Objective Scatter Index: Working Toward a New Quantification of Cataract?

Abstract

Purpose: 

To investigate the associations between clinical cataract classifications, quality of life (QOL), and the objective loss of ocular transparency in patients presenting with clinical cataracts.

Methods:

In this prospective, multicenter, cross-sectional study, 1,768 eyes of 1,768 patients (mean age: 72.5 years; range: 28 to 93 years) referred for cataract assessment were enrolled. Visual acuity was measured before slit-lamp examination to determine the severity of lens opacification using the Lens Opacities Classification System III. Patients were asked to complete the Visual Function Index (VF-14) questionnaire. Ocular transparency was quantified by ObjectiveScatter Index (OSI) and was measured by the HD Analyzer (Visiometrics SL, Terrassa, Spain). Association and categorical data analysis were performed between each measured parameter alongside cross-tabulation analyses to determine sensitivity and efficiency of the HD Analyzer.

Results:

High OSI levels corresponded slightly with a lower visual acuity value and corresponded better with lower VF-14 scores. OSI scores were strongly associated with cataract classification and severity. Cross-tabulation analysis revealed a high sensitivity and efficiency index for the OSI with these clinically validated parameters illustrating good agreement overall for theOSI in determining cataract.

Conclusions:

The OSI measured by the HD Analyzer is a sensitive and efficient tool to be considered in the early detection of cataract in patients.

[J Refract Surg. 2016;32(2):96-102.]

Cataracts, a consequence of lenticular opacification,lead to degradation of qualitative vision, loss of con-trast sensitivity, and visual disturbances such asglare. Postoperative outcomes after intraocular lens implanta-tion have markedly improved in the past decade. This has ledto a high reduction in postoperative spectacle dependency.The major factors contributing to this advancement have beenthe vast improvements in intraocular lens design and phaco-emulsification techniques and early detection techniques.

Cataract assessment is usually based on visual acuity andlifestyle impact. More recently, ophthalmologists have beenlooking toward quantifying the severity of lenticular opacifi-cation to determine lifestyle and visual function impact. TheHD Analyzer (Visiometrics SL, Terrassa, Spain) is the onlysystem that objectively measures visual quality and scatter inthe eye. Based on the double-pass technique, the HD Analyzerrecords the retinal image of a point source of light obtainedafter centration of an infrared signal.1-3 The double-pass tech-nique has been reported as being useful to examine forward-scattered light, which causes degradation of retinal images ineyes with cataract.4 The HD Analyzer determines the ObjectiveScattering Index (OSI) using a point spread function, whichdetermines how a point source of light is imaged on the retina.A high OSI score indicates degradation of the quality of the pa-tient’s vision. Moreover, the retinal image corresponds to thedouble-pass point spread function, enabling calculation of themaximum theoretical visual acuity and determination of themodulation transfer function. This has been used previouslyto understand retinal images as a function of contact lens use,age, or implantation of intraocular lenses.5-8

From the Department of Ophthalmology, Centre Hospitalier Universitaire, Brest,France (FG, BC); and Medeuronet UK Ltd., London, United Kingdom (SRP).

Submitted: August 4, 2015; Accepted: November 20, 2015

The authors have no financial or proprietary interest in the materials pre-sented herein.
Drs. Galliot and Cochener contributed equally to this work and should be considered as equal first authors.

Correspondence: Béatrice Cochener, MD, PhD, Department of Ophthalmology, Centre Hospitalier Universitaire, Brest, France. E-mail: beatrice.cochener@ophtalmologie-chu29.fr

doi:10.3928/1081597X-20151222-02​

Some of our concurrent work has also investigated correlations between the OSI and functional markers. This work has furthered existing reports in the literature by determining the use of the OSI against visual acuity and the Visual Function Index-14 (VF-14) questionnaire. This study investigated associations and specifically the sensitivity and specificity to complement these results by understanding the use of the OSI better.

In the current study of a large population of patients, the correlation between the clinical classification of cataract via slit-lamp examination (Lens Opacities Classification System [LOCS] III), the subjective quality of vision assessed by the VF-14, the visual acuity, and the objective loss of ocular transparency as measured by the HD Analyzer were investigated to determine any clinically and statistically significant relationships.

