Evaluation of corneal thickness and topography in normal ... ?· Evaluation of corneal thickness and…

Download Evaluation of corneal thickness and topography in normal ... ?· Evaluation of corneal thickness and…

Post on 28-Jul-2018




0 download


  • Evaluation of corneal thickness and topography innormal eyes using the Orbscan cornealtopography system

    Zuguo Liu, Andrew J Huang, Stephen C Pflugfelder

    AbstractAimsTo map the thickness, elevation(anterior and posterior corneal surface),and axial curvature of the cornea innormal eyes with the Orbscan cornealtopography system.Methods94 eyes of 51 normal subjectswere investigated using the Orbscan cor-neal topography system. The anterior andposterior corneal elevation maps wereclassified into regular ridge, irregularridge, incomplete ridge, island, and un-classified patterns, and the axial powermaps were grouped into round, oval, sym-metric bow tie, asymmetric bow tie, andirregular patterns. The pachymetry pat-terns were designated as round, oval,decentred round, and decentred oval.ResultsThe thinnest point on the corneawas located at an average of 0.90 (SD 0.51)mm from visual axis and had an averagethickness of 0.55 (0.03) mm. In 69.57% ofeyes, this point was located in the infero-temporal quadrant, followed by the su-perotemporal quadrant in 23.91%, theinferonasal quadrant in 4.35%, and thesuperonasal quadrant in 2.17%. Amongthe nine regions of the cornea evaluated(central, superotemporal, temporal, in-ferotemporal, inferior, inferonasal, nasal,superonasal, and superior) the centralcornea had the lowest average thickness(0.56 (0.03) mm) and the superior corneahad the greatest average thickness (0.64(0.03) mm). The mean simulatedkeratometry (SimK) was 44.24 (1.61)/43.31(1.66) dioptres (D) and the mean astigma-tism was 0.90 (0.41) D. Island (71.74%) wasthe most common elevation pattern ob-served in the anterior corneal surface, fol-lowed by incomplete ridge (19.57%),regular ridge (4.34%), irregular ridge(2.17%), and unclassified (2.17%). Island(32.61%) was the most common topo-graphic pattern in the posterior cornealsurface, following by regular ridge(30.43%), incomplete ridge (23.91%), andirregular ridge (13.04%) patterns. Sym-metric bow tie was the most common axialpower pattern in the anterior cornea(39.13%), followed by oval (26.07%), asym-metric bow tie (23.91%), round (6.52%),and irregular (4.53%) patterns. In thepachymetry maps, 47.83% of eyes had anoval pattern, and round, decentred oval,

    and decentred round were observed in41.30%, 8.70%, and 2.18% of eyes, respec-tively.ConclusionThe information on regionalcorneal thickness, corneal elevation andaxial corneal curvature obtained with theOrbscan corneal topography system fromnormal eyes provides a reference for com-parison with diseased corneas. The Orb-scan corneal topography system is a usefultool to evaluate both corneal topographyand corneal thickness.(Br J Ophthalmol 1999;83:774778)

    Measurement of shape, refractive power, andthe thickness of the cornea are very importantfor designing vision correction surgeries anddiagnosing corneal diseases. The majority ofthe commercial computerised videokeratogra-phy instruments measure the topography ofthe anterior corneal surface, which accountsfor the majority of the refractive power of thecornea. However, the posterior surface of thecornea also contributes to its refractive power.1

    Corneal thickness is an important factor toevaluate corneal barrier and endothelial pumpfunction.24 Accurate measurement of cornealthickness is helpful in the diagnosis of cornealdiseases and avoidance of keratorefractive sur-gery complications. Hence, an instrument thatevaluates all of these factors will be very usefulfor clinical practice.

    The Orbscan corneal topography system is arecently developed device that evaluates ante-rior and posterior corneal surface topographyas well as the thickness of the entire cornea.This instrument has the potential to greatlyincrease our knowledge about corneal shapeand function in health and disease. Thepurpose of this study was to establish normalvariables for elevation and curvature topogra-phy of the anterior and posterior cornea, aswell as thickness of the entire cornea with theOrbscan corneal topography system.

    Subjects and methodsThis study evaluated 94 eyes (46 right eyes, 48left eyes) of 51 subjects (24 male and 27female). Normal subjects were defined as thosewho had no complaints of ocular irritation, nohistory of contact lens use, no anterior segmentabnormality on biomicroscopic examination,and a manifest refractive error of less than6.00 dioptres of myopia and less than 2.00dioptres of astigmatism with a best correctedvisual acuity of 20/20 or better. They wererecruited from employees of the Bascom

    Br J Ophthalmol 1999;83:774778774

    Ocular Surface andTear Center, BascomPalmer Eye Institute,University of MiamiSchool of Medicine,Miami, USAZ LiuA J HuangS C Pflugfelder

    ZhongshanOphthalmic Center,Sun Yat-sen Universityof Medical Sciences.Guangzhou 510060,P R ChinaZ Liu

    Correspondence to:Stephen C Pflugfelder,Bascom Palmer Eye Institute,900 NW 17th Street, Miami,FL 33136, USA.

