Early Retinal Microcirculation in Nonfunctioning Pituitary Adenomas Without Visual Field Defects Using Optical Coherence Tomography Angiography

For patients with nonfunctioning pituitary adenoma (NFPA) without manifesting visual acuity impairment or visual field defect (VFD), more sensitive and objective assessment methods will allow earlier detection before irreversible damage to the visual system. This study aimed to evaluate retinal vessel densities (VDs) alterations in these patients using optical coherence tomography angiography and to determine its diagnostic abilities.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Early Retinal Microcirculation in Nonfunctioning Pituitary Adenomas Without Visual Field Defects Using Optical Coherence Tomography Angiography Yuyu Chou, MD, Xuqian Wang, MD, Ye Wang, PhD, Linyang Gan, MD, Bing Xing, MD, Hui You, MD, Jin Ma, PhD, Yong Zhong, PhD Background: For patients with nonfunctioning pituitary adenoma (NFPA) without manifesting visual acuity impairment or visual field defect (VFD), more sensitive and objective assessment methods will allow earlier detection before irreversible damage to the visual system. This study aimed to evaluate retinal vessel densities (VDs) alterations in these patients using optical coherence tomography angiography and to determine its diagnostic abilities. Methods: Between patients with NFPA without VFDs and age-matched, sex-matched healthy control individuals, comparisons of visual field metrics, retinal structural thickness, and microcirculation were conducted after adjusting for axial length (AL) and signal index of scans. Receiver operating characteristic (ROC) curves were further depicted to assess the diagnostic performance of significant parameters. To explore the impact of symptom duration, Department of Ophthalmology (YC, XW, LG, JM, YZ), Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China; Key Laboratory of Ocular Fundus Diseases (YC, XW, LG, JM, YZ) and Chinese Academy of Medical Sciences, Beijing, China; Department of Epidemiology and Statistics (YW), Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, School of Basic Medicine, and Peking Union Medical College, Beijing, China; Department of Neurosurgery (BX) and Radiology (HY), Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China; China Pituitary Disease Registry Center (YC, XW, BX, HY, JM, YZ), Beijing, China; and China Pituitary Adenoma Specialist Council (XW, BX, HY, JM, YZ), Beijing, China. The authors report no conflicts of interest. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the full text and PDF versions of this article on the journal’s Web site (www. jneuro-ophthalmology.com). This study is in accordance with the tenets of the Declaration of Helsinki. Its protocol was approved by the Institutional Review Board of Peking Union Medical College Hospital, Chinese Academy of Medical Sciences. Owing to the retrospective nature of this study, patient consent was waived. Address correspondence to Yong Zhong, PhD, No. 1 Shuaifuyuan, Dongcheng District, Beijing 100730, China; E-mail: zhongy_pumch@ 126.com Chou et al: J Neuro-Ophthalmol 2022; 42: 509-517 tumor size, and axial length on the significant parameters, multivariate regression analysis was conducted. Results: This cross-sectional study reviewed 107 patients with NFPA. Twenty-seven eyes of patients with NFPA without VFDs and 27 eyes of healthy controls were enrolled. Compared with healthy controls, patients with NFPA without VFDs had similar foveal avascular zone areas and perimeters, macular ganglion cell complex (mGCC) and peripapillary retinal nerve fiber layer thicknesses, and macular VDs. Only the VD in the radial peripapillary capillary (RPC) segment of the inferior temporal (IT) sector was much lower in the patient group. The 2 largest area under the ROC curves were the focal loss volume (FLV) of the mGCC and the VD in the RPC of the IT sector, both of which were significantly related to symptom duration and tumor size. Conclusions: At the early stage of NFPA before VFD and retinal thickness thinning, fundus microcirculation impairments may occur initially in the microvascular density of the RPC segment of the IT sector. The FLV and the VD of RPC at the IT sector may provide a basis for the early diagnosis of NFPA without VFD in clinical practice. Journal of Neuro-Ophthalmology 2022;42:509–517 doi: 10.1097/WNO.0000000000001562 © 2022 by North American Neuro-Ophthalmology Society N onfunctioning pituitary adenomas (NFPAs) are characterized by the absence of excessive hormones secretion (1). NFPAs can present as incidentalomas, and some patients manifest asymptomatic. These tumors are usually detected by MRI or CT by accident when scans are conducted for other