Patients and Methods

Study Population

This was a prospective, multicenter (10-center) study. All participants were made aware of the nature of the study and informed consent was sought before enrollment. The tenets of the Declaration of Helsinki were adhered to throughout the study. The study was approved by the local institutional review board.

The main inclusion criteria were a positive diagnosis of cataract (early to mature) and a competent comprehension of the French language. Exclusion criteria included a previous history of retinal and ocular pathology, ocular surface disease, and previous surgical history other than lens opacities, cataract suspicion, or both.

The visual acuity was recorded for all participants and slit-lamp examination performed to note the grade and severity of cataract using the LOCS III.9 All cataracts were classified and graded in four groups: nuclear opacity, nuclear color, cortical opacity, and posterior subcapsular opacity. The severity was graded from 1 (early cataract) to 6 (mature cataract) for the nuclear color and nuclear opacity and 1 to 5 for the cortical and posterior subcapsular opacities.

VF-14

Participants were also asked to complete the VF-14 questionnaire. The VF-14 is a brief questionnaire designed to measure visual function impairment in patients with cataract. It consists of 18 questions covering 14 aspects of visual function affected by cataracts. The VF-14 shows high internal consistency and is a reliable, valid instrument providing information not conveyed by visual acuity or general health status measures.10-12 A score of 100 indicates that the patient is able to do all applicable activities, whereas a score of 0 indicates that the patient is unable to do all applicable activities because of vision quality.

HD Analyzer

OSI measurements were determined for each eye. The HD Analyzer measures the combined effect of higher order aberrations and scattered light using a double-pass system, thus providing information on the optical quality of the eye. A point source image is formed on the patient’s retina using a 780-nm laser diode. Double-pass images of every eye were acquired in focus, corrected internally by an optometer that ranges from -8.00 to +6.00 diopters. Any refractive astigmatism was corrected by placing the appropriate cylindrical lens in front of the eye. Pupil alignment was monitored with an additional camera.

A charge-coupled device camera records the doublepass images from the retina and beam splitter. A personal computer was used to process the retinal images and to collect the data. The double-pass instrument provides several measurements, including the OSI, the modulation transfer function, and the maximum visual acuity predicted for objects with 100%, 50%, and 9% contrast. The point spread function recorded by the asymmetric double-pass system represents the image projected onto the retina from a point light source. The point spread function was calculated as the mean of six individual acquisitions. In the absence of optical aberrations or light scattering, the point spread function is quasi point and the image perceived is unaltered (if the diffraction eventually caused by a large pupil diameter is moderate).

The OSI is an objective evaluation of the scattering degree caused by the loss of transparency of one or more of the ocular structures, such as corneal opacities or cataract. This index is defined as the ratio between the integrated light in the periphery (between 12 and 20 minutes of arc) and a circular area of 1 minute of arc around the central peak of the double-pass image. The higher the OSI value, the higher the level of intraocular scattering. The resulting figure is proportional to the light diffusion rates. Normal values are less than 0.5 for a young person with a healthy eye,13 between 1.45 and 4 for an early-stage cataract, and greater than 4 for mature cataract. The OSI parameter can be affected by uncorrected refractive errors (defocus and astigmatism). Hence, all patients had a refractive examination and all measurements were performed with best-corrected sphere and cylinder to avoid these artifacts.

Statitical Analysis

To explore the OSI as an indicator for defining cataracts and against visual acuity and quality of life, associations were explored using 2 × 2 tables and categorical data analysis (sensitivity, specificity, and efficiency indices)against defined cut-off points (Table 1). Once the cutoffpoints for each variable were determined, populationtables were constructed to give an account for how manyeyes existed in each category. These were then crossclarifiedagainst each other to provide 2 × 2 tables thatwere used for categorical data analysis to determine thesensitivity (ie, proportion of correct clinical disease outcomes[or positives] identified by a measurement test [ie,OSI] or questionnaire), specificity (proportion of correctnon-disease outcomes identified), and efficiency (overallproportion of correct diagnoses identified by a test or questionnaire) of the variable in determining cataract. Abinomial test of proportions was used to examine signifi-cance. This is more accurate and powerful than the chisquaretest. Fisher’s exact test was also used, althoughit was unnecessary because numbers in the contingencytables considerably exceeded 50.

