Abstract:

We present four operated cataract cases with extreme axial lengths where the biometric formulas used for intraocular lens dioptre calculation have made surprising or unusual predictions. The four operated cases were part of a study on 1192 eyes that compared the accuracy of five biometric formulas. Case 2 and 3 had an unexpected increase in both keratometric and refractive cylinder, compared to the preoperative cylinder. Case 1 with short axial length showed unexpectedly poor results for Hoffer Q formula and Case 4 showed good results for all five formulas, even Hoffer Q that is not recommended for long axial lengths. Refraction prediction errors altered by test-to-test variations in keratometry or surgically induced astigmatism, as well as other erroneous measured variables, could influence the results of studies analysing accuracy of biometric formulas, if the number of cases is great enough, and should not be ignored

Keywords: