| OCR Text |
Show Journal of Clinical Neuro- ophlhalmology 11( 4): 280- 283. 1991. Pituitary Tumor Volume as a Predictor of Postoperative Visual Field Recovery Quantitative Analysis Using Automated Static Perimetry and Computed Tomography Morphometry Henry Hudson, M. D., Colleen Rissell, D. O., William J. Gauderman, M. S., and Steven E. Feldon, M. D. © 1991 Raven Press, Ltd., New York We attempted to define the relationship between the volume of pituitary adenomas and postoperative visual field recovery. Fourteen consecutive patients ( 27 eyes) were studied who presented with visual complaints and who were confirmed to have pituitary macroadenomas at the time of resection. For each eye, preoperative visual field loss in each quadrant as well as whole field loss were analyzed. Tumor volume was determined from computed tomography scans by summing serial axial sections. Statistically significant correlations ( p < 0.05) of preoperative superonasal and whole visual field loss with tumor volume were found. Postoperative visual field recovery was determined for each patient by subtracting the postoperative visual field loss from the preoperative field loss. Postoperative inferonasal field recovery was significantly correlated ( p < 0.01) with whole tumor volume, with markedly greater recovery in patients with tumor volumes of 5 cc or less. In general, tumor volume proved to be a poor predictor of postoperative visual field recovery. Key Words: Pituitary neoplasm- Visual fieldsPerimetry chiasm- Suprasellar mass- Hypophysectomy- Computerized tomography. From the Departments of Ophthalmology ( H. H., C. R., W. T. G., S. E. F.) and Neurological Surgery ( S. E. F.), University of Southern California School of Medicine, and the Doheny Eye Institute ( H. H., C. R., W.}. G., S. E. F.), Los Angeles, California, U. S. A. Address correspondence and reprint requests to Dr. S. E. Feldon at Doheny Eye Institute, 1355 San Pablo Street, Los Angeles, CA 90033, U. s. A. 280 The major neuro- ophthalmic manifestation of pituitary macroadenoma with suprasellar extension is visual field loss caused by optic nerve compression, chiasmal compression, or both ( 1-- 6). The anatomic relationship between the diaphragm sella and the anterior visual pathway suggests that tumor extension of greater than 10 mm above the diaphragm sella is necessary for the anterior visual system to be compressed. Quantitative assessments of preoperative visual loss and postoperative visual recovery ( 7- 11) have met with limited success because of the use of arbitrary systems of grading visual recovery. Only a few preoperative factors, such as duration of symptoms, severity of preoperative visual loss, presence or absence of optic atrophy, and age of the patient ( 12- 17), have proven predictive of postoperative visual field recovery. In addition, the presence of a prefixed or postfixed chiasm, the exact position of the tumor with respect to the chiasm, dehiscence of the diaphragm sellae, and vascularity of the tumor may play important but as yet unquantifiable roles in determining visual field recovery. Size of the tumor and degree of suprasellar extension are measurable factors that may predict the degree of preoperative visual field loss and/ or of postoperative visual field recovery. However, there are conflicting reports on the correlation of tumor size and visual field recovery ( 7,8), perhaps related to measurement in arbitrary units ( 7) or in millimeters of suprasellar extension ( 8). In this study, we quantitatively analyze the relationship of whole pituitary tumor volume and suprasellar PITUITARY TUMOR VOLUME 281 volume to preoperative visual field deficit and postoperative visual field recovery. MATERIALS AND METHODS Patients Fourteen consecutive patients were studied who presented to the Doheny Eye Institute with visual complaints and who were later confirmed, at the time of resection, as having pituitary macroadenomas. Visual field measurements were obtained on both eyes of 13 patients, but on only 1 eye of 1 patient, where the other eye was blind. The mean age of these patients was 48 years, with a range of 22- 72 years; 10 patients were men, and 4 were women. None of the patients received postoperative radiotherapy or bromocriptine treatment prior to the time of postoperative evaluation. Tumor Volume We obtained computerized tomography for each patient preoperatively. Slices were 1.5 mm, 3 mm, or 5 mm thick, with the exception of 2 patients in whom only 10- mm and 13- mm sections were available. Camera lucida images of all sections were constructed and then traced onto a digitizing bit pad. Total tumor volume, corrected for magnification and overlap, if present, was then determined using a volume summation technique ( 18,19). Visual Fields All visual fields were determined preoperatively and postoperatively for each eye using the OCTOPUS ( TM, Interzeag AG, Schlieren, Switzerland) automated static perimeter. Visual fields were measured postoperatively within 1 month in 9 patients, within 1 year in 3 patients, and