- Email: [email protected]
Ultrasound Biomicroscopy in Strabismus Reoperations
Ultrasound Biomicroscopy in Strabismus Reoperations Shuan Dai, MBBS, FRANZCO,a,b Stephen P. Kraft, MD, FRCSC,a David R. Smith, MD, FRCSC,a and J. Raymond Buncic, MD, FRCSCa Purpose: To evaluate the accuracy of ultrasound biomicroscopy (UBM) for measuring the insertional distance from the limbus of horizontal extraocular muscles (EOM) that had previous surgery. Methods: This prospective, masked, observational study compared the distance of the horizontal EOM insertion from the limbus as measured by 50-MHz UBM, either preoperatively or at the time of the anesthesia, with that measured by surgical caliper intraoperatively. Results: Forty-three horizontal rectus muscles in 23 subjects were measured using the UBM. Of these 43 muscles, 41 (22 medial rectus and 19 lateral rectus muscles) were successfully imaged by UBM and also received intraoperative surgical measurements. Two medial rectus muscles of one patient were inserted too far from the limbus (beyond 12 mm) to be detected by UBM. The mean measured distances of the horizontal muscle insertions from the limbus were 9.0 mm with UBM and 9.3 mm at surgery (P ⫽ 0.0001, showing no significant difference). For 33 (80.5%) of the muscles, the 2 methods agreed within 1 mm (P ⬍ 0.0001). There was no difference in accuracy for lateral rectus compared with medial rectus muscles or when comparing the muscles imaged by UBM under topical versus general anesthesia. Conclusion: The 50-MHz UBM can accurately measure the horizontal EOM insertion distances from the limbus for muscles in patients that had previous surgery. The limit of detection was 12 mm from the limbus for the MR and 14 mm for the LR. Depending on the patient’s age and level of cooperation, the UBM study can be done under either general anesthesia or topical anesthesia. (J AAPOS 2006;10:202-205) he distance of the extraocular muscle (EOM) insertion from the limbus is a major factor influencing the amount of surgery (recession or resection) required for a given angle of deviation. Eye muscle surgery alters the normal insertion of the EOM,1 and information from previous surgery may not be readily available, which makes the planning of appropriate surgical procedures for those patients more difficult and may increase the risk of perioperative complications. Computed axial tomography,2 magnetic resonance imaging (MRI),3,4 cine MRI,5 and conventional A and B scan ultrasonography6,7 all have been used to study the size and position of the EOM. None of these modalities can accurately locate the insertion of the EOM to the globe and the insertion distance from limbus. In a recent pilot study,8 the
T
From the aDepartment of Ophthalmology and Vision Sciences, The Hospital for Sick Children and the University of Toronto, Toronto, Ontario, Canada bDepartment of Ophthalmology, University of Auckland, Auckland, New Zealand Presented in part as a poster at the 31st Annual Meeting of the American Association for Pediatric Ophthalmology and Strabismus, Orlando, Florida, March 9-13, 2005. Submitted July 1, 2005. Revision accepted December 27, 2005. Reprint requests: Stephen Kraft, MD, Deparment of Ophthalmology & Vision Sciences, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, Canada M5G 1X8 (e-mail: [email protected]). Copyright © 2006 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/2006/$35.00 ⫹ 0 doi:10.1016/j.jaapos.2006.01.209
202
June 2006
high-frequency ultrasound biomicroscope (UBM) was able to reliably estimate the horizontal rectus muscle insertion distance from limbus in the eyes of patients undergoing primary strabismus surgery. The purpose of this study was to determine whether UBM could accurately measure the insertion distance from the limbus of horizontal EOMs in patients that had undergone previous surgery.
