Pak J Ophthalmol. 2021, Vol. 37 (3): 300-305 300 Original Article Inter-Observer Reproducibility of Axial Ocular Measurements with Non-contact HAAG-STRAIT Biometer Muhammad Suhail Sarwar 1 , Sehrish Shahid 2 , Muhammad Arslan Ashraf 3 , Shaista Kanwal 4 1-4 Department of Ophthalmology, Mayo Hospital, Lahore ABSTRACT Purpose: To check inter observer reproducibility of axial ocular measurements i.e. central corneal thickness (CCT), anterior chamber depth (ACD), aqueous depth (AD), lens thickness (LT), anterior segment lens (ASL), vitreous length (VL) and axial length (AL) with non-contact HAAG-STRAIT biometer. Study Design: Comparative Reproducibility Analysis. Place and Duration of Study: College of ophthalmology and allied vision sciences (COAVS), Mayo Hospital, Lahore. Methods: This study included 66 healthy students (132 eyes). Data was collected through self-designed proforma by 2 operators independently. SPSS 21 was used for data analysis. Interclass correlation was applied for agreement between the two readings. Interclass coefficient (ICC) value greater than 0.7 was considered as excellent correlation. Results: The mean CCT, AD, ACD, LT, ASL, VL, and AL were 526.47 ± 35.72 µm and 526.47 ± 36.06 µm (ICC = 0.92); 2.93 ± 0.29 mm and 2.93 ± 0.29 mm (ICC = 0.81); 3.45 ± 0.30 mm and 3.46 ± 0.30 mm (ICC = 0.79); 3.58 ± 0.28 mm and 3.56 ± 0.22 mm (ICC = 0.76); 7.03 ± 0.30 mm and 7.02 ± 0.27 mm (ICC = 0.80); 16.56 ± 0.85 mm and 16.62 ± 0.81 mm (ICC = 0.72); and 23.59 ± 0.85 mm and 23.64 ± 0.87 mm (ICC: 0.76) of observer 1 and 2, respectively. Conclusion: Non-contact Biometer (HAAG-STRAIT) has high inter-observer reproducibility with strong interclass coefficient of greater than 0.72. Key Words: Biometry, Axial Length, Central Corneal Thickness, Anterior Chamber Depth. How to Cite this Article: Sarwar MS, Shahid S, Ashraf MA, Kanwal S. Inter-Observer Reproducibility of Axial Ocular Measurements with Non-contact HAAG-TRAIT Biometer. Pak J Ophthalmol. 2021, 37 (3): 300-305. Doi: 10.36351/pjo.v37i3.1239 INTRODUCTION In last few decades, modernizations such as phacoemulsification, ocular biometry and intraocular lens (IOL) power estimate formulas have improved the Correspondence: Muhammad Arslan Ashraf Department of Ophthalmology, Mayo Hospital, Lahore Email: rajkumararslan@yahoo.com Received: March 08, 2021 Accepted: May 10, 2021 refractive outcomes of cataract surgery. 1-3 To encounter these prospects, consideration to precise biometry reading is critical. In recent cataract surgery and corneal refractive surgery, the biometric parameters like corneal curvature, CCT (central corneal thickness), ACD (anterior chamber depth), LT (lens thickness) and AL (axial length), ASL, VL are the most significant to achieve good refractive results. 4,5 Like contact biometer, Optical biometry gives IOL power calculation which is the key to get an emmetropic outcome after the surgery. 6-10 OPEN ACCESS Muhammad Suhail Sarwar, et al 301 Pak J Ophthalmol. 2021, Vol. 37 (3): 300-305 Non-contact optical biometry devices use the principle of partial coherence interferometry (PCI). It uses a 780-nm semiconductor diode laser. Besides AL, it can also measure ACD and keratometry (K) based on 6 points of reference in a 2.3 mm zone. It has an accuracy of ± 0.02 mm for AL measurement; with excellent reproducibility compared with ultrasound devices. 