## Purpose

Age-related changes in lens elasticity and ciliary muscle contractility can affect how ocular parameters respond to cycloplegia, and therefore intraocular lens (IOL) power measurements calculated by formulas using anterior chamber depth (ACD), lens thickness (LT), or white-to-white (WtW) for effective lens position prediction can vary. In response, using swept-source optical biometry in prepresbyopic and presbyopic eyes, we investigated changes in ocular parameters and IOL power calculations attributable to cycloplegia.

## Design

Cross-sectional study.

## Methods

In 38 prepresbyopic and 42 presbyopic eyes, we measured pupil diameter, radius of corneal curvature values, central corneal thickness, WtW, ACD, LT, and axial length both before and after cycloplegia. We determined IOL power calculations with the Sanders-Retzlaff-Kraff/theoretical, Holladay 2, and Haigis formulas. To pinpoint the effect of cycloplegia, we recorded refractive predictions in pre- and postdilation conditions according to the same IOL power calculations, even if postdilation IOL power calculations had changed.

## Results

With cycloplegia, pupil diameter changed significantly more in presbyopic eyes ( *P *< .001). Central corneal thickness decreased in prepresbyopic eyes ( *P *= .048), whereas WtW increased in presbyopic eyes ( *P *= .02). In both groups, ACD and LT changed significantly ( *P *< .001). IOL power calculations according to the Holladay 2 formula differed in prepresbyopic eyes ( *P *= .042), and refractive predictions with the Holladay 2 and Haigis formulas differed significantly in prepresbyopic eyes ( *P *= .043 and *P *= .022, respectively).

## Conclusion

Surgeons should consider the effect of cycloplegia on refractive prediction errors and IOL power calculations determined with Haigis and Holladay 2 formulas, especially in prepresbyopic ages.

Though primarily used in detailed fundus examinations, cycloplegic agents are also used in surgery, refraction, and the treatment of some ocular diseases. Meaning “paralysis of the ciliary muscle” (CM), cycloplegia results in the backward movement of the crystalline lens and decreased lens thickness (LT). Because of changes in crystalline lens dynamics, anterior chamber depth (ACD) increases. Because of the change in anterior chamber parameters, especially in lens elasticity, CM morphology, and contractility with age, the response to cycloplegia can vary.

Various studies have reported changes in anterior segment parameters after pupil dilation with different devices: Pentacam, anterior segment optical coherence tomography, ultrasound biomicroscopy, A-scan ultrasonography, AL-Scan, Lenstar LS 900, and IOLMaster 500. Designed with specific age groups in mind, most of these studies reported a significant increase in mean ACD and decrease in LT with cycloplegia.

The accurate measurement of ocular biometry and intraocular lens (IOL) power is essential for achieving the target refraction after cataract surgery. To that end, measurements of axial length (AL), ACD, and keratometry (K) values are important parameters for determining effective lens position (ELP) and IOL power calculations. Commonly used 2-variable formulas, such as Hoffer Q, Holladay 1, and Sanders-Retzlaff-Kraff/theoretical (SRK/T) predict the ELP based on AL and K. The Haigis formula uses AL, ACD, and corneal radii of curvature in millimeters for IOL calculation. Olsen’s formula, a 5-variable predictor, uses AL, K, ACD, LT, and preoperative refraction. The Holladay 2 formula uses 7 parameters (ie, those used by Olsen, plus patient age and the horizontal white-to-white [WtW] measurement) with variable relative importance for more accurate ELP estimation. However, using IOLMaster 500, Srivannaboon and colleagues demonstrated high accuracy of Holladay 2 formula even when used in the absence of an LT value. In the development of the Holladay 2, measurements under cycloplegia were the recommended protocol for biometry measurements. As such, the difference in ACD and LT induced by cycloplegia in Haigis, Olsen, and Holladay 2 formulas might affect IOL power calculations.

Accordingly, we investigated how age affects responses to cycloplegia in terms of anterior chamber parameters and IOL power calculations using swept-source optical biometry (IOLMaster 700; Carl Zeiss Meditec AG, Jena, Germany).

## Methods

## Participants and Protocol

This study was approved by the ethics committee of Mugla University and conducted according to the Declaration of Helsinki. Each participant was informed of the purpose of the study and signed a written consent form. A sample of 38 eyes of 20 men and 18 women, with a mean age of 30.1 ± 6.8 years (range 24–39 years), and 42 eyes of 19 men and 23 women, with a mean age of 56.6 ± 7.3 years (range 47–64 years) participated in this cross-sectional study. The exclusion criteria were spherical equivalent refractive error > ±1.0 diopter (D), previous ocular surgery, corneal opacities, severe dry eye, history of contact lens wear, retinal disease (eg, cystoid macular edema), elevated scars, glaucoma, a narrow iridocorneal angle, cataract, an inability to open eyelids widely, ocular inflammatory disease, and poor ocular fixation.

All measurements were taken between 9 AM and 11 AM to eliminate the effect of diurnal variation on measurements. All the noncycloplegic and cycloplegic measurements were performed within 1 hour on the same day by the same experienced examiner according to the manufacturer’s recommendations. Three measurements were acquired using swept-source optical biometry before and after cycloplegia, and the average of the each parameter was used for analysis. The quality control criteria were used in accordance with the manufacturer recommendations; on the IOLMaster 700, the operator can see the whole scan image and visually check the eye geometry and axis of the measurements; also the foveal scan checks the correct fixation. Standard deviation (SD) values of the ACD and LT measurements are calculated, and the device warns the operator of low-quality results, such as the SD for ACD > 0.021 mm, for LT > 0.038 mm, and AL > 0.027 mm.

Participants were asked to blink just before measurements were taken to create an optically smooth tear film over the cornea. All measurements were taken in room light condition. Cycloplegia was achieved using 3 drops of an eye solution containing cyclopentolate 1% every 5 minutes. After cycloplegia, the pupillary light reflex was checked, and about 45 minutes later, the measurements were performed.

We recorded pupil diameter (PD), radius of corneal curvature values (R _{1 }, R _{2 }, and R _{m }), central corneal thickness (CCT), WtW, ACD, LT, and AL both before and after cycloplegia. We completed IOL power calculations with the SRK/T, Holladay 2, and Haigis formulas using a target refraction of −0.50 D. IOLMaster 700 included the User Group for Laser Interference Biometry–optimized IOL constants for each IOL power formula within the device software. A 1-piece AcrySof SA60AT (Alcon Laboratories, Inc., Fort Worth, TX) was used to calculate IOL power with a 118.8 (A) constant for the SRK/T, +5.426 (ACD) constant for Holladay 2, and −0.111, +0.249, and +0.179 (a0, a1, and a2) constants for the Haigis. We evaluated changes in IOL power calculations according to changes in IOL power after cycloplegia, and to pinpoint the effect of cycloplegia, we recorded refractive predictions according to predilation IOL power both before and after cycloplegia.

## Statistical Analysis

We used SPSS software for statistical analysis (version 18.0; SPSS Inc., Chicago, IL), as well as the Kolmogorov–Smirnov test to gauge normality in particular. We evaluated changes before and after dilation with a paired samples *t *test and compared parameters between groups with an unpaired samples *t *test. *P *< .05 was considered statistically significant.

Based on the previous studies with IOLMaster, setting the level of α = 0.05 and β = 80%, a sample size calculation indicates that a minimum of 36 subjects for each group would be required to give an 80% power at 5% 2-sided significance level for the statistical test to detect the difference.

## Results

Tables 1–8 show the results of R _{1 }, R _{2 }, R _{m }, CCT, WtW, ACD, LT, AL, and PD measurements, as well as of IOL power calculations, according to the SRK/T, Holladay 2, and Haigis formulas with and without cycloplegia in both prepresbyopic and presbyopic eyes.

Parameter | Noncycloplegia | Cycloplegia | Mean Difference ± SD | P Value ^{a } | ||
---|---|---|---|---|---|---|

Mean ± SD | Range | Mean ± SD | Range | |||

R _{1 } |
7.89 ± 0.25 | 7.24–8.21 | 7.90 ± 0.25 | 7.25–8.22 | −0.004 ± 0.02 | .52 |

R _{2 } |
7.69 ± 0.35 | 6.83–8.18 | 7.69 ± 0.35 | 6.82–8.17 | −0.001 ± 0.03 | .65 |

R _{m } |
7.79 ± 0.30 | 7.06–8.2 | 7.79 ± 0.30 | 7.04–8.2 | 0.0 ± 0.02 | .89 |

CCT | 0.556 ± 0.046 | 0.502–0.636 | 0.559 ± 0.045 | 0.508–0.635 | −0.003 ± 0.003 | .048 |

WtW | 11.9 ± 0.68 | 10.8–13.2 | 11.9 ± 0.70 | 10.8–13.2 | 0.0 ± 0.13 | 1.0 |

ACD | 3.43 ± 0.39 | 2.78–4.17 | 3.56 ± 0.36 | 2.9–4.23 | −0.13 ± 0.08 | <.001 |

LT | 3.7 ± 0.25 | 3.29–4.13 | 3.61 ± 0.21 | 3.25–4.0 | 0.09 ± 0.05 | <.001 |

AL | 23.25 ± 0.67 | 22.18–23.98 | 23.25 ± 0.67 | 22.17–23.99 | 0.0 ± 0.007 | 1.0 |

PD | 5.91 ± 1.21 | 4.0–7.0 | 7.90 ± 0.73 | 6.2–8.7 | −1.98 ± 1.06 | <.001 |

a *P *value for comparison of parameters in noncycloplegia and cycloplegia states.

Parameter | Noncycloplegia | Cycloplegia | Mean Difference ± SD | P Value ^{a } | ||
---|---|---|---|---|---|---|

Mean ± SD | Range | Mean ± SD | Range | |||

R _{1 } |
7.88 ± 0.32 | 7.36–8.22 | 7.87 ± 0.32 | 7.34–8.20 | 0.01 ± 0.01 | .42 |

R _{2 } |
7.80 ± 0.31 | 7.36–8.17 | 7.79 ± 0.32 | 7.31–8.14 | 0.005 ± 0.02 | .64 |

R _{m } |
7.67 ± 0.22 | 7.36–8.19 | 7.67 ± 0.22 | 7.33–8.17 | 0.0 ± 0.02 | .85 |

CCT | 0.540 ± 0.036 | 0.509–0.612 | 0.541 ± 0.039 | 0.512–0.609 | −0.001 ± 0.005 | .57 |

WtW | 11.8 ± 0.42 | 11.6–12.8 | 11.9 ± 0.39 | 11.7–12.9 | −0.05 ± 0.11 | .02 |

ACD | 3.11 ± 0.29 | 2.94–3.68 | 3.20 ± 0.29 | 3.15–3.85 | −0.08 ± 0.03 | <.001 |

LT | 4.52 ± 0.43 | 4.01–5.02 | 4.50 ± 0.43 | 3.95–5.0 | 0.02 ± 0.01 | <.001 |

AL | 23.09 ± 0.6 | 22.4–24.66 | 23.09 ± 0.6 | 22.4–24.64 | −0.004 ± 0.01 | .15 |

PD | 3.85 ± 0.56 | 3.0–5.5 | 7.07 ± 0.73 | 7.0–8.6 | −3.21 ± 0.91 | <.001 |

a *P *value for comparison of parameters in noncycloplegia and cycloplegia states.

Parameter | Prepresbyopia | Presbyopia | Mean Difference ± SD | P Value ^{a } | ||
---|---|---|---|---|---|---|

Mean ± SD | Range | Mean ± SD | Range | |||

R1 | −0.004 ± 0.02 | −0.05 to 0.04 | 0.01 ± 0.01 | −0.01 to 0.03 | 0.006 ± 0.02 | .27 |

R2 | −0.001 ± 0.03 | −0.04 to 0.05 | 0.005 ± 0.02 | −0.04 to 0.05 | −0.006 ± 0.01 | .32 |

Rm | 0.0 ± 0.02 | −0.04 to 0.02 | 0.0 ± 0.02 | −0.02 to 0.03 | 0.006 ± 0.006 | .36 |

CCT | −0.003 ± 0.003 | −0.006 to 0.003 | −0.001 ± 0.005 | −0.003 to −0.009 | −0.002 ± 0.002 | .38 |

WtW | 0.02 ± 0.09 | −0.2 to 0.4 | −0.05 ± 0.11 | −0.2 to 0.1 | 0.07 ± 0.03 | .29 |

ACD | −0.13 ± 0.08 | −0.05 to −0.35 | −0.08 ± 0.03 | −0.12 to −0.26 | 0.04 ± 0.02 | .03 |

LT | 0.09 ± 0.05 | 0.03–0.24 | 0.02 ± 0.01 | 0.02–0.13 | 0.07 ± 0.01 | <.001 |

AL | 0.0 ± 0.007 | −0.01 to 0.01 | −0.004 ± 0.01 | 0.0–0.02 | 0.004 ± 0.02 | .36 |

PD | −1.98 ± 1.06 | −4.1 to −0.3 | −3.21 ± 0.91 | −2.4 to −5.0 | 1.22 ± 0.32 | <.001 |

a *P *value for comparison of parameters in noncycloplegia and cycloplegia states.

Intraocular Lens Formula | Intraocular Lens Power | Mean Difference ± SD | P Value | |||
---|---|---|---|---|---|---|

Noncycloplegia | Cycloplegia | |||||

Mean ± SD | Range | Mean ± SD | Range | |||

Sanders-Retzlaff-Kraff/theoretical | 22.72 ± 2.16 | 20.0–25.5 | 22.66 ± 2.18 | 20.0–25.5 | 0.05 ± 0.29 | .43 |

Holladay 2 | 21.86 ± 2.41 | 18.5–25.0 | 21.97 ± 2.39 | 18.5–25.0 | −0.11 ± 0.21 | .042 |

Haigis | 21.97 ± 2.36 | 18.0–25.0 | 22.02 ± 2.39 | 18.0–25.0 | −0.05 ± 0.16 | .093 |

Intraocular Lens Formula | Intraocular Lens Power | Mean Difference ± SD | P Value | |||
---|---|---|---|---|---|---|

Noncycloplegia | Cycloplegia | |||||

Mean ± SD | Range | Mean ± SD | Range | |||

Sanders-Retzlaff-Kraff/theoretical | 22.53 ± 1.23 | 20.0–25.0 | 22.53 ± 1.23 | 20.0–25.0 | 0.0 ± 0.19 | 1.0 |

Holladay 2 | 21.72 ± 1.25 | 18.5–24.5 | 21.81 ± 1.2 | 18.5–25.0 | −0.09 ± 0.19 | .072 |

Haigis | 21.38 ± 1.17 | 18.0–25.0 | 21.27 ± 1.24 | 18.0–25.0 | 0.01 ± 0.21 | .66 |

Intraocular Lens Formula | Refractive Predictions (Diopters) ^{a } |
Mean Difference ± SD | P Value ^{b } | |||
---|---|---|---|---|---|---|

Noncycloplegia | Cycloplegia | |||||

Mean ± SD | Range | Mean ± SD | Range | |||

Sanders-Retzlaff-Kraff/theoretical | −0.49 ± 0.1 | −0.37 to −0.65 | −0.5 ± 0.21 | −0.26 to −0.70 | 0.02 ± 0.17 | .68 |

Holladay 2 | −0.5 ± 0.09 | −0.40 to −0.68 | −0.42 ± 0.12 | −0.17 to −0.63 | −0.08 ± 0.01 | .005 |

Haigis | −0.52 ± 0.1 | −0.33 to −0.69 | −0.43 ± 0.1 | −0.23 to −0.63 | −0.09 ± 0.1 | .001 |

a Target refraction was −0.50 D.

b *P *value for comparison of parameters in noncycloplegia and cycloplegia states.

Intraocular Lens Formula | Refractive Predictions (Diopters) ^{a } |
Mean Difference ± SD | P Value ^{b } | |||
---|---|---|---|---|---|---|

Noncycloplegia | Cycloplegia | |||||

Mean ± SD | Range | Mean ± SD | Range | |||

Sanders-Retzlaff-Kraff/theoretical | −0.42 ± 0.25 | −0.30 to −0.54 | −0.49 ± 0.12 | −0.27 to −0.61 | 0.07 ± 0.26 | .13 |

Holladay 2 | −0.44 ± 0.12 | −0.36 to −0.57 | −0.44 ± 0.11 | −0.17 to −0.56 | 0.01 ± 0.11 | .70 |

Haigis | −0.50 ± 0.1 | −0.27 to −0.59 | −0.49 ± 0.14 | −0.24 to −0.57 | −0.01 ± 0.12 | .68 |

a Target refraction was −0.50 D.

b *P *value for comparison of parameters in noncycloplegia and cycloplegia states.