Gonioscopic grading

In addition to IOP-related codes (see separate thread), another area of discussion from the CRG meeting on 3/23/23 was the identification of gaps in gonioscopic grading. Some felt that these exam areas are not populated consistently, but it sounds like in the UK there is fairly consistent documentation, and that standardized coding in this area would be high-yield. Currently, SNOMED has codes such as "narrow angle of anterior chamber", "wide open angle", and "moderate open angle", but there are not codes to indicate exact grading specification or quadrant of observation. 

Elaine Wooler asked to see what gonioscopic grading typically looks like in the EHR, so I have copied over the interface from Epic Kaleidoscope as an example: 

For each eye (right and left), there are 4 boxes/fields corresponding to each quadrant: superior, nasal, temporal, and inferior. The "method" field allows the clinician to indicate the method of gonioscopic grading used. 

Laterality: Similar to IOP, these would apply to each eye (left eye, right eye). In a further level of specification beyond laterality, each gonioscopic observation has a quadrant designation. Thus, you would have observations for right eye superior, right eye nasal, right eye temporal, and right eye inferior, and similar for left eye. 

Associated Units: None. There is not really a standard "unit" of measurement for these gradings (some assessment of degrees occurs, but typically this will correlate into a grading - see "range of observed values" section below). 

Method of gonioscopic grading: Here are some examples of gonioscopic grading classification systems: 

• Spaeth
• Scheie
• Shaffer
• Van Herick
• Non-codified (just reporting what angle structures are visible) 

During our discussion at the meeting on 3/23, it was felt that some of these may be quite complex (particularly Spaeth). The Spaeth classification system has several components that can be combined together, resulting in a large list of possibile gradings. It was felt that perhaps we should start with codifying the simpler classification systems first, and then if there is user demand for Spaeth, we can deal with those at a later time. 

Range of Observed Values: I am listing the range of observed values below for a few of the classification systems above. Others, please correct if I have made any mistakes. Some references: https://www.aao.org/education/disease-review/gonioscopic-grading-systems

https://www.semanticscholar.org/paper/The-Spaeth-Gonioscopic-Grading-System-Assessing-the-Marsh-Cantor/39bf462444a6f4bc205ea40c6d0f518f6c90f217

• Scheie: Wide (Wide open), I (Slightly narrowed), II (Apex not visible), III (Posterior half of trabeculum not visible), IV (None of angle visible) 
• Shaffer: Grade 4 (45 degree to 35 degree angle; Wide open); Grade 3 (35 degree to 20 degree angle; Wide open), Grade 2 (20 degree angle, Narrow), Grade 1 (less than or equal to 10 degree angle, Extremely narrow), Slit (0 degree angle, narrowed to slit) 
• Van Herick: Grade 4 (Open), Grade 3 (Open), Grade 2 (Narrow), Grade 1 (Angle Closure likely)
• Non-codified (just reporting what angle structures are visible): none, Schwalbe's line, anterior trabecular meshwork, posterior trabecular meshwork, scleral spur, ciliary body band

An additional consideration is grading angle pigmentation, which is codified in Scheie. We did not specifically discuss this in the meeting and I'm not sure how often this is documented or contained in structured form in EHRs, but including here to try to be comprehensive.

• Scheie angle pigmentation: None, I, II, III, IV  (larger numbers indicating greater pigmentation)

Again, Elaine Wooler let me know if any additional info is needed, and thanks everyone for you input on this!

I agree - this is complex!  Perhaps one high priority that is pragmatic:  is an angle is occludable or not?  I.e. 180+ degrees of apposition with posterior trabecular meshwork?  I believe that was the definition used in the EAGLE and ZAP trials (though we should check!).  PAS also important.  Also new vessels and level of pigment may be relevant.  Which quadrant and extent could be post-coordinated?  Just thoughts....