reasons (2). With tumor growth, the mass effect of NFPAs may directly damage surrounding structures and functions, causing pituitary dysfunction and visual impairments. As previously reported, 50%–60% of people with NFPAs suffer from visual impairments, including vision loss and visual field defects (VFDs) (3,4), posing serious risks to people’s daily life and work capability. Ancillary ophthalmologic assessments for patients with NFPA mainly include visual acuity, perimetry, 509 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution and visual evoked potential (VEP) testing (5). However, acuity testing and perimetry are psychophysical measurements that are relatively subjective with unstable reliability, and the VEP testing is time-consuming and complicated operating. Supporting this notion, a faster, more sensitive, and objective detection method is needed. Moreover, partly because of the improvement and widespread application of radiographic techniques, pituitary incidentalomas were easier to be discovered. For these patients, visual field (VF) and VEP testing are not sufficient for assessing ophthalmologic impairments and assisting in early diagnosis (6). Under these circumstances, many studies have explored the clinical efficacy of optical coherence tomography (OCT) in NFPAs and have found that structural and functional changes in the ganglion cell complex (GCC) could be seen before VFDs were discernible (7). The NFPA-induced changes in OCT parameters were even earlier than the optic chiasma compression detected by MRI (8). Nevertheless, the GCC loss is irreversible and can gradually damage the VF (9,10). Considering the theory that ganglion cells have both high the metabolism and energy demands and they are sensitive to the supply of energy sources delivered by blood flow, we hypothesized that at the early stage of NFPAs retinal microcirculation might have changed along with GCC functional and/or structural alteration or even before GCC thickness changes. In this study, we conducted a cross-sectional study to evaluate retinal microcirculation and structural changes in patients with NFPA before VFDs by using optical coherence tomography angiography (OCTA). METHODS This study was in accordance with the tenets of the Declaration of Helsinki and approved by the Institutional Review Board of Peking Union Medical College Hospital (PUMCH), Chinese Academy of Medical Sciences. Study Design and Subjects The medical information of naive patients with NFPA without VFD and age-matched, sex-matched healthy control individuals were comprehensively reviewed at PUMCH in China from June 2019 to August 2020. We extracted the medical records before surgery, including chief complaints, symptom duration (defined as the period from the onset of self-reported symptoms to the diagnosis), ophthalmic examinations, pituitary MRIs, other demographic information, and systemic test results. The inclusion criteria for patients were older than 18 years and consistent with the diagnostic criteria of NFPA referring to the World Health Organization classification (11). Patients with NFPA without VFD were defined as having a mean deviation (MD) between 22.0 dB and 2.0 dB without obvious abnormalities of probability 510 graph methods (12). The control individuals were healthy PUMCH colleagues who had no abnormalities in craniocerebral MRI scans and ophthalmic tests. The exclusion criteria were (1) subjects with other ophthalmic or systemic comorbidities that might affect fundus microcirculation or imaging quality, such as diabetes and severe cataract; (2) subjects having received ocular or craniocerebral operation, radiotherapy, and endocrine therapy; (3) subjects with abnormal hormone secretion or with other sellar mass lesions such as craniopharyngioma; and (4) spherical equivalents larger than 3.0 diopters (D) as hyperopia or less than 26.0 D as high myopia. Suprasellar Extension Measurement The methods to measure suprasellar extension (SSE) were described in our previous article (13). In brief, sagittal SSE was accessed from the highest point of mass to the line drawn between the tuberculum sellae and dorsum sellea. Coronal SSE was defined as the distance between the highest point of NFPA and the level of the upper surfaces of the bilateral internal carotid artery. Ocular Examinations The outcomes of ocular examinations were recorded, including the best-corrected visual acuity (BCVA), axial length (AL), intraocular pressure (IOP), comprehensive slitlamp examinations, static VF examination, and OCT/OCTA scans. After correcting refractive errors, static automated perimetry was conducted with the Octopus 900 perimeter (Haag-Streit, Köniz, Switzerland). The reliability of perimetry was regarded as both the false