METHODS This was a prospective, masked, observational study comparing the distance of the horizontal EOM insertion from the limbus as measured by UBM (Humphrey Instruments, San Leandro, CA) either preoperatively or at the time of the anesthesia with that measured by surgical caliper intraoperatively. The study was conducted between April 2004 and March 2005 at The Hospital for Sick Children in Toronto. This project had the approval of the hospital’s research ethical board, and informed consent was obtained for all the subjects or the guardians of patients who participated in this study. Subjects Twenty-three subjects (11 boys, 12 girls; mean age 16.9 years, range 7-41 years) with a previous history of horizontal muscle surgery and who underwent further surgery on horizontal rectus muscle(s) that were previously operated upon were recruited into the study. Subjects whose eyes had structural abnormalities such as microphthalmia Journal of AAPOS
Journal of AAPOS Volume 10 Number 3 June 2006
FIG 1. UBM image of insertion of rectus muscle and caliper used in measurement. White arrowhead indicates EOM insertion (anterior extent of potential space under muscle) and black arrowhead indicates position of surgical caliper.
or with existing scleral buckles were excluded from the study. UBM Measurements All the UBM measurements were performed by 1 examiner (S.D.). For patients younger than 14 years of age, UBM was performed under general anesthesia in the operating room immediately before the planned surgery but before preoperative preparation of the patient. For patients older than 14 years of age, UBM was performed using topical anesthesia in the outpatient clinic during the preoperative assessment visit. After adequate anesthesia was obtained, a lid speculum was inserted and tear gel was used as coupling agent without the need of a container. The UBM transducer was held tangentially over the muscle to be scanned and moved to and from the limbus, with the oscillations perpendicular to the limbus until a cross section of the muscle insertion was observed. A surgical caliper was introduced, with the tip of one arm fixed at the limbus and the other arm adjusted until the echo of its tip was seen in the UBM image (Figure 1). The echo from this tip of the caliper served as a landmark to facilitate the measurement of the distance of muscle insertion from the limbus, as the horizontal field of view of the 50-MHz UBM was limited to less than 5 mm. The area of the muscle insertion was characterized by a potential space located between the global surface of the muscle and the sclera, and which invariably terminated just behind the insertion into the sclera (Figure 1). The anterior extent of this potential space was identified as the insertion position in the same manner as described in our original UBM study in primary surgeries.8 Three images were obtained for each muscle being measured, and the image with the best quality was used for calculation of the distance of the muscle insertion
Dai et al
203
to the tip of the caliper. This distance between the black and white arrow (Figure 1), in millimeters, was added to the reading on the surgical caliper to produce the measured distance of muscle insertion from the limbus. The UBM screen provides a scale accurate to the nearest 0.1 mm and surgical caliper allowed a measurement to the nearest 0.5 mm. When UBM measurement was performed using topical anesthesia subjects were instructed to abduct the examined eye when the medial rectus (MR) muscle was measured and to adduct the examined eye when the lateral rectus muscle (LR) was measured.8 When UBM was performed under general anesthesia, the examined eye was held in abduction with a fixation forceps for the MR measurement and in adduction for the LR measurement. The UBM measurement was interpreted and stored in the hard drive of the UBM machine before the start of the surgery and the surgeon was masked to the values obtained. Measurements at Surgery During surgery, upon satisfactory exposure of the muscle insertion and after insertion of a muscle hook, the distance from the limbus to the midpoint of the muscle hook was measured by the surgeon (S.P.K., D.R.S., J.R.B.) using the caliper in the same manner as that used for UBM measurements. Measurements were recorded to the nearest 0.5 mm. This information was recorded on a separate data collection sheet designed for the study and was not revealed to the examiner who performed the UBM study (S.D.). We arbitrarily decided that the 2 methods were in agreement if the difference between UBM and intraoperative measurements of a given horizontal rectus muscle insertion distance was no more than 1 mm. The intraoperative surgical caliper measurement was considered as the “gold standard,” and the study was designed to test the proportion of UBM measurements equal to, or within, 1 mm from those obtained intraoperatively (either 1 mm more or less than the reading at surgery). Our null hypothesis was that the difference between the 2 measurement modalities for a given rectus muscle was greater than 1 mm. The data obtained from the two methods were analyzed using paired t-test, and a value of .05 was set as the level of significance.
