Advertisement

A Nomenclature to Describe the Sequence of Visual Field Defects in Progressive Thyroid Eye Disease–Compressive Optic Neuropathy (An American Ophthalmological Society Thesis)

  • Suzanne K. Freitag
    Correspondence
    Inquiries to Suzanne K. Freitag, Ophthalmic Plastic Surgery Service, Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, 243 Charles St, Boston, MA 02114, USA
    Affiliations
    Ophthalmic Plastic Surgery Service, Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, Massachusetts
    Search for articles by this author
  • Thidarat Tanking
    Affiliations
    Department of Ophthalmology, Somdech Phrapinklao Hospital, Royal Thai Navy, Bangkok, Thailand
    Search for articles by this author
Published:December 13, 2019DOI:https://doi.org/10.1016/j.ajo.2019.12.005

      Purpose

      To create a novel nomenclature to characterize the longitudinal sequence of visual field (VF) defects in patients with progression of thyroid eye disease–compressive optic neuropathy (TED-CON).

      Methods

      A retrospective review of records from 1 institution identified patients with progressive Humphrey VF defects secondary to TED-CON. The VF defects were analyzed by 2 independent reviewers and classified into 1 of 10 categories, divided into 3 stages that reflect the observed progression pattern, plus a miscellaneous category (stage X). Stage 1 VF defects are the earliest detectable and involve the inferior visual field with 3 levels of severity. Stage 2 VF defects include 2 distinguishable levels of severity and occur as the inferior defects advance above the horizontal midline to involve the superior VF. Stage 3 involves progression of stage 2 VF defects to complete loss of inferior and superior hemifields.

      Results

      Of 234 VFs in 37 eyes of 23 subjects, inferior defects were most common, including stage 1a (small inferior paracentral defect) in 22 of 234 VFs (9.4%), stage 1b (large inferior paracentral defect) in 112 of 234 VFs (47.9%), and stage 1c (inferior altitudinal defect) in 11 of 234 VFs (4.7%). Stage 2a (inferior altitudinal with superior advancement above the horizontal meridian) occurred in 41 of 234 VFs (17.5%), stage 2b (inferior altitudinal with superior arcuate) occurred in 6 of 234 VFs (2.6%), and stage 3 (total loss) occurred in 5 of 234 VFs (2.1%). The longitudinal sequence of VF defects from the 37 eyes of 23 patients was analyzed. Thirty-one of 37 eyes (83.8%) demonstrated a predictable progression pattern from least to more severe: stage 1a, stage 1b, stage 1c, stage 2a, stage 2b, and stage 3. A reverse order of VF defect progression was noted in 15 eyes with improving TED-CON. A minority of progression patterns (16.2%) originated from stage X (central/paracentral, enlarged blind spot, and scatter).

      Conclusions

      Humphrey VF defects resulting from TED-CON are most often inferior, often have a predictable pattern of progression, and can be categorized into a novel descriptive nomenclature system. NOTE: Publication of this article is sponsored by the American Ophthalmological Society.
      Thyroid eye disease (TED) is strongly associated with thyroid autoimmune disease, and affects 25%-50% of patients with Graves disease.
      • Victores A.J.
      • Takashima M.
      Thyroid eye disease: optic neuropathy and orbital decompression.
      The pathogenesis of TED is in part caused by the binding of autoantibodies to orbital antigens resulting in orbital inflammation. Nunery and associates described 2 subtypes of disease: type 1, involving increased orbital fat volume with no extraocular muscle (EOM) enlargement, and type 2, involving increased EOM volume and restrictive motility leading to diplopia.
      • Nunery W.R.
      • Nunery C.W.
      • Martin R.T.
      • Troung T.V.
      • Osborn D.R.
      The risk of diplopia following orbital floor and medial wall decompression in subtypes of ophthalmic Graves’ disease.
      Thickening of EOMs in the orbital apex, which is seen in about 70% of patients with TED,
      • Wiersinga W.M.
      • Regensburg N.I.
      • Mourits M.P.
      Differential involvement of orbital fat and extraocular muscles in Graves’ ophthalmopathy.
      also contributes to compressive optic neuropathy (CON).
      The specific pathogenesis of EOM enlargement in TED has not yet been elucidated. Histopathology reveals infiltration of CD4+ T cells with some B cells, plasma cells, and macrophages.
      • Bahn R.S.
      Graves’ ophthalmopathy.
      Glycosaminoglycan deposition also plays a role in EOM edema, which is thought to occur late in the disease process. EOM fibroblasts show specific properties that distinguish them from other skeletal muscle, including a difference in the complement of sarcomeric myosin heavy chain isoforms.
      • Mascarello F.
      • Toniolo L.
      • Cancellara P.
      • et al.
      Expression and identification of 10 sacromeric MyHC isoforms in human skeletal muscles of different embryologic origin. Diversity and similarity in mammalian species.
      Calsequestrin, a calcium-binding protein, is thought to play a role in EOM enlargement because its gene is 4.7 times more expressed in EOM than other skeletal muscles. In addition, the roles of several antibodies targeting EOM proteins have been recently considered. Autoantibodies targeting protein antigens G2s, the terminal end of transcription factor FOXP1, and Fp, the flavoprotein subunit of succinate dehydrogenase, are sensitive markers of EOM damage in TED, although neither is specific to EOM or TED.
      • Mizokami T.
      • Salvi M.
      • Wall J.R.
      Eye muscle antibodies in Graves’ ophthalmopathy: pathogenic or secondary epiphenomenon?.
      Regardless of the specific etiology, it is established that an increase in volume of the EOMs at the orbital apex can lead to compression of the optic nerve. COM is uncommon, occurring in 5%-8.6% of patients with TED, but urgent recognition and treatment are required in order to prevent permanent visual loss.
      • Verity D.H.
      • Rose G.E.
      Acute thyroid eye disease (TED): principles of medical and surgical management.
      ,
      • Neigel J.M.
      • Rootman J.
      • Belkin R.I.
      • et al.
      Dysthyroid optic neuropathy. The crowded orbital apex syndrome.
      The diagnosis of CON in TED is based on a constellation of clinical features, including decreased visual acuity, dyschromatopsia, relative afferent pupillary defect, optic disc edema, visual field (VF) defects, apical crowding on orbital imaging, and reduced amplitude on pattern electroretinogram or visual evoked potentials.
      • Przemyslaw P.
      • Janusz M.
      • Alina B.L.
      • Maria G.
      Pattern electroretinogram (PERG) in the early diagnosis of optic nerve dysfunction in the course of Graves’ orbitopathy.
      • Perez-Rico C.
      • Rodriguez-Gonzalez N.
      • Arevalo-Serrano J.
      • Blanco R.
      Evaluation of multifocal visual evoked potentials in patients with Graves’ orbitopathy and subclinical optic nerve involvement.
      • Goncalves A.C.
      • Silva L.N.
      • Gebrim E.M.
      • Matayoshi S.
      • Monteiro M.L.
      Predicting dysthyroid optic neuropathy using computed tomography volumetric analyses of orbital structures.
      Automated VF testing is a sensitive method of monitoring optic nerve function and is routinely used in patients with TED who are considered to be at risk for CON. Choi and associates showed that the most common pattern of Humphrey VF change in TED-CON is a blob-shaped defect located in the inferior hemifield.
      • Choi C.J.
      • Oropesa S.
      • Callahan A.B.
      • et al.
      Patterns of visual field changes in thyroid eye disease.
      This current study arises from our clinical observations of patients with TED and hypothesizes that there is a typical pattern of progression of VF defects in patients with TED-CON. This study longitudinally evaluates patients with worsening or improvement in a series of Humphrey VFs to categorize the defects and to characterize the typical progression of VF defects in TED-CON from mild to advanced loss. Our previous publication on this topic demonstrated that the existing nomenclature systems for VF loss do not fit the changes commonly seen in TED-CON
      • Choi C.J.
      • Oropesa S.
      • Callahan A.B.
      • et al.
      Patterns of visual field changes in thyroid eye disease.
      ; therefore, another purpose of this study is to establish a nomenclature system that accurately describes the various VF defects seen with worsening and improving TED-CON.

