Types of Surgery Performed and Reoperation Rate for Congenital Superior Oblique Palsy: a Claims Database Study

Abstract

Purpose:

To investigate types of surgeries performed to treat a presumed congenital superior oblique palsy (SOP) and the reoperation rate.

Methods:

This was a population-based retrospective cohort study using claims data from the United States. Patients who underwent strabismus surgery for a presumed congenital SOP with ≥ 3 months of continuous enrollment after the initial surgery were included. We investigated age, surgical methods, and the time interval between the initial surgery and reoperation. The hazard ratios for reoperation were estimated according to the surgical methods using Cox regression analysis.

Results:

A total of 3,998 patients underwent surgery for presumed congenital SOP; 2,981 (74.6%) on only one vertical muscle (excluding superior oblique). Reoperation was performed on 427 patients (10.7%). Compared to patients who underwent unilateral surgery on one vertical muscle (excluding superior oblique muscle), patients who underwent surgery that included the superior oblique muscle (unilateral 2.08; 95% CI, 1.61–2.67, p < 0.001; bilateral 2.44; 95% CI, 1.40–4.28, p = 0.002) and two or more vertical muscles (excluding the superior oblique muscle) (unilateral 2.99; 95% CI, 2.00–4.49, p < 0.001; bilateral 1.68; 95% CI, 1.23–2.28, p = 0.001) had increased hazard ratios for reoperation. The median period between the initial surgery and reoperation was 168.0 [Q1–Q3 84.0–407.8] days and negatively correlated with patient age at initial surgery (r = −0.199, p < 0.001).

Conclusion:

The reoperation rate for presumed congenital SOP was 10.7%. Patients who underwent surgery on two or more vertical muscles or the superior oblique muscle had an increased risk of reoperation.

INTRODUCTION

Superior oblique palsy (SOP) is the most common palsy of the cyclovertical muscles (Knapp 1971). Most cases are believed to be congenital. Children with congenital SOP often have a noticeable hypertropia and adopt a compensatory head tilt. Adults with congenital SOP are often diagnosed after they develop vertical diplopia or asthenopia. Decompensation of a congenital SOP is believed to be due to the progressive breakdown of vertical fusional control (Mansour & Reinecke 1986). While some patients may be successfully treated with prismatic correction, strabismus surgery is often required due to the incomitance of the deviation and the presence of excyclotorsion. Patients with a SOP secondary to trauma or a microvascular event may recover spontaneously (Khaier, Dawson & Lee 2012, Mollan et al. 2009).

In most cases of presumed congenital SOP, the exact pathophysiology is ambiguous. Studies using high-definition magnetic resonance imaging have suggested two different pathogenic mechanisms. In some patients, the trochlear nerve is absent, and the superior oblique muscle is atrophied. This condition has been grouped by some as a type of congenital cranial dysinnervation syndrome. In other patients, the trochlear nerve is normal, but the tendon of the superior oblique is presumed to be abnormal (Lee et al. 2014, Yang, Kim & Hwang 2012). In congenital SOP, the large patient-to-patient variations in vertical fusion amplitudes may lead to unpredictable surgical outcomes. In contrast, patients with acquired SOP typically have a smaller range of vertical fusion amplitudes. Most previous studies of SOP included both congenital and acquired cases because these studies were conducted at a single center with a limited number of patients (Helveston et al. 1996, Knapp 1974, Nash et al. 2017, Simons et al. 1998).

Many options exist for the surgical treatment of congenital SOP (Khawam, Scott & Jampolsky 1967, Knapp 1974). However, the reoperation rate in a large cohort of patients with presumed congenital SOP has not been previously investigated.

We conducted this study to investigate types of surgeries performed to treat a presumed congenital SOP and the reoperation rate using two United States different claims databases.

