Visual Acuity Outcomes after Cataract

Surgery
High-Volume versus Low-Volume Surgeons
Jacob T. Cox, MD, MPhil,1 Ganesh-Babu B. Subburaman, MS, MBA,2 Beatriz Munoz, MS,1
David S. Friedman, MD, PhD,1 Ravilla D. Ravindran, MS, DO2

Purpose: The primary objective was to assess associations between annual surgeon case volume and visual
acuity outcomes after cataract surgery. Secondary objectives included (1) assessing associations between other
case and patient characteristics and visual acuity outcomes and (2) assessing associations between annual
surgeon case volume and complication rates.
Design: Database study.
Participants: All adult eyes that underwent small-incision cataract surgery (SICS) or phacoemulsification
cataract extraction (PECE) with intraocular lens placement at the Aravind Eye Hospital, Madurai, India,
during 2015.
Methods: Descriptive statistics were used to characterize the study population. Uncorrected visual acuity
(UCVA) at follow-up was assessed relative to annual surgeon case volume and other case and demographic
factors using bivariate linear regression with random effects modeling. Factors with P values of less than 0.20 on
bivariate regression were included in multivariate linear regression with random effects modeling.
Main Outcome Measures: Postoperative UCVA after cataract surgery.
Results: Of 91 084 surgeries, 35 880 eyes were included in this study. Cases were performed by 69 sur-
geons, who varied in annual case volume from 76 to 2900 cases during the study period. Increasing annual
surgeon case volume was independently associated with a statistically significant but clinically modest
improvement in UCVA in PECE but not in SICS. This association was most pronounced when comparing sur-
geons with case volumes of 350 PECE/year or fewer; surgeons with more than 350 PECE/year had similar results
to one another. Similarly, increased annual case volume was associated with significantly lower complication
rates, both in PECE and SICS. Younger patient age was independently associated with improved visual acuity
outcomes and lower complication rates in both PECE and SICS. Greater surgeon experience was associated with
lower complication rates in PECE, but not SICS, and there was no significant association with visual acuity
outcomes.
Conclusions: High-volume cataract surgeons showed improved visual acuity outcomes in PECE and lower
complication rates in PECE and SICS. These findings further support the benefit of high-output ophthalmology
clinics wherein individual surgeons perform a high number of cataract extractions annually, particularly in
developing nations where there is a large backlog of untreated cataracts and the cataract patient-to-surgeon ratio
is high. Ophthalmology 2019;-:1e10 ª 2019 by the American Academy of Ophthalmology

Cataract is the leading cause of blindness globally, with low-
and middle-income countries (LMICs) showing a dispropor-
tionate burden of disease.1 The World Health Organization
projects that an annual cataract surgery rate of 4000 cases
per 1 million people is needed to eliminate cataract-induced
blindness, and a higher rate still would be needed to elimi-
nate moderate to severe visual impairment resulting from
cataract.2 However, many LMICs currently fall significantly
short of this target, leading to an increasing backlog of cases
within those nations. Implementation of high-volume cata-
ract surgery models potentially could ease the growing prev-
alence of cataract-induced blindness in developing regions.
Such a model has been implemented successfully at the
Aravind Eye Care System (AECS) in India and elsewhere.3e5
Founded in 1976, AECS currently consists of 12 ophthalmic
hospitals in Tamil Nadu, India, where approximately 470 000
eye surgeries, laser procedures, and intraocular injections were
performed from 2016 through 2017, 285 000 of which were
cataract extractions with intraocular lens (IOL) placement.3e5
Many AECS ophthalmologists perform more than 2000
cataract surgeries per year.3,4 The primary techniques used
are small-incision cataract surgery (SICS) and phacoemulsi-
fication cataract extraction (PECE). For readers not familiar
with the technique, SICS involves removal of the undivided
cataract nucleus. Thus, it involves use of a self-sealing scle-
rocorneal tunnel that corresponds to the size of the nucleus, as

ª 2019 by the American Academy of Ophthalmology https://doi.org/10.1016/j.ophtha.2019.03.033 1