positive and false negative rates ,20%. MD and mean sensitivity (MS) were extracted. OCTA and OCT scanning were performed by AngioVue RTVue XR Avanti (Optovue, Inc, Fremont, CA). AngioVue software predefined measured areas and automatically calculated vessel densities (VDs). Only highquality scans, of which signal strength index (SSI) $7/10, were enrolled in this study, including the angio retina scanning (HD 6 · 6 mm) at macular superficial and deep capillary plexus segments, the angio disc scanning (HD 4.5 · 4.5 mm) at radial peripapillary capillary (RPC) segment, macular GCC (mGCC) scanning, and optic nerve head program scanning. According to the Garway–Heath method, the peripapillary area was divided into 8 sectors (temporo inferior, TI; temporo superior TS; superior temporal, ST; superior nasal, SN; nasal superior, NS; nasal inferior, NI; inferior nasal, IN; and inferior temporal, IT). Based on the fovea-centered Early Treatment Diabetic Retinopathy Study grid, the angio retina scans were divided into the foveal zone, parafoveal, and perifoveal areas. Choroidal blood flow was considered as the ratio of the flow area in the choroid segment to the selected area (within the circle with a diameter of 3 mm) (Fig. 1). Chou et al: J Neuro-Ophthalmol 2022; 42: 509-517 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 1. Comparisons of visual field, vessel density, foveal avascular zone structure, macular ganglion cell complex, and peripapillary retinal nerve fiber layer thickness between NFPA eyes without visual field defect (2 right columns) and healthy control eyes (2 left columns). DCP, deep capillary plexus; FAZ, foveal avascular zone; MD, mean deviation; mGCC, macular ganglion cell complex; NFPA, nonfunctioning pituitary adenoma; pRNFL, peripapillary retinal nerve fiber layer thickness; RPC, radial peripapillary capillary; SCP, superficial capillary plexus. Statistical Analysis Continuous variables were recorded as the mean ± SD while categorical variables were presented as frequencies. We accessed the normality assumption using the visual method and the Shapiro–Wilk test. Independent sample t tests and Mann–Whitney U tests were used to compare the differences between patients and controls. Multivariate regression analysis after adjusting for the AL and the SSI of scans was performed to further confirm these differences. Moreover, receiver operating characteristic (ROC) curves were depicted to assess the diagnostic ability of significant parameters according to the aforementioned statistical analyses and clinical experience. We also recorded the area under the ROC curve (AUC) values, optimal cutoff value (determined when Youden Chou et al: J Neuro-Ophthalmol 2022; 42: 509-517 index [sensitivity + specificity 2 1] maximum), and the corresponding sensitivity and specificity. To explore the impact of symptom duration, tumor size, and AL on the significant parameters, multivariate regression analysis was conducted. P values less than 0.05 were considered statistically significant. Statistical analyses were performed using SPSS software. RESULTS We comprehensively reviewed 107 patients with NFPA during the study period. Twenty-seven eyes of 15 patients with NFPA without VFD met the inclusion criteria. Meanwhile, 27 eyes of 14 healthy controls were also included. In the NFPA without the VFD 511 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 1. Demographic and clinical information Age, yrs Sex (male/female) IOP, mm Hg AL of eyes, mm BCVA, logMAR MD, dB MS, dB SSI (HD disc 4.5 · 4.5 mm) SSI (HD retina 6 · 6 mm) Symptom duration, yrs Coronal SSE, cm Sagittal SSE, cm Chief complaints NFPA Without VFD, Mean (SD) Control, Mean (SD) 39.96 (12.56) 11/16 eyes 14.64 (2.25) 23.89 (1.00) 20.02 (0.05) 0.87 (1.09) 27.56 (1.36) 8.63 (0.79) 7.96 (0.76) 0.38 (0.35) 0.93 (0.32) 0.70 (0.19) No symptoms: 66.67% Visual impairment: 0% Headache: 25.93% NA: 7.40% 40.26 (12.86) 11/16 eyes 15.31 (2.08) 24.05 (1.17) 20.01 (0.06) 0.37 (0.72) 28.24 (0.83) 8.78 (0.70) 7.81 (0.83) P 0.876* 0.232* 0.656† 0.532* 0.05† 0.043† 0.370* 0.536* *Measured by the Mann–Whitney U test. † Measured by the independent samples t test. AL, axial length; BCVA, best-corrected visual acuity; IOP, intraocular pressure; MD, mean deviation of perimetry; MS, mean sensitivity of perimetry; NFPA, nonfunctioning pituitary adenoma; SSE, suprasellar extension; SSI, signal strength index; VFD, visual field defect. TABLE 2. Comparisons of structure of FAZ, optic disc, mGCC, and pRNFL thickness between the NFPA without the VFD group and the control group NFPA Without VFD, Mean (SD) FAZ area, mm2 FAZ perimeter, mm FAZ acircularity index FD-300 area density (%) Area C/D ratio Horizontal C/D ratio Vertical C/D ratio Cup volume, mm3 mGCC