Predictive value of retinal nerve fibre layer thickness for postoperative visual improvement in patients with pituitary macroadenoma

Objective To determine the usefulness of retinal nerve fibre layer (RNFL) thickness and a reliable cut-off value that can predict postoperative visual function improvement in patients with pituitary macroadenoma. Methods and Analysis This was a prospective observational study. Preoperative optical coherence tomography of the RNFL was performed in patients with pituitary macroadenoma. Postoperative visual function changes (acuity and visual fields) were identified using predefined criteria. Receiver operating characteristic curves were constructed for RNFL values to define the ideal cut-off value that predicted improvement. Other variables including preoperative visual acuity, mean deviation, visual field index and tumour volume were also analysed. Results Twenty-nine eligible subjects (58 eyes) were recruited. The mean (+/- SD) age was 43.9 (+/- 12.85) years and 65.5% were male. The mean (+/- SE) follow-up duration was 20.8 (+/- 6.42) months. RNFL thickness was significantly thinner in eyes with visual dysfunction and optic disc pallor. Better preoperative logarithmic minimum angle of resolution (logMAR) visual acuity, higher RNFL thickness and smaller tumour volume were associated with postoperative visual field improvement on univariate analysis; however, only mean RNFL thickness had significant association on multivariate analysis. None of the preoperative variables showed significant association with improvement in visual acuity. The best cut-off of mean RNFL thickness for visual field improvement was estimated at 81 mu m with 73.1% sensitivity and 62.5% specificity. Conclusion Preoperative RNFL thickness can be an objective predictor of visual field outcomes in patients undergoing surgery for pituitary macroadenomas, with moderate sensitivity and specificity. It is, however, not a good predictor of visual acuity outcome.