Impact of cataract surgery on depression and cognitive function: Systematic review and meta-analysis

Methodological quality of the review: Medium confidence

Author: Pellegrini M, Bernabei F, Schiavi C, Giannaccare G.

Region: United States of America (USA), China, Japan, Vietnam, United Kingdom (UK), Iran, France, Greece and Australia

Sector: Impact

Subsector: Quality of life

Equity focus: Not stated

Review type: other review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background:
Brief description of the interventions and motivation for the review. What is the problem? What is the intervention? How does it aim to have an impact on outcomes of interest?

Cataract is a major cause of visual impairment and blindness worldwide. Adequate vision is an important aspect for physical and cognitive function. There is a growing body of evidence that visual impairment in the elderly contributes to cognitive decline. Moreover, the loss of vision caused by senile cataract has shown a strong association with depressive symptoms. In elderly people, depression often coexists with cognitive impairment, worsening the outcome of many medical illnesses and increasing disability and mortality.

Objectives:
To evaluate the effect of cataract surgery on depression and cognitive status in the elderly.

Main findings:
In total, authors included 16 studies in the review. These were published between 1996 and 2018 and included 1,308 patients who underwent cataract surgery. Five of them were conducted in the US, two in China, two in Japan, two in Vietnam, one in the UK, one in Iran, one in France, one in Greece and one in Australia. Duration of follow-up ranged between one to 12 months. Overall, 14 studies presented data on depression, nine studies reported data on cognitive function and 10 studies had a control group. The total quality score of studies ranged between 11 and 22 of a maximum score of 16 for uncontrolled studies and 24 for controlled studies.

In the 14 studies reporting the change in depression before and after surgery, the overall SDM was 0.460 (95% CI: 0.223-0.697; P<.001), indicating a significant reduction in depression after surgery. Authors identified significant heterogeneity (Q=413.1, I²=96.9%) and after removing four studies with high risk of bias, the SDM was 0.248 (95% CI: 0.097-0.399; P=.001). No significant publication bias was detected by the authors, by both visual evaluation of funnel plots and Egger’s test (P=.688). The subgroup analysis revealed a significant difference in the SDM among studies that used different tools to evaluate depression (P<.001). Authors noted that the reduction in depression after surgery remained statistically significant within the subgroup of studies using CES-D (SDM=0.197; 95% CI: 0.031-0.364; P=.020) and within the subgroup of studies using GDS (SDM=0.275; 95% CI: −0.007 to 0.558; P=.056). Meta-regression found a significant association between the reduction in depression and the baseline visual acuity (B=0.656; 95% CI: 0.133-1.180; P=.014). Conversely, no significant association of depression improvement with mean age of patients (P=.059), length of follow up (P=.236) or final visual acuity (P=.298) was reported by the authors.

In the nine studies reporting change in cognitive function before and after surgery, the overall SDM was 0.254 (95% CI: 0.120-0.388; P<.001), indicating a significant improvement of cognitive function after surgery. Authors identified a significant heterogeneity among these studies (Q=42.1; I²=81.0%) and after sensitivity analysis the SDM was 0.238 (95% CI: 0.071-0.404; P=.005). Egger’s test was positive (P=.005), suggesting the possibility of publication bias. The subgroup analysis revealed a significant difference in the SDM among studies that used different tools to evaluate depression (P<.001).

In the six studies reporting the change in depression in both the surgery group and the control group, the between-group SDM was 0.161 (95% CI: 0.027-0.295; P=.019), indicating higher reduction in depression in the surgery group. No heterogeneity and no significant publication bias was reported by the authors.

In the six studies reporting the change in cognitive function in both the surgery group and the control group, the between-group SDM was 0.188 (95% CI: 0.002-0.374; P=.048), indicating higher cognitive improvement in the surgery group. Heterogeneity was reported as low and no significant publication bias was detected by the authors.

Methodology:
Authors searched PubMed, Scopus and Web of Science until September 2019 for relevant studies. In addition, authors manually screen bibliographies of relevant studies as part of the search strategy. Studies were included if they met the following criteria: (1) prospective design; (2) population: patients with age-related cataract; (3) intervention: cataract surgery; (4) outcome: report of a measure of depression or cognitive impairment before and after surgery; and (5) language: English. Conference proceedings and abstracts, letters, reviews, editorials, cross-sectional or case-control studies and full texts without raw data available for retrieval were excluded. Studies focused on specific subsets of patients, such as those with Alzheimer’s disease, were also excluded.

Screening of studies and data extraction were conducted by two reviewers independently. To assess the methodological quality, authors used the Methodological Index for Non-Randomized Studies (MINORS) score system.

Data analysis included a random-effects model; the effect size for the change in depression and cognitive function before and after surgery was estimated using the standardised difference in mean values (SDM).

The Q and I² tests were used to evaluate heterogeneity. Authors also performed sensitivity analysis and subgroup analysis (grouping studies that used the same tool). Publication bias was assessed by both visual evaluation model of funnel plots and Egger’s test.

Applicability/external validity:
Authors note that the presence of unmeasured confounders and the methodological differences among studies resulted in high heterogeneity, and this limits the generalisability of findings from this review.

Geographic focus:
Studies included in the review focused on different regions, however, authors did not discuss the generalisability of findings from studies conducted in low and middle income countries.

Summary of quality assessment:
Overall, medium confidence was attributed to the findings of this review as important limitations were identified. Authors did not conduct a thorough search of the literature to ensure that all relevant studies were included in the review, and it is not clear from the review if authors covered a comprehensive search period to ensure that relevant literature was unlikely to be omitted..