Prevalence and types of strabismus in cerebral palsy: A global and historical perspective based on a systematic review and meta-analysis

Abstract

Purpose.

Strabismus is more frequent in cerebral palsy (CP) than in the normal population, but reports differ how much it is increased. We here examined the global prevalence and types of strabismus in CP, whether esotropia or exotropia is more frequent, and whether the prevalence differs between ethnicities and/or country income levels, and between generations.

Methods.

We compiled in a systematic review and meta-analysis the results of 147 CP studies that report the prevalence of strabismus or the ratio of esotropia to exotropia, and we conducted subgroup analyses for region (income level) and ethnicity. We performed a pooled analysis for the CP strabismus prevalence, and estimated the global number of CP cases with strabismus.

Results.

The pooled prevalence of strabismus in CP is 49.8% in high-income countries and 39.8% in lower-income countries. We estimate the global number of strabismus cases in CP as 12.2 million, with 7.6 million males and 4.6 million females, based on current estimates of 29.6 million global CP cases. Esotropia is more frequent than exotropia in Caucasians, while exotropia is more frequent than esotropia in Hispanic and in some Asian and African populations. The strabismus prevalence in CP increases with increasing country income levels.

Conclusion.

Generational changes in strabismus prevalence appear to reflect a transition of CP types and an increase in prevalence as countries attain higher income and more effective maternal health care. The distribution of esotropia and exotropia in CP patients largely reflects the horizontal strabismus type that is predominant in the subject’s ethnicity.

Introduction

Strabismus is one of the most frequent co-morbidities of cerebral palsy (CP).^1–6 While there is a consensus that strabismus is much more frequent in CP than in the normal, non-CP population (where it is estimated at 2–3%),^7,8 reports of the prevalence of CP-associated strabismus differ considerably, ranging from 15% or less^9–13 to over 90%.^14–16 The prevalence of strabismus in CP differs not only between studies, but may differ between geographic regions or ethnicities. The only previous systematic review of strabismus in CP⁶ estimated the prevalence of strabismus at 48%, but Caucasians in high-income countries were over-represented, while Africans and Asians in lower-income countries were under-represented.

It is not known how many cases of CP with strabismus exist worldwide. Some of this uncertainty is due to controversies about the global prevalence of CP. CP is well-studied in Caucasian populations, in high-income countries, where the prevalence is about 2 per 1,000,^17–20 but the CP prevalence in lower-income countries (which is the large majority of the world population) is uncertain, with estimates ranging between 2 and 10 per 1,000.^{5,16,18,20–32} Since CP types differ between developing and high-income countries,^26,33–35 and most authors conclude that the type of CP associates with strabismus prevalence,^{5,6,10,11,28,33,35–47} it needs to be explored whether socioeconomic factors and/or ethnicity are significant variables for the strabismus prevalence among CP patients.

Another issue of contention is whether esotropia (ET) or exotropia (XT) is more frequent in CP. Initial studies, most of them examining populations of European ancestry, report more ET than XT, although some studies noted a lower ET/XT ratio in CP than in the general population.^15,48–53 Several of the more recent studies, many of them from East Asia and the South of the Indian subcontinent, reported more XT than ET in CP patients.^54–59 It has not been established whether the type of horizontal strabismus in CP is associated with the type of the underlying brain lesion, or whether it rather reflects the type of horizontal strabismus generally seen within the same ethnicity. We sought to answer the following questions:

1. What is the global prevalence of strabismus in people with CP? How much more frequent is it compared to the non-CP population? Can the global number of CP cases with strabismus be estimated?

2. Is the prevalence of strabismus in CP the same in Caucasians as in other ethnicities or regions (lower- vs high-income countries)?

3. Is the ET/XT ratio in CP significantly different between ethnicities?

4. Does the type of brain lesion in CP influence the type of horizontal strabismus (ET or XT), or does the direction of deviation reflect the type of horizontal strabismus that is predominant in the same ethnic population?

5. Are there longitudinal trends of strabismus prevalence in CP over generations?

To resolve these issues, we performed a systematic review and meta-analysis of strabismus in CP that was initially posted as a preprint.⁶⁰ Our review provides a comprehensive compilation of relevant studies, and we map their geographic distribution. We offer a global and historical perspective by exploring the role of ethnicity and socioeconomic factors.