RESULTS

Baseline Characteristics

We included 1,768 eyes of 1,768 patients in thisstudy. As illustrated in Table 2, the mean age of thesample population was 72.5 years (range: 28 to 93years). The average LOCS III scores for the populationwere posterior subcapsular 0.87 (range: 0 to 6), nuclearopalescence/nuclear color 5.53 (range: 0 to 12), andcortical cataract 1.04 (range: 0 to 5).

Association Analysis

The study looked at the number of eyes within thestudy population against the determined cut-off values indicated in Table 1.

Table 3 outlines the population of eyes when lookingat the OSI against the visual acuity, VF-14, and cataractgrading cut-off points. It is evident that there was a goodassociation, with a reasonable number of eyes (n = 876)from the total population, between high OSI (better than1.45) and lower visual acuity value (0.7 or worse). A highOSI value also corresponded well to low VF-14 scores (n= 624 eyes). There were also good associations betweenthe number of populated eyes between each cataractgrade (posterior subcapsular opacity, nuclear opalescence/nuclearcolor, and cortical cataract) and LOCS IIIwith an OSI value greater than 1.45 (n = 842 eyes).

Table 4 outlines the number of eyes when investigatingVF-14 values against the intended cut-off points. As compared to the OSI, there was a stronger association between visual acuity and VF-14 (n = 876 vs n = 1,044 eyes). However, a weaker association was observed between posterior subcapsular (n = 57 eyes) and cortical cataracts (n = 62 eyes) with VF-14 scores.

Table 5 outlines the number of eyes from the total population when using visual acuity against cataract severity. The analysis revealed better associations than for OSI and VF-14, between cortical and posterior subcapsular cataracts with visual acuity values (n= 671 and 517, respectively). Moreover, nuclear color was more associated with visual acuity than other cataract types (n = 1,299 eyes).

Pivot Table Analyses

We analyzed and compared three methods of cataractscreening: the HD Analyzer assessment, the VF-14questionnaire, and visual acuity measurement. We calculatedfor these three tests their sensitivity, specificity,and efficacy (Table 6) to compare the HD Analyzerand VF-14 examinations with visual acuity (worse than0.5), which is the reference test for cataract detection.

OSI Measurement

The OSI is a sensitive examination for cataract detection,as found by the calculated sensitivity scores of 96.5%for cortical cataracts, 93.6% for nuclear cataracts, and100% for posterior subcapsular cataracts for an overallsensitivity of 96% for all types of cataracts. The OSI hasa specificity of 76.7% for the three cataract subtypes combined and is effective (89.9% for cortical cataracts, 89.4% for posterior subcapsular, and 80.4% for nuclear cataracts).

VF-14 Quetionnaire

The VF-14 questionnaire was slightly less sensitive than the OSI examination. We calculated a sensitivity of 65.3% for cortical cataracts, 61.3% for nuclear cataracts, and 78.1% for posterior subcapsular cataracts. The VF-14 had a specificity of 68% for all types of cataracts combined. Its efficiency was 66.2% for cortical cataracts, 61.3% for nuclear cataracts, and 74.1% for posterior subcapsular cataracts.

Visual Acuity

We voluntarily tested two visual acuity thresholds: 0.5 or worse (reference) and 0.7 or worse. Visual acuity of 0.5 or worse was a poorly sensitive test (34% for cortical cataracts, 43% for nuclear cataracts, and 46% for posterior subcapsular cataracts), with a specificity of 93%, effective in only 54% for cortical cataracts, 46% for nuclear cataracts, and 47% for posterior subcapsular cataracts. Visual acuity of 0.7 or worse had signifi- cantly better sensitivity with 88% for cortical cataracts, 90% for nuclear cataracts, and 88% for posterior subcapsular cataracts. The specificity was the same (92%) and the efficiency was 89% for cortical and posterior subcapsular cataracts and 90% for nuclear cataracts.

General Analysis

Of patients with cataract who had a visual acuity of 0.7 or worse, 51.8% presented a subjective change in their quality of vision (VF-14 ≥ 84). We also observed that 83% of patients with cataract (LOCS III ≥ 3) with pathological light scattering (OSI > 1.45) reported impaired functioning in the questionnaire. Of patients with cataract who had a visual acuity of 0.7 or worse, 81.2% had increased ocular light scattering (OSI > 1.45).

