Article Text

Download PDFPDF

Understanding diagnostic disagreement in angle closure assessment between anterior segment optical coherence tomography and gonioscopy
  1. Natalia Porporato1,
  2. Mani Baskaran1,
  3. Tin A Tun1,
  4. Rehena Sultana2,
  5. Marcus Tan3,
  6. Joanne HuiMin Quah4,
  7. John C Allen2,
  8. Shamira Perera1,
  9. David S Friedman5,
  10. Ching Yu Cheng1,
  11. Tin Aung1
  1. 1 Singapore National Eye Centre, Singapore Eye Research Institute, Singapore
  2. 2 Centre for Quantitative Medicine, Duke-NUS Graduate Medical School, Singapore
  3. 3 Ophthalmology, National University Hospital, Singapore
  4. 4 Outram Polyclinic, SingHealth Polyclinics, Singapore
  5. 5 Glaucoma, Massachusetts Eye and Ear Infirmary, Boston, Massachusetts, USA
  1. Correspondence to Dr Tin Aung, Singapore National Eye Centre, Singapore Eye Research Institute, Singapore 169856, Singapore; aung.tin{at}snec.com.sg

Abstract

Background/aims Although being a more objective tool for assessment and follow-up of angle closure, reliability studies have reported a moderate diagnostic performance for anterior segment optical coherence tomography (OCT) technologies when comparing with gonioscopy as the reference standard. We aim to determine factors associated with diagnostic disagreement in angle closure when assessed by anterior segment swept source OCT (SS-OCT, CASIA SS-1000; Tomey, Nagoya, Japan) and gonioscopy.

Methods Cross-sectional study. A total of 2027 phakic subjects aged ≥50 years, with no relevant previous ophthalmic history, were consecutively recruited from a community polyclinic in Singapore. Gonioscopy and SS-OCT (128 radial scans) for the entire circumference of the angle were performed for each subject. A two-quadrant closed gonioscopic definition was used. On SS-OCT images, angle closure was defined as iridotrabecular contact (ITC) to the extent of ≥35%, ≥50% and ≥75% of the circumferential angle. Diagnostic disagreements between both methods, that is, false positives or overcalls and false negatives or undercalls were defined, respectively, as gonioscopic open/closed angles inversely assessed as closed/open by SS-OCT.

Results Two hundred and seventy-two (14.7%) resulted in overcall results (false positives) when ≥50% of the angle circumference was closed using SS-OCT. These eyes had significantly wider (anterior chamber width, 11.7 vs 11.6 mm, p<0.001) and deeper (anterior chamber depth (ACD), 2.4 vs 2.2 mm, p<0.001) anterior chambers than eyes assessed by both methods as closed (true positives). Deeper ACD (OR 9.31) and lower lens vault (LV) (OR 0.04) were significantly associated with a false positive diagnosis in the multivariate analysis. Most of these cases had short (52.6%) or irregular (39%) ITC in SS-OCT images.

Conclusions We found that anterior chamber dimensions, determined by ACD and LV, were factors significantly associated with diagnostic disagreement between anterior segment SS-OCT and gonioscopy in angle closure assessment.

  • angle
  • anterior chamber
  • glaucoma
  • imaging

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

The number of subjects with primary angle closure glaucoma (PACG) worldwide is estimated to increase to 32 million by 2040, with Asia accounting for 76% of PACG cases.1 The current gold standard for angle closure assessment is gonioscopy, a contact method developed in 1800s that requires anaesthesia and patient cooperation. Gonioscopy is subjective, and its findings may vary with pressure on the cornea, lighting conditions, pigmentation of the angle and convexity of the iris. Ten per cent of the time, an examiner at the second visit will disagree with his original gonioscopic classification on the same patient seen previously.2 The lack of objective documentation of gonioscopic findings makes it inappropriate for long follow-up of angle closure suspects with or without intervention.