Agree that these elements are important, but I think there is already some level of representation in SNOMED. For example, there is an existing SNOMED code called "occludable anterior chamber of eye" (https://athena.ohdsi.org/search-terms/terms/4227977).  There are also several synechiae-related codes in SNOMED, including "peripheral anterior synechiae" (https://athena.ohdsi.org/search-terms/terms/441016). There are also some related to new vessels ("rubeosis iridis", "neovascularization of the iris and/or ciliary body", "proliferative diabetic retinopathy with iris neovascularization due to diabetes mellitus"). So it looks like these have some level of representation/codification in SNOMED already, but the specific gonioscopic gradings do not. 

Great - so we should ensure that the best of these terms are in our suggested "preferred list".  We may need to clarify / update the definition of occludable to an international consensus if not already aligned.  Thanks!

Great, agree worth having a pragmatic initial approach.

FYI, in OpenEyes (which is an ophthalmic EPR very widely used in the UK and also increasingly internationally) there are two options for angle classification:

Basic - each angle quadrant having a yes or no option for being graded as open;

Shaffer grade - each angle quadrant can be graded 0-4

Additionally, it is possible to grade the overall angle pigmentation (per eye) with the following options (I appreciate there will not be any clinical standardisation of these pigmentation levels):

• very light
• light
• moderate
• heavy
• very heavy

And it is possible to grade the overall iris configuration (per eye) with the following options:

• flat
• plateau
• concave
• bombé
• steep

The only other clinical finding that I can think of that don't seem to have a SNOMED finding code is angle recession (although there is a disorder code (Angle recession (disorder) SCTID: 193525001)

I had to think about gonioscopy grading this week as I was editing the AAO glaucoma book, and so made an attempt to harmonize existing grading systems with a proposed standard:

https://docs.google.com/spreadsheets/d/1BBEU6M8MA7BB5xL27dmC7QXBrqI218SkaFRc1Jp_JS0/edit?usp=sharing

Based on this, we could consider 4 concepts to include in SNOMED (see link).  I am most unclear about how to specify the iris approach (may be similar enough to iris configuration that we don't need it?) Comments welcome!

Thanks Michael Boland I think this is a really interesting framework. Are you engaging with a broader group in the process of harmonizing the grading systems, either via AAO or AGS or something? Since it seems like for this, we need kind of a unified standardized framework as opposed to just submitting individual terms (like we would for maybe a missing diagnosis or something), it may be worthwhile to make sure there is broader buy-in beyond our smaller group here. It would be cool to see if we could then take the proposed standard (once agreed upon) not only to standards bodies like SNOMED, but also back to EHR vendors and see if they could implement and update their interfaces accordingly so there is greater standardization in data entry. 

I am actually proposing something much less ambitious than harmonizing grading systems or creating a new one from scratch. Rather, I am proposing we merely identify the key concepts included in existing systems and then ensure there is a place to capture those in SNOMED. That way, clinicians can grade angles with whatever scheme they like and there will be a place to put it. We have multiple grading schemes now, and none has been adopted widely, so I have no reason to believe that a new system would fare any better.  As I was reviewing grading for the BCSC book, it did occur to me that the major grading schemes had concepts in common and that is what I summarized in the table linked above.  I agree some consensus from interested glaucoma informatics folks is important and this group seemed a good place to start. We should also engage the US glaucoma informatics group (we are meeting this week so I will send this around).

Many thanks Michael Boland  I really like your very wise and sensible approach. The document mapping the grades between the different classifications is super useful.

I agree with Eric Brown  - would be sensible to have the SNOMED examination finding term in plain English (i.e. "Closed anterior chamber angle") with the synonyms for the corresponding classification systems (i.e "Sheie Grade IV angle" / "Saffer Grade 0 angle" / "Spaeth Grade A angle"). With ?post-co-ordination for each quadrant for Right & Left eye.

I would say that other than "Steep Iris approach" there is enough overlap with iris configuration that I think we could do without, especially if the Shaffer grades are included as synonymous terms to the angle openness terms. 