Published by Elsevier Inc. ISSN 0161-6420/19
 Ophthalmology Volume -, Number -, Month 2019
well as a larger capsulorrhexis (6.0e7.0 mm) than that used in
PECE. It is a commonly used technique in lower-resource
settings because it is less expensive, generally has a shorter
operating time, and has been associated with lower compli-
cation rates than PECE among surgical trainees in developing
contexts.6,7 Further detail on the specifics of this technique
and management of its complications are well described in the
literature.8e11
Although it is well documented that increased annual
case volume is associated with decreased complication rates,
data regarding the impact of case volume on visual acuity
outcomes are limited.12,13 The primary aim of this study was
to identify any associations between surgeon case volume
and visual acuity outcomes. Secondary aims were to identify
associations between (1) other case characteristics and
visual acuity outcomes and (2) surgeon case volume and
complication rates.
Methods
Patient Selection
This study complied fully with the tenets of the Declaration of
Helsinki and received approval from the AECS Institutional Ethics
Committee. As a retrospective chart review in which all data were
devoid of patient identifiers, we did not seek informed consent from
study participants. Patients were selected using the AECS cataract
surgery database for retrospective review of all patients who un-
derwent SICS or PECE with intended IOL placement at the Aravind
Eye Hospital in Madurai over the course of 2015. Exclusion criteria
included: (1) surgeries conducted by residents or other trainees; (2)
surgeries conducted by physicians who had completed fewer than
100 cataract extractions as an attending-level physician by the
beginning of the study period; (3) postoperative uncorrected visual
acuity (UCVA) measurements not recorded in the medical record;
(4) visual acuity data collected fewer than 3 weeks into the post-
operative period; (5) presence of the following on baseline exami-
nation for patients in whom the posterior pole could be visualized:
glaucoma, pseudoexfoliation syndrome, guttata, small pupil, dia-
betic retinopathy, age-related macular degeneration, previous
penetrating ocular trauma, or previous ocular surgery; (6) patient
age younger than 16 years; (7) annual surgeon case volume fewer
than 50 cases per year; and (8) an IOL was not placed or there was
incongruence between patient IOL power measurements and those
of the IOL placed. Cases were categorized as level 1, 2, or 3 based on
case complexity. Level 1 cases were defined as PECE with dilated
pupil and immature cataract or SICS with dilated pupil and mature
or immature cataract. Level 2 and 3 cases were defined by mature
cataract (PECE only), hypermature cataract, hard cataract, posterior
polar cataract, scleral- or iris-fixated IOL, IOL power less than 10
diopters, axial length less than 20 mm, multifocal IOL, toric IOL,
use of capsular tension ring, or use of femtosecond laser. Because
high-volume surgeons performed a significantly larger proportion of
complex cases (P < 0.001), only cases with level 1 complexity were
included for analysis to allow for greater comparability between
surgeons across case volume levels. Of note, surgeon case volume
was determined based on the total number of cataract surgeries
performed by the surgeon, rather than simply the number of cases
that met inclusion criteria.
Visual Acuity Measurements
The intended final refraction in all eyes was emmetropia. Uncor-
rected visual acuity was measured as part of routine clinical care in
2
a variety of settings using Snellen charts, and the results were
converted to the logarithm of the minimum angle of resolution
(logMAR) scale for analytical purposes. Thus, 20/20 vision cor-
responded to a score of 0.0 logMAR, 20/40 vision corresponded to
a score of 0.3 logMAR, and so on. Counting fingers vision was
recorded as 1.8 logMAR, hand movement vision was recorded as
2.3 logMAR, light perception was recorded as 2.8 logMAR, and no
light perception was recorded as 3.0 logMAR.14,15 Uncorrected
visual acuity was selected as the primary outcome because, unlike
best-corrected visual acuity (BCVA), UCVA also enabled us to
account for surgically induced astigmatism and thus more
comprehensively represented surgical outcomes. Moreover, the
lack of access to corrective eyewear for many patients in devel-
oping contexts makes UCVA a more appropriate measurement of
true day-to-day vision.
Complications
For analytical purposes, complications comprised both intra-
operative and postoperative complications. Intraoperative compli-
cations included: detachment of Descemet’s membrane (excluding
small, visually insignificant detachment at the main or side port);
iridodialysis of more than 3 clock hours; disruption of the posterior
capsule or separation of zonules from the capsular or ciliary
attachment leading to communication between the anterior and
posterior chambers, with or without disturbance of the anterior
hyaloid face of the vitreous membrane; inability to implant an IOL
(e.g., because of inadequate capsular support or expulsive
choroidal hemorrhage); and dislocation of crystalline lens matter
into the vitreous; or dislocation of the IOL into the vitreous with
subsequent failure of immediate retrieval. Postoperative compli-
cations were determined within 24 hours of surgery by slit-lamp
examination and included: wound leakage secondary to a poorly
constructed wound, wound burn, large wound size, or a broken or
loose suture; iris prolapse; iris incarceration within the wound;
vitreous extrusion through the wound; corneal edema of 2þ or
more; detachment of Descemet’s membrane that involved more
than one third of the total cornea, that involved the visual axis, that
required reoperation, or that was associated with postoperative
visual acuity of 5/60 or worse; iritis with 15 or more cells evident
in a 21-mm slit beam in the anterior chamber; organization of
fibrin exudate with inflammatory cells in the anterior chamber;
iritis with hypopyon; retained cortical material significant enough
to cause sequelae (e.g., inflammation, increased intraocular pres-
sure, compromised visual acuity, cystoid macular edema, corneal
decompensation, or required reoperation); hyphema occupying one
third or more of the anterior chamber; presence of vitreous in the
anterior chamber; or IOL decentration that was visually significant
or required further intervention.
Statistical Methods
Descriptive statistics were used to characterize the study popula-
tion; these were reported as proportions, means with standard de-
viations, or medians with interquartile ranges as appropriate.
Analyses were split into 2 subsets based on surgical technique:
SICS or PECE. Within each subset, bivariate linear regression
analyses were run with random effects modeling to weigh each
surgeon equally despite differences in surgical volume; post-
operative UCVA was the dependent variable in these regression
analyses. All covariates approaching significance (P < 0.20) on
bivariate analysis were incorporated into multivariate regression
analyses (mutivariate linear regression with random effects
modeling). Logistic regression was performed to compare UCVA
outcomes between high- and low-volume PECE surgeons; logistic
regression diagnostics were performed accordingly, including the
 Cox et al 
Cataract Outcomes by Surgeon Case Volume
Hosmer and Lemeshow’s goodness-of-fit test. The variance infla-
tion factor was calculated for each covariate in all multivariate
regression analyses to assess for multicollinearity.
In SICS analyses, both SICS and PECE cases were included
when determining each surgeon’s annual case volume, whereas only
PECE cases were included when determining annual case volume in
PECE analyses. This distinction was made under the premise that
surgical experience with either technique may enhance a surgeon’s
SICS capabilities, whereas the additional steps and equipment
involved in PECE may render a surgeon’s SICS experience less
relevant. Surgeons who performed fewer than 50 cataract extrac-
tions (SICS and PECE combined) in the study period were excluded
from SICS analyses, and surgeons who performed fewer than 50
PECE in the study period were excluded from PECE analyses.
We sought to account for surgeon experience using 2 measures:
(1) surgical experience in terms of number of years in practice
(termed years of experience) and (2) surgical experience in terms of
number of prior cases (termed prior case load). Years of experi-
ence refers to the number of years each surgeon has been in
practice since completing residency; although this is communi-
cated in units of years in this manuscript, these data were recorded
in months and were updated throughout the study period (2015) to
reflect the number of months of experience each surgeon had at the
time of each case. Prior case load is an estimated proxy for the
number of cataract surgeries each surgeon has performed in their
career; this was calculated by referencing the number of cases each
surgeon performed for the 5-year period preceding the study period
(i.e., 2010e2014), calculating the average annual case volume for
each surgeon based on those data, and then multiplying that
average by the surgeon’s number of years in practice as an
attending-level ophthalmologist. Similar to determining annual
case volume, only PECE cases were included when determining a
surgeon’s prior case load for PECE analyses, but both PECE and
SICS cases were included when determining a surgeon’s prior case
load for SICS analyses.
Coefficients were reported in tables outlining linear regression
outcomes to demonstrate the magnitude of associations. These
coefficients represent the magnitude of change in UCVA (in log-
MAR units) for every 1-unit increase in the given independent
variable. For example, the coefficient for multivariate regression
analysis of the association between surgeon volume and post-
PECE UCVA outcomes among lower-volume surgeons (defined as
surgeons with 50e350 PECE/year) is e0.0238, which represents
an improvement of 0.0238 logMAR (1.5 ETDRS letters) per 100-
case increase in surgeon volume. All statistical analyses were
carried out using Stata software version 13 (StataCorp LLC, Col-
lege Station, TX).
Results
Aggregate Data
A total of 91 084 cataract extractions with IOL placement were
performed at the Aravind Eye Hospital in Madurai, India, during the
study period. Data from 35 880 eyes met the inclusion criteria and
were analyzed. Overall mean UCVA at follow-up was 0.320.24
logMAR (Snellen equivalent: w 20/40; Table 1). Postoperative
UCVA was better for PECE than SICS (P  0.001), with mean
UCVA after PECE of 0.140.17 logMAR (Snellen equivalent:
w 20/25) and mean UCVA after SICS of 0.410.22 logMAR
(Snellen equivalent: w 20/50). Most cases were SICS (65.8%).
The average patient age was 60.59.1 years, and slightly more
than half of patients were women (54.7%). Surgeon experience as
measured by years in practice correlated roughly with annual case
volume. Surgeons with fewer than 250 cases/year and surgeons
with 251e500 cases/year had the least experience, with a median
of 2.8 and 3.2 years in practice, respectively; those with
1501e2000 cases/year were the most experienced, with a median
of 4.7 years in practice. All remaining case-volume categories had
similar experience levels (median: 3.9e4.1 years in practice).
Surgeon experience as measured by the estimated number of prior
cataract surgeries performed also generally increased as annual case
volume increased; the lowest 3 case-volume categories showed
medians between 254 and 557 prior cases as an attending-level
physician, whereas the highest 3 case-volume categories showed
medians that were an order of magnitude higher, ranging from 4527
to 8417 prior cases. Generally, UCVA outcomes improved as sur-
geon volume increased. However, this effect seemed to plateau at
approximately 500e1000 cases/year. Surgeons performing
1501e2000 cases/year did not fit this UCVA trend given that their
visual acuity outcomes were worse than all other surgeon groups
aside from the lowest-volume cohorts (50e250 cases/year and
251e500 cases/year). Although higher-volume surgeons showed a
significantly higher proportion of high-complexity cases (P <
0.001), we included only cases with level 1 complexity to allow for
greater comparability between surgeons across case-volume levels,
as described in Methods.
When reviewing Table 1, it is worth noting that roughly 60% of
all cases did not meet inclusion criteria for this study. The
percentage of cataract extractions that met inclusion criteria
generally increased as surgical volume increased. By case-
volume category: (1) 50e250 case/year surgeons completed
6571 cases (15% met inclusion criteria); (2) 251e500 case/year
surgeons completed 6573 cases (34% met inclusion criteria); (3)
501e1000 case/year surgeons performed 8899 surgeries (40%
met inclusion criteria); (4) 1001e1500 case/year surgeons
completed 17 184 cases (40% met inclusion criteria); (5)
1501e2000 case/year surgeons performed 21 981 cases (45%
met inclusion criteria); and (6) 2001 cases/year completed 29
876 cataract extractions (41% met inclusion criteria).
The most common reasons for exclusion were (1) high case
complexity (level 2 or 3 complexity), especially among higher-
volume surgeons; (2) surgeons having performed fewer than 100
prior cases as an attending-level physician, especially among
lower-volume surgeons; (3) visual acuity data collected fewer than
3 weeks into the postoperative period; and (4) ocular comorbidities
that may directly alter visual acuity or complicate cataract extrac-
tion (e.g., age-related macular degeneration, diabetic retinopathy,
glaucoma, pseudoexfoliation syndrome, or prior ocular surgery).
Small-Incision Cataract Surgery
Bivariate regression analysis of SICS cases found that annual case
volume did not achieve a statistically significant association with
UCVA outcomes (P ¼ 0.33; Table 2). Surgeon experience as measured
by years since residency and by prior case load was not significantly
associated with improved postoperative acuity on multivariate
analysis (P ¼ 0.27 and P ¼ 0.55, respectively). However, lower
patient age was independently associated with improved UCVA
outcomes (P < 0.001). Of note, there was no concern for
multicollinearity given that the tolerance for each covariate
(calculated as 1 / uncentered variance inflation factor) was greater
than 0.1.
Phacoemulsification Cataract Extraction
Figure 1 is a locally weighted scatterplot smoothing (LOWESS)
curve fitted to UCVA outcomes for PECE cases based on annual
case volume. This fitted curve demonstrated an asymptotic
relationship between UCVA and surgical volume with an
inflection point at 350 PECE/year. Thus, the association between
3
4

