Ophthalmic Surgery, Lasers and Imaging Retina

Clinical Science 

Intra-Observer and Inter-Observer Reliability and Reproducibility of Slit-Lamp–Adapted Optical Coherence Tomography for Evaluation of Anterior Chamber Depth and Central Corneal Thickness

Shlomit F. Sandler, BA; Joseph R. Zelefsky, MD; Syril Dorairaj, MD; Stella N. Arthur, MD, MSPH; Robert Ritch, MD; Jeffrey M. Liebmann, MD

Abstract

Background and Objective:

To evaluate the intra-observer and inter-observer reproducibility of anterior chamber depth (ACD) and central corneal thickness (CCT) measurements taken by slit-lamp–adapted optical coherence tomography (SL-OCT) and to evaluate concordance of CCT and ACD values obtained by SL-OCT with gold standard devices.

Patients and Methods:

Three images of each eye were taken by two different operators using SL-OCT. ACD and CCT values obtained by each operator were determined and compared to one another and to ultrasonic pachymetry (CCT) and axial OCT biometry (ACD) values obtained by a third, masked operator. Intra-observer and inter-operator reproducibility was assessed using the intraclass correlation coefficient (ICC).

Results:

Forty-one eyes of healthy subjects were enrolled. Mean CCT was 0.556 ± 0.020, 0.557 ± 0.019, and 0.532 ± 0.032 and mean ACD was 3.12 ± 0.36, 3.13 ± 0.38, and 3.50 ± 0.44 for operators 1, 2, and 3, respectively (P < .001, for both CCT and ACD). Intra-operator reproducibility for CCT (ICC of 0.948 to 0.995) and ACD (ICC of 0.972 to 0.987) was excellent.

Conclusion:

SL-OCT provides highly reproducible measures for both CCT and ACD. SL-OCT yields higher values for CCT compared to ultrasonic pachymetry and lower values for ACD compared to axial OCT biometry. Caution should be taken when extrapolating these data into clinical practice until the differences between SL-OCT, axial OCT biometry, and ultrasonic pachymetry can be further elucidated.

Abstract

Background and Objective:

To evaluate the intra-observer and inter-observer reproducibility of anterior chamber depth (ACD) and central corneal thickness (CCT) measurements taken by slit-lamp–adapted optical coherence tomography (SL-OCT) and to evaluate concordance of CCT and ACD values obtained by SL-OCT with gold standard devices.

Patients and Methods:

Three images of each eye were taken by two different operators using SL-OCT. ACD and CCT values obtained by each operator were determined and compared to one another and to ultrasonic pachymetry (CCT) and axial OCT biometry (ACD) values obtained by a third, masked operator. Intra-observer and inter-operator reproducibility was assessed using the intraclass correlation coefficient (ICC).

Results:

Forty-one eyes of healthy subjects were enrolled. Mean CCT was 0.556 ± 0.020, 0.557 ± 0.019, and 0.532 ± 0.032 and mean ACD was 3.12 ± 0.36, 3.13 ± 0.38, and 3.50 ± 0.44 for operators 1, 2, and 3, respectively (P < .001, for both CCT and ACD). Intra-operator reproducibility for CCT (ICC of 0.948 to 0.995) and ACD (ICC of 0.972 to 0.987) was excellent.

Conclusion:

SL-OCT provides highly reproducible measures for both CCT and ACD. SL-OCT yields higher values for CCT compared to ultrasonic pachymetry and lower values for ACD compared to axial OCT biometry. Caution should be taken when extrapolating these data into clinical practice until the differences between SL-OCT, axial OCT biometry, and ultrasonic pachymetry can be further elucidated.

From Mount Sinai School of Medicine (SFS), New York; New York University School of Medicine (JRZ, JML), New York; Manhattan Eye, Ear, and Throat Hospital (JRZ, JML), New York; Einhorn Clinical Research Center (SD, SNA, RR, JML), New York Eye and Ear Infirmary, New York; and New York Medical College (RR), Valhalla, New York.

