2023-10-31 - Eye Care CRG Meeting

Hi Anthony Khawaja Elaine Wooler do we have a meeting link for today's meeting?

Yes https://snomed.zoom.us/my/opthamology

008589

It's possible to get an "academic" membership/login to GMDN and see what glaucoma devices are enumerated. Unfortunately, it's quite limited. I'll try to find others, but so far the categories appear to be quite high-level such as "Eye valve" (42526), "Glaucoma shunt" (61127), "Glaucoma suupraciliary implant" (63567), and "Glaucoma micro-stent" (62945).

Each term has a definition. Here's the one for the micro-stent:

A non-bioabsorbable device designed to be implanted in Schlemm’s canal (within the eye) for the restoration of aqueous humour outflow and subsequent reduction of intraocular pressure as part of treatment for open angle glaucoma. It is a metallic device which may be intended to dilate, support, and/or create a channel into Schlemm’s canal; disposable devices associated with implantation (e.g., delivery system) may be included.

I personally don't view this as sufficiently precise to be useful for monitoring what type of glaucoma device (or surgery) was performed. Since there is no distinction between devices such as iStents or Hydrus implants. And other procedures that don't leave a device in the eye (such as a goniotomy or trabeculectomoy) might not be covered at all.

Thanks Eric.  My view is that it is ridiculous we need to make some long-winded description of something to infer a branded device, when there may still be some ambiguity due to varying interpretation.  It goes against every scientific bone in my body - good science requires us to be as precise, specific and replicable as possible.

Jane Millar Elaine Wooler - is there a procedure/implants specialist at SNOMED we could maybe start a discussion with?  Even if we add in "e.g. iStent Inject W" at the end of a description, for example, that would aid correct use and clarity.  I think we risk failing the principles of why we need SNOMED by not referring to specific implants/devices.

Many thanks

Hi Anthony Khawaja Could you give us a few examples of the procedures and how you record them at present and we can take it from there.  It may be that my implementation colleagues can provide some help here on linking the procedure and device and I can certainly speak to our Chief Terminologist for advice.

So I was wondering for iStent Inject that a procedure such as 439751005 |Transluminal dilation of aqueous outflow canal using stent (procedure)| or something like that and the device recorded separately.  NICE refer to it as - Trabecular stent bypass microsurgery.  The UK have added a concept in their extension - 819871000000106 |Insertion of bypass stent through trabecular meshwork (procedure)|

But if I can see what you record now that should take us forward and I can bring colleagues in.

Thanks Elaine.  For example, I would want to know which if the following trabecular bypass stenting procedures was carried, and that would be best defined by the device name:

• iStent G1
• iStent Inject
• iStent Inject W
• iStent Infinity
• Hydrus microstent

Another good example that was brought up at our meeting was trabeculotomy.  We would want to know whether this was done with iTrack, prolene, OMNI, Kahood Dual Blade, Bent needle etc etc.

Thanks

I fully agree with Anthony Khawaja - I really think that to ultimately be able to make use of these SNOMED termed operations, similar to coding diagnoses, they need to be as specific, unambiguous and easy to use as possible otherwise the data quality will be so poor it will not be worth the effort of analysing. 

I think that the operation coding needs to allow capture the broad procedure type e.g.:

• trabecular bypass,
• internal trabeculectomy/goniotomy,
• tube shunt/glaucoma drainage device)

AND then the specific procedure method (which could just be structured as children concepts of the parent term) - e.g.:

• trabecular bypass:
  • trabecular bypass with iStent G1 
  • trabecular bypass with iStent G2
  • trabecular bypass with Hydrus microstent
• internal trabeculectomy/goniotomy:
  • internal trabeculectomy/goniotomy with OMNI surgical system
  • internal trabeculectomy/goniotomy with Kahook Dual Blade
  • internal trabeculectomy/goniotomy with Prolene suture
• insertion of tube shunt/glaucoma drainage device:
  • insertion of Baerveldt 101-350 tube shunt/glaucoma drainage device
  • insertion of Baerveldt 103-250 tube shunt/glaucoma drainage device
  • insertion of Ahmad FP7 tube shunt/glaucoma drainage device

As well as this info we would need to consider how we capture:

• eye laterality (right eye, left eye);
• location within the eye (i.e. superior nasal quadrant, inferior nasal quadrant, superior temporal quadrant, inferior temporal quadrant); and
• number of devices simultaneously implanted (as potentially multiple of the same device can be implanted simultaneously)
• (combinations of different procedures performed simultaneously will need to be handled differently)

Thanks Elaine Wooler for looking into all this further. Interestingly when looking at the UK SNOMED browser for medications, it seems that branded medications / trade names are commonly used:

so hopefully we could do something similar with glaucoma procedures?

Hi Ian Rodrigues Similar to the discussion we had on branded medications, device brand names are also represented in National extensions not the International Release. This is because proprietary names may refer to different products depending on the country and the meaning of these names are dependent on the country or jurisdiction in which the product is approved. 

So in the UK you can record the procedure and additionally the brand of device used from the 49062001 |Device (physical object)| that is in the UK Extension e.g. 12234601000001103 |MAGEC internal spinal bracing and distraction system (Ellipse Technologies Inc.) (physical object)| is the device recorded for the procedure 773133007 |Surgical distraction of externally adjustable spinal deformity correction system (procedure)|.  The UK do not pre-coordinate these devices in to the procedure but the device could be post-coordinated with the procedure however you would need to speak to the UK SNOMED National Release Centre as I don't know what their plans are in this area.

So I think this may be an individual country implementation solution but we can work to ensure the non-branded procedure is represented in the International Edition to enable post-coordination with the specific device in countries National Extensions.  Additionally if you would like me to ask my implementation colleagues if they have experience on how this is achieved in other countries such as the US then happy to do that.

Thanks Ian and Elaine.  Can have the devices in the International edition, and it can be noted that the name might vary between countries?  The aim is that we can easily collect data on the same device across countries?

Thanks Anthony Khawaja - agree that would be ideal - the different brand names could appear as synonymous under the same SNOMED concept?

As Elaine Wooler mentioned, the current editorial policy for the International release of SNOMED CT (and has been for decades) is to not add branded products, for the reasons Elaine outlined.  As branded products do have different names and also go through various changes and improvements over time, it is not possible for the International release to manage the evolution of branded products, which would lead to inconsistencies and divergence from the current status of a product.  This is why it has been delegated to the individual national release centers.  While we do sometimes pre-coordinate the class of device used in a procedure, it is recommended that the specific brand of device be recorded separately (as a separate data element) as (at least in the US) they would have a GUID (see https://www.fda.gov/medical-devices/unique-device-identification-system-udi-system/global-unique-device-identification-database-gudid), which could be used to aggregate data across countries.