Long‐term results of cataract surgery in type 2 diabetics, a 20‐year prospective longitudinal study

To investigate the 20‐year outcome, in best‐corrected visual acuity (BCVA), and subjective visual function (VF‐questionnaire, VF‐14) after cataract surgery in type 2 diabetics, compared with non‐diabetics.


| I N T RODUC T ION
Diabetes mellitus (DM) is among the most morbid chronic diseases.The prevalence and burden are sharply increasing worldwide (GBD-019, 2020).The number of diagnosed patients rose from ~333 to 537 million during the 10 years 2005-2021, and the number is expected to rise to 783 million by 2045 (IDF Diabetes Atlas, 2021).
It is well-documented that there is an association between cataract and especially late-onset type 2 DM.Patients with diabetes have an increased risk of especially cortical and posterior subcapsular cataract, and consequently a need for cataract surgery (Chang et al., 2011).Cataract also occurs at an earlier age in diabetics (Klein et al., 1995).Overall, up to 20% of cataract extractions are estimated to be performed on diabetic patients (Rossi et al., 2020).In addition to improving visual acuity (VA), it is also important to perform cataract surgery in diabetics with significant cataract, to facilitate adequate screening and treatment of diabetic retinopathy (DR).
In contrast to older surgical procedures such as extracapsular cataract surgery, modern small-incision phacoemulsification (PE) has been shown to provide good and safer outcomes regarding the postoperative progression of DR and/or macular oedema (Romero-Aroca et al., 2006;Squirrell et al., 2002).
In most diabetics, it is well documented that VA improves after cataract surgery (Danni et al., 2019;Han et al., 2020;Liu et al., 2019).Little is known about the long-term visual outcome, as the follow-up is less than 10 years in most published prospective studies (Lundqvist & Mönestam, 2012).Even though the mortality rate in diabetics is still higher than in the general population, life span has increased also in diabetics (Laatikainen et al., 2016).Therefore, long-time data on visual function after surgery is important knowledge.
The purpose of this study is to prospectively assess the long-term (20-year), longitudinal outcome in bestcorrected VA (BCVA) and subjective visual function following PE surgery in type 2 diabetics, compared with non-diabetics.Visual function was assessed by a questionnaire on the self-estimated ability to perform visiondependent activities.

| Participants
Details of the study design and methods of this prospective, population-based study have been published elsewhere (Mönestam, 2016;Mönestam et al., 2005).The study was based on a geographically defined cohort of cataract surgery patients, followed from before surgery to 20 years after surgery.Before the study started, the institutional review board/ethics committee of Umeå University approved the protocol, and informed consent was obtained from all participants at the beginning of each examination.The study was performed in accordance with the tenets of the Declaration of Helsinki.
Briefly, 810 patients, aged 30 to 96 years, had surgery for senile and presenile cataract from 1 June 1997 to 31 May 1998.Only the first surgical eye was included in patients who had surgery on both their eyes (n = 27) during the study period.
All patients underwent a standardized, clinical, dilated eye examination including BCVA before and a few months after surgery.A visual function questionnaire (VF-14) on their self-perceived difficulties in performing visually demanding tasks was answered (Steinberg et al., 1994).
The same ophthalmologist (EM) examined all patients.Diabetic retinopathy was graded by clinical assessment based on the International Clinical Diabetic Retinopathy Severity Scales (Wilkinson et al., 2003).The 4-stage grading of DR severity used in the present paper was; no DR, mild non-proliferative DR (NPDR), moderate to severe NPDR and Proliferative DR (PDR) past or present.The patients were classified according to their level of DR in the eye to be operated on at the examination before surgery.Patients with a history of proliferative retinopathy before surgery were also classified into the proliferative DR group even though the PDR was quiet at cataract surgery, Figure 1.The presence and grading of any diabetic macular oedema was not assessed.
Surviving patients willing to participate, had the examination and questionnaire repeated at 5, 10, 15 and 20 years after surgery.At each follow-up, a few patients unable to travel to the clinic participated in the questionnaire only.
It was also checked if all patients who were classified as non-diabetics before surgery had a new diagnosis of DM at any time during the follow-up period of 20 years.Diabetes was defined as having a continuous treatment with insulin and/or oral glucose-lowering medications.