Great point about Iris Bombé - currently there is a SNOMED "disorder" code for this [Iris bombé (disorder) SCTID: 75895005] but no SNOMED "finding" code - @Elaine Wooler is there any established convention within SNOMED of having both a finding and disorder code where a clinical feature could be either an examination finding and/or a diagnosis? Do we need both, or is there some kind of hierarchy between the different concept categories?

Hi All,

Just catching up on the discussions so apologies for any delay.  We can add the examination findings if they are not already there - if you can review the content under 366019007 |Finding of angle patency (finding)| for any gaps.  However we would not be able to add synonyms with the classification system used as they are not true synonyms. 

An alternative way to do this would be to have observable entity concepts in SNOMED - these are essentially the question concepts which take values similar to 363983007 |Visual acuity (observable entity)|, 386708005 |Visual acuity - left eye (observable entity)|, 413077008 |LogMAR visual acuity left eye (observable entity)| and we may be able to pre-coordinate with each quadrant and laterality e.g. Gonioscopy grading of superior quadrant of retina of left eye (observable entity).

The grading systems would be the techniques used and could be added under 254291000 |Staging and scales (staging scale)|. If you look under 277457005 |Histological grading systems (staging scale)| for examples of specific systems used in histology.  

Whether we could also add the specific grading systems pre-coordinated with the observable I'm not sure e.g. Spaeth gonioscopy grading of superior quadrant of retina of left eye (observable entity) I'm not sure though it can be modeled as we have the body structures under this hierarchy 314857008 |Structure of quadrant of retina (body structure)|.

I'll take this for discussion with our Chief Terminologist and get back to you.

I also like the approach of a pragmatic mapping.  Though, we will need some kind of international clinical approval for this - which is what the CRG is all about.  If we agree with the active CRG that this approach is a good one, we can then seek approval from AGS, EGS, AAO, together with other items/changes we are approving.  For me, "Occludable" may be better than "closed", as >180 degrees apposition may not be fully "closed", but that is a minor point.  Agree that varying degrees of open is not so helpful.  I also think that iris configuration may not be that important other than for plateau?  We also need PAS, NVA, but maybe that is already captured.  Thanks all

HI

It is up to you all to decide what you want to include in SNOMED CT and Elaine will get advise on how things can be achieved. But I need to look at it from the perspective of intellectual property so if you reach a point of agreeing internationally, for example, on a specific scale to be included 

1. Authoring need to discuss how to manage
2. We need to seek permission to use said content and put an agreement in place - I am happy to work with you on this, aim to join the next call

Jane

Hi everyone,

I wanted to provide an update here for visibility based on our discussions after the August 2023 CRG meeting. Apologies for the delayed update. Here were some of the key decisions re gonioscopic grading, and I will put together into a summary handout that will be attached to the meeting materials/documents as well for easy detection later. Anthony Khawaja please correct me if I misrepresented anything below: 

1. We decided on representing the gonioscopic exam findings as observable entities, as suggested by Elaine Wooler. There was a consensus to pre-coordinate quadrant and laterality for the following features: degree of angle openness, degree of angle pigmentation, presence of neovascularization of the angle, and presence of synechiae. The pre-coordinated options would be as follows for those features (the underscore is a placeholder for the feature itself): 
   1. ____, superior quadrant, right eye
   2. ____, inferior quadrant, right eye
   3. _____, nasal quadrant, right eye
   4. _____, temporal quadrant, right eye
   5. ____, unspecified quadrant, right eye
   6. ____, superior quadrant, left eye
   7. ____, inferior quadrant, left eye
   8. _____, nasal quadrant, left eye
   9. _____, temporal quadrant, left eye
   10. ____, unspecified quadrant, left eye
2. For iris configuration and iris approach, the group recommended to just have laterality (no quadrant specification needed), i.e.: 
   1. Anteriorly bowed iris configuration, right eye
   2. Anteriorly bowed iris configuration, left eye
   3. Posteriorly bowed iris configuration, right eye
   4. Posteriorly bowed iris configuration, left eye
   5. Flat iris configuration, right eye
   6. Flat iris configuration, left eye
   7. Steep iris approach, right eye
   8. Steep iris approach, left eye
   9. Shallow iris approach, right eye
   10. Shallow iris approach, left eye 
3. Regarding plateau iris specifically, there was consensus to retire the "plateau iris" code (SNOMED 232080006), because this did not offer any additional value given that there already exists "Plateau iris configuration" (404630001) and "Plateau iris syndrome" (404631002). 
   1. On a related note, "Plateau iris configuration" (404630001) needs laterality added, i.e.:
      1. Plateau iris configuration, right eye
      2. Plateau iris configuration, left eye 

