Introduction

Cataract surgery is among the most frequently performed surgical procedures.1, 2, 3, 4, 5 Several studies have reported improved visual function and quality of life (QoL) in up to 90% of patients who underwent cataract surgery .6, 7, 8, 9 In general, persons less likely to experience improvement are older and have ocular comorbidities.2, 10, 11, 12

Both age-related maculopathy (ARM) and cataract are strongly age related13, 14, 15, 16 and often present concurrently in older persons.16 Previously, patients with ARM or other ocular comorbidities were considered unlikely to benefit from cataract surgery.2, 10, 11 Recent studies, however, indicate that ARM patients can achieve significant gains in visual function and QoL within months to a year after surgery.17, 18, 19, 20 Few studies, however, have assessed long-term (at least 1 year) surgery outcomes in patients with pre-existing ARM.20 Some population-based reports have suggested an increased long-term risk of developing late ARM in eyes after cataract surgery.21, 22, 23

This study aimed to assess mid- to long-term visual and health-related quality of life (HRQoL) outcomes in a hospital-based sample of patients who had cataract surgery at least 1 year earlier, and to compare outcomes between patients with and without preoperative ARM, plus factors associated with poor outcomes.

Methods

Westmead Hospital is the major public hospital in western Sydney (Australia) and provides public eye services to local residents. In this study, we invited consecutive patients aged 60+ years who had undergone cataract surgery at Westmead Hospital in the last 2 years to return for follow-up examinations. Study procedures were approved by the hospital Human Ethics Committee and informed consent was obtained.

Over a period of 2 years till June 2003, 941 routine cataract procedures were performed, including 718 procedures on 671 patients aged 60+ years with available medical records; as 49 died before the follow-up examinations, 454/622 survivors (73%) were examined.

Preoperative information was obtained retrospectively, by reviewing medical records, including demography, systemic and ocular comorbidities, smoking, preoperative visual acuity (VA) and postoperative surgical complications. To define persons with and without ARM before surgery, the 4-week postoperative clinical notes were used to confirm the preoperative assessment in most patients.24 We considered that the improved postoperative fundal view would provide a more reliable diagnosis of ARM lesions. The early ARM lesions recorded were soft drusen or macular retinal pigment epithelial (RPE) abnormalities (hyper- or hypo-pigmentation); the late lesions recorded were geographic atrophy and neovascular ARM. Using the 4-week postoperative clinical notes did not change any late ARM assessments but did result in a small number (only six cases, 6.1%) of changes to early ARM assessments.24

Participants were re-examined during March–December 2004. VA was measured under standard illumination using Snellen charts with distance correction, if worn, then with pinhole. We defined the following visual impairment levels: no impairment if VA≥6/12; mild if VA<6/12–≥6/24; moderate if VA<6/24–≥6/60; severe if VA<6/60. Structured interviews using VF-14 and SF-12 questionnaires were conducted.

Vision-related QoL was measured using the Visual Function Index (VF-14),9, 25, 26 with scores ranging from 0 (complete loss of visual ability) to 100 (no loss of visual ability).9, 26 The SF-12 (SF-36 short version) was used to assess HRQoL. SF-12 scores were highly correlated with SF-36 summary measures,27 and summarised into mental and physical component scores (MCS, PCS). In general, each component's population mean score is around 50; scores below 50 indicate disability (mild 40–49; moderate 30–39; severe<30). Each question's response is tagged with weights, differently for MCS, PCS. These weights are added to a constant to give the final score of 100.27

Characteristics assessed in relation to surgical outcomes included age, sex, systemic (eg ischaemic heart disease, stroke, diabetes) and ocular comorbidities, other than ARM (eg diabetic retinopathy, glaucoma, cataract type), potential sight-threatening surgical and postoperative complications (eg posterior capsular tear, cystoid macular oedema). The following poor surgical outcomes were defined: lowest quintile VF-14 and SF-12 scores; VA<6/12 in the operated eye; no VA improvement (equal to or worse than preoperative VA).