Table 1. Patient and Case Characteristics According to Surgeon’s Annual Case Volume (n ¼ 35 880)

Ophthalmology Volume -, Number -, Month 2019

Surgeon’s Annual Case Volume
Variable 50e250 251e500 501e1000 1001e1500 1501e2000 2001þ Total
Surgeon characteristics
No. of surgeons 10 13 9 11 12 14 69
Years of experience, median (IQR) 2.8 (2.7e2.8) 3.2 (2.0e3.7) 3.9 (2.1e12.0) 4.0 (1.7e11.9) 4.7 (2.7e8.2) 4.1 (3.2e5.0) 3.7 (2.7e7.2)
Prior case load, median (IQR) 435 (348e481) 254 (138e2448) 557 (202e5060) 8417 (1418e21 515) 5237 (1611e15 674) 4527 (3763e13 610) 2200 (299e9692)
Case characteristics
Total no. of cases meeting 985 (15%) 2255 (34%) 3602 (40%) 6918 (40%) 9889 (45%) 12 231 (45%) 35 880 (41%)
inclusion criteria (% of all cases)
UCVA at follow-up, meanSD 0.390.22 0.370.23 0.290.23 0.280.23 0.350.25 0.300.24 0.320.24
Patient age (yrs), meanSD 59.98.1 60.38.4 60.38.6 60.59.0 60.89.1 60.69.5 60.59.1
Patient gender (%)
Male 37.3 43.1 44.0 44.9 44.9 47.2 45.3
Female 62.7 56.9 56.0 55.1 55.1 52.8 54.7
Surgery type (%)
SICS 95.5 83.6 59.4 57.3 71.7 61.9 65.8
PECE 4.5 16.4 40.6 42.7 28.3 38.1 34.2
Any complications (%)
No 97.6 97.4 98.7 97.8 97.9 98.3 98.1
Yes 2.4 2.6 1.3 2.2 2.1 1.7 1.9

