Reoperations in strabismus surgery are common.1 Unfortunately, records from previous surgery are frequently unavailable. An imaging modality that can reliably detect extraocular muscle (EOM) insertion in patients with recent or remote history of strabismus surgery would be a useful clinical tool, decreasing the need for surgical exploration, and therefore, decreasing time under anesthesia and excessive conjunctival scarring.
Anterior segment optical coherence tomography (AS-OCT) is a non-contact imaging method that can provide accurate measurements of EOM to limbus insertion distances.2,3 Studies that have investigated this technique in muscles that underwent prior surgical procedures demonstrated less success when compared to attempts on surgically naïve muscles.5–8 Most studies have compared preoperative AS-OCT measurements to intraoperative measurements in the setting of muscle resection or standard recessions.4–8
Patients and Methods
All adults undergoing horizontal muscle surgery were screened for eligibility. Exclusion criteria included inability to comply with AS-OCT imaging or any anatomical variation that would affect OCT images. Institutional review board approval was obtained and informed consent was acquired for all participants.
The hang-back technique and standard recessions or resections were performed by two surgeons (LBN, KBG). Intraoperative measurements were performed using Castroviejo calipers from the limbus to the newly reattached muscle insertion. The muscle insertion was defined as the midpoint where the anterior fibers of the muscle adhered to or touched the sclera with all slack removed from the suture in the case of hang-back procedures. Caliper measurements were recorded in 0.5-mm increments.
AS-OCT images in the transverse plane with an anterior segment lens (version 9.5.2; Zeiss Cirrus 5000 HD OCT) were obtained at the 1- to 2-week postoperative visit (Figure 1). During OCT image collection, patients were instructed to look in the opposite direction of the operated muscle. The best quality image was selected for analysis and uploaded to ImageJ software (National Institutes of Health). The muscle insertion was defined as the posterior face of the rectus muscle insertion on the sclera. A masked examiner measured the distance from the iris root to the muscle insertion three separate times using ImageJ. One millimeter was added to the measurement because, anatomically, the iris root lies approximately 1 mm posterior to the corneoscleral junction.9 The average of the three measurements was calculated. Intraclass correlation coefficients (ICCs) were used to calculate agreement between intraoperative and AS-OCT measurements. Good and excellent agreement was defined as 0.16 to 0.80 and less than 0.80, respectively.10 A difference of ±1 mm between AS-OCT and surgical measurements was considered clinically acceptable. Bland-Altman analysis was completed.
Optical coherence tomography image. The measurement is marked with a line from anterior angle to muscle insertion.
Thirty-one eye muscles from 28 patients were included in the study. Twenty patients were women. The average age was 46 ± 15 years. There were 14 medial rectus muscle surgeries (7 recessions, 7 re-sections) and 17 lateral rectus muscle surgeries (12 recessions, 5 resections).
EOM insertion was identifiable on AS-OCT for 14 (45%) eye muscles from 13 (46%) patients. The difference between the average measurement of non-identifiable muscles (12.0 mm; CI: 10.6 to 14.0 mm) and identifiable muscles (7.5 mm; CI: 5.5 to 9.6 mm) was statistically significant (P < .001). A comparison of measurements for caliper and AS-OCT by muscle and procedure type is listed in Table 1.
Caliper vs AS-OCT Measurements for Muscle Insertions Identified on AS-OCT
The ICC between all AS-OCT and surgical measurements was 0.771. For lateral and medial rectus muscle measurements alone, the ICC was 0.668 and 0.879, respectively. For all muscles that had insertions less than 8 mm from the limbus, the ICC was 0.887. For larger caliper measurements (≥ 8 mm), the ICC was small (< .001). ICC values are summarized in Table 2. Of the 14 AS-OCT measurements obtained, 8 (57.1%) were within ±1 mm of caliper values. All AS-OCT measurements that were within ±1 mm of caliper values had a caliper measurement less than 8 mm. As demonstrated by the Bland-Altman plot (Figure 2), 13 (92.8%) of the OCT measurements were within ±1.96 standard deviations of the mean difference between AS-OCT and caliper measurements, which is under the 95% cut-off value to classify as good agreement. It is apparent from the Bland-Altman plot that differences between AS-OCT and caliper measurements increased with increased length from limbus to muscle insertion.
ICC Measurements Overall and by Subgroup
Bland-Altmann plot for differences between anterior segment optical coherence tomography (AS-OCT) and caliper measurements. The differences between AC-OCT and caliper measurements (y axis) are plotted as a function of the average between AS-OCT and caliper measurements (x axis). The middle line represents the average of differences between AS-OCT and caliper measurements, and the upper and lower lines represent 1.96 standard deviations from the average.
AS-OCT can be a valuable tool to help identify the location of an EOM insertion in some patients with a history of strabismus surgery. In our study, only 45% of muscles were identified in the postoperative period using AS-OCT. Other studies on previously operated muscles have reported identifying 78% to 94% of the EOM with OCT.5,6 In the early postoperative period, Lee et al11 also reported difficulty identifying EOM insertion due to postoperative tissue swelling, until 3 months postoperatively. The current study also supports that AS-OCT may not be a useful tool to accurately identify muscle insertions in the acute postoperative setting, perhaps limiting its usefulness in confirming slipped muscles.
We compared AS-OCT and caliper measurements quantitatively for muscles with insertion sites greater than and less than 8 mm from the limbus. Agreement between AS-OCT and surgical measurements was better for muscles inserted less than 8 mm from the limbus. From the Bland-Altman analysis, the AS-OCT underestimated caliper measurements less than 8 mm and overestimated larger caliper measurements. The imaging frame of the AS-OCT used in the current study has a maximum of 15 mm. Approximately 3 mm is needed on either side to accurately identify the EOM insertion and limbus. Therefore, this technology limits the feasibility of AS-OCT to measure the EOM insertion from the limbus to a maximum of approximately 9 mm.