      Methods

      This retrospective, noncomparative chart review began with all adult patients with TED who underwent automated VF testing at the Massachusetts Eye and Ear Infirmary (MEEI) from January 1, 2007 to March 31, 2017. The study received approval from the meei institutional review board (IRBNet ID 1042134-2). Patients were identified through billing codes with subsequent review of electronic and paper medical records. Collection and evaluation of protected patient health information followed the rules and regulations of the Health Insurance Portability and Accountability Act. All interventions involved human participants in this study and conformed to the ethical standards of the institutional research committees and with the 1964 Helsinki declaration and its later amendments, or comparable ethical standards.
      Included participants were ≥18 years of age and were diagnosed with Graves disease with TED at an outside facility and were referred to MEEI for management. The time from diagnosis until the first visit at MEEI ranged from 5 days to 14 months. The diagnosis of TED with CON was confirmed at MEEI by either an ophthalmic plastic or neuro-ophthalmic physician. Included patients underwent >2 Humphrey VF (HVF) 24-2 or 30-2 tests during the course of the disease that showed abnormalities that were interpreted to be a result of their CON. While some patients had VF testing performed at outside facilities, only VF tests performed at MEEI were included in the study.
      Exclusion criteria included a history of another cause of optic neuropathy unrelated to TED, such as glaucoma, neurologic disease, or ischemic optic neuropathy, a coexisting ocular or eyelid condition that affected the interpretation of the VF tests, such as significant ptosis or dermatochalasis, corneal opacity including significant keratopathy from dry eye, cataract, or retinal disease, and a physical or mental disability impeding reliable VF testing.
      Participants underwent automated HVF testing using Swedish Interactive Threshold Algorithm Fast Test (SITA FAST) 24-2 or 30-2 test pattern on the Humphrey Field Analyzer (HFA; Carl Zeiss Meditec, Dublin, CA) with stimulus Goldmann size III, on a dim background (31.5 apostilb). Instructions were clarified with participants and the appropriate correction of refractive error for HVF testing was provided. HVFs were required, with 1 exception described below, to meet reliability criteria of <33% fixation losses, <33% false positives, and <33% false negatives in keeping with previously established standards from the Ocular Hypertension Treatment Study.
      • Keltner J.L.
      • Johnson C.A.
      • Cello K.E.
      • et al.
      Classification of visual field abnormalities in the ocular hypertension treatment study.
      Fixation losses are the most common errors made by normal and pathologic subjects. In addition to true fixation loss, they may result from incorrect initial mapping of the blind spot or other artifacts. Repeat testing can assist in verifying genuine fixation losses.
      • Katz J.
      • Sommer A.
      • Witt K.
      Reliability of visual field results over repeated testing.
      In this study, there were multiple follow-up tests in each subject's HVF timeline. Therefore, we made an exception to allow fixation losses of slightly >33% if the false positive and false negatives were ≤5%, and if the location and characteristics of the defect correlated with the other reliable VFs in the sequence.
      An abnormal VF was defined as meeting ≥1 of the 2 following criteria: 1) having a single point worse than the 0.5% pointwise probability level on the total or pattern deviation probability plots and 2) 3 clustered points beyond normal limits (P < .05) and ≥1 point worse than the 1% level on the total or pattern deviation plot.
      Besides these probability criteria and reliability criteria of fixation loss, false positive, and false negative as mentioned above, first-occurring abnormal VFs underwent confirmation on the next retest. The HVF defect was considered to be real if there were 2 consecutive abnormal VFs with the defect in the same or related location in next retest in every enrolled HVF. Any single VF test that was not compatible with the other VFs in the timeline was excluded.
      The categories for classification of the VF defects were developed after careful review of the VFs in order to create a meaningful and descriptive terminology for the characteristic changes found in these patients with TED-CON. Pre-existing standard terminology was used when it was applicable, including inferior altitudinal, central/paracentral, and enlarged blind spot. The classification categories included the following 10 options, divided into 3 stages that reflect the observed progression pattern, plus a miscellaneous category.
      Stage 1 VF defects are the earliest detectable in the progression of TED-CON VF loss and involve predominantly the inferior VF, with 3 distinguishable levels of severity. The smallest defect, stage 1a, involves 1-4 consecutive points in the inferior paracentral hemifield and is referred to as a “small inferior paracentral hemifield defect.” Stage 1b occurs when this VF defect increases in size to involve 5-29 consecutive points, but not the entire altitudinal inferior hemifield, and is referred to as a “large inferior paracentral hemifield defect.” Stage 1c, also known as “inferior altitudinal defect,” is seen with a defect involving the entire inferior hemifield.
      Stage 2 VF defects involve 2 distinguishable levels of severity and occur as the inferior altitudinal defect in stage 1c advances above the horizontal midline to involve the superior VF in addition to the inferior VF. Stage 2a defect includes the entire inferior hemifield plus 1-15 points of the superior hemifield periphery, either nasally, temporally, or both sides, and is called “inferior altitudinal defect with superior advancement.” Stage 2b involves the entire inferior hemifield plus progression of the VF loss above the horizontal midline into arcuate defects, resulting in an “inferior altitudinal plus superior arcuate defect.”
      Stage 3 VF loss in TED-CON is a progression of stage 2b into a complete loss of inferior and superior VF, called “total loss.” Stage X refers to the other, less common patterns of VF loss that were not part of the progression pattern, including: superior defect, central/paracentral defect, enlarged blind spot, and “scatter,” which is defined as diffuse VF loss including ≥3 quadrants with ≥1 point at <2% in ≥2 quadrants but not fitting into any of the other 9 categories described herein. Table 1 describes the specific criteria for each of these categories. The sequence of VF defect progression for each patient was recorded.
      Table 1Classification of Visual Field Abnormalities in Thyroid Eye Disease–Compressive Optic Neuropathy
      StageDescription
      IaSmall inferior paracentral hemifield abnormality from 1-4 consecutive points. At least 1 point at <0.5% or <1%
      IbSightly larger central/paracentral hemifield abnormality involving 5-29 consecutive points, not involving the entire inferior hemifield. At least one point at <1%
      IcSevere visual loss throughout the entire inferior hemifield that respects the horizontal midline. Points >70% in the hemifield have P < .5 on pattern deviation plot.
      IIaVisual fileld defect involving the entire inferior hemifield mostly respecting the horizontal midline, most points with <0.5% level, with 1-15 points crossing the horizontal midline into the superior hemifield on the nasal or temporal side or both, outside 15° on the meridian
      IIbDefect involving the entire inferior hemifield and with superior extension across contiguous abnormal points from the blind spot to ≥1 point outside 15° adjacent to nasal meridian
      IIISevere widespread visual field loss in 4 quadrants (MD >20.00 dB)
      Superior defectVisual field defect located predominantly in the superior hemifield, not compatible with nerve fiber bundle abnormalities
      Central/ParacentralVisual field loss that is predominantly at macular or perimarcular region which is inside 15° of the fixation. Generally not contiguous with blindspot
      Enlarged blindspotVisual field defect that is contiguous with the blindspot
      ScatterDiffuse visual field loss including ≥3 quadrants. At least 1 point at <2% in ≥2 quadrants
      These series of HVFs for each patient were reviewed and categorized independently by 2 ophthalmic plastic surgeons. The classification for each VF was accepted if the 2 readers agreed; however, if 2 readers disagreed, the VF was reviewed by a third ophthalmic plastic surgeon and there was discussion between the 3 readers to reach a final consensus. A commercially available statistical software package (Stata for Windows version 13.1; StataCorp LP, College Station, TX) was used to perform a kappa statistic for interrater agreement.