Discussion

We observed a mean OSI score of 3.7 for early-stage cataract, 4.8 for mild cataract, 6.2 for moderate cataract, and 10.2 for white cataract. The OSI score therefore increases with cataract severity, as previously shown in the literature. Artal et al.7 observed that an OSI score of less than 1 corresponds to the absence of cataract, whereas a score of 2 is correlated to early-stage cataract and a score of greater than 2 corresponds to a moderate to mature cataract. Our study also shows that even early-stage cataract causes an alteration of the ocular diffusion, and therefore a deterioration in the patient’s quality of vision (OSI average of 3.7 in early-stage cataract).

Previous studies have demonstrated the advantage of using a double-pass system, and in particular the ability to test under different contrast conditions. Furthermore, the system demonstrates good intraobserver reproducibility.14 Another study demonstrated good correlation between OSI and cataract grading15 and the current study has further established this by investigating associations. Pujol et al.16 also conducted a preliminary study on the use of OSI with subjective measures such as visual acuity and slit-lamp examination and the current study extends that work by additionally investigating VF-14. The obvious disadvantage to using the widely used LOCS system is that it is somewhat subjective, and therefore the use of imaging techniques such as the HD Analyzer or the Pentacam allow more quantitative evaluation of lens opacification.

Preliminary analysis in this study showed significant associations between the OSI, VF-14, and clinical outcomes. Further in-depth analysis estimated the sensitivity, specificity, and efficiency indices of both parameters. These demonstrated the superiority of the OSI over the VF-14, both clinically and statistically (P < .001).

We conducted our analysis on three cataract screening tests: visual acuity (gold standard), light scattering (OSI score), and functional impairment with the VF-14 questionnaire. These three tests all demonstrated effectiveness in the detection of cataract as already demonstrated in the literature.17,18 Ours is so far the only study to have evaluated the sensitivity and specificity of these tests. Furthermore, the efficiency index is more relevant than a linear correlation coefficient alone, because the variables are not continuous sets of data that assume a linear relationship (ie, a change of one point in a questionnaire score does not necessarily lead consistently to a unit change in OSI or visual acuity).

OSI

The OSI was found to be a sensitive test for cataract detection. Indeed, we observed a sensitivity of 96%, which means that there is a 96% probability of having an OSI score of greater than 1.45 if the patient has a cataract (LOCS III score ≥ 3).

Our study also showed that the OSI is a specific test. The specificity of the OSI score according to each type of cataract (cortical, nuclear, and posterior subcapsular) was calculated at 76.74%. However, even minimal lens clouding (LOCS III < 3) may induce an increase in light scattering. The positive likelihood ratio was approximately 4 for each type of cataract; this measurement represents an added value to the diagnosis and the OSI is a good tool for the detection of early cataract.

VF-14

The results of this study suggest that the VF-14 is a less sensitive tool than the OSI for cataract screening. Indeed, the sensitivity of the VF-14 questionnaire is 65.95% for the total LOCS III score and 61.29%,65.95%, and 78.08% for nuclear, cortical, and posterior subcapsular cataracts, respectively. The VF-14was found to be 68% specific for cortical, nuclear, and posterior subcapsular cataracts.

This questionnaire comprises a subjectivity bias. Indeed,elderly patients often minimize their functional signs and become accustomed to the degradation of their quality of vision generated by the cataract. TheVF-14 questionnaire is therefore useful to quantify functional impairment in patients, but contains a subjectivity bias that does not allow the test to be as sensitive as an objective examination.

The VF-14 questionnaire is useful for the diagnosis of early cataract in young adults but contributes weakly to the diagnosis of advanced cataracts in the elderly.However, it can quantify the subjective impairment experienced by patients in their daily lives.

Visual Acuity

We selected two visual acuity thresholds: 0.5 or worse (current definition of cataract according to theFrench Medicines Agency) and 0.7 or worse. We observed that the test using a visual acuity of 0.5 or worse was less sensitive (30% to 40%) but as specific (> 90%)compared to a visual acuity cut-off of 0.7 or worse(sensitivity = 90%).

