Journal of Neuro-Ophthalmology, December 1991, Volume 11, Issue 4

Visual outcomes of pituitary adenoma surgery. St. Vincent's Hospital 1968-1987.

Pituitary tumors are a significant cause of visual morbidity. The medical records of 45 patients with histologically verified pituitary adenoma and visual defects (fields and/or acuity) were reviewed. The patients' ages ranged from 19 to 80 years with a mean of 52.4 years. Ophthalmologists referred 76% of the patients for neurosurgical opinion. Bitemporal field defects were present in 89%. Twenty-three patients underwent transfrontal craniotomy prior to 1984, and the remaining twenty-two subsequently had transsphenoidal microsurgery. Of those eyes with abnormal visual acuity preoperatively, 74% had improvement noted postoperatively (p less than 0.001). Of those eyes with preoperatively abnormal visual fields, 68% improved (p less than 0.001). For transsphenoidal surgery the rate of improvement for fields was 81%. Preoperative visual acuity, visual field score, and patient age, were not predictive of postoperative acuity. Optic disc pallor was associated with poorer postoperative fields and acuity compared to eyes without preoperative pallor. Transsphenoidal surgery is effective treatment for visual compromise due to pituitary adenoma, and the need for early diagnosis is emphasized.

Journal of Clillical Nt'u", · oplrthall/ lolo~ y 11( 4): 262- 267. 1991 © 1991 Raven Press, Ltd., New York Visual Outcomes of Pituitary Adenoma Surgery St. Vincent's Hospital 1968- 1987 L. J. Sullivan, M. B., B. S., J. O'Day, M. B., B. S., F. R. A. C. O., F. R. A. C. S., F. R. A. C. P., and P. McNeill, M. B., B. S., F. R. A. C. S. Pituitary tumors are a significant cause of visual morbid­ity. The medical records of 45 patients with histologically verified pituitary adenoma and visual defects ( fields and/ or acuity) were reviewed. The patients' ages ranged from 19 to 80 years with a mean of 52.4 years. Ophthal­mologists referred 76% of the patients for neurosurgical opinion. Bitemporal field defects were present in 89%. Twenty- three patients underwent transfrontal cranioto­my prior to 1984, and the remaining twenty- two subse­quently had transsphenoidal microsurgery. Of those eyes with abnormal visual acuity preoperatively, 74% had improvement noted postoperatively ( p < 0.001). Of those eyes with preoperatively abnormal visual fields, 68% improved ( p < 0.001). For transsphenoidal surgery the rate of improvement for fields was 81 %. Preopera­tive visual acuity, visual field score, and patient age, were not predictive of postoperative acuity. Optic disc pallor was associated with poorer postoperative fields and acuity compared to eyes without preoperative pal­lor. Transsphenoidal surgery is effective treatment for visual compromise due to pituitary adenoma, and the need for early diagnosis is emphasized. Key Words: Pituitary adenoma- Surgery- Trans­sphenoidal surgery- Visualloss-- Visual fields From the Royal Victorian Eye and Ear Hospital ( L. j. S.); De­partments of Ophthalmology ( J. O'D.) and Neurosurgery ( p. McN.), St. Vincent's Hospital, Melbourne, Australia. Address correspondence and reprint requests to Dr L. J. Sul­livan at Royal Victorian Eye and Ear Hospital, 32 Gisborne ric, d le, t ~."' lh"'. lnl<.' iOO". Australia. 262 Pituitary adenomas make up 15% of all intracra­nial tumors and have long been known to have the potential to impair vision ( 1). Classically, large tu­mors have produced a visual triad of visual field defects ( particularly bitemporal hemianopia), loss of central visual acuity, and optic disc pallor ( 2). Pituitary tumor surgery offers rapid and gratify­ing recovery of visual function for those affected. Optimal surgical management of these tumors has been controversial until recent times. In the 1980s, transsphenoidal surgery using the operating mi­croscope and image intensifier became the pre­ferred mode of removal, and has largely replaced the earlier transfrontal, transcranial approach for most tumors (~ 5). This study examines the presentation, modes of referral, and visual results of surgery for pituitary macroadenoma with visual compromise per­formed