       Results

      Two hundred thirty-four VFs from 37 eyes of 23 patients with TED-CON who underwent HVF testing at 1 institution (MEEI) from January 1, 2007 to March 31, 2017 satisfied the general and VF inclusion criteria and were analyzed. The mean ± standard deviation (SD) age of the patients was 60 ± 8.9 years (range, 45-77 years), and 17 of 23 patients (73.9%) were female. The follow-up period ranged from 1-59 months (mean ± SD, 17.3 ± 17.4 months; median, 9 months).
      First, each of the 234 included VFs was reviewed by 2 independent readers and placed into 1 of the 10 categories. The interreader agreement was strong in VF categorization between the 2 readers (94.58% agreement, kappa 0.92 ± 0.04). Single fields that did not meet the reliability criteria of <33% fixation loss but had <5% false positives and 5% false negatives and a defect location compatible to those of the reliable VF timeline accounted for 9 of 234 VFs (3.8%).
      Stage 1 inferior VF defects were the most common found in patients with TED-CON and accounted for 145 of 234 VFs (62.0%). These were categorized into stage 1a small inferior paracentral hemifield defects (22/234; 9.4%), stage 1b large inferior paracentral hemifield defects (112/234; 47.9%), and stage 1c inferior altitudinal defects (11/234; 4.7%). Stage 2a inferior altitudinal defects with superior advancement above the horizontal meridian were found in 41 of 234 VFs (17.5%). The other 6 defects ranged from 0.9%-10.7% (Figure 1). Examples of these VF defects are shown in Figure 2, Figure 3.
      Figure thumbnail gr1
      Figure 1Frequency of pattern defects found in Humphrey visual fields in thyroid eye disease–compressive optic neuropathy (n = 234).
      Figure thumbnail gr2
      Figure 2Examples of the most common visual field defects in thyroid eye disease–compressive optic neuropathy: stages 1a (top row), 1b (middle rows), and 1c (bottom).
      Figure thumbnail gr3
      Figure 3Examples of the most common visual field defects in thyroid eye disease–compressive optic neuropathy: stages 2a (top), 2b (third row), 3 (fourth row), and X (bottom).
      Next, the longitudinal sequence of VF defects from each of the 37 eyes was analyzed in chronological order. The most common pattern of HVF progression in TED-CON, in order from least to more severe, was: small inferior paracentral hemifield defect, large inferior paracentral hemifield defect, inferior altitudinal defect, inferior altitudinal defect with superior advancement above the horizontal meridian (either temporal, nasal, or both temporal and nasal), inferior altitudinal with superior arcuate defect, and total loss. Thirty-one of 37 eyes (83.8%) shared this similar progression pattern, although not all progressed to the most severe spectrum of these deficits. The reverse order of VF defect progression was noted with improving TED-CON. Sixty-five percent of eyes had both improving and worsening VF changes in their timeline, while 8% had only worsening VF changes reflecting TED-CON progression, and 27% had only improving VF defect progression reflecting disease improvement. Table 2, Table 3 demonstrate the detailed progression pattern in each eye, correlating with therapies received. Examples of grayscale and pattern/total deviation plots of the most common VF progressions are shown in Figure 4, Figure 5, Figure 6. A minority of progression patterns in 6 of 37 eyes (16.2%) were outside of this typical pattern and originated from scatter and superior defects.
      Table 2Visual Field Progression Pattern for Each Eye with Thyroid Eye Disease–Compressive Optic Neuropathy Correlating with Treatment Timing and Modality (Patients 1-12)
      Month
      Patient No.1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859
      1LS
      1b1a1a1b
      2RSS
      STSTSTSTSSTSTSSTSTST
      1b1b1bXXX2a1c2a1b1b1b1bLF
      LSS
      1b1a1b1b1b1bLF
      3RS
      STSTSTSTSTSTSTSTSTSTSTSTST
      1a1a2a1a2a1c2a2a2a2a2a2a2a2aLF
      LSS
      1b1b1bX1b1bLF
      4LS
      STSTSTST
      XX1b1b1a
