: metastatic potential and clinical risk factors. Br J Ophthalmol 1993; 77:544-548.[Abstract/Free Full Text]
  8. Baez KA, Ulbig MW, Cater J, Shields CL, Shields JA. Iris neovascularization, increased intraocular pressure and vitreous hemorrhage as risk factors for invasion of the optic nerve and choroid in children with retinoblastoma. Ophthalmologe 1994; 91:796-800.[Medline]
  9. Shields CL, Shields JA, Baez K, Cater JR, De Potter P. Optic nerve invasion of retinoblastoma: metastatic potential and clinical risk factors. Cancer 1994; 73:692-698.[CrossRef][Medline]
  10. Karcioglu XA, Gordon RA, Karcioglu GL. Tumor seeding in ocular fine needle aspiration biopsy. Ophthalmology 1985; 92:1763-1767.[Medline]
  11. Shields JA, Shields CL. Differentiation of Coats’ disease and retinoblastoma. J Pediatr Ophthalmol Strabismus 2001; 38:262-266.[CrossRef][Medline]
  12. Roth DB, Scott IU, Murray TG, et al. Echography of retinoblastoma: histopathologic correlation and serial evaluation after globe-conserving radiotherapy or chemotherapy. J Pediatr Ophthalmol Strabismus 2001; 38:136-143.[Medline]
  13. Mafee MF, Goldberg MF, Cohen SB, et al. Magnetic resonance imaging versus computed tomography of leucokoric eyes and use of in vivo proton magnetic resonance spectroscopy of retinoblastoma. Ophthalmology 1989; 23:144-147.
  14. De Potter P, Flanders AE, Shields JA, Shields CL, Gonzales CF, Rao VM. The role of fat-suppression technique and gadopentetate dimeglumine in magnetic resonance imaging evaluation of intraocular tumors and simulating lesions. Arch Ophthalmol 1994; 112:340-348.[Abstract]
  15. Smirniotopoulos JG, Bargallo N, Mafee MF. Differential diagnosis of leukokoria: radiologic-pathologic correlation. RadioGraphics 1994; 14:1059-1079.[Abstract]
  16. De Potter P, Shields JA, Shields CL. Tumors and pseudotumors of the retina. In: De Potter P, Shields JA, Shields CL, eds. MRI of the eye and orbit. Philadelphia, Pa: Lippincott, 1995; 93-116.
  17. Atlas SW, Galetta SL. The orbit and visual system. In: Atlas SW, eds. Magnetic resonance imaging of the brain and spine. 2nd ed. Philadelphia, Pa: Lippincott Williams & Wilkins, 1996; 1018-1092.
  18. Barkhof F, Smeets M, van der Valk P, et al. MR imaging in retinoblastoma. Eur Radiol 1997; 7:726-731.[Medline]
  19. Kaufman LM, Mafee MF, Song CD. Retinoblastoma and simulating lesions. Radiol Clin North Am 1998; 36:1101-1117.[CrossRef][Medline]
  20. Villablanca JP, Mafee MF, Kaufman LM, Greenwald M, Naidich TP. Facies to remember: retinoblastoma, Coats disease, and toxocariasis. Int J Neuroradiol 1998; 4:41-50.
  21. Galluzzi P, Venturi C, Cerase A, et al. Coats disease: smaller volume of the affected globe. Radiology 2001; 221:64-69.[Abstract/Free Full Text]
  22. Shields JA, Shields CL, Suvarnamani C, et al. Retinoblastoma manifesting as orbital cellulitis. Am J Ophthalmol 1991; 112:442-449.[Medline]
  23. Brisse HJ, Lumbroso L, Fréneaux PC, et al. Sonographic, CT, and MR imaging findings in diffuse infiltrative retinoblastoma: report of two cases with histologic comparison. AJNR Am J Neuroradiol 2001; 22:499-504.[Abstract/Free Full Text]
  24. Desjardins L, Levy C, Lumbroso L, et al. Current treatment of retinoblastoma: 153 children treated between 1995 and 1998. J Fr Ophtalmol 2000; 23:475-481. [French].[Medline]
  25. Imhof SM, Moll AC, Schouten-Van Meeteren AY. Intraocular retinoblastoma: new therapeutic options. Ned Tijdschr Geneeskd 2001; 145:2165-2170. [Dutch].[Medline]
  26. Moffat K, Blumenkranz MS, Hernandez E. The lens and rubeosis iridis: an angiographic study. Can J Ophthalmol 1984; 19:130-133.[Medline]
  27. Walton DS, Grant WM. Retinoblastoma and iris neovascularization. Am J Ophthalmol 1968; 65:598-599.[Medline]
  28. Spaulding AG. Rubeosis iridis in retinoblastoma and pseudoglioma. Trans Am Ophthalmol Soc 1978; 76:564-609.
  29. Pe’er J, Neufeld M, Baras M, Gnessin H, Itin A, Keshet E. Rubeosis iridis in retinoblastoma: histologic findings and the possible role of vascular endothelial growth factor in its induction. Ophthalmology 1997; 104:1251-1258.[Medline]
  30. Miyake T, Kayazawa F. Blood-aqueous barrier in eyes with retinal vein occlusion. Ophthalmology 1992; 99:906-910.[Medline]
  31. Nguyen NX, Kuchle M. Aqueous flare and cells in eyes with retinal vein occlusion: correlation with retinal fluorescein angiographic findings. Br J Ophthalmol 1993; 77:280-283.[Abstract/Free Full Text]
  32. Küchle M, Nguyen NX, Naumann GOH. Quantitative assessment of the blood-aqueous barrier in human eyes with malignant or benign uveal tumors. Am J Ophthalmol 1994; 117:521-528.[Medline]
  33. Moriarty AP, Spalton DJ, Moriarty BJ, Shilling JS, Ffytche TJ, Bulsara M. Studies of blood-aqueous barrier in diabetes mellitus. Am J Ophthalmol 1994; 114:768-771.