With the development of anterior segment optical coherence tomography (AS-OCT), objective non-contact imaging of the anterior chamber angle (ACA) has become widely available, and qualitative and quantitative analysis of anterior segment parameters is now possible. However, earlier time-domain devices could only obtain a single cross-sectional scan of the anterior segment at a time, which meant that the remainder of the angle could not be visualised.3 Anterior segment swept source OCT (SS-OCT, CASIA SS-1000, Tomey, Nagoya, Japan) provides a wide scanning range of 16 mm, which allows an entire cross-section of the anterior chamber to be captured simultaneously. The SS-OCT’s low density three-dimensional angle scan simultaneously obtains 128 meridional radial scans of the anterior chamber over the entire circumference of the angle.4 We recently evaluated the Tomey SS-OCT device for its ability to detect angle closure and found it moderately accurate in diagnostic performance compared with gonioscopy, using either its built in semiautomated iridotrabecular contact (ITC) index algorithm (area under curve (AUC) 0.83)5 or via manually graded images (AUC 0.84).6

Besides a moderate diagnostic performance, AS-OCT technologies have showed fair to moderate agreement with gonioscopy5 6 and a tendency to classify more angles as closed than gonioscopy.7 Thus, the aim of this study was to determine factors leading to this disagreement in angle closure assessment between SS-OCT angle imaging and gonioscopy, the current reference standard.

Methods

Study population and recruitment

Subjects aged ≥50 years with no previous history of glaucoma, intraocular surgery or ocular trauma were consecutively recruited from a Singapore community polyclinic providing primary healthcare services. A total of 2038 consecutive subjects were eligible and screened for this study, after obtaining written informed consent, from June to September 2013. We previously reported that 92% (1857) of the subjects included had gradable SS-OCT images, 87.8% were Chinese with a mean age was 61.8±6.7 years and 63.5% were women.6 From the 170 SS-OCT images discarded due to poor quality for grading (~8%), 6 eyes had gonioscopic angle closure and 164 eyes had open angles in gonioscopy.

SS-OCT imaging

All subjects underwent SS-OCT imaging before any contact procedure, under dark room conditions. The upper eyelid was gently elevated and the lower eyelid was gently pulled down by the operator so that the ACA could be seen in the scan window, taking care to avoid inadvertent pressure on the globe. Patients were asked to focus on an internal fixation target, and once the patient had been optimally positioned, each eye was scanned with the three-dimensional angle analysis scan (which takes 2.4 s) using the auto alignment function. The algorithm requires 128 consecutive meridional scans, each consisting of 512 A-scans across the anterior chamber.

All angle images were assessed by a single trained ophthalmologist (NP) who was masked to gonioscopy findings. We have previously reported good intraobserver and interobserver agreement for image grading using SS-OCT.8 A closed angle on an SS-OCT image was defined as contact between the iris and any part of the angle wall anterior to the scleral spur in ≥35%, ≥50% and ≥75% of the circumference, as reported in a previous study.5 6 Diagnostic disagreements between both methods, that is, false positives or overcalls and false negatives or undercalls were defined, respectively, as gonioscopic open/closed angles incorrectly inversely assessed as closed/open by SS-OCT. These ‘false’ SS-OCT results were assessed a second time to identify short ITC and steep irides cases, as reported previously.7 Short ITC corresponds to SS-OCT scans where the iris does not have contact with the trabecular meshwork entirely (from scleral spur to Schwalbe line). In our analysis, an image with short ITC was considered closed if the five adjacent scans had ITC from scleral spur to Schwalbe line (figure 1). A steep iris profile was present when there was a pronounced convex profile of the anterior surface of the iris with an open angle (figure 2). We also identified cases of irregular ITC, characterised by the presence of uneven iris profile along the circumference, where some areas show ITC while other areas within the quadrant do not show it (figure 3).

Figure 1

Swept-source OCT scans from different study patients showing short iridotrabecular contact (A) and iridotrabecular contact from scleral spur to Schwalbe line (B). OCT, optical coherence tomography.

