Role of ultrasonography in childhood eye diseases in a tertiary care setting: indications and scope

Abstract

We sought to evaluate the role and diagnostic potential of ocular B-scan ultrasonography in childhood eye disease in an observational cross-sectional study; 1091 patients with a total of 1445 eyes examined were studied. Cataract was the single most common indication for ultrasound followed by corneal pathology, ocular trauma, posterior segment pathology, primary congenital glaucoma, leukocoria, orbital pathology and other disease. Ultrasonography resulted in a change in diagnosis in 198 cases (18%). We conclude that B-scan ultrasonography plays an important adjunctive role in the management of childhood eye disease.

Keywords

Ophthalmic ultrasound B mode children paediatric ophthalmology diagnostic ultrasonography

Introduction

Ultrasonography (USG) is a well-established, simple and cost-effective modality for aiding in diagnoses of assorted ophthalmic pathology.^1,2 B-scan USG is particularly useful to rule out gross posterior segment abnormality when examination is not possible owing to media opacity but also to characterise posterior segment or orbital lesions. It may also be helpful in assessing anterior segment structures, optic nerve status, measurement of gross axial length or in cases where clinical examination is inconclusive.^3–5

B-scan USG is especially helpful in children as a detailed and meticulous examination may not always be possible; as it does not require any sedation or additional anaesthesia, it lends itself to use in children. Furthermore, it can be easily performed and is rapidly learnt. There is surprisingly little literature regarding its diagnostic potential in the young, especially in uncooperative or mentally challenged children.

Materials and methods

Our study was an observational, cross-sectional study in a tertiary care eye hospital in North India. All patients aged <18 years referred during a two-month period (16th November 2018 to 15th January 2019) were included. Their demographic details (age and sex), laterality of involvement, clinical diagnosis, indication for USG and respective findings, and diagnosis as noted by the USG (EZ Scan AB-5500, Sonomed Inc.) were documented. Data were recorded after taking an informed consent from the guardian, wherever possible. We adhered to the Helsinki guidelines.⁶ All USGs were performed by a single, well-trained ophthalmologist. If any patient underwent a repeat USG during the study period, details of the first visit were noted for the purpose of analysis.

Results

B-scan ultrasound was performed for a total of 1091 patients: their demographic details are shown in Table 1. Cataract was the single most common indication (n = 259, 23.7%), followed by assorted corneal pathology (n = 223, 20.4%) and ocular trauma (n = 183, 16.7%). Other findings were posterior segment pathology (n = 128, 11.7%), primary congenital glaucoma (n = 75, 6.9%), leukocoria (n = 92, 8.4%), orbital pathology (n = 43, 3.9%) and others (n = 88, 8.1%) (Figures 1–3). The latter included cases of phthisis bulbi, visual axis opacification, congestion, uveitis, posterior scleritis, neurocysticercosis, optic nerve head drusen and optic nerve avulsion.

Figure 1.

Clinical picture of a one-month-old boy with bilateral cryptophthalmos (a) and B-scan ultrasonography showing right and left grossly normally formed globes (b, c).

Table 1.

Demographic details of the study population.

Variable	n (%)
Gender
Boys	744 (68)
Girls	347 (32)
Age (years)
£1	252 (23)
2–5	276 (25)
6–10	202 (19)
11–15	249 (23)
≥16	112 (10)
Mean age (±SD) (years)	6.94 (±5.84)
Median age (range) (years)	6 (2 days–18 years)
Laterality (n)
Right eye (OD)	370
Left eye (OS)	367
Both eyes (OU)	354
Total eyes examined (n)	1445
Total patients examined (n)	1091

Values are presented as n (%) unless otherwise specified.

OD: right eye; OS: left eye; OU: both eyes.

USG was of particular benefit in 14 mentally handicapped and uncooperative children. Out of the 259 cases of cataract, approximately one-third had congenital or developmental cataract (n = 90, 35%) while another third had traumatic cataract (n = 89, 34%). The remaining number had various causes (Figure 4). A change in diagnosis based on ultrasound findings was seen in 50/259 (21.2%); in 43, a retinal detachment not evident on clinical examination was found. Unexpected USG findings included six persistent hyperplastic primary vitreous (PHPV), four retinoblastoma, four posterior staphyloma, and one phthisis bulbi. Furthermore, USG could measure and compare bilateral axial lengths as well as look for posterior capsule (PC) status, especially in traumatic cataracts.

Figure 2.

Clinical picture of a seven-year-old girl with ocular surface disorder with severe photophobia (a) with a normal posterior segment B-scan ultrasound (b).

Figure 3.

(a) A mobile cystic structure in the vitreous cavity (arrow) in a five-year-old boy with neurocysticercosis, (b) fundal coloboma with a closed funnel retinal detachment (arrow), (c) choroidal detachment (arrow) after glaucoma surgery in a six-month-old girl with primary congenital glaucoma, (d) retinal detachment with a subretinal localised mass lesion (arrow) with high spikes in a one-year-old girl suggestive of retinoblastoma, and (e) retained intra-ocular foreign body (arrow) with post-acoustic shadowing in a six-year-old boy with penetrating trauma to the globe.

Figure 4.

Clinical picture of an eight-year-old boy with unilateral complicated cataract (a) with a retinal detachment (yellow arrow) along with a posterior vitreous detachment (blue arrow) as seen on B-scan ultrasound (b).