RESULTS Fourteen subjects were younger than 14 years of age and had the UBM measurement performed under general anesthesia whereas 9 patients older than the age of 14 years had the UBM measurement performed using topical anesthetic. A total of 43 muscles were evaluated, including 24 MR muscles and 19 LR muscles. In one male patient, both of his previously recessed MR muscles could not be detected by the UBM. At surgery, they were located more than 12 mm from the limbus. These two muscles were
Journal of AAPOS Volume 10 Number 3 June 2006
204 Dai et al TABLE 1. EOM insertion distance from limbus measured by UBM and at surgery (mean ⫹ standard deviation, in mm) UBM, ultrasound biomicroscopy Overall (41 muscles) Medial rectus (22) Lateral rectus (19)
UBM
At surgery
p value
9.0 ⫾ 2.3 8.4 ⫾ 1.4 9.7 ⫾ 2.4
9.3 ⫾ 2.5 8.6 ⫾ 1.6 9.8 ⫾ 2.5
0.0001 0.0001 0.0001
Delta (mm)
2
1
0
-1
-2 5
6
7
8
9
10
11
12
13
14
Mean (mm) FIG 2. Scatterplot of “Delta” difference (between UBM and surgical measurements) for all 41 muscles vs. “Mean” (distance to muscle insertion). TABLE 2. UBM measurements under general anesthesia and topical anesthesia (mean ⫹ standard deviation, in mm) UBM, ultrasound biomicroscopy General anesthesia Topical anesthesia
UBM
At surgery
P value
9.3 ⫾ 1.9 8.7 ⫾ 2.9
9.7 ⫾ 1.9 8.8 ⫾ 3.1
0.0001 0.0001
excluded from the analysis because no UBM data were available. Thus, 22 MR muscles and 19 LR muscles were used for analysis. The mean distance of horizontal muscle insertions from limbus measured was 9.0 mm with UBM and 9.3 mm at surgery (Table 1). For 33 (80.5%) of the muscles measured, the two methods agreed within one millimeter (P ⬍ 0.0001; Table 1 and Figure 2). In no case was the difference greater than 2 mm. The measurement data for UBM and surgery for MR and LR muscles are listed in Table 1. The differences between means of the 2 measurements were not significant for either muscle. The means of UBM measurements performed under topical anesthesia and general anesthesia did not differ significantly from their respective mean intraoperative measurements (Table 2). The longest horizontal muscle insertion distance from limbus measurable by UBM was 12 mm for the MR and 14 mm for LR.
DISCUSSION This study demonstrates that the 50-MHz UBM can provide accurate quantitative measurement of the horizontal rectus muscle insertion distance when previous surgery has been performed on the muscle. There have been many studies using computed axial tomography, MRI, and cine MRI scans to image the size and location of the EOM in normal eyes and in eyes with diseases such as thyroid orbitopathy.9,10 Although those techniques, especially MRI, provide high-quality images, they cannot provide accurate measurements of the distances of EOM insertion from limbus. In addition, it is not possible to validate these measurements unless EOM surgery is performed on these patients. In a recent study, Tamburrelli et al7 used a 10-MHz conventional B scanner to study the EOM insertion distance in both primary strabismus patients and patients undergoing reoperation for horizontal muscle recessions. They found the 10-MHz B scanner was not particularly accurate for quantitative measurements of the distance of EOM muscle insertions. This result may be attributed to 2 factors. First, they used the estimated location instead of the actual position of the limbus for their measurement of the muscle insertion distance. Second, the resolution of the 10-MHz B scanner was limited (600 m vs 20-60 m of 50-MHz UBM) for imaging the EOM insertions. However, the 50-MHz UBM used for this study has a resolution 10 times higher than that of the 10-MHz conventional B scanner and produces superior image quality of the potential space between the EOM and the sclera,8 which makes the muscle insertion readily identifiable in the image. In 2 or 3 cases, the UBM was able to differentiate pseudotendon (scar tissue found at surgery extending from the true muscle insertion to the original insertion) from the true insertion. The muscle insertion was characterized by a potential space located between the global surface of the muscle and the sclera, and which invariably terminated just behind the insertion into the sclera (Figure 1). In contrast, we found that when we retrospectively reviewed the preoperative images in these cases, the reflectivity of the pseudotendons was not distinguishable from that of overlying soft tissue and underlying sclera, such that there was no potential space that was imaged by the UBM. Therefore, in no cases did the UBM yield a false insertion location in the presence of a pseudotendon. Our main goal was to evaluate the accuracy of the 50-MHz UBM for measuring the horizontal rectus muscle insertion distance from the limbus in those eyes that had previous surgery on the same muscles. The horizontal rectus muscles were selected for this study because data on the horizontal rectus muscle insertions using the 50-MHz UBM had been validated by a previous study. In our practice, there are more cases of additional surgery for residual or consecutive horizontal strabismus than for ver-