METHODS

Study design and data source

This population-based retrospective cohort study was performed using claims from the Optum SES Medical Claims dataset from Optum’s De-identified Clinformatics^® Data Mart (CDM) (2003–2020) and IBM^® MarketScan^® Research Databases (2007–2016) that were analyzed separately. The CDM Database is a de-identified commercial and Medicare Advantage claims database. The dataset includes claims information on approximately 15 to 18 million patients annually covered lives, from a geographically diverse population in the United States spanning all 50 states. The Optum dataset from CDM provides demographic and medical claims data for inpatient and outpatient services including surgery. The MarketScan Databases include more than 240 million patients insured by 350 unique health carriers. These databases include health insurance claims across the continuum of care (inpatient, outpatient, outpatient pharmacy, and carve-out behavioral healthcare) and enrollment data from large employers and health plans across the United States. Both medical claims datasets include the International Classification of Disease, 9th and 10th revisions (ICD-9-CM and ICD-10-CM, respectively), diagnosis codes, and Current Procedural Terminology (CPT) Version 4 procedure codes.

Data access for this project was provided by the Stanford Center for Population Health Science Data Core, which is supported by the National Institutes of Health National Center for the Advancing Translational Science Clinical and Translational Science Award (UL1 TR001085) and internal Stanford funding. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The Stanford University School of Medicine Institutional Review Board (IRB) determined that this study did not require IRB approval because all data were deidentified.

Study population

Patients were included in the study if they had been diagnosed with SOP and underwent vertical extraocular muscle surgery (CPT codes: 67314, one vertical muscle excluding superior oblique; 67316, two or more vertical muscles excluding superior oblique; 67318, superior oblique muscle; modifier code 50, bilateral surgery). We only included patients who had continuous enrollment for ≥ 3 months from the time of initial vertical muscle surgery for SOP. Exclusion criteria encompassed the following: patients with ICD codes for mechanical strabismus, Brown syndrome, Duane retraction syndrome, thyroid-associated ophthalmopathy, myasthenia gravis, and oculomotor or abducens nerve palsy. To exclude patients with traumatic superior oblique palsy, patients with ICD codes for traumatic brain injuries and fractures of skull, orbit or facial bones within 6 months before they had ICD codes for SOP were excluded. To exclude patients with superior oblique palsy secondary to microvascular disease or other reasons, patients with ICD codes for hypertension, diabetes mellitus, ischemic heart disease, peripheral vascular disease, cerebrovascular disease, transient cerebral ischemia, neoplasm of brain or eye, and inflammatory or infectious disease of central nervous system within 6 months before they had ICD codes for SOP were also excluded. Patients with reoperation CPT codes 67332 in the initial surgery for SOP and the patients who underwent another strabismus surgeries before the diagnosis for SOP were also excluded (Table 1 and Figure 1). The following data were collected separately from the databases: date, sex, and age at the initial surgery and reoperation, surgical methods, those who underwent combined horizontal muscle surgery (CPT codes: 67311 and 67312), and continuous enrollment period after the initial surgery.

Table 1.