Statistical Analysis System (V8.2 SAS Institute, Cary, NC) was used for statistical analyses, including χ2 statistics, general linear, and logistic regression models. We converted Snellen VA to logMAR VA for the purpose of statistical analysis as follows: Snellen VA 6/6=logMAR VA 0; 6/7.5=0.1; 6/9=0.2; 6/12=0.3; 6/15=0.4; 6/18=0.5; 6/24=0.6; 6/30=0.7; 6/36=0.8; 6/48=0.9; and 6/60=1.0. General linear models were used to compare mean VA, VF-14, SF-12 scores in patients with/ without preoperative ARM, after adjusting for age, sex, systemic, and ocular comorbidities. Logistic regression models assessed factors associated with poor surgical outcomes, while adjusting for age, sex, preoperative systemic, and ocular comorbidities. Proportions, adjusted means, P-values, odds ratios (OR), and 95% confidence intervals (CI) are presented.

Results

Mean postoperative follow-up period of the 454 patients was 2.8 years (range 0.9–3.4 years). Preoperative characteristics of surviving patients examined at follow-up (n=454) were compared with those of persons not examined (n=168) or deceased (n=49) before follow-up (Table 1). Patients who had moved away or declined further participation were not significantly different from study participants by age or major illness history. Those who had died (n=49) were older or more likely to have had ischaemic heart disease. Mean age of participants (74% Caucasian) was 76 (SD±7.0) years. Mean follow-up VA was 0.21 or 6/9 and the mean VA gain after surgery was 0.28 (approximately three lines on the Snellen chart).

Table 1 Comparison of baseline and follow-up characteristics of participants and non-participants
Full size table

Figure 1 shows proportions with visual impairment before and after surgery. Visual impairment improved by at least one category in 201/454 patients (44.3%), as highlighted in the diagonal boxes, was unchanged in 234 (51.5%), and deteriorated in 19 eyes (4.2%). Overall, postoperative VA was 6/12 or better in 87.0% of eyes. Relatively similar proportions were observed in patients with (80.2%) or without preoperative ARM lesions (88.8%).

Figure 1
figure 1

Comparison of pre- and postoperative visual impairment (VI) levels in patients 1–3.5 years after cataract surgery (n (%)). Definitions: No VI (visual acuity (VA) ≥6/12); mild VI (VA<6/12-≥6/24); moderate VI (VA<6/24-≥6/60); and severe VI (VA<6/60).

Full size image

Adjusted means for VA, VF-14, SF-12 scores and VA gain after surgery in patients with/ without early/ late ARM lesions, are shown in Table 2. After adjusting for age, sex, systemic, and other ocular comorbidities, patients with late ARM before surgery had significantly lower postoperative mean VF-14 scores and mean VA, and experienced negative VA gain, than patients without late ARM. Patients with early ARM before surgery had nonsignificantly lower mean VF-14 and SF-12 scores, lower mean VA and less VA gain postoperatively than patients without ARM.

Table 2 The adjusteda mean VF-14 and SF-12 scores, visual acuity (VA) and VA gain follow-up by presence or absence of early or late ARM preoperatively
Full size table

Table 3 show the proportions with different adverse surgical outcomes by potential predictors. Patients older or with ocular comorbidities were more likely to have low follow-up VF-14 scores, while systemic comorbidities led to significantly lower SF-12 mental, physical component scores. Old age and ocular comorbidities or late ARM were significantly associated with postoperative visual impairment.

Table 3 Proportions with poor surgical outcome (lowest quintile of the VF-12 and SF-12 scores, follow-up visual impairment, and no visual improvement at follow-up) by preoperative and operative factors
Full size table

Table 4 shows factors associated poor surgical outcomes after adjusting for age, sex, preoperative VA, ocular and systemic comorbidities, and ARM. Patients with non-ARM ocular comorbidities were around three times more likely to have low VF-14 (OR 2.77, CI 1.65–4.64), visual impairment (OR 3.59, CI 1.98–6.52) or no visual improvement (OR 2.72, CI 1.53–4.84) postoperatively. Systemic comorbidities were predicted by both low SF-12 MCS (OR 2.25, CI 1.37–3.68) and PCS (OR 2.59, CI 1.56–4.28). Preoperative early ARM was not associated with any measured adverse surgical outcomes.

