Note: This study was conducted by first author at Department of Ophthalmology, Rangaraya Medical College, Kakinada, India in the year 2002-2003.

Introduction

Conventional cycloplegic retinoscopy is widely used method to determine the refractive errors in patients of all age groups. Ophthalmologists and opticians in India are increasingly using Non-cycloplegic autorefractor measurements to determine refractive errors in patients of all age groups. Their reliability is questionable especially in young children.

The present study is an attempt to determine the reliability of cycloplegic automated refractor measurements as an effective substitute for conventional cycloplegic retinoscopy in young children.

Material and Methods

It is a prospective study involving 200 school children aged 8 to 15 years. They were evaluated for refractive errors over a period of one year. Visual acuity is tested using Snellen's acuity chart. Automated refraction was performed prior to the administration of cycloplegics. It is followed by cycloplegic conventional retinoscopy and cycloplegic autorefraction. 4% lignocaine topical drops were used to minimise the irritation with cyclopentolate eye drops. 1% cyclopentolate eye drops was used for the purpose. Cyclopentolate eye drops were administered at 0 minutes, after 5 minutes and after 20 minutes. The pupillary reflex and dilatation were inspected at 35 minutes from the first drop of administration. After adequate papillary dilatation (6 mm or more), cycloplegic conventional retinoscopy followed by cycloplegic autorefraction was performed.

Retinoscopy

Automated refraction

Procedure

The child is comfortably seated in front of the Autorefractor. Instrument is adjusted so that the pupil is concentric with the inner alignment. The operation lever is fine adjusted until a clear bright dot appears in the centre of the inner alignment.

Figure 1

Figure 1

When the bright dot is clearest in the centre alignment ring and eye is in proper focus, measurement starts automatically. After measurements are made, the standard value will be automatically calculated and displayed within brackets. The measurement in the other eye was done in the similar way. After both eyes are measured, the results were printed automatically. Autorefractor readings were taken before and after cycloplegia.

Fundus Examination

Post-Mydriatic Test

Results

We used Chi test to test the level of significance. Post-mydriatic test values were taken as gold standard and compared with that of cycloplegic autorefractor measurements, non-cycloplegic autorefractor measurements and conventional cycloplegic retinoscopy.

Figure 2

Table 1: Percentage of measurements in 400 eyes of spherical power by two methods that agree

In our study 70% of cycloplegic autorefraction spherical measurements agreed within ± 0.25 D of post-mydriatic test values and 97% agreed within ±0.50D

93% of Cycloplegic retinoscopy spherical power measurements agreed within ± 0.25 D of post-mydriatic test values and 98% agreed within ±0.50 D.

52% of non-cycloplegic autorefraction spherical power measurements agreed within ± 0.25 D of post-mydriatic test values and 70% within ±0.50 D.

There is a significant difference between cycloplegic autorefractor spherical power measurements and Cycloplegic retinoscopy values with in the range of ±0.25D (P<0.001), but not with in the range of ±0.50D (P= 0.8398).

There is a significant difference between non-cycloplegic autorefractor spherical power measurements and Cycloplegic retinoscopy values with in the range of ±0.25D (P<0.001) and also with in the range of ±0.50D (P<0.001).

Figure 3

Table 2: Percentage of measurements in 400 eyes of cylindrical power by two methods that agree

In our study 95% of cycloplegic autorefraction cylindrical power measurements agreed within ±0.25 D of post-mydriatic test values and 98% within ± 0.50 D.

96% of cycloplegic retinoscopy cylindrical power measurements agreed within ±0.25 D of post-mydriatic test values and 99% within ±0.50D

85% of non-cycloplegic autorefraction cylindrical power measurements agreed within ±0.25 D of post-mydriatic test values and 89% within ±0.50 D.

There is no significant difference between cycloplegic autorefractor cylindrical power measurements and Cycloplegic retinoscopy values with in the range of ±0.25D (P=0.838) and also with in the range of ±0.50D (P=0.801).

There is a significant difference between non-Cycloplegic autorefractor cylindrical power measurements and Cycloplegic retinoscopy values with in the range of ±0.25D (P=0.02) and also with in the range of ±0.50D (P=0.04).

Figure 4

Table 3: Percentage of measurements in 400 eyes of cylindrical axis by two methods that agree

In our study 95% of cycloplegic autorefraction cylindrical axis measurements agreed within ≤ 10° of post-mydriatic test values and within 20° in 98% of eyes.

73% of cycloplegic retinoscopy cylindrical axis measurements agreed within ≤ 10° of post-mydriatic test values and within 20° in 91% of eyes.

75% of non-cycloplegic autorefraction cylindrical axis measurements agreed within ≤ 10° of post-mydriatic test values and within 20° in 89% of eyes.

There is a significant difference between cycloplegic autorefractor cylindrical axis measurements and Cycloplegic retinoscopy values with in the range of ≤ 10° (P<0.001) and also with in the range of 11° to 20° (P =0.033).

There is a significant difference between non-cycloplegic auto refractor cylindrical axis measurements and post-mydriatic test values with in the range of ≤ 10° (P =0.639) and also with in the range of 11° to 20&176; (P =0.831).

Discussion

There is remarkable agreement between Cycloplegic autorefractor spherical power measurements and Cycloplegic retinoscopy with in the range of ±0.50D is comparable.

Non-cycloplegic autorefractor spherical power measurements and Cycloplegic retinoscopy with in the range of ±0.5D are not comparable.

There is remarkable agreement between Cycloplegic autorefractor cylindrical power measurements and Cycloplegic retinoscopy with in the range of ±0.25D is comparable.

Out of the all the methods used, cycloplegic autorefractor cylindrical axis measurements is the best. It is even better than cycloplegic retinoscopy with in the range of ≤ 20°.

Non-cycloplegic autorefractor cylindrical axis measurements also can be comparable with cycloplegic retinoscopy values.

There are a few published studies concluding that the most reliable measure of spherical and astigmatic refractive error was cycloplegic autorefraction followed by non-cycloplegic autorefraction and cycloplegic subjective refraction₁, ₂, ₃, ₅, ₆.

Conclusion

Cycloplegic autorefraction spherical and cylindrical measurements can be substituted for conventional cycloplegic retinoscopy in young children. It is even better than conventional cycloplegic retinoscopy for cylindrical axis measurements.

The least reliable measure was non-cycloplegic auto refraction. Despite the cost of equipment, autorefraction with cycloplegia can be comparable to conventional cycloplegic retinoscopy in accuracy in children, can be run by an ophthalmic technician and therefore eliminates the ophthalmologist's examination time required for retinoscopy. Autorefraction without cycloplegia cannot substitute for the conventional cycloplegic retinoscopy in young children.

Correspondence to

Dr. Srinivas B. Pedamallu, 7, Dryclough Close, Halifax, HX30LF, England Email: pbs6vas@yahoo.com