International classification of disease (ICD) codes used to identify patients

Diseases	ICD −9 codes	ICD-10 codes
Included
Superior oblique palsy	378.53	H49.1, H49.1x
Excluded
Mechanical strabismus	378.6, 378.60, 378.62, 378.63	H50.6, H50.60, H50.69
Brown syndrome	378.61	H50.61, H50.611, H50.612
Duane retraction syndrome	378.71	H50.81, H50.811, H50.812
Thyroid associated ophthalmopathy	242.00, 242.01, 242.90	E05.00, E05.01, E05.09
Myasthenia gravis	358.0, 358.00, 358.01	G70.0, G70.00, G70.01
Oculomotor nerve palsy	378.51, 378.52	H49.0, H49.0x
Abducens nerve palsy	378.54	H49.2, H49.2x
Traumatic brain injuries / Fracture of skull, orbit or facial bones	310.2, 800, 800.x, 800.xx, 801, 801.x, 801.xx, 802, 802.x, 802.xx, 803, 803.x, 803.xx, 804, 804.x, 804.xx, 850, 850.x, 850.11, 850,12, 851, 851.x, 851.xx, 852, 852.x, 852.xx, 853, 853.x, 853.xx, 854, 854.x, 854.xx, 959.0, 959.01, 995.55	S06, S06.x, S06.xx, S06.xxx, S06.xxxx, S07.1, S07.1xxx, S09.9, S09.9x, S09.9xxx, T74.4, T74.4xxx, S02.0, S02.0xxx, S02.1, S02.1x, S02.1xx, S02.1xxx, S02.3, S02.3x, S02.3xx, S02.3xxx, S02.4, S02.4x, S02.4xx, S02.4xxx, S02.8, S02.8x, S02.8xx, S02.8xxx, S02.9, S02.9x, S02.9xxx,
Hypertension	401, 401.x, 402, 402.x, 402.xx, 403, 403.x, 403.xx, 404, 404.x, 404.xx, 405, 405.x, 405.xx	I10, I11, I11.x, I12, I12x, I13, I13.x, I13.xx, I15, I15.x, I16, I16.x
Diabetes mellitus	250, 250.x, 250.xx	E08, E08.x, E08.xx, E08.xxx, E08.xxxx, E09, E09.x, E09.xx, E09.xxx, E09.xxxx, E10, E10.x, E10.xx, E10.xxx, E10.xxxx, E11, E11.x, E11.xx, E11.xxx, E11.xxxx, E13, E13.x, E13.xx, E13.xxx, E13.xxxx
Ischemic heart disease	410, 410.x, 410.xx, 411, 411.x, 411.xx, 412, 413, 413.x, 414, 414.x, 414.xx	I20, I20x, I21, I21.x, I21.xx, I22, I22.x, I23, I23.x, I24, I24.x, I25, I25.x, I25.xx, I25.xxx
Peripheral vascular disease	440, 440.x, 440.xx, 443, 443.x, 443.xx, 447, 447.x, 447.xx, 448, 448.x	I70, I70.x, I70.xx, I70.xxx, I73, I73.x, I73.xx, I78, I78.x, I79, I79.x
Cerebrovascular disease	430, 431, 432, 432.x, 433, 433.x, 433.xx, 434, 434.x, 434.xx, 435, 435.x, 436, 437, 437.x, 438, 438.x, 438.xx	I60, I60.x, I60.xx, I61, I61.x, I62, I62.x, I62.xx, I63, I63.x, I63.xx, I63.xxx, I65, I65.x, I65.xx, I66, I66.x, I66.xx, I67, I67.x, I67.xx, I67.xxx, I68, I68.x, I69, I69.x, I69.xx, I69.xxx
Transient cerebral ischemia	435, 435.8, 435.9	G45.8, G45.9
Neoplasm of brain or eye	190, 190.x, 191, 191.x, 192, 192.x, 198.3, 224, 224.x, 225, 225.x,	C69, C69.x, C69.xx, C70, C70.x, C71, C71.x, C71.x, C72, C72.x, C72.xx, C79.x, C79.xx, D32, D32.x, D33, D33.x, D42, D42.x, D43, D43.x
Inflammatory or infectious disease of central nervous system	046, 046.x, 046.xx, 047, 047.x, 048, 049, 049.x	G00, G00.x, G01, G02, G03, G03.x, G04, G04.x, G04.xx, G05, G05.x, G06, G06.x, G07, G09, A80, A80.x, A80.xx, A81, A81.x, A81.xx, A82, A82.x, A83, A83.x, A84, A84.x, A84.xx, A85, A85.x, A86, A87, A87.x, A88, A88.x, A89

Figure 1.

Flow diagram illustrating the path taken to identify patients who underwent surgery for presumed congenital superior oblique palsy.

Statistical analysis

The chi-square test was used for comparison of sex, age, surgical methods, reoperation rate, and combined horizontal muscle surgery between patients who underwent reoperation and those who did not. This was followed by a secondary post hoc analysis for surgical methods. The Mann–Whitney U test was used to compare age at the time of the initial surgery, the time period between the initial surgery and subsequent reoperations, and the continuous enrollment period after the initial surgery between the two groups. Cox proportional hazards regression analyses were performed to evaluate the effects of sex, age, combined horizontal muscle surgery, and surgical methods at the initial surgery on reoperation. The Pearson correlation coefficient between the patient age at initial surgery and the period between the initial surgery and reoperation was calculated. A p-value of < 0.05 was considered statistically significant. All statistical analyses were conducted using R software (version 4.0.0).

RESULTS

A total of 7,601,339,218 claims for 66,228,965 patients were analyzed in the Optum SES Medical Claims dataset (version 4.0) between 2003 and 2020. Among these patients, 1,540 patients underwent surgery for presumed congenital SOP with a continuous postoperative enrollment period ≥ 3 months. A total of 8,713,134,185 claims for 123,637,719 patients were analyzed in the MarketScan Databases (version 2.0) and 2,458 of these patients underwent surgery for presumed congenital SOP with a continuous postoperative enrollment period ≥ 3 months. A total of 3,998 patients were included in the analysis.

The median age of the patients studied was 19.0 [Q1–Q3 6.0–46.0] years and 44.2% of the patients were female. The median continuous enrollment period after the initial surgery was 22.0 [Q1–Q3 10.0–42.0] months. Surgery was performed on only one vertical muscle (excluding the superior oblique) in 2,981 patients (74.6%). Surgery was performed on the superior oblique muscle alone or combined with one or more vertical muscles in 508 patients (12.7%). Surgery was performed on both horizontal and vertical muscles in 666 patients (16.7%). Of the 3.998 patients included in this analysis, 427 (10.7%) underwent one or more reoperations. Reoperations were performed more often on patients who underwent unilateral surgery on two or more vertical muscle (excluding superior oblique muscle) and unilateral surgery on the superior oblique muscle than on patients who underwent unilateral surgery on one vertical muscle (Table 2).

Table 2.

Demographics and characteristics of patients who underwent surgery for a presumed congenital superior oblique palsy

	Total patients (n = 3,998)	Patients undergoing a single operation (n = 3,571, 89.3%)	Patients undergoing ≥2 operations (n = 427, 10.7%)	p-value†
Sex (n (%))				0.41
Female	1,769 (44.2%)	1,588 (44.5%)	181 (42.4%)
Male	2,229 (55.8%)	1,983 (55.5%)	246 (57.6%)
Age (yrs) *	19.0 [6.0;46.0]	20.0 [6.0;46.0]	16.0 [3.0;47.0]	0.002
Surgical methods				<0.001
Unilateral
One vertical muscle (excluding SO muscle)	2,981 (74.6%)	2,723 (76.3%)	258 (60.4%)‡
Two or more vertical muscles (excluding SO muscle)	117 (2.9%)	91 (2.5%)	26 (6.1%)‡
SO muscle ± other vertical muscle(s)	439(11.0%)	360 (10.1%)	79 (18.5%)‡
Bilateral
Two or more vertical muscles (excluding SO muscle)	392 (9.8%)	341 (9.5%)	51 (11.9%)
SO muscle ± other vertical muscle(s)	69 (1.7%)	56 (1.6%)	13 (3.0%)
Continuous enrollment period after the initial surgery (months) *	22.0 [10.0;42.0]	21.0 [10.0;41.0]	29.0 [15.0;51.0]	<0.001

SO = superior oblique

^*Median value [quartile 1;quartile 3]

^†Comparison between patients undergoing a single operation and patients undergoing ≥2 operations

^‡Statistical significance by Bonferroni post-hoc tests

Table 3 shows the hazard ratios for reoperation. There were no differences in sex, age, and combined horizontal and vertical muscle surgery. Compared to patients who underwent unilateral surgery on one vertical muscle (excluding superior oblique muscle), patients who underwent surgery that included the superior oblique muscle (unilateral 2.08; 95% CI, 1.61–2.67, p < 0.001; bilateral 2.44; 95% CI, 1.40–4.28, p = 0.002) and two or more vertical muscles (excluding the superior oblique muscle) (unilateral 2.99; 95% CI, 2.00–4.49, p < 0.001; bilateral 1.68; 95% CI, 1.23–2.28, p = 0.001) had increased hazard ratios for reoperation.

Table 3.

Multivariate Cox regression analysis evaluating the effect of surgical methods on reoperation rate.

	Hazard ratios (95% CI)	p-value
Sex
Female	0.91 (0.75–1.11)	0.35
Male	1.0 (ref)
Age	1.00 (0.99–1.00)	0.11
Surgical methods
Unilateral
One vertical muscle (excluding SO muscle)	1.0 (ref)
Two or more vertical muscles (excluding SO muscle)	2.99 (2.00–4.49)	<0.001
SO muscle ± other vertical muscle(s)	2.08 (1.61–2.67)	<0.001
Bilateral
Two or more vertical muscles (excluding SO muscle)	1.68 (1.23–2.28)	0.001
SO muscle ± other vertical muscle(s)	2.44 (1.40–4.28)	0.002
Combined vertical and horizontal muscle surgery	0.80 (0.61–1.06)	0.12

CI = confidence interval; SO = superior oblique

The median period between the initial surgery and reoperation was 168.0 [Q1–Q3 84.0–407.8] days and the length of time between surgeries negatively correlated with patient age at the time of the initial surgery (r = −0.199, p < 0.001).