I believe with the above changes, in addition to the prior discussions regarding representation of angle openness, we should now have standard representation of the various gonioscopic grading systems, including an ability to represent the affected quadrant. 

Others in the CRG, please chime in if I didn't capture everything. I will go ahead and prepare a tentative summary document based on the above and attach to the meeting link for the August 2023 CRG meeting, but will revise/update if anyone else has any suggested edits. Thanks!

My apologies, I realized that my last post failed to include angle pigmentation. The closest code is 251728001 ("Angle meshwork pigmentation"), but this does not reflect the degree of pigmentation nor laterality information. The proposal was to delineate the following severity levels: none, mild, moderate, heavy, and unspecified severity.

Thanks for this Sally.  What great progress for a very complex topic.  I agree with all of the above. Ian Rodrigues - any comments before Sally formalises in a document?  Thanks

Thanks very much Sally Baxter - fantastic work! I agree with Anthony Khawaja - it all looks great!

For the degree of angle pigmentation, I think we should maybe refer to it as pigmentation "level" rather than "severity" (to avoid confusion with disease severity scales). Perhaps "light" would be better than "mild" (if we are using "heavy" instead of "severe"). We could also use "medium" instead of "moderate", but I actually think "moderate" probably sounds better!

Regarding the Iris approach/configuration options - currently we have right & left versions for:

• Anteriorly bowed iris configuration
• Posteriorly bowed iris configuration
• Flat iris configuration
• Steep iris approach
• Shallow iris approach

...Just wondering if we could get away with reducing it to the following options (also to be more consistent with the Spaeth classification)?:

• Steep iris configuration (with synonymous terms: "Convex iris configuration" and "Anteriorly bowed iris configuration")
• Regular iris configuration (with synonymous terms: "Flat iris configuration" and "Shallow iris approach")
• Concave iris configuration ((with synonymous terms: "Queer iris configuration" and "Posteriorly bowed iris configuration")

or am I missing something about the subtleties between iris approach and iris configuration that we need to differentiate between??

I'm happy with that, and if no disagreement here, we can then put it to the AGS/EGS/AAO, as planned?

Do you want me to hold off authoring this content until put to the associations?

Hi Anthony Khawaja  and team - I have drafted the document and placed it here as a google doc. I'll also copy here below my message for ease of review, but feel free to add comments/edits into the Google Doc directly as well. Also please add any CRG members I may have forgotten - I included those who I remember being active in some of these discussions but may have inadvertently left someone out.

I tried to keep it brief as to not make it overly onerous for review by the societies. For angle pigmentation, I did change "mild" to "light" per Ian Rodrigues suggestion.

I saw a bunch of comments here regarding "flat" vs. "steep" vs. "shallow"... and admit I am not totally sure of the best way to approach this. You can see in the summary table that this in part stems from the summary/overview that Michael Boland previously provided. There is a distinction between iris configuration and iris approach based on components from the Spaeth and Shaffer classification systems, respectively. If we (or those we consult) feel it would be better to consolidate some of these components, I am totally open to that. I think having those involved in angle closure research comment would certainly be helpful. 

Thanks! Copying document below as well if you want to browse (same as what's in the google doc).

