For an object to be perceived with clarity, it is imperative to ensure a steadfast fixation of its image onto the fovea of the retina. Nystagmus, an involuntary rhythmic oscillation of the eyes, can disrupt the fixation and result in blurred vision and illusory motion (Thurtell and Leigh, 2011). The manifestation of these visual symptoms can induce considerable distress and result in substantial disability, impeding various aspects of an individual’s self-esteem, social functioning, and quality of life (QoL) (Pilling et al., 2005; McLean et al., 2012). The sign of nystagmus is rhythmic, regular, and sustained compared to saccadic intrusions (Rucker, 2019). The mechanisms of nystagmus remain incompletely understood. However, based on the waveform characteristics and, primarily, the onset time, nystagmus is categorized into two main types: infantile nystagmus (IN) (previously termed congenital nystagmus) and acquired nystagmus (Papageorgiou et al., 2014).

Typically, IN is present and diagnosed from the age of 3 ~ 6 months onwards while acquired nystagmus appears later (Papageorgiou et al., 2014; Ospina, 2018; Fossataro et al., 2024). It is common for IN to manifest as an isolated symptom, whereas the majority of cases in acquired nystagmus exhibit a confluence of additional indicative factors that point toward the underlying etiology (Ospina, 2018). Acquired nystagmus often arises from lesions affecting the visual feedback loop, the vestibular-ocular reflex circuits, and the neural eye position integrators, while IN is characterized by different but enigmatic etiologies (Rucker, 2019; Leigh et al., 2002). Various hypotheses about the pathophysiological mechanisms of IN have been proposed, including instability within the optokinetic system, dysfunction of the neural integrator, aberrations in the foveal pursuit system, and conflicting interplay between the primitive subcortical optokinetic pathway and cortical smooth pursuit system (Yee et al., 1980; Optican and Zee, 1984; Jacobs and Dell’Osso, 2004; Brodsky and Dell’Osso, 2014). Mutations in FRMD7, TUBB3, or L-dopachrome tautomerase were demonstrated to cause IN (Tarpey et al., 2006; Jin et al., 2021; Volk et al., 2021). The heterogeneity of the causes adds up to the challenge of distinguishing IN from symptoms secondary to other conditions, complicating the clinical decisions and its systematic overview.

Although the exact pathophysiological underpinnings of IN are yet to be elucidated, clinical treatment is needed, as IN represents the predominant form of nystagmus, accounting for the majority of the cases (58%), particularly in pediatric practice (75–83%) (Sarvananthan et al., 2009; Ehrt, 2012; Hertle et al., 2022). Patients with IN often do not receive treatment, either due to an absence of complaints or the prevailing belief that individuals must adapt to living with the condition (Oogfonds, 2023). For patients in need, treatment options do exist and various academic studies have explored different treatment approaches for IN (Hertle, 2022; Self et al., 2020). However, the available literature summaries in this field are either focusing on acquired nystagmus or nystagmus generally (Thurtell and Leigh, 2010; Tegetmeyer, 2014; Thurtell, 2015; Käsmann-Kellner, 2017; Strupp et al., 2021), limited to a single treatment modality (Zahidi et al., 2017; Cham et al., 2021; Chang et al., 2023), outdated (Abel, 2006; Richards and Wong, 2015), constrained in geographic locations (Self et al., 2020), or written in a non-English language (Pan, 2021). Moreover, almost none of these summaries encompass the significant body of behavioral studies that have demonstrated promising potential in the treatment of IN in recent years.

Before reviewing treatment methods, it is important to highlight key characteristics of IN: Key clinical features include predominant horizontal oscillations, with vertical and torsional oscillations being less common (Ospina, 2018). IN exacerbates during distance fixation and diminishes when the patient engages in near-vision tasks that involve convergence (Rucker, 2019; Ospina, 2018). A null zone, where nystagmus intensity decreases, is common. If offset from the primary position, individuals may adopt an anomalous head posture (AHP) to enhance foveation (Ospina, 2018). This posture commonly involves a head turn or head tilt. IN can occur in association with ocular disease, neurological disorders, or genetic syndromes, though many cases are idiopathic (Self et al., 2020; Käsmann-Kellner, 2017). A Danish study of 103 patients found a 32% prevalence of idiopathic infantile nystagmus (IIN) among children with IN (Hvid et al., 2020).

