Detection of ocular hypertelorism among Indian children

Irregularities of growth and shape in craniofacial region disturb the harmonious structural relationship between the cranium and face of the growing children1. Therefore, for the detection of aberrant growth patterns in the facial region of children beyond normal variation, understanding of the auxological dynamics of the ocular apparatus becomes important. Ocular hypertelorism (representing an increased distance between centres of both the pupils) is known to be constituent of many clinical syndromes including, Crouzan’s disease, Cleidocranial dysostosis, Apert, Hurler’s, Cri-du-chat, Oto-palato-digital and Basal Cell Nevus syndromes, etc23. While, ocular hypotelorism (depicting decreased distance between centres of the pupils of two eyes) is associated with Mechel, Coffin-Siris and, William’s syndromes, and chromosomal disorders involving some partial trisomy etc3. It may present as an unusual morphologic feature without any medical or cosmetic consequence3. An optical illusion produced by a flat nasal bridge, lateral displacement of inner canthi or short palpebral fissures, epicanthal folds, widely spaced eyebrows, often makes evaluation of ocular hypertelorism imprecise and misleading. To diagnose true ocular hypertelorism accurately, the distance between both inner and outer canthi should be increased. Furthermore, hypertelorism must be differentiated from telecanthus or pseudohypertelorism in which the lateral displacement of the inner canthi may give a false impression of widely spaced eyes4. Therefore, a need to objectively evaluate hypertelorism becomes obvious.

Inter-pupillary distance (IPD) is considered to be the most accurate and reliable measure for the quantification of ocular hypertelorism as the centre of the pupil remains unaffected by the soft tissues surrounding the eye5. Available anthropological evidence confirms the existence of significant differences in the craniofacial growth patterns of children representing various ethnic and racial groups. Available data on IPD of Turkish6, Nigerian7, American89, Iranian10, Ghanaian11 populations exhibit substantial inter-population variation and hence, cannot be used to evaluate and compare the growth of IPD among children of Indian origin. However, detection of ocular hypo/hypertelorism in terms of inter-pupillary index derived from IPD and fronto-occipital circumference is another method suggested by Evereklioglu et al12. Furthermore, different methodologies employed previously to measure and calculate IPD not only make inter-population comparison difficult but are also scientifically imprecise13. It was, therefore, emphasized that for the purpose of making a precise comparison, norms derived from homogenous racial stocks should be available5. To pick any possible anomaly, understanding of the growth dynamics of the craniofacial region, particularly affecting ocular apparatus becomes important.

Available anthropometric data sets on IPD of Indian children were either drawn from a small sample size1415 or the values were merely presented for the first year of life2. Owing to the absence of any authentic age and gender-specific anthropometric data on the pattern of growth of IPD, the present study aimed to check for the auxological dynamics of IPD among normal, healthy children residing in northwestern parts of India.

Material & Methods

A total of 3622 (2239 males and 1383 females) normal, healthy children aged one month to 14 yr who were born to parents from mixed socioeconomic strata, and residing in northwestern parts of India constituted sample for this cross-sectional study. These children were consecutively enrolled over a period of 24 years between April 1994 and September 2018 from the Growth Clinic/Growth Laboratory of the Advanced Paediatrics Centre, Postgraduate Institute of Medical Education & Research (PGIMER), Chandigarh, where they visited for either scheduled immunization, routine health check-up, or nutritional growth assessment etc. Children suspected with any disease of the central nervous system or any chronic disease or congenital anomalies were excluded from the study. Furthermore, severe under-nutrition, physical or craniofacial anomalies were also factors for exclusion. Informed written consent from parents of the participating children was obtained. The study protocol was approved by the Institutional Ethics Committee.

The direct measurement of IPD in awake and unco-operative infants and young children is difficult, which often yields faulty data. To overcome difficulties encountered in the measurement of this dimension, the formula by Feingold and Bossert (1974)13 was used to derive IPD from outer canthal distance (OCD) and inner canthal distance (ICD). All children were measured for ICD (i.e. distance between the two endocanthion points) and OCD (i.e. distance between the two ectocanthion points) using the digimatic sliding caliper (least count: 0.01 mm, Mitutoyo, Japan) in the study centre following standardized anthropometric measurements2. The measurement in children less than two years was taken with the caliper on a supine child, with and head arms restrained by an attendant. All anthropometric measurements were taken by the well-trained staff of the unit for whom intraclass correlation coefficient to check reliability of different measurements were periodically calculated on a sub-sample of 50 subjects and were found to be 0.795 for OCD (excellent agreement) and 0.816 for ICD (almost perfect agreement).

The IPD (i.e. distance between the centre of pupils of the two eyes) for each individual at each age was then calculated in centimetres from the ICD and OCD measured in each child by using the following formula:

IPD (cm)=0.7+ 0.41 (OCD)+0.59 (ICD)

Statistical analysis: The normality of data was checked using Kolmogorov–Smirnov tests of normality. Since the data were normally distributed mean in the present study, standard deviation (SD) were computed for the parameters namely IPD, ICD and OCD at each age among male and female children. Student’s unpaired t test was employed to compute gender differences. The inter-population difference for mean IPD among children representing different population groups was calculated using online GraphPad calculator. A P<0.05 was considered as significant.

A total of seven age- and sex-specific percentiles (i.e. 3^rd, 10^th, 25^th, 50^th, 75^th, 90^th and 97^th)16 were obtained for IPD from ages one month to 14 yr using the formulae given by Tanner et al17.

