Objective Evaluation of Inferior Oblique Overaction Using an Interocular Level Difference Gauge

Abstract

Purpose

The conventional method for evaluating inferior oblique overaction (IOOA) relies on subjective, qualitative grading. This study aimed to introduce and perform a pilot evaluation of a novel, objective, and quantitative measurement method for IOOA using a newly designed device, the interocular level difference gauge (IOLDG).

Methods

Sixty-two consecutive patients (124 eyes) with intermittent exotropia, IOOA or superior oblique palsy were recruited. The degree of IOOA was measured using two methods: (1) the conventional subjective grading scale (+0 to +4) by two examiners (A and B); and (2) the objective IOLDG method by two examiners (C and D). The IOLDG utilizes a spectacle-like frame with a 1 mm-spaced horizontal grid to measure the interocular height difference of the inferior corneal margin in millimeters. Interobserver repeatability was assessed using the paired t-test and the intraclass correlation coefficient (ICC).

Results

The conventional method showed a statistically significant difference between examiners (paired t-test, p = 0.002) and moderate repeatability (ICC, 0.749). In contrast, the IOLDG method showed no significant difference between examiners (paired t-test, p = 0.064) and demonstrated excellent repeatability (ICC, 0.903; p < 0.001). The correlation analysis established quantitative equivalents for the conventional grades: grade +1 corresponded to 1.04 ± 0.75 mm, grade +2 to 1.75 ± 0.60 mm, and grade +3 to 2.81 ± 0.75 mm.

Conclusions

The IOLDG provides a highly repeatable, objective, and quantitative measurement for IOOA, largely independent of the examiner’s subjective judgment. This simple, chair-side method offers superior reliability compared to the conventional grading scale.

Inferior oblique overaction (IOOA) is a common clinical finding in strabismus patients, defined as the excessive upward rotation (elevation) of the eye during adduction [1–4]. IOOA, which may be unilateral or bilateral, is frequently associated with horizontal strabismus, particularly intermittent exotropia, and is a major consideration in surgical planning for both comitant and incomitant deviations [5–7]. The accurate and reliable assessment of the degree of IOOA is crucial for determining the appropriate surgical procedure and dosage to achieve successful and lasting alignment [8–10].

Traditionally, the severity of IOOA has been quantified using a subjective, observational grading system, typically ranging from grade +1 to +4 [2,7]. This conventional method relies on the clinician’s subjective visual estimation of the degree of hyperdeviation in adduction, often referenced against standard illustrations. While widely accepted, this method inherently suffers from poor interobserver reliability and high variability [11–13], as it is nonquantitative and heavily dependent on the individual examiner’s judgment. Furthermore, this method provides only qualitative data (grades) rather than a precise, quantitative measurement (e.g., in millimeters or prism diopters), which limits its utility for objective surgical dose-response correlations and rigorous clinical research [11,13,14].

To overcome the limitations of subjective assessment, this study proposes a novel, objective, and quantitative method for evaluating IOOA using a newly designed instrument, the interocular level difference gauge (IOLDG). This device is designed to measure the vertical deviation between the eyes in millimeters, providing a standardized, objective metric for IOOA. We aimed to present a new method using an IOLDG and to compare its performance with that of the conventional subjective grading system in clinical cases.

Materials and Methods

Ethics statement

The study adhered to the tenets of the Declaration of Helsinki and was approved by the Institutional Review Board of Chungnam National University Hospital (No. 2025-10-048). Informed consent was waived due to the use of deidentified data and the retrospective nature of the study.

Subjects

This study included cooperative patients aged 4 years or older with intermittent exotropia or primary IOOA or superior oblique palsy who visited the Strabismus Clinic at Chungnam National University Hospital from July 2018 to August 2018. Patients with concurrent strabismus other than the primary diagnosis or a best-corrected visual acuity of less than 0.1 were excluded. A total of 62 patients (124 eyes) were consecutively recruited for the study.

IOOA measurement methods

All recruited patients were assessed for the degree of IOOA using two distinct methods: the conventional subjective grading method and the objective measurement method using the IOLDG.