IN treatments generally include causal and symptomatic approaches. This review focuses on symptomatic treatments, often used for IIN or when causal options are unavailable, as in albinism or optic nerve hypoplasia (ONH) (Lohmüller et al., 2017; Liu et al., 2021). Symptomatic treatments are mostly considered in patients with abnormal gaze, restricted head posture, or decreased visual acuity (VA) (Self et al., 2020; Richards and Wong, 2015). Common evaluation of patients with IN includes nystagmography, gaze and posture, best-corrected visual acuity (BCVA), reading performance, and QoL (Self et al., 2020; Käsmann-Kellner, 2017). Articles published before 2015, including work by Bedell, Dell’Osso, and others, have been extensively reviewed. This updated review focuses on studies since 2015. After pre-screening, we identified four categories of current interventions: surgical, pharmacological, optical, and behavioral. Clinically, refractive correction is always the first step in addressing low vision in IN (Self et al., 2020). Thus, only articles emphasizing optical correction as the primary intervention to reduce nystagmus or improve VA are included in the optical modality section.

The initial search terms resulted in 97 articles and an additional 7 articles ranging from 2004 to 2011 were included by citation searching. After careful screening, 60 journal articles were selected in the full-text review based on their abstracts according to the inclusion criteria. Of the included articles for full-text review, 8 articles were excluded after examination. The remaining 52 articles comprised 8 review articles, 43 journal articles, and 1 correspondence. The journal articles were categorized by the primary type of intervention. The results were as follows: 23 studies were found on surgical interventions (from 2015 to 2023), 7 covering pharmacological therapy (from 2007 to 2023), 5 studies about optical correction (from 2004 to 2023), 7 studies about behavioral training (from 2015 to 2023), and 1 article with no intervention (2017). For an overview of the selection process, see the PRISMA flow chart (Figure 1). Additionally, this review comprehensively examined 8 comprehensive review articles ranging from 2015 to 2023, with two focusing on surgery for IN (Cham et al., 2021; Chang et al., 2023), one on optometry (Zahidi et al., 2017), two written in German (Tegetmeyer, 2014; Käsmann-Kellner, 2017), and one in Chinese (Pan, 2021), one general review paper on IN (Richards and Wong, 2015), and one broader review on nystagmus (Ospina, 2018).

Supplementary Tables 1–4 present information across the four treatment modalities in IN, including the study design, outcomes in VA, nystagmography, QoL, and complications. The 42 journal articles with intervention comprise of randomized control trials (n = 7), non-randomized intervention studies (n = 14), case–control study (n = 1), cross-sectional studies (n = 3) and retrospective case report/series (n = 17), with no cohort study (n = 0). Numbers of different types of studies under each modality can be found in Table 2. Studies often only report specific outcome measures (Figure 2). Significant methodological heterogeneity across the three types of outcomes hinders the comparison of treatment effects between studies (Supplementary Tables 1–4). An example in nystagmus parameters is shown in Figure 3.

The present review comprehensively examines the current state of research on the symptomatic treatment of IN. By critically evaluating the available evidence, this discussion aims to shed light on the effectiveness, limitations, and future directions of various treatment approaches. In this section, we will delve into the main findings, address unresolved questions, and explore the implications of these findings for clinical practice and future research.

Broadly speaking, the primary objective of surgical and pharmacological interventions (Supplementary Tables 1, 2) is to reduce nystagmus, whereas the main goal of optical and behavioral approaches is to enhance visual function (Supplementary Tables 3, 4). In the case of the former interventions, the observed improvements in VA were relatively modest. However, there are indications that visual functions, such as recognition time, contrast sensitivity and motion processing improve by improving the nystagmus waveform, even if VA does not increase measurably (Käsmann-Kellner, 2017). Based on the gathered information, surgical intervention remains the predominant treatment modality for addressing IN. It was advised that it should be avoided in the cases of children under 5 years old because the pre-surgery measurement and assessment are challenging due to a lack of cooperation (Käsmann-Kellner, 2017). Meanwhile, surgery on patients under 2 years old demonstrated promising outcome in ocular motor and visual system (Hertle et al., 2009). Even though there are multiple surgical methods, there is no targeted pathology. The same situation also stands in medication. Currently, most of the pharmaceutical studies are off-label and there was no study on the long-term effect of medication for IN, which impeded its clinical use (Ospina, 2018). Additionally, the pharmacological treatment for nystagmus is typically reserved for adults who present with distressing visual symptoms, such as oscillopsia. Given the rarity of these symptoms in early-onset nystagmus, pharmacological intervention is often deemed unnecessary (Richards and Wong, 2015). Among the medications examined, brinzolamide exhibited promising outcomes in enhancing VA, with the added advantage of its administration via eyedrops, which contributes to its safety profile (Hertle et al., 2015; Dell'Osso et al., 2011; Nieves-Moreno et al., 2017; Aygit et al., 2018; Yadav et al., 2024). As previously mentioned, a longitudinal study without any intervention demonstrated a slight improvement in VA throughout a lifetime tested before 18 (Balzer et al., 2018). This observation naturally leads us to think that the self-correcting influence of daily experiences may have a positive impact on individuals with IN and perceptual training can be seen as an accelerated form of this natural process. The reduction in population receptive field (pRF) size in V2 and the alleviation in visual orientation crowding provided additional evidence supporting the efficacy of perceptual training and its impact on neuronal plasticity (He et al., 2019). However, as the measurement results (e.g., VA) vary among individuals and the follow-up periods in clinical cases are often limited, we also suggest researchers take caution and do not rely solely on the significance (e.g., p-value) of a univariate approach (Morey et al., 2016).

Although we provide a systematic review of recent symptomatic IN treatments, the articles reviewed had several overarching limitations. Firstly, there was a lack of standardization in outcome measurements (Figures 2, 3, Supplementary Tables 1–4). Due to the heterogeneity of the outcome measurements, conducting a quantitative comparison of treatment types or cross-benefit analysis was impossible. Different units or measurements for certain parameters, such as VA represented by either logMAR units, Snellen fraction, Snellen line gain, or decimal acuity, complicated the comparison between studies and treatment options (Wesemann et al., 2020) (Supplementary Tables 1–4). The eXpanded Nystagmus Acuity Function (NAFX) index has been proposed as a measure of VA (Khanna and Dell'Osso, 2006). However, there is evidence of limitations in using NAFX and other eye movement measurements for patients with apparent visual issues and its inefficacy in patients with albinism (Hertle and Dell’Osso, 2023). Moreover, those measurements are not often direct indicators of visual function (Hertle and Dell’Osso, 2023). The main goal of symptomatic treatments of nystagmus is to improve the QoL of the patients, raising the question of whether more emphasis should be placed on patients’ well-being and real-life performance or their nystagmus signs. Even though the reduction the nystagmus intensity could improve QoL via improving visual functions, the QoL evaluation was absent in most of the studies other than perceptual learning (Figure 2). We recommend that clinicians prioritize the primary objective of improving the overall well-being and daily functioning of patients, assess and select treatment options accordingly, and report QoL evaluation and related functional factors such as binocular VA, gaze dependent VA, contrast sensitivity, visual task performance including recognition time, etc. For future trials and clinical studies, it is highly suggested that we include, and report detailed settings in a unified way. One example would be, for VA, specifying near or distant and crowded or non-crowded VA would different aspects of participants’ sensory states. This would also help further quantitative comparison across studies since if the test methods are sparse, the analysis would be underpowered. Moreover, only after a systematic report of standardized nystagmus measurements, questions such as which nystagmus trait affects VA the most could be studied.

The second limitation of the reviewed articles is that most of the studies were clinical trials with small sample sizes and short follow-up times. More long-term studies are needed in the field since a lasting effect or an adverse effect would largely alter the benefit of the intervention (Fossataro et al., 2024). In that way, if the effect is similar, non-surgical intervention would be preferred which contradicts the status quo.

The third limitation of the current research is the lack of information of personalization and categorization of the patients. IN is complicated and it relates to multiple other illnesses which makes handling of screening and diagnosing data across studies a challenge. If a systematic report of personalization and categorization procedure were available in every study, a better insight into which intervention would fit which type of patients would be feasible when researchers and clinicians line up the inclusion criteria and treatment methods (Self et al., 2020; Käsmann-Kellner, 2017).

It is nevertheless important to note that many of the articles reviewed were retrospective studies, and there may be some overlapping in the statistics presented. We classified them into different categories to give general insight into treatment options, but they can be essentially divergent. Furthermore, this review exclusively includes articles concerning studies in humans, while nystagmus models in zebrafish and mice could offer valuable insights into the fundamental nature of the disease (Bögli et al., 2017; Winkelman et al., 2019). On the other hand, integrating mechanical (Beh et al., 2014) or magnetic (Nachev et al., 2017) force has shown potential in dampening acquired nystagmus, which can also be inspiring for future applications for IN treatment. Lastly, it is often unclear whether the articles examined have adopted the definition of IN as outlined by the Classification of Eye Movement Abnormalities and Strabismus (CEMAS) Working Group in 2001, which distinguishes it from other types of early-onset nystagmus such as fusion maldevelopment nystagmus syndrome (FMNS), spasmus nutans syndrome, etc. (CEMAS, 2001). Readers should be mindful of this consideration.

Methodological heterogeneity in outcome measures is ubiquitous in the current research field of symptomatic treatments of infantile nystagmus. We conclude that the field needs to adopt standardized outcome measures to enable quantitative comparison of intervention approaches and meta-analyses. To support patient well-being, we further suggest that functional vision measurements and quality of life should be included in future research.