after 1 year in 2 patients. Stationary stimuli of constant size ( Goldmann III) were presented pseudorandomly to different locations within the field of vision. Retinal sensitivity was tested by comparing the threshold luminance intensity at which the patient detected the target with that of age- matched controls. Sensitivity was expressed on a logarithmic decibel ( dB) scale. For this study, 30 0 of spectacle- corrected central field were analyzed and retinal sensitivity was determined every 6 0 ( program 32 or 34). The horizontal and vertical meridians were straddled by the test points. The mean loss of sensitivity for the entire field and for each quadrant of the visual field was obtained both pre- and postoperatively. Statistical Methods Preliminary analyses showed that right and left eye preoperative visual field measurements within the same patient were significantly correlated ( mean r = 0.69, P < 0.005). Thus, the regression model of Rosner ( 20), which adjusts for dependent outcomes, is used to test for correlations between tumor volume and visual fields. Correlations between postoperative visual field recovery and tumor volume are also adjusted for preoperative field loss. Correlations are reported in terms of preoperative R2 , which is the amount of variation in visual field measurements that can be explained by tumor volume. All reported p values are twosided; the statistical significance level is 0.05. RESULTS Whole tumor volume and suprasellar volume had similar correlations with preoperative visual loss and postoperative recovery. Regression analyses demonstrated that whole tumor volume was a better predictor than was suprasellar volume. Thus all correlations will be reported and discussed with regard to whole tumor volume only. Preoperative Visual Field Loss Statistically significant correlations ( p < 0.05) were found between whole tumor volume and preoperative superonasal visual field loss ( R2 = 16.5%) and preoperative whole field loss ( R2 = 13.7%) ( Table 1). Tumor volume was not significantly correlated with preoperative visual field loss in the inferonasal, superotemporal, and inferotemporal quadrants. Postoperative Visual Field Recovery No significant relationship was found between tumor volume and superotemporal, inferotempo- TABLE 1. Correlation between whole tumor volume and preoperative visual field loss Slope of Field R- squared8 p value regression line SN 16.5% 0.03 0.58 IN 6.4% 0.19 0.33 ST 10.9% 0.08 0.69 IT 8.0% 0.14 0.58 Whole field 13.7% 0.04 0.55 SN, superonasal; IN, inferonasal; ST, superotemporal; IT, inferotemporal. 8 The amount of variation in visual field loss that can be explained by whole tumor volume. I Clin Neuro- ophthalmol, Vol. 11, No. 4, 1991 282 H. HUDSON ET AL. TABLE 2. Correlation between whole tumor volume and postoperative visual field recovery Slope of Field R- squaredB p value regression line SN 0.4% 0.83 - 0.02 IN 24.1% 0.006 - 0.21 ST 2.7% 0.37 - 0.22 IT 3.1% 0.46 - 0.17 Whole field 3.5% 0.32 - 0.13 SN, superonasal; IN, inferonasal; ST, superotemporal; IT, inferotemporal. B The amount of variation in visual field recovery that can be explained by whole tumor volume. ral, superonasal, or whole visual field recovery ( Table 2). However, as shown in Fig. 1, tumor volume was a significant predictor of inferonasal visual field recovery ( R2 = 24,1%, P < 0.01). Figure 1 also demonstrates that inferonasal visual field recovery was clinically important only in patients with a tumor volume of 5 cc or less. DISCUSSION Perimetric analysis of visual field defects remains an important component in the diagnosis and management of patients with suprasellar extension of pituitary adenomas ( 21). Temporal and bitemporal hemianopias are the most common defects, being present in up to 96% of patients with pituitary tumors that affect the visual pathways ( 8). Although magnetic resonance is now the preferred procedure by which these tumors are imaged, the large tumor volumes seen in these patients can accurately be assessed by computerized tomography scans ( 22). Using quantitative techniques, we have confirmed the poor correlation between tumor volume and preoperative temporal field loss that has been suggested on a qualitative basis by other studies ( 7,8,23). However, preoperative whole field loss and especially preoperative superonasal field loss correlated significantly with pituitary tumor volume. This is expected, as the superonasal visual field is typically involved late in the course of this disease ( 1,24). Of importance to the surgeon and to the patient is the absence of correlation between postoperative visual field recovery and tumor volume. The lack of correlation between temporal visual field recovery and tumor volume might be expected, because the chiasmal crossing fibers are typically compressed first by the enlarging tumor ( 1,24). However, it is also not possible to preoperatively predict superonasal or whole visual field recovery, b · ' jCpC! nn hJmor volume. Only inferonasal visual JClin Neuro- ophthalmol, Vol. 11, No. 4. 