Eye drops instillation just before biometry in candidates to cataract surgery may influence the planning of future toric IOL position.

Abstract

At the time of planning surgery, keratometry (K) measurements determine whether a patient may require a toric intraocular lens (t-IOL) and, if needed, the optimal axis for t-IOL orientation in the capsular bag. However, measures of anterior corneal curvatures could be influenced by the status of the tear film, as suggested by the influence of osmolarity (Epitropoulos et al. 2015) and the impact of the viscosity of artificial tears (Röggla et al. 2021). Since ocular biometry is usually performed immediately after the pre-operative clinical examination, eye drops used for pupil dilation (fundus examination) and/or ocular surface checking may temporarily modify the biometric measurements, with potential consequence for t-IOL planning. We evaluated whether instillation of eye drops immediately before performing biometric measurements can modify the planning of intraocular lens (IOL) in cataract candidates. The biometric measures (mean K, corneal cylinder, cylinder axis and power of the emmetropic IOL) of 30 right eyes of 30 consecutive candidates to cataract surgery (mean age ± SD: 71.7 ± 7.5 years), with no dry eye syndrome, were retrospectively analysed (ethics approval by IRB 00008855, Société Française d’Ophtalmologie IRB#1). All included patients had been tested twice for ocular biometry (AL-Scan optical biometer, V1.11.01, NIDEK Co. Ltd., Gamagori, Japan): the first at arrival before instillation of eye drops (before slit lamp examination) when being taken in charge by the resident (M.A.), and the second at the end of the journey, after senior’s confirmation of surgical decision (M.L.), i.e. just after the last 0.9% saline eye drop (fluorescein washout). Differences in power and axis of cylinder were evaluated by vectorial analysis (Holladay et al. 1998; Naeser & Hjortdal 1999). We also evaluated the axis of cylinder separately, notwithstanding the power of the cylinder as in the specific case of t-IOL planning, its future orientation is mostly based on the anterior astigmatism axis. For these eyes characterized by a mean K at 44.17 ± 1.67 D, an anterior astigmatism at 1.03 ± 0.71 D (<0.75D: 40.0%; 0.75D–1.50D: 36.7%; >1.50D: 23.3%), and a mean emmetropic IOL power at 21.48 ± 1.76 D, saline instillation did not statistically modify the mean K (variation of − 0.03 ± 0.08 D for before vs. after, p = 0.43), nor for the emmetropic IOL power (0.03 ± 0.09 D, p = 0.51). Vectorial analysis of the change of astigmatism induced by eye drops instillation indicated a mean absolute magnitude of 0.46 ± 0.28 D (SD). No significant variation in cylinder power was observed (0.07 ± 0.15 D, p = 0.34). However, the absolute value of rotation of the steep axis analysed as an isolated variable, indicated an average difference of 10.1 ± 3.2° according to saline instillation, with less than 5° in 43.3% of eyes but, in the opposite, more than 10° in 50% of eyes, and even more than 20° in 10.0% of eyes (see distribution in Figure 1). Subset analysis of eyes with corneal astigmatism above 1.50 D (n = 7) versus below or equal to 1.50 D (n = 23) showed no significant difference between groups for changes of mean K (p = 0.57), cylinder power (p = 0.11) or emmetropic IOL power (p = 0.71), but axis changes were significantly less important among eyes with astigmatism > 1.5 D (p = 0.01) (full details are given in Table 1). Even if changes seem to be less important in eyes with the higher anterior astigmatism, these outcomes are clinically important given the observation from previous studies that a 10° misalignment of a t-IOL results in a 30% loss of the expected toric correction in the targeted axis and that an error of more than 30° nullifies the cylinder correction in that specific axis (Jin et al. 2010), likely resulting in a fuzzier perception than if no toric IOL had been used. In that context, our results indicate that half of t-IOL implantations may be under-optimized due to the variability in anterior astigmatism measurements as a function of the ocular surface conditions. One limitation of the present study, due to medical records reviewed retrospectively, is the absence of analysis of intra-individual variability in the axis of astigmatism (without saline drop). However, previous studies already suggested that modern optical biometer devices such as the AL-Scan provide precise and highly reproducible biometry measurements (Kola et al. 2014). This study strongly suggests that any diagnostic eye drop (mydriatic, fluorescein, saline…) used in every day clinical practice at the time of cataract surgery planning may influence the choice of t-IOL position, according to the delay since the last eye drop instillation.