      5LST
      1b1b
      6LS
      STST
      2a1c2a1b
      7RS
      STSTST
      2b2b2b33
      LS
      1b2a2a1b1b
      8RS
      STSTSTSTSTSTST
      2a1b1b1b1b1b1a1a
      LS
      2a2a1b1b1b1c1b1b1b1b
      9RS
      STSSTST
      33X
      LS
      X3XXXX
      10LS
      STSTST
      1b1c2a1b
      11LS
      ST
      XXX1b1b
      12RS
      STSTST
      1bXXXXXX
      LS
      1b1b1b1a1bX1a
      1a = stage 1a disease; 1b = stage 1b disease; 1c = stage 1c disease; 2a = stage 2a disease; 2b = stage 2b disease; L = left; LF = lost to follow-up; R = right; S = surgery; ST = steroid treatment; SST = self-stop steroid.
      Table 3Visual Field Progression Pattern for Each Eye with Thyroid Eye Disease–Compressive Optic Neuropathy Correlating with Treatment Timing and Modality (Patients 13-23)
      Month
      Patient No.123456789101112131415161718192021222324252627
      13RS
      STSTSTSTST
      XXXXXXX2aX
      LS
      XXXXXXX
      14RSS
      STSTST
      1b2a1a1a
      LS
      2b2a2a2b1b
      15RS
      STSTSTST
      1a1b1a1b1a1b
      LS
      1b1b1b
      16LS
      ST
      1b1a1a
      17RS
      STSTSTST
      1c1c1c1b
      LS
      1b1b1b1b
      18RS
      STSTSTSTST
      2a2a2a1b1b1b
      LS
      1a1b1b
      19RS
      STST
      1b1b
      20RS
      STSTSTSTST
      2a2a2a2a1b1c2a2a
      LS
      2a2a2a1c2a1c1b
      21RS
      STSTSTSTSTSTSTSTSTSTSTSTSTST
      1b1b1b1b1b1b1b1b1b2a1b1b1b1b1b1b1a1a
      LSST
      1b1b2a1b1b1b2a1b1b1b1b1b1b1b1b1b1b1b1b
      22RSTSTSTST
      1b1b1b1b1b
      23RRA
      STST
      1b1b1b
      RA
      L2a1b1a
      1a = stage 1a disease; 1b = stage 1b disease; 1c = stage 1c disease; 2a = stage 2a disease; 2b = stage 2b disease; L = left; LF = lost to follow-up; R = right; RA = radiation; S = surgery; ST = steroid treatment; SST = self-stop steroid.
      Figure thumbnail gr4
      Figure 4Progressive worsening of Humphrey visual field defects in thyroid eye disease–compressive optic neuropathy in a 61-year-old woman's right eye demonstrated in grayscale and pattern deviation: (A) stage 1b, (B) larger stage 1b, (C) stage 2a, and (D) stage 2b.
      Figure thumbnail gr5
      Figure 5Progressive improvement of Humphrey visual field defects in thyroid eye disease–compressive optic neuropathy in a 76-year-old woman's right eye before and after oral steroid and orbital decompression. (A and B) Stage 2a decreases in size after steroid treatment and orbital decompression to become stage 1c with a small area of temporal creeping in (C). (D-F) Stage 1b decreases in size over time.
      Figure thumbnail gr6
      Figure 6Progression of Humphrey visual field defects in a 57-year-old woman with advanced thyroid eye disease–compressive optic neuropathy. (A-C) Stage 2b progression, as the patient declined any intervention. She was lost to follow-up for 3 years, and when she returned, the changes had progressed to stage 3 total loss (D).
      This observed progression of VF pattern deviation defects with TED-CON worsening or improvement was confirmed by correlation with the decreasing or increasing mean deviation (MD) values of the VF categories. Figure 7 shows a box and whisker plot of the mean (vertical line inside blue box), the 25th and 75th percentiles (blue box), the 10th and 90th percentiles (vertical line outside blue box), and outliers (dots) of MD values. As expected, the small inferior paracentral hemifield defect had the least negative MD value, followed by large inferior paracentral hemifield defect, inferior altitudinal defect, inferior altitudinal defect with superior advancement, inferior altitudinal with superior arcuate defect, and the worst MD value, total loss. The results of this MD value plot correlate with our descriptive classification system ordering of the most common progression pattern of HVF defects in TED-CON.
      Figure thumbnail gr7
      Figure 7Box and whisker plot of the mean deviation (MD) values of the most common progression patterns—mean, 10th, 25th, 75th, and 90th percentiles, and outliers.