Materials and Methods

Search Strategy

For our systematic review of the literature, we adhered to the PRISMA guidelines.⁶¹ Reports of studies were identified through a search of three databases: Google Scholar, PubMed, and CNKI (China National Knowledge Infrastructure), with unrestricted years. We used the keywords “cerebral palsy”, “strabismus”, “esotropia” and “exotropia” in Google Scholar, and “cerebral palsy” and “strabismus”, as well as “cerebral palsy” and “squint” in PubMed. Only English terms were used for the search strategy of the first two databases. We searched the CNKI database using the Chinese words for “cerebral palsy” and “strabismus” as key words. Reference lists from eligible articles were examined to find additional relevant studies. Studies were also identified by searching papers that cited relevant sources. All titles were screened, and when potentially relevant, the abstract was evaluated to decide whether the paper should be obtained for full-text reading to verify eligibility (Fig. 1). We failed to obtain an abstract and full text in two cases (2/708).

Fig. 1.

Flowchart of the Literature Search and Screening Strategy.

Inclusion/Exclusion Criteria

To be eligible for the systematic review, studies had to report the numerical prevalence of strabismus in humans with cerebral palsy (CP) and/or provide the ratio of esotropia vs exotropia in a CP cohort. CP was defined as a “disorder of movement and posture due to a defect or lesion of the immature brain”, a definition that was subsequently refined and expanded.^62,63 We excluded studies that reported on children with various other causes of visual impairment, unless they reported specific data for a CP cohort. We excluded disability conditions that can be similar to CP, such as premature birth, hydrocephalus, microcephalus, meningitis, developmental delay, Down syndrome, epilepsy, or autism spectrum disorders. We also excluded reviews only, case studies, abstracts at meetings when later published as a peer-reviewed paper, and papers focused on surgical outcomes. We included 54 studies on Caucasians (most of them in Europe, North America and Australia),^{9,10,14,36–42,44,48–53,64–101} 11 studies from the Middle East,^{5,16,23,102–109} 9 on Hispanics (most of them from Latin America),^15,110–117 41 from East Asia,^{12,13,43,46,56,58,118–152} 23 from South Asia,^{33,35,45,47,54,57,153–169} and 5 from Africa^11,170–174 (Supplemental Table 1).

Data Extraction and Analyses

Data were extracted by using pre-designed tables, including date of publication, first author name, country, geographic region, cohort size, number of cases of strabismus, and, when available, age range and the type of strabismus: horizontal vs. vertical, and among the horizontal strabismus, how many cases of esotropia, and how many cases of exotropia. We compiled information about gender distribution in the cohort and in strabismus cases when such data was reported. The percentage of strabismus cases was calculated from the number of cases per cohort. We conducted subgroup analyses between ethnicities, as well as between high-income and lower-income countries. Income levels were defined by the GDP as per the World Bank Report (1987 to 2022).¹⁷⁵ Because of ethnic differences between populations, the prevalence for each major ethnicity (European ancestry, African, Middle East, East Asian, South Asian, Latino/Hispanic) was estimated separately and weighted by population size to generate an overall estimate of the current global strabismus prevalence in CP. This was also necessary to prevent bias: the largest fraction of available studies examined people with European ancestry (in high-income countries). There were sufficient data for Caucasians, South Asians, and East Asians to assess generational (longitudinal) trends.

Statistical Analyses

The primary purpose of our meta-analysis was to generate a more precise and reliable estimate of the prevalence of strabismus among people with CP and to predict the global number of CP cases with strabismus. Pooled analyses were performed for strabismus prevalence in CP and the ET/XT ratio. The heterogeneity among studies was evaluated by Cochran’s Q test and the I² index.^176,177 The random-effect models were used to conservatively diminish the heterogeneity between studies.¹⁷⁷ The study weights were obtained based on the DerSimonian-Laird method.¹⁷⁷ A continuity correction of 0.5 was applied to studies with zero cases.¹⁷⁶ Subgroup pooled analyses were conducted by region/ethnicity, separately for prevalence and for the ET/XT ratio, to assess differences between Caucasians and other ethnicities. Meta-regression analyses were performed to test associations between risk factors and study outcomes. The risk of publication bias was evaluated using funnel plots and Egger’s test.¹⁷⁸ The significance level was set to 0.05. All meta-analyses were performed using the Stata SE 16.0 software (StataCorp, TX, USA).