In addition, larger caliper measurements had larger differences from AS-OCT measurements in general. Although some studies have reported identifying the lateral rectus muscle up to 14 mm and the medial rectus muscle 12 mm from the limbus, the vast majority of identified muscles are within 11 mm of the limbus.5 Different OCT machines use different wavelengths, allowing greater length of penetration into tissue.8 In the current study, in the acute postoperative period, EOM insertions were identified accurately within 8 mm of the limbus, although Lee et al11 reported recessions of the lateral rectus muscle up to 8 mm or 11.6 mm from the limbus could be identified on OCT within 6 months postoperatively.
Previous studies that have compared AS-OCT and surgical measurements in muscles with a history of prior operations have ICC ranges from 0.73 to 0.90, report 50% to 92% of AS-OCT measurements within 1 mm of caliper measurements, and have successfully identified reoperated muscles in 37.5% to 94% of cases.5–8 Our ICC (0.771), percentage of ASOCT measurements within ±1 mm of caliper measurements (57.1%), and percentage of identifiable muscle insertions (45%) are comparable to values of the previous studies. No prior study included hang-back recessions. The majority of recessions in our study were done by hang-back technique.
There are several limitations in our study. The number of measurements obtained is notably small despite reaching statistical significance. Further studies with larger sample sizes are needed to confirm the limitations of the distance from the limbus at which postoperative muscles can be identified. Additionally, caliper measurements were to the nearest 0.5 mm, and AS-OCT measurements were to the nearest 0.01 mm, which may have created a discrepancy when comparing the measurements. AS-OCT measurements were obtained at one postoperative visit, within the first 1 to 2 weeks after surgery. Our measurement landmarks of the corneoscleral junction and posterior face of the muscle insertion may be particularly affected by postoperative swelling.11 Muscle insertions may have been more easily identifiable if AS-OCT measurements were taken at a longer interval postoperatively.11 More studies of the utility of AS-OCT in identifying EOM previously recessed more than 8 mm from the limbus are needed.
- Repka MX, Lum F, Burugapalli B. Strabismus, strabismus surgery, and reoperation rate in the United States: analysis from the IRIS Registry. Ophthalmology. 2018;125(10):1646–1653. doi:10.1016/j.ophtha.2018.04.024 [CrossRef]
- Park KA, Lee JY, Oh SY. Reproducibility of horizontal extra-ocular muscle insertion distance in anterior segment optical coherence tomography and the effect of head position. J AAPOS. 2014;18(1):15–20.
- Salcedo-Villanueva G, Paciuc-Beja M, Harasawa M, et al. Identification and biometry of horizontal extraocular muscle tendons using optical coherence tomography. Graefes Arch Clin Exp Ophthalmol. 2015;253(3):477–485. doi:10.1007/s00417-014-2862-5 [CrossRef]
- Liu X, Wang F, Xiao Y, Ye X, Hou L. Measurement of the limbus-insertion distance in adult strabismus patients with anterior segment optical coherence tomography. Invest Ophthalmol Vis Sci. 2011;52(11):8370–8373. doi:10.1167/iovs.11-7752 [CrossRef]
- Ngo CS, Smith D, Kraft SP. The accuracy of anterior segment optical coherence tomography (AS-OCT) in localizing extraocular rectus muscles insertions. J AAPOS. 2015;19(3):233–236.
- Rossetto JD, Cavuoto KM, Allemann N, McKeown CA, Capó H. Accuracy of optical coherence tomography measurements of rectus muscle insertions in adult patients undergoing strabismus surgery. Am J Ophthalmol. 2017;176:236–243. doi:10.1016/j.ajo.2017.01.025 [CrossRef]
- Pihlblad MS, Reynolds JD. Anterior segment optical coherence tomography of previously operated extraocular muscles. Am Orthopt J. 2017;67(1):61–66. doi:10.1080/0065955X.2017.12023634 [CrossRef]
- de-Pablo-Gómez-de-Liaño L, Fernández-Vigo JI, Ventura-Abreu N, Morales-Fernández L, García-Feijóo J, Gómez-de-Liaño R. Agreement between intraoperative measurements and optical coherence tomography of the limbus-insertion distance of the extraocular muscles. Arch Soc Esp Oftalmol. 2016;91(12):567–572. doi:10.1016/j.oftale.2016.09.002 [CrossRef]
- Bron A, Tripathi R, Tripathi B. Wolff's Anatomy of the Eye and Orbit, 8th ed. Chapman & Hall; 1997.
- Bland M. An Introduction to Medical Statistics, 4th ed. Oxford University Press; 2015.
- Lee JY, Park KA, Lyu IJ, Oh SY. Postoperative change in lateral rectus muscle insertion measured by anterior segment optical coherence tomography. Eye (Lond). 2017;31(11):1556–1561. doi:10.1038/eye.2017.89 [CrossRef]
Caliper vs AS-OCT Measurements for Muscle Insertions Identified on AS-OCT
|Case No.||Muscle||Procedure||Caliper Measurement (mm)||AS-OCT Measurement (mm)|
ICC Measurements Overall and by Subgroup
|Subgroup||No. of Patients||ICC||CI||Grading|
|Overall||13 (14 eyes)||0.771||0.44 to 0.92||Good|
|Caliper < 8 mm||10||0.886||0.63 to 0.97||Excellent|
|Caliper ≥ 8 mm||4||0.000977||N/A||Poor|
|Lateral rectus muscle||6||0.6667||0.00 to 0.94||Good|
|Medial rectus muscle||8||0.878||0.55 to 0.97||Excellent|