| The questionnaire
The VF-14 questionnaire is a well-known and widely used patent-reported outcome instrument in ophthalmology.It is highly responsive and useful for assessing longitudinal changes in visual function after cataract surgery (Bilbao et al., 2009).Activity limitations in 14 vision-specific activities performed in everyday life give an overall estimate of visual function.The VF-14 scores range from 0 to 100 and are calculated as a summary score as originally described (Steinberg et al., 1994).A score of 0 denotes a very low visual function with no ability to perform activities such as reading, driving, etc.A score of 100 represents unrestricted/unaffected subjective ability to perform the 14 vision-dependent activities.The same questionnaire was answered a few weeks before surgery, 2-4 months postoperatively, and at all points of follow-up at 5,10,15 and 20 years, to enable direct comparison.
Summary scores of the original VF-14 questionnaire were used at each examination, to make it possible to longitudinally compare scores from 1997 to 1998 and onward.Rasch-analysis-based revision of the VF-14 had not been developed at the time when the study started.

| Ten years after surgery
Ten years after surgery, 396 of 807 patients remained alive (49%).Thirty-five of 396 (9%) had preoperatively known diabetes.Seventeen new patients of 396 (4%) had been diagnosed with diabetes during the 10 years that had passed after surgery, and 344 of 396 (87%) had no DM, Figure 1.

| Twenty years after surgery
133 of 807 patients (16%) of the original cohort of 1997-1998, were still alive 20 years after surgery.Six of these 133 (5%) had DM preoperatively, 20 of 133 (15%) new patients had a diagnosis of diabetes during the 10-20 years that had passed after surgery, and 107 of 133 (80%) had no DM.
The 15 patients who participated in the questionnaire only, and did not have an eye examination, were significantly older and more often females than those who had an eye examination (n = 97), Tables 1 and 2. The mean age of the patients with the questionnaire only was 88.0 years (SD = 10.1) vs. 79.6 years (SD = 10.4) (p = 0.04).There was no significant difference in gender (p = 0.09).

| Best cases
In an effort to refine the long-time impact of diabetes on visual function after cataract surgery, 2 subsets of patients were selected.One group was patients who never pre-or postoperatively had a diagnosis of diabetes at any time, and no other ocular disease, with a follow-up of 10 years or more (n = 306; control group).The second group had a diagnosis of type 2 diabetes with or without DR at surgery but no other ocular disease, and had a follow-up of 10 years or more (n = 38; diabetics).

| Surgical procedure
In most patients, the surgical technique consisted of a standard sutureless clear corneal PE with 3.2 mm temporal incision, and a posterior chamber foldable intraocular lens (IOL).The majority of patients had an acrylic hydrophobic IOL (Alcon Acrysof® MA60BM) implanted (95%).

| Statistical methods
VA measured on ETDRS-charts was scored as the total number of letters read correctly, and transformed into the logMAR (logarithm of the minimum angle of resolution) scale.Patients who failed to read any letters were tested using counting fingers (CF), hand movements (HM) and light perception (P).For VA less than CF 0.5 m the following arbitrary logMAR values were used; CF in front of the eye = logMAR 2.2, HM = logMAR 2.5, and no light perception = logMAR 3, in a similar manner as previously used (Mönestam, 2016).One letter on the ETDRS chart corresponds to 0.02 logMAR.Postoperative change in BCVA was calculated by subtracting the 1-2 months postoperative BCVA logMAR score, and the BCVA logMAR score at each follow-up after surgery, a positive value indicating improved VA compared with 1-2 months after surgery.The same method was used for calculating change in VF-14 total score and a positive value indicated improved subjective visual function.
All data were tested for normality.Data which did not follow a normal distribution were examined using non-parametric analyses.For continuous variables, values were displayed as mean ± standard deviation (SD) for normally distributed data, and as median with corresponding min-max values for skewed distributions.For Cataract surgery in 810 patients, 3 patients with type 1-diabetes were excluded, 807 patients remained.
10 years after surgery, 396 patients were still alive (49%), 35 of these with type-2 diabetes preoperatively (9% categorical variables, values were presented as counts and proportions in percentages.Independent samples median test with Bonferroni correction for multiple tests and Kruskal-Wallis tests were used to analyse VA data.Independent sample ttests were used to compare age differences and differences in VF-14 scores.To analyse tables with qualitative data, Yate's corrected chi-square tests or Fisher's tests were used.The chi-square for trend was used to compare the survival of non-diabetics and diabetics at surgery.Repeated measures tests were used to compare longitudinal changes in BCVA (logMAR) and VF-14 scores between the follow-up examinations.
All tests were two-sided and p-values of less than 0.05 were considered statistically significant.SPSS (Statistical Package for the Social Sciences for MS Windows, software 27.0, SPSS Inc.) was used for all data analyses.