at St. Vincent's Hospital, Melbourne, be­tween 1968 and 1987 and compares these results to other centers ( 6,7). Also, a comparison is made of transsphenoidal versus transfrontal surgery. An assessment is also made of various preoperative findings, such as preoperative acuity and field scores, age of patient, and presence of optic atro­phy to ascertain correlation with the final visual outcome. MATERIALS AND METHODS Case Selection A retrospective review was conducted of 300 consecutive patients who had undergone pituitary surgery at St. Vincent's Hospital, Melbourne, be­tween 1968 and 1987. To qualify for inclusion, pa­tients had to fulfill the follOWing criteria: preoper­ative visual impairment ( of either fields or acuity); histologically verified pituitary macroadenoma VISION AND PITUITARY ADENOMA 263 ( not craniopharyngioma); and adequate documen­tation of pre- and post- operative visual status. 45 patients met these criteria and are the basis of this study. Twelve patients met the criteria except for adequate documentation; they were excluded for this reason. Data Recorded Records were searched for data relating to type of surgery, presenting features, histopathology, patient age, diagnostic imaging, treatment, recur­rence, and visual status. Preoperative best cor­rected visual acuity and fields were recorded, along with Ishihara color scores, and the results of ophthalmoscopy, looking for optic disc pallor or atrophy. Postoperative fields and acuity were re­corded. Visual fields were assessed before and after sur­gery in all 45 patients ( 90 eyes). Postoperative acu­ities were not recorded for 3 patients; therefore, this sample is 84 eyes. Results were evaluated for transsphenoidal and transfrontal surgery sepa­rately and together. Snellen acuity was recorded unaltered, because it was felt this would provide a sensitive quantitative comparison of pre- and post­operative acuity. Visual fields were assigned a nu­merical score according to the following schema: The plotted field was divided into quadrants, and each quadrant subdivided into a central 10° area and a peripheral area. Each subdivision was as­signed a potential score of 3, where 3 = full field in that sector, 2 = partial decrease in field « V3 area), 1 = large decrease in field « 213 area), 0 = loss of field in that sector. The maximum score possible was 24; a complete hemianopia would score 12. As far as possible, it was ascertained that comparable targets and meth­ods were used for pre and post operative field as­sessment. The relative weighting for the central 10° of field reflects the relative functional importance of this region compared to the periphery. It was felt that this schema, which is a modification and simplification of previous schemata, would pro­vide a sensitive quantitative approach to assess­ment of field loss changes ( 8,9). RESULTS Presenting Features The average age at presentation was 52.4 years. Patient ages ranged from 19 to 80 years. There were 18 women and 27 men. The most frequent presenting symptom was vi­sual impairment, with decreased visual acuity be­ing particularly frequent ( 89% of cases), and visual field loss being noted by 58%. Diplopia was noted in 4%. Symptoms of endocrine dysfunction were noted at presentation in 29%. Headache was a rel­atively infrequent complaint in this series ( 18%). One patient was diagnosed on routine sinus roent­genogram. Referral Sources Ophthalmologists referred 34 ( 76%) of the pa­tients for neurosurgical assessment. General prac­titioners referred 3 patients ( 7%) and neurologists 2 patients ( 4%). Visual Field Defects All patients had a field defect. As would be ex­pected, a classical bitemporal hemianopia was present in the majority of cases ( 67%) ( 10- 12), al­though variations on this theme were also noted. Asymmetry was a feature of many presenting field defects, and 2 patients had a junctional scotoma ( Table 1). Color vision testing was only recorded for 16 patients, but was abnormal in 14. Unilateral optic atrophy was noted in 9, bilateral optic atrophy in 12. Diagnostic Tests Air encephalography was the routine imaging mode until 1977. Cerebral angiography was also commonly performed up until this time. After 1976 contrast- enhanced CT scanning became routine and replaced the earlier studies. Tumor size was estimated according to Hardy's classification ( 10). All tumors were macroade- TABLE 1. Visual field defects Type of Surgery Until 1983 transfrontal craniotomy was the pre­ferred method of surgery, and 23 patients under­went this operation. During 1983 the microsurgical transsphenoidal approach was adopted and used in the subsequent 22 patients. Bitemporal hemianopia Bitemporal superior quadrantanopia Hemianopia/ quadrantanopia Junctional scotoma/ quadrantanopia Junctional scotoma/ hemianopia Quadrantanopia one eye only Homonymous hemianopia Blind ( NPL) both eyes 30 ( 67%) 6 ( 13%) 4( 9%) 1 ( 2%) 1 ( 2%) 1 ( 2%) 1 ( 2%) 1 ( 2%) JClin Neuro- ophthalmol, Vol. 11, No. 4, 1991 264 L. J. SULLIVAN ET AL. nomas (> 10 mm height), average size being 26.6 mm above the sella turcica. Preoperative endocrinological testing revealed hypopituitarism in 8 patients, ( 18%) of which 3 ( 7%) were panhypopituitary. Hypersecretion was noted in 10 patients ( 22%). Seven tumors pro­duced elevated prolactin levels, three produced ex­cess growth hormone, and one produced both pro­lactin and growth hormone in excess. Histopathology Chromophobe adenomas accounted for 32 tu­mors ( 71 %). There were 7 prolactinomas ( 16%), 3 acidophil adenomas ( 7%), 1 basophil, and 1 mixed cell type ( 2% each). Tumor Size The average height of the tumors in the trans­frontal group was 2.77 cm, compared to 2.61 cm for the transsphenoidal. VISUAL OUTCOMES Visual Acuity Preoperatively, visual acuity was worse than 20/ 20 in 73% of eyes ( 61/ 84). Postoperative improve-ment was noted in 56% of eyes overall, and in 74% ( 45/ 61) of eyes which had decreased acuity ( less than 20/ 20) preoperatively. Over the whole sam­ple, postoperative acuity was normal or improved in 77% ( 65/ 84) of eyes. When comparison was made between the trans­frontal and transsphenoidal approaches, the im­provement rates were 63% ( 25/ 40) and 50% ( 22/ 44), respectively. These figures appear to favor the transfrontal approach. However, in the transsphenoidal group, 16 eyes ( 36%) had normal preoperative acuity compared to 7 eyes ( 18%) in the transfrontal group. When eyes with normal preoperative acuity are excluded, comparative improvement rates of 76% ( transfron­tal) and 71% ( transsphenoidal) are obtained, ( X2 = 0.15, P = 0.92 ns). These results are summarized in Fig. 1. Five eyes with normal preoperative acuity dete­riorated. All had transfrontal surgery. Three eyes of transsphenoidal patients with abnormal preop­erative acuity deteriorated, 1 went from 20/ 20 to 20/ 40; 1 from 20/ 40 to 20/ 80; and 1 from 201120 to Perception of Light. Visual Fields Preoperatively, visual fields were full in 3 eyes. Postoperatively, visual fields were improved in Zll. Zll. Zll. CF HM PL 80 120 200 FIG. 1. Results: visual acuity. POST OPERATIVE PL HM CF 20/ 200 0 20/ 120 0 20/ 80 0 • 20/ 60 0 20/ 40 20/ 30 20/ 20 • " worse area" o o 0 • • " no change" diagonal i " improved area" PRE OPERATIVE J Clin Neuro- ophtlullmol. Vol. 11, No. 4, , CF = count fingers at 10 feet HM = hand movements at 3 feet PL = perception of light o = Transfrontal • = Transsphenoidal VISION AND PITUITARY ADENOMA 265 66% ( 59/ 90) of eyes overall, and were full in 46% ( 41/ 90) of eyes. If preoperatively normal eyes are excluded, there was improvement in 68% ( 59/ 87). When comparison was made between transfrontal and transsphenoidal results, the improvement rates were 56% ( 25/ 45) and 81% ( 34/ 42), respec­tively, when eyes with normal preoperative fields were excluded. No eyes with normal preoperative fields deteriorated. Results for visual fields are summarized in Fig, 2. Preoperative Vision and Outcome Preoperative visual acuity better than or equal to 20/ 100 was associated with an improvement rate of 77% ( 27/ 35), compared to 69% ( 18/ 26) for eyes with initial acuity less than 20/ 100 ( X 2 = 0.16, P = 0.689 ns). Twenty- three eyes with preoperatively normal acuity ( 20/ 20 or better) were excluded, as they could not improve, and would bias these figures. Preoperative visual field scores of 16 or better were associated with an improvement rate of 75% ( 21/ 28) versus 64% ( 38/ 59) for eyes with initial field scores less than 16, when eyes with preoperatively abnormal fields are considered. These differences are not statistically significant ( X 2 = 0.55, P = 0.458). Disc Pallor/ Optic Atrophy and Outcome Disc pallor or frank atrophy was recorded in 21 patients, being bilateral in 12 and unilateral in 9. For the patients with bilateral pallor or atrophy, improvement rates were 65% for acuity and 63% for fields ( only 20 eyes had abnormal preoperative acuity). With unilateral pallor or atrophy, the rates for improvement postoperatively were 63% for acuity and 55% for fields. These results contrast with those for eyes with no pallor or atrophy, in which the rates of im­provement were 85% for acuity and 72% for fields. The samples here are too small to attain statistical significance ( acuity, i = 2.43 and p = 0.12; fields, X2 = 0.79 and p = 0.37). Table 2 summarizes the results for all sample subgroups. Age Patient age was not predictive of amount of im­provement in visual acuity or fields postopera­tively, Recurrence Tumor recurrence was reported in 8 patients, 7 of which had had transfrontal surgery. The aver- POST OPERATIVE 4 6 8 2 improved area" " worse area" " no change', diagonal ~ po ~. 0 0 1= ~ u •• ,", v • •• • • ... 1- • 1= • i. r-~.. i" i.. ~ it- 24 I-' 20 18 22 12 14 16 10 FIG. 2. Results: visual fields. 24 22 20 18 16 14 12 10 8 6 4 2 PRE OPERATIVE o = Transfrontal • = Transsphenoidal 1Clin Neuro- ophthalmol, Vol. 11, No. 4, 2992 266 L. J. SULLlVAN ET AL. DISCUSSION TABLE 2. Visual outcome for sample subgroups age time interval to recurrence was 4 years post­operatively. Pituitary adenomas, particularly if they are non­secreting, may not be symptomatic until they cause visual morbidity via the chiasmal syndrome ( 2). The important role played by ophthalmologists in the diagnosis and appropriate referral of these patients is emphasized in this study, because 76% of cases were referred from this source. However, some delay in diagnosis has proven to be a not uncommon feature, in our experience. The classical bitemporal hemianopia was the most frequent field defect encountered, but asym­metry is the rule rather than the exception ( 10- 12). The fact that junctional scotomata were present in two patients underlines the importance of careful field examination of the fellow eye in patients pre­senting with an apparently unilateral scotoma. Headache was an infrequent presenting feature in our series ( 18% of cases), although others have re­ported this symptom in as many as 48% of cases ( 7,11). Surgery for these tumors is very effective, but the late onset of recurrences ( average 4 years post­operatively) requires prolonged regular follow- up by means of serial neuro- ophthalmic testing and neuroradiological imaging. In this study group, patients are seen every 3 months for the first year, and annually thereafter in the pituitary review clinic. Good results are achieved with surgery, and our series compares favorably with other operative se­ries, notably two large American series, both of which reported visual results for transsphenoidal surgery for pituitary adenoma ( 6,7). The results are summarized in Table 3. Postoperative visual acuity was improved in 74% of eyes with abnormal preoperative acuity, and postoperative visual fields were improved in 68% of eyes with preoperative field defects. The improvement in visual fields was more impressive for transsphenoidal surgery, with 81% of eyes im­proving their preoperative abnormal fields. Trans­sphenoidal surgery is faster, safer, more effective, and associated with less operative morbidity than craniotomy ( 7). The only open question on trans­sphenoidal surgery is that of long- term effective­ness and rate of recurrence ( 5). In our series only one patient had recurrence following transsphe­noidal surgery and was reoperated. However, the average length of follow- up for our transsphenoi­dal cases was just 1.4 years. Only long- term fol­low- up will answer the question of long- term ef­fectiveness. A prospective study is now in