Presented in part at the Association for Research and Vision in Ophthalmology annual meeting, May 6–10, 2007, Fort Lauderdale, Florida.

Supported in part by the Ephraim and Catherine Gildor Fund of the New York Glaucoma Research Institute, New York, New York.

Address correspondence to Jeffrey M. Liebmann, MD, 310 East 14th Street, New York, NY 10003.

Accepted: October 20, 2007

Introduction

Devices that provide quantitative information regarding the anatomy of the anterior segment are increasingly being applied to the care of patients and in ophthalmic research. Current gold standards include ultrasonic pachymetry for measurement of central corneal thickness and ultrasound biomicroscopy for evaluation of anterior and posterior chamber anatomy, both of which require contact with the eye. Slit-lamp–adapted optical coherence tomography (SL-OCT) is being evaluated as an alternative, non-contact tool for obtaining this information and the assessment of patients with glaucoma.1–3

Currently, there is little information available describing the inter-operator and intra-operator measurement reproducibility for these technologies. The purpose of this study was to evaluate the reproducibility of SL-OCT images with respect to central corneal thickness and anterior chamber depth and to assess the concordance of these measurements with gold standard devices.

Patients and Methods

The study protocol was approved by the Institutional Review Board for Human Research of the New York Eye and Ear Infirmary. Normal subjects were enrolled after informed consent was obtained. Subjects with a history of ocular disease and prior incisional surgery were excluded, as was one eye for which collected data were incomplete.

Inclusion criteria were age 18 to 75 years and no ocular pathology apart from refractive error. Each eye was scanned three times each by operators 1 (SD) and 2 (SFS) using SL-OCT (1,310 nm; Heidelberg Engineering, GmBH, Dossenheim, Germany). Cross-sectional images of the cornea and views of the anterior chamber, iris, and anterior lens were obtained.

The anterior chamber depth and central corneal thickness were calculated from each image using software provided by the manufacturer (Figure). Operator 3 (SNA), masked to the prior results, obtained central corneal thickness and anterior chamber depth measurements using ultrasonic pachymetry (Pachette GDH 500; DGH Technology, Inc., Philadelphia, PA) and OCT biometry (IOL Master; Carl Zeiss Meditec, Inc., Dublin, CA), a non-contact optical device that measures the anterior chamber depth by means of a slit-illumination system with subsequent image evaluation.4 All ultrasonic pachymetry measurements were based on standard speed-of-sound through the cornea of 1,640 m/s. The anterior chamber depth was established as the distance between the posterior surface of the central cornea and the crystalline lens (Zeiss IOL Master Manual; Carl Zeiss Meditec, Inc.). The SL-OCT measurements were evaluated for intra-operator and inter-operator reproducibility and measurements from all three techniques were evaluated for concordance.

Slit-Lamp–Adapted Optical Coherence Tomography Image Depicting Anterior Chamber Depth (red Line) and Central Corneal Thickness (white Block).

Figure. Slit-Lamp–Adapted Optical Coherence Tomography Image Depicting Anterior Chamber Depth (red Line) and Central Corneal Thickness (white Block).

Descriptive statistics were used to analyze demographic data. Repeated measures analysis of variance was used to compare the measurements of the three operators. Intraclass correlation coefficient (ICC) was used to assess individual and collective reproducibility. Pearson product-moment correlation (r) was used to evaluate correlation between the measurements. P values less than .05 were considered statistically significant.

