Introduction

In all, 95% of the sensory input to the brain needed for driving comes from vision.1 Alterations in visual acuity (VA) thus comprise an important factor in determining fitness for driving. Decrease in VA for the purposes of driving is considered significant if there exists the inability to read in good light (with the aid of corrective lenses if necessary), a registration mark fixed to a motor vehicle and containing letters and figures 79.4 mm high at a distance of 20.5 m, (Section 92 (2) of the 1988 Road Traffic Act).2 This corresponds to a binocular VA of approximately 6/10 on the Snellen chart.3 The number plate standard is absolute in law and is not open to interpretation, so also the statutory number plate test. The Driver and Vehicle Licensing Agency (DVLA) recognises that an adequate field of vision is necessary for driving and has defined the minimum visual field for safe driving of at least 120 on the horizontal meridian measured by the Goldmann perimeter on the III4e settings (or equivalent perimetry).2 Pupillary dilatation adversely affects vision as compared to the undilated state. Factors related to glare, decreased ability for accommodation (in some instances) and refractive aberrations all contribute to decreasing visual quality and quantity.4,5 Thus, the DVLA recommends that a patient must not drive after pupillary dilatation till the effect of the drops wears off. Patients driving unaccompanied and requiring pupil dilatation (ie both eye dilatation, or one eye dilatation if the other eye has less than adequate vision) may require to return another day either by public transport, or with another driver.

This study aims to ascertain the subjective variations in driving vision post-tropicamide 1% dilatating drop application.

Methods

A total of 28 patients aged 19–76 years were included in the study (mean 48, SD=16). The patients were randomly selected from those attending outpatient clinic. All had one drop proxymetacaine hydrochloride administered followed by one drop of tropicamide6 1%. Postdilatation measurements were made a standard 20 min after instillation of the dilating drop. Repeated instillation of drops was avoided to help standardise the process. The proxymetacaine drop was used to eliminate the effect of variability in tear washout because of tropicamide-induced reflex lacrimation. Patients who were not actively driving, who had binocular corrected vision worse than 6/12, and those who had pre-existing pupillary abnormalities were excluded.

The equipment used included, standard Snellen's and near vision charts and standard licence plates (yellow and white). The Sekonic FML 308-B2 light meter was used for the objective glare measurements. The light metre glare-recording surface was held close to the patient's eyes and facing the direction of the target that the patient was viewing. This allows for a measure of the glare that the patient's eyes are receiving. Measurements were read in electron volts (eV).

Testing steps:

  1. 1

    Best-corrected vision assessed by Snellen at 6 m (each eye and binocularly).

  2. 2

    Near vision assessed—Near test chart (each eye and binocularly).

  3. 3

    Licence plate reading at 20 m indoor assessed (binocularly).

  4. 4

    Ambient glare levels objectively measured with the light metre (indoor and outdoor).

  5. 5

    Licence plate reading at 20 m outdoor assessed (binocularly).

  6. 6

    Process repeated post dilatation.

Results are analysed using the Student's t-test (paired, two-tailed) and χ2-test.

Results

The pupil size changed significantly by the dilating process indicated above. The mean pupil size predilation was 3 mm in both eyes (SD=0.4), and the mean pupil size postdilation was 5 mm in the right eye (SD=1.3) and 6 mm in the left eye (SD=1.2). The P-value of the change in pupil size is <0.001 (by paired two-tailed t-test) for both eyes, and this is statistically significant.

The Snellen distance VA changed as indicated in Table 1, and the near VA changed as indicated in Table 2, from the undilated to the dilated state. There is a statistically significant change in the distance6,7,8 and near vision pre- and postdilation.

Table 1 Comparison of undilated to dilated Snellen VA
Full size table
Table 2 Comparison of undilated to dilated near VA
Full size table

The distance at which the number plate could be read correctly predilation was a mean of 20 m (SD=0), which is the standard required for driving. Thus, it is established that all patients in this study had vision adequate for driving predilation. Postdilation, the distance at which the number plates could be read dropped to a mean of 19 m (SD=2), both indoors and outdoors (Table 3). Comparison between licence plate recognition distance dilated vs undilated showed P=0.07 (indoor), P=0.11 (outdoor), both not significant—that is, there is no significant change in the distance at which the number plate could be read. The DVLA however is strict about its 20 m reading of licence plate criteria (number plate test). The number of patients who could not read the licence plate at 20 m indoors was six out of 28 (22%) and the number of patients who could not read the plate outdoors was four out of 28 (14%). Two patients were able to see the plate outdoors but not indoors. The P-value (χ2-test) of the decrease in the number of patients able to read the licence plate at 20 m is <0.05 outdoors and <0.01 indoors (both statistically significant). Thus, a significant number of patients failed the number plate test (reading at 20 m), postdilatation.

Table 3 Analysis of ability to read licence plate at 20 m or below, before and after dilatation
Full size table

Glare sensitivity was subjectively graded and scored as in Table 4. Prior to predilation no patient reported glare. Postdilation, the mean glare score was 0.8, corresponding to a less than mild symptomatology. Light metre readings were a constant 6 eV indoors and a mean of 11eV (SD 1 eV) outdoors—for both dilated and undilated states. This was carried out to ensure that ambient glare levels were constant during the testing process and hence was not a factor to affect subjective assessment. Five out of 28 patients (1.8%) had bad but tolerable glare in the postdilated state (P<0.025 χ2-test is statistically significant).Table 5

Table 4 Glare sensitivity dilated vs undilated
Full size table
Table 5 Data spreadsheet for all patients
Full size table

The patients were asked at the end of the test to answer yes or no to the question—‘can you drive comfortably and safely now under the influence of the drops?’ Four of 28 (14%) reported ‘no’ (ie they would not be able to drive comfortably and safely). This was statistically significant (P<0.05 by χ2-test).

Discussion

This study addresses the debatable question of whether to disallow driving after pupillary dilation (of both eyes or the one better eye). Proponents of allowing driving argue that there is no significant change in the accident rate following dilation in our present experience.9 This, however, cannot be conclusively determined because of the varying practices adopted across different units. In the UK the present common practice is that patients are advised not to drive postdilation. The study confirms the validity of this practice by demonstrating significant decrease in Snellen's vision, near vision, and licence plate reading ability at 20 m when the pupil size is significantly changed. More relevant however, is that a significant 14% of patients themselves feel unsafe and uncomfortable to drive following dilation. In this study, only one drop of tropicamide 1% was used to standardise the process, whereas in the clinical scenario stronger and increased number of drops may be used for achieving maximal dilation, and this may further decrease the visual capacity of the patient with regard to driving.

Despite this study and others like it, of importance is the fact that these analyses are based on simulating driving conditions. The validity of such studies can be increased greatly by using actual driving situations in future studies. Moreover, the pre-existing refractive state of the patient may have had a bearing on the results. Hypermetropes who often accommodate for distance may have been affected by cycloplegia more than others. Additionally, the effects of dilation in various clinical scenarios need to be looked into, for example, effect of dilation in patients with cataracts of various types, etc.

Recommendations against driving postpupillary dilatation have been made on reasonable expectations of decrease in visual ability to qualify for safe driving. Evidence, as provided in this study, confirms the decrease in visual function below the present acceptable standard. Visual standards however are under constant review because of the inevitable difficulties in standardising the visual criteria with relevance to safety standards. Difficulties arise in accommodating the effects of clinical variables that complicate analyses (eg patients with certain cataracts, postdilatation may in fact have better vision than predilatation). Large-scale controlled studies are required to include the entire spectrum of patients undergoing pupil dilation, and this needs to be correlated to specific measurable reduction in safety standards. Until then the recommendations for allowing or disallowing driving would continue to be based on reasonable expectation, corroborated by evidence as available in this study and others like it.