Convergence insufficiency (CI) is a syndrome in which the ability to turn both eyes inward (converge) when focusing on a near target is reduced, making it impossible to maintain binocular fusion. It is a condition in which convergence movements are insufficient due to accommodative dysfunction, resulting in inadequate accommodative convergence (convergence that occurs with the activation of accommodative effort) and fusional convergence (convergence that occurs to align images from both retinas). It is characterized by a receded near point of convergence (NPC), reduced convergence amplitude, and exophoria at near (typically greater than 10 prism diopters [∆]).
Convergence insufficiency is classified into the following three types.
| Type | Definition | Characteristics |
|---|---|---|
| Functional convergence insufficiency | A syndrome characterized by receded near point of convergence, reduced convergence amplitude, and near exophoria | Most common. May be accompanied by accommodative dysfunction |
| Convergence paralysis | An organic disease with acute onset that completely prevents convergence | Caused by dorsal midbrain syndrome such as pineal tumor. Head MRI is essential. |
| Convergence spasm | Convergence abnormality with excessive adduction and miosis of both eyes. Often psychogenic. | Differentiation from pseudo-6th nerve palsy is important. |
It occurs in almost all age groups but is most common in young adults. Prevalence varies widely between studies, ranging from 1.7% to 33%, and the incidence in the general population is estimated at 0.1% to 0.2%. CI is found in 11% to 19% of children with exophoria. There is no sex difference. CI accounts for approximately 15.7% of adult-onset strabismus. It also tends to occur in VDT workers, those engaged in near work, and schoolchildren and students.
In general, CI is unlikely to improve naturally. However, the severity of symptoms varies depending on the amount of near work. CI after a concussion may improve over time.
The annual incidence of CI is estimated at 8.4 per 100,000 people. CI is found in 11–19% of children with exophoria, and symptoms often become apparent during the period when near work increases after starting school (ages 7–10). In modern society where VDT work has become common, adult CI is also on the rise5).
It is most common in young adults but can occur across a wide age range from children to the elderly. Among children with exotropia, 11–19% have CI, and CI accounts for approximately 15.7% of new-onset strabismus in adults. It is more frequent in VDT workers and those engaged in near work, and also occurs in schoolchildren and students.
Subjective symptoms of CI worsen with near work. They become noticeable with prolonged use of reading, computers, and smartphones.
In functional convergence insufficiency, patients complain of severe eye strain during near work. Because accommodation and convergence ability are reduced, prolonged near work leads to exophoria at near, resulting in crossed diplopia, sensory abnormalities, and eye strain.
The CISS (Convergence Insufficiency Symptom Survey) developed by the CITT group uses 15 questions answered on a Likert scale to quantify symptom severity with a score from 0 (best) to 60 (worst). A score of 16 or higher is considered significant. Its reliability has been validated in children aged 9 to 18 years and in adults.
Unlike functional convergence insufficiency, convergence paralysis has an acute onset with inability to converge, resulting in exotropia only at near and thus diplopia. Adduction is possible and there is no restriction of eye movements, but convergence is not possible. Therefore, exotropia occurs at near, causing crossed diplopia. An important distinguishing point is that diplopia occurs only at near and not at distance.
The CISS, developed by the CITT group, is a standard patient-reported outcome measure that quantifies the severity of convergence insufficiency symptoms. It consists of 15 items rated on a Likert scale (0 to 4 points), with a total score ranging from 0 to 60 points1).
Representative questionnaire items:
A CISS score of 16 or higher is considered the threshold for significant symptoms. The reliability of repeated measurements in the same patient (ICC 0.87) and sensitivity to changes before and after treatment intervention have been confirmed, and it is used in both clinical trials and daily practice 2). Using the amount of score improvement (treatment effect indicator: a decrease of 6 points or more) when explaining to patients helps in sharing treatment goals.
The following four items are used as diagnostic criteria for CI.
| Test findings | Criteria for abnormality |
|---|---|
| Convergence near point (CNP) recession | ≥6 cm (pre-presbyopic) / ≥10 cm (presbyopic) |
| Exophoria at near | >10∆ |
| Decreased fusional vergence (PFV) | Near <15–20∆ |
| CISS score | ≥16 points (subjective symptoms) |
The Convergence Insufficiency Symptom Survey (CISS) is a 15-item Likert scale questionnaire that quantifies the severity of CI symptoms on a scale of 0 to 60. A score of 16 or higher is considered the threshold for suspecting CI, and it is also used to evaluate treatment effectiveness. Its reliability has been validated in children aged 9 to 18 years and in adults.
Primary CI
Congenital imbalance of convergence and divergence: Due to differences in innervation, the ability to converge for near vision is limited.
Fusional convergence insufficiency: CI mainly occurs due to incomplete fusional convergence (the convergence that occurs to align the retinal images of both eyes).
Acquired CI
Convergence insufficiency with accommodative dysfunction: Excessive near work and VDT tasks reduce accommodative function, leading to insufficient accommodative convergence and fusional convergence.
Fatigue and excessive near work: This is likely to occur in individuals who engage in prolonged VDT work (technostress eye strain) or near work.
Drugs and systemic diseases: Anticholinergic drugs, uveitis, post-concussion syndrome, and CNS diseases such as Parkinson’s disease can be causes.
Trauma and other causes: Head trauma, glasses that induce a base-out prism effect, and encephalitis can also be causes.
Convergence consists of four components (Maddox classification).
| Types of convergence | Description |
|---|---|
| Accommodative convergence | Convergence induced by accommodative effort. Expressed as the AC/A ratio. |
| Fusional convergence | Voluntary convergence to align the retinal images of both eyes |
| Proximity convergence | Convergence due to proximity perception of the target (other than disparity) |
| Tonic convergence | Convergence for sustained eye position at rest |
In CI, fusional convergence is mainly insufficient, but it may also be accompanied by decreased accommodative convergence (CI with accommodative dysfunction). The relationship between convergence and accommodation is not proportional but exists within a certain range. Prolonged near work in inappropriate environments can disrupt this relationship, leading to persistent decline in accommodation and convergence functions.
Convergence paralysis is caused by organic lesions. The dorsal midbrain syndrome, primarily due to tumors near the cerebral aqueduct (especially pineal tumors), demyelination, inflammation, and vascular lesions, is well known. Unlike functional convergence insufficiency, it is an acute-onset emergency requiring neuroimaging.
Although a direct causal relationship has not been established, prolonged VDT work causes a decline in convergence and accommodation functions, worsening CI symptoms. It is recognized as technostress eye strain, and it is recommended to limit continuous VDT work to a maximum of one hour, followed by a 10-15 minute break, and to arrange the environment accordingly.
The diagnosis of CI is based on subjective symptoms and the following clinical examination findings. A comprehensive sensorimotor examination, assessment of refractive status, and dilated fundus examination are recommended.
The main examination methods are shown below.
| Examination | Overview of method | Reference range for abnormal values |
|---|---|---|
| Convergence near point test (NPC) | Move fixation target from 40–50 cm toward the nose | ≥6 cm (pre-presbyopic) / ≥10 cm (presbyopic) |
| Fusional vergence test (PFV) | Measured with base-out prism | Near <15–20∆ |
| AC/A ratio | Heterophoria method or gradient method | <2:1 |
Move a target such as a finger or toy from a position 40–50 cm in front of the face, slightly below eye level, slowly toward the bridge of the nose. Measure the distance from the bridge of the nose to the point where the target begins to appear double or either eye deviates outward (break point). The normal value is approximately 6–8 cm. Repeated measurement of accommodative and convergence function is useful for diagnosis, and gradual prolongation of the near point with repeated testing also suggests CI.
It is important to perform the measurement under full refractive correction. In the heterophoria method, the AC/A ratio is calculated from the difference in strabismus angle between distance (5 m) and near (33 cm). The normal value is approximately 4±2, and it shows a low value in CI.
Using a major amblyoscope, rotary prism, Bagolini striated lens, etc., the convergence range that can maintain binocular single vision while keeping accommodation constant is measured. In measuring the fusion range with a major amblyoscope or base-out prism, even cases where initial fusional convergence is sufficient may show an extension of the convergence near point.
In convergence paralysis, adduction is possible during eye movement, but convergence movement is completely absent. An important distinguishing point from functional convergence insufficiency is that when measuring the fusion range using a major amblyoscope or base-out prism, the fusion range in the convergence direction is almost unmeasurable.
The pupillary light reflex is normal, so light-near dissociation (the pupillary light reflex is normal, but the near response [miosis, accommodation, convergence] is absent) is observed. In acute-onset convergence paralysis, prompt neuroimaging (e.g., head MRI) is required to rule out intracranial lesions.
| Disease | Key points for differentiation from CI |
|---|---|
| Convergence paralysis | Acute onset, fusion range nearly unmeasurable, light-near dissociation, head MRI required |
| Convergence spasm | Bilateral extreme adduction, miosis, monocular adduction movement with limited resolution, associated accommodation spasm |
| Divergence paralysis | Increased esotropia at distance. Near deviation within normal range |
| Abducens nerve palsy | Monocular abduction limitation (CI involves binocular convergence insufficiency) |
| Dry eye, accommodative insufficiency | Near vision symptoms and eye strain are similar, but NPC and fusional vergence are normal |
Treatment for CI is performed stepwise according to severity and disease type. First, the disease type is confirmed, and the presence or absence of dysregulation is evaluated before determining the treatment strategy.
If accommodative dysfunction is present, convergence training may worsen eye strain. Therefore, it is important not to perform convergence training. Prioritize environmental improvements, and instruct that continuous VDT work should be limited to a maximum of one hour, followed by a 10- to 15-minute break.
Refractive correction is performed in all cases. After conducting a refraction test using cycloplegic agents (e.g., Mydrin-P ophthalmic solution), prescribe near-vision-only glasses tailored to the actual VDT working distance. Intermediate-near progressive lenses are acceptable, but distance-near bifocals or distance-near progressive lenses are not recommended because the near portion is too small. If dry eye is also present, use artificial tears or hyaluronic acid eye drops concomitantly.
For functional convergence insufficiency without accommodative dysfunction, refractive correction is the foundation, combined with convergence training, prism glasses, and surgery.
In CI without accommodative insufficiency, convergence training improves fusional convergence. It is important to perform it daily, even for a short time. Home convergence training (such as pencil push-ups and convergence cards) tends to be inferior to in-office therapy, but it has been shown that home training alone can improve symptoms in children with symptomatic CI 6).
Home Training
Pencil push-ups: Focus on a small target, and while maintaining binocular single vision, slowly bring the target toward the nose.
Convergence card: Hold the card at the bridge of the nose and gradually shift gaze from the farthest point to a nearer target.
Stereogram: Two images separated horizontally are cross-viewed to produce a third fused image in the center.
In-hospital training
In-office vision therapy: Intentionally and controllably manipulating target blur, disparity, and proximity to eliminate suppression and normalize convergence and accommodation.
Computer vergence training (CVS): A program that uses random dot stereograms to gradually increase the required vergence amount. Progress can be monitored.
A Cochrane systematic review by Scheiman et al. (2020) (12 RCTs, 1289 participants) showed “high-certainty evidence” that in-office therapy with home reinforcement in children leads to better convergence ability compared to pencil push-ups alone or computer therapy 2). Children treated with base-in prism reading glasses did not show significant improvement 2). In adults, base-in prism glasses improved symptoms but did not improve convergence ability 2). In young adults aged 19–30 years, in-office training was more effective than home training in improving positive fusional vergence (PFV), but there was no significant difference in NPC or patient symptoms 1).
Success rate and long-term outcomes of convergence training
The reported success rate of in-office convergence training is 70–80%, and many patients remain asymptomatic even one year after discontinuing treatment 1). However, recurrence rates have also been reported, and regular follow-up is recommended, especially in cases of high-grade CI, post-concussion CI, and CI with accommodative dysfunction.
In the CITT study (2005), 73% of children achieved a CISS score ≤15 (normalization) with 4–8 weeks of in-office convergence training (1 session per week plus 12–24 hours of home training per week), whereas only 43% achieved this with pencil push-ups alone 4). This result provides evidence that supervised in-office convergence training is superior to home-based training alone.
The Cochrane review indicates that evidence for treatment of CI in adults is limited compared to children, and further research is needed2). For post-concussion CI, no standardized protocol has been established, and individual balancing of active training and rest is necessary1).
Prescribed when convergence training does not improve symptoms or in cases of convergence paralysis. Use the minimum amount of prism necessary to achieve comfortable binocular single vision at near. The optimal prism power is determined by incorporating 2–4∆ base-in in each eye (total 4–8∆ correction) into the near refractive correction glasses and performing a wear test 7). Long-term wear may be necessary regardless of the underlying disease. In adults with presbyopia, base-in prisms can be effective 7).
In convergence paralysis, improvement can be expected if treatment of the underlying disease is effective. In the meantime, base-in prism glasses for near vision are prescribed to symptomatically reduce diplopia. For underlying diseases such as pineal tumors, demyelinating diseases, and vascular lesions, treatment in collaboration with neurology and neurosurgery is necessary.
Detailed Protocol for Convergence Training
A standard protocol for convergence training for functional CI without accommodative dysfunction is shown below2).
| Training type | Method | Target frequency |
|---|---|---|
| Pencil push-ups | While fixating both eyes on a pen, bring it closer to the bridge of the nose. Stop just before double vision appears, hold for 5 seconds, and return. 10 repetitions per set. | 3 to 5 sets per day |
| Convergence card (Brock string) | Three beads are placed at equal intervals on a 70 cm string; the patient sequentially fixates on each bead from the nearest to train convergence step by step | Twice a day, 5 minutes each |
| In-hospital vision training | Gradual loading of convergence and accommodation. Using Vectograph, convergence cards, and Brewster-type stereoscope | 1–2 times per week (total 12–16 weeks) |
| Computer training (HTS) | Gradually increase convergence using a binocular vision program on screen; progress confirmed by automatic recording. | 15–20 minutes per day |
It is important to perform convergence training daily, and continuing even for short periods is key to long-term success. The treatment goal is to improve the CISS score to 16 or less, achieve an NPC of 5 cm or less, and a PFV of 20∆ or more 1).
CI complicated by accommodative dysfunction should be treated as a separate condition1).
Additional checklist items for diagnostic criteria:
In the treatment of convergence insufficiency with accommodative dysfunction, environmental improvements and spectacle correction are prioritized before performing convergence training. Near-vision glasses (with an addition of +0.75 to +1.25 D) may reduce accommodative load and contribute to symptom improvement.
This is indicated for refractory CI and CI accompanied by intermittent exotropia.
Indications for surgery: when manifest strabismus is present at distance, symptoms are consistent, and prism glasses are ineffective.
The main surgical procedures include bilateral lateral rectus recession (based on distance deviation angle), bilateral medial rectus resection (based on near deviation angle), and unilateral lateral rectus recession with medial rectus resection 8). Botulinum toxin injection is also an option for refractory cases.
The reported success rate of convergence training is 70–80%, and most patients remain asymptomatic even one year after discontinuing treatment. However, convergence training is contraindicated in convergence insufficiency with accommodative dysfunction, and environmental improvements and spectacle prescription should be prioritized. Even when training is performed, the maintenance of effects varies among individuals, so regular follow-up is recommended.
The exact disease mechanism of CI has not been fully elucidated, but the neural centers controlling convergence movements have been identified.
This condition is based on accommodative dysfunction, where both accommodative convergence and fusional convergence become insufficient, resulting in inadequate convergence movement. The relationship between convergence and accommodation is not proportional but exists within a certain range. Prolonged near work in inappropriate environments gradually leads to persistent decline in accommodative and convergence functions. Technostress eye syndrome caused by VDT work is a typical example.
Improving the VDT work environment, wearing appropriate glasses, and instilling artificial tears are effective.
Convergence paralysis is primarily an organic disorder caused by the dorsal midbrain syndrome. Lesions near the cerebral aqueduct impair the convergence center. While the pupillary light reflex remains normal, the near response (miosis, accommodation, convergence) is impaired, resulting in light-near dissociation. Unlike functional convergence insufficiency, it is characterized by the inability to measure the fusional convergence amplitude.
Divergence insufficiency is characterized by increased esotropia at distance, contrasting with the near exophoria of CI. The adult strabismus PPP clearly outlines a differential diagnosis flow for divergence insufficiency and convergence paralysis, and both conditions require exclusion of organic lesions (elevated intracranial pressure, central nervous system disorders) 1).
In divergence paralysis, the main complaint is horizontal diplopia at distance, and symptoms are reduced or disappear at near. This contrasts with CI, where diplopia occurs at near, making history-taking the first step in differentiation. Additionally, in differentiating from abducens nerve palsy, the presence or absence of monocular abduction limitation is important, and it should be noted that both CI and divergence paralysis involve impairment of binocular convergence and divergence movements1).
CI develops in association with several neurological diseases.
Convergence movements are integratively controlled by multiple nuclei in the midbrain and pons. The main neural structures involved in convergence are as follows.
In convergence paralysis, as part of the dorsal midbrain syndrome (Parinaud syndrome), it may be accompanied by upward gaze palsy, convergence-retraction nystagmus, and light-near dissociation. Pineal tumors, demyelination, hemorrhage, and trauma are the main causes, and for acute-onset convergence paralysis, detailed evaluation with head MRI is essential 1).
In functional convergence insufficiency, no organic lesion is observed, but the core pathophysiology is thought to be ciliary muscle fatigue due to prolonged near work and functional overload of the convergence center. In the modern era of rapidly increasing digital device use, the rise of CI in the context of technostress eye strain (VDT syndrome) has become a concern.
The Convergence Insufficiency Treatment Trial – Attention and Reading Trial (CITT-ART) was a randomized multicenter clinical trial that investigated whether treatment of symptomatic convergence insufficiency improves reading performance in children aged 9 to 14 years.
Participants were randomly assigned to either in-office convergence and accommodation therapy or in-office placebo therapy. After 16 weeks, there was no significant difference in CISS scores between the two groups, and the results indicated that in-office convergence and accommodation therapy was not more effective than placebo therapy in improving reading comprehension in children with symptomatic CI3).
This result suggests that even if CI treatment improves convergence ability and symptoms, it may not directly lead to improved reading comprehension.
Regarding the 2005 CITT, multiple ophthalmology experts have pointed out methodological limitations4). The in-office treatment group was prescribed significantly longer treatment times than the other groups (inequality of treatment dose). Additionally, there is criticism that “pencil push-ups” do not accurately represent conventional vision therapy, which includes a variety of exercises using accommodative targets4).
Spontaneous remission of symptoms has been reported in CI patients. Therefore, it is considered important to include a placebo group when evaluating treatment effects5). CI is observed in 11–19% of children with exophoria, and the annual incidence of CI is estimated at 8.4 per 100,000 people5).
The CITT-ART (2019) showed that convergence/accommodation therapy did not improve reading comprehension3), but this does not mean that treatment for CI is ineffective. Improvements in CISS scores and near point of convergence were observed; the impact on reading comprehension, a composite outcome, was merely limited. In-office convergence training significantly improved CISS scores (treatment group: 16.0→9.0, placebo group: 16.0→12.5), and this improvement directly contributes to patient quality of life3).
The fact that CITT-ART used improvement in reading comprehension in children as its primary endpoint is a unique strength, but there is also criticism that reading comprehension is influenced by multiple factors such as learning environment, cognitive function, and attention, making it inherently difficult to capture improvement through convergence training alone 1).
Post-concussion CI following sports-related concussion (SRC) is increasingly recognized as a special subtype of symptomatic CI. The characteristics of post-concussion CI are as follows1).
It is important to incorporate CI assessment and management into the return-to-sports protocol, and collaboration between ophthalmology and sports medicine departments is beneficial 1).
American Academy of Ophthalmology. Adult Strabismus Preferred Practice Pattern. Ophthalmology. 2023.
Scheiman M, Kulp MT, Cotter SA, Lawrenson JG, Wang L, Li T. Interventions for convergence insufficiency: a network meta-analysis. The Cochrane database of systematic reviews. 2020;12(12):CD006768. doi:10.1002/14651858.CD006768.pub3. PMID:33263359; PMCID:PMC8092638.
Convergence Insufficiency Treatment Trial Study Group.. Randomized clinical trial of treatments for symptomatic convergence insufficiency in children. Arch Ophthalmol. 2008;126(10):1336-1349. doi:10.1001/archopht.126.10.1336. PMID:18852411; PMCID:PMC2779032.
Scheiman M, Mitchell GL, Cotter S, Cooper J, Kulp M, Rouse M, Borsting E, London R, Wensveen J, Convergence Insufficiency Treatment Trial Study Group.. A randomized clinical trial of treatments for convergence insufficiency in children. Arch Ophthalmol. 2005;123(1):14-24. doi:10.1001/archopht.123.1.14. PMID:15642806.
Rouse MW, Borsting E, Hyman L, et al. Frequency of convergence insufficiency among fifth and sixth graders. The convergence insufficiency and reading study (CIRS) group. Optom Vis Sci. 1999;76:643-649. doi:10.1097/00006324-199909000-00022.
Pediatric Eye Disease Investigator Group. Home-Based Therapy for Symptomatic Convergence Insufficiency in Children: A Randomized Clinical Trial. Optometry and vision science : official publication of the American Academy of Optometry. 2016;93(12):1457-1465. doi:10.1097/OPX.0000000000000975. PMID:27575992; PMCID:PMC5118058.
Teitelbaum B, Pang Y, Krall J. Effectiveness of base in prism for presbyopes with convergence insufficiency. Optometry and vision science : official publication of the American Academy of Optometry. 2009;86(2):153-6. doi:10.1097/OPX.0b013e318194e985. PMID:19156012.
Yang HK, Hwang JM. Surgical outcomes in convergence insufficiency-type exotropia. Ophthalmology. 2011;118(8):1512-7. doi:10.1016/j.ophtha.2011.01.004. PMID:21474185.