| R E SU LT S
Of the 810 participants enrolled in the study, 112 (14%) had a diagnosis of diabetes at the time of surgery in 1997-1998.Three patients of 112 (3%) had type 1 DM recorded and were excluded, and 109 of 112 diabetics (97%) had type 2 DM, Figure 1.
At the 15-year follow-up, 35 patients of 359 (9.7%) remaining in the study had type 2 diabetes at surgery.At the 20-year follow-up, 133 patients (133/807;16%) were still alive.Of these, 6 patients had a diagnosis of type 2 diabetes preoperatively, and 20 patients had a new diagnosis of DM after cataract surgery, but before the final examination at 20 years.The remaining 107 survivors had no diagnosis of DM, Figure 1, Table 1.Thus, 6 of 113 patients (5.3%) diabetics and non-diabetics, had known type 2 diabetes at surgery.The patients with known type 2 diabetes at surgery had significantly lower survival compared with the non-diabetics at each follow-up (p = 0.003, chi-square for trend).

| Changes in visual acuity of the operated eye, all patients (n = 807 at surgery)
There was no significant difference in preoperative median BCVA of the operated eye, between patients with type 2 diabetes and the non-diabetics at surgery (0.7 vs. 1.0; p = 0.83, Kruskal-Wallis test).
Figure 2 shows the change over time in the median BCVA of the operated eye from preoperatively over 20 years after surgery.From 10 years or more after surgery, the longitudinal BCVA of the diabetics at surgery was slightly but not significantly better than in the non-diabetics ( p = 0.45, p = 0.44 and p = 0.83, at 10, 15 and 20 years, respectively).Further analysis with repeated measures test controlling for age did not change the results.There was no significant difference in age at 10, 15 and 20 years after surgery between the remaining patients, diabetics at surgery and the nondiabetics, p = 0.89, p = 0.93 and p = 0.53, respectively, independent samples t-test, Table 1, 15-year data not shown.
At 10 and 20 years after surgery, the patients who had a new diagnosis of type 2 diabetes 1-10 years after surgery were significantly older than those with a new  The numbers in brackets refer to the number of patients contributing with VF-14 total score data in relation to the number of survivors in each patient group.

MÖNESTAM
diagnosis 11-20 years surgery, p = 0.02 and p = 0.026, respectively, Table 1.There was no significant difference in BCVA between these 2 groups at 10 years as well as at 20 years, p = 0.41 and p = 0.17, respectively.

| Changes in visual acuity of the operated eye, best cases
The BCVA stratified by the 2 selected subgroups of best cases with or without diabetes is shown in Table 1.At 20 years a larger proportion of the control group (63/306;21%) vs. the best cases diabetics (6/38;16%) remained in the study, this difference was not significant (p = 0.49, chi-square test).
All best case patients, diabetics and non-diabetics, had their BCVA improved by the cataract surgery.There was no significant difference in median BCVA between the best cases diabetics or controls at any examination, preoperatively p = 0.96, postoperatively p = 0.87, 10 years after surgery p = 0.45, and 20 years after surgery p = 0.81, Table 1.Further analysis with repeated measures test controlling for age did not change the results.
Ten, 15 and 20 years after surgery, participating diabetics had lost fewer letters than the non-diabetics, but this difference was not significant, p = 0.40, p = 0.24 and p = 0.50, respectively.At 15 years after surgery the remaining diabetics had improved their median BCVA with one letter compared with the results from 10 to 20 years after surgery, Figure 3.
Ten years after surgery, BCVA was equal or improved compared with postoperative levels, in 10 of 17 (58%) of the best cases of diabetics, and in 88 of 170 (52%) of the best cases non-diabetics, p = 0.58.The corresponding figures at 20 years after surgery were 2 of 6 (33%) in the diabetics and 22 of 63 (35%) of the non-diabetics, p = 0.94.
All (100%) examined best cases diabetics at 10 and 20 years after surgery had better BCVA of the operated eye than it had been before surgery.The corresponding figures for non-diabetics were 162 of 170 (95%) and 59 of 63 (94%), p = 1.

