Comprehensive analysis of strabismus reoperations: clinical insights and progression factors

Abstract

Background

To compare and analyze clinical characteristics of patients undergoing two surgeries and multiple surgeries and explore relevant factors to lay the foundation for clinical prediction.

Methods

A retrospective analysis was conducted on clinical data from all patients who underwent twice and multiple strabismus surgeries at Tianjin Eye Hospital between October 2012 and September 2021. Patients were divided into Group A (two surgeries) and Group B (more than two surgeries) based on the cumulative number of surgeries performed. Clinical details at the first recurrence, including sex, age, native place, overall medical history, onset time, visual acuity, affected muscle(s), etc., were documented. Non-parametric tests and chi-square tests were used to analyze clinical characteristics in each group. Binary and ordered logistic regression analysis assessed parameters associated with multiple reoperations. A linear mixed-term model observed factors impacting affected muscle(s) during surgery. Researchers examined clinical traits related to secondary strabismus variables.

Results

Among the 910 included patients, 840 required two surgeries (Group A) and 70 underwent more than two surgeries (Group B). Significant differences were found in age, onset time, interval time, and secondary factors. Regression analysis highlighted the significant impact of interval time on the reoperation rate, effectively predicting outcomes in patients with concomitant strabismus. Other ophthalmoplegia and secondary factors significantly influenced reoperation rates in patients with non-concomitant strabismus. Interval time, esotropia, and exotropia were linked to concomitant secondary strabismus patients, while the number of surgeries, DVD, esotropia, exotropia, and esotropia V-pattern were associated with non-concomitant secondary strabismus patients. In a longitudinal study, patients with multiple surgeries showed a correlation between the vertical deviation angle magnitude and the number of involved extraocular muscles. Regression analysis revealed that in patients with concomitant strabismus, interval time, exotropia, and esotropia influenced the total number of muscles during surgery. For patients with non-concomitant strabismus, interval time, secondary factors, and SOP impacted the total number of muscles during surgery.

Conclusions

Interval time in patients with concomitant strabismus, as well as secondary and other ophthalmoplegia in non-concomitant strabismus, are the main factors for multiple reoperations.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12886-024-03712-2.

Introduction

Strabismus, a condition affecting visual function and binocular movement, is categorized based on variations in ocular movement and deviation angle at different fixation gaze, distinguishing between concomitant and non-concomitant strabismus [1, 2]. The etiology of strabismus involves factors such as neurodevelopmental abnormalities, deficiencies in motor function mechanisms, anatomical abnormalities, and disruptions in perceptual fusion [3–7]. While surgical intervention proves effective for a majority of strabismus cases, aimed at improving ocular position and binocular movement, the condition’s inherent characteristics introduce the potential for reoperations and long-term complications post-operation. Therefore, individuals with strabismus may confronted with the risk of undergoing additional surgeries, sometimes even multiple procedures [8, 9].

The reoccurrence and/or reoperation of strabismus is influenced by various factors, including a high accommodative convergence to accommodation ratio, lateral incomitancy, age at the time of surgery, and the timing for surgical reintervention [10, 11]. Nevertheless, there remain considerable controversies on the recurrence of postoperative strabismus. Long-term follow-up after strabismus correction surgery for about 1–8 years showed a recurrence rate of 20 -40% for esotropia and 22 -59% for exotropia [12, 13]. While Superior oblique surgery、botulinum injection、a history of prior strabismus surgery were associated with increased odds of reoperation [14, 15]. Vertical and cyclotorsion strabismus, driven by their etiology, ocular movement patterns, and other characteristics, may also necessitate multiple surgical interventions [16, 17]. Common contributors to secondary/residual strabismus encompass amblyopia, restricted eye movement, and dissociated vertical deviation (DVD) and/or A-pattern or V-pattern [18]. Therefore, summarizing the attributes and characteristics of patients with secondary/residual strabismus undergoing reoperations holds paramount significance for conducting timely observations and providing guidance to patients, thereby avoiding key factors associated with strabismus reoccurrences. The timely identification of latent factors serves the dual purpose of pinpointing crucial risk indicators for strabismus reoperations and facilitating early intervention to predict the likelihood of relapses in patients.

This investigation is designed to provide an in-depth analysis of the etiological underpinnings and clinical profiles of individuals undergoing singular and repetitive strabismus operations. A comparison of the distinctive features between patients undergoing twice and multiple surgeries was conducted, with the aim of identifying statistically significant correlates. Robust regression models were meticulously crafted to elucidate the factors intricately linked to secondary strabismus. Employing both longitudinal and cross-sectional methodologies, this study meticulously delineated the primary determinants contributing to the necessity for repetitive strabismus operations. The envisaged outcome is to furnish clinicians with a nuanced understanding, thereby offering informed guidance to mitigate the incidence of strabismus reoperations.

Methods

Patients and methods

This study strictly adheres to the Helsinki Declaration and has been approved by the Medical Ethics Committees of Tianjin Eye Hospital (KY-2024001). Before the study started, all patients (minor patients including parents) signed the informed consent forms. Retrospective collection and analysis of medical records were conducted on all patients who underwent second or multiple strabismus surgery at Tianjin Eye Hospital from October 2012 to September 2021 (a total of 910 people). According to the number of surgeries, patients are divided into Group A (twice surgeries) and Group B (> twice surgeries). twice surgeries represent long-term follow-up after strabismus surgery (at the time of article reporting) with binocular alignment. The data encompassed various parameters, including patient demographics (sex, age, place of origin), comprehensive medical history, onset timeline, etiological factors, visual acuity (LogMar visual acuity chart), and the enumeration of affected muscles during the second surgery.

The study excluded individuals meeting the following criteria: patients experiencing initial recurrence who did not visit at our hospital, those with surgery for the null zone of nystagmus, and individuals who underwent botulinum toxin injections. The population density data is based on the 1999 report on urban infrastructure levels in various regions released by the National Bureau of Statistics of China (https://www.stats.gov.cn/zt_18555/ztsj/hjtjzl/1999/202303/t20230302_1923294.html).

Statistical analysis

All analyses were conducted using SPSS 26.0. Kolmogorov Smirnov test was used to check the normal distribution of the results. Non-parametric rank sum test for quantitative variables, described using median M (P25, P75). The comparison of categorical variables is conducted using the chi square test or Fish exact probability statistics. Binary and ordered logistic regression analysis was conducted to test all independent variables, and it was found that P < 0.05 was significant, while the Hosmer Lemeshow test was completed. The Spearman rank correlation test was performed on continuous independent variables. The linear mixed-term model was used to analyze multiple follow-up data.

Result

Reoperations patients’ demographics

This retrospective analysis investigated the etiology and clinical characteristics of patients experiencing reoperations of ocular deviation who underwent either single or multiple surgical interventions at Tianjin Eye Hospital from October 2012 to September 2021. The study included 910 patients, with 840 individuals undergoing a second surgery (Group A), and 70 patients undergoing multiple surgeries (Group B). Their average ages are 10.0 years (5.0,20.8) and 6.0 years (3.0,13.0), with a statistically significant difference (P < 0.001). Within this cohort, 92.31% underwent two surgeries, 7.03% underwent three surgeries, 0.55% underwent four surgeries, and 0.11% required five surgeries. The onset time and interval time in Group A and Group B were 3.0 years (1.0, 6.0) and 1.0 (0.5, 3.0), respectively; 3.0 years (1.0,7.0), 2.0 (0.8,5.0), with statistically significant differences (P < 0.001, = 0.002). Congenital factors constituted a higher proportion (86.70%, 7.36%) among the etiological factors, significantly surpassing acquired factors (5.60%, 0.33%). Residual factors (80.77%, 6.04%) were also higher than secondary factors (11.54%, 1.65%), demonstrating significant differences (P = 0.042). In Group A, specific diagnoses included 7 cases of Brown syndrome, 7 cases of exotropia triad (exotropia A-pattern, dissociated vertical deviation (DVD), and superior oblique overaction), 13 cases of thyroid associated ophthalmopathy (TAO), 23 cases of Congenital Cranial Dysinnervation Disorders (CCDDs), 1 case of torsional strabismus, 1 case of orbit apex syndrome, and 1 patient with Myasthenia (Table 1).

Table 1.

Comparison of characteristics of patients with group A and B

(Variables)	A	B	P value
No. of patients	840(92.31)	70(7.69)
Onset time, years	3.0(1.0,6.0)	1.0 (0.5,3.0)	< 0.001
Interval time, years	3.0 (1.0,7.0)	2.0 (0.8,5.0)	0.002
Sex
Male	426(46.81)	31(3.41)	0.321
Female	414(45.49)	39(4.29)
Age, years	10.0 (5.0,20.8)	6.0 (3.0,13.0)	< 0.001
Etiology
Congenital	789(86.70)	67(7.36)	0.792
Acquired	51(5.60)	3(0.33)
Diagnosis
Residual	735(80.77)	55(6.04)	0.042
Secondary	105(11.54)	15(1.65)
Concomitant strabismus	361(39.67)	31(3.41)
Esotropia	98(10.77)	10(1.10)	0.562
Sensory esotropia	3(0.33)	0(0.00)	1.000
Exotropia	214(23.52)	13(1.43)	0.250
Sensory Exotropia	7(0.77)	0(0.00)	1.000
Vertical Deviations	49(5.38)	8(0.88)	0.072
Non-concomitant Strabismus	479(52.64)	39(4.29)
Paralytic strabismus	220(24.18)	21(2.31)	0.573
DVD	141(15.49)	12(1.32)	1.000
A-V pattern	51(5.60)	4(0.44)	1.000
Special Motility Disorders	53(5.82)	2(0.22)
Brown syndrome	7(0.77)	0(0.00)	1.000
Exotropia triad	7(0.77)	2(0.22)	0.148
TAO	13(1.43)	0(0.00)	0.615
CCDDs	23(2.53)	0(0.00)	0.248
Torsional strabismus	1(0.11)	0(0.00)	1.000
Orbit Apex Syndrome	1(0.11)	0(0.00)	1.000
Myasthenia	1(0.11)	0(0.00)	1.000
With eyelid disorders	21(2.31)	1(0.11)	1.000
LogMar visual acuity (R)	0.0 (0.0,0.0)	0.0 (0.0,0.0)	0.708
LogMar visual acuity (L)	0.0 (0.0,0.0)	0.0 (0.0,0.0)	0.547
Affected muscle	1.0 (1.0,2.0)	1.0 (1.0,2.0)	0.457