=======================================================================

SNOMED International Eye Care Clinical Reference Group (CRG) Proposal for Representing Gonioscopic Exam Findings

October 2023

CRG Members (alphabetical): Sally Baxter, MD; Michael Boland, MD, PhD; Eric Brown, MD, PhD; Anthony Khawaja, MBBS; Ian Rodrigues, MBBS; Joshua Stein, MD, MS; Benjamin Xu, MD, PhD; Sophia Wang, MD, MS

Background/Rationale: Gonioscopic findings that enable characterization of the angle are important components of the eye exam, particularly for glaucoma. A systematic analysis comparing eye exam fields in electronic health record (EHR) systems in both the United States and the United Kingdom against existing concepts in SNOMED revealed that there were several gaps in representation in this area. Formal and informal surveys of glaucoma specialists from multiple institutions revealed that this comprises a high-priority area for standardized representation. Over the last several months, CRG members have discussed these items both synchronously via teleconferences and asynchronously on the SNOMED Eye Care CRG Confluence space.

Current Gaps in Representation: Although SNOMED already contained some codes related to angle characterization, such as “narrow angle of anterior chamber,” “wide open angle” and “moderate open angle,” key gaps included the following:

• Inability to delineate the laterality or quadrant of observation
• Inability to delineate the grading/classification system being used
• Inability to delineate exact angle features, including:
  • Extent of angle openness/visible structures
  • Degree of angle pigmentation
  • Iris configuration (with the exception of plateau iris, which did have associated codes already but some were duplicated)
  • Iris approach

Proposal to Address Gaps: The CRG proposed the identification of key concepts included in existing gonioscopic grading systems and ensuring that there is adequate representation of these in SNOMED. This recognizes the wide variation in how clinicians document gonioscopic exams and acknowledges that clinicians may use different grading schemes.

Based on discussion with SNOMED terminologist Elaine Wooler, it was determined to be possible to represent these exam findings as observable entity concepts in SNOMED, with pre-coordination of each quadrant and laterality. The grading systems themselves could be represented in the techniques used and added under 254291000 |Staging and scales (staging scale)|.

The CRG generated an overview of various gonioscopic grading systems to help understand what concepts may require representation:

Following this, the CRG’s proposals to enhance representation of gonioscopic exam findings in SNOMED include the following aspects.

• Grading System: Enable specification of the grading system used, e.g. Scheie, Shaffer, Spaeth, visible structure
• Pre-coordinate laterality and quadrant for the items below, allowing specification of superior/nasal/inferior/temporal quadrants as well as “unspecified quadrant” for each eye:
  • Degree of angle openness
  • Degree of angle pigmentation
  • Presence of neovascularization of the angle
  • Presence of synechiae
• Pre-coordinate laterality only (no quadrant specification needed) for the following:
  • Iris configuration
  • Iris approach
• Degree of angle openness: Concepts should be developed to represent each possible grading value from the various systems (e.g. I to IV, 0 to 4, A to E, various structures)
• Degree of angle pigmentation: Similar to the above, concepts should be developed to represent each possible grading value from the various systems (e.g. I to IV, 0 to 4) as well as generic representation of “light,” “moderate”, “heavy,” and “unspecified.”
• Iris configuration: Concepts should be able to represent the Spaeth classification (f, b, c, p) as well as a generic representation of “flat,” “anteriorly bowed,” “posteriorly bowed,” and “plateau iris” (of note, the latter – plateau iris configuration – is an existing code in SNOMED; see below)
  • Plateau Iris: Regarding plateau iris specifically, there was consensus to retire the "plateau iris" code (SNOMED 232080006), because this did not offer any additional value given that there already exists "Plateau iris configuration" (404630001) and "Plateau iris syndrome" (404631002).
• Iris approach: Concepts should be able to represent the Shaffer grading system (e.g. numeric value) as well as generic representation of “steep” or “shallow”

Anticipated Impact: With the changes proposed above, gonioscopic exam findings would be able to be represented in structured form in the SNOMED International terminology. This may provide an additional layer of validation above diagnosis/condition codes alone. In addition, these could be subsequently used for defining relevant cohorts, such as those with narrow/occludable angles or angle closure glaucoma. This may improve the utility of electronic health record data for secondary research in glaucoma.