| Subjective visual function (VF-14 questionnaire)
Table 2 shows the demographics and the VF-14 score, in relation to time of type 2 DM diagnosis, in patients who participated with the VF-14 questionnaire.
There was no significant difference in self-perceived visual function between the diabetics and the nondiabetics at any time, including VF-14 score before surgery (p = 0.72, repeated measures test).As a result of the cataract surgery, the diabetics as well as the non-diabetics had a significant improvement in VF-14 score, p < 0.001, all groups.There was no significant difference between the groups in Table 2, regarding gender and the percentage of patients who had surgery on their first eye when the study started, p = 0.55 and p = 0.96, respectively.
Figure 4 shows the longitudinal change in VF-14 score from before cataract surgery to 20 years after surgery in the 109 patients who had diabetes at surgery and the 651 non-diabetics.Even though the mean VF-14 score was better in the diabetics at 10, 15 and 20 years after surgery this difference was not significant, p = 0.72, p = 0.20 and p = 0.78, respectively.

| Visual acuity results in relation to severity of diabetic retinopathy
This analysis is based on patients with diagnosis of type 2 diabetes at the time of cataract surgery (n = 109).As shown in Figure 1, 66 of 109 patients (61%) had no DR, 18 patients (16%) had mild NPDR, 14 patients (13%) had moderate to severe NPDR and 11 patients (10%) had proliferative DR past or present at the time of cataract surgery, Figure 1. as having moderate to severe DRP at surgery survived more than 10 years.At 20 years after surgery, only 6 patients with type 2 diabetes preoperatively remained in the study.
There was no significant difference in median BCVA between the different DRP groups at any examination before or after surgery, p-values displayed in Figure 5.
However, there is a trend from 10 years after surgery and onward that patients with no DRP at baseline lost fewer letters during the 20 years than those with retinopathy.Surviving patients with DRP had only a small decline in BCVA, at 20 years, Figure 5.

| DI SC US SION
The current study investigates the 20-year visual outcome after cataract surgery in diabetics, compared with non-diabetics.This study shows the majority of diabetics have, as well as the non-diabetics, a substantial gain in BCVA after cataract surgery.The longterm results of BCVA and subjective visual function (VF-14 questionnaire) in diabetics are comparable to non-diabetics.This favourable outcome remains up to 20 years postoperatively for the majority of surviving patients.This is consistent with previous research showing good results with improved visual function also in diabetics, especially those without retinopathy (Danni et al., 2019;Liu et al., 2019;Ostri et al., 2011).Several studies report worse postoperative BCVA and less VA gain in diabetics with increasing severity of retinopathy and age (Gupta et al., 2018;Han et al., 2020).Particularly the vision-threatening stages of DR, have a substantial negative impact on questionnaire assessed overall vision-dependent functioning (Gupta et al., 2018).
In the present report, DR if any, was graded at the time of the cataract surgery.The study was not designed to assess the possible influence of cataract surgery on the postoperative change in DR, and potential effects on the visual results.Too few patients with retinopathy and 15-20 years of survival would have given insufficient power for studying this aspect of the longtime outcome.