Note Important findings (P < 0.05) are shown in bold. The value of continuous variable is the median (P25, P75), and the dummy variable is the number of cases and proportion (%).DVD, dissociated vertical deviation; TAO, thyroid associated ophthalmopathy; CCDDs, Congenital Cranial Dysinnervation Disorders

Ordered logistic regression analysis based on patient age

Based on previous research, the age of patients undergoing reoperation was divided into 0–3, 4–6, and 7–12 years, and significant influencing factors were observed in different age groups [19]. Using ordered logistic regression analysis, investigate the relationship between multiple reoperations, exotropia, esotropia, secondary, esotropia V-pattern, and exotropia V-pattern in different age groups of patients with both concomitant and non-concomitant strabismus. Obtained odds ratio (OR) values for multiple reoperations, exotropia, and secondary in patients with concomitant strabismus were 0.383, 2.045, and 2.463, respectively, all of which were significant (P = 0.031, 0.043, 0.029); The OR values for the multiple reoperations, esotropia V-pattern, and exotropia V-pattern in patients with non-concomitant strabismus were 0.437, 3.237, and 3.982, respectively, all of which were significant (P = 0.02, 0.047, 0.011); This implies that esotropia, esotropia V-pattern, exotropia V-pattern, and secondary factors are expected to positively influence age groups, while multiple reoperations are anticipated to have a negative impact across different age groups (Table 2).

Table 2.

Ordered logistic regression of patients with concomitant and non-concomitant strabismus based on age

Predictors	Concomitant strabismus		Predictors	Non-concomitant strabismus
	OR (95% CI)	P Value		OR (95% CI)	P Value
Multiple reoperations	0.383(0.16–0.918)	0.031	Multiple reoperations	0.437(0.218–0.878)	0.020
Exotropia	2.045(1.021–4.094)	0.043	Esotropia V-pattern	3.237(1.014–10.335)	0.047
Esotropia	0.65(0.289–1.463)	0.298	Exotropia V-pattern	3.982(1.364–11.625)	0.011
Secondary	2.463(1.096–5.534)	0.029	Secondary	1.212(0.528–2.783)	0.651

Correlation analysis between patients with concomitant strabismus and multiple reoperations

Comparison of clinical characteristics of patients with concomitant strabismus between Group A and Group B

As mentioned above, the proportion of patients with concomitant and non-concomitant strabismus varies between secondary and multiple surgeries. Therefore, a relationship was observed between concomitant strabismus patient characteristics and the total number of surgeries according to the etiology. Significant differences (P = 0.01, 0.005, 0.013) were observed in the onset time, interval time, and age between patients in Group A and Group B with concomitant strabismus (Table 3). This suggests a relationship between multiple reoperations and the age, interval time, and onset time of individuals with concomitant strabismus. In comparison to Group A, patients in Group B exhibit a shorter onset time and interval, as well as a younger age.

Table 3.

Clinical characteristics analysis of patients with concomitant strabismus

(Variables)	A	B	P value
No.of patients	361(92.09)	31(7.91)
Onset time, years	3.0 (1.0,6.0)	1.0 (0.5,3.0)	0.010
Interval time, years	4.0 (2.0,8.0)	1.5 (0.5,5.0)	0.005
Sex
Male	163(41.58)	13(3.32)	0.851
Female	198(50.51)	18(4.59)
Age, years	11.0 (7.0,22.5)	8.0 (3.0,17.0)	0.013
Etiology
Congenital	338(86.22)	29(7.40)	1.000
Acquired	23(5.87)	2(0.51)
Trauma	12(3.06)	1(0.26)	1.000
Diagnosis
Residual	284(72.45)	24(6.12)	1.000
Secondary	77(19.64)	7(1.79)
Esotropia	98(25)	10(2.55)	0.676
Sensory esotropia	3(0.77)	0(0.00)	1.000
Exotropia	214(54.59)	13(3.32)	0.087
Sensory Exotropia	7(1.79)	0(0.00)	1.000
Vertical Deviations	49(12.5)	8(2.04)	0.105
With Eyelid disorders	10(2.55)	0(0.00)	1.000
LogMar visual acuity (R)	0.0 (0.0,0.0)	0.0 (0.0,0.0)	0.249
LogMar visual acuity (L)	0.0 (0.0,0.0)	0.0 (0.0,0.0)	0.860

Note Important findings (P < 0.05) are shown in bold. The value of continuous variable is the median (P25, P75), and the dummy variable is the number of cases and proportion (%)

Binary logistic regression analysis of patients with concomitant strabismus

A binary logistic regression analysis was performed, considering interval time, acquired factors, sex, and secondary factors in patients with concomitant strabismus. Continuous independent variables have completed Spearman rank correlation test (eTable 1). The OR value for interval time was 0.901 (95% confidence interval: 0.815–0.995) with a significance (P = 0.039). This indicates that for each unit increase in the interval time, the likelihood of undergoing more than two surgeries decreases by a factor of 0.901. Consequently, a shorter interval between the first surgery in patients with concomitant strabismus is associated with an elevated probability of requiring more than two surgeries (Table 4).