The effectiveness of the screening test using a visual acuity of 0.7 or worse was two times more efficient than the reference test. Therefore, the strong sensitivity scores for OSI with visual acuity versus VF-14 that contribute to good efficiency scores (Table 7) in determining cataract in this large cohort suggest that this measure is good at detecting lenticular opacification. However, it is important to note that this study is applying associations and categorical data analysis to determine the efficiency of the OSI in identifying cataract. Whether this overall agreement is due to cause and effect is yet to be determined.Because the double-pass system measures within a small visual angle, OSI values can be highly dependent on capsule remnants or dense posterior capsule opacification in the periphery, therefore presenting a shortcoming, especially in those with large pupils. This may explain why some associations, especially with visual acuity, were weaker than expected. Although it would be expected that light scatter would be a better determinant,this is also consistent with previous reports.14 The issue raised by Ginis et al., investigating the optical intraocular light scatter using an infrared system, also highlights a limitation of the system.19

Nonetheless, this study does highlight particular uses for the OSI as a good indicator for cataract alongside visual acuity. In particular, the approach using pivot analyses shows good sensitivity and efficiency scores with visual acuity. This goes further than previous studies in indicating the potential clinical application of the OSI in defining and determining cataract.The sensitivity issues with the validated VF-14 may be due to the subjective nature of questionnaires and the influence of human error.

The results presented in Table 7 are important.There were 65 patients with an OSI score of 1.45 orgreater, and total LOCS score of 3 or greater, and yetthey had good vision (visual acuity better than 0.7).This is because a patient can have a cataract and yetstill maintain good vision as judged by the usual visualacuity measurement. The OSI detects the cataract, butthis does not always imply poor vision and vice versa.This may account for its poor specificity with regard tovisual acuity. Therefore, a suggestion for future workmight be to use different OSI critical points for differentvariable outcomes.

The authors also recognize that the sample size isa limitation of this study (ie, different eyes presenting with different types and severity of cataracts) because the type of analysis conducted would be much more robust with a larger sample of cataract severities. However, it is important to note that there are limited data available on using a double-pass system with only a small cohort of patients. This study sets a precedent to elaborate on in larger multi-site studies to investigate the clinical application and association of the OSI with the incidence of cataract. Although the study shows strong associations, whether there are correlations between these measures remains to be seen.

Conclusion

The OSI assessment was the most sensitive and most specific test for the detection of cataract, even at an early stage. A visual acuity threshold of 0.7 or worse appears to be the most sensitive and the most effective. This suggests that the associated subjective change in quality of vision is experienced early on in cataract development, despite having visual acuity excellent for driving.

Even minor cataract can cause an increase in light scattering with a pathological OSI score, thus significantly affecting a patient’s quality of life. In our study, we found that 81.2% of patients with cataract who had a visual acuity of 0.7 or worse had increased light scattering; more than half of these patients reported an impaired quality of life in the questionnaire. Thus, the OSI could be used as an additional surrogate measure of validating patient complaints, especially in pay-model ophthalmology practice.

Overall, these initial results suggest that the OSI can provide clinicians with an overall clue and determination of lenticular changes as suggested by the high sensitivity (96% for the LOCS) and efficiency (89%) as compared to visual acuity alone. These results suggest a rationale for using the OSI and VF-14 to quantify and define visual degradation with early lenticular changes.

New clinical investigations such as the OSI allow the objective evaluation of the quality of vision in patients and should, in the future, be able to assist the clinician in the detection of early cataract. To date, the distinction between true cataract and clear lens remains at the sole discretion of the surgeon. New instruments to measure the quality of vision both subjectively (VF-14 questionnaire) and objectively (HD Analyzer) should, in the future, be part of the recommendations for good surgical practice in cataract management.

Author Contributions

Study concept and design (FG, BC); data collection (FG, BC); analysis and interpretation of data (FG, SRP, BC); writing the manuscript (FG, SRP, BC); critical revision of the manuscript (SRP); administrative, technical, or material support (SRP, BC); supervision (BC)

References

1. Westheimer G, Liang J. Evaluating diffusion of light in the eye by objective means. Invest Ophthalmol Vis Sci. 1994;35:2652-2657.