F I G U R E 4
Longitudinal change in mean VF-14 total score for patients participating with the questionnaire from preoperatively to 20 years after surgery, in the 109 patients who had diabetes at surgery and the non-diabetics (n = 651).VF-14 = Visual functional 14 (questions) questionnaire.Type 2 diabetes is associated with the patient not being aware of the diagnosis.In industrialized countries, it is estimated that more than onethird of people with diabetes are undiagnosed (Wong & Sabanayagam, 2019).It is also not uncommon that patients have developed DR at the time of diagnosis, implying several years of undiagnosed type 2 diabetes (Laatikainen et al., 2016).Diabetes duration is a strong predictor of DR, and it has been estimated that more than 60% of those with type 2 diabetes will develop some degree of DR after 20 years with DM (Cheung et al., 2010;Gupta et al., 2018).
As the time span of this study comprises up to 20 years after surgery, it was necessary to separately report patients who were diagnosed with diabetes mellitus after the cataract surgery.The inclusion of these patients would bias the long-time results in a positive direction.As shown in Tables 1, 2 these patients were younger and had a better outcome both regarding BCVA and the VF-14 results.This is in agreement with previous research, that visual outcome was slightly better in patients with a duration of diabetes less than 10 years vs. more than 10 years (Liu et al., 2019).
Not surprisingly, patients with ocular comorbidity other than DR and cataract, for instance, age-related macular degeneration (AMD), glaucoma and epiretinal membrane, have lower odds of achieving a postoperative BCVA of 20/20 or better.This is probably caused by poor macular or optic nerve function limiting the visual potential.Therefore, to further refine the results, a separate analysis was made on "best cases patients," who were patients with or without a diagnosis of diabetes before surgery and had no other ocular disease except cataract.There was also no significant difference in long-term visual outcome, when the "best cases patients", diabetics vs. non-diabetics, were compared, Table 1, Figure 3.
Even though insignificant, the participating "best cases diabetics" in this study had lost fewer letters than the "best cases non-diabetics" 10, 15 and 20 years after surgery, Figure 3.One explanation for this finding might be that the study cohort of diabetics comprised patients with good compliance with medications and treatments.They were also willing to be examined at the study follow-up every 5 years, which might bias the diabetic's group towards those with very good compliance.It may also play a role that recurrent retinal screening examinations increase the opportunity to detect ocular comorbidity other than DR in the diabetic population.Screening compliance has been shown to result in better visual outcomes (Zoega et al., 2005).This may explain why in the present study, the preoperative BCVA of the diabetics was slightly better than in the non-diabetics.
Diabetes care of today in industrialized countries includes expanded screening programs for early detection and treatment of DR, as well as marked improvements in eye care for the ocular complications of diabetes.The relative incidence and progression of DR, PDR and associated blindness have declined over the past decades, at least in high-income countries (Wong & Sabanayagam, 2019).However, DR is still an important cause of vision loss (Thomas et al., 2015).
The presence and increasing severity of DR have repeatedly been associated with decreased survival (Knudtson et al., 2006).As a result of advancements in monitoring and treatment of high blood glucose levels, blood pressure and serum lipid levels, life span has increased also in diabetics.The mortality rate is, however, still greater than in the general population (Cheung et al., 2010;Laatikainen et al., 2016).Survival is even more compromised among diabetics with complications from the disease.This might explain why a significantly smaller percentage of the diabetics diagnosed before cataract surgery were alive at 10, 15 and 20 years (p = 0.003), compared with non-diabetics, Table 1.Comparatively fewer diabetics live more than 10 years after surgery, although no significant difference in age at surgery.It might be the least affected diabetics who survive and therefore have better outcomes.
Pseudophacic macular oedema (ME) remains a common cause of reduced VA gain after cataract surgery.It is well-known there is an association between the degree of DR and the development of diabetic macular oedema (DME) after cataract surgery (Romero-Aroca et al., 2006;Ylinen et al., 2017).The present study was not designed to assess the impact and development of DME in association with cataract surgery.As the most important clinical result caused by DME is low BCVA postoperatively, it was believed that reporting BCVA results would be sufficient.Only a small proportion, ~1% of cataract surgery patients develop clinically significant pseudophacic macular oedema (Chu et al., 2016), and the occurrence of DME might not increase after cat surgery in diabetics without DR (Ylinen et al., 2017).
The major strength of this study is the prospective, long-term, longitudinal study design.The long follow-up time, as well as the large number of operated patients, who were followed closely and systematically, enabled high follow-up rates with low rates of drop-outs and refusals.The population-based design reduces the risk of referral bias.The study has the advantage of a resource-rich, integrated health care delivery system setting, in which everyone had access to comprehensive eye care.
A key concern with observational studies is the potential for unmeasured confounding, or selective unavailability in which patients with diabetes are different from those without diabetes in ways that are not recorded and cannot be adjusted.The high participation rate of the survivors included in the study (84%, Table 2), reduces the impact of selective unavailability.The only reasons for not participating at least with the questionnaire, were dementia and inability to locate the patient.
Another strength of the study is that even though the cataract surgeries were performed 20 years earlier, the current surgical technique was comparable to methods used in cataract surgery today.Most PE surgery of today uses a smaller incision size of 2.2 mm or less instead of 3.2 mm, and IOLs are implanted using injectors and are not folded with forceps as in 1997-1998.However, modern diabetic care has improved compared with 20 years earlier, for example by preoperative administration of anti-VEGF.This development would however not the results, but rather improve them.
There are several limitations of the study.The relatively small number of surviving diabetics from 10 years or more after surgery, causes small samples and less power to detect meaningful relationships.The mean age at surgery in 1997-1998 was ~75 years, which is a common mean age in most cataract surgery cohorts, and the main reason for drop-out was death.Although no evidence of worse visual functional outcomes in diabetics compared with non-diabetics was found, there could nevertheless be a possibility of lack of power to detect such associations.It is also the diabetics with compromised health who have the shortest survival.Therefore, the results are probably influenced by selection bias or healthy patient bias.The remaining participants still alive (and thus included and remaining in the study) might be more health-conscious and have a higher level of health-care utilization than the dropouts and deceased.
These findings imply the results can be used for surgical decision-making and preoperative counselling of diabetics facing cataract surgery.Knowledge of these long-term outcomes provides essential information for healthcare providers and patients, as knowledge of prognostic factors for BCVA is valuable.Rising life expectancy and greater dependence on digital devices have increased patients' demand for good functional vision and well-being.Surgeons are recommended to set realistic expectations of postoperative visual function in patients with moderate to severe non-proliferative DR and proliferative DR.Further research may be necessary to examine other factors that may affect the outcomes of cataract surgery in type 2 diabetics.
In conclusion, this population-based, longitudinal study provides insight into the long-term outcome of cataract surgery in diabetics.Timely cataract surgery is beneficial in most diabetics who postoperatively in many cases have excellent subjective visual function.Diabetics with significant cataract can be advised there is good chance the visual prognosis might be sustained as long as 20 years after cataract surgery in patients with good adherence to follow-up eye care.

