Results

Of 703 questionnaires mailed 489 replies were received. Sixteen of the respondents excluded themselves because they did not consider the questionnaire relevant to their current practice. The commonest reason was that they did not routinely operate on the age group specified. A total of 473 of the respondents considered themselves eligible for inclusion in the study, of which 471 completed the section on lens morphology. Question 1, which concerned classes of lens opacity, was completed by 467 surgeons. The results from question 1 are presented in Table 2.

A broad range of lens opacities was assessed, with differences between surgeons for some opacities with high prevalences in the population, particularly cortical opacities. Because of the possibility that individual lens features were being disregarded on the questionnaire because other associated opacities had already been endorsed the responses were grouped into categories of “subcapsular”, “nuclear sclerosis”, and “cortical” (Table 3). Grouping the responses made little difference to the results for cortical opacities. Twenty nine per cent of surgeons did not endorse any of the questionnaire options for cortical spokes or other related opacities.

A total of 347 (74%) of 467 surgeons indicated that they assessed one or more of the following lens opacities: anterior subcapsular cataract, vacuoles, water clefts, coronary flakes, focal dots, retrodots, fibre folds, which have been disregarded in many research studies. Other lens features were mentioned infrequently. Other features named were the zones of discontinuity (one surgeon), polychromatic lustre (one), lens shape—for example, lenticonus (two) and central nuclear morphology (one). Some surgeons also included features related to the ease of surgery—for example, lens stability.

The vast majority of surgeons (433 of 469) considered proximity to the visual axis to be an important feature. Opinion was divided as to the importance of the anteroposterior location of opacities (excluding posterior subcapsular cataract); 252 of 466 surgeons considered this feature. Ten of 466 routinely used a recognised grading system in their clinics. A further two used a recognised system occasionally and a further surgeon used part of the LOCS system.

Discussion

The highest endorsement frequency was for posterior subcapsular cataract (97%). There is little doubt that posterior subcapsular cataract impairs vision. The class of opacity which is arguably the least likely to impair vision (fibre folds) was assessed by the smallest number of surgeons (12%). The appearances of fibre folds may be caused by reflection6 and may not necessarily cause degradation of the retinal image. Between the above extremes there was a range of endorsement frequencies for the various lens features.

The questionnaire response to cortical spokes was interesting. Some investigators have reported associations between cortical opacity and visual impairment78 but this finding has not been consistent across all studies.910 A relatively low prevalence of “pure” cortical cataract was found in cases listed for surgery in Parma, Italy.1 Elliottet al found that some of their patients with extensive cortical opacity retained excellent visual acuity and attributed this to the ability to look through a “hole” in the cataract.11 Rouhiainen et alfound that a decrease in Snellen acuity was associated with increasing nuclear opacification but not with early cortical opacity.10 Pesudovs and Coster found that a score of visual disability in cataract patients was related to nuclear opacity and to posterior subcapsular cataract but not to cortical cataract.12

It is also noteworthy that 74% of surgeons assessed lens opacities which have been disregarded in many research studies. It is commonly assumed that there are three major types of cataract—nuclear, cortical, and posterior subcapsular.13 This simplified scheme may be satisfactory for studies of severe cataract but it clearly does not take into account the subtleties of lens examination in clinical ophthalmic practice.

The results raise the question of which of the less frequently investigated classes of opacity cause visual impairment. If some, but not all, classes of opacity interfere with vision, careful examination of the lens at the slit lamp becomes crucial. Vision tests and visual symptoms may not be helpful in the presence of other coexisting ocular pathology. Discovery of lens opacities which are not visually significant would lead to a search for other causes of visual impairment. A possible means of avoiding the problem of classifying opacities is to assess the clarity of the clinician's view of the retina through the cataractous lens. Target resolution ophthalmoscopes have been devised for this purpose,14 but these instruments provide an incomplete description of the severity of opacity. Nor do they discriminate between the optical effects of cataract and those of other structures within the eye.15Likewise, it is uncertain whether assessment of the general clarity of the red reflex (for example, by direct ophthalmoscopy or retinoscopy) will reliably discriminate between visually significant early lens opacities and visually insignificant opacities. Some abnormalities are poorly visible on retroillumination.

The present survey has several implications for cataract research. Researchers frequently recruit cases of cataract from surgical waiting lists for a variety of purposes including psychophysical and epidemiological studies. Occasionally lenses are obtained after cataract extraction or from postmortem eyes for the purposes of biochemical or photometric analyses. The present results demonstrate the potential pitfalls of this approach if the required classes of opacity are not subsequently verified using standardised methods of lens examination. The presence of aphakia or pseudophakia may be included when estimating the prevalence of cataract in epidemiological studies.1617 This may lead to the anomalous situation where certain classes of opacity are discounted in phakic individuals but may be counted as cataract in those having received surgery. The event of cataract surgery has also be assumed to indicate progression of the opacity type measured in a longitudinal study.18This assumption is not valid if coexisting but unmeasured classes of opacity have developed or progressed.

Although some classes of opacity are associated with one another19 it is presently uncertain whether the associations are strong enough to eliminate the potential measurement errors caused by ignoring particular opacities. Recently the Melton Eye Study has provided information about the prevalence of 11 clinicopathological classes of lens opacity.20 Such information, when combined with future knowledge of the vision impairing capabilities of a broader range of opacities, may help both in determining which lens features should be assessed and in evaluating the effects of ignoring specific classes of opacity.