IQR ¼ interquartile range; PECE ¼ phacoemulsification cataract extraction; SD ¼ standard deviation; SICS ¼ small-incision cataract surgery; UCVA ¼ uncorrected visual acuity.
Both PECE and SICS cases are included. Uncorrected visual acuity measurements presented in logarithm of the minimum angle of resolution scale.
Data included both SICS and PECE cases and only reflect cases performed as an attending-level physician. The data do not include surgeries performed as a resident, fellow, or other trainee.
 Cox et al 
Cataract Outcomes by Surgeon Case Volume
improved visual acuity outcomes and increasing surgeon volume is
most pronounced when the annual case volume is less than 350
PECEs/year (P ¼ 0.02; Table 3). Conversely, UCVA outcomes
largely are constant for any surgical volume of more than 350
PECEs/year (P ¼ 0.80). Comparing the 2 groups, UCVA
outcomes were statistically significantly better among >350
PECEs/year surgeons than a 350 PECEs/year surgeons (P ¼
0.01), as determined using multivariate logistic regression
controlled for patient age, surgeon experience in years, and
surgeon prior case load. As with our SICS analyses, there was no
concern for multicollinearity in the multivariate regression
analyses (both linear and logistic), given that the tolerance for
each covariate included (calculated as 1 / uncentered variance
inflation factor) was greater than 0.1.
Complication Rates
The average complication rate for all surgeons, accounting for both
intraoperative and postoperative complications, was 1.91.0%.
The mean SICS complication rate was 1.71.0%, with a compli-
cation rate of 2.41.9% among lower-volume surgeons (500
cases/year) and 1.60.8% among higher-volume surgeons (>500
cases/year). The mean PECE complication rate was 2.41.8%,
with a rate of 4.43.2% among lower-volume surgeons (350
PECEs/year) and 2.01.1% among higher-volume surgeons
(>350 PECEs/year). Multivariate linear regressiondadjusted for
patient age, patient gender, surgeon experience in years, and sur-
geon prior case loaddfound a statistically significant decrease in
complication rates as annual case volume increased; this was true
for SICS and PECE procedures (P ¼ 0.02 for SICS, P < 0.001 for
PECE). This volume-associated improvement in complication rates
was more pronounced in PECE than SICS procedures, with esti-
mated decreases in complication rates of 0.1% and 0.03% for every
100-case increase in annual case volume, respectively. This
multivariate analysis also found that younger patient age was sta-
tistically significant for lower complication rates in both surgical
techniques (P ¼ 0.002 for SICS and P < 0.001 for PECE);
increased prior case load was also statistically significant for lower
complication rates in PECE (P ¼ 0.01) but not SICS (P ¼ 0.17)
(Fig 2).
Discussion
This large study (35 880 eyes) of UCVA outcomes after
cataract surgery at the Aravind Eye Hospital in Madurai,
India, found that visual acuity was significantly better after
PECE than SICS (P < 0.001). Annual case volumes (PECE
and SICS combined) ranged from 76 to 2900 surgeries.
When considering PECE and SICS cases in aggregate, vi-
sual acuity outcomes and complication rates improved as
case volume increased, but this association did not achieve
statistical significance. When analyzing the 2 surgical
techniques separately, SICS showed a nonsignificant trend
toward improvement as volume increased whereas PECE
had a statistically significant association between increased
surgical volume and improved UCVA outcomes. This as-
sociation seems to apply mainly to surgeons with fewer than
350 PECEs/year, as outcomes remained largely constant at
annual case volumes higher than this threshold. Higher
annual case volume was also associated with lower
complication rates for both PECE and SICS. Although this
volume-associated improvement in complication rates was
more pronounced in PECE than SICS, its clinical magnitude
was modest in both techniques.
Our finding that use of the phacoemulsification technique
was associated independently with improved visual acuity
outcomes compared with SICS (P < 0.001) is similar to
findings reported in a 2014 Cochrane Review of randomized
control trials.10 That review reported a potential short-term
UCVA benefit with phacoemulsification and no significant
difference in BCVA outcomes between the 2 techniques.10
We speculate that, as in the Cochrane review, the UCVA
benefit of phacoemulsification may be short term and the
result of subclinical postoperative cystoid macular edema
experienced by SICS patients.10,16,17 This effect may still
impact visual acuity during the timeframe when many par-
ticipants in this study underwent postoperative UCVA
measurements given that 84.8% of UCVA measurements
were collected 21 to 35 days after surgery. However, SICS
is also associated with a slightly higher rate of astigmatism
than PECE, which may further explain the difference in
UCVA outcomes.18,19 Unlike the Cochrane Review, our
study does not assess BCVA outcomes. As mentioned in
Methods, we chose to focus on UCVA rather than BCVA
because this enabled us to account for surgically induced
astigmatism, thus providing a more comprehensive view of
surgical outcomes. Moreover, many individuals in LMICs
lack access to corrective eyewear, making UCVA a more
accurate reflection of true day-to-day vision. Uncorrected
visual acuity was considered a valid measure of acuity
outcomes because the goal was emmetropia in all eyes.
It is worth noting that although the volume-associated
improvement seen in post-PECE UCVA was statistically
significant, the clinical significance was modest. Our ana-
lyses indicate that performing 50 PECEs/year, which is the
lowest surgical-volume threshold for inclusion in this study,
would correlate with a UCVA outcome of 0.17 logMAR
(Snellen equivalent: w 20/30), whereas performing 350 or
more PECEs/year would correlate with a UCVA outcome of
0.14 logMAR (Snellen equivalent: w 20/25). Nevertheless,
the statistical significance on multivariate regression and the
locally weighted scatterplot smoothing plot in Figure 1
suggest that these findings are reliable and represent a true
association between postoperative acuity and surgical
volume. These findings also may underestimate the
magnitude of this association, given that our data were
limited to uncomplicated cataract cases. We speculate that
factors such as surgical volume and surgeon experience
may play a greater role in outcomes when cases are more
complex. This also poses the question of whether 350
PECEs/year would still serve as the plateau point for
visual acuity improvement in higher-complexity cases or if
the plateau point would be at a higher volume, given that
added surgical volume beyond 350 cases/year may continue
to provide benefit in higher-complexity cases. Given these
questions, further research is warranted into the effects of
surgical volume and surgeon experience on outcomes in
complex cases.
Our findings differ from those of Habib et al,20 who
compared high-volume and low-volume surgeons perform-
ing PECE and found no association between surgeon volume
and postoperative visual acuity. However, their study divided
5
 Ophthalmology Volume -, Number -, Month 2019