average, mm mGCC superior, mm mGCC inferior, mm FLV (%) GLV (%) pRNFL average, mm pRNFL NU, mm pRNFL NL, mm pRNFL IN, mm pRNFL IT, mm pRNFL TL, mm pRNFL TU, mm pRNFL ST, mm pRNFL SN, mm 0.34 2.23 1.09 55.20 0.33 0.61 0.51 0.20 98.85 99.10 98.61 0.62 2.22 105.37 81.56 71.89 113.73 149.56 70.59 85.86 148.51 121.25 (0.10) (0.37) (0.02) (3.66) (0.18) (0.22) (0.20) (0.19) (7.09) (7.43) (6.97) (0.05) (2.91) (10.31) (16.06) (9.44) (16.24) (16.86) (10.53) (13.39) (18.92) (20.65) Control, Mean (SD) 0.32 2.17 1.09 55.66 0.40 0.68 0.59 0.23 98.56 98.49 98.64 0.42 1.97 103.75 82.24 73.47 115.02 143.47 69.71 86.00 137.78 118.76 (0.09) (0.30) (0.03) (2.98) (0.13) (0.15) (0.12) (0.20) (6.06) (6.23) (6.34) (0.34) (2.75) (6.09) (12.75) (10.96) (12.51) (20.04) (11.03) (14.47) (18.54) (21.59) P 0.384* 0.460* 0.877* 0.618* 0.143* 0.191† 0.083* 0.446† 0.842† 0.647† 0.910† 0.039† 0.775† 0.210† 0.723† 0.575* 0.745* 0.232* 0.539† 0.924† 0.033† 0.910† The peripapillary area to measure pRNFL thickness was evenly divided into 8 subsectors by Optovue: nasal upper (NU), nasal lower (NL) inferior nasal (IN), inferior temporal (IT), temporo lower (TL), temporo upper (TU), superior temporal (ST), and superior nasal (SN) sectors. The bold values indicate P values ,0.05, which is considered statistically significant. *Measured by the independent samples t test. † Measured by the Mann–Whitney U test. C/D, cup-to-disc ratio; FAZ, foveal avascular zone; FD-300, foveal density-300 mm; FLV, focal loss of volume of the mGCC; GLV, general loss of volume of the mGCC; mGCC, macular ganglion cell complex thickness; NFPA, nonfunctioning pituitary adenoma; pRNFL, peripapillary retinal nerve fiber layer thickness; VFD, visual field defect. 512 Chou et al: J Neuro-Ophthalmol 2022; 42: 509-517 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution group, 1 eye with the epiretinal membrane and 2 eyes suffering from glaucoma were excluded. One eye of healthy individuals was excluded because of the low SSI. As Table 1 presents, there was no significant difference in age, sex, IOP, AL, or BCVA between patients and healthy controls. Compared with those in healthy subjects, patients had worse MS (P = 0.043) and were slightly different in MD with marginal significance (P = 0.05). However, after adjusting for AL, the differences in MS and MD between patients and healthy individuals became nonsignificant (all P . 0.2, Table 5). All included patients were with optic chiasm compression to varying degrees, except for 1 participant. The coronal SSE and sagittal SSE were 0.93 ± 0.32 cm and 0.70 ± 0.19 cm, respectively. The symptom duration of patients was 0.38 ± 0.35 years. A total of 66.67% of patients declared no symptoms or signs, and 25.93% of patients presented headache as their chief complaint. None of patients complained of visual impairments. Comparisons of the structure of the foveal avascular zone (FAZ), optic disc, mGCC, and peripapillary retinal nerve fiber layer thickness (pRNFL) were conducted between the NFPA without the VFD group and the control group (Table 2). No significant difference was found in FAZ metrics and optic disc structure. Most mGCC and pRNFL thickness parameters were also similar between the patient group and the control group (all P . 0.05), whereas the focal loss volume of the mGCC (FLV) and the pRNFL thickness of the ST sector of patients were obviously larger (P = 0.039 and P =0.033, respectively). After adjusting for AL, those significant differences became nonsignificant (all P . 0.085) (Table 5). TABLE 3. Comparison of the vessel density centered on the fovea between the NFPA without the VFD group and the control group NFPA Without VFD, Mean (SD)% Superficial capillary plexus Whole image Superior hemiretina Inferior hemiretina Fovea Parafovea Inferior Nasal Superior Temporal Perifovea Inferior Nasal Superior Temporal Deep capillary plexus Whole image Superior hemiretina Inferior hemiretina Fovea Parafovea Inferior Nasal Superior Temporal Perifovea Inferior Nasal Superior Temporal Choroidal blood flow Control, Mean (SD)% P 49.95 50.14 49.77 16.94 52.11 51.69 51.62 52.79 52.34 50.74 50.58 54.56 50.83 47.01 (2.51) (2.73) (2.41) (7.18) (2.75) (3.13) (3.47) (3.66) (2.71) (2.57) (2.63) (2.63) (3.16) (2.65) 49.34 49.47 49.20 16.80 51.86 52.40 50.42 52.65 51.96 50.02 50.24 54.24 50.06 45.58 (2.22) (2.07) (2.51) (5.48) (2.65) (3.45) (3.52) (2.99) (2.38) (2.32) (2.75) (2.42) (2.63) (2.59) 0.349* 0.318* 0.399* 0.937* 0.732* 0.207† 0.213* 0.878* 0.622* 0.286* 0.640* 0.639* 0.335* 0.076† 48.35 49.30 47.48 32.01 54.47 53.03 55.69 53.88 55.26 49.32 47.04 48.06 48.99 53.18 71.24 (4.66) (4.68) (4.79) (7.76) (3.52) (4.28) (3.54) (4.20) (3.44) (5.20) (6.59) (5.46) (5.99) (4.05) (2.28) 47.72 48.43 47.09 34.14 54.37 52.86 