Results

Our analyses are based on 147 studies, with 136 of them reporting the prevalence of strabismus in CP (total cohort number: 21,449), 102 partially overlapping studies reporting the esotropia/exotropia (ET/XT) ratio in CP cases (total cohort number: 13,613), and 90 partially overlapping studies reporting the gender distribution in the CP cohorts (total cohort number: 15,728). Altogether, these studies examined 21,753 CP cases in cohorts from 29 countries on five continents.

Global prevalence of CP and estimate of the total number of CP cases worldwide

While the prevalence of CP in developed (high-income) countries is well-established at about 2/1,000,^{2,19,20,24,26,44,49,64,71,179} there is less information about the prevalence of CP in lower-income countries.^{17,25,27–31} Initially, it was assumed that the CP prevalence was similar between developed and developing countries.^17–19,24 Recent studies have revealed a much higher prevalence of CP in many lower-income countries.^{20,25,26,30,31} For example, the prevalence of CP per 1,000 in India was reported as 2.8,¹⁸⁰ in rural Uganda it is 3,²⁷ in Bangladesh 3.4,²⁸ in Egypt 3.6,¹⁸¹ in Mexico 4.4,³² in Turkey between 4.4 and 5.6,^22,182 and in rural South Africa as 10.^21,25 These prevalences are thought to be underestimates.^20,28 Some authors noted that the prevalence of CP in lower-income countries is twofold larger than in high-income countries.^20,26,28 Very recent studies, based on new data and statistical modeling, estimated the prevalence and total number of global CP cases significantly greater, at about 6 per 1,000,^20,27,28,30 equivalent to about 50 million global cases.^29,31 However, the statistical modeling appears to include cases with motor dysfunction similar to CP, but not strictly with a CP diagnosis.²⁹ The much higher CP prevalence in lower-income countries has important implications for the estimates of the global prevalence of CP.

In our approach to estimate global CP numbers, we applied a prevalence of 2/1,000 for high-income countries, and 4/1,000 for lower-income countries (distinction according to The World Bank).¹⁷⁵ We estimate the CP cases in high-income countries (population of about 1.2 billion) to be 2.4 million (1.2 billion × 0.2%), and in lower-income countries (population of about 6.8 billion) to be 27.2 million (6.8 billion × 0.4%), which combines for a global total of 29.6 million CP cases. This estimate is intermediate between the previous, too low notion of 18 million global CP cases,¹⁸³ and the likely too high 50 million global cases.²⁹

Global prevalence of strabismus in CP: estimate of total numbers of CP cases with strabismus

The geographic distribution of studies reporting the prevalence of strabismus in CP is shown in Fig. 2A, with the prevalence indicated by a color gradient and the size of the cohort reflected by the size of the circles. The prevalence of strabismus in CP seems lower in some regions (Africa, for example) when compared with other regions. We therefore performed subgroup analyses to determine whether socioeconomic conditions or ethnicity are significant factors. Subgroup analyses for the major ethnicities were not significantly different, except for Caucasians vs. East Asians (p=0.001, Fig. 3A). Subgroup analysis by country income level showed a significantly higher strabismus prevalence for high income (49.8%, 95% confidence interval CI=44.3–55.2%) than for lower income (39.8%, CI=34.4–45.1%, p=0.004). The global strabismus prevalence was 41.3% when adjusted for the population size of high-income countries (1.2 billion) and lower-income countries (6.8 billion, Table 1A). When the differences in CP cases between regions as well as the differences between prevalence of strabismus in CP cases are taken into account, we can use the above estimate of the global number of CP cases (29.6 million) to calculate the total number of CP cases with strabismus worldwide – as 12.2 million (29.6 × 41.3%).

Fig. 2.