DISCUSSION

In this claims-based analysis, approximately 75% of patients with congenital SOP underwent unilateral surgery on only one vertical muscle other than the superior oblique muscle. Although no CPT code exists explicitly for inferior oblique muscle surgery, we presumed this one vertical muscle surgery was an inferior oblique weakening procedure in most cases. The reoperation rate was 10.7%. Age and combined horizontal and vertical muscle surgery did not affect the reoperation rate. Patients who underwent unilateral or bilateral surgeries including surgery on the superior oblique muscle, or two or more vertical muscle excluding the superior oblique muscle had increased hazard ratios for reoperation.

Previous studies that included patients of all ages showed that the most common type of isolated SOP was congenital, followed by palsies secondary to trauma and microvascular diseases (Dosunmu et al. 2018, von Noorden, Murray & Wong 1986). Isolated SOP secondary to intracranial tumors is rare as these tumors typically affect multiple cranial nerves (Richards, Jones & Younge 1992, Rush & Younge 1981). We only included patients with presumed congenital SOP (including late decompensation) in our analysis and excluded patients with ICD codes for intracranial disease and microvascular, traumatic and restrictive strabismus. The median age of presentation in our series was 19.0 years. Others have reported a similar age of presentation for congenital SOP. In a series of 89 patients with congenital SOP from Thailand, the mean age of presentation was 24.1 years. While most of the patients in this series presented during the first decade of life, 20% had an age of presentation when they were 50 years of age or older presumably because of a late decompensation of their congenital SOP (Lekskul, Wuthisiri & Tangtammaruk 2021). Similarly, the mean age of presentation of 231 patients with SOP from the United States was 24.7 years (von Noorden, Murray & Wong 1986).

There are various options for the surgical treatment of SOP and most surgeons individualize their treatment plan based on the range of under- and overaction of the oblique muscles, forced ductions, and the pattern of deviation in various gaze positions (Khawam, Scott & Jampolsky 1967, Knapp 1974). Particularly important in formulating this plan is the degree of inferior oblique overaction, superior oblique tendon laxity, and superior rectus contracture (Plager 1999). If the hyper-deviation is greater than 15–20 prism diopter (PD) in primary position, surgeons often operate on two muscles; either two in one eye or one in both eyes (Knapp 1974, Nejad et al. 2013). For patients with relatively symmetric bilateral SOP, both inferior oblique muscles may be weakened or both superior obliques tendons may be tightened. If there is symptomatic excyclotorsion with minimal hypertropia in side gaze, bilateral superior oblique strengthening procedures may be performed (Price et al. 1987). Our results show that most patients with congenital SOP (74.6%) underwent surgery on only one vertical muscle in one eye, presumably an inferior oblique weakening procedure, and 12.7% underwent surgery on the superior oblique muscle, presumably because there was tendon laxity.

Aoba K and coworkers (Aoba et al. 2015) investigated clinical factors affecting the reoperation rate in 212 patients with congenital SOP after one muscle surgery. They reported an increased risk of reoperation for larger vertical deviations, cyclotorsional deviations, and younger age. Nash and coworkers (Nash et al. 2017) compared 1- vs 2-muscle surgery for moderate-angle hyper-deviation due to presumed unilateral SOP. In 46 patients with 1-year outcome data, reoperations were performed in 8/29 (28%) patients undergoing 2-muscle surgery and 3/17 (18%) patients undergoing 1-muscle surgery. The contributions of under or overcorrection after the initial surgery to reoperation were not reported, but most overcorrected patients underwent 2-muscle surgery. In our study, patients who underwent surgeries including superior oblique muscle or two or more vertical muscle excluding the superior oblique muscle had increased hazard ratios for reoperation.