55.68 53.82 55.11 48.67 47.13 46.99 47.60 52.98 69.64 (5.31) (5.55) (5.23) (6.69) (3.90) (5.22) (3.41) (4.47) (3.60) (6.02) (6.76) (6.73) (6.98) (4.90) (4.38) 0.646* 0.536* 0.776* 0.284* 0.923* 0.893* 0.897† 0.962* 0.876* 0.675* 0.957* 0.523* 0.436* 0.870* 0.122† The parafoveal and perifoveal areas were divided into 4 quadrants in accordance with the Early Treatment Diabetic Retinopathy Study grid. The bold values indicate P values ,0.05, which is considered statistically significant. *Measured by the independent samples t test. † Measured by the Mann–Whitney U test. NFPA, nonfunctioning pituitary adenoma; VFD, visual field defect. Chou et al: J Neuro-Ophthalmol 2022; 42: 509-517 513 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution Among different sectors of foveal, parafoveal, and perifoveal areas, the alterations of VDs in SCP or DCP segments between the patient group and the control group were nonsignificant (all P . 0.05) (Table 3). Our results demonstrated that the choroidal blood flow of patients was larger than that of healthy individuals, but the difference was not statistically significant (P = 0.122). Centered on the optic disc, the capillary density inside the disc of patients obviously increased (P = 0.035), and the VD in the RPC segment of the IT sector of patients was much lower (P = 0.009) (Table 4). No significant changes in the density of all vessels (both large and capillary vessels) were observed. After adjusting for AL and SSI, we found that the VD in the RPC segment at the IT subsector of patients was still significantly decreased compared with that of healthy controls (P = 0.010), and the difference in capillary density inside disc became nonsignificant (P = 0.168) (Table 5). The ROC curves were conducted in MD, MS, BCVA, mGCC, pRNFL, and RPC. Only the FLV and the VD in the RPC of the IT sector presented their significant diagnostic ability (P = 0.04 and P = 0.02, respectively, Fig. 2). As presented in Table 6, the AUC of the FLV was 0.66 (95% confidence interval [CI], 0.52–0.81), and its cutoff value was 0.54%. For VD in the RPC segment of the IT sector, the AUC was 0.68 (95% CI, 0.54–0.82), and the cutoff value was 59.84% (Table 6). The results also revealed that the FLV was positively related to symptom duration (P , 0.001), coronal SSE (P = 0.01), and AL (P , 0.001), whereas the microcirculation in the RPC of the IT sector was negatively associated with symptom duration, coronal SSE, and AL (all P , 0.001) (Table 7). CONCLUSIONS This study is the first to explore the fundus microcirculation of patients with NFPA before VFD and to compare the diagnostic ability of several structural and functional parameters. Compared with healthy controls, patients with NFPA without VFD had similar FAZ structure, mGCC and pRNFL thicknesses, and macular VD after adjusting for AL and SSI. Only the VD in the RPC segment of the IT sector was much lower in the patient group. The 2 largest AUCs indicating the diagnostic performance were the FLV and the VD in the RPC of the IT sector, both of which were significantly related to symptom duration and tumor size. No significant differences were found in MD and retinal structural measurements between patients and TABLE 4. Comparison of the vessel density centered on optic disc between the NFPA without the VFD group and the control group NFPA Without VFD, Mean (SD)% Capillary density Whole image Inside disc Peripapillary average Superior hemi Inferior hemi RPC IT RPC IN RPC NI RPC NS RPC SN RPC ST RPC TS RPC TI All vessel density Whole image Inside disc Peripapillary average Superior hemi Inferior hemi Control, Mean (SD)% P 50.20 53.02 52.14 52.16 52.11 57.60 51.25 47.63 47.79 49.42 56.25 57.13 53.61 (2.46) (3.53) (3.05) (3.31) (3.00) (3.66) (4.20) (4.41) (3.84) (4.97) (5.22) (2.94) (4.06) 50.49 50.56 53.25 53.49 52.98 60.08 52.36 46.97 49.43 50.81 57.32 57.87 54.57 (1.64) (4.71) (1.94) (2.19) (2.21) (3.03) (3.86) (5.18) (2.84) (3.20) (3.95) (3.21) (3.39) 0.609* 0.035* 0.116* 0.087* 0.231* 0.009* 0.313* 0.616* 0.079* 0.229* 0.403* 0.390* 0.364† 56.54 61.78 58.48 58.70 58.26 (2.66) (2.93) (3.12) (3.22) (3.16) 57.07 60.14 59.82 60.28 59.31 (1.55) (3.77) (1.65) (1.80) (1.83) 0.373* 0.081* 0.055* 0.084† 0.140* According to the “Garway Heath” method, the peripapillary area of angio disc scan was divided into 8 subsectors: inferior temporal (IT), inferior nasal (IN), nasal inferior (NI), nasal superior (NS), superior nasal (SN), superior temporal (ST), temporo superior (TS), and temporo inferior (TI). The bold values indicate P values ,0.05, which is considered statistically significant. *Measured by the independent samples t test. † Measured by the Mann–Whitney U test. NFPA, nonfunctioning pituitary adenoma; RPC, radial peripapillary capillary; VFD, visual field defect. 