11 It also measures CCT which is important in vision improvement surgeries e.g. laser in situ keratomileusis (LASIK), as well as in glaucoma diagnosis and other corneal diseases. In addition, it can also provide measurements for LT. 11-13 This non-contact technique is associated with increased patient comfort and decreased risk for corneal complications when compared with immersion ultrasound biometry. It also allows for patient fixation during the measurement process, which increases the likelihood of the AL measurement being directly aligned to the fovea. However, obtaining measurements can be tough and less reliable in the human eyes with corneal opacities, dense posterior sub-capsular cataracts (PSC), macular disease, and poor fixation. 11,14 This study was done to find out the repeatability of axial ocular measurements i.e. CCT (central corneal thickness), ACD (anterior chamber depth), LT (lens thickness), anterior segment lens (ASL), vitreous length (VL) and AL (axial length) measured with non- contact biometer in patients visiting Mayo Hospital Lahore. METHODS It was a comparative reproducibility analysis and 132 was the sample size of healthy individuals who were students of college of ophthalmology and allied vision sciences (COAVS), Mayo hospital, Lahore. The mean age of males was 20.73 ± 2.337 and females was 21.17 ± 2.514 (Table 1). The sampling technique used in this study was non-probability convenient sampling. Patients with poor fixation, any opacity other than cataract or any other ocular pathology were excluded. Equipment used was pen torch, slit lamp and non- contact Biometer (Haag Streit model: LS 900). Log MAR visual acuity chart was used for visual acuity. Patients with visual acuity of 0.5 Log MAR or better were included. Age, gender and literality were independent variables while axial ocular parameters like CCT, ACD, AD, LT, ASL, VL and AL were dependent variables. Quantitative variables like age, CCT, AD, ACD, LT, ASL, VL and AL were presented as mean ± SD. SPSS 21 software was used for data analysis. Interclass correlation was applied for agreement between the two readings. Interclass coefficient (ICC) value greater than 0.7 was considered as excellent correlation. RESULTS Table 2 shows the mean axial ocular measurements, measured by observer 1 and 2. Interclass correlation showed excellent correlation between the two CCT readings (0.921), as well as between two readings of AD (0.813), ACD (0.792), LT (0.757), ASL (0.795), VL (0.719) and AL readings (0.759). Table 1: Descriptive statistics of Age distribution among gender. Descriptive Statistics Gender N Minimum Maximum Mean Std. Deviation Statistic Statistic Statistic Statistic Std. Error Statistic Female Age 74 18 28 20.73 .272 2.337 Male Age 58 18 28 21.17 .330 2.514 Inter-Observer Reproducibility of Axial Ocular Measurements with Non-contact HAAG-STRAIT Biometer Pak J Ophthalmol. 2021, Vol. 37 (3): 300-305 302 Table 2: Descriptive statistics of CCT, AD, ACD, LT, ASL, VL and AL measured by observer I and II. Descriptive Statistics Minimum Maximum Mean Std. Deviation Intra-class Correlation Mean Diff. Std. Deviation Statistic Statistic Statistic Std. Error Statistic Single Measures Average Measures CCT1 432 601 526.4697 3.10887 35.71824 .921 a .959 c 0 14.355 CCT2 430 610 526.4697 3.13861 36.05984 AD1 2.23 3.84 2.935 0.02529 0.29052 .813 a .897 c 0.001 0.17912 AD2 2.25 3.86 2.934 0.02552 0.29322 ACD1 2.49 4.37 3.4522 0.02574 0.29574 .792 a .884 c -0.008 0.18911 ACD2 2.72 4.38 3.4602 0.02518 0.28932 LT1 2.7 4.43 3.5752 0.02062 0.23693 .757 a .862 c 0.0135 0.16016 LT2 3.06 4.43 3.5617 0.01935 0.22237 ASL1 5.65 7.92 7.0273 0.02596 0.29826 .795 a .886 c 0.0055 0.18216 ASL2 6.32 7.99 7.0219 0.02337 0.26847 VL1 14.28 18.75 16.5575 0.07441 0.85495 .719 a .837 c -0.0574 0.62425 VL2 15.32 18.94 16.6148 0.0706 0.81115 AL1 21.63 25.71 23.58482 0.07425 0.853064 .759 a .863 c -0.0519 0.59786 AL2 21.99 25.8 23.6367 0.07568 0.86948 a. The estimator is the same, whether the interaction effect is present or not. b. Type A intra-class correlation coefficients using an absolute agreement definition. c. This estimate is computed assuming the interaction effect is absent, because it is not estimable otherwise. Figure 1: scatter chart showing regression value (0.846), strong relationship between both measurements of CCT. Figure 2: scatter chart showing regression value (0.676), moderate relationship between both measurements of ACD. Muhammad Suhail Sarwar, et al 303 Pak J Ophthalmol. 2021, Vol. 37 (3): 300-305 Figure 3: Scatter chart showing regression value (0.576), moderate relationship between both measurements of AL. Figure 4: Scatter chart showing regression value (0.575), moderate relationship between both measurements of LT. Figure 5: Scatter chart showing regression value (0.519), moderate relationship between both measurements of VL. Figure 6: Scatter chart showing regression value (0.637), moderate relationship between both measurements of ASL. DISCUSSION Optical biometry is being widely used by ophthalmologists to measure axial ocular measurement of eyes and to calculate the intraocular lens power excluding 5 to 10 percent of those eyes with dense cataract or poor fixation. With the help of biometer we can measure the CCT, AD, LT, AL and IOL power of eye. The accuracy of all parameters that can be measured by optical biometer is imperative for exact intraocular lens power calculation. In this study, like in some previous studies CCT, AD, ACD, ASL, VL and AL measurements have been performed by 2 observers. Andrew KC et al. showed a study to assess the repeatability and accuracy of non-contact device. The AL and ACD were measured by two practitioners independently by using non-contact biometer followed by ultrasound. There was good repeatability of AL and ACD. There was no difference on AL and ACD between the two practitioners. 15 Andrew Carkeet et al. also found the AL and ACD measurements with non- contact showed better repeatability. The mean difference of AL and ACD between the readings 2 and 1 was -0.006 mm and 0.009 mm, respectively. 16 L P J Cruysberg and co-workers evaluated the reproducibility with non-contact biometer of the Inter-Observer Reproducibility of Axial Ocular Measurements with Non-contact HAAG-STRAIT Biometer Pak J Ophthalmol. 2021, Vol. 37 (3): 300-305 304 Lenstar LS 900. CCT, ACD, LT and AL were attained to regulate the reproducibility of the Lenstar. The reproducibility of the Lenstar was more than 0.9%; for CCT, ACD, LT, K values and AL measurements. Even though all correlations were highly significant (p, 0.001). The reproducibility of the Lenstar was excellent. 17 In another study, the exactness of axial length measurements was tremendously high with ICC of 0.759. 18 Some of the measurements can be little different when taken by different instruments and technicians, but some of these measurements should be firmly checked in cases like central corneal thickness and cases of glaucoma or refractive surgery evaluation. This study measured the mean CCT of observer 1 and 2 as 526.47 ± 35.72 µm and 526.47 ± 36.06 µm, respectively. Interclass correlation (ICC) showed excellent correlation between the two reading (ICC: 0.921). Ramazan Yagc et al, also reported that the assortment of agreement for reproducibility was great for the measurements of central corneal thickness (1.610 and 3.077 for normal eyes and for the eyes with keratoconus, respectively). 