Results

Forty-one eyes (21 right, 20 left) of 21 subjects (13 women, 8 men) were enrolled. Mean subject age was 32.0 ± 11.0 years (range: 18 to 58 years). Mean values for central corneal thickness were 0.556 mm for operator 1 (range: 0.518 to 0.619 mm), 0.557 mm for operator 2 (range: 0.522 to 0.604 mm), and 0.532 mm for operator 3 (range: 0.469 to 0.595 mm). Mean values for anterior chamber depth were 3.12 mm for operator 1 (range: 2.38 to 4.00 mm), 3.13 mm for operator 2 (range: 2.36 to 4.05 mm), and 3.50 mm for operator 3 (range: 2.28 to 4.34 mm) (Table 1). All operators obtained highly reproducible values for both central corneal thickness (ICC = 0.952, 0.948, and 0.995 for operators 1, 2, and 3, respectively) and anterior chamber depth (ICC = 0.972, 0.987, and 0.977 for operators 1, 2, and 3, respectively) measurements (ICC = 95% confidence interval of 0.80 to 0.92).

Means (± Standard Deviation), P Values, and Intraclass Correlation Coefficients (ICCs) for Central Corneal Thickness (CCT) and Anterior Chamber Depth (ACD) Measurements Obtained by Three Different Operators

Table 1: Means (± Standard Deviation), P Values, and Intraclass Correlation Coefficients (ICCs) for Central Corneal Thickness (CCT) and Anterior Chamber Depth (ACD) Measurements Obtained by Three Different Operators

Operators 1 and 2 obtained higher central corneal thickness measurements and lower anterior chamber depth measurements than operator 3. Overall differences between the means were significant (P < .001, for both central corneal thickness and anterior chamber depth). The highest correlation was found between operators 1 and 2 for both central corneal thickness and anterior chamber depth measurements (r = 0.845 and 0.951, respectively), whereas the correlations between operators 1 and 3 and operators 2 and 3 were notably lower for both central corneal thickness and anterior chamber depth measurements (Table 2).

Interitem Pearson Correlation Matrix for Mean Central Corneal Thickness (CCT) and Anterior Chamber Depth (ACD) Values

Table 2: Interitem Pearson Correlation Matrix for Mean Central Corneal Thickness (CCT) and Anterior Chamber Depth (ACD) Values

Discussion

SL-OCT is a rapid, reproducible, non-contact method of imaging anterior segment anatomy, and its performance while the patient is in the sitting position makes it a desirable alternative to ultrasound biomicroscopy imaging.1–3,5 Software from the manufacturer allows for calculation of central corneal thickness and anterior chamber depth from the images produced by the device. Our data demonstrate excellent reproducibility of measurements obtained by SL-OCT with minimal differences between operators.

The measurement of central corneal thickness has become increasingly important in the treatment of patients with glaucoma6 and refractive surgery candidates.7 Ultrasonic pachymetry is the current gold standard for measuring central corneal thickness. Although studies have shown that ultrasonic pachymetry is a reliable and reproducible method for measuring central corneal thickness,8 Shildkrot et al.9 demonstrated that central corneal thickness measured by ultrasonic pachymetry can vary by at least 20 microns in up to 20% of subjects when measured on two separate visits. There are multiple potential sources to account for this variability, some of which are subject related (change in true central corneal thickness due to medication, contact lens wear, and age) and others that are examiner related (angle of the probe and site of applanation on the cornea). Because corneal thickness increases toward the periphery, imprecise probe placement leads to inaccurate central corneal thickness values using ultrasonic pachymetry.

OCT is a non-invasive imaging modality that can be used to image the anterior and posterior segment with great detail. Recently, OCT has been used to obtain measurements of central corneal thickness and has theoretical advantages over ultrasonic pachymetry. OCT is capable of making repeated cross-sectional measurements at the exact central part of the cornea, perpendicularity onto the central cornea is assured, and corneal touch is not required.10

Although central corneal thickness measurements were nearly identical between both SL-OCT operators in our study, concordance with ultrasonic pachymetry was low. Our results are consistent with prior reports suggesting that central corneal thickness measurements taken by OCT pachymetry are higher than those taken by ultrasonic pachymetry.10,11 A recent meta-analysis found that mean central corneal thickness was 14 microns less in OCT pachymetry than ultrasonic pachymetry.12 Fakhry et al. suggested a correction factor of 0.92 when comparing OCT pachymetry to ultrasound pachymetry.13 Leung et al.10 explained this inconsistency by noting that ultrasonic pachymetry requires contact with the surface of the cornea. As such, the applanation force of the probe tip may displace the tear film and compress the corneal epithelium, thus accounting for the discrepancy between OCT pachymetry and ultrasonic pachymetry measurements.