514 Chou et al: J Neuro-Ophthalmol 2022; 42: 509-517 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 5. Multivariable analyses between the NFPA without the VFD group and the control group Parameters pRNFL ST, mm mGCC FLV (%) MD, dB MS, dB VD inside optic disc (%) RPC IT (%) Estimate Standard Error 29.29 20.24 20.37 0.38 21.86 3.00 5.31 0.13 0.29 0.34 1.33 1.10 P 0.088 0.085 0.217 0.272 0.168 0.010 Multivariable regression analyses were conducted after adjusting for axial length of the eyes and signal strength index of scans to confirm the differences in VDs between patients and healthy controls. The bold values indicate P values ,0.05, which is considered statistically significant. GCC FLV, focal loss of volume of macular ganglion cell complex thickness; MD, mean deviation of perimetry; MS, mean sensitivity of perimetry; NFPA, nonfunctioning pituitary adenoma; pRNFL ST, peripapillary retinal nerve fiber layer thickness of superior temporal sector; RPC IT, radial peripapillary capillary density of inferior temporal sector; VD, vessel density; VFD, visual field defect. healthy individuals, whereas the differences in MS, the FLV, and the pRNFL thickness of the ST sector were significant. Although the inclusion criteria strictly restricted MD between 22 dB and 2 dB, the sensitivity of perimetry in patients was worse than that in controls, which was hardly noticed by the patients themselves in daily life. In accordance with previous studies (8,14,15), our results also showed that the attenuation of retinal thickness could be earlier than VF dysfunction, especially mGCC metrics. However, after adjusting for AL, the differences in those parameters mentioned above became nonsignificant. The AL of the eyeball has already been proven to be negatively related to retinal VDs (16). Thus, considering the confounding effect of AL on fundus microcirculation, our cross-sectional study was possibly conducted at a relatively early stage of the disease before VFD and structural thickness thinning. In addition, between patients with NFPA without VFD and healthy subjects, only the VD in the RPC segment of the IT sector demonstrated obvious differences, whereas the alterations in macular microvascular densities were not statistically significant. Unlike previous reports, Cennamo et al and Lee et al showed that macular VDs of patients remarkably decreased (17–19). This might suggest that microvascular alterations in NFPA begin at the peripapillary area. We also conducted a subgroup analysis only among the patients with optic chiasm compression. After excluding the patient without chiasmal compression, the peripapillary VD and the RPC layer VD of the IT sector of the patients were found to be significantly decreased compared with controls (see Supplemental Digital Content, Tables 1–5, http://links. lww.com/WNO/A568).This could indicate that at the early stage of NFPA before VFD and retinal thickness thinning, the ocular impairment initially occurs in microcirculation of the RPC segment at the IT sector. The VD in this sector was also significantly attenuated in the early stage of glaucoma and related to the progression of DR (20,21). It could be explained by that the capillaries in this area have narrower diameters and larger perfusion volumes (22). By contrast, Dalloroto et al reported no FIG. 2. Receiver operating characteristic (ROC) curves of 2 significant parameters between healthy controls and nonfunctioning pituitary adenoma without visual field defect patients. A. The ROC of focal loss volume of macular ganglion cell complex; (B) The ROC of vessel density in radial peripapillary capillary segment of inferior temporal sector. Chou et al: J Neuro-Ophthalmol 2022; 42: 509-517 515 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 6. Receiver operating characteristic (ROC) curves of different measurements between patients with NFPA without VFD and healthy individuals AUC (SE) MD MS BCVA mGCC average mGCC FLV pRNFL average RPC average RPC IT 0.64 0.65 0.54 0.52 0.66 0.60 0.61 0.68 (0.08) (0.08) (0.08) (0.08) (0.07) (0.08) (0.08) (0.07) 95% CI P Cutoff 0.49–0.79 0.50–0.79 0.39–0.70 0.36–0.68 0.52–0.81 0.44–0.76 0.46–0.76 0.54–0.82 0.08 0.07 0.60 0.84 0.04 0.21 0.16 0.02 — — — — 0.54% — — 59.84% Sensitivity Specificity — — — — 51.85% — — 74.07% — — — — 81.48% — — 59.26% The bold values indicate P values ,0.05, which is considered statistically significant. AUC, the area under the receiver operating characteristic curve; AUC, the area under the receiver operating characteristic curve; BCVA, best-corrected visual acuity; CI, confidence interval; MD, mean deviation of perimetry; mGCC FLV, focal loss of volume of mGCC; mGCC, macular ganglion cell complex