Table 2. Bivariate and Multivariate Linear Regression Analyses for Uncorrected Visual Acuity in Small-Incision Cataract Surgery

Bivariate Regression Multivariate Regression

Variable Coefficient (95% Confidence Interval) P Value Coefficient (95% Confidence Interval) P Value
Surgeon characteristics
Annual case volume (100 cases/year) e0.0010 (e0.0029 to 0.0010) 0.33 d d
Experience (yrs) e0.0040 (e0.0081 to e0.0001) 0.049 e0.0032 (e0.0088 to 0.0024) 0.27
Prior case load (100 cases) e0.0002 (e0.0004 to e0.0001) 0.037 e0.0001 (e0.0004 to 0.0002) 0.55
Patient characteristics
Age (yrs) 0.0039 (0.0036e0.0041) <0.001 0.0038 (0.0035e0.0041) <0.001
Gender* 0.0026 (e0.0031 to 0.0083) 0.38 d d

d ¼ covariates did not achieve P < 0.20 on bivariate analysis, which was necessary for inclusion in multivariate analysis.
Linear regression analyses with random-effects modeling to account for surgeon case volume. The dependent variable is uncorrected visual acuity in log-
arithm of the minimum angle of resolution scale.
Data include both small-incision cataract surgery and phacoemulsification cataract extraction cases and reflect only cases performed as an attending-level
physician. These data do not include surgeries performed as a resident, fellow, or other trainee.
*Gender was coded as 1 for female and 2 for male. Therefore, the negative coefficient values in these regression analyses indicate improved uncorrected
visual acuity outcomes among male patients.

surgeons only broadly into high-volume and low-volume
groups, rather than assessing them on a continuous scale, as
was done in our analyses.20 Additionally, the previous study
included a smaller range of annual case volumes (48e1161
cases/year in their study vs. 76e2900 cases/year in ours).20
These design differences could explain why we discerned a
statistically significant difference in PECE outcomes and
they did not. It is interesting to note that our findings
demonstrate an asymptotic distribution wherein UCVA
outcomes among surgeons with more than 350 PECEs/year
remained largely constant, averaging 0.14 logMAR
(Snellen equivalent: w 25/20). This suggests that whatever
benefit a surgeon gains from increased surgical volume is
essentially maximized at approximately 350 PECEs/year,
and any additional increase in case volume no longer
provides an added benefit to postoperative UCVA.
We also sought to account for the potentially con-
founding factor of surgeon experience, both in terms of
surgeons’ years in practice and estimated number of prior
surgeries. (See Methods for details on calculating these
values.) When considering the difference in outcomes be-
tween high- and low-volume surgeons, it is worth noting the
differences in experience between the 2 groups. Surgeons
with fewer than 350 PECEs/year were relatively early in
their practice, with a median of 3.7 years since residency
(interquartile range, 2.1e5.7 years; range, 1.7e11.4 years).
Surgeons with 350 or more PECE/year had been in practice
longer (median, 7.7 years since completing residency;

Table 3. Bivariate and Multivariate Linear Regression Analyses for Uncorrected Visual Acuity in Phacoemulsification Cataract Extraction

Bivariate Regression Multivariate Regression

Variable
Lower-volume PECE surgeons (300 PECE/yr)
Surgeon characteristics
Annual case volume (100 PECE/yr)
Experience (yrs)
Prior case load (100 PECEs)
Patient characteristics
Age (yrs)
Gender*
Higher-volume PECE surgeons (>300 PECE/yr)
Surgeon characteristics
Annual case volume (100 PECE/yr)
Experience (yrs)
Prior case load (100 PECEs)
Patient characteristics
Age (yrs)
Gender*
Coefficient (95% Confidence Interval) P Value Coefficient (95% Confidence Interval) P Value
e0.0322 (e0.0515 to e0.0128) <0.001 e0.0238 (e0.0447 to e0.0029) 0.02
e0.0100 (e0.0167 to e0.0034) 0.003 e0.0033 (e0.0093 to 0.0026) 0.27
e0.0013 (e0.0035 to 0.0009) 0.24 d d
0.0036 (0.0025e0.0046) <0.001 0.0038 (0.0027e0.0048) <0.001
e0.0206 (e0.0378 to e0.0033) 0.02 e0.0279 (e0.0450 to e0.0107) 0.001
e0.0001 (e0.0007 to 0.0005) 0.80 d d
e0.0002 (e0.0007 to 0.0004) 0.58 d d
0.0000 (e0.0001 to 0.0001) 0.48 d d
0.0026 (0.0022e0.0029) <0.001 0.0027 (0.0023e0.0030) <0.001
e0.0066 (e0.0129 to e0.0003) 0.04 e0.135 (e0.0201 to e0.0069) <0.001

PECE ¼ phacoemulsification cataract extraction; d ¼ covariates did not achieve P < 0.20 on bivariate analysis, which was necessary for inclusion in
multivariate analysis.
Case volume included only PECE cases; small-incision cataract surgery cases were not included. Data represent linear regression analyses with random-effects
modeling to account for surgeon case volume. The dependent variable is uncorrected visual acuity in logarithm of the minimum angle of resolution scale.
Data include only include PECE cases performed as an attending-level physician. These data do not include surgeries performed as a resident, fellow, or other
trainee.
*Gender was coded as 1 for female and 2 for male. Therefore, the negative coefficient values in these regression analyses indicates improved uncorrected
visual acuity outcomes among male patients.