Miscellaneous corneal pathology (n = 223, 20.4%) was found, approximately one-third being cases of infective keratitis (active or healed) (n = 67, 30%) or congenital causes (n = 63, 28%). The remainder had traumatic opacities (n = 30, 13%), iatrogenic injury (n = 22, 10%), ocular surface disorder (n = 18, 8%), chemical injury (n = 14, 6%) or other diagnoses (n = 11, 5%). Here, in 24 (10.9%) cases, a different diagnosis was made after USG than presumed beforehand. Again, USG could look at lens status and optic nerve head cupping.

The third leading indication was ocular trauma (n = 183, 16%), a majority of cases being open globe injuries (n = 157, 86%), where USG was performed after primary repair. Other cases were from blunt trauma (n = 26, 14%) with dislocated lens, hyphaema and vitreous haemorrhage, inter alia. The latter was seen in 100 cases, but USG was normal (anechoic vitreous cavity) in 83 cases. A change in diagnosis followed in 18 (9.8%), as a result of detection of retinal detachment (RD) or an intra-ocular foreign body (IOFB).

Among the 128 cases of posterior segment pathology, USG resulted in change of diagnosis in 29 (23.7%) with distribution as follows: endophthalmitis (n = 32, 25%); ROP: Retinopathy of Prematurity (n = 29, 18%); operated vitreoretinal surgery (n = 19, 15%); RD (n = 13, 10%); IFC: Iridofundal Coloboma (n = 11, 9%); vitreous haemorrhage (n = 9, 7%); and others (n = 13, 10%). For all diagnoses of leukocoria (n = 92), USG revealed findings suggestive of retinoblastoma in 53 (58%), retinal detachment in 12 (13%) and various other findings such as retinopathy of prematurity, persistent fetal vasculature, vitreous haemorrhage, coloboma or choroidal detachment in 6 (6%). However, the posterior segment was within normal limits with an anechoic vitreous cavity in 16 (17%) and there was a diagnostic dilemma in 5 (5%) cases.

Overall, a change in diagnosis ensued in 198 of the total 1091 patients as a result of USG findings.

In our study, there were 60 cases where the USG was not diagnostic and there was a dilemma about these findings. The most common scenario was confirming the presence of RD, present in 38 cases. Out of these, it was difficult to differentiate this from a posterior vitreous detachment in 16, from PHPV in seven and from a choroidal detachment in one. In other cases, there was a doubt regarding the presence of a shallow or peripheral retinal detachment or a closed funnel RD in small phthisical globes. Other dilemmas included the presence of a mass and/or calcification (n = 8), a fundal coloboma (n = 2), a posterior staphyloma (n = 2), a doubtful cyst or extra-ocular mass (n = 4), IOFB (n = 2) and/or other causes such as optic disc coloboma and optic nerve head drusen (4).

Discussion

A wide set of indications prompted USG, which was not only very useful but easily and successfully performed in all patients, barring 14 who were slightly uncooperative for the procedure. Nonetheless, USG was performed and it was notably better tolerated than slit lamp and fundus examination in difficult cases, particularly in those with photophobia.

USG was used both for prognosis in ruling out any posterior segment abnormality in cases of media haze due to cataract and corneal opacities, and for diagnosis, for example, to determine the cause of leukocoria or the nature of an orbital mass.

At our hospital, a single USG machine is used for all routine cases and a major proportion of emergency cases; this may have affected the overall distribution of indications. Furthermore, cases of traumatic globe injuries, cataract and corneal opacities due to trauma were relatively more common than would be expected, as were cases of retinoblastoma, most likely because ours is a tertiary care referral ophthalmic centre.

USG forms one of the most important investigations in retinoblastoma and other causes of leukocoria in children (Figure 5).⁷ USG also helps in localising suspicious lesions in the contralateral eye of patients with suspected unilateral retinoblastoma.

Figure 5.

Clinical picture of an 18-month-old boy diagnosed clinically as OD post-traumatic endophthalmitis (a); ultrasound of the posterior segment, however, reveals the presence of diffuse moderate to high amplitude spikes in the vitreous cavity, signifying the possibility of retinoblastoma (b) OD: right eye.

However, it is important to understand that USG has its limitations and there can be diagnostic dilemmas. It can be challenging to differentiate a thick posterior vitreous detachment from RD. Similarly, distinguishing a closed funnel RD from a PHPV stalk may not always be possible. Therefore, clinical assessment and appropriate correlation is of utmost importance in all cases.

Further, it may not always be possible to confirm the presence of a mass or calcification or IOFB; however, a well-performed USG along with clinical examination is almost always able to provide sufficient information so as to guide further management.

Of note, all USGs in our study were independently, easily and successfully performed by a single ophthalmologist instead of a radiologist. In addition, USG could even be performed in children as young as two days old.

Limitations of our study are that it represents data from a tertiary care eye hospital and that it was a cross-sectional study where additional ophthalmic imaging was performed but not individually followed up.

USG can be easily interpreted by the treating ophthalmologist and, although being operator-dependent, is still highly sensitive and specific, thereby acting as a significant universal diagnostic modality. We therefore recommend that all ophthalmologists should be trained to perform ophthalmic USG and all ophthalmic clinics should be equipped with the necessary equipment.