Table 4.

Binary logistic regression analysis of Group A and Group B in patients with concomitant strabismus

Predictors		A VS B
	β	SE	WaldX2	OR (95% CI)	P Value
Interval time	-0.105	0.051	4.245	0.901(0.815–0.995)	0.039
Acquired	-0.198	0.776	0.065	0.82(0.179–3.753)	0.798
Sex	-0.157	0.384	0.168	0.854(0.403–1.812)	0.682
Secondary	0.193	0.457	0.178	1.213(0.495–2.971)	0.673

Correlation analysis between patients with non-concomitant strabismus and multiple reoperations

Comparison of clinical characteristics of patients with non-concomitant strabismus between Group A and Group B

Significant differences (P = 0.016, 0.003, 0.003) were observed in the onset time, age, and presence of secondary factors between patients in Group A and Group B with non-concomitant strabismus. Patients in Group B exhibited a shorter onset time, a younger age, and a higher proportion of secondary factors compared to Group A (Table 5).

Table 5.

Clinical characteristics analysis of patients with non-concomitant strabismus

(Variables)	A	B	P value
No.of patients	479(92.47)	39(7.53)
Onset time, years	3.0 (1.0,6.0)	1.0 (1.0,3.0)	0.016
Interval time, years	3.0 (1.0,7.0)	2.0 (1.0,4.0)	0.114
Sex
Male	263(50.77)	18(3.47)	0.319
Female	216(41.70)	21(4.05)
Age, years	8.0 (5.0,19.0)	5.0 (3.0,10.0)	0.003
Etiology
Congenital	451(87.07)	38(7.34)	0.715
Acquired	28(5.41)	1(0.19)
Trauma	13(1.43)	0(0.00)
Diagnosis
Residual	451(87.07)	31(5.98)	0.003
Secondary	28(5.41)	8(1.54)
Paralytic strabismus
Other ophthalmoplegia	18(3.47)	4(0.77)	0.075
SOP	193(37.26)	17(3.28)	0.736
Monocular elevation deficiency	7(1.35)	0(0.00)	1.000
Abducens nerve paralysis	13(2.51)	0(0.00)	0.612
Oculomotor nerve paralysis	3(0.58)	0(0.00)	1.000
Trochlear nerve paralysis	3(0.58)	1(0.19)	0.270
DVD	141(27.22)	12(2.32)	0.856
DHD	2(0.39)	0(0.00)	1.000
A-V pattern
Esotropia A-pattern	3(0.58)	1(0.19)	0.270
Esotropia V-pattern	15(2.90)	2(0.39)	0.371
Exotropia A-pattern	15(2.90)	1(0.19)	1.000
Exotropia V-pattern	21(4.05)	0(0.00)	0.391
With eyelid disorders	11(2.12)	1(0.19)	1.000
LogMar visual acuity (R)	0.0 (0.0,0.0)	0.0 (0.0,0.0)	0.626
LogMar visual acuity (L)	0.0 (0.0,0.0)	0.0 (0.0,0.1)	0.349

Binary logistic regression analysis of patients with non-concomitant strabismus

Consequently, a binary logistic regression analysis was conducted, considering variables such as age, onset time, secondary factors, sex, acquired factors, Superior Oblique Palsy (SOP), DVD, trochlear nerve paralysis, and other ophthalmoplegia in non-concomitant strabismus patients. The OR values and 95% confidence intervals (CI) for secondary factors and other ophthalmoplegia in this model were 4.849 (1.932–12.171) and 4.424 (1.254–15.605), respectively (P = 0.001, 0.021). These results indicate that the presence of secondary factors and other ophthalmoplegia significantly increases the likelihood of second or subsequent surgeries in patients with non-concomitant strabismus (Table 6).

Table 6.

Binary logistic regression of Group A and Group B in patients with non-concomitant strabismus

Predictors				A VS B
	β		SE		WaldX2	OR (95% CI)		P value
Age	-0.022		0.023		0.907	0.978(0.934–1.024)		0.341
Onset time	-0.032		0.045		0.512	0.968(0.887–1.057)		0.474
Secondary	1.579		0.47		11.303	4.849(1.932–12.171)		0.001
Sex	-0.496		0.352		1.986	0.609(0.305–1.214)		0.159
Acquired	0.077		1.189		0.004	1.08(0.105–11.099)		0.949
SOP	0.335		0.416		0.649	1.398(0.619–3.161)		0.420
DVD	0.231		0.419		0.304	1.26(0.554–2.863)		0.581
Trochlear nerve paralysis	1.895		1.214		2.438	6.651(0.616–71.751)		0.118
Other ophthalmoplegia	1.487		0.643		5.346	4.424(1.254–15.605)		0.021

Analysis of related factors of secondary reoperation strabismus

Clinical characteristics analysis of patients with residual and secondary strabismus

Secondary causes significantly contribute to the increased frequency of surgeries, with notable clinical distinctions observed in patients with secondary strabismus compared to those with residual strabismus. There are significant differences in multiple reoperations, interval time, age, esotropia, sensory esotropia, exotropia, vertical deviations, other ophthalmoplegia, SOP, and esotropia V-pattern between the two groups (P = 0.042, 0.015, 0.015, < 0.001, 0.047, 0.006, < 0.001, < 0.001, < 0.001, 0.004). Patients with secondary strabismus, in contrast to those with residual strabismus, undergo more surgeries, experience longer intervals, are older in age, and exhibit associated esotropia, exotropia, sensory esotropia, other ophthalmoplegia, SOP, and a higher prevalence of esotropia V-pattern (Table 7).

Table 7.

Clinical characteristics analysis of patients with residual and secondary strabismus

(Variables)	Residual	Secondary	P value
Single reoperations	735(80.77)	105(11.54)	0.042
Multiple reoperations	55(6.04)	15(1.65)
Onset time, years	3.0(1.0,5.3)	3.0(1.0,9.0)	0.235
Interval time, years	3.0(1.0,6.0)	4.0(1.0,10.0)	0.015
Sex
Male	406(44.62)	51(5.60)	0.078
Female	384(42.20)	69(7.58)
Age, years	9.0(5.0,19.0)	12.0(7.0,23.0)	0.015
Etiology
Congenital	740(81.32)	116(12.75)	0.297
Acquired	50(5.49)	4(0.44)
Trauma	22(2.42)	4(0.44)	0.767
Diagnosis
Concomitant strabismus
Esotropia	144(15.82)	54(5.93)	< 0.001
Sensory esotropia	1(0.11)	2(0.22)	0.047
Exotropia	278(30.55)	58(6.37)	0.006
Sensory exotropia	6(0.66)	3(0.33)	0.103
Vertical deviations	161(17.69)	5(0.55)	< 0.001
Non-concomitant strabismus
Other ophthalmoplegia	21(2.31)	1(0.11)	< 0.001
SOP	203(22.31)	7(0.77)	< 0.001
Monocular elevation deficiency	7(0.77)	0(0.00)	0.603
Abducens nerve paralysis	13(1.43)	0(0.00)	0.237
Oculomotor nerve paralysis	14(1.54)	0(0.00)	0.236
Trochlear nerve paralysis	4(0.44)	0(0.00)	1.000
DVD	138(15.16)	15(1.65)	0.192
DHD	2(0.22)	0(0.00)	1.000
A-V pattern
Esotropia A-pattern	3(0.33)	1(0.11)	0.433
Esotropia V-pattern	10(1.10)	7(0.77)	0.004
Exotropia A-pattern	15(1.65)	1(0.11)	0.709
Exotropia V-pattern	19(2.09)	2(0.22)	1.000
LogMar visual acuity (R)	0.0(0.0,0.0)	0.0(0.0,0.0)	0.335
LogMar visual acuity (L)	0.0(0.0,0.0)	0.0(0.0,0.0)	0.749

Logistic regression models based on secondary and residual factors in patients with concomitant and non-concomitant strabismus

Subsequently, additional research delved into the correlation between pertinent factors and the progression of secondary strabismus. For patients with concomitant strabismus, binary logistic regression analysis was applied to variables such as sex, age, interval time, multiple reoperations, esotropia, exotropia, and trauma. The OR values for interval time, esotropia, and exotropia were determined as 1.0127, 36.586, and 13.284, respectively, all of which achieved statistical significance (P = 0.005, 0.001, 0.012). This implies that an increase in interval time, presence of esotropia, and presence of exotropia elevate the likelihood of developing secondary strabismus in patients (Fig. 1A).

Fig. 1.

Logistic regression models based on secondary and residual factors. (A) Binary logistic regression analysis of patients with concomitant secondary strabismus. (B) Binary logistic regression analysis of non-concomitant secondary strabismus patients. Yellow represents OR > 1, blue represents OR < 1

Subsequently, binary logistic regression analysis was executed on patients with non-concomitant strabismus, encompassing variables such as age, onset time, sex, multiple reoperations, SOP, DVD, other ophthalmoplegia, esotropia, exotropia, esotropia A-pattern, esotropia V-pattern, exotropia A-pattern, and exotropia V-pattern. All continuous independent variables have completed the Spearman rank correlation test (eTable 2). The OR values for multiple reoperations, DVD, esotropia, exotropia, and esotropia V-pattern were calculated as 3.802, 2.864, 13.738, 9.319, and 24.114, respectively, all of which demonstrated statistical significance (P = 0.015, 0.02, < 0.001, < 0.001, < 0.001). This implies that the presence of multiple reoperations, DVD, esotropia, exotropia, and esotropia V-pattern significantly increases the likelihood of developing secondary strabismus in patients (Fig. 1B).

Analysis of factors affected muscle(s) in patients with reoperations strabismus

Longitudinal observation of factors affected muscle(s) in patients with multiple reoperations

Patients were selected who have undergone 3 or more surgeries to construct an LME model (patients can complete prism examination). A panel model was constructed with age, onset time, vertical and horizontal deviation at distance and near as explanatory variables, and vertical and horizontal muscle quantity as dependent variables. The mixed POOL model was selected: onset time and vertical deviation at distance, with regression coefficient values of 0.148 and 0.043, respectively. Among three or more hospitalizations, onset time and vertical deviation at distance have a significant positive impact on the number of muscles involved in vertical surgery.

The application of the FE model found that the results of horizontal deviation at near and distance were not related to the number of horizontal muscles involved during surgery, but onset time had a positive impact on it. Therefore, for patients with multiple surgeries, the vertical deviation angle is correlated with the number of extraocular muscles involved in the surgery (Table 8).

Table 8.

Linear mixed effects models of clinical characteristics of patients with multiple reoperations strabismus

Predictors		Horizontal			Predictors		Vertical
	t		Std. Err	P Value		t		Std. Err	P Value
Age	-1.169		0.057	0.251	Age	0.991		0.008	0.326
Onset time	2.567		0.076	0.015	Onset time	2.901		0.051	0.005
33cmPD	-0.505		0.032	0.617	33cmPD	-1.207		0.019	0.232
5mPD	0.873		0.033	0.389	5mPD	2.176		0.02	0.034

An ordered regression model was constructed based on the affected muscle(s) of reoperation patients

The factors were further observed that affect the total amount of muscle involved during surgery. Using multiple reoperations, interval time, secondary, exotropia, esotropia, SOP, and DVD as independent variables, and affected muscle(s) (1, 2, 3, greater than 3) as dependent variables, an ordered logistic regression analysis was conducted to investigate the impact of related factors on affected muscle(s) during secondary surgery. The OR values and 95% CI for the interval time, horizontal exotropia, and esotropia in the patient of concomitant strabismus were 1.055 (1.02–1.091), 3.348 (1.632–6.868), and 2.57 (1.175–5.62), respectively (P = 0.002, 0.018, 0.001); It means that the interval time, exotropia and esotropia, have a positive impact on the affected muscle(s) of patients with concomitant strabismus. The OR values and 95% CI for interval time, secondary, and SOP in the model of non-concomitant strabismus patients were 1.094 (1.059–1.131), 3.499 (1.827-6.7), and 0.395 (0.269–0.58), respectively (all P < 0.001); This indicates that the interval time and secondary factors positively influence the affected muscle(s) in non-concomitant strabismus patients, while SOP (Standard Operating Procedure) has a negative impact (Table 9).

Table 9.

Ordered logistic regression of patients with concomitant and non-concomitant strabismus stratified by affected muscle(s)

Predictors	Concomitant strabismus		Predictors	Non-concomitant strabismus
	OR (95% CI)	P Value		OR (95% CI)	P Value
Multiple reoperations	1.03(0.472–2.247)	0.941	Multiple reoperations	0.805(0.412–1.576)	0.527
Interval time,	1.055(1.02–1.091)	0.002	Interval time,	1.094(1.059–1.131)	< 0.001
Secondary	1.387(0.841–2.287)	0.200	Secondary	3.499(1.827-6.7)	< 0.001
Exotropia	3.348(1.632–6.868)	0.018	SOP	0.395(0.269–0.58)	< 0.001
Esotropia	2.57(1.175–5.62)	0.001	DVD	1.139(0.776–1.671)	0.505

Clinical characteristics of esotropia V-pattern patients

As previously discussed, individuals with esotropia V-pattern tend to exhibit a higher likelihood of developing secondary strabismus, an increased number of affected muscles, and a younger age. Consequently, clinical characteristics were closely examined for all patients displaying esotropia V-pattern. Within this group, patients with V-pattern exhibit patients with V-pattern exhibit a younger age and longer interval time, specifically 3 (1.25, 7.25) and 4.5 (3, 7.5), respectively. Additionally, there is a higher prevalence of congenital and secondary causes (100%, 43.75%), along with an increased count of affected muscles ranging from 1 to 2.75 (Table 10).

Table 10.

Analysis of clinical characteristics of patients with internal oblique V-pattern

NO.	Number of surgeries	Age of first reoperations	Sex	native place	Onset time	Interval time	Residual / Secondary	SOP	DVD	Affected muscle
1	2	5	M	Tianjin	3	3	0	0	0	3
2	2	7	M	Henan	0.5	0.5	1	0	0	4
3	2	26	M	Guizhou	3	3	0	0	0	1
4	2	5	F	Tianjin	3	4	0	0	0	2
5	2	10	F	Tianjin	8	8	0	0	0	1
6	2	28	F	Tianjin	9	9	0	0	0	1
7	2	8	F	Tianjin	5	5	0	1	0	2
8	2	7	F	Hebei	0.5	0.5	0	1	0	2
9	2	3	M	Hubei	3	2	0	0	0	5
10	2	18	F	Inner Mongolia	11	13	1	0	0	2
11	2	8	M	Anhui	1	3	1	0	0	1
12	2	8	M	Henen	3	5	0	0	1	2
13	2	9	F	Jilin	2	4	1	0	0	2
14	2	18	M	Hubei	10	10	1	0	1	1
15	3	6	F	Hebei	2	5	1	0	0	2
16	3	10	F	Hebei	0.5	6	1	0	0	4

Diagnosis for patients with multiple surgeries

Among all reoperation patients, patients who have been hospitalized more than 3 times will undergo etiological analysis in order to observe the causes of recurrent strabismus in patients. The majority of patients undergoing reoperation receive diagnoses of SOP and DVD before subsequent surgeries, with a notable prevalence of secondary cases among those undergoing multiple reoperations (Table 11).

Table 11.

List of etiological analyses of multiple surgeries

No of surgeries	First		Second		Third		Fourth		Fifth
No.of patients	Age	Diagnosis	Age	Diagnosis	Age	Diagnosis	Age	Diagnosis	Age	Diagnosis
NO.1	1.4	esotropia、SOP	2	Residual esotropia	2.5	SOP	3	SOP	3.5	SOP
NO.2	-		5	Residual esotropia、DVD	7	DVD	9	Residual exotropia、DVD	-
NO.3	1.5	esotropia	2.5	Secondary exotropia、DVD	3	Secondary exotropia、DVD	3	Secondary exotropia、DVD	-
NO.4	-		4	SOP、DVD	5	DVD	6	DVD	-
NO.5	-		6	SOP	6.5	SOP	8	SOP	-
NO.6	1.7	exotropia	2	Secondary esotropia、SOP	3	Secondary esotropia	4	DVD	-

Distribution of hometowns of reoperation patients

The second surgery patients mainly come from Tianjin, Hebei, Shandong, Henan, Liaoning and other places. Patients who have undergone three or more surgeries mainly come from Tianjin, Hebei, Henan and other places (Fig. 2). The population densities of Tianjin, Hebei, and Henan are 1381, 1769, 1326 (persons/ sq.km), which are 2.99, 3.83, and 2.87 times higher than the national population density (462 persons/ sq.km), respectively. Therefore, even though the reoperation patients in this study mainly come from the North China region, the population density in the North China region accounts for a relatively large proportion of the national population density, indicating that the results of this study have a certain degree of reliability.

Fig. 2.

Distribution of hometowns of reoperation patients. (A) Distribution of patients undergoing twice surgeries. Tianjin and Hebei have the largest number of reoperation patients. (B) Distribution of patients undergoing third surgeries. Tianjin and Hebei have the largest number of reoperation patients. (C) Distribution of patients with more than third surgeries. Tianjin, Hebei and Henan have the largest number of reoperation patients

Discussion

Residual or secondary strabismus that requires additional surgical intervention is commonly encountered in clinical practice, and potential risk factors for strabismus reoperations include patient age, the presence of nystagmus, and the onset time of strabismus [10, 12]. Previous studies have shown that younger age is associated with a higher likelihood of reoperation [19]; Bilateral lateral rectus recession (BLR) and unilateral lateral recession combined with medial resection (R&R) procedure are linked to recurrence rates in patients, although this remains a point of controversy [20, 21]. Infantile esotropia has been reported to have a higher reoperation rate [22]. Therefore, this study aimed to summarize the relevant risk factors and develop a predictive model by comparing sex, age, onset time, interval time, and secondary characteristics in patients undergoing two surgeries versus multiple surgeries.

The study included 457 men and 453 women, and no significant differences in gender were observed between the two groups. Additionally, multiple reoperations were not correlated with systemic illness. In terms of etiological factors, no significant difference was found between congenital and acquired causes. However, an increased proportion of secondary causes was noted in patients with multiple reoperations. Comparing patients with two surgeries to those with multiple surgeries, significant differences were observed in both onset time and interval time. Patients undergoing multiple reoperations had shorter intervals between surgeries, underscoring the critical importance of timing for subsequent procedures. Previous studies suggest that, where possible, delaying the second surgery for at least 8 to 12 weeks is advisable to facilitate tissue healing and reduce inflammation [11, 23]. Our results align with this, as we found that shorter intervals between surgeries significantly increased the likelihood of multiple surgeries in patients with concomitant strabismus. Slight surgical overcorrection appears to yield satisfactory outcomes in patients with intermittent exotropia, potentially reducing the need for further surgeries [11, 24]. Therefore, when muscle slippage is not the issue, appropriate post-surgical intervention and close observation can help reduce the need for strabismus reoperations.

In non-concomitant strabismus patients, the likelihood of reoperation increases significantly if secondary factors such as ophthalmoplegia are present. This may be attributed to the imbalance in extraocular muscle strength, which affects binocular fusion and increases the probability of postoperative intervention. The reoperation rate for strabismus varies across age groups. Therefore, this study stratified patients into three age groups: 0–3 years, 4–6 years, and 7–12 years [19]. Ordered logistic regression was used to analyze the relevant factors across age groups. The results indicate that multiple reoperations have a negative impact on younger age groups, while exotropia, secondary strabismus, esotropia with a V-pattern, and exotropia with a V-pattern have a positive impact. This is consistent with the finding that the older the age at onset, the lower the postoperative recurrence rate [12]. Furthermore, the factors associated with reoperation vary across different age groups.

Previous studies have shown that factors associated with infant reoperations differ from those in adults. For infants, larger preoperative deviation angles, younger age at initial surgery, high hyperopia, developmental delay, and early-onset esotropia are key risk factors, while for adults, only some of these indicators are relevant [19, 25, 26]. These findings are important for predicting early surgical needs and developing tailored treatment strategies for different age groups.

Understanding the causes of secondary strabismus is crucial for early intervention to reduce its incidence. Compared with patients with residual strabismus, those with secondary strabismus undergo more surgeries, experience longer intervals between surgeries, are older, and are more likely to present with esotropia, exotropia, sensory esotropia, ophthalmoplegia, superior oblique palsy (SOP), and esotropia with V-pattern. This suggests that there are distinct patterns in the age and strabismus type distributions among patients with secondary strabismus. The coexistence of multiple types of strabismus may increase the complexity of treatment and the difficulty of surgery, exacerbating secondary strabismus.

Regression analysis showed that shorter interval time, esotropia, and exotropia increased the likelihood of secondary strabismus in patients with concomitant strabismus Regarding etiology, previous studies have found that postoperative exotropia recurs, which is consistent with our findings, suggesting that overcorrection or older age may be contributing factors [27, 28]. In non-concomitant strabismus patients, multiple reoperations, dissociated vertical deviation (DVD), esotropia, exotropia, and esotropia with V-pattern were associated with an increased risk of secondary strabismus. This is because complex conditions such as DVD and V-pattern esotropia disrupt extraocular muscle balance, affecting binocular fusion. In our study, the presence of V-pattern esotropia during reoperation was a significant risk factor for secondary strabismus [29]. Most patients who underwent multiple surgeries had SOP or DVD, and one-third of patients required multiple surgeries due to secondary strabismus. This suggests that paralytic strabismus disrupts eye movement balance and triggers compensatory mechanisms. For these patients, treatment plans should be adjusted according to the specific type of strabismus [30].

Further analysis of the number of muscles involved in surgery, using both longitudinal and cross-sectional data, revealed that among patients with three or more surgeries, the duration of disease onset and the angle of vertical deviation at a distance had a significant positive impact on the number of affected muscles. However, horizontal deviation factors had no significant impact, leading us to propose that as the number of reoperations increases, the horizontal deviation angle is no longer entirely positively correlated with the number of affected muscles, possibly due to changes in muscle tension. Ordered logistic analysis showed that an increase in the number of affected muscles did not correspond to an increased likelihood of reoperation, which contrasts with findings in strabismus patients with thyroid eye disease [31]. The main factors associated with this phenomenon include interval time, secondary factors, exotropia, esotropia, and SOP.

Geographic factors also influence the incidence of strabismus [32]. n this study, the distribution of patients with single versus multiple recurrences was examined. The majority of patients requiring second, third, or subsequent surgeries were geographically dispersed, particularly in the Tianjin, Hebei, and Henan provinces. These provinces are located in North China, a region with a large land area and high population density, lending credibility to the data in this study. To establish a comprehensive understanding of the incidence of strabismus and the proportion of patients requiring surgery, further population-based research is needed, particularly considering that most patients were from Tianjin and its surrounding areas. A significant number of these patients underwent multiple reoperations [33].

Our research presents several limitations. Firstly, as all samples were collected from the same hospital, the study lacks a comprehensive description of the overall population of strabismus cases. Secondly, if some mildly affected patients have the potential for reoperation but opt for non-surgical methods, their reoperation rates cannot be calculated. Additionally, the study’s statistical analysis is based on a specific population, limiting the generalizability of the results.

In summary, this study analyzed patients undergoing two or more surgeries and found that the interval time in concomitant strabismus patients, along with secondary factors such as ophthalmoplegia in non-concomitant strabismus patients, were the primary contributors to multiple reoperations. The age at surgery was also correlated with the recurrence rate. Clinicians should pay particular attention to maintaining eye movement balance during reoperation, especially in patients presenting with V-pattern esotropia or DVD, to reduce the incidence of secondary strabismus.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

RN and PPL collected data and completed data analysis. RN and PPL jointly wrote the manuscript. WZ and RH contributed to the concept and study design. WZ and RH guided statistical methods. WZ and RH reviewed and revised the manuscript.

Funding

The present study was supported by General Project of Tianjin Health Science and Technology Fund (grant no. TJWJ2021MS041), Tianjin Key Medical Discipline (Specialty) Construction Project (grant no. TJYXZDXK‑016 A). Tianjin Eye Hospital Science and Technology Fund Youth Cultivation Project (grant no. YKPY2201), Open Project of Nankai University Optometry and Vision Science Institute (2024430HJ0347).

Data availability

The data used to support the findings of this study are available from the corresponding author upon request.

Declarations

Ethics approval and consent to participate

This study strictly adheres to the Helsinki Declaration and has been approved by the Medical Ethics Committees of Tianjin Eye Hospital (KY-2024001). Before the study started, all patients (minor patients including parents) signed the informed consent forms.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Clinical trial number

Not applicable.