  34. Nguyen NX, Küchle M, Strunk W. Tyndallometrie zur verlaufskontrolle nach Jod-125-brachytherapie uvealer melanoma. Klin Monatsbl Augenheilkd 1996; 209:25-30.[Medline]
  35. Ino-ue M, Azumi A, Shirabe H, Yamamoto M. Iridopathy in eyes with proliferative diabetic retinopathy: detection of early stage of rubeosis iridis. Ophthalmologica 1998; 212:15-18.[CrossRef][Medline]
  36. Frank JA, Dwyer AJ, Girton M, et al. Opening of blood-ocular barrier demonstrated by contrast-enhanced MR imaging. J Comput Assist Tomogr 1986; 10:912-916.[Medline]
  37. Kolodny NH, Freddo TF, Lawrence BA, Suarez C, Bartels S. Contrast-enhanced magnetic resonance imaging confirmation of an anterior protein pathway in normal rabbit eyes. Invest Ophthalmol Vis Sci 1996; 37:1602-1607.[Abstract/Free Full Text]
  38. Bert R, Freddo T, Caruthers SD, Jara H, Kolodny N, Melham ER. Confirmation of an anterior large-molecule diffusion pathway in the normal human eye. Invest Ophthalmol Vis Sci 1999; 40(suppl):S198.
  39. Freddo TF. Shifting the paradigm of the blood-aqueous barrier. Exp Eye Res 2001; 73:581-592.[CrossRef][Medline]
  40. Manfrè L, Midiri M, Giuffrè G, et al. Blood-ocular barrier damage: use of contrast-enhanced MRI. Eur Radiol 1997; 7:110-114.[CrossRef][Medline]
  41. Chen CJ. Intraocular contrast enhancement in Adams pattern III hypoxic brain damage: MRI. Neuroradiology 2000; 42:54-55.[CrossRef][Medline]
  42. Warwick R, Williams PL. The peripheral visual apparatus Gray’s anatomy. 35th ed. Philadelphia, Pa: Saunders, 1973; 1096-1122.
  43. Mafee MF, Ainbinder D, Afshani E, Mafee RF. The eye. Neuroimaging Clin North Am 1996; 6:29-59.
  44. Pavlin CJ, Foster FS. Ultrasound biomicroscopy: high-frequency ultrasound imaging of the eye at microscopic resolution. Radiol Clin North Am 1998; 36:1047-1058.[CrossRef][Medline]
  45. Bill A. Blood circulation and fluid dynamics in the eye. Physiol Rev 1975; 44:383-417.
  46. Rapoport SI. Osmotic opening of blood-brain and blood-ocular barriers. Exp Eye Res 1977; 25(suppl):499-509.
  47. Cunha-Vaz JG. The blood-retinal barriers. Doc Ophthalmol 1976; 41:287-327.[CrossRef][Medline]
  48. Bill A. The blood-aqueous barrier. Trans Ophthalmol Soc U K 1986; 105:149-155.
  49. Noske W, Stamm CC, Hirsch M. Tight junctions of the human ciliary epithelium: regional morphology and implications on transepithelial resistance. Exp Eye Res 1994; 59:141-149.[CrossRef][Medline]
  50. Jacquemin C, Karcioglu ZA. Detection of optic nerve involvement in retinoblastoma with enhanced computed tomography. Eye 1998; 12:179-183.
  51. Hayreh SS. Structure and blood supply of the optic nerve. In: Heilmann K, Richardson KT, eds. Glaucoma: conceptions of a disease. Philadelphia, Pa: Saunders, 1978; 78-103.
  52. Orgul S, Cioffi GA. Embryology, anatomy and histology of the optic nerve vasculature. J Glaucoma 1996; 5:285-294.[Medline]
  53. Aiello LP, Avery RL, Arrig PG, et al. Vascular endothelial growth factor in ocular fluid of patients with diabetic retinopathy and other retinal disorders. N Engl J Med 1994; 331:1480-1487.[Abstract/Free Full Text]
  54. Ehrenberg M, Mc Cuen BW, II, Schindler RH, Machemer R. Preoperative iris fluorescein angiography and periocular steroids. Ophthalmology 1984; 91:321-325.[Medline]
  55. Ainbinder DJ, Haik BG, Frei DF, Gupta KL, Mafee MF. Gadolinium enhancement: improved MRI detection of retinoblastoma extension into the optic nerve. Neuroradiology 1996; 38:778-781.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 RadioGraphicsHome page
E. M. Chung, C. S. Specht, and J. W. Schroeder
From the Archives of the AFIP: Pediatric Orbit Tumors and Tumorlike Lesions: Neuroepithelial Lesions of the Ocular Globe and Optic Nerve
RadioGraphics, July 1, 2007; 27(4): 1159 - 1186.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2283020466v1
228/3/683    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Galluzzi, P.
Right arrow Articles by Venturi, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Galluzzi, P.
Right arrow Articles by Venturi, C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE