Convergence insufficiency (CI) is a common binocular vision disorder characterized by a greater exophoria at near than at distance, a receded near point of convergence (NPC), and reduced positive fusional vergence (PFV). These features lead to symptoms such as eye strain, headaches, blurred or double vision, and difficulty with reading or near tasks [
1-
3]. The prevalence of CI ranges from 2.25% to 17.6%, and it can significantly affect quality of life, especially in children and adolescents, by impairing academic performance and concentration [
2]. Despite its frequency and clinical significance, CI is often underdiagnosed due to its nonspecific symptoms and the lack of standardized diagnostic criteria [
1,
2]. The underlying pathophysiology involves a mismatch in neural control of accommodation and vergence. This imbalance results in inefficient vergence adaptation and excessive accommodative effort, which together contribute to the clinical manifestations of CI [
1].
Among patients with intermittent exotropia, a subset may present with CI-type characteristics, known as CI-type exotropia, which is characterized by a greater exodeviation at near than at distance [
4]. These patients often exhibit symptoms similar to those of classic CI, such as asthenopia and difficulty with near tasks, but differ in that they display intermittent manifest exotropia rather than pure phoria [
4,
5]. CI-type exotropia represents a distinct clinical variant within the spectrum of intermittent exotropia but is often underrecognized due to its overlapping features with other forms of exotropia and the lack of standardized diagnostic criteria.
Nonsurgical interventions, particularly orthoptic exercises (also known as vision therapy), including convergence exercises have been widely prescribed as first-line treatment for CI [
6-
8]. These exercises aim to enhance the function of the vergence and accommodative systems through repetitive, targeted visual tasks, utilizing neuroplasticity to improve binocular coordination and relieve symptoms [
1,
2]. Orthoptic therapy can be administered in various forms, including home-based pencil push-ups, computer-based programs, and office-based vergence/accommodative therapy with home reinforcement [
6-
8]. However, there remains a lack of consensus a mong eye care professionals regarding the most effective treatment modality, partly due to differences in study design, patient populations, and outcome measures [
2].
Among the home-based interventions, pencil push-ups are the simplest and most accessible for patients with CI. This technique involves focusing on a small target, such as the tip of a pencil, while gradually bringing it closer to the nose to stimulate convergence. It improves convergence ability and enhances neuromuscular control of the vergence system [
1,
2]. Pencil push-ups are inexpensive, easy to perform, and do not require any specialized equipment, which makes them a commonly used first-line treatment in clinical settings [
9]. However, their effectiveness may be limited due to poor patient compliance and the absence of therapist supervision, as reported in several studies [
7].
Recent high-quality randomized controlled trials and network meta-analyses have demonstrated that office-based vergence/accommodative therapy with home reinforcement is more effective than home-based therapies in children with CI [
2]. However, while the pediatric evidence is robust, studies in adults remain limited and inconclusive [
10]. Furthermore, CI-type exotropia in adults has been particularly understudied, as most clinical trials have narrowly focused on classic CI and frequently excluded this variant from their inclusion criteria. Uncertainties remain about the optimal type, intensity, and duration of therapy, as well as the influence of individual patient factors and adherence on treatment outcomes [
2,
11].
Given these uncertainties, further investigation is needed to assess the effectiveness of orthoptic exercises in patients with CI-type exotropia. This study aims to evaluate the long-term effects of consistent pencil push-up exercises in adults with CI-type exotropia. By comparing patients with high and low compliance, we aim to assess not only clinical efficacy but also the role of adherence in treatment outcomes. The findings are expected to inform tailored therapeutic strategies for this underrecognized subgroup of exotropia patients.
Materials and Methods
Ethics statement
This study was approved by the Institutional Review Board of Kosin University Gospel Hospital (No. 2025-06-026). The requirement for informed consent was waived due to the use of deidentified data and the retrospective nature of the study. All study procedures were performed in accordance with the principles of the Declaration of Helsinki.
Participants
Thirty participants diagnosed with CI-type exotropia were categorized into two groups according to their compliance with home-based pencil push-up exercises: a highly compliant group (pencil push-ups group) and a poorly compliant group (noncompliant group). Participants were not randomized but were instead allocated to groups based on self-reported adherence, reflecting real-world patient behavior and feasibility. This study was conducted as a retrospective observational comparison.
Inclusion criteria for CI-type exotropia included symptomatic exodeviation at near that was at least 10 prism diopters (PD) greater than at distance, a receded NPC with a break point of ≥6 cm, based on evidence that such findings are predictive of convergence dysfunction and response to convergence therapy [
12-
14]. Although PFV and Convergence Insufficiency Symptom Survey (CISS) score were not included as formal criteria, the diagnosis was based on near-distance disparity and receded NPC, which are widely accepted diagnostic markers in CI-type exotropia. Exclusion criteria included a history of strabismus surgery, amblyopia, nystagmus, neurological disorders, or any ocular pathology affecting binocular vision or convergence.
The pencil push-ups group were instructed to perform pencil push-up exercises using a standard pencil with a visible letter on its side as the fixation target (
Fig. 1). Each session consisted of 15 repetitions per set. For each repetition, patients were instructed to slowly move the pencil with a visible target from arm’s length toward the nose until they experienced diplopia or could no longer maintain a single image. They were advised to hold the pencil at the closest point where single vision could be maintained for 5 seconds before returning to the starting position. Patients performed four sets per day (a total of 60 repetitions per day), spaced evenly throughout waking hours, for a period of 6 months. Instruction and technique were demonstrated by a clinician during the initial visit, and patients were reminded to avoid head movements and to focus only with both eyes during the exercise.
Adherence to the exercise regimen was assessed through verbal self-reports during follow-up visits. Patients in the pencil push-ups group performed the exercise at least 5 days per week, whereas the noncompliant group did fewer. The compliance threshold (≥5 days per week) was pragmatically set based on the Convergence Insufficiency Treatment Trial (CITT) [
4].
Clinical assessments
All participants underwent baseline assessment of best-corrected visual acuity (BCVA), Intraocular pressure, deviation angle, NPC and stereopsis. BCVA was recorded using logarithm of the minimum angle of resolution (log-MAR) values. The angle of deviation was measured using the alternate prism cover test at both near (33 cm) and distance (6 m) fixation. PFV, accommodative amplitude, or symptom surveys such as the CISS were not included in the evaluation. After a 6-month intervention period, changes in the angle of deviation, NPC, and stereopsis were reassessed and compared between the two groups to evaluate the efficacy of consistent pencil push-ups performance. Stereopsis was assessed using the Titmus near stereotest (Stereo Optical). Standardized assessments across all subjects were conducted at baseline and at 6 months, which served as the primary end point for analysis. All measurements and assessments were conducted under standardized conditions at fixed distances (near, 33 cm; distance, 6 m) by a single experienced examiner using standard clinical protocols. This ensured consistency and minimized inter-observer variability.
Statistical analysis
The results of the two groups were compared using GraphPad Prism ver. 10.4.2 (GraphPad Software). Continuous variables were tested for normality using the Shapiro-Wilk test. As the data did not meet the assumption of normality, nonparametric statistical tests were applied. The Wilcoxon signed rank test was used to compare pre- and post-intervention values within groups, and the Mann-Whitney U-test was employed for comparisons between the pencil push-ups group and noncompliant group. Statistical significance was defined as p < 0.05.
Results
General characteristics
A total of 30 patients were included and arranged into pencil push-ups group and noncompliant group. The pencil push-ups group consisted of 20 patients (11 men and 9 women) who demonstrated good compliance (more than 5 days a week) with the prescribed pencil push-ups exercise regimen, while the noncompliant group comprised 10 patients (4 men and 6 women) who did not adequately perform the pencil push-ups. There was no statistically significant difference in the sex distribution between the two groups (
p = 0.699). The mean age was 35.65 ± 10.44 years in the noncompliant group and 32.50 ± 11.16 years in the pencil push-ups group, with no statistically significant difference between the two groups (
p = 0.463). There were no statistically significant differences in BCVA between the two groups for either eyes (right eye,
p = 0.704; left eye,
p = 0.792). Refraction also showed no statistically significant difference between the two groups in either eye (right eye,
p = 0.687; left eye,
p = 0.736). Intraocular pressure showed no statistically significant difference between the two groups in either eye (right eye,
p = 0.856; left eye,
p = 0.588) (
Table 1).
Angle of exodeviation
The pencil push-ups group showed a significant reduction in the angle of exodeviation, from 24.65 ± 5.46 to 14.40 ± 5.71 PD at distance (
p < 0.001) and from 35.25 ± 5.36 to 23.85 ± 6.68 PD at near (
p < 0.001). In contrast, the noncompliant group showed no significant change in deviation at either distance fixation (
p = 0.403) or near fixation (
p = 0.307) (
Table 2 and
Fig. 2A-2D).
Near point of convergence
NPC significantly improved in the pencil push-ups group, decreasing from 22.6 ± 5.2 to 8.6 ± 3.1 cm (
p < 0.001). In contrast, no significant change was observed in the noncompliant group (
p = 0.510) (
Table 3 and
Fig. 3).
Stereopsis
No participants in either group demonstrated measurable stereopsis better than 400 arcseconds at baseline, and this threshold remained unchanged after 6 months of intervention, suggesting limited effect of pencil push-up exercises on stereoacuity in this adult CI-type exotropia population (
Table 4).
Discussion
This study demonstrated that consistent convergence exercises using pencil push-ups significantly improved both the angle of exodeviation and the NPC in patients with CI-type exotropia. These findings are consistent with previous reports indicating that convergence training is an effective nonsurgical intervention for CI [
10,
12,
14].
The CITT and other randomized controlled studies have established the clinical efficacy of orthoptic therapy in reducing symptoms and improving fusional vergence in classic CI, especially in pediatric populations [
7]. However, the CITT specifically addressed classic CI and did not include patients with CI-type exotropia, thus leaving a notable gap in evidence for this subgroup.
In contrast, Singh et al. [
15] demonstrated that orthoptic treatment, including convergence exercise was beneficial in small-angle CI-type exotropia (≤25 PD), resulting in improved binocular function and symptomatic relief. Furthermore, Kuwera and Hariharakumar [
16] reported a rare case of bilateral internuclear ophthalmoplegia accompanied by CI type exotropia, in which the patient achieved complete resolution of diplopia and measurable reduction of exodeviation solely through convergence exercises. This case underscores the potential effectiveness of targeted convergence therapy even in complex neuro-ophthalmologic conditions. These findings collectively support the potential efficacy of convergence exercises specifically in CI-type exotropia patients, and highlight the need for further targeted research to better define optimal treatment strategies for this subgroup.
The present study builds upon previous findings by demonstrating that consistent adherence to home-based pencil push-ups can also yield substantial improvements in adult patients. A notable strength of our study is the high compliance rate observed in the pencil push-ups group, with participants performing four sets of 15 pencil pushups daily over 6 months without deviation from the protocol. Given the intensive nature and the extended 6-month duration of this regimen, the observed improvements highlight the crucial role of patient motivation and adherence in achieving favorable outcomes. These results suggest that, under conditions of high patient compliance, home-based therapy may approximate the effectiveness of office- based interventions.
While office-based vergence/accommodative therapy is considered the most effective intervention, particularly in children, access and cost remain significant barriers to its widespread use [
2,
3]. Our findings support the feasibility of home-based therapy as a meaningful alternative, especially for motivated adult patients who receive adequate instruction. This aligns with previous studies suggesting that, while the efficacy of home-based exercises can be variable, beneficial outcomes are achievable in selected patients [
3].
Recent advances in technology have introduced digital and virtual tools for vision therapy, including computerized vergence training systems, virtual reality-based platforms, and mobile applications [
2,
17]. These modalities may help address challenges related to compliance and monitoring. Nevertheless, our findings indicate that simple, low-cost interventions such as pencil push-ups remain valuable, particularly when performed with sufficient frequency and consistency.
Although pencil push-ups are primarily designed to improve near convergence, a modest yet statistically significant reduction in distance exodeviation was also noted. This may be attributed to enhanced tonic vergence and improved neuromuscular coordination of ocular alignment as a result of repetitive convergence training, as well as enhanced patient awareness and voluntary control developed during the exercise period [
1,
18].
Despite significant improvements, posttreatment deviation angles remained above clinical thresholds in many patients, suggesting that pencil push-ups may be more appropriate as an adjunct or bridging therapy rather than a definitive treatment in cases with large-angle CI-type exotropia. This underscores the importance of individualized treatment planning based on the severity of deviation, symptom burden, and patient-specific factors. Although NPC significantly improved following the intervention, the mean posttreatment values remained above the normal range. This partial recovery may reflect structural or adaptive limitations in adult patients, emphasizing the need for early intervention to maximize therapeutic outcomes [
19]. Despite improvements in convergence and motor alignment, no measurable enhancement in stereopsis was observed in either group. This likely reflects long-standing suppression or reduced sensory plasticity in the adult CI-type exotropia population. These findings suggest that motor alignment alone may not be sufficient to restore binocular depth perception, especially in patients with chronic deviation [
13].
Several limitations should be acknowledged. First, the grouping was based on patients’ adherence to the pencil push-up regimen, rather than randomized allocation. The compliance-based grouping may have introduced selection bias, as well as confounding due to individual differences in motivation, health-seeking behavior, or baseline visual habits. This design limits the strength of causal inference and reduces the generalizability of our findings to broader clinical populations. Second, the sample size was relatively small, which may have reduced the statistical power and robustness of the results. Third, adherence to the exercise protocol was assessed through patient self-report during clinic visits, which introduces the possibility of recall bias and overestimation of actual compliance. As a result, misclassification of group assignment may have occurred, potentially influencing the validity of the observed differences between the compliant and noncompliant groups. Future studies should consider using objective monitoring tools, such as digital logs or adherence tracking applications, to enhance data reliability. Fourth, standardized symptom questionnaires or functional assessments, such as the CISS, were not employed, which may limit both the clinical relevance and the ability to evaluate subjective treatment outcomes. While participants anecdotally reported symptoms such as eye strain and fatigue during near work fatigue, the severity of these symptoms was not systematically quantified using validated instruments like the CISS. Future studies should incorporate validated symptom assessment tools like the CISS to more accurately capture patients’ symptomatic changes. Fifth, although a significant improvement in NPC was observed after pencil push-up exercises, the mean NPC remained above the normal range (<6 cm) in most patients. This indicates that while the intervention enhanced convergence ability, it did not fully normalize near function, highlighting a limitation in overall treatment efficacy. Lastly, due to the observational design, we were unable to determine the definitive impact of pencil push-up exercises on the need for surgical intervention.
In conclusion, convergence exercises using pencil push-ups may provide a viable nonsurgical treatment option for CI-type exotropia with receded NPC, particularly when performed consistently and with high adherence. Clinically, pencil push-ups may serve as a practical first-line option or an adjunctive approach in adults with CI-type exotropia, particularly in those with moderate deviation and high motivation. While not a replacement for surgery in cases of large-angle deviation, they may help delay the need for surgical intervention or improve post-operative outcomes by enhancing convergence ability preoperatively. However, further research is necessary to establish the optimal frequency, intensity, and duration of such exercises, and to clarify their potential to reduce or delay the need for surgical correction.