Visual Field Outcome Reporting in Neurosurgery: Lessons Learned from a Prospective, Multicenter Study of Transsphenoidal Pituitary Surgery

OBJECTIVE: Visual field (VF) outcomes are commonly reported in neurosurgical case series; however, substantial variability can exist in VF testing and outcome reporting. We aimed to evaluate the challenges of VF testing and to develop detailed recommendations for VF outcome reporting by analyzing results from an ongoing, multicenter study of transsphenoidal pituitary surgery. METHODS: VF testing results were collected during a prospective, multicenter clinical trial evaluating patient outcomes after transsphenoidal surgery for nonfunctioning pituitary adenomas (TRANSSPHER). Two independent ophthalmologists reviewed reliability and outcomes of all VF studies. Preoperative and postoperative VF studies were evaluated individually and as preoperative/postoperative pairs. RESULTS: Suboptimal perimetry field settings were reported in 37% of VF studies. Automated reliability criteria flagged 25%-29% of VF studies as unreliable, whereas evaluation by 2 independent ophthalmologists flagged 16%-28%. Agreement between automated criteria and raters for VF reliability was inconsistent (K coefficients = 0.55-0.83), whereas agreement between the 2 raters was substantial to almost perfect (K coefficients = 0.78-0.83). Most patients demonstrated improvement after surgery (rater 1, 67%; rater 2, 60%), with substantial rater agreement on outcomes for paired examinations (K coefficient = 0.62). CONCLUSIONS: VF outcome studies demonstrated significant variability of test parameters and patient performance. Perimetry field settings varied among patients and for some patients varied preoperatively versus postoperatively. Reliance on automated criteria alone could not substitute for independent ophthalmologist review of test reliability. Standardized guidelines for VF data collection and reporting could increase reliability of results and allow better comparisons of outcomes in future studies.