Figure 2

Swept-source optical coherence tomography (SS-OCT) scan corresponding to a steep iris profile (‘over-the-hill’ configuration type) that was assessed as closed in gonioscopy but open in SS-OCT, representing a false negative or undercall case.

Figure 3

Irregular iridotrabecular contact (ITC). (A) Swept-source optical coherence tomography (SS-OCT) scan showing an uneven iris profile leading to irregular ITC. (B) Graphical representation of the qualitative assessment of the SS-OCT scans (0=open, 1=closed) corresponding to the superior quadrant (45°–135° of the circumference) of the same patient, that was assessed as open by gonioscopy and closed by SS-OCT resulting in a false positive or overcall case.

Gonioscopy

Indentation gonioscopy was performed in the dark by a glaucoma fellowship-trained ophthalmologist (MT) using a Sussman 4-mirror goniolens (Ocular Instruments, Bellevue, Washington, USA) who was masked to imaging results. A weighted kappa of 0.82 was achieved in the assessment of angle grading when comparing with and the readings of an ophthalmologist with subspecialty glaucoma training (TA). A 1 mm light beam was reduced to a narrow slit, and the vertical beam was offset horizontally for evaluating nasal and temporal angles and maintained vertically for assessing superior and inferior angles. The examination was performed with the subject’s eye in the primary position of gaze. Care was taken to avoid light from falling on the pupil and to avoid inadvertent indentation during examination. In some cases, the gonioscopy lens was tilted slightly to allow a view of the angle over the convexity of the iris, avoiding ocular distortion. The angle in each quadrant was graded per the modified Scheie grading system, according to the anatomic structures observed during gonioscopy. The ACA was considered ‘closed’ in a quadrant if the posterior pigmented trabecular meshwork could not be seen in the primary position without indentation. An eye was classified as having angle closure if there were two or more closed quadrants.

Statistical analysis

Demographic and SS-OCT parameters were summarised by mean (SD or median (range) as appropriate for continuous variables and frequency (proportion) for categorical variables. The two-sample Student’s t-test and Fisher’s exact test, respectively, were used to compare continuous and categorical variables between groups. Univariate and multivariable logistic regression analyses were used to identify risk factors. AUC receiver operating characteristic with 95% CI was calculated to assess performance of variables identified as predictors of false positive results in angle closure disease assessed by SS-OCT, using gonioscopy as the reference standard.

Statistical significance was set at p<0.05. Statistical analyses were performed using SAS V.9.4 (SAS Institute).

Results

Table 1 shows the total number of agreements and disagreements in angle closure diagnosis between gonioscopy and SS-OCT. Using a two-quadrant gonioscopic angle closure definition (7.5% prevalence of angle closure in our community-based sample), a false positive misdiagnosis occurred in up to almost 15% of eyes assessed with SS-OCT, compared with a 2% false negative rate. The overcalls were lower in eyes with more circumferential closure on SS-OCT, viz. 20%, 14.7% and 8.5% for angle closures of ≥35%, ≥50% and ≥75%, respectively (p=0.001), and the undercalls increased only slightly with higher percentages of circumferential closure, viz. 1.3%, 1.83% and 2.9% when ≥35%, ≥50% and ≥75% of the circumference was closed on SS-OCT, respectively (p=0.007).

Table 1

Number of diagnostic agreements and disagreement between two quadrants closed gonioscopic definition and the different cut-off values of the extension of circumferential angle closure by swept-source optical coherence tomography (SS-OCT)

Compared with patients where both methods agree in the diagnosis of angle closure, false positives had deeper (2.4 vs 2.2 mm, p<0.001) and wider (11.7 vs 11.6 mm, p<0.001) anterior chambers (table 2). The multivariable analysis (table 3) showed that deeper anterior chamber depth (ACD) (adjusted OR 9.31, 95% CI 1.45 to 59.714, p<0.05) and higher lens vault (LV) (adjusted OR 0.038, 95% CI 0.004 to 0.314, p<0.05) were factors significantly associated with diagnostic disagreement between both methods. Compared with eyes seen by SS-OCT and gonioscopy as open, false negatives had smaller anterior chamber characteristics, viz. shallower ACD (2.2 vs 2.7 mm, p<0.001) and narrower anterior chamber width (ACW) (11.6 vs 11.8 mm, p<0.001), with no differences in pupil diameter (table 4).