Thanks Sally Baxter  Similar to the severity stages we would need to ensure there are no intellectual property issues with representing the named grading systems.

Jane Millar and Ian Green I think you wanted to help with this also.

Great, thanks!  I will circulate to our local angle-closure glitterati for their opinion and get back (hopefully they will directly comment on the Google Doc).

Update so all are aware: Flora Lum sent the document to the AAO Preferred Practice Pattern chair for glaucoma (Steve Gedde, MD, Bascom Palmer Eye Institute), and he also conferred with David Friedman, MD, angle closure expert from Mass Eye and Ear. They did not have any revisions for the document, and she relayed that they felt very positive about the work. Thanks!

Brilliant!  I have not had any response from my local angle-closure gurus.  I will chase once it has been a couple weeks.  Thanks.

Thanks Sally Baxter This all looks really great. 

I think the difference between iris configuration and iris approach are nicely delineated in the table.

For the Iris Approach options of 11 to 19 and at 20 to 34 degrees there currently is no label. I suspect we will need to come up with something for the name of the new SNOMED concepts that Elaine will need to create. What do you suggest? ... 11-19deg = "moderately steep" and 20-34 deg = "moderately shallow"??

Similarly for the degree of angle pigmentation we have two moderate labels. Again, I think we would need to distinguish these with different names... II/2 = "medium-light" and III/3 = "medium-heavy"??

Hi all - some interesting comments from Paul Foster (pasted below).  He believes that a code for 180+ degrees of ITC would be most meaningful, and should be a mandatory field for glaucoma examination (I agree, and I know we discussed this - it could be an addition to what we are doing).  I don't think Paul is suggesting deviation from what we are proposing though, just that addition as being the most pragmatic / clinically meaningful.

Hi Anthony 

Yes, of course, please share.

I do agree that the division between 180° closed or not it’s probably the key physical sign and this should have its own snomed code, and I would hope that this would become mandatory in a glaucoma assessment.

Best wishes, 
Paul

From: Khawaja, Anthony <anthony.khawaja@ucl.ac.uk>
Sent: Sunday, October 29, 2023 12:38:14 PM
Subject: Re: SNOMED CT Gonioscopic grading

 
Thanks very much, Paul.  Would you mind if I shared these comments with our group?  In short, these codes will not define clinical practice, but are a way that we can help harmonise data collected across different systems.  It seems you do not have an objection to the codes we are suggesting, though do question whether they are clinically impactful? 

We did have talk about pragmatically having a code for >180 degrees ITC (i.e. inclusion criterion for the major RCTs we base our management decisions on) - do you think that would be helpful?

The other thing we are trying to do is have codes for what is already being captured in a structure way in the major EMRs, i.e. EPIC, Medisoft, OE.

Thanks again,

Anthony

From: Foster, Paul <p.foster@ucl.ac.uk>
Sent: 29 October 2023 12:23
Subject: Re: SNOMED CT Gonioscopic grading

Hi Anthony 

Thanks very much for sending this to me. This is an interesting project. However, I'm a little pessimistic that it will achieve a useful standardisation. The primary reason for this is that I think the evidence that is needed to guide such a revision and to make an advance in the standard of clinical care, is currently lacking. there are really no useful population studies, looking at gonioscopy characteristics in relation to outcomes aside from the Mongolia Singapore and studies. I think to a certain extent EAGLE also helps, but the gonioscopic characterisation was relatively less well standardised and prone to a variety of interpretations. ZAP. With Mongolia, Singapore and ZAP, careful attempts were made to standardise the observations.