2. Güell JL, Pujol J, Arjona M, Diaz-Douton F, Artal P. Optical Quality Analysis System: instrument for objective clinical evaluation of ocular optical quality. J Cataract Refract Surg. 2004;30:1598-1599.

3. Saad A, Saab M, Gatinel D. Repeatability of measurements with a double-pass system. J Cataract Refract Surg. 2010;36:28-33.

4. Nam J, Thibos LN, Bradley A, Himebaugh N, Liu H. Forward light scatter analysis of the eye in a spatially-resolved doublepass optical system. Opt Express. 2011;19:7417-7438.

5. Guirao A, Gonzalez C, Redondo M, Geraghty E, Norrby S, Artal P. Average optical performance of the human eye as a function of age in a normal population. Invest Ophthalmol Vis Sci. 1999;40:203-213.

6. Guirao A, Redondo M, Geraghty E, Piers P, Norrby S, Artal P. Corneal optical aberrations and retinal image quality in patients in whom monofocal intraocular lenses were implanted. Arch Ophthalmol. 2002;120:1143-1151.

7. Artal P, Benito A, Pérez GM, et al. An objective scatter index based on double-pass retinal images of a point source to classify cataracts. PLoS One. 2011;6:e16823.

8. Pujol J, Gispets J, Arjona M. Optical performance in eyes wearing two multifocal contact lens designs. Ophthalmic Physiol Opt. 2003;23:347-360.

9. Chylack LT Jr, Wolfe JK, Singer DM, et al. The Lens Opacities Classification System III. The Longitudinal Study of Cataract Study Group. Arch Ophthalmol. 1993;111:831-836.

10. Steinberg EP, Tielsch JM, Schein OD, et al. The VF-14. An index of functional impairment in patients with cataract. Arch Ophthalmol. 1994;112:630-638.

11. Cassard SD, Patrick DL, Damiano AM, et al. Reproducibility and responsiveness of the VF-14: an index of functional impairment in patients with cataracts. Arch Ophthalmol. 1995;113:1508-1513.

12. Alonso J, Espallargues M, Andersen TF, et al. International applicability of the VF-14. An index of visual function in patients with cataracts. Ophthalmology. 1997;104:799-807.

13. Martínez-Roda JA, Vilaseca M, Ondategui JC, et al. Optical quality and intraocular scattering in a healthy young population. Clin Exp Optom. 2011;94:223-229.

14. Hirnschall N, Crnej A, Gangwani V, Findl O. Comparison of methods to quantify posterior capsule opacification using forward and backward light scattering. J Cataract Refract Surg. 2014;40:728-735.

15. Lim SA, Hwang J, Hwang KY, Chung SH. Objective assessment of nuclear cataract: comparison of double-pass and Scheimp- flug systems. J Cataract Refract Surg. 2014;40:716-721.

16. Pujol J, Vilaseca M, Salvadó A, et al. Cataract evaluation with an objective scattering index based on double-pass image analysis. Invest Ophthalmol Vis Sci. 2009;50:E-Abstract 6127.

17. Cabot F, Saad A, McAlinden C, Haddad NM, Grise-Dulac A, Gatinel D. Objective assessment of crystalline lens opacity level by measuring ocular light scattering with a double-pass system. Am J Ophthalmol. 2013;155:629-635.

18. Ortiz D, Alió JL, Ruiz-Colechá J, Oser U. Grading nuclear cataract opacity by densitometry and objective optical analysis. J Cataract Refract Surg. 2008;34:1345-1352.

19. Ginis H, Sahin O, Pennos A, Artal, P. Compact optical integration instrument to measure intraocular straylight. Biomedical Optics Express. 2014;5:3036-3041.

From the Department of Ophthalmology, Centre Hospitalier Universitaire, Brest,France (FG, BC); and Medeuronet UK Ltd., London, United Kingdom (SRP).

Submitted: August 4, 2015; Accepted: November 20, 2015

The authors have no financial or proprietary interest in the materials pre-sented herein.
Drs. Galliot and Cochener contributed equally to this work and should be considered as equal first authors.

Correspondence: Béatrice Cochener, MD, PhD, Department of Ophthalmology, Centre Hospitalier Universitaire, Brest, France. E-mail: beatrice.cochener@ophtalmologie-chu29.fr

doi:10.3928/1081597X-20151222-02​

About the Author