A, B. World maps showing the distribution of studies reporting the prevalence of strabismus in cerebral palsy (A) and the esotropia/exotropia (ET/XT) ratio in cerebral palsy (B). The cohort size is indicated by the circle size and the prevalence or ratio is indicated by the color gradient. In A, note that lower-income countries (Africa and India) have a somewhat lower prevalence than high-income countries. In B, the ET/XT ratio below 0.9 is indicated in red, a nearly equal distribution (0.9 to 1.1) is indicated in tan, and a ratio above 1.1 is indicated in blue. Note that regions with Caucasian populations mostly have a high ET/XT ratio, while some populations in Africa, East Asia, Southern India and some Hispanic populations have a low ET/XT ratio.

Fig. 3.

A, B: Forest Plots of the Prevalence of Strabismus in Cerebral Palsy. A. Sorted by ethnicity. B. Sorted by lower-income countries vs high-income countries – a statistically significant difference (p=0.004, see Supplemental Fig. 2A). Note the shift in high-income countries towards a higher prevalence of strabismus. The diamonds show the pooled prevalence for the listed studies.

Table 1.

Prevalence of strabismus in cerebral palsy (CP): Data and estimates of numbers (A). Esotropia (ET) and exotropia (XT) in CP: Data and estimates of numbers (B).

A.
Ethnicity/Region	Population	Studies	Cohort	CP with Strabismus in cohort	Cohort: CP Strabismus Prevalence	CP with Strabismus, Cases in population (estimate)	CP Cases in population (estimate)
	billion	number	number	number	percent	million	million
Caucasian	1.2	53	6,329	2,711	48.8	1.17	2.4
Middle East	0.5	9	584	300	48.6	0.97	2.0
Hispanic	0.7	6	1,016	677	60.8	1.70	2.8
African	1.3	5	736	263	31.5	1.64	5.2
East Asian	2.4	40	7,182	2,287	35.7	3.43	9.6
South Asian	1.9	23	5,602	1,990	43.0	3.27	6.0
Global (Total)	8.0	136	21,449	8,228	41.3	12.2	29.6
B.
Ethnicity/Region	Population	Studies	Cohort	Cohort CP cases with ET	Cohort CP cases with XT	ET/XT ratio in Cohort	CP with ET, entire population (estimate)	CP with XT, entire population (estimate)
	billion	number	number	number	number		million	million
Caucasian	1.2	42	5,167	2,747	1,481	1.76	0.733	0.417
Middle East	0.5	9	590	235	135	1.56	0.579	0.371
Hispanic	0.7	9	1,098	303	331	0.94	0.809	0.861
African	1.3	4	587	84	155	0.85	0.739	0.871
East Asian	2.4	22	3,281	590	451	1.40	1.966	1.404
South Asian (North)	1.5	6 7	1,800	496	120	2.99	1.754	0.586
South Asian (South)	0.4	9	1,090	228	267	0.94	0.417	0.443
Global (Total)	8.0	102	13,613	4,683	2,940		6.997	4.953

Footnote: The number of studies differs between prevalence and ET/XT ratio, because not all studies reported overall prevalence and ET/XT ratios, although most studies reported both. The estimates for global numbers of ET and XT cases in CP do not include the estimated number of pure vertical deviations which are about 0.2 million cases. We distinguish between South Asians in the North and South for the ET/XT ratio, because of the significant difference between these populations as discussed in the text and illustrated in Fig. 2B.

Frequencies, numbers and ratios of esotropia and exotropia cases in CP

The large majority of strabismus cases in CP are horizontal deviations, either esotropia (ET) or exotropia (XT), with a much smaller percentage of strabismus cases that have an exclusive vertical deviation (1.8%, CI=1.2–2.4%), based on 5,092 CP cases with information on hypertropia, Supplemental Table 1). The predominance of horizontal over vertical deviations is similar in non-CP and in CP populations. In the rest of our review, we will focus on horizontal strabismus. Most studies (74.5%) report more ET than XT in cases of CP (Fig. 2B; Supplemental Table 1). Populations with more ET than XT include Caucasians in Europe and North America, and most of the populations in the Middle East, South Asia, and East Africa. Populations with more XT than ET include parts of East Asia, the South of the Indian subcontinent, West Africa, and Hispanics in Central America (Figs. 2B; 4). Differences in the ET/XT ratio reached significance in Caucasians vs Hispanics (p=0.036) and was borderline for South Asians (South) vs Caucasians (p=0.062). Based on the estimate of the global prevalence of CP (see section above), we calculated the total number of CP cases with ET or with XT. Weighted by population size, the total global number of CP cases with ET is 7.0 million, and the total global number of CP cases with XT is 5.0 million (Table 1B).

Fig. 4:

Forest Plot of Esotropia/Exotropia (ET/XT) Ratios in Cerebral Palsy sorted by Ethnicity. Ethnicities with higher ET/XT ratios (higher than 1.5) are compiled on the left side, ethnicities with lower ET/XT ratios (below 1.5, several below 1.0) are compiled on the right side of the forest plot. The diamonds show the pooled ratios for the listed studies.

Gender distribution of strabismus in CP

Among our eligible CP studies, 89 report the gender distribution in the cohort, but only two of these studies also report the gender of the cases with strabismus.^166,167 In CP cohorts, males nearly always exceed females, overall by a factor of about 1.5 to 1.¹⁸⁴ The same is true for “our” CP cohorts – the ones that have been examined for strabismus (n=15,728), with 9,808 males and 5,920 females), which is a male/female ratio of 1.66 to 1 (7.6 million males and 4.6 million females). Two of these studies disclosed gender among strabismus cases (total cohort of 484)^166,167; they indicate that there is no gender difference – males and females appear to contribute equally to the strabismus cases in the CP cohort, with the male and female prevalence of strabismus not being significantly different (p=0.440). In other words, there are more males with CP-associated strabismus than females, but only because more males than females have CP.

Longitudinal analysis: strabismus prevalence in CP has increased over decades

To determine whether the prevalence of CP-associated strabismus has changed between generations, we examined the prevalence data in three ethnicities: Caucasians, East Asians, and South Asians (Supplemental Table 1). There was a significant trend in Caucasians from about 35% to 55% prevalence of strabismus in CP between 1950 and 1995, and the trendline slightly decreased from about 55% to 50% between 1995 and 2022 (Fig. 5A). The data for East Asians also showed an increasing trendline (Fig. 5B), from about 28% to 37%, but a nearly flat trendline for South Asians (Fig. 5C). These trends are consistent with the notion that strabismus prevalence in CP is associated with socioeconomic factors (see Discussion).

Fig. 5.

A-C. Trends over decades in the prevalence of strabismus in cerebral palsy (CP) in Caucasians (A), East Asians (B), and South Asians (C). Note the increasing trend from about 35% to 55% (significant with p=0.039) in Caucasians from 1950 to 1995, and a near stable trend thereafter (A). East Asians show a more gradual increase from 1995 to 2022 (B), while the trendline for South Asians is nearly flat (C). These data are consistent with the notion that with more effective maternal health care, the prevalence of strabismus in CP increases, likely due to fewer cases of CP with perinatal etiology and increased numbers of surviving premature babies, and then reaches a plateau.

Discussion

An association between CP and strabismus was first noted in the middle of the 19^th century; for a review of the early literature, see Smith.¹ In the second half of the 20^th century, a series of more detailed studies from Europe and North America reported on the prevalence of strabismus in CP (Supplemental Table 1). Until the year 2000, most studies (39/49=79.6%) were carried out on Caucasian populations, thereby creating a Eurocentric bias. More recent studies examined populations from other regions of the world, notably in South Asia, East Asia and the Middle East (Table 1A; Supplemental Table 1), allowing to gain a better global perspective.

Global prevalence of strabismus in CP

We estimate the total number of CP cases worldwide at 29.6 million (Table 1A). Given the incomplete ascertainment of CP cases as well as survival bias in developing countries, our estimate of 29.6 million CP cases globally may be an underestimate.^{20,25,27,28,34} How many of these CP patients have strabismus? The only previous systematic review estimated a 48% strabismus prevalence in CP which was based on 17 studies, with a combined cohort of 1,734.⁶ We find a lower global prevalence, of 41.3%. Our estimate is based on a much larger number of CP cases (21,449 cases), nearly 13-fold larger than the cohort size in the previous estimate. The higher prevalence estimates^3–6,185 are likely due to bias towards Caucasian populations in high-income countries.