    Accepted for publication13 January 1999

    on 27 July 2018 by guest. Protected by copyright.



    r J Ophthalm

    ol: first published as 10.1136/bjo.83.7.774 on 1 July 1999. Dow

    nloaded from


  • Palmer Eye Institute and patients presentingfor routine eye examination.

    The Orbscan corneal topography system(Orbscan, Inc, Salt Lake City, UT, USA) wasperformed by one individual in the followingmanner. The subjects chin was placed on thechin rest and the forehead rested against theforehead strap. The subject was instructed tolook at a blinking red fixation light. The exam-iner adjusted the optical head using a joystickto align and focus the eye so that the corneawas centred on the video monitor and both halfslits were seen on the cornea. The video imagewas then captured.

    The Orbscan corneal topography systemmeasures anterior and posterior corneal eleva-tion (relative to a best fit sphere), surfacecurvature, as well as corneal thickness using ascanning optical slit device. The optical acqui-sition head scans the eye using light slits thatare projected at a 45 degree angle. Twenty slitsare projected sequentially on the eye from theleft and 20 slits from the right side for a total of40 slits. The instruments software analyses upto 240 data points per slit and calculates theaxial curvature (mm or dioptres) of theanterior and posterior corneal surface. It alsocalculates the elevation of the anterior andposterior surface of the cornea (relative to abest fit sphere) as well as the corneal thicknessof the entire cornea.

    Pachymetry is determined by this instru-ment from the diVerence in elevation between

    the anterior and posterior surface of thecornea. This instrument averages pachymetryin nine circles of 2 mm diameter that arelocated in the centre of the cornea and at eightlocations in the mid-peripheral cornea (supe-rior, superotemporal, temporal, inferotempo-ral, inferior, inferonasal, nasal, superonasal),each located 3 mm from the visual axis. TheOrbscan corneal topography system softwarealso identifies the thinnest point on the corneaand marks its distance from visual axis and itsquadrant location (superotemporal, inferotem-poral, superonasal, and inferonasal).

    The colour coded maps showing the eleva-tion pattern of the anterior and posteriorsurface of cornea were classified according to aclassification scheme for anterior elevationmaps of normal corneas made with the PARcorneal topography system.5 This scheme clas-sifies maps into regular ridge, irregular ridge,incomplete ridge, island, and unclassifiedpatterns as shown in Figure 1. The axial curva-ture maps of the anterior corneal surface wereclassified into round, oval, symmetric bow tie,asymmetric bow tie, and irregular patterns,using a proposed classification scheme ofcorneal topography obtained with the TMS-1corneal topography system.6

    Because there was no existing method forclassifying corneal pachymetry maps, thefollowing classification system was devised.The warmest colour in the pachymetry map,identifying the thinnest area of the cornea, wasused to designate one of four diVerentpatterns: round, oval, decentred round, anddecentred oval. These patterns are depicted inFigure 2. The following objective criteriadefined the categories used for this classifi-cation:

    Round: the ratio of the shortest to the longestdiameter of the warmest colour zone was twothirds or greater, and the warmest colour waseither entirely located at the centre of the mapor more than half of the area of the warmestcolour entered the central 3 mm diameterzone.

    Oval: the ratio of the shortest to the longestdiameter at the warmest colour zone was lessthan two thirds, and the warmest colour waseither entirely located at the centre of map ormore than half of area of the warmest colourentered the central 3 mm diameter zone.

    Decentred round: the warmest colour wasround, and more than half of its area was out-side of the central 3 mm diameter zone or itwas entirely located in the peripheral cornea.

    Decentred oval: the warmest colour was oval,and more than half of its area was outside thecentral 3 mm diameter zone or it was entirelylocated in the peripheral cornea.


    Statistical analysis was performed by MrWilliam Feuer, a biostatistician in our depart-ment. The mean and standard deviation of themean was used for various descriptive quanti-ties. DiVerence in topographic measurementsbetween eyes was calculated using t test.

    Figure 1 Elevation patterns of the posterior corneal surface: (A) regular ridge; (B)irregular ridge; (C) incomplete ridge; (D) island; (E) unclassified.

    Evaluation of corneal thickness and topography in normal eyes using the Orbscan corneal topography system 775

    on 27 July 2018 by guest. Protected by copyright.



    r J Ophthalm

    ol: first published as 10.1136/bjo.83.7.774 on 1 July 1999. Dow

    nloaded from


  • ResultsA total of 94 eyes of 51 subjects (both eyes of43 subjects and one eye of eight subjects) wereinvestigated using the Orbscan corneal topo-graphy system. The fellow eye of eight subjectswas excluded from this study was because ofprevious surgery (four eyes), contact lens wear(three eyes), and eyelid abnormality (one eye).The mean age of these subjects was 47.32(SD14.06) years (range 2378 years). Theright eye of normal subjects (n=46) was used

    for the statistical analysis and the classificationof topographic patterns.