Correlation of ICD and OCD with IPD was calculated using Pearson correlation-coefficient at each age level.

Results

The mean (SD) for IPD, ICD, OCD are presented in Table I. Percentile grids for IPD among male and female children are depicted in Table II and III while, mean increase for OCD, ICD and IPD is shown in Table IV. A regular increase in IPD similar to its constituents, i.e. ICD and OCD among both boys and girls was recorded throughout the period of the study. However, this increase was found to be faster during the first year of life after which, these parameters grew slowly and uninterruptedly until 14 yr (Figs 1 and 2). The mean OCD, ICD and IPD were 65.8 (4.25) mm, 21.6 (1.79) mm and 4.68 (0.21) cm in boys at one month of age, respectively. These increased to 89.8 (4.91) mm, 30.7 (3.62) mm and 6.19 (0.36) cm by 14 years of age, respectively.

Close

Fig. 1

Inter-pupillary distance of male children representing different population stocks.

Export to PPT

Close

Fig. 2

Inter-pupillary distance of female children representing different population stocks.

Export to PPT

The corresponding figures for the girls measured 63.9 (4.65) mm for OCD, 21.5 (2.04) mm for ICD, 4.59 (0.26) cm for IPD at one month and 88.6 (4.14) mm, 29.6 (1.93) mm, 6.08 (0.25) cm at 14 yr, respectively. Boys in general, possessed larger IPD, ICD and OCD than girls. The magnitude of gender differences became significant at 11, 12, 13-15, 16-18 and 22-24 months, respectively for OCD and at 10, 11 and 19-21 months, 4.5 and five years, respectively for ICD (Table I). While, for IPD, significance (P≤0.05) in terms of gender differences was observed at 10, 11, 16-18 and 22-24 months, five and 10 yr of age, respectively.

Discussion

For precise detection of ocular hyper/hypotelorism, the use of authentic population-specific growth reference data becomes a paramount necessity as besides IPD, ICD and OCD are also known to exhibit substantial degree of racial and ethnic variation. This study presents cross-sectional reference data for IPD which has been derived from OCD and ICD measured among apparently normal healthy male and female children inhabiting northwestern parts of India over the last two decades.

IPD among the study participants demonstrated a regular increase throughout the study period. This can be attributed to the sustained increase recorded for both inner canthal and OCD which were used to calculate IPD. The mean increase in OCD, ICD and IPD in male subjects (24.0 mm, 9.1 mm, 1.51 cm) was marginally higher than for females (24.7 mm, 8.1 mm, 1.49 mm). The net per cent increase for IPD from one month to 14 yr was almost similar among male (32.3%) and female (32.5%) children. The per cent increase for OCD measured lesser in males (36.47%) as compared to females (38.6%). While, for ICD, it remained higher in male (42.1%) than the female children (37.7%). Around 48 per cent of the adult OCD, ICD and IPD were attained by two years of age and around 85 per cent by 10 yr.

The co-relation coefficient calculated among our study subjects confirmed a strong inter dependency of IPD on both OCD and ICD as the values were statistically significant throughout the age range. However, the magnitude of this correlation was found to be slightly more with OCD than ICD. A strong correlation of IPD with inner-outer ICD in Iranian individuals younger than 20 yr has also been reported18.

Relatively larger OCD, ICD and IPD noticed among the male children than females of same age depicted almost a similar pattern of gender differences reported earlier among Black children819, as also with those with Chinese20, Nigerian6 and American7 ancestry. However, Gupta et al21 reported no gender influence (children as well as adults) on OCD, ICD and IPD values in the normal Indian population.

IPD of the children in the present study measured larger than infants and children of American origin714, White males12 as well as those representing central22 and north India23. The magnitude of this inter-population differential between the present study and American7 children, however, remained significant only during infancy. As compared to their central22 Indian counterparts, this measure was significantly larger (P≤0.05) beyond three years of age and at 14 yr (P≤0.001) when compared with those from Amritsar, Punjab23. On the contrary, IPD in children measured significantly (P≤0.01) shorter in this study than Black children8, throughout. Since shape and cranial proportions affect the orbits, Pryor12 suggested that patients should preferably be compared with their own age, gender and racial norms for evaluating deviations from normal. These norms were reported for IPD among Whites, Asians and Mexican-Americans; however, their mean IPD values are lesser than that reported in the present study. This aforementioned size-related differential confirms the existence of a substantial degree of racial/ethnic variation in the growth of IPD among children belonging to different population stocks. Existence of ethnic variation for OCD, ICD and IPD in different parts of the world among Black8 individuals in contrast to those belonging to White, Asian, Mexican and American racial stocks has also been reported12. However, no racial differences were observed while comparing subjects of Chinese ancestry24 with those representing Caucasian25 population.

In view of the existence of substantial degree of inter-population variability, there is a need to establish age, gender, geographic, racial, ethnicity specific data sets of male and female individuals to arrive at precise and accurate clinical inferences. Therefore, it is suggested that the percentile grids presented from two months to 14 yr among male and female children of north-western India (Table II & III) may be used for comparative purposes to detect ocular hypotelorism and hypertelorism to corroborate diagnosis of suitable syndromic conditions. The patients having IPD less than 3^rd percentile should be treated as cases of hypotelorism while those exceeding 97^th percentile as cases of hypertelorism.