Table 2

Demographics and anterior segment quantitative parameters of true positives and false positives, at different cut-off values of circumferential angle closure assessed by swept-source optical coherence tomography (SS-OCT)

Table 3

Determinants of false positive angle closure diagnosis by swept-source optical coherence tomography with gonioscopy as reference standard

Table 4

Demographics and anterior segment quantitative parameters of true negatives and false negatives, at different cut-off values of circumferential angle closure assessed by swept-source optical coherence tomography

Further, we identified short ITC in 52.57% and irregular ITC in 39% of the false positive cases while 66.7% of the false negatives corresponded to steep irides profiles and 33.3% irregular ITC (table 5). Eyes with short ITC had deeper (2.6 vs 2.4 mm, p<0.001) but narrower (11.6 vs 11.8 mm; p<0.05) anterior chambers than irregular ITC cases. In both types of disagreements eyes with irregular ITCs had similar ACDs (≈2.4 mm) but different ACW and LVs (11.8 vs 11.5 mm and 0.5 vs 0.8 mm, respectively; p<0.01).

Table 5

Quantitative SS-OCT parameters of short iridotrabecular contact (ITC) and irregular ITC cases among false positives (gonioscopy open and SS-OCT closed), and steep iris profile and irregular ITC among false negatives (gonioscopy closed and SS-OCT open)

Discussion

Considering the diagnostic discordance between AS-OCT and gonioscopy when assessing angle closure, we found that overcalls rated significantly higher than undercalls and showed that parameters related with anterior chamber dimensions (ACD and LV), represented associated factors for this disagreement, in this large community-based study. These data are indicative of rates and directionality of angle closure diagnostic disagreement in clinical practice among the current reference standard, gonioscopy and AS-OCT imaging.

As previously stated, we found a high false positive rate using SS-OCT (15%), although lower than that obtained with Visante time-domain AS-OCT in previous studies (36.3%).5 9 This would explain the improved AUC and specificity obtained with SS-OCT over Visante in assessing angle closure (AUC, 0.76 vs 0.84; specificity, 62.9% vs 78.5%, respectively).10 Nolan et al 3 showed that the rate of false positive disagreements between gonioscopy and Visante decreased from dim light to darkness (36%–30%) while the rate of false negatives increased (4%–10%), explaining the variability that exists due to pupil reaction to light. Rigi et al 11 analysed gonioscopy versus SS-OCT interinstrument disagreement (n=85) and found a similar range of overcalls (9%–22% vs 8.5%–20%, in our analysis) but higher undercalls rate (~5% to 10% compared with our 1.34% to ~3%). This suggests that SS-OCT diagnostic outcomes among different settings are highly comparable, which is important information for planners of multicenter studies.

