Purpose
To explore the changes in axial length and refractive error after noninvasive normalization of intraocular pressure (IOP) from elevated levels.
Design
A prospective observational study.
Methods
We enrolled 51 consecutive patients with abnormally elevated unilateral IOP (≥10 mm Hg compared with that of the fellow eye, in which the IOP was ≤21 mm Hg). In all patients, the keratometric value and axial length were obtained with the aid of an IOLMaster before and after IOP normalization (defined as attainment of an IOP difference of ≤3 mm Hg compared with the fellow eye, with or without topical application of ocular hypotensive therapy). We focused principally on IOP, axial length, the keratometric value, and the predicted refractive difference (the predicted refractive error after IOP normalization upon placement of an IOL with a power for emmetropia correction determined prior to IOP normalization).
Results
The axial length was significantly reduced from 23.5 to 23.3 mm after IOP normalization, from 45.9 mm Hg to 14.3 mm Hg ( P < .001). The change in IOP correlated with that of the axial length (r = 0.826, P < .001), but not with the change in the keratometric value ( P = .618). The change in axial length per 10 mm Hg IOP decrease was −0.06 mm ( P < .001). The IOP change was correlated with the predicted refractive difference (r = 0.693, P < .001); the predicted refractive difference per 10 mm Hg IOP decrease was +0.15 diopter ( P < .001).
Conclusions
The axial length decreased and the predicted refractive difference increased (hyperopia) as IOP decreased. Therefore, a possible risk of postoperative hyperopic shift should be considered when biometric examination for IOL power calculation is performed in a patient with an abnormally elevated IOP.
Cataract surgery by phacoemulsification and posterior chamber intraocular lens (IOL) implantation is indicated not only for patients with cataracts but also for patients with angle-closure glaucoma with uncontrolled intraocular pressure (IOP). Cataract surgery is sometimes combined with trabeculectomy or minimally invasive glaucoma surgery, such as a trabecular micro-bypass stent (iStent) and trabeculotomy ab interno (trabectome), in patients with concurrent glaucoma and cataracts for medically uncontrolled IOP. Additionally, cataract surgery is now frequently combined with pars plana vitrectomy in patients with various vitreoretinal diseases often associated with preoperative IOP elevation caused by conditions such as uveitis, hyphema (ghost cell glaucoma), and/or angle neovascularization (neovascular glaucoma).
To meet postoperative expectations of improved vision after cataract surgery (alone or combined with other surgery), one of the most important considerations is accurate prediction of the postoperative refractive error. This error is calculated using an IOL power formula incorporating preoperatively measured biometric data, including the axial length and keratometric value. These data are key when choosing the lens power.
Many studies have verified that axial length shortening is associated with a fall in IOP following trabeculectomy, glaucoma implant surgery, cataract surgery, and combined cataract surgery/vitrectomy. However, in such situations, it is impossible to exclude potential dimensional changes in the eyeball caused by surgery; medical treatment does not trigger such changes. Therefore, we suggest that information on the relationship between IOP and axial length following noninvasive normalization of IOP in patients with abnormally elevated IOP may aid the selection of IOLs of powers affording optimal refractive outcomes after cataract or combined cataract surgery.
The primary purpose of our study was thus to explore the relationships between changes in ocular biometric variables and IOP changes following IOP normalization. Our secondary aim was to evaluate the effect of an abnormally elevated IOP on predicted refractive error.
Methods
This prospective observational study was approved by the Institutional Review Board of Chungnam National University Hospital and conducted in accordance with all relevant requirements of the Declaration of Helsinki. Patients with elevated unilateral IOP (≥10 mm Hg compared with the fellow eye, which had an IOP in the normal range of ≤21 mm Hg in the absence of any ocular hypotensive therapy) were consecutively enrolled, regardless of the cause of IOP elevation, in our Department of Ophthalmology from July 1, 2013 to December 31, 2014. Written informed consent was obtained from each patient prior to participation. The exclusion criteria were best-corrected visual acuity of <0.3 (the decimal measure), macular disease, corneoscleral disease, connective tissue disease, a history of corneal or scleral surgery, and a history of any other ocular surgery within the prior 6 months in either eye.
Both eyes were examined in terms of autorefraction and best-corrected visual acuity and subjected to slit-lamp biomicroscopy and dilated fundus examination. Partial coherence interferometry (IOLMaster; Carl Zeiss Meditec, Dublin, California, USA) was used to obtain keratometric values and axial lengths. IOPs were measured via Goldmann applanation tonometry. After IOP normalization, all examinations were repeated.