thickness; MS, mean sensitivity of perimetry; NFPA, nonfunctioning pituitary adenoma; pRNFL, peripapillary retinal nerve fiber layer thickness; RPC IT, RPC of inferior temporal sector; RPC, radial peripapillary capillary density; SE, standard error; VFD, visual field defect. significant difference between pituitary patients and healthy individuals (23). Nevertheless, their study only enrolled 8 participants, and the measured area was not divided into subsectors. The sex and age of the control group did not match those of the patient group. For diagnostic performance, the FLV and the VD of RPC at the IT sector were much better than functional measurements (BCVA, MS, and MD) and structural parameters (pRNFL and mGCC thickness). Although the diagnostic ability of these 2 metrics was limited, they may provide a basis for the early diagnosis of NFPA without VFD in clinical practice. For suspected patients without a history of NFPAs, the finding of abnormal FLV and sectoral VD may prompt further investigations, such as neuroimaging. For the patients with NFPAs, these findings may hint the adjustment of the intervals of the clinical follow-up. Besides, the precise pathogenesis of ganglion cell impairment affected by optic chiasm compression and the causality between retinal microcirculation and ganglion cell loss remains poorly understood (24). This cross-sectional study observed that retinal microcirculation was first affected by NFPA before VFD and retinal thickness thinning. Admittedly, this study had some limitations. Partly to do with the patients with NFPA without VFD seldom attending to hospitals and ophthalmologists lack awareness of the efficacy of OCTA in NFPA, our study sample size was relatively small. The AL of patients did not match that of healthy individuals, but we used multivariate regression analysis to adjust for those confounding factors. We did not have sufficient data to evaluate the patients’ color vision and contrast sensitivity, which were also significant to the diagnosis of optic neuropathies. Consequently, large-scale, prospective, and high-quality studies are required to confirm the alterations in VDs before patients with NFPA experience VFD and retinal structural loss. In summary, this study was possibly conducted at a relatively early stage of NFPA before VFD and retinal thickness loss. In this case, fundus vessel impairments may occur initially in the VD of the RPC segment of the IT sector. The VD in this area and FLV could provide information to assist the early diagnosis of NFPA without VFD. In addition, these 2 parameters were significantly associated with sellar mass size and symptom duration. TABLE 7. Multivariable analyses of risk factors for mGCC FLV and RPC IT in the NFPA without the VFD group mGCC FLV Symptom duration Coronal SSE Sagittal SSE AL of eyes RPC IT Estimate SE P VIF Estimate SE P VIF 2.05 1.50 0.60 0.33 0.32 0.37 0.92 0.07 ,0.001 0.01 0.54 ,0.001 2.51 2.46 1.97 2.89 228.85 224.52 217.40 24.86 2.95 3.40 8.50 0.63 ,0.001 ,0.001 0.10 ,0.001 2.51 2.46 1.97 2.89 The bold values indicate P values , 0.05, which is considered statistically significant. AL, axial length; mGCC FLV, focal loss of volume of macular ganglion cell complex; NFPA, nonfunctioning pituitary adenoma; RPC IT, radial peripapillary capillary density of inferior temporal sector; SE, standard error; SSE, suprasellar extension; VFD, visual field defect; VIF, variance inflation factor. 516 Chou et al: J Neuro-Ophthalmol 2022; 42: 509-517 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: Y. Zhong and Y. Chou; b. Acquisition of data: X. Wang, B. Xing, and H. You; c. Analysis and interpretation of data: Y. Wang, L. Gan, and Y. Chou. Category 2: a. Drafting the manuscript: Y. Chou; b. Revising it for intellectual content: B. Xing, H. You, L. Gan, and J. Ma. Category 3: a. Final approval of the completed manuscript: Y. Chou, X. Wang, Y. Wang, L. Gan, B. Xing, H. You, J. Ma, and Y. Zhong. REFERENCES 1. Greenman Y, Stern N. Non-functioning pituitary adenomas. Best Pract Res Clin Endocrinol Metab. 2009;23:625–638. 2. Galland F, Vantyghem MC, Cazabat L, Boulin A, Cotton F, Bonneville JF, Jouanneau E, Vidal-Trécan G, Chanson P. Management of nonfunctioning pituitary incidentaloma. Ann Endocrinol (Paris). 2015;76:191–200. 3. Ferrante E, Ferraroni M, Castrignanò T, Menicatti L, Anagni M, Reimondo G, Del MP, Bernasconi D, Loli P, Faustini-Fustini M, Borretta G, Terzolo M, Losa M, Morabito A, Spada A, BeckPeccoz P, Lania AG. Non-functioning pituitary adenoma database: a useful resource to improve the clinical management of pituitary tumors. Eur J Endocrinol. 2006;155:823–829. 