      Discussion

      One of the most urgent and severe complications of TED is optic neuropathy. Early detection of optic nerve compression is crucial to prevent permanent visual loss. It is diagnosed by clinical examination and ancillary studies, such as VF testing, pattern electroretinogram, or visual evoked potentials.
      • Przemyslaw P.
      • Janusz M.
      • Alina B.L.
      • Maria G.
      Pattern electroretinogram (PERG) in the early diagnosis of optic nerve dysfunction in the course of Graves’ orbitopathy.
      • Perez-Rico C.
      • Rodriguez-Gonzalez N.
      • Arevalo-Serrano J.
      • Blanco R.
      Evaluation of multifocal visual evoked potentials in patients with Graves’ orbitopathy and subclinical optic nerve involvement.
      • Goncalves A.C.
      • Silva L.N.
      • Gebrim E.M.
      • Matayoshi S.
      • Monteiro M.L.
      Predicting dysthyroid optic neuropathy using computed tomography volumetric analyses of orbital structures.
      ,
      • Bothun E.D.
      • Scheurer R.A.
      • Harrison A.R.
      • Lee M.S.
      Update on thyroid eye disease and management.
      Clinical ophthalmic examination alone can delay the detection of TED-CON because good visual acuity with normal appearing optic nerves and absence of relative afferent pupillary defect do not exclude CON. A previous study reported that 50%-70% of eyes with confirmed TED-CON had a visual acuity of 20/40 or better, 76% of cases were bilateral with no relative afferent pupillary defect, and only 20%-50% had optic disc swelling.
      • Dickinson A.J.
      • Perros P.
      Controversies in the clinical evaluation of active thyroid-associated orbitopathy: use of a detailed protocol with comparative photographs for objective assessment.
      In addition, TED-CON seems to not correlate with the degree of proptosis. A previous study showed that many patients with TED-CON did not have significant proptosis, and that in fact the lack of proptosis precipitated a rise in orbital pressure that is thought to be a factor in CON. In addition, the absence of severe proptosis in patients with TED-CON has been reported.
      • McKeag D.
      • Lane C.
      • Lazarus J.H.
      • et al.
      Clinical features of dysthyroid optic neuropathy: a European Group on Graves’ Orbitopathy (EUGOGO) survey.
      ,
      • Mourits M.P.
      • Lombardo S.H.
      • van der Sluijs F.A.
      • Fenton S.
      Reliability of exophthalmos measurement and the exophthalmometry value distribution in a healthy Dutch population and in Graves’ patients. An exploratory study.
      These findings make ancillary testing, such as automated perimetry, helpful. VF testing is an excellent modality for following the progression of TED-CON and monitoring response to therapy.
      • Gasser P.
      • Flammer J.
      Optic neuropathy of Graves’ disease. A report of a perimetric follow-up.
      Goldmann perimetry and frequency-doubling technology have been reported to detect TED-CON
      • Trobe J.D.
      • Glaser J.S.
      • Laflamme P.
      Dysthyroid optic neuropathy. Clinical profile and rationale for management.
      • Monteiro M.L.
      • Portes A.L.
      • Moura F.C.
      • Regensteiner D.B.
      Using frequency-doubling perimetry to detect optic neuropathy in patients with Graves’ orbitopathy.
      • Hedges Jr., T.R.
      • Scheie H.G.
      Visual field defects in exophthalmos associated with thyroid disease.
      ; nevertheless, the most commonly used perimetry testing remains HVF.
      Previous studies that mention VF defect patterns in TED-CON demonstrate variability but an overall preponderance of “inferior arcuate” VF defects as well as “central scotomas.”
      • Neigel J.M.
      • Rootman J.
      • Belkin R.I.
      • et al.
      Dysthyroid optic neuropathy. The crowded orbital apex syndrome.
      ,
      • Trobe J.D.
      • Glaser J.S.
      • Laflamme P.
      Dysthyroid optic neuropathy. Clinical profile and rationale for management.
      ,
      • Hedges Jr., T.R.
      • Scheie H.G.
      Visual field defects in exophthalmos associated with thyroid disease.
      • Mensah A.
      • Vignal-Clermont C.
      • Mehanna C.
      • et al.
      Dysthyroid optic neuropathy: atypical initial presentation and persistent visual loss.
      • Soares-Welch C.V.
      • Fatourechi V.
      • Bartley G.B.
      • et al.
      Optic neuropathy of Graves disease: results of transantral orbital decompression and long-term follow-up in 215 patients.
      However, on careful analysis of the VF defects published in some of these series, the nomenclature used by the authors to describe the defects does not adequately fit the defects. For example, the Goldmann VF defects presented by Trobe and associates,
      • Trobe J.D.
      • Glaser J.S.
      • Laflamme P.
      Dysthyroid optic neuropathy. Clinical profile and rationale for management.
      which are described as central scotomas, are almost completely below the horizontal meridian, as was found in our series and named a small inferior paracentral hemifield defect. The VF defects described as “severe defects involving fixation with inferior altitudinal depression” are classic large inferior paracentral hemifield defects in our nomenclature system. Another series resorted to using a complicated system of “combinations” of VF nomenclature to attempt to describe the defects found in their patients.
      • Neigel J.M.
      • Rootman J.
      • Belkin R.I.
      • et al.
      Dysthyroid optic neuropathy. The crowded orbital apex syndrome.
      Recently, the senior author and her colleagues analyzed a large series of TED-CON HVFs and found that most common pattern is an inferior blob-shaped defect that is unable to be classified into the Ocular Hypertension Study categories.
      • Choi C.J.
      • Oropesa S.
      • Callahan A.B.
      • et al.
      Patterns of visual field changes in thyroid eye disease.
      In fact, many of the VF defects found in this study did not fit into pre-existing nomenclatures because the systems were created to describe VF defects resulting from glaucomatous or optic nerve fiber layer–related damage.
      To create a useful and understandable classification system, we reviewed HVFs in patients with TED-CON and created a descriptive nomenclature system based on the shape and location of the defects. When previously described VF defect names applied to defects in the current study, these terms were adopted for ease of use and understanding. The VFs were evaluated not only in isolation but also in the context of longitudinal review of individual patient VF testing. This additional analysis provided the authors with an understanding of individual defect patterns, as well as the progression of VF worsening or improvement with changes in the patient's TED-CON. The VF defect patterns progressed through a predictable sequence with worsening of disease. Conversely, when an effective intervention, such as medical therapy or surgical orbital decompression, occurred and resulted in an improvement of CON, there was a similar and predictable progression of reversal of the VF defects. Knowledge of this classification system and nomenclature allows the clinician to not only technically describe the defects but also to have an understanding of where on the spectrum of progression a patient's VF defects fall. Knowledge of whether a VF defect is early or late in the spectrum may be useful in clinical decision making.