To decrease IOP, we prescribed topical ocular hypotensive eye drops, miotics, systemic therapy (a carbonic anhydrase inhibitor and/or a hyperosmotic agent), and/or laser iridotomy according to individual patient needs and ocular conditions (diagnoses). IOP normalization was defined as no further need for systemic therapy with or without topical ocular hypotensive eye drops except miotics and an IOP difference of ≤3 mm Hg compared with the fellow eye.
All IOLMaster examinations were performed by a single experienced operator. The IOLMaster is currently the gold standard for IOL power calculation; axial length measurement errors are much smaller than those of contact-type ultrasound biometry because the IOLMaster is a noncontact technique. When the IOP was abnormally elevated, calculation of the IOL power for emmetropia was achieved using the SRK/T formula with an A-constant of 119.3. The IOLMaster printouts provided these values. We next calculated each predicted refractive difference, defined as the predicted refractive error after IOP normalization when an IOL of a power determined before IOP normalization was placed (ie, when the IOP was abnormally elevated). All statistical analyses were performed using SPSS version 18.0 (SPSS Inc, Chicago, Illinois, USA). Fellow eyes lacking IOP elevation served as controls. Comparisons of ocular variables before and after IOP normalization in the same eye and between eyes (ie, subject eyes with IOP elevations and fellow [control] eyes) before and after IOP normalization were analyzed using the paired t test. Correlation and linear regression analyses were used to define clinical variables associated with IOP change, biometric values, and the predicted refractive difference.
Results
Fifty-one patients (18 female, 33 male; mean age, 59.8 ± 18.0 years) with unilaterally elevated IOP were consecutively enrolled. Twenty eyes (39.2%) had open-angle glaucoma, 16 (31.4%) had angle-closure glaucoma, 8 (15.7%) had neovascular glaucoma, 4 (7.8%) had uveitis, and 3 (6.0%) had traumatic microhyphema ( Table 1 ).
| Age (y) | 59.8 ± 18.0 |
| Male sex | 33 (64.7) |
| Right laterality | 18 (35.3) |
| Diagnosis | |
| Open-angle glaucoma | 20 (39.2) |
| Angle-closure glaucoma | 16 (31.4) |
| Neovascular glaucoma | 8 (15.7) |
| Uveitis | 4 (7.8) |
| Traumatic microhyphema | 3 (6.0) |
Significant differences in mean IOP and axial length were evident between the subject and fellow eyes (45.9 vs 15.6 mm Hg and 23.5 vs 23.3 mm, respectively; P < .001 for both). These differences disappeared after IOP normalization (14.3 vs 15.3 mm Hg and 23.3 vs 23.3 mm; P = .090 and P = .197, respectively). The best-corrected visual acuity of subject eyes was 0.63 (decimal measure; 0.2 logMAR) before IOP normalization and 0.8 (0.2 logMAR) after normalization; this difference was not statistically significant ( P = .168). IOP in the subject eye decreased significantly after IOP-lowering therapy from 45.9 to 14.3 mm Hg ( P < .001), and the axial length became significantly shorter when the IOP was normalized (23.5 to 23.3 mm, P < .001). The spherical equivalent significantly increased (hyperopic shift) after IOP normalization; the change was 0.4 diopter (D) ( P = .031). The anterior chamber depth and keratometric value did not change significantly (3.2 vs 3.1 mm and 44.5 vs 44.3 D, respectively; P = .235 and P = .186, respectively). IOP normalization of the subject eye did not affect the best-corrected visual acuity, IOP, axial length, anterior chamber depth, keratometric value, or spherical equivalent in the fellow eye. The differences in all parameters, except best-corrected visual acuity and anterior chamber depth, before and after IOP normalization of subject eyes were significantly higher than those of fellow eyes (all P values <.05). The predicted refractive difference of subject eyes was 0.5 D, thus significantly more hyperopic than that of fellow eyes (0.0 D, P < .001) ( Table 2 ). Additionally, we calculated another predicted refractive error in the subject’s eye after IOP normalization, when an IOL with a calculated power based on axial length of the fellow eye measured before IOP normalization was inserted. As a result, this predicted refractive error was 0.1 D ± 0.1 D, which was significantly smaller than the above predicted refractive difference of 0.5 D ( P < .001).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