1) Conventional grading method

IOOA was graded by two independent examiners (examiner A and B) who are experienced strabismus specialists, according to the conventional scale. The degree of IOOA was classified as +0 for a normal eye (no hyperdeviation in adduction), +1 for slight overaction (hyperdeviation appearing only on attempted upgaze in adduction), +2 for moderate overaction (slight hyperdeviation present in primary gaze and becoming more evident in adduction), +3 for marked overaction (definite hyperdeviation in adduction), and +4 for severe overaction (large hyperdeviation in adduction).

2) New objective measurement method (IOLDG)

The IOLDG is a spectacle-like instrument constructed with an aluminum alloy frame. The device features a set of horizontally placed monofilament nylon strings, spaced 1 mm apart, which serve as a precise horizontal reference grid. The strings were colored in a sequence, ensuring that strings at the same vertical level on both frames share the same color, facilitating the easy comparison of interocular height differences. A central acrylic septum was installed between the two frames to prevent the eye being measured from fixating the target during the deviation test (Fig. 1A, 1B).

Fig. 1.

The interocular level difference gauge (IOLDG). (A) Frontal view of the IOLDG. The device features an aluminum alloy frame with precisely spaced (1 mm apart) horizontal monofilament nylon strings. These strings are colored to facilitate easy differentiation of vertical levels between the two frames. A central acrylic septum is installed to prevent fixation by the eye being measured. (B) Side view of the IOLDG, illustrating the spectacle-like design and the position of the horizontal grid.

3) Measurement procedure

The measurement was performed by two independent examiners (examiners C and D) following a systematic, chair-side procedure. Initially, the patient was seated, and the IOLDG was placed on their face. The examiners first established the horizontal reference plane by carefully adjusting the instrument’s horizontal tilt until the corneal reflex of both eyes was visually aligned with a string of the same color. The patient was then instructed to maintain binocular fixation on a target positioned 50 cm away. While fixation was maintained, the target was moved maximally into the field of inferior oblique action, specifically 45 degrees superior and temporal to the fixing eye. As the fixing eye followed the target, the eye being measured deviated superiorly and nasally (up and in) due to the induced IOOA. Critically, the central acrylic septum prevented the measured eye from fixating the target, ensuring the measurement was taken in a dissociated state. Finally, the height difference between the inferior corneal margins (inferior limbus) of the two eyes was objectively quantified in millimeters (mm) by counting the difference in string levels, with color differences used to facilitate accurate visualization (Fig. 2).

Fig. 2.

Objective measurement of inferior oblique overaction (IOOA) using the interocular level difference gauge (IOLDG). The photograph illustrates the objective measurement procedure during the patient fixates on a target positioned 45 degrees superior and temporal to the fixing eye (right eye in this image). The IOLDG, featuring a 1 mm-spaced horizontal grid, is aligned to ensure the corneal reflexes are positioned on the same color-coded string. In this dissociated position, the height difference between the inferior corneal margins (inferior limbus) of the fixing eye (abducted) and the measured eye (adducted and elevated) is quantified in millimeters (mm) by counting the difference in string levels based on the inferior limbus of both eyes. The patient provided written informed consent for publication of the clinical image.

Statistical analysis

Statistical analysis was performed using PASW SPSS ver. 18.0 (IBM Corp.). The interobserver variability between the two examiners within each method (conventional, A vs. B; new, C vs. D) was assessed using the paired t-test, with a p-value of <0.05 considered statistically significant. The repeatability of each method was evaluated using the intraclass correlation coefficient (ICC). A one-way analysis of variance (ANOVA) was used to determine the corresponding quantitative measurement (in mm) of the IOLDG method for each conventional grade (+0 to +3) assigned by examiner A, who is a strabismus specialist.

Results

A total of 124 eyes from 62 patients were analyzed. The cohort consisted of 29 male patients (46.8%) and 33 female patients (53.2%), with a mean age of 10.29 ± 5.07 years.

Comparison of conventional method

Examiners A and B evaluated IOOA using the conventional grading system. The distribution of grades was as follows: grade 0, 58 eyes (A) and 68 eyes (B); grade +1, 40 eyes (A) and 42 eyes (B); grade +2, 18 eyes (A) and 12 eyes (B); grade +3, 8 eyes (A) and 2 eyes (B); and grade +4, 0 eyes (A) and 0 eyes (B) (Fig. 3). The mean grade was 0.80 ± 0.93 for examiner A and 0.58 ± 0.73 for examiner B. The paired t-test revealed a statistically significant difference between the two examiners (p = 0.002), indicating poor interobserver agreement. The ICC for the conventional method was 0.749 (Table 1).