place to examine this factor. Several preoperative parameters were examined for correlation with postoperative results. Others had suggested positive predictive value for preop­erative visual acuity and an inverse correlation with age, duration of symptoms, and presence of optic atrophy ( 7). This study did not demonstrate a 75 64 63 55 72 66 68 81 56 Improved fields (%) 65 63 85 56 74 71 76 73 69 Improved acuity (%) Entire sample Exclude preop. normal Transsphenoidal Transfrontal Preop VA ~ 20/ 100 Preop VA < 20/ 100 Preop VF ~ 16 Preop VF < 16 Bilateral pallor/ atrophy Unilateral pallor/ atrophy No pallor/ atrophy TABLE 3. Results of transsphenoidal surgery NYMC ( 7) 1974- 84 ( transsphenoidal) MAYO ( 6) 1971- 82 ( transsphenoidal) SVH 1968- 87 ( whole sample) SVH 1983-- 87 ( transsphenoidal) Visual acuity ( preoperative normals excluded) Better 63% ( 65) 46% ( 53) 74% ( 45) Same 35% ( 36) 52% ( 60) 16% ( 10) Worse 3% - ill 2% -. i? l 16% . ill ( 104) ( 115) ( 61) Visual Fields ( preoperative normals excluded) Better 73% ( 134) 73% ( 168) 68% ( 59) Same 27% ( 50) 23% ( 52) 25% ( 22) Worse 0% - lJ 4% J..!. Ql 7% . ill ( 184) ( 230) ( 87) 71% ( 20) 18% ( 5) 11% ~ ( 28) 81% ( 34) 17% ( 7) 2% . J! l ( 42) .' · c · ". - I,,. ~ I ,:' .., nl~ r. MAYO. Mayo Clinic; SVH, St. Vincent's Hospital. VISION AND PITUITARY ADENOMA 267 positive correlation between visual improvement and preoperative acuity and fields. Preoperative optic atrophy was associated with less postopera­tive improvement of acuity and fields, but these figures did not reach statistical significance be­cause of small sample size. Age was not a signifi­cant predictor of visual outcome. This study supports the view that transsphenoi­dal hypophysectomy is effective in relieving visual morbidity due to anterior visual pathway compro­mise. Local surgical experience compares favor­ably with that of major overseas centers. Ophthalmologists are responsible for diagnosis in the majority of cases and the possibility of early diagnosis rests with them. Early diagnosis and sur­gery before the occurrence of optic atrophy could be expected to improve visual outcomes and should be the goal of all clinicians. Acknowledgment: We would like to thank Professor Hugh Taylor and Mr. K. Henderson for reviewing the manuscript, and Ms. S. Yaffe and Mr. C. Seater for help­ing prepare and type the manuscript. REFERENCES 1. Cushing H, Walker CB. Distortions of the visual fields in cases of brain tumour. Brain 1915; 37: 341- 400. 2. Miller NR, ed. Walsh and Hoyt's clinical neuroophthalmology. Baltimore: Williams & Wilkins, 1982: 119- 27. 3. Hardy J. Transsphenoidal hypophysectomy. J Neurosurg 1971; 34: 582- 594. 4. Ciric I, Mikhael M, Stafford T, et al. Transsphenoidal mi­crosurgery of pituitary macroadenomas with long- term fol­low- up results. J Neurosurg 1983; 569: 395-- 401. 5. Newman S. Advances in diagnoses and treatment of pitu­itary tumours. lnt Ophthal Clin 1988; 26: 28:>- 300. 6. Trautmann Je. Laws ER. Visual status after transsphenoi­dal surgery at the Mayo Clinic, 1971- 1982. Am J Ophthalmol 1983; 96: 200- 8. 7. Cohen AR, Cooper PR, Kupersmith MJ, et al. Visual recov­ery after transsphenoidal removal of pituitary adenomas. Neurosurgery 1985; 17: 446-- 452. 8. Repka MX, Miller NR, Miller M. Visual outcome after sur­gical removal of craniopharyngiomas. Ophthalmol 1989; 96: 195-- 9. 9. 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 ( Wien) 1983; 68: 17:>- 86. 10. Hardy J. Transsphenoidal surgery of hypersecreting pitu­itary tumours. In: Kohler P, Ross GT, eds. Diagnosis and treatment of pituitary tumors; proceedings of a conference. Ex­cerpta Medica. New York: Americal Elsevier Publishing, 1973: 179- 94. 11. Hollenhorst RW, Younge BR. Ocular manifestations pro­duced by adenomas of the pituitary gland; analysis of 1,000 cases. In: Kohler P, Ross GT, eds. Diagnosis and treatment of pituitary tumors; proceedings of a conference. Excerpta Medica. New York: Americal Elsevier Publishing, 1973: 53- 64. 12. Chamlin M, Davidoff LM, Feiring EH. Ophthalmologic changes produced by pituitary tumors. Am J Ophthalmol 1955; 40: 353-- 68. ] Clin Neuro- ophthalmol, Vol. 11, No. 4, 1991