      In this study, the inferior hemifield was the most common location of HVF defects. In order to be more specific for the classification system, defects in this region were divided into 3 stages based primarily on the size of the defect. In order of increasing size, they are: small inferior paracentral hemifield defect, large inferior paracentral hemifield defect, and inferior altitudinal defect. It is logical that the majority of the VF defects fell into these less severe categories, because fewer patients had advanced disease and therefore fewer VF defects advanced to cross the horizontal meridian and involve the superior VF. Five VFs from 3 eyes of 2 patients had total loss HVF defects. The first patient was lost to follow-up for many months with known severe TED, and the second had severe TED-CON resistant to medical therapy but delayed orbital decompression surgery against medical advice because of her job responsibilities.
      It is remarkable that 31 of 37 eyes in the study had HVF progression follow a similar pattern originating from an inferior defect, which in cases of severe disease progressed to involve the superior VF. This suggests that the findings are not random and that there is likely an anatomic basis for these observations. There are several existing theories regarding the mechanism of optic neuropathy in TED. Perhaps the most widely accepted theory is that of direct compression of the optic nerve by the inflamed and enlarged EOMs at the orbital apex.
      • Goncalves A.C.
      • Silva L.N.
      • Gebrim E.M.
      • Matayoshi S.
      • Monteiro M.L.
      Predicting dysthyroid optic neuropathy using computed tomography volumetric analyses of orbital structures.
      ,
      • Garrity J.A.
      • Bahn R.S.
      Pathogenesis of graves ophthalmopathy: implications for prediction, prevention, and treatment.
      • Khong J.J.
      • McNab A.A.
      • Ebeling P.R.
      • Craig J.E.
      • Selva D.
      Pathogenesis of thyroid eye disease: review and update on molecular mechanisms.
      • Feldon S.E.
      • Lee C.P.
      • Muramatsu S.K.
      • Weiner J.M.
      Quantitative computed tomography of Graves’ ophthalmopathy. Extraocular muscle and orbital fat in development of optic neuropathy.
      • Le Moli R.
      • Pluchino A.
      • Muscia V.
      • et al.
      Graves’ orbitopathy: extraocular muscle/total orbit area ratio is positively related to the Clinical Activity Score.
      Based on the anatomy of the orbital apex (Figure 8), the superior portion of the optic nerve lies in close proximity to the inferior portion of the superior rectus muscle in the annulus of Zinn, making it most susceptible to compression resulting in an inferior VF defect. Several published reports demonstrate that the superior rectus–levator muscle complex or levator muscle are the most commonly enlarged EOMs in TED.
      • Nugent R.A.
      • Belkin R.I.
      • Neigel J.M.
      • et al.
      Graves orbitopathy: correlation of CT and clinical findings.
      ,
      • Davies M.J.
      • Dolman P.J.
      Levator muscle enlargement in thyroid eye disease-related upper eyelid retraction.
      As the EOMs at the orbital apex become larger, the medial and lateral portions of the optic nerve are the next to be compressed, resulting in superior creeping of the inferior defects. The relationship of the inferior portion of optic nerve to the inferior rectus muscle at the orbital apex makes it less likely to be involved until late stages, despite the inferior rectus being more commonly affected in TED. Therefore, superior hemifield involvement implies a later stage VF defect.
      • Nagy E.V.
      • Toth J.
      • Kaldi I.
      • et al.
      Graves’ ophthalmopathy: eye muscle involvement in patients with diplopia.
      Another less plausible theory suggests that vascular compromise secondary to orbital congestion causes an increase in orbital pressure, leading to impeded venous drainage and arterial hypoperfusion resulting in optic nerve damage.
      • Thyparampil P.
      • Yen M.T.
      Compressive optic neuropathy in thyroid eye disease.
      Another theory suggests direct neurotoxicity to the optic nerve from thyrotoxicosis, which has little supporting evidence to date.
      • Soares-Welch C.V.
      • Fatourechi V.
      • Bartley G.B.
      • et al.
      Optic neuropathy of Graves disease: results of transantral orbital decompression and long-term follow-up in 215 patients.
      Neither of these latter 2 theories explain the consistent pattern of VF changes found in this study.
      Figure thumbnail gr8
      Figure 8Coronal computed tomography scans taken near the orbital apex. (A) The optic nerve (white asterisk) lies closer to the superior rectus muscle than the other extraocular muscles in the normal orbital apex. (B) When the extraocular muscles become enlarged in thyroid eye disease, resulting in apical crowding, there is contact between the superior aspect of the optic nerve and the superior rectus muscle. (C) A more severe stage of thyroid eye disease shows profound orbital apex crowding with nearly obliterated fat planes, again with contact between the optic nerves and the superior rectus muscles and with a small space remaining inferior to the optic nerves.
      Analysis was performed on the 6 eyes in 4 patients that did not follow the stereotypical progression pattern in order to determine whether there were any unique features to these cases. Patient 2's right eye was initially found to have an enlarged blind spot on VF testing, followed soon after by a central scotoma. The clinical examination showed optic disc edema in this eye. After starting oral prednisone therapy, the disc edema improved and her VF defect became typical stage 1b inferior loss that worsened as she self-tapered her steroid prematurely and then improved to a smaller stage 1a spot after resuming steroid therapy with a longer taper. For 2 patients, scatter pattern VF loss was the bilateral baseline VF defect during their episodes of optic neuropathy. The scatter pattern defects worsened and improved along with the other parameters of their disease and neither patient had optic disc edema. There was nothing noted to be unique about the clinical presentation or orbital imaging of these 2 patients compared with the others in the cohort. Finally, patient 12's right eye had 1 VF along his course of improvement showing a scatter pattern. He initially had a stage 1b inferior defect that decreased to a stage 1a defect after oral prednisone therapy. He then had orbital decompression surgery and before his VF returned to normal, his penultimate VF showed a scatter pattern. There did not appear to be a unique circumstance associated with this isolated scatter pattern VF defect.
      In conclusion, this is the first study to show and systematically categorize the longitudinal pattern of HVF defect progression in patients with CON secondary to TED. HVF defects in TED-CON often show a specific pattern of progression and should be recognized by physicians as they may be helpful for detection, monitoring, and decision-making regarding treatment of patients with TED-CON. The novel nomenclature system described herein parallels the observed progression of VF defects in TED-CON and provides a simple and useful lexicon for clinicians caring for patients with this challenging condition.
      All authors have completed and submitted the ICMJE form for disclosure of potential conflicts of interest. Funding/Support: The authors indicate no financial support. Financial Disclosures: Dr Freitag receives textbook royalties from Thieme and Springer and does medicolegal consulting for private individuals/law firms. All authors attest that they meet the current ICMJE criteria for authorship. We thank Hokyung Choung, MD for advice and interpretation of visual fields.

