A while ago I made a couple of posts about the amount of several refractive surgery terms found in the title or abstract of papers found in PubMed to have an idea where was going most of the research effort and the evolution over time of them. The articles are this one and this one. This time I will do the same but with wikipedia pages which could be seen as an indicator of the general public interest.

The first one shows the number of visits per day to different wikipedia pages about several terms related with refractive surgery since 2008, it's only for the english version pages:




We can see that since around mid 2012 there is interest in multifocal IOLs and phakic IOLs but take into account the Y axis, for the multifocal IOLs goes up to 30 visits, for the phakic IOLs up to almos 150 visits, for other terms this axis goes up to several thousands of visits per day. There are four terms though with some outliers, IntraLasik, Lasik, PRK and refractive surgery. I'll have a closer look at them to see if we can see any trend. For Lasik there is an outlier going up to 300.000 +, lets have a look when we rearrange the Y axis just up to 5.000:




We can see now more detail, let's do the same for IntraLasik, PRK and Refractive Surgery but with the Y axis up to 1.500:




From the three graphics I will say that since 2013 the trend is going down for all of them except for multifocal IOLs and phakic IOLs that are starting to be looked for , this could mean that there is less interest in refractive surgery in general or it could be something else. I won't run the time series analysis here but it is interesting to compare some of them with the first post for some terms in PubMed.




In the following program I have put all the graphics together from the last three posts. You can choose from different graphics and options to limit the observations by different factors and in some of them you can click on the points to see more information. I recommend to click here to see it in full screen in another tab.



Up to now we have seen how the patients were before and after refractive surgery and related the spherical equivalents to each other but at the moment we have no information about the astigmatism, we haven't made any description graphic and we have no idea what is happening to the axis. One way to include somehow the cylinder of the astigmatism is to take as postop value in the graphic the defocus equivalent (DEQ). The DEQ correlates much better with the visus that the SE and is calculated with the following formula:

DEQ = absolute value of SE + ½ absolute value of the cylinder

 The DEQ is always positive and if we relate it to the SE preop we wil get a graphic like this one:




The points in that graphic are jittered by 0.025 to avoid overplotting, that means they are moved randomly 0.025 D horizontally and vertically. In this graphic we still have no idea about the axis or the magnitude of the cilinder. I will show now a graphic to describe the cylinder and the axis of the astigmatism:




In this graphic the axis of the astigmatism is doubled. The reason to do this is that the 180º and 0º axis are in the opposite side of the graphic and doubling the angles we see them at the same point. There are other reasons for doubling the angle (vector analysis) but I won't go through them in this post. You can find further information about double angle plots in this paper by Holladay.

In the graphic each circle is -0.50 D of cylinder and the outter circle is -5 D, the number in parenthesis is the original axis of the astigmatism. The points that fall in the green area have astigmatism with the rule (WTR), in the blue area are against the rule (ATR) and in the read area the oblique astigmatism. The red cross is the mean cylinder considering the axis, an example to explain it: a cylinder of -1 x 90º and a cylinder of -1 x 180º the mean would be 0 since to correct a cylinder of -1.00 x 90º we need a +1.00 x 90º or a +1.00 -1.00 x 180º. The shaded ellipse around the red cross is the standard deviation of the x and y values in cartesian coordinates and we can interpretate it as the ratio between the ATR + WTR astigmatisms and oblique astigmatisms, when the horizontal axis is larger than the vertical there are more ATR + WTR than obliques astigmatisms, if the vertical is larger then there are more obliques astigmatisms.

In the next graphic I will plot the astigmatism postop related to how it was preop (ATR, WTR or oblique):




Now we see that the astigmatism is much less, as expected, but we can also see if we under or over corrected or if we treated another axis. The blue points were ATR preop, so if they fall in the blue area we undercorrected, if they are in the green area we overcorrected and if they are in the red area we treated the wrong axis, equivalently for the green points (WTR preop) if they fall in the green area we undercorrected, if they are in the blue area we overcorrected and if they are in the red area we treated the wrong axis. For the oblique (red points) we cannot see if we over or under corrected but we can see if we treated the right axis. We could separate the oblique astigmatism points in 30º-60º and 120º-150º giving them other colors and we could see if we over or undercorrected.

We can divide the graphic as with the SE by any factor we want, as an example the following graphic divides them by the amount of cilinder preop, more or less than 2 D:




Here we can see that for the cylinders preop under 2 D the points are closer to 0 and it seems we got more oblique astigmatisms, there is quite a lot of overplotting so we should change the scale of the plot making the outer ring -1.00 D for example and we will see much more detail.

This was the last post of the series to get a better insight of refractive surgery results. Remember, if you want any of this graphics but don't know how or don't have time, I can do them for you.