Further studies investigating this disparity will be necessary before such data could be applied to clinical practice. For example, the Ocular Hypertension Treatment Study risk calculator uses central corneal thickness values based on ultrasonic pachymetry to determine the risk for developing glaucoma. A modification would be necessary if OCT-based pachymetry were used to determine central corneal thickness.

Measurement of anterior chamber depth has important clinical value, both in the assessment of the patient with glaucoma and in the evaluation of the phakic intraocular lens candidate. In our study, SL-OCT had excellent intra-operator and inter-operator reproducibility when measuring anterior chamber depth and yielded lower values than the IOL Master. Previous evaluations of anterior chamber depth measurement using numerous devices, including IOL Master, AC-Master (Carl Zeiss Meditec, Inc.), Pentacam (Oculus, Inc., Lynnwood, WA), Jaeger Slit-Lamp Pachymeter (Haag-Streit AG, Inc., Koeniz, Switzerland), and Orbscan (Orbtek Inc, Salt Lake City, UT), have similarly revealed excellent reproducibility for each specific instrument but significant differences between mean and median anterior chamber depth values obtained by the different devices.4,14 The authors attribute these differences to different physical measuring principles and different instrument-specific correction factors related to each device.

This study was limited to eyes with no ocular disease or prior incisional surgery. Further studies investigating the performance of SL-OCT in the presence of corneal and anterior segment pathology are necessary before these data can be applied to clinical practice.

SL-OCT offers excellent reproducibility for central corneal thickness and anterior chamber depth measurements. These values differ from those of ultrasonic pachymetry and OCT biometry, a finding that is consistent with previous studies investigating the measurement of central corneal thickness and anterior chamber depth. The ability of SL-OCT to generate qualitative and quantitative information for anterior segment analysis makes it a potentially interesting research tool for assessing anterior chamber depth and central corneal thickness.