Table 4 Adjusted odds ratios (OR) and 95% confidence intervals (CI) for the preoperative predictors of poor surgical outcome, indicated by the lowest quintile VF-14 and SF-14 scores, follow-up visual impairment, or no visual improvement
Full size table

Discussion

In this hospital-based sample of 454 cataract surgery patients aged 60+ years, follow-up examinations were conducted for an average 2.8 years (range 1.0–3.5 years) after surgery. We found that 44.3% of patients achieved improvement in visual impairment, with 87% achieving VA≥6/12. Relatively similar proportions of patients with (80.2%) and without (88.8%) preoperative ARM improved postoperatively. Preoperative late ARM led to significantly poorer VA and vision-related QoL, whereas preoperative early ARM gave similar postoperative mean VA and QoL as those without ARM. Patients of older age, or with ocular comorbidities before surgery were more likely to have poorer postoperative vision-related QoL (VF-14) and VA outcomes, while those with systemic comorbidities were more likely to have poorer HRQoL (SF-12).

The proportions of our patients with and without ARM achieving VA≥6/12, 1–3 years after surgery, were similar to previously reported short-term cataract surgery visual outcomes.2, 12, 28, 29, 30 Using this same cohort, we previously reported that 82% of persons with ARM achieved VA≥6/12, 4 weeks after surgery.24 Several studies have reported short-term (up to 1 year) postoperative best-corrected VA.2, 12, 28, 29 In general, VA≥6/12 was achieved overall in 90% of eyes, and in 95% of eyes without pre-existing ocular disease. In the Auckland Cataract Study cohort of 97 patients aged 76.3 (±9.9) years, randomly selected from 480 participants, 95% of eyes maintained a best-corrected VA≥6/12, 2 years after surgery.31 Our sample had relatively high prevalence of ocular comorbidities and diabetes,30 a longer follow-up period, and used pinhole VA. All these factors could have influenced postoperative VA levels and contributed to the slightly lower proportion with long-term VA≥6/12 in our sample.

Several preoperative characteristics associated with poor surgical outcomes identified in this study are similar to findings from earlier studies.2, 10, 11, 12 Previous studies have also reported poor mid- to long-term outcomes in patients with surgical/ postoperative complications.10 Our failure to find such associations suggests that surgical complications may predominantly influence only short-term surgical outcomes in this sample.

We found that patients with late ARM before surgery had poor vision-related QoL and a higher proportion had postoperative visual impairment. Presence of preoperative early ARM, however, had a marginal effect on postoperative VA, vision-related and HRQoL over this period, but none of these outcomes were statistically significantly different from patients without early ARM. Our findings are in agreement with previous reports of short-term surgical outcomes.17, 18, 19, 20 Shuttleworth et al.18 reported that two-thirds of ARM patients felt that cataract surgery was worthwhile. Armbrecht et al.17 reported that patients with mild or moderate ARM had significant improvements in HRQoL and visual function 3–5 months after surgery, particularly with moderately severe cataract.

Overall, our study achieved reasonable participation and follow-up rates. Weaknesses include its hospital-based sample, use of medical records to document preoperative ARM signs and the use of pinhole VA as a surrogate measure of best-corrected VA.

In summary, we report that 87% of public hospital-derived patients achieved VA at least 6/12, 1–3 years after cataract surgery. Increasing age and preoperative ocular comorbidities including late ARM, were factors associated with low VF-14 scores and failure of VA improvement. Systemic comorbidities were also related to poorer VA and low SF-12 scores. Preoperative early ARM was not significantly associated with low postoperative VA or low VF-14 scores.

The benefit from cataract surgery to patients with early ARM after more than 3 years, however, was not examined in our study. This is particularly relevant as a number of reports have suggested that increased progression to late ARM may be a long-term phenomenon.21