AC K NO W L E DGE M E N T S
Funding: Grants from the Västerbottens County Council Research Fund, Umeå, Sweden, the Swedish Government ("Agreement concerning Research and Education of Doctors" ALF), Ögonfonden, Gothenburg, Sweden and Capio Medocular, Stockholm, Sweden are acknowledged.The sponsor or funding organization had no role in the design or conduct of this research.

CON F L IC T OF I N T E R E ST STAT E M E N T
The author has no financial disclosures and no possible conflicts of interest to report.

F
Flowchart on study population showing the longitudinal flow of patients who had cataract surgery for senile or presenile cataract during June 1st 1997 to May 31st 1998 at Norrlands University Hospital, Umeå Sweden.DM, diabetes mellitus; DR, diabetic retinopathy; NPDR, nonproliferative diabetic retinopathy.
Demographics and pre-and postoperative mean VF-14 score, in patients who participated in the VF-14 questionnaire, in relation to time of diagnosis of type 2 diabetes.

Figure 5
shows the longitudinal 20-year change in median BCVA in relation to preoperative grade of DRP.The number of patients examined from 10 years or more postoperatively are few.No patient classified F I G U R E 2 Longitudinal change in median best corrected visual acuity in the operated eye from preoperatively over 20 years after surgery, in diabetics at surgery and non-diabetics.The vertical scale shows logMAR visual acuity.Approximate Snellen values are; logMAR VA 0 = 20/20, 0.2 = 20/29, 0.4 = 20/50, 0.6 = 20/80, 1.0 = 20/200.logMAR, logarithm of the minimum angle of resolution; VA, visual acuity.Postoperative change in median BCVA in all "best cases" patients, diabetics and non-diabetics.BCVA-change was calculated by subtracting the 1-2 months postoperative BCVA (logMAR VA), and the BCVA (logMAR VA) score at each followup examination after surgery, a positive value indicating improved visual acuity.Change in BCVA is expressed in logMAR units and 0.02 represents one letter on the logMAR chart.BCVA, Best corrected visual acuity; LogMAR, logarithm of the minimum angle of resolution; VA, visual acuity.
(109/807;14%) No DR (66/109;61%) Mild NPDR (18/109; 16%) Moderate to severe NPDR (14/109;13%) Proliferative DR (11/109; 10%) | 61 MÖNESTAM T A B L E 1 Preoperative and postoperative median best-corrected visual acuity and age, in relation to time of diagnosis of type 2 diabetes.Characteristic, Note: Five-and 15-year data not shown.The numbers in brackets refer to the number of patients contributing with BCVA data.Age is defined as age in years at cataract surgery and at the examination at 10 and 20 years, respectively.Abbreviations: BCVA, best corrected visual acuity; logMAR, logarithm of the Minimum Angle of Resolution. a Snellen geometric mean computed by converting the average logMAR to Snellen (Holladay JT, Prager TC.Mean visual acuity.Am J Ophthalmol 1991;111:372-4).bBestcasesdiabetics, are defined as patients who at surgery had known type 2 diabetes with or without diabetic retinopathy, but no other ocular disease but cataract.cBest cases control group had no other ocular disease than cataract.
The numbers in brackets refer to the number of patients in each group contributing to VF-14 total score data.
a b