The CP prevalence is known to increase with lower socioeconomic status.^{18,25,28,34,186–188} The recent studies from Africa and from South and East Asia are consistent with this notion. We show that lower-income countries (LICs) have a lower prevalence of CP-associated strabismus than high-income countries (HICs) (Fig. 3B). The key to understanding this surprising finding likely is that the timing of the brain insult and the mechanism of CP (and possibly strabismus) differ between LICs and HICs (Supplemental Fig. 2B). In LICs, the majority of CP cases are caused by peri- and postnatal events: asphyxia during delivery, or post-partum infection, while in HICs, the majority of CP cases are due to preterm births – when the immature brain is more vulnerable to lesions of the visual pathways.^{26,28,33,34,185,189,190} Additional factors complicate this basic pattern, in that very preterm babies rarely survive in LICs, but are more likely to survive in HICs, and that severe cases of CP are also more likely to survive in HICs, while in LICs, many may die prior to the age of CP diagnosis.^24,27,28,34 This explanation is consistent with the observed data of a lower CP prevalence in HICs, but a higher prevalence of strabismus in CP cases in HICs than in LICs.

The trendlines of strabismus prevalence in CP (Fig. 5) are consistent with the prediction that developing countries will undergo similar shifts in CP types as developed countries did several decades ago – when more effective maternal health care systems were implemented. With an increasing number of studies from Asian, African and Hispanic populations, it now is possible to estimate the global prevalence of CP-associated strabismus and the global number of such cases. Accordingly, we estimate about 12.2 million such cases worldwide. This allows us to calculate how many cases of strabismus, due to CP, need to be added to the number of strabismus cases in the normal (non-CP) population. The normal (non-CP) population is thought to have a strabismus prevalence of 2–3%.^7,8 Since most of the epidemiological studies rely on surveys of normal schools,^8,191 many of the CP-associated strabismus cases (the ones with more severe CP) would have been missed. The ~12.2 million CP strabismus cases comprise 6% of the total global number of strabismus cases (2–3% of a population of 8 billion people translates to 160–240 million cases).^7,8

Variation between studies

There is considerable variability in the prevalence of CP-associated strabismus between studies, ranging from less than 15% to over 90%, I²=97.9 (Table 1A; Fig. 3A,B; Supplemental Table 1). Multiple factors, not mutually exclusive, may contribute to this variability. First, the severity of CP differs between cohorts, and severity of CP is associated with strabismus prevalence. When cohorts primarily comprise mild cases of CP, they tend to have a lower prevalence of strabismus, while cohorts with severe cases have a higher prevalence of strabismus^9,37,192. Second, the types of brain lesions differ between studies and between populations, due to differences in maternal health care as mentioned above. Third, ethnicity may be a significant variable, due to differences between populations in CP prevalence.^193,194 Whether socioeconomic factors can fully account for such ethnic differences is controversial.^193,195 Ethnicity may affect the type of strabismus (ET vs XT, see below), as well as the overall prevalence of strabismus.¹⁴² Fourth, methodological issues likely play a role, especially the criteria used to identify and diagnose CP cases which varies between countries and studies.^19,29,34,188 Finally, there likely is natural variation between populations and also between risk factors for both CP and strabismus that may change over time – e.g., maternal smoking is a major risk factor for strabismus,^196,197 regardless whether the offspring has CP or not. All of these factors likely contribute to the observed variation between studies.

Is strabismus prevalence associated with CP types?

Strabismus prevalence is associated with the severity of CP, visual impairment, and also with the extent of intellectual capacity.^{3,5,37,38,45,46,48,58,70,115,153,159,198–202} Relatively few authors disagree with this conclusion.^39,74,101 Some authors noted that strabismus is most frequent in spastic tetraplegia (or triplegia) and somewhat less in diplegia, still less in hemiplegia, and rare in athetoid or ataxic CP,^{5,11,36,39,40,45,47,64,70,101,105,149,163} while others disagree with this notion.^{15,41–43,52,56,67,78,100,111,203} Authors concur that there is no association between the type of strabismus (esotropia, exotropia, hypertropia) and the type of CP.^{15,46,51,52,67,100,114}

The esotropia/exotropia (ET/XT) ratio in CP

Our work has revealed differences between populations in how prevalent ET and XT are in CP (Figs. 2B, 4). Why do Caucasians, North Indians and people in the Middle East have more ET than XT, while populations in West Africa, Hispanics in Central America, and South Indians have more XT than ET? CP increases the likelihood of strabismus, but whether the strabismus will be ET or XT depends largely on the ethnicity, presumably due to differences in orbital anatomy, as first proposed by Holm²⁰⁴ and Waardenburg.²⁰⁵ The ET/XT ratio appears to be changed in CP (closer to 1), as has been noted by several authors for Caucasians^{15,48,50–53} and for East Asians.⁵⁶ It seems that CP “tempers” the extremes and makes the ET/XT ratio more balanced than it is in the non-CP population (of the same ethnicity).

Gender distribution

In the CP cohorts examined for gender, males dominate over females by a ratio of 1.66:1. This ratio is similar to the known male dominance in CP (about 1.5 to 1).¹⁸⁴ Among CP cases, the strabismus prevalence appears to be the same between males and females (although this is based on sparse data). Because of the male dominance in CP, there are more males with CP-associated strabismus than females. In the normal (non-CP) population, the prevalence of strabismus is the same between males and females.²⁰⁶ Our data suggest that there is a male predominance of CP-associated strabismus (7.6 million males vs 4.6 million females), which would indicate a 3 million male/female difference in the global cases of all strabismus, when CP-associated strabismus cases are included in the count.

Pathogenesis of strabismus in CP

It is currently controversial what causes strabismus in CP; three different explanations have been proposed: (1) spasticity of extraocular muscle, (2) lesion of cortical or subcortical motor circuits, and (3) lesion of cortical or subcortical visual circuits. We will discuss these scenarios in sequence and evaluate their merits.

Extraocular muscle spasticity (similar to the limb spasticity seen in CP) as a mechanism for strabismus was discussed by Smith,^1,70 and proposed by Sandfield-Nielsen et al.²⁰⁰ This notion was essentially refuted by Smith^1,70 and Kalbe et al.⁵¹ It was pointed out that extraocular muscles are not spastic in CP, and there is no reason why any such “spasticity” should selectively affect certain extraocular muscles and spare others. While strabismic extraocular muscles do exhibit abnormal extracellular matrix gene and protein expression,²⁰⁷ which resembles abnormalities in CP limb skeletal muscle,²⁶ such dysregulation may be a general response of muscle tissue to abnormal functional demands, and does not necessarily imply similarity in pathogenesis.

Some authors state that subcortical motor lesions (midbrain, brainstem) may cause strabismus in CP.^{1,45,54,94,155,167} Lesions of basal ganglia and cerebellum (dyskinetic and ataxic CP) less often associate with strabismus.^{10,36,37,47,105,189}

Other authors implicate lesions of cortical or subcortical visual pathways as the primary cause of strabismus in CP.^{1,46,56,56,189,208,209} The significance of impaired visual pathways (optic radiations and/or cortical areas involved in visual motion processing) for strabismus in CP is supported by imaging studies (CT, MRI).^{189,208,210–214}. Disturbance of motor pathways appears to play a lesser role in the pathogenesis of strabismus in CP than compromised visual pathways. It was noted that the lesion of subcortical visual pathways – more frequent in preterm births due to the immaturity of the pathways – appears to cause more esotropia, while lesions of visual cortex more often cause exotropia^189,214,215 – but some authors disagree.²⁰⁹ Such differences in the direction of horizontal misalignment are consistent with prenatal lesions being more frequent in high-income countries, and perinatal lesions being more frequent in lower-income countries. Overall, the direction of horizontal strabismus in CP appears to be determined primarily by ethnic/genetic factors rather than socioeconomic factors or timing of lesions (prenatal vs perinatal). Regions with a lower ET/XT ratio (South India) utilize maternal health care more frequently than regions with a higher ET/XT ratio (North of India),²¹⁶ so a shift to more prenatal insults than perinatal insults cannot explain the lower ET/XT ratio in South India (Fig. 2B).

Conclusion

Differences in maternal health care between lower-income and high-income countries are important to understand global and regional differences in CP and to predict trends in CP prevalence as well as the prevalence of strabismus in CP.