18 Bengu E. found correlation coefficient to be 99.3% for Lenstar and 99.2% for UP (ultrasound pachymetry). The measurements taken by the two different technicians seemed to agree in a high level for both Lenstar (r = 0.993) and ultrasound pachymetry (r = 0.957). The actual importance of this study was that sample size was large and the interobserver unpredictability was estimated for both OLCR (optical low-coherence reflectometry) and UP (ultrasound pachymetry). 19 In our study, mean AD was 2.9350 ± 0.291 mm and 2.934 ± 0.293 mm of observer 1 and 2, respectively. ICC showed excellent correlation between two reading (ICC: 0.813). The mean ACD of observer 1 and 2 was 3.452 ± 0.296 mm and 3.460 ± 0.289 mm of observer 1 and 2, respectively. ICC showed excellent correlation between two reading (ICC: 0.792). According to a former study of Lenstar device, the accuracy of measurement of anterior chamber depth was high and the assortment of agreement was 0.025 millimeter and 0.069 millimeter in normal (emmetropic) eye and the eye with keratoconus, respectively. According to the assessment of Haigis formula, which uses the preoperative measurement of anterior chamber depth in the calculation of intraocular lens power, a difference of 0.06 millimeter in ACD affects the ultimate refraction by only 0.05 D. 18 J. S Shammas et al. also found that, with ICC of 0.946 the accuracy of the ACD measurements was high. 20 In our study, ICC showed excellent correlation between two LT readings (ICC: 0.757). H. John Shammas found high accuracy of the measurement of LT, with an ICC of 0.963. Ramazan Yagc et al, found that the non-contact biometer attained brilliant reproducibility for the measurements of axial length (assortment of agreement 0.038 and 0.041 for normal eyes and eyes having keratoconus, respectively). In a usual eye, a difference of 0.04 millimeter affects the final refraction by almost 0.10 D. 18 Limitation of this study are small sample size and it was a single center study. More data for our population is needed for further evaluation. This study can be improved with the participation of more than two observers. Moreover, comparison of reproducibility and repeatability of non-contact with contact biometer can also be done. CONCLUSION It is concluded that non-contact biometer (HaigStrait) has high reproducibility. The interclass coefficient value for CCT, AD, ACD, LT, ASL, VL and AL is greater than 0.7. Ethical Approval The study was approved by the Institutional review board/ Ethical review board. (Ref No. COAVS/276/2021) Conflict of Interest Authors declared no conflict of interest. REFERENCES 1. Eleftheriadis H. IOL Master biometry: refractive results of 100 consecutive cases. Br J Ophthalmol. 2003; 87 (8): 960-963. 2. Li Y, Li H-X, Liu Y-C, Guo Y-T, Gao J-M, Wu B, et al. Comparison of immersion ultrasound and low coherence reflectometry for ocular biometry in cataract patients. Intern J Ophthalmol. 2018; 11 (6): 966. 3. McAlinden C, Gao R, Yu A, Wang X, Yang J, Yu Y, et al. Repeatability and agreement of ocular biometry measurements: Aladdin versus Lenstar. Br J Ophthalmol. 2017; 101 (9): 1223-1229. Muhammad Suhail Sarwar, et al 305 Pak J Ophthalmol. 2021, Vol. 37 (3): 300-305 4. Goyal R, North RV, Morgan JE. Comparison of laser interferometry and ultrasound A-scan in the measurement of axial length. Acta Ophthalmol Scand. 2003; 81 (4): 331-335. 