6
 Cox et al 
Cataract Outcomes by Surgeon Case Volume

Figure 1. Fitted plots showing uncorrected visual acuity (UCVA) out-

comes versus surgeon case volume in phacoemulsification cataract extrac-
tion (PECE) cases, where annual case volume is based on the number of
PECE cases performed by the surgeon that year. The red line is a locally
weighted scatterplot smoothing (LOWESS) curve. The thin vertical line
indicates the inflection point (350 PECE/year). Both black dashed lines
represent the lines of best fit above and below the inflection point.
Uncorrected visual acuity measurements are in logarithm of the minimum
angle of resolution (logMAR) scale. The x-axis minimum is 50 PECE/year
because this was the minimal case-volume needed to meet study inclusion
criteria.

interquartile range, 3.9e11.7 years; range, 0.3e27.7 years),

and 1 of the 39 high-volume PECE surgeons was in his or
her first year of practice. Importantly, surgeon experience,
both in terms of years in practice and prior case load, was
accounted for in all multivariate regression analyses and did
not have a significant association with outcomes. This lack
of association may be surprising but further supports the
notion that the difference in outcomes between low- and
high-volume surgeons was driven by surgical volume and
not differences in experience. Although the 2 measures of
surgeon experience may sound similar, collinearity did not
account for their lack of statistical significance on multi-
variate analysis given that tolerance, as calculated by 1 /
uncentered variance inflation factor, was more than 0.1 for
all covariates. We hypothesize that surgeon experience may
demonstrate a significant effect on outcomes in high-
complexity cases, which if true, would not have been
appreciated in this study because our data were limited to
low-complexity cases in an effort to improve standardiza-
tion and comparability across groups.21
It is unclear why annual case volume showed a statisti-
cally significant association with postoperative visual acuity
in PECE but not in SICS. This may indicate that SICS
proficiency is achieved and maintained more readily than in
PECE, as suggested by prior research demonstrating that
trainees have lower complication rates in SICS than PECE6;
thus, the experience gained from higher case volumes
simply may not provide as meaningful a benefit in
low-complexity cases like those analyzed in this study.
Considering that higher-volume surgeons showed lower
complication rates in SICS, one might expect that improved
Figure 2. Graphs showing complication rates according to annual surgeon
case volume. A, Complication rate in small-incision cataract surgery
(SICS) cases, where annual case volume is based on the number of SICS
and phacoemulsification cataract extraction (PECE) cases performed by the
surgeon that year. B, Complication rate in PECE cases, where annual case
volume is based on the number of PECE cases performed by the surgeon
that year. The dashed red lines represent 95% confidence intervals. Hori-
zontal black line represent the average complication rate. The x-axis
minimum is set to 50 cases per year because surgeons with fewer than 50
cases per year were excluded from analysis.
visual acuity outcomes would have been seen. However,
with an average SICS complication rate of only 1.7%, it is
likely that this effect alone was not large enough to cause a
statistically significant association between surgical volume
and visual acuity outcomes.
Several studies have assessed cataract surgery compli-
cation rates by annual case volume and report a significant
decrease in complication rates as surgical volume increases,
which is consistent with our findings.12,20,22 Using multi-
variate linear regression with random effects modeling, our
cohort demonstrated an independent association between
complication rates and surgical volume after adjusting for
patient age, patient gender, and surgeon experience (both in
terms of years in practice and prior case load); this was true
for both SICS and PECE. Greater surgeon experience also
was significantly associated with lower complication rates in
PECE but not in SICS. The association with annual case