Journal of AAPOS Volume 10 Number 3 June 2006
tical strabismus. Concern of recruiting an adequate sample size within the time frame available to us was another contributing factor for this decision. The agreement between the UBM and intraoperative measurements was very good to excellent, with more than 80% of the readings agreeing within our arbitrary limit of 1 mm. However, in no case was the difference in readings between the 2 modalities more than 2 mm. We feel that even this error is not necessarily clinically significant. In this small percentage of cases, the surgeon may need to take more than one pass of the muscle hook to secure the muscle at surgery, but this should not add much time or trauma to the procedure. The maximum EOM insertion distance from limbus measured by 50-MHz UBM was 12 mm for the MR and 14 mm for the LR. The factor limiting the maximum distance of EOM insertion that can be accurately measured is the positioning of the probe. The examiner must ensure that the transducer remains tangential to the plane of the muscle insertion during the measurement. In detecting the position of the MR, the interference by the position of the patient’s nose can limit the range of insertions detectable, which also explains the difference between the MR and LR in the maximum EOM insertion distance measured by UBM in the current study. In one of the studied subjects, neither MR muscle was visible on the UBM examination when the transducer was placed approximately 12 mm from the limbus. At surgery, the MR muscles were found inserted 14 mm from the limbus. Although these data were not included in our analysis, they provided useful information to the surgeon, indicating the horizontal rectus muscle was not within 12 mm from the limbus. The time taken to perform UBM measurement for an individual rectus muscle averaged 10 minutes, compared with the 5 minutes it took to perform the same measurements in patients with primary strabismus.8 This reflects the fact that it took longer to locate the muscle insertion in eyes that had previous surgery when the insertion site was not known. With increasing distance from the limbus, the site becomes progressively more difficult to image, contributing to the increased time required. With the current transducer probe size and less flexible arm, a reliable image of the horizontal rectus muscle insertion cannot be obtained beyond 12 mm and 14 mm from the limbus for the MR and LR, respectively. These ranges may expand in the near future when a smaller transducer probe with a more flexible arm becomes available. The 50-MHz UBM can be a useful tool for the assessment of a patient with a suspected lost or slipped muscle after surgery. One of our studied cases was a 7-year-old girl with limited in adduction of the right eye 2 months after previous MR recession surgery. The referring sur-
Dai et al
205
geon was concerned about a possible slipped or lost right MR muscle. The UBM study performed under topical anesthesia identified the MR 11 mm from limbus, avoiding the need for an MRI scan. The accuracy of the UBM measurement of the horizontal rectus muscle insertion was not affected by the modality of anesthesia. In our study, equal accuracy was achieved for measurements obtained under general and topical anesthesia. One would expect that the muscle relaxant used during general anesthesia might interfere with the accuracy of the UBM measurement because this could potentially reduce the size of the submuscular space and theoretically cause a “false” posterior shift of the muscle insertion. This was not the case in our experience. In conclusion, the 50-MHz UBM can accurately measure the distance from the limbus of a horizontal EOM that has undergone previous surgery. Depending on the patient’s age and level of cooperation, the procedure can be performed under either general or topical anesthesia. In cases where details of the previous surgical procedure are not available, the information provided by the UBM preoperatively can be valuable for the surgeon in planning the site of conjunctival incision, selecting the muscle upon which to operate, and determining the amount of muscle surgery to perform. UBM is also a useful tool for identifying whether EOM may have slipped. The limits within which the muscle can be detected with the current probe size are 12 mm from the limbus for the MR and 14 mm for the LR.