The present study provides novel information on ACD variations in the disagreement of gonioscopic angle closure relative to SS-OCT that were not found in previous studies. ACD was found to be a factor highly associated with diagnostic disagreement (OR 9.31) and we observed that gonioscopy and SS-OCT methods tend to agree in the angle status diagnosis when these eyes have either shallow or deep anterior chambers—the mean ACD was 2.2 mm for true positives and 2.7 mm for true negatives (tables 2 and 4). However, they seem to disagree when the ACD is around 2.4 mm (mean ACD for overcalls, table 2), that is, an observer performing gonioscopy may see these angles as open while on SS-OCT it may be closed. Interestingly, in a population-based study in Singapore, Aung et al 12 found that ACD between 2.2 and 2.6 mm had three to five times more risk of peripheral anterior synechiae (PAS) (OR 3.7 for ACD between 2.4 and 2.6 mm and OR 5.0 for ACD between 2.2 and 2.39 mm). Also, Lavanya et al 10 identified ACDs <2.7 mm as a major risk factor for angle closure (OR 42.5). Foster et al 13 reported that between 0.3% and 1.7% of people with Shaffer grading of 30° and 40° angles (gonioscopic open angles) had PAS. While open in gonioscopy, we do not know if those eyes could have corresponded to SS-OCT closed eyes. In a similar community-based study in Singapore, Baskaran et al 9 showed that 17% of the false positive disagreements between gonioscopy and Visante at baseline were finally assessed as closed by both methods after 4 years of follow-up. No information of AS-OCT anterior chamber quantitative parameters was provided to compare with our analysis.

Gonioscopy and AS-OCT have different landmarks for the grading of angle closure and so are inherently different. Fortunately, in most cases the structures of the trabecular meshwork and scleral spur are sufficiently similar to be good surrogates. However, in the cases where there is a discordance, these small distances become more important. In our analysis, 90% of the false positive disagreements are explained by short and irregular ITCs. This represents a novel finding that is now better understood with the 360° analysis by SS-OCT. Short ITC was also observed in 71% of quadrants that resulted overcalls on Visante.7 In our analysis, an image with short ITC was considered closed if the five adjacent scans had ITC from scleral spur to Schwalbe line. In relatively deeper ACD eyes, short ITC can inadvertently be opened by indentation gonioscopy or bright light. On the contrary, a high LV may result in substantial ITC and lead to lesser false positive results, as found in our study. Given an uneven iris profile, some areas of the angle circumference can show ITC while other areas within the quadrant do not show it. This scenario could be considered either open or closed by the examiner as it may be difficult to calculate how many clock hours within the quadrant are closed when the trabecular meshwork structures are seen in patches (figure 3).

As for false negative outcomes (up to 2% of the study population), we showed that the main disagreement factor is related to cases of steep irides (called over-the-hill configuration type),14 that in spite of having no ITC, may be interpreted as closed angles on gonioscopy. Among the false negative images found with Visante in comparison with gonioscopy, the presence of these type of irides configuration was determined in 51% of these quadrants.7 On the other hand, when comparing false negatives and true positives (data no shown), eyes from the ≥50% and ≥75% groups had similar biometric characteristics with exception of the pupil diameter that resulted significantly smaller in the false negative group (p<0.05). A smaller pupil diameter could lead to artificial opening of the angle on SS-OCT images.

Our study has limitations, such as the potential error in manual grading of images, reliance on a single person’s gonioscopy, and use of only a Sussman lens as opposed to a Goldmann lens (which may have decreased the rate of identified angle closure). Finally, as this was a community-based study, there exists the potential for some inherent sampling bias as we included patients in a primary care setting in Singapore.

In summary, the data presented here outline the scope and direction of diagnostic disagreement in the diagnosis of angle closure between gonioscopic examination and AS-OCT technology. In this large community-based study, we found that anterior chamber dimensions, determined by ACD and LV, were factors significantly associated with this diagnostic disagreement.

References

Footnotes

  • Contributors Conception and design of the work: NP, MB, SP, DSF, CYC and TA. Acquisition, analysis or interpretation of data for the work: NP, MB, TAT, RS, MT, JHQ, JCA and TA. Drafting the work: NP Revising the work: NP, MB, SP, DSF, CYC, TA, TAT, RS, MT, JHQ and JCA. All authors have approved the version to be published and agreed to be accountable for all aspects of the work.

  • Funding This work was supported by National Medical Research Council and Biomedical Research Council, Singapore (grant No. 10/1/35/19/674).

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval The study had the approval of the SingHealth ethics review board.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available on request.

Linked Articles

  • At a glance
    Keith Barton James Chodosh Jost B Jonas