5. Haddad JS, Barnwell E, Rocha KM, Ambrosio Jr R, Waring IV GO. Comparison of biometry measurements using standard partial coherence interferometry versus new Scheimpflug tomography with integrated axial length capability. Clin Ophthalmol. (Auckland, NZ). 2020; 14: 353. 6. Cao X, Hou X, Bao YJJoo. The ocular biometry of adult cataract patients on lifeline express hospital eye- train in rural China. J Ophthalmol. 2015; 2015. 7. Aydin R, Erdur SK, Cabuk KS, Karahan E, Kaynak SJE. Comparision of Optical Low Coherence Reflectometry Versus Ultrasonic Biometry in High Hypermetropia. Eye Contact Lens, 2018; 44: S115- S117. 8. Çınar Y, Cingü AK, Şahin M, Şahin A, Yüksel H, Türkcü FM, et al. Comparison of optical versus ultrasonic biometry in keratoconic eyes. J Ophthalmol. 2013; 2013. 9. Erb-Eigner K, Hirnschall N, Hackl C, Schmidt C, Asbach P, Findl OJIo, et al. Predicting lens diameter: ocular biometry with high-resolution MRI. Invest Ophthalmol Vis Sci. 2015; 56 (11): 6847-6854. 10. Trivedi RH, Wilson MEJO. Axial length measurements by contact and immersion techniques in pediatric eyes with cataract. Ophthalmology, 2011; 118 (3): 498-502. 11. Rodrigo Guimara˜es de Souza M, Ildamaris Montes de Oca M, Isi Esquenazi M, Zaina Al-Mohtaseb M, Mitchell P. Weikert M. Updates in Biometry. Intern Ophthalmol Clin. 2017; 57 (3): 115-124. 12. Huang J, Lu W, Savini G, Chen H, Wang C, Yu X, et al. Comparison between a new optical biometry device and an anterior segment optical coherence tomographer for measuring central corneal thickness and anterior chamber depth. J Ophthalmol. 2016; 2016. 13. Ashraf MA, Sarwar MS, Afzal MA, Khalid I, Shahid S. Comparison of Axial Ocular Measurements with Contact and Non-Contact Biometry. Pak J Ophthalmol. 2020; 36(1). DOI: https://doi.org/10.36351/pjo.v36i1.922 14. de Souza RG, de Oca IM, Esquenazi I, Al-Mohtaseb Z, Weikert MP. Updates in Biometry. Intern Ophthalmol Clin. 2017; 57 (3): 115-24. 15. Lam AK, Chan R, Pang PCJO, Optics P. The repeatability and accuracy of axial length and anterior chamber depth measurements from the IOLMaster™. Ophthalmic Physiol Opt. 2001; 21 (6): 477-483. 16. Carkeet A, Saw S-M, Gazzard G, Tang W, Tan DTJO. Repeatability of IOL Master biometry in children. Optom Vis Sci. 2004; 81 (11): 829-834. 17. Cruysberg LP, Doors M, Verbakel F, Berendschot TT, De Brabander J, Nuijts RM. Evaluation of the Lenstar LS 900 non-contact biometer. Br J Ophthalmol. 2010; 94 (1): 106-110. 18. Yağcı R, Güler E, Kulak AE, Erdoğan BD, Balcı M, Hepşen İF. Repeatability and reproducibility of a new optical biometer in normal and keratoconic eyes. JCataract Refract Surg. 2015; 41 (1): 171-177. 19. Koktekir BE, Gedik S, Bakbak B. Comparison of central corneal thickness measurements with optical low-coherence reflectometry and ultrasound pachymetry and reproducibility of both devices. Cornea, 2012; 31 (11): 1278-1281. 20. Shammas HJ, Hoffer KJ. Repeatability and reproducibility of biometry and keratometry measurements using a noncontact optical low- coherence reflectometer and keratometer. Am J Ophthalmol. 2012; 153 (1): 55-61. e2. Authors’ Designation and Contribution Muhammad Suhail Sarwar; MS: Concepts, Design, Manuscript preparation, Manuscript review. Sehrish Shahid; Consultant Ophthalmologist: Literature search, Data acquisition, Data analysis, Manuscript preparation, review. Muhammad Arslan Ashraf; Consultant Ophthalmologist: Literature search, Data acquisition, Data analysis, Statistical analysis, Manuscript review.. Shaista Kanwal; Consultant Ophthalmologist: Data acquisition, Manuscript preparation, Manuscript review. .…  …. https://doi.org/10.36351/pjo.v36i1.922