7
 Ophthalmology Volume -, Number -, Month 2019
volume may indicate that higher-volume surgeons possess
particularly fine-tuned surgical skills that further reduce the
likelihood of complications. This association also may be
the result of more experienced surgeons having seen first-
hand the various pitfalls of cataract surgery, making them
better able to avoid them.
As has been reported previously, increasing age was
associated with poorer visual acuity outcomes and higher
complication rates.23e25 This is likely because of several
factors. Age-associated neurologic changes, such as dimin-
ished retinal nerve fiber thickness, may lead to diminished
visual acuity relative to younger patients even after cataracts
are removed. Older individuals often show increased nuclear
cataract density and decreased corneal endothelium density,
which may lead to more difficult and potentially traumatic
cataract extraction, as well as increased corneal edema in
PECE cases. It is worth noting that although age was not
independently associated with increased complication rates
in PECE, the average complication rate was nonetheless
higher among older patients. Older patients are also more
likely to have coexisting ocular pathologic features, such as
macular degeneration; however, our study sought to account
for this by excluding patients with known comorbid ocular
disease.
Current literature indicates varied results regarding the
association between patient gender and acuity outcomes after
cataract surgery in LMICs. When assessing PECE, we found
that there was a significant association between being male
and improved UCVA. This association aligns with several
studies, including several in India.26e29 It has been postu-
lated that this gender-based discrepancy in outcomes may
stem from diminished access to care in India’s male-oriented
society, leading to more mature cataracts at the time of
surgery, potentially undiagnosed ocular comorbidities, and
increased likelihood that a male patient’s cataract surgery
will be performed by a more senior, experienced oph-
thalmolgist. Conversely, our SICS data indicated no signifi-
cant association between outcomes and gender, which is also
consistent with several studies in similar contexts.30,31 In
fact, some studies in India have found female gender to be
significantly associated with improved postoperative acuity,
attributing this to potentially improved postoperative self-
care among women or lower rates of ocular comorbid-
ities.32,33 These differing and, at times, contradictory out-
comes clearly indicate that the interplay between gender and
visual acuity after cataract surgery is complex and multi-
faceted. Our study was not designed to fully characterize this
association, and further research is warranted.
It is important to consider the clinical context of our
findings. In developed nations, outcomes of cataract surgery
generally are quite good, and a substantial proportion of
cataract surgeons perform fewer than 350 cataract extrac-
tions annually.22,34,35 Yet even in this setting, some studies
have documented slightly lower complication rates in the
highest-volume surgeons.20,36 In LMICs, the backlog of
cataract cases requires high volumes of surgery, and our
research supports focusing on making sure that surgeons
perform a large number of surgeries annually to minimize
complications and maximize outcomes. At first glance, our
findings also may seem to suggest that outcomes would
8
improve with increased use of PECE, but as discussed
earlier, we suspect that the lower UCVA seen in SICS was
likely a short-term difference resulting from subclinical
cystoid macular edema.16,17
Small-incision cataract surgery (SICS) is also associated
with slightly higher rates of postoperative astigmatism than
PECE, which would lead to long-lasting differences in vi-
sual acuity.18,19 However, studies demonstrating a differ-
ence in rates of postoperative astigmatism between the 2
techniques have not shown a significant difference in visual
acuity outcomes, which suggests that the effect is small.19
Thus, it is sensible that a large proportion of cases are
performed using the SICS technique in LMICs given that
it is faster, more affordable, and offers comparable long-
term visual acuity outcomes.10
Because of the retrospective nature of this research, our
findings are subject to potential selection and population
biases. For example, it is AECS’s policy to have higher-
complexity cases performed by more experienced sur-
geons; this would bias our results toward worse outcomes
for these surgeons. Although we attempted to adjust for this
policy by including only level 1 (lowest complexity) cases,
not all level 1 cases are the same, and it is possible that
higher-volume surgeons performed a higher proportion of
the relatively complex level 1 cases. If true, then we may
have underestimated the true association between increasing
surgeon volume and improvement in postoperative visual
acuity. Other differences between surgeons presumably
would be random and not concentrated within any given
case-volume group, meaning they should not meaningfully
impact our findings. It is also worth noting that post-
operative UCVA is affected by biometry, which is per-
formed by technicians rather than surgeons at AECS.
However, these technicians receive standardized training at
AECS, which should minimize inter-technician differences.
Moreover, technicians are assigned to patients at random
based on availability, meaning any inter-technician differ-
ences should be distributed randomly across varying sur-
geons’ outcomes, and thus should not affect any given
surgeon’s outcomes in a systematic way. As discussed
earlier, post-SICS UCVA is known to be affected adversely
by cystoid macular edema for the first several weeks after
surgery.16,17 Given that most of our visual acuity data were
collected within that time frame, the SICS and overall visual
acuity results in this study are likely reported as being
poorer than they truly were in the long term. It is also
worth noting that our estimate of each surgeon’s career
case volume is based on their average annual case vol-
ume from 2010 through 2015; thus, this approximation
failed to account for possible fluctuations in a surgeon’s
surgical volume before 2010. Our findings are limited
further by the fact that only approximately 40% of cases
performed in the study period were included in our an-
alyses. This was an unfortunate consequence of our
relatively strict inclusion criteria and may limit the
generalizability of our findings to more complex cases or
those performed by less experienced ophthalmologists.
However, these criteria were necessary to achieve
reasonable comparability across cases, and our sample
size remained robust at 35 880 cases.
 Cox et al 
Cataract Outcomes by Surgeon Case Volume
In conclusion, increased surgical volume is significantly
associated with improved postoperative acuity outcomes for
patients undergoing PECE. This effect is most appreciable
among surgeons with fewer than 350 PECE per year. Surgeons
with volumes higher than this have reached an apparent
asymptote with UCVA outcomes of approximately 0.14 log-
MAR (Snellen equivalent: w 25/20). Higher-volume surgeons
also show significantly lower complication rates. Our findings
support the benefit of high-output ophthalmology clinics in
LMICs, wherein individual surgeons can perform a high
number of cataract extractions annually. Younger patient age
was also consistently associated with better visual acuity out-
comes and lower complication rates, whereas greater surgeon
experience in terms of number of prior cases was significant for
lower PECE complication rates but not for improved visual
acuity outcomes. These findings may inform the design and