4. Di Somma C, Scarano E, de Alteriis G, Barrea L, Riccio E, Arianna R, Savastano S, Colao A. Is there any gender difference in epidemiology, clinical presentation and co-morbidities of non-functioning pituitary adenomas? A prospective survey of a national referral center and review of the literature. J Endocrinol Invest. 2021;44:957–968. 5. Abouaf L, Vighetto A, Lebas M. Neuro-ophthalmologic exploration in non-functioning pituitary adenoma. Ann Endocrinol (Paris). 2015;76:210–219. 6. Messerer M, Dubourg J, Raverot G, Bervini D, Berhouma M, George I, Chacko AG, Perrin G, Levivier M, Daniel RT, Trouillas J, Jouanneau E. Non-functioning pituitary macro-incidentalomas benefit from early surgery before becoming symptomatic. Clin Neurol Neurosurg. 2013;115:2514–2520. 7. Chou Y, Zhang B, Gan L, Ma J, Zhong Y. Clinical efficacy of optical coherence tomography in sellar mass lesions: a metaanalysis. Pituitary. 2020;23:733–744. 8. Cennamo G, Auriemma RS, Cardone D, Grasso LF, Velotti N, Simeoli C, Di Somma C, Pivonello R, Colao A, de Crecchio G. Evaluation of the retinal nerve fibre layer and ganglion cell complex thickness in pituitary macroadenomas without optic chiasmal compression. Eye (Lond). 2015;29:797– 802. 9. Tieger MG, Hedges TR, Ho J, Erlich-Malona NK, Vuong LN, Athappilly GK, Mendoza-Santiesteban CE. Ganglion cell complex loss in chiasmal compression by brain tumors. J Neuroophthalmol. 2017;37:7–12. 10. Monteiro M. Macular ganglion cell complex reduction preceding visual field loss in a patient with chiasmal compression with a 21-month follow-up. J Neuroophthalmol. 2018;38:124–127. Chou et al: J Neuro-Ophthalmol 2022; 42: 509-517 11. Sarkar S, Philip VJ, Cherukuri SK, Chacko AG, Chacko G. Implications of the world health organization definition of atypia on surgically treated functional and non-functional pituitary adenomas. Acta Neurochir (Wien). 2017;159:2179–2186. 12. Jones PR, Lindfield D, Crabb DP. Using an open-source tablet perimeter (eyecatcher) as a rapid triage measure for glaucoma clinic waiting areas. Br J Ophthalmol. 2021;105:681–686. 13. Gan L, Ma J, Feng F, Wang Y, Cui J, Guo X, Zhang X, You H, Wang Z, Yin Z, Zhong Y, Xing B. The predictive value of suprasellar extension for visual function evaluation in Chinese patients with nonfunctioning pituitary adenoma with optic chiasm compression. World Neurosurg. 2018;116:e960– e967. 14. Yum HR, Park SH, Park HY, Shin SY. Macular ganglion cell analysis determined by cirrus hd optical coherence tomography for early detecting chiasmal compression. PLoS One. 2016;11:e153064. 15. Zhang X, Ma J, Wang YH, Gan LY, Li L, Wang XQ, Li DH, Xing B, Feng M, Zhu HJ, Lu L, Feng F, You H, Zhang ZH, Zhong Y. The correlation of ganglion cell layer thickness with visual field defect in non-functional pituitary adenoma with chiasm compression [in Chinese]. Zhonghua Yan Ke Za Zhi. 2019;55:186–194. 16. You QS, Chan J, Ng A, Choy B, Shih KC, Cheung J, Wong J, Shum J, Ni MY, Lai J, Leung GM, Cheung C, Wong TY, Wong I. Macular vessel density measured with optical coherence tomography angiography and its associations in a large population-based study. Invest Ophthalmol Vis Sci. 2019;60:4830–4837. 17. Lee GI, Park KA, Oh SY, Kong DS. Parafoveal and peripapillary perfusion predict visual field recovery in chiasmal compression due to pituitary tumors. J Clin Med. 2020;9:697. 18. Cennamo G, Solari D, Montorio D, Scala MR, Melenzane A, Fossataro F, Somma T, Tranfa F, Cavallo LM. Early vascular modifications after endoscopic endonasal pituitary surgery: the role of oct-angiography. PLoS One. 2020;15:e241295. 19. Lee GI, Park KA, Oh SY, Kong DS. Analysis of optic chiasmal compression caused by brain tumors using optical coherence tomography angiography. Sci Rep. 2020;10:2088. 20. Mansoori T, Sivaswamy J, Gamalapati JS, Balakrishna N. Radial peripapillary capillary density measurement using optical coherence tomography angiography in early glaucoma. J Glaucoma. 2017;26:438–443. 21. Custo GE, Brigell M, Cao F, Levine ES, Peters K, Moult EM, Fujimoto JG, Waheed NK. Macular and peripapillary optical coherence tomography angiography metrics predict progression in diabetic retinopathy: a sub-analysis of time-2b study data. Am J Ophthalmol. 2020;219:66–76. 22. Yu PK, Cringle SJ, Yu DY. Correlation between the radial peripapillary capillaries and the retinal nerve fibre layer in the normal human retina. Exp Eye Res. 2014;129:83–92. 23. Dallorto L, Lavia C, Jeannerot AL, Shor N, Jublanc C, Boch AL, Bodaghi B, Tadayoni R, Touitou V, Bonnin S. Retinal microvasculature in pituitary adenoma patients: is optical coherence tomography angiography useful? Acta Ophthalmol. 2020;98:e585–e592. 24. Danesh-Meyer HV, Yoon JJ, Lawlor M, Savino PJ. Visual loss and recovery in chiasmal compression. Prog Retin Eye Res. 2019;73:100765. 517 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.