Eight steps to better corneal topography

Corneal topography is your main diagnostic tool for Orthok lens fitting. Get this right and everything else becomes so much easier. Do it badly, especially at first fit, and the rest of the journey will become harder. Fortunately, it’s not that hard a skill to learn and once you have the hang of it you will easily be able to teach your support staff if that’s how your practice works. These eight steps will show you how:

1.      Check tear film before attempting topography image capture

Corneal topographers measure the profile of the tear film meaning that any dry spots, gunk in the tear film, etc are going to interfere with your captured images. For this reason it is worth checking the anterior corneal surface with the slit-lamp before starting topography image capture and instilling a non-viscous lubricating drop prior to measurement if the tear film is not stable.

2.      Capture multiple maps

I recommend taking six maps for each eye at initial fit and four maps at follow up visits, with the intention to end up with four and two maps respectively. Averaging four maps at initial fit will help improve accuracy of any extracted corneal shape descriptors used for lens fitting while two in the bank at follow up will give you a backup if a difference map ends up looking wrong for whatever reason. Capturing more maps than you intend to keep in the first instance gives you some scope to remove any bad maps without subsequently having to bring the patient back to the instrument.

3.      Review image quality at capture stage

Keeping to the theme of avoiding patient inconvenience and quicker acquisition of images try to review the maps immediately after capture while your patient is still by the instrument so that you can easily capture additional maps if you end up deleting a whole bunch for being bad quality. How you do this will vary by instrument, but in general you need to open each captured image in turn and view the black and white raw image to assess quality of image capture and then the built curvature heat map to assess extent of data capture.

4.      Check for distorted mires

Topography software establishes a model of corneal shape by tracing each of the mire rings. It follows therefore, that if the mires are distorted the fitted model of corneal shape will also be distorted. The simple check is to examine the quality of the reflected image eyes and delete any maps where you see any wobbles in the mires. In the example below distortion to the image mires indicated by the arrows has caused a blemish in the constructed topography axial curvature map that could interfere with modelling if used for choosing first lens fit or difference map interpretation if used for a response visit.

5.      Mind the gap(s)

Gaps in the image mires can also be problematic as they may cause the computer software to jump across to a neighbouring mire ring while tracing the mire outlines. Unless you purposely try to displace lashes during image capture it is almost inevitable that you will see breaks from lash shadows, particularly from the upper lid. The important question is what should be considered acceptable and when you should attempt to move lashes out of the way – something that shouldn’t be considered lightly as the very act of displacing the lids can cause you to put pressure onto the eye that results in greater degree of corneal topography distortion than that you are trying to avoid!

Having evaluated many analyses I have come to realise that corneal topography software overall is generally good at tracing across gaps where there is not too much interference, e.g. from a single lash, but can become confused where there is a large gap to bridge or only small parts of the ring visible within larger gaps, e.g. from clumped lashes. Also, check for any trichiasis as this will cause tear film distortion which should be avoided as already covered above. What you choose to accept or delete is a personal choice but I’m hoping the links I have included here will help you reach your own conclusions.

6.      Maximise area of image capture

You want to have as much image as possible to work with at any visit, however, this is particularly important at baseline because at all subsequent visits you will be using one of these to create the post wear difference map. Difference maps are created by subtracting post from pre-wear curvature data on a point by point basis. No data at any point on either visit will result in a null data point causing the extent of your difference map image to be limited by whatever is the smallest extent of data. It doesn’t matter if your post wear map is the size of China if the baseline map is the size of England – as in this case, the difference map will also be the size of England!

The extent of data needed at baseline will depend on lens design and is covered in more detail in a separate post. In other posts, I have provided tips on increasing image capture area in problematic eyes.

7.      Only retain useful maps

Tempting as it is to retain more maps than you require I suggest you instead get into good housekeeping habits now and only keep the best four maps at baseline for each eye, two for post wear and delete all others. It may not cost much in computer space to store them all but a few years down the line you will get fed up trawling through the multiple maps you have for each patient to locate those that you previously established to be the best. Having multiple maps under each patient can also slow processing depending on which software you are using.

You should always retain all four baseline maps in case you ever need to compare baseline metrics at some point in the future. Difference maps, however, are not so important so once you have established which of the two post wear maps you have captured provides the best difference map, keep things tidy by deleting the other – this will make it far easier to quickly track through difference maps over time to establish how fit varies across visits.

8.      Add notes to maps

Fresh as it is in your mind now, you will soon find it difficult to recollect how you decided to interpret a map at a given visit, or which out of the group you considered the best to use for building difference maps. Adding notes directly onto the map, where this option is available, will make your future life that much easier. Yes, of course, you will have your practice management software notes, but it will take effort to move back and forth cross-correlating these with each topography map. Some software like the Medmont Studio supporting the Medmont E300 also allows short labels that appear next to the file name – in this case rather than label all four maps baseline maps for each eye as ‘baseline’, only add this label to the map with the biggest area (Step 6) – this gives enough information to identify the remaining baseline maps by date if needed while making it much easier at post wear visits to select which baseline map from which to build the difference map.

A final word

All up you can only do the best that you can with what you are presented – if despite your best efforts you are still left with suboptimal images you have little choice but to retain and use the best of the bad bunch. That said if you follow the steps you shouldn’t go far wrong.

About Paul

Dr Paul Gifford is a co-founder of Eyefit, an information resource to assist contact lens practitioners in all modes of practice. Learn more about him here.