    The mean simulated keratometry (SimK)measurement was 44.24 (1.61)/43.31 (1.66)dioptre (D), which was not statistically diVer-ent (p>0.2) from the mean SimK values in astudy of normal corneal topography with theTMS-1 reported by Bogan et al.6 The meankeratometric astigmatism was 0.90 (0.41) D,which was also not significantly diVerent fromthat reported by Bogan et al6 (p>0.1) andanother investigation of normal corneal topo-graphy with PAR Corneal Topography Systemreported by Naufal et al 5 (p >0.5). Mean astig-matism in the 3 mm, 5 mm, and 7 mmdiameter zones was 1.22 (1.08) D, 0.94 (0.44)D, and 1.51 (0.91) D, respectively. Table 1shows the astigmatism, irregularity, and meankeratometric power of the anterior and poste-rior corneal surfaces and the combination ofthe anterior and posterior corneal surfaces inthe 3 mm, 5 mm, and 7 mm diameter zones.There was a decrease in the refractive power ofthe cornea and an increase in the irregularityfrom the centre to the periphery of the corneain both the anterior and posterior corneal sur-faces. No significant diVerence was noted in allof above factors between eyes.

    Figure 2 Pachymetric patterns in maps generated with the Orbscan corneal topography system: (A) oval; (B) round; (C) decentred round; (D) decentredoval.

    Table 1 Distribution of astigmatism, irregularity, and mean power in anterior andposterior corneal surface and combination of anterior and posterior corneal surface in3 mm, 5 mm, and 7 mm diameters of 46 subjects

    3 mmMean (SD)

    5 mmMean (SD)

    7 mmMean (SD)

    Astigmatism (D) 1.22 (1.08) 0.94 (0.44) 1.51 (0.91)Anterior corneal surface

    Irreg* 1.16 (0.53) 1.73 (0.75) 2.89 (1.03)Mean (D) 49.30 (1.91) 48.78 (1.68) 47.76 (1.55)

    Posterior corneal surfaceIrreg* 0.37 (0.18) 0.52 (0.25) 1.04 (0.70)Mean (D) 6.30 (0.21) 6.16 (0.23) 5.95 (0.25)

    Combination of anterior and posterior corneal surfaceIrreg* 1.05 (0.48) 1.57 (0.69) 2.57 (1.00)Mean (D) 44.29 (1.72) 43.82 (1.54) 42.89 (1.39)

    *Irregularity = maps represent deviation of both the mean and astigmatic curvature in adjacentsmall areas.Mean power = Gaussian power proportional to the mean curvature of the optical surface andbased on physiological refractive indices.

    776 Liu, Huang, Pflugfelder

    on 27 July 2018 by guest. Protected by copyright.



    r J Ophthalm

    ol: first published as 10.1136/bjo.83.7.774 on 1 July 1999. Dow

    nloaded from


  • The most common anterior corneal eleva-tion pattern was the island (71.74%), followedby incomplete ridge, regular ridge, irregularridge, and unclassified patterns (Table 2). Theisland was the most commonly observedposterior corneal elevation pattern (32.61%),followed by regular ridge, incomplete ridge,and irregular ridge (Table 2). No unclassifiedpattern appeared in posterior corneal elevationmaps. The symmetric bow tie was the mostcommonly observed topographic pattern in theaxial power maps of anterior corneal surface(39.13%), following by oval, asymmetric bowtie, round, and irregular (Table 3).

    The thinnest site on the entire cornea was anaverage of 0.55 (0.03) mm thick and waslocated at an average of 0.90 (0.51) mm fromthe visual axis. This site was most commonlylocated in the inferotemporal quadrant(69.57%), followed by the superotemporal,inferonasal, and superonasal quadrants (Fig 3).Among the nine regions evaluated, the centralcornea was found to be the thinnest (0.56(0.03) mm). The superior cornea was thethickest (0.64 (0.03) mm), followed in order ofdecreasing thickness by the superonasal, infe-ronasal, inferior, superotemporal, nasal, infero-temporal and temporal zones (Fig 4). Pachy-metry colour coded maps were classified intoround, oval, decentred round, and decentredoval patterns. The majority of eyes had eitheroval (47.83%) or round (41.30%) patterns(Table 4). There was no statistical diVerence incorneal thickness of the thinnest site or the

    central and mid-peripheral measured sitesbetween eyes.

    DiscussionClinicians have a number of methods formeasuring corneal power and/or corneal shapeincluding keratometry, keratoscopy, photo-keratography, interferometry, computer assistedvideokeratography, and rasterstereography.712

    Before the introduction of the Orbscan cornealtopography system, placido based computeredassisted videokeratoscopy and rasterstereogra-phy were the only commercially availablemodalities to clinically evaluate corneal topogra-phy. The computered assisted videokeratoscopeuses a collimating cone to reflect 25 or 30 ringsoV the corneal surface.6 These reflected ringsyield as many as 8000 data points for computeranalysis. This technique detects curvature andrefractive power of the anterior corneal surface.The PAR corneal t...


View more >