Superior oblique tucking is performed when superior oblique tendon laxity is noted (Bhola, Velez & Rosenbaum 2005, Helveston & Ellis 1983, Li & Zhao 2014). Unilateral two or more vertical muscle surgery excluding the superior oblique muscle is performed where there is a hyper-deviation of greater than 15–20 PD in primary position in patients with unilateral SOP (Caca et al. 2012, Nash et al. 2017, 2019). The results of our study suggest that patients with superior oblique tendon laxity and/or large vertical deviations have a higher risk of reoperation compared with patients who have normal superior oblique muscles and/or those with comparatively small vertical deviations. We presume that most patients who underwent unilateral two or more vertical muscle surgery excluding the superior oblique muscle underwent inferior oblique weakening and a superior rectus recession. A previous study reported that the success rate of combined inferior oblique myectomy and superior rectus recession was 50% (Caca et al. 2012). The success rate increased to 80% when an adjustable suture was used for superior rectus recession combined with inferior oblique disinsertion. Nonetheless, overcorrections occurred in 20% of these patients (3 out of 15 patients) (Özkan, Akyuz Unsal & Kagnici 2019). Our results showed a 22.2% (26/117) reoperation rate in patients who underwent unilateral surgery on two or more vertical muscles excluding the superior oblique muscle. Surgery on the superior oblique muscle was associated with a higher reoperation risk than surgery on one vertical muscle excluding the superior oblique muscle. We presumed that most superior oblique muscle surgeries consisted of superior oblique tucking to remove the laxity of the superior oblique tendon. Helveston and Ellis (Helveston & Ellis 1983) reported that acquired Brown’s syndrome required take-down in 17% of patients an average of 9.1 months after superior oblique tucking and myectomy of the antagonist inferior oblique. In a study of 123 patients with superior oblique palsy, poor outcomes (>7 prism diopter) were reported in 5% (1/22) of patients after an inferior oblique weakening surgery, compared to 21% (7/34) of patients following superior oblique tuck. Our results are consistent with these other studies.

In patients who underwent reoperation, the median time between the initial surgery and the reoperation was approximately 5.5 months. Older age was associated with a shorter time interval between the initial surgery and reoperation. A reoperation after strabismus surgery is a joint decision between a patient and their surgeon. Patients with presumed congenital SOP who are undercorrected, are often asymptomatic due to their high vertical fusional amplitudes whereas even a small overcorrection can result in debilitating diplopia. It is not surprising that older patients who are overcorrected would choose to have a reoperation sooner than children given the difficulty most adults have in suppressing one image when overcorrected.

Combined horizontal muscle surgeries were performed in 17.8% of the patients in our series. Helveston et al (Helveston et al. 1996) reported that a concomitant horizontal deviation was present in 36% of SOP cases and 10% of patients with SOP underwent combined horizontal surgery. In a previous study that included both unilateral congenital and acquired SOP with large deviations, the reoperation rate was higher in patients who required simultaneous horizontal muscle surgery than patients who did not require combined surgery (Nejad et al. 2013). Khawam et al (Khawam, Scott & Jampolsky 1967) has only recommended combined vertical and horizontal muscle surgery in patients with SOP and an esotropia. Combining horizontal and vertical muscle surgery did not affect the reoperation rate in our series.

This study has several limitations. First, it was conducted using claims data, which does not provide information on the angle of deviation, amount of cyclotorsion, presence of abnormal head posture, fixation preference, oblique muscle overaction or underaction, visual acuity, or stereoacuity. Furthermore, it was not possible to distinguish between unilateral and bilateral SOP in most cases. Second, we excluded the patients with acquired SOP due to trauma, microvascular disease, and diseases of the central nervous system or orbit to include only patients with presumed congenital SOP in our study. However, patients with idiopathic acquired SOP may have been included. Third, miscoding may have occurred if a provider submitted the wrong diagnosis or procedure code. Lastly, we could not investigate patient data outside of the enrollment period; hence, we may have missed patients who were diagnosed and underwent surgery for SOP before enrollment. Furthermore, the initial surgery during the enrollment period may have been a reoperation. To control for this possibility, we excluded patients with the add-on CPT codes 67332, which would suggest prior a strabismus surgery.

In conclusion, based on medical claims data that reflect real-world practice, the reoperation rate for presumed congenital SOP was 10.7%. Patients who underwent unilateral or bilateral surgeries that included the superior oblique muscle or surgery involving two or more vertical muscles excluding the superior oblique muscle showed increased hazard ratios for reoperation. Older age was associated with a shorter time interval between initial surgery and reoperation.