References 1. Helveston E. Reoperation in strabismus. Ophthalmology 1979;86: 1379-88. 2. Ozgen A, Alp MN, Ariyurek M, et al. Quantitative CT of the orbit in Graves’ disease. Br J Radiol 1999;72:757-62. 3. Demer JL, Kerman BM. Comparison of standardized echography with magnetic resonance imaging to measure extraocular muscle size. Am J Ophthalmol 1994;118:351-61. 4. Ettl A, Kramer J, Daxer A, Koornneef L. High-resolution magnetic resonance imaging of the normal extraocular musculature. Eye 1997; 11:793-7. 5. Cadera W, Viirre E, Karlik S. Cine magnetic resonance imaging of ocular motility. J Pediatr Ophthalmol Strabismus 1992;29:120-2. 6. Gillies WE, McIndoe A. The use of ultrasonography in determining the amount of extraocular muscle surgery in strabismus. Aust J Ophthalmol 1982;10:191-4. 7. Tamburrelli C, Salgarello T, Vaiano AS, et al. Ultrasound of the horizontal rectus muscle insertion sites: implications in preoperative assessment of strabismus. Invest Ophthalmol Vis Sci 2003;44:618-22. 8. Watts P, Smith D, Mackeen L, et al. Evaluation of the ultrasound biomicroscope in strabismus surgery. J of AAPOS 2002;6:187-90. 9. Tian S, Nishida Y, Isberg B, Lennerstrand G. MRI measurements of normal extraocular muscles and other orbital structures. Graefes Arch Clin Exp Ophthalmol 2000;238:393-404. 10. Ozgen A, Aydingoz U. Normative measurements of orbital structures using MRI. J Comput Assist Tomogr 2000;24:493-6.
T
From the aDepartment of Ophthalmology and Vision Sciences, The Hospital for Sick Children and the University of Toronto, Toronto, Ontario, Canada bDepartment of Ophthalmology, University of Auckland, Auckland, New Zealand Presented in part as a poster at the 31st Annual Meeting of the American Association for Pediatric Ophthalmology and Strabismus, Orlando, Florida, March 9-13, 2005. Submitted July 1, 2005. Revision accepted December 27, 2005. Reprint requests: Stephen Kraft, MD, Deparment of Ophthalmology & Vision Sciences, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, Canada M5G 1X8 (e-mail: [email protected]). Copyright © 2006 by the American Association for Pediatric Ophthalmology and Strabismus. 1091-8531/2006/$35.00 ⫹ 0 doi:10.1016/j.jaapos.2006.01.209
202
June 2006
high-frequency ultrasound biomicroscope (UBM) was able to reliably estimate the horizontal rectus muscle insertion distance from limbus in the eyes of patients undergoing primary strabismus surgery. The purpose of this study was to determine whether UBM could accurately measure the insertion distance from the limbus of horizontal EOMs in patients that had undergone previous surgery.
METHODS This was a prospective, masked, observational study comparing the distance of the horizontal EOM insertion from the limbus as measured by UBM (Humphrey Instruments, San Leandro, CA) either preoperatively or at the time of the anesthesia with that measured by surgical caliper intraoperatively. The study was conducted between April 2004 and March 2005 at The Hospital for Sick Children in Toronto. This project had the approval of the hospital’s research ethical board, and informed consent was obtained for all the subjects or the guardians of patients who participated in this study. Subjects Twenty-three subjects (11 boys, 12 girls; mean age 16.9 years, range 7-41 years) with a previous history of horizontal muscle surgery and who underwent further surgery on horizontal rectus muscle(s) that were previously operated upon were recruited into the study. Subjects whose eyes had structural abnormalities such as microphthalmia Journal of AAPOS
Journal of AAPOS Volume 10 Number 3 June 2006
FIG 1. UBM image of insertion of rectus muscle and caliper used in measurement. White arrowhead indicates EOM insertion (anterior extent of potential space under muscle) and black arrowhead indicates position of surgical caliper.