      Supplemental Data

      References

        • Victores A.J.
        • Takashima M.
        Thyroid eye disease: optic neuropathy and orbital decompression.
        Int Ophthalmol Clin. 2016; 56: 69-79
        • Nunery W.R.
        • Nunery C.W.
        • Martin R.T.
        • Troung T.V.
        • Osborn D.R.
        The risk of diplopia following orbital floor and medial wall decompression in subtypes of ophthalmic Graves’ disease.
        Ophthal Plast Reconstr Surg. 1997; 13: 153-160
        • Wiersinga W.M.
        • Regensburg N.I.
        • Mourits M.P.
        Differential involvement of orbital fat and extraocular muscles in Graves’ ophthalmopathy.
        Eur Thyroid J. 2013; 2: 14-21
        • Bahn R.S.
        Graves’ ophthalmopathy.
        N Engl J Med. 2010; 362: 726-738
        • Mascarello F.
        • Toniolo L.
        • Cancellara P.
        • et al.
        Expression and identification of 10 sacromeric MyHC isoforms in human skeletal muscles of different embryologic origin. Diversity and similarity in mammalian species.
        Ann Anat. 2016; 207: 9-20
        • Mizokami T.
        • Salvi M.
        • Wall J.R.
        Eye muscle antibodies in Graves’ ophthalmopathy: pathogenic or secondary epiphenomenon?.
        J Endocrinol Invest. 2004; 27: 221-229
        • Verity D.H.
        • Rose G.E.
        Acute thyroid eye disease (TED): principles of medical and surgical management.
        Eye. 2013; 27: 308-319
        • Neigel J.M.
        • Rootman J.
        • Belkin R.I.
        • et al.
        Dysthyroid optic neuropathy. The crowded orbital apex syndrome.
        Ophthalmology. 1988; 95: 1515-1521
        • Przemyslaw P.
        • Janusz M.
        • Alina B.L.
        • Maria G.
        Pattern electroretinogram (PERG) in the early diagnosis of optic nerve dysfunction in the course of Graves’ orbitopathy.
        Klin Oczna. 2013; 115: 9-12
        • Perez-Rico C.
        • Rodriguez-Gonzalez N.
        • Arevalo-Serrano J.
        • Blanco R.
        Evaluation of multifocal visual evoked potentials in patients with Graves’ orbitopathy and subclinical optic nerve involvement.
        Doc Ophthalmol. 2012; 125: 11-19
        • Goncalves A.C.
        • Silva L.N.
        • Gebrim E.M.
        • Matayoshi S.
        • Monteiro M.L.
        Predicting dysthyroid optic neuropathy using computed tomography volumetric analyses of orbital structures.
        Clinics (Sao Paulo). 2012; 67: 891-896
        • Choi C.J.
        • Oropesa S.
        • Callahan A.B.
        • et al.
        Patterns of visual field changes in thyroid eye disease.
        Orbit. 2017; 36: 201-207
        • Keltner J.L.
        • Johnson C.A.
        • Cello K.E.
        • et al.
        Classification of visual field abnormalities in the ocular hypertension treatment study.
        Arch Ophthalmol. 2003; 121: 643-650
        • Katz J.
        • Sommer A.
        • Witt K.
        Reliability of visual field results over repeated testing.
        Ophthalmology. 1991; 98: 70-75
        • Bothun E.D.
        • Scheurer R.A.
        • Harrison A.R.
        • Lee M.S.
        Update on thyroid eye disease and management.
        Clin Ophthalmol. 2009; 3: 543-551
        • Dickinson A.J.
        • Perros P.
        Controversies in the clinical evaluation of active thyroid-associated orbitopathy: use of a detailed protocol with comparative photographs for objective assessment.
        Clin Endocrinol (Oxf). 2001; 55: 283-303
        • McKeag D.
        • Lane C.
        • Lazarus J.H.
        • et al.
        Clinical features of dysthyroid optic neuropathy: a European Group on Graves’ Orbitopathy (EUGOGO) survey.
        Br J Ophthalmol. 2007; 91: 455-458
        • Mourits M.P.
        • Lombardo S.H.
        • van der Sluijs F.A.
        • Fenton S.
        Reliability of exophthalmos measurement and the exophthalmometry value distribution in a healthy Dutch population and in Graves’ patients. An exploratory study.
        Orbit. 2004; 23: 161-168
        • Gasser P.
        • Flammer J.
        Optic neuropathy of Graves’ disease. A report of a perimetric follow-up.
        Ophthalmologica. 1986; 192: 22-27
        • Trobe J.D.
        • Glaser J.S.
        • Laflamme P.
        Dysthyroid optic neuropathy. Clinical profile and rationale for management.
        Arch Ophthalmol. 1978; 96: 1199-1209
        • Monteiro M.L.
        • Portes A.L.
        • Moura F.C.
        • Regensteiner D.B.
        Using frequency-doubling perimetry to detect optic neuropathy in patients with Graves’ orbitopathy.
        Jpn J Ophthalmol. 2008; 52: 475-482
        • Hedges Jr., T.R.
        • Scheie H.G.
        Visual field defects in exophthalmos associated with thyroid disease.
        AMA Arch Ophthalmol. 1955; 54: 885-892
        • Mensah A.
        • Vignal-Clermont C.
        • Mehanna C.
        • et al.
        Dysthyroid optic neuropathy: atypical initial presentation and persistent visual loss.
        Orbit. 2009; 28: 354-362
        • Soares-Welch C.V.
        • Fatourechi V.
        • Bartley G.B.
        • et al.
        Optic neuropathy of Graves disease: results of transantral orbital decompression and long-term follow-up in 215 patients.
        Am J Ophthalmol. 2003; 136: 433-441
        • Garrity J.A.
        • Bahn R.S.
        Pathogenesis of graves ophthalmopathy: implications for prediction, prevention, and treatment.
        Am J Ophthalmol. 2006; 142: 147-153
        • Khong J.J.
        • McNab A.A.
        • Ebeling P.R.
        • Craig J.E.
        • Selva D.
        Pathogenesis of thyroid eye disease: review and update on molecular mechanisms.
        Br J Ophthalmol. 2016; 100: 142-150
        • Feldon S.E.
        • Lee C.P.
        • Muramatsu S.K.
        • Weiner J.M.
        Quantitative computed tomography of Graves’ ophthalmopathy. Extraocular muscle and orbital fat in development of optic neuropathy.
        Arch Ophthalmol. 1985; 103: 213-215
        • Le Moli R.
        • Pluchino A.
        • Muscia V.
        • et al.
        Graves’ orbitopathy: extraocular muscle/total orbit area ratio is positively related to the Clinical Activity Score.
        Eur J Ophthalmol. 2012; 22: 301-308
        • Nugent R.A.
        • Belkin R.I.
        • Neigel J.M.
        • et al.
        Graves orbitopathy: correlation of CT and clinical findings.
        Radiology. 1990; 177: 675-682
        • Davies M.J.
        • Dolman P.J.
        Levator muscle enlargement in thyroid eye disease-related upper eyelid retraction.
        Ophthal Plast Reconstr Surg. 2017; 33: 35-39
        • Nagy E.V.
        • Toth J.
        • Kaldi I.
        • et al.
        Graves’ ophthalmopathy: eye muscle involvement in patients with diplopia.
        Eur J Endocrinol. 2000; 142: 591-597
        • Thyparampil P.
        • Yen M.T.
        Compressive optic neuropathy in thyroid eye disease.
        Int Ophthalmol Clin. 2016; 56: 51-67