References

  1. : Wirbelauer C, Karandish A, Haberle H, Pham DT. Noncontact goniometry with optical coherence tomography. Arch Ophthalmol. 2005;123:179–185. doi:10.1001/archopht.123.2.179 [CrossRef]
  2. : Baikoff G, Lutun E, Wei J, Ferraz C. Static and dynamic analysis of the anterior segment with optical coherence tomography. J Cataract Refract Surg. 2004;30:1843–1850. doi:10.1016/j.jcrs.2004.05.024 [CrossRef]
  3. : Hoerauf H, Wirbelauer C, Scholz C, et al. . Slit-lamp-adapted optical coherence tomography of the anterior segment. Graefes Arch Clin Exp Ophthalmol. 2000;238:8–18. doi:10.1007/s004170050002 [CrossRef]
  4. : Meinhardt B, Stachs O, Stave J, Beck R, Guthoff R. Evaluation of biometric methods for measuring the anterior chamber depth in the non-contact mode. Graefes Arch Clin Exp Ophthalmol2006;244:559–564. doi:10.1007/s00417-005-0103-7 [CrossRef]
  5. : Radhakrishnan S, Goldsmith J, Huang D, et al. . Comparison of optical coherence tomography and ultrasound biomicroscopy for detection of narrow anterior chamber angles. Arch Ophthalmol2005;123:1053–1059. doi:10.1001/archopht.123.8.1053 [CrossRef]
  6. : Gordon MO, Beiser JA, Brandt JD, et al. . The Ocular Hypertension Treatment Study: baseline factors that predict the onset of primary open-angle glaucoma. Arch Ophthalmol2002;120:714–720.
  7. : Price FW Jr, Koller DL, Price MO. Central corneal pachymetry in patients undergoing laser in situ keratomileusis. Ophthalmology1999;106:2216–2220. doi:10.1016/S0161-6420(99)90508-0 [CrossRef]
  8. : Miglior S, Albe E, Guareschi M, Mandelli G, Gomarasca S, Oralesi N. Intraobserver and interobserver reproducibility in the evaluation of ultrasonic pachymetry measurements of central corneal thickness. Br J Ophthalmol2004;88:174–177. doi:10.1136/bjo.2003.023416 [CrossRef]
  9. : Shildkrot Y, Liebmann JM, Fabijanczyk B, Tello CA, Ritch R. Central corneal thickness measurement in clinical practice. J Glaucoma2005;14:331–336. doi:10.1097/01.ijg.0000176929.83734.b4 [CrossRef]
  10. : Leung DY, Lam DK, Yeung BY, Lam DS. Comparison between central corneal thickness measurements by ultrasound pachymetry and optical coherence tomography. Clin Experiment Ophthalmol2006;34:751–754. doi:10.1111/j.1442-9071.2006.01343.x [CrossRef]
  11. : Bechmann M, Thiel MJ, Neubacher AS, et al. . Central corneal thickness measurement with a retinal optical coherence tomography device versus standard ultrasonic pachymetry. Cornea2001;20:50–54. doi:10.1097/00003226-200101000-00010 [CrossRef]
  12. : Doughty MJ, Zaman ML. Human corneal thickness and its impact on intraocular pressure measures: a review and meta-analysis approach. Surv Ophthalmol. 2000;44:367–408. doi:10.1016/S0039-6257(00)00110-7 [CrossRef]
  13. : Fakhry MA, Artola A, Belda JI, Ayala MJ, Alió JL. Comparison of corneal pachymetry using ultrasound and Orbscan II. J Cataract Refract Surg. 2002;28:248–252. doi:10.1016/S0886-3350(01)01277-9 [CrossRef]
  14. : Buehl W, Stojanac D, Sacu S, Drexler W, Findl O. Comparison of three methods of measuring corneal thickness and anterior chamber depth. Am J Ophthalmol2006;141:7–12. doi:10.1016/j.ajo.2005.08.048 [CrossRef]

Means (± Standard Deviation), P Values, and Intraclass Correlation Coefficients (ICCs) for Central Corneal Thickness (CCT) and Anterior Chamber Depth (ACD) Measurements Obtained by Three Different Operators

VariableCCTACD
Operator 10.556 ± 0.020a3.12 ± 0.36a
Operator 20.557 ± 0.019a3.13 ± 0.38a
Operator 30.532 ± 0.032b3.50 ± 0.44c
Pd< .001< .001
ICC0.7670.845

Interitem Pearson Correlation Matrix for Mean Central Corneal Thickness (CCT) and Anterior Chamber Depth (ACD) Values

VariableOperator 1Operator 2Operator 3
CCT
  Operator 11.000.8450.688
  Operator 20.8451.000.780
  Operator 30.6880.7801.00
ACD
 Operator 11.000.9510.812
 Operator 20.9511.000.762
 Operator 30.8120.7621.00
Authors

From Mount Sinai School of Medicine (SFS), New York; New York University School of Medicine (JRZ, JML), New York; Manhattan Eye, Ear, and Throat Hospital (JRZ, JML), New York; Einhorn Clinical Research Center (SD, SNA, RR, JML), New York Eye and Ear Infirmary, New York; and New York Medical College (RR), Valhalla, New York.

Address correspondence to Jeffrey M. Liebmann, MD, 310 East 14th Street, New York, NY 10003.

10.3928/15428877-20080701-15

Sign up to receive

Journal E-contents