1991 Visual Field Recovery ( dB) 10 8 * 2 R · 24.1% 6 * * ** <& l * 2 * 0 * - 2 - ----- ~ _<& L- .. L....- -- L- -'-- *--' o 5 10 15 20 Tumor Volume ( cc) FIG. 1. Visual field recovery ( dB) in the inferonasal field versus tumor volume ( cc). The R2 value is adjusted for preoperative visual field loss and for the correlation between eyes. Note the markedly greater recovery of visual field with tumor volumes of 5 cc or less. field recovery was significantly correlated with tumor volume, and such recovery tended to occur only if the tumor volume was 5 cc or less. This study shows that caution should be exercised in predicting visual field recovery following hypophysectomy. Unmeasured parameters, such as the presence of a prefixed or postfixed chiasm, the exact position of the tumor with respect to the chiasm, an incompetent diaphragm sellae, and the vascularity of the tumor may be overriding factors in determining preoperative visual field loss and the extent of postoperative recovery. Acknowledgment: This research was supported in part by an unrestricted grant from Research to Prevent Blindness, Inc., New York, New York, and by Core Grant EY03040 ( Biometry Module) from the National Institutes of Health. REFERENCES 1. Cushing H, Walker ca. Distortions of the visual fields in cases of brain tumour. Brain 1915; 37: 341- 400. 2. Chamlin M, Davidoff LM. Ophthalmic criteria in diagnosis and management of pituitary tumors. Symposium on PituItary Tumors III. 1961; 9- 18. 3. Chamlin M, Davidoff LM, Feiring EH. Ophthalmologic changes produced by pituitary tumors. Am J Ophthalmol 1955; 40: 353-- 68. PITUITARY TUMOR VOLUME 283 4. Elkington SG. Pituitary adenoma: preoperative symptomatology in a series of 260 patients. Br' Ophthalmol 1968; 52: 322-- 8. 5. Klauber A, Rasmussen P, Lindholm J. Pituitary adenoma and visual function: the prognostic value of clinical, ophthalmological and neuroradiological findings in 51 patients subjected to operation. Acta Ophthalmol 1968; 56: 252--- 63. 6. Meller PM, Hvid- Hansen O. Chiasmal visual fields. Acta Ophthalmol 1970; 48: 678-- 84. 7. Findlay G, McFadzean RM, Teasdale G. Recovery of vision following treatment of pituitary tumours: application of a new system of assessment to patients treated by transsphenoidal operation. Acta Neurochir 1983; 68: 175-- 86. 8. Symon L, Jakubowski J. Transcranial management of pituitary tumours with suprasellar extension. , Neurol Neurosurg Psych 1979; 42: 123- 33. 9. Fahlbusch R. Optic nerve compression by pituitary adenomas. In: Samii M, Janetta P, eds. Cranial nerves. Proceedings of the Symposium on cranial nerves, ' une 1980. Heidelberg: Springer, 1981: 140- 47. 10. Jefferson A. The treatment of chromophobe pituitary adenomas by means of transfrontal surgery, radiation therapy and supportive hormone therapy. In: Fahlbusch R, Werder KV, eds. Treatme1lt of pituitary ademonas: 1st European workshop. Stuttgart: Theime, 1978. 11. Laws ER, Trautmann JC, Hollenhorst RW. Transsphenoidal decompression of the optic nerve and chiasm: visual results in 62 patients. , Neurosurg 1977; 46: 717- 22. 12. Symon L, Jakubowski J, Kendall B. Surgical treatment of giant pituitary adenomas. J Neurol Neurosurg Psychiatr 1979; 42: 973-- 82. 13. Safran AB, Safran E, Glaser JS, David NJ. Adenomes pituitaires et meningiomes supra- sellaires: pronostic visuel post- operatoire. Neurochirurgie 1980; 26: 39-- 45. 14. Rougier J, Lelievre- Clapies C. Evolution campimetrique apres interventions des adenomes hypophysaires ( a propos de 190 cas). Bull Soc Ophtalmol Fr 1981; 81: 491- 5. 15. Cohen AR, Cooper PR, Kupersmith MJ, et al. Visual recovery after transsphenoidal removal of pituitary adenomas. Neurosurgery 1985; 17: 446- 52. 16. Bynke H. Pituitary adenomas with ocular manifestations: incidence of cases and clinical findings 1946- 1984. Neuroophthalmology 1986; 6: 303- 11. 17. Miller NR, Calvert JS, Zinreich, Long D. Quantitative analysis of visual field results following transsphenoidal surgery for pituitary adenomas. Neuro- ophthalmol Jpn 1986; 3: 431. 18. Penn RD, Belanger MG, Yasnoff WA. Ventricular volume in man computed from CAT scans. Ann NeuroI1978; 3: 21623. 19. Pentlow KS, Rottenberg DA, Deck MDF. Partial volume summation: a simple approach to ventricular volume determination from CT. Neuroradiology 1978; 16: 130- 2. 20. Rosner B. Multivariate methods in ophthalmology with application to other paired- data situations. Biometrics 1984; 40: 1025- 35. 21. Vlahovitch B, Reynaud C, Rhiati J, et al. Treatment and recurrences in 135 pituitary adenomas. Acta Neurochir 1988; 42( suppl): 120- 3. 22. Mikhael MA, Ciric IS. MR imaging of pituitary tumors before and after surgical and/ or medical treatment. J Comput Assist Tomogr 1988; 12: 441- 5. 23. Wilson P, Falconer MA. Patterns of visual failure with pituitary tumours: clinical and radiological correlations. Br , OphthalmoI1968; 52: 9~ 110. 24. Tindall GT, Barrow DL. Disorders of the pituitary. St. Louis: CV Mosby, 1986: 124-- 8. I CIin Neuro- ophthalmol, Vol. 11, No. 4, 1991 |
