Related Experiment Video
Updated: Aug 29, 2025

Subjective Refraction Test Using a Smartphone for Vision Screening
Published on: October 18, 2024
Leftover Astigmatism: The Missing Link Between Measured and Calculated Posterior Corneal Astigmatism
Purpose:
To quantify the total eye astigmatism that is not accounted for by measurement of anterior corneal astigmatism and posterior corneal astigmatism and knowledge of intraocular lens (IOL) astigmatism and assess whether it is correlated with candidate sources of or correlates with leftover astigmatism.
Methods:
Vector subtraction of anterior corneal, posterior corneal, and IOL astigmatism from total eye astigmatism as represented by spectacle astigmatism to yield a value of "leftover" astigmatism that is neither corneal nor lenticular. This value was derived in a series of eyes following cataract surgery. This novel entity was examined for correlation with candidate sources of or correlates with leftover astigmatism.
Results:
In 103 pseudophakic eyes with known IOL toricity, mean leftover astigmatism was 0.71 ± 0.43 diopters. This was significantly correlated with against-the-rule anterior corneal astigmatism (P < .001).
Conclusions:
Leftover astigmatism is clinically substantial. Because it is included in IOL cylinder power calculations based on refractive outcome, it may explain why methods of IOL cylinder power calculation using refractive outcome-based adjustments to anterior corneal astigmatism (previously described as adjustments for "posterior corneal astigmatism") are more successful than adjustment on the basis of measured posterior corneal astigmatism. [.
Related Concept Videos
Calibration Curves: Correlation Coefficient
Standard Deviation of Calculated Results
A broad Gaussian distribution curve has a wider standard deviation, representing a data set with...
Uncertainty in Measurement: Reading Instruments
Calculating and Interpreting the Linear Correlation Coefficient
Testing a Claim about Standard Deviation
The hypothesis testing for the claim of population standard deviation (or variance) requires the data and samples to be random and unbiased. The population distribution also must be normal. There is no specific requirement on the sample size as the estimation is based on the chi-square distribution.
As a first step, the hypothesis (null and alternative) concerning the claim about...
Calibration Curves: Linear Least Squares
For data that follow a straight line, the standard method for fitting is the linear...