or with existing scleral buckles were excluded from the study. UBM Measurements All the UBM measurements were performed by 1 examiner (S.D.). For patients younger than 14 years of age, UBM was performed under general anesthesia in the operating room immediately before the planned surgery but before preoperative preparation of the patient. For patients older than 14 years of age, UBM was performed using topical anesthesia in the outpatient clinic during the preoperative assessment visit. After adequate anesthesia was obtained, a lid speculum was inserted and tear gel was used as coupling agent without the need of a container. The UBM transducer was held tangentially over the muscle to be scanned and moved to and from the limbus, with the oscillations perpendicular to the limbus until a cross section of the muscle insertion was observed. A surgical caliper was introduced, with the tip of one arm fixed at the limbus and the other arm adjusted until the echo of its tip was seen in the UBM image (Figure 1). The echo from this tip of the caliper served as a landmark to facilitate the measurement of the distance of muscle insertion from the limbus, as the horizontal field of view of the 50-MHz UBM was limited to less than 5 mm. The area of the muscle insertion was characterized by a potential space located between the global surface of the muscle and the sclera, and which invariably terminated just behind the insertion into the sclera (Figure 1). The anterior extent of this potential space was identified as the insertion position in the same manner as described in our original UBM study in primary surgeries.8 Three images were obtained for each muscle being measured, and the image with the best quality was used for calculation of the distance of the muscle insertion
Dai et al
203
to the tip of the caliper. This distance between the black and white arrow (Figure 1), in millimeters, was added to the reading on the surgical caliper to produce the measured distance of muscle insertion from the limbus. The UBM screen provides a scale accurate to the nearest 0.1 mm and surgical caliper allowed a measurement to the nearest 0.5 mm. When UBM measurement was performed using topical anesthesia subjects were instructed to abduct the examined eye when the medial rectus (MR) muscle was measured and to adduct the examined eye when the lateral rectus muscle (LR) was measured.8 When UBM was performed under general anesthesia, the examined eye was held in abduction with a fixation forceps for the MR measurement and in adduction for the LR measurement. The UBM measurement was interpreted and stored in the hard drive of the UBM machine before the start of the surgery and the surgeon was masked to the values obtained. Measurements at Surgery During surgery, upon satisfactory exposure of the muscle insertion and after insertion of a muscle hook, the distance from the limbus to the midpoint of the muscle hook was measured by the surgeon (S.P.K., D.R.S., J.R.B.) using the caliper in the same manner as that used for UBM measurements. Measurements were recorded to the nearest 0.5 mm. This information was recorded on a separate data collection sheet designed for the study and was not revealed to the examiner who performed the UBM study (S.D.). We arbitrarily decided that the 2 methods were in agreement if the difference between UBM and intraoperative measurements of a given horizontal rectus muscle insertion distance was no more than 1 mm. The intraoperative surgical caliper measurement was considered as the “gold standard,” and the study was designed to test the proportion of UBM measurements equal to, or within, 1 mm from those obtained intraoperatively (either 1 mm more or less than the reading at surgery). Our null hypothesis was that the difference between the 2 measurement modalities for a given rectus muscle was greater than 1 mm. The data obtained from the two methods were analyzed using paired t-test, and a value of .05 was set as the level of significance.
RESULTS Fourteen subjects were younger than 14 years of age and had the UBM measurement performed under general anesthesia whereas 9 patients older than the age of 14 years had the UBM measurement performed using topical anesthetic. A total of 43 muscles were evaluated, including 24 MR muscles and 19 LR muscles. In one male patient, both of his previously recessed MR muscles could not be detected by the UBM. At surgery, they were located more than 12 mm from the limbus. These two muscles were
Journal of AAPOS Volume 10 Number 3 June 2006
204 Dai et al TABLE 1. EOM insertion distance from limbus measured by UBM and at surgery (mean ⫹ standard deviation, in mm) UBM, ultrasound biomicroscopy Overall (41 muscles) Medial rectus (22) Lateral rectus (19)
UBM
At surgery
p value
9.0 ⫾ 2.3 8.4 ⫾ 1.4 9.7 ⫾ 2.4
9.3 ⫾ 2.5 8.6 ⫾ 1.6 9.8 ⫾ 2.5
0.0001 0.0001 0.0001
Delta (mm)
2
1
0
-1
-2 5
6
7
8
9
10
11
12
13
14
Mean (mm) FIG 2. Scatterplot of “Delta” difference (between UBM and surgical measurements) for all 41 muscles vs. “Mean” (distance to muscle insertion). TABLE 2. UBM measurements under general anesthesia and topical anesthesia (mean ⫹ standard deviation, in mm) UBM, ultrasound biomicroscopy General anesthesia Topical anesthesia
UBM
At surgery
P value
9.3 ⫾ 1.9 8.7 ⫾ 2.9
9.7 ⫾ 1.9 8.8 ⫾ 3.1
0.0001 0.0001
excluded from the analysis because no UBM data were available. Thus, 22 MR muscles and 19 LR muscles were used for analysis. The mean distance of horizontal muscle insertions from limbus measured was 9.0 mm with UBM and 9.3 mm at surgery (Table 1). For 33 (80.5%) of the muscles measured, the two methods agreed within one millimeter (P ⬍ 0.0001; Table 1 and Figure 2). In no case was the difference greater than 2 mm. The measurement data for UBM and surgery for MR and LR muscles are listed in Table 1. The differences between means of the 2 measurements were not significant for either muscle. The means of UBM measurements performed under topical anesthesia and general anesthesia did not differ significantly from their respective mean intraoperative measurements (Table 2). The longest horizontal muscle insertion distance from limbus measurable by UBM was 12 mm for the MR and 14 mm for LR.
DISCUSSION This study demonstrates that the 50-MHz UBM can provide accurate quantitative measurement of the horizontal rectus muscle insertion distance when previous surgery has been performed on the muscle. There have been many studies using computed axial tomography, MRI, and cine MRI scans to image the size and location of the EOM in normal eyes and in eyes with diseases such as thyroid orbitopathy.9,10 Although those techniques, especially MRI, provide high-quality images, they cannot provide accurate measurements of the distances of EOM insertion from limbus. In addition, it is not possible to validate these measurements unless EOM surgery is performed on these patients. In a recent study, Tamburrelli et al7 used a 10-MHz conventional B scanner to study the EOM insertion distance in both primary strabismus patients and patients undergoing reoperation for horizontal muscle recessions. They found the 10-MHz B scanner was not particularly accurate for quantitative measurements of the distance of EOM muscle insertions. This result may be attributed to 2 factors. First, they used the estimated location instead of the actual position of the limbus for their measurement of the muscle insertion distance. Second, the resolution of the 10-MHz B scanner was limited (600 m vs 20-60 m of 50-MHz UBM) for imaging the EOM insertions. However, the 50-MHz UBM used for this study has a resolution 10 times higher than that of the 10-MHz conventional B scanner and produces superior image quality of the potential space between the EOM and the sclera,8 which makes the muscle insertion readily identifiable in the image. In 2 or 3 cases, the UBM was able to differentiate pseudotendon (scar tissue found at surgery extending from the true muscle insertion to the original insertion) from the true insertion. The muscle insertion was characterized by a potential space located between the global surface of the muscle and the sclera, and which invariably terminated just behind the insertion into the sclera (Figure 1). In contrast, we found that when we retrospectively reviewed the preoperative images in these cases, the reflectivity of the pseudotendons was not distinguishable from that of overlying soft tissue and underlying sclera, such that there was no potential space that was imaged by the UBM. Therefore, in no cases did the UBM yield a false insertion location in the presence of a pseudotendon. Our main goal was to evaluate the accuracy of the 50-MHz UBM for measuring the horizontal rectus muscle insertion distance from the limbus in those eyes that had previous surgery on the same muscles. The horizontal rectus muscles were selected for this study because data on the horizontal rectus muscle insertions using the 50-MHz UBM had been validated by a previous study. In our practice, there are more cases of additional surgery for residual or consecutive horizontal strabismus than for ver-
Journal of AAPOS Volume 10 Number 3 June 2006
tical strabismus. Concern of recruiting an adequate sample size within the time frame available to us was another contributing factor for this decision. The agreement between the UBM and intraoperative measurements was very good to excellent, with more than 80% of the readings agreeing within our arbitrary limit of 1 mm. However, in no case was the difference in readings between the 2 modalities more than 2 mm. We feel that even this error is not necessarily clinically significant. In this small percentage of cases, the surgeon may need to take more than one pass of the muscle hook to secure the muscle at surgery, but this should not add much time or trauma to the procedure. The maximum EOM insertion distance from limbus measured by 50-MHz UBM was 12 mm for the MR and 14 mm for the LR. The factor limiting the maximum distance of EOM insertion that can be accurately measured is the positioning of the probe. The examiner must ensure that the transducer remains tangential to the plane of the muscle insertion during the measurement. In detecting the position of the MR, the interference by the position of the patient’s nose can limit the range of insertions detectable, which also explains the difference between the MR and LR in the maximum EOM insertion distance measured by UBM in the current study. In one of the studied subjects, neither MR muscle was visible on the UBM examination when the transducer was placed approximately 12 mm from the limbus. At surgery, the MR muscles were found inserted 14 mm from the limbus. Although these data were not included in our analysis, they provided useful information to the surgeon, indicating the horizontal rectus muscle was not within 12 mm from the limbus. The time taken to perform UBM measurement for an individual rectus muscle averaged 10 minutes, compared with the 5 minutes it took to perform the same measurements in patients with primary strabismus.8 This reflects the fact that it took longer to locate the muscle insertion in eyes that had previous surgery when the insertion site was not known. With increasing distance from the limbus, the site becomes progressively more difficult to image, contributing to the increased time required. With the current transducer probe size and less flexible arm, a reliable image of the horizontal rectus muscle insertion cannot be obtained beyond 12 mm and 14 mm from the limbus for the MR and LR, respectively. These ranges may expand in the near future when a smaller transducer probe with a more flexible arm becomes available. The 50-MHz UBM can be a useful tool for the assessment of a patient with a suspected lost or slipped muscle after surgery. One of our studied cases was a 7-year-old girl with limited in adduction of the right eye 2 months after previous MR recession surgery. The referring sur-
Dai et al
205
geon was concerned about a possible slipped or lost right MR muscle. The UBM study performed under topical anesthesia identified the MR 11 mm from limbus, avoiding the need for an MRI scan. The accuracy of the UBM measurement of the horizontal rectus muscle insertion was not affected by the modality of anesthesia. In our study, equal accuracy was achieved for measurements obtained under general and topical anesthesia. One would expect that the muscle relaxant used during general anesthesia might interfere with the accuracy of the UBM measurement because this could potentially reduce the size of the submuscular space and theoretically cause a “false” posterior shift of the muscle insertion. This was not the case in our experience. In conclusion, the 50-MHz UBM can accurately measure the distance from the limbus of a horizontal EOM that has undergone previous surgery. Depending on the patient’s age and level of cooperation, the procedure can be performed under either general or topical anesthesia. In cases where details of the previous surgical procedure are not available, the information provided by the UBM preoperatively can be valuable for the surgeon in planning the site of conjunctival incision, selecting the muscle upon which to operate, and determining the amount of muscle surgery to perform. UBM is also a useful tool for identifying whether EOM may have slipped. The limits within which the muscle can be detected with the current probe size are 12 mm from the limbus for the MR and 14 mm for the LR.
References 1. Helveston E. Reoperation in strabismus. Ophthalmology 1979;86: 1379-88. 2. Ozgen A, Alp MN, Ariyurek M, et al. Quantitative CT of the orbit in Graves’ disease. Br J Radiol 1999;72:757-62. 3. Demer JL, Kerman BM. Comparison of standardized echography with magnetic resonance imaging to measure extraocular muscle size. Am J Ophthalmol 1994;118:351-61. 4. Ettl A, Kramer J, Daxer A, Koornneef L. High-resolution magnetic resonance imaging of the normal extraocular musculature. Eye 1997; 11:793-7. 5. Cadera W, Viirre E, Karlik S. Cine magnetic resonance imaging of ocular motility. J Pediatr Ophthalmol Strabismus 1992;29:120-2. 6. Gillies WE, McIndoe A. The use of ultrasonography in determining the amount of extraocular muscle surgery in strabismus. Aust J Ophthalmol 1982;10:191-4. 7. Tamburrelli C, Salgarello T, Vaiano AS, et al. Ultrasound of the horizontal rectus muscle insertion sites: implications in preoperative assessment of strabismus. Invest Ophthalmol Vis Sci 2003;44:618-22. 8. Watts P, Smith D, Mackeen L, et al. Evaluation of the ultrasound biomicroscope in strabismus surgery. J of AAPOS 2002;6:187-90. 9. Tian S, Nishida Y, Isberg B, Lennerstrand G. MRI measurements of normal extraocular muscles and other orbital structures. Graefes Arch Clin Exp Ophthalmol 2000;238:393-404. 10. Ozgen A, Aydingoz U. Normative measurements of orbital structures using MRI. J Comput Assist Tomogr 2000;24:493-6.