In addition, the knowledge and interpretation of the various schemes that you list Scheie, Shaffer and Spaeth), are very variable. I would wager that less than 50% of the panel involved in this project, fully understand or appreciate the variation in grading technique. My experience has been that nobody outside a very small circle of gonioscopy fanatics, understands the difference in grading specification between Scheie and Shaffer. The Spaeth classification is more broadly understood, because of George's incredible reach and diligence in training, such a large number of fellows who then went on to become influential in US and global ophthalmology. 

Muchof what is proposed remains based on received wisdom and is not based on hard evidence. I think really all we know is that contact between the Iris and trabecula meshwork can, and to result in significant elevation of interocular pressure. The apparent risk of having a narrow but open-angle is much lower than previously suspected (ZAP study), and therefore identifying a narrow angle is really largely pointless. 

Clearly, angle and iris neovascularisation are an important sign that need to be detected. However, degree of pigment in the angle is really less well codified and I'm not sure that anybody would really be able to produce hard evidence linking particular level of pigmentation with a particular prognosis for patients. Peripheral anterior synechiae are not mentioned, and this is justifiable as again we don't really have an idea of how they impact the visual prognosis for an individual patient. My feeling is that they are not a particularly important sign now. Most important things seem to be the level of pressure and that is directly determined by the amount of iris that is in contact with the trabecular meshwork. 

A further problem is that gonioscopy does not easily identify the key physical sign for glaucoma prognosis. This is irido-trabecular contact.

Really all that the observer can identify is that they cannot see the trabecular meshwork. Contrast this with OCT examination where it is possible to definitively Identify contact between Iris and trabecular meshwork, and I think this clinches the key problem with the gonioscopic examination in the modern age.

I'd be very happy to comment further or input any on any supplementary document. I suppose the one benefits of this may be producing a reference standard that researchers can use in designing clinical studies that will then provide the evidence needed. However, I would not be particularly enthusiastic myself to move forward in this regard because  I think we have really into the age where OCT is the primary clinical examination and gonioscopy, in the western hemisphere is not a particularly helpful clinical examination in quantifying a patient's prognosis.

ATB

Paul

Paul J Foster BMedSci(Hons) BMBS PhD FRCS(Ed) FRCOphth FRCS(Eng)
UCL Professor of Ophthalmic Epidemiology and Glaucoma Studies
Honorary Consultant Ophthalmologist, Glaucoma Service, Moorfields Eye Hospital 

And a follow-up response from Gus Gazzard, along similar lines:

I’d hesitated to respond as I feared my response would be too ’negative’ - along the lines of ‘am I bovvered?’ 

We all 3 (with others) used to spend a great deal of energy and time in training and disseminating gonio’ teaching - EDP4 etc. It’s highly likely that Paul and Winnie’s great efforts on their return from Mongolia drove a hugely increased pick-up and diagnosis of PACD that previously went untreated. 
However, we know now that ITC in the absence of signs of GON or high IOP may not matter, even differentiation between PAS and iris processes (let alone elegant distinctions of differential pigment deposition - see Paul’s lovely Archives photo essay) may not matter - or at least we don’t know if they do. 

So I think we are left with ITC  (with a proper definition, but goodness only knows who outside of a geeky few really abide by ‘ visible contact between iris and posterior, usually pigmented, presumed functional TM in primary position with no indentation on high magnification dark-room gonioscopy’) 

… and that being (probably) relevant only with 180 degrees or more … [although I’m not certain how robust even that cut-off is - PAUL? - as I recall that some of this was inferred from PAS incidence rates with different degrees of closure in TPGS/Mongolia].

We certainly don’t monitor patients for the risk of 'angles narrowing in the future' as some once did, and the test-retest reliability and inter-observer reliability of gonio’ grading is probably sufficiently poor that even IF it were relevant and even IF people did it carefully in the right conditions and even IF we recorded it properly in a consistent manner we may not be able to draw sufficiently accurate  inferences to be useful!