workflow of ophthalmology clinics further, particularly in re-
gions with a high burden of cataract-related visual impairment
and high patient-to-surgeon ratios.
References
1. Bourne RR, Stevens GA, White RA, et al. Causes of vision
loss worldwide, 1990e2010: a systematic analysis. Lancet
Glob Health. 2013;1(6):339.
2. World Health Organization. Programme for the Prevention of
Deafness. Global initiative for the elimination of avoidable
blindness. Available at: http://www.who.int/iris/handle/10665/
63748; 2000. Accessed January 2, 2018.
3. Le HG, Ehrlich JR, Venkatesh R, et al. A sustainable model for
delivering high-quality, efficient cataract surgery in southern
India. Health Aff (Millwood). 2016;35(10):1783e1790.
4. Mehta PK, Shenoy S. Infinite Vision: How Aravind Became
the World’s Greatest Business Case for Compassion. San
Francisco: Berrett-Koehler Publishers; 2011.
5. Natchiar G, Thulasiraj R, Sundaram RM. Cataract surgery at
Aravind eye hospitals: 1988e2008. Community Eye Health.
2008;21(67):40e42.
6. Haripriya A, Chang DF, Reena M, Shekhar M. Complication
rates of phacoemulsification and manual small-incision cata-
ract surgery at Aravind Eye Hospital. J Cataract Refract Surg.
2012;38(8):1360e1369.
7. Tabin G, Chen M, Espandar L. Cataract surgery for the
developing world. Curr Opin Ophthalmol. 2008;19(1):55e59.
8. Henderson BA, SpringerLink. Manual Small Incision Cataract
Surgery. 2016th ed. Cham: Springer International Publishing;
2016:188.
9. Gills JP, Sanders DR. Small-Incision Cataract Surgery:
Foldable Lenses, One-Stitch Surgery, Sutureless Surgery,
Astigmatic Keratotomy. Thorofare, NJ: Slack, Inc.; 1990:271.
10. Riaz Y, de Silva SR, Evans JR. Manual small incision cataract
surgery (MSICS) with posterior chamber intraocular lens versus
phacoemulsification with posterior chamber intraocular lens for
age-related cataract. Cochrane Database of Systematic Reviews.
Available at: https://www.cochranelibrary.com/cdsr/doi/10.1002/
14651858.CD008813.pub2/full; 2013(10). Accessed 25.07.17.
11. Ruit S, Tabin G, Chang D, et al. A prospective randomized
clinical trial of phacoemulsification vs manual sutureless
small-incision extracapsular cataract surgery in Nepal. Am J
Ophthalmol. 2007;143(1):32e38.
12. Bell CM, Hatch WV, Cernat G, Urbach DR. Surgeon volumes
and selected patient outcomes in cataract surgery: a population-
based analysis. Ophthalmology. 2007;114(3):405e410.
13. Fang YT, Chien LN, Ng YY, et al. Association of hospital and
surgeon operation volume with the incidence of postoperative
endophthalmitis: Taiwan experience. Eye (Lond). 2006;20(8):
900e907.
14. Chang JW, Kim JH, Kim SJ, Yu YS. Congenital aniridia:
Long-term clinical course, visual outcome, and prognostic
factors. Korean J Ophthalmol. 2014;28(6):479e485.
15. Schulze-Bonsel K, Feltgen N, Burau H, et al. Visual acuities
“hand motion” and “counting fingers” can be quantified with
the Freiburg visual acuity test. Invest Ophthalmol Vis Sci.
2006;47(3):1236e1240.
16. Chaudhary C, Bahadhur H, Gupta N. Study of cystoid macular
edema by optical coherent tomography following uneventful
cataract surgery. Int Ophthalmol. 2015;35(5):685e691.
17. Miyake K, Masuda K, Shirato S, et al. Comparison of diclo-
fenac and fluorometholone in preventing cystoid macular
edema after small incision cataract surgery: a multicentered
prospective trial. Jpn J Ophthalmol. 2000;44(1):58e67.
18. Gogate P, Ambardekar P, Kulkarni S, et al. Comparison of
endothelial cell loss after cataract surgery: phacoemulsification
versus manual small-incision cataract surgery: six-week results of
a randomized control trial. J Cataract Refract Surg. 2010;36(2):
247e253.
19. George R, Rupauliha P, Sripriya AV, et al. Comparison of
endothelial cell loss and surgically induced astigmatism
following conventional extracapsular cataract surgery, manual
small-incision surgery and phacoemulsification. Ophthalmic
Epidemiol. 2005;12(5):293e297.
20. Habib M, Mandal K, Bunce CV, Fraser SG. The relation of
volume with outcome in phacoemulsification surgery. Br J
Ophthalmol. 2004;88(5):643e646.
21. Schallhorn JM, Ciralsky JB, Yeu E. Resident and young
physician experience with complex cataract surgery and new
cataract and refractive technology: results of the ASCRS 2016
Young Eye Surgeons Survey. J Cataract Refract Surg.
2017;43(5):687e694.
22. Sparrow JM, Taylor H, Qureshi K, et al. The Cataract National
Dataset electronic multi-centre audit of 55,567 operations: risk
indicators for monocular visual acuity outcomes. Eye.
2012;26(6):821e826.
23. Xiaochun L, Xiaoguang C, Xianru H, Yongzhen B. The corre-
lation of age and postoperative visual acuity for age-related
cataract. BioMed Research International. 2016;Arte D:7147543.
24. McKee M, Whatling JM, Wilson JL, Vallance-Owen A.
Comparing outcomes of cataract surgery: challenges and op-
portunities. J Public Health. 2005;27(4):348e352.
25. Panchapakesan K, Rochtchina E, Mitchell P. Five-year change
in visual acuity following cataract surgery in an older commu-
nity: the Blue Mountains Eye Study. Eye. 2004;18(3):278e282.
26. Matta S, Park J, Palamaner Subash Shantha G, et al. Cataract
surgery visual outcomes and associated risk factors in sec-
ondary level eye care centers of L V Prasad Eye Institute,
India. PLoS One. 2016;11(1):e0144853.
27. Yan X, Guan C, Mueller A, et al. Outcomes and projected
impact on vision restoration of the China million cataract sur-
geries program. Ophthalmic Epidemiol. 2013;20(5):294e300.
28. Hashemi H, Mohammadi SF, Z-Mehrjardi H, et al. The role of
demographic characteristics in the outcomes of cataract surgery
and gender roles in the uptake of postoperative eye care: a
hospital-based study. Ophthalmic Epidemiol. 2012;19(4):
242e248.
9
 Ophthalmology Volume -, Number -, Month 2019
29. Venkatesh R, Muralikrishnan R, Balent LC, et al. Outcomes of
high volume cataract surgeries in a developing country. Br J
Ophthalmol. 2005;89(9):1079e1083.
30. Murthy GV, Vashist P, John N, et al. Prevalence and vision-
related outcomes of cataract surgery in Gujarat, India.
Ophthalmic Epidemiol. 2009;16(6):400e409.
31. Lam DS, Congdon NG, Rao SK, et al. Visual outcomes and
astigmatism after sutureless, manual cataract extraction in rural
China: Study of Cataract Outcomes and Up-Take of Services
(SCOUTS) in the caring is hip project, report 1. Arch Oph-
thalmol. 2007;125(11):1539e1544.
32. Gogate P, Vakil V, Khandekar R, et al. Monitoring and
modernization to improve visual outcomes of cataract surgery
in a community eyecare center in western India. J Cataract
Refract Surg. 2011;37(2):328e334.
Footnotes and Financial Disclosures
Originally received: June 1, 2018.
Final revision: March 13, 2019.
Accepted: March 22, 2019.
Available online: ---. Manuscript no. 2018-1125.
1
Dana Center for Preventive Ophthalmology, Wilmer Eye Institute, Johns
Hopkins Hospital, Baltimore, Maryland.
2
Aravind Eye Hospital, Madurai, Tamil Nadu, India.
Financial Disclosure(s):
The author(s) have no proprietary or commercial interest in any materials
discussed in this article.
HUMAN SUBJECTS: Human subjects were included in this study. The
human ethics committees at Aravind Eye Care System approved the study.
All research adhered to the tenets of the Declaration of Helsinki. As a
retrospective dataset review that did not involve any identifying informa-
tion, we did not receive informed consent for this project from patients
whose clinical data were used.
No animal subjects were included in this study.
Author Contributions:
Conception and design: Cox, Subburaman, Friedman, Ravindran
10
33. Khandekar RB, Jain BK, Sudhan AK, Pandey KP. Visual
acuity at 6 weeks after small incision cataract surgery and role
of audit in predicting visual acuity. Eur J Ophthalmol.
2010