Fig. 3.

Distribution of inferior oblique overaction (IOOA) grading by conventional method (examiners A and B).

Table 1.

Interobserver variability and repeatability of the conventional and new measurement methods (IOLDG), assessed by paired t-test and ICC

Method	Grade	p-value*	ICC (95% CI)‡
Conventional method		0.002†	0.749 (0.645–0.819)
Examiner A	0.80 ± 0.93
Examiner B	0.58 ± 0.73
IOLDG		0.064	0.903 (0.857–0.935)
Examiner C	0.94 ± 0.97
Examiner D	0.85 ± 0.97

Values are presented as mean ± standard deviation.

IOLDG = interocular level difference gauge; ICC = intraclass correlation coefficient; CI = confidence interval.

^*Paired samples t-test was used for within-group comparisons;

^†Statistically significant (p < 0.05);

^‡Interpretation of ICC values: poor, <0.50; moderate, 0.50 to 0.75; good, 0.75 to 0.90; and excellent, >0.90.

Evaluation of new objective method (IOLDG)

Examiners C and D assessed the IOOA using the IOLDG, providing measurements in millimeters. The distribution of measurements was as follows: 0 mm, 50 eyes (C) and 58 eyes (D); 1 mm, 41 eyes (C) and 37 eyes (D); 2 mm, 24 eyes (C) and 20 eyes (D); 3 mm, 8 eyes (C) and 8 eyes (D); and 4 mm, 1 eye (C) and 1 eye (D) (Fig. 4). The mean measurement was 0.94 ± 0.97 mm for examiner C and 0.85 ± 0.97 mm for examiner D. The paired t-test showed no statistically significant difference between the two examiners (p = 0.064). Crucially, the ICC for the new method was 0.903 (p < 0.001), indicating high interobserver repeatability (Table 1).

Fig. 4.

Distribution of inferior oblique overaction measurements by new method using the interocular level difference gauge (examiners C and D).

Correlation between methods

The objective measurements obtained by the new method were correlated with the conventional grades assigned by the strabismus specialist (examiner A). This analysis established quantitative equivalents for the subjective IOOA grades using IOLDG measurements. The corresponding average millimeter measurements for each conventional grade were as follows: grade 0, 0.27 ± 0.53 mm; grade +1, 1.04 ± 0.75 mm; grade +2, 1.75 ± 0.60 mm; and grade +3, 2.81 ± 0.75 mm (Table 2). The measurements exhibited a clear, sequential increase, ranging from 0.27 mm to 2.81 mm, corresponding to the grade rising from 0 to +3.

Table 2.

Corresponding quantitative measurements of the new objective method (IOLDG; examiners C and D and their average) to the grades of conventional IOOA grading method (examiner A)

IOOA grade (examiner A)	Measurement (mm)
	Examiner C	Examiner D	Average
0	0.27 ± 0.56	0.26 ± 0.51	0.27 ± 0.53
+1	1.09 ± 0.75	0.98 ± 0.74	1.04 ± 0.75
+2	1.82 ± 0.62	1.68 ± 0.59	1.75 ± 0.60
+3	2.81 ± 0.75	2.81 ± 0.75	2.81 ± 0.75

Values are presented as mean ± standard deviation. This table establishes the correlation between the conventional subjective IOOA grades (grade 0 to +3) and the objective interocular height difference measurements (mm) obtained using the IOLDG.

IOLDG = interocular level difference gauge; IOOA = inferior oblique overaction.

Discussion

The surgical management of strabismus is critically dependent on the accurate and reliable assessment of preoperative measurements [8–10]. IOOA is a key diagnostic factor, yet its conventional assessment remains a clinical challenge due to its inherent subjectivity [5,7]. The definitions of the conventional grading scale are based on vague, nonquantitative descriptors, leading to the risk of subjective interpretation and low interobserver agreement [11,13,15]. Our findings corroborate the limitations of this traditional grading system, showing a statistically significant difference between two experienced examiners (p = 0.002) and yielding a moderate ICC of 0.749 (Table 1). This moderate reliability aligns with previous literature highlighting the inherent variability in subjective strabismus measurements.

Previous research has attempted to overcome this subjectivity by developing quantitative methods. Lim et al. [11] utilized frontal gaze photographs for quantitative analysis of IOOA. While demonstrating high interobserver reliability (ICC, 0.986), their method was reported to be difficult to apply in young, noncooperative children and in patients with small palpebral fissures, which can obscure anatomical landmarks necessary for measurement. Similarly, Yoon et al. [15] proposed an objective method that analyzed the corneal contour from nine-gaze photographs to determine the angular difference between the two eyes. However, this photographic approach had limitations, including the possibility that nine-gaze photographs might not fully capture the complete extent of the muscle dysfunction, potentially requiring repeated image capture to correct for positional errors. Lou et al. [14] quantitatively analyzed IOOA using a deep learning-based automatic photographic analysis system and demonstrated high reliability (ICC = 0.975 between automated and manual measurements) and reproducibility (ICC = 0.998 for repeated automated measurements). However, the method faced limitations in full automation, as automatic ellipse fitting was difficult when corneal exposure was insufficient, requiring human supervision. This requirement for human intervention in severe IOOA cases compromises the system’s objective nature, introducing a degree of subjective error back into the measurement process. Crucially, a common disadvantage shared by these photographic methods is the requirement for specialized software analysis and post-processing, making them inconvenient and time-consuming for immediate, chair-side use in a busy clinical environment [11,13,14].

The IOLDG was specifically designed to provide a standardized, chair-side, and quantitative solution. By employing a precisely spaced (1 mm) horizontal reference grid and a standardized measurement landmark (the inferior corneal margin), the IOLDG transforms the assessment of IOOA into a quantitative metric (mm) that is immediately available. The validation results strongly support this objective approach: the paired t-test showed no statistically significant difference between the two examiners using the IOLDG (p = 0.064), and the ICC was notably high at 0.903 (p < 0.001). An ICC value exceeding 0.90 indicates excellent interobserver reliability, suggesting that the IOLDG provides a highly repeatable measurement largely independent of the examiner’s experience or subjective judgment (Table 1). While a precise fixation instrument was not utilized, its future integration is anticipated to further optimize the already high precision and consistency demonstrated by the current measurement protocol. This is a clear improvement in practicality over previous methods that relied on external image processing. Furthermore, unlike the photographic methods, the IOLDG does not appear to exhibit the same limitations regarding difficulty in measuring patients with small palpebral fissures or requiring repeated tests to correct for measurement errors.

This study also established a strong correlation between the conventional subjective grades and the new objective millimeter measurements. The determined correspondences (e.g., grade 0 to approximately 0.27 mm, grade +1 to approximately 1.04 mm, grade +2 to approximately 1.75 mm, and grade +3 approximately 2.81 mm) provide a quantitative reference bridge, which is invaluable for standardizing the existing grading system and applying objective metrics to surgical planning.

Despite these advances, the current study has several limitations. First, the patient cohort predominantly exhibited mild degrees of IOOA, with severe cases of grade +4 being absent and grade +3 cases being limited. Future studies should include a larger number of patients with higher-grade IOOA to ensure the generalizability of the IOLDG across the full spectrum of muscle dysfunction. Second, the study did not assess intra-observer repeatability, a crucial metric for evaluating the consistency of a single examiner over time. Third, by using separate examiner groups for the two methods, we were unable to perform a direct within-examiner comparison of the two measurement techniques. While this design minimizes potential cross-method bias, future studies could explore a crossover design to compare the methods within the same observer. Fourth, for statistical comparison, conventional IOOA grading, an ordinal scale, was treated as an interval scale, a common but recognized limitation in strabismus research. We did not use the Mann-Whitney U-test because it is designed for independent samples, whereas our data involved paired, nonindependent observations from two examiners (A vs. B) on the same eyes (n = 124). Although the Wilcoxon signed-rank test is statistically rigorous for paired ordinal data, we adhered to the convention of using the paired t-test to assess interobserver differences, as is frequently reported in strabismus literature. We acknowledge this treatment of ordinal data remains a methodological limitation.

In conclusion, the IOLDG represents a practical and significant advancement in the objective assessment of IOOA. This novel method is objective, quantitative, and demonstrates high interobserver repeatability. Its simplicity and independence from the examiner’s subjective skill or external software analysis suggest that the IOLDG can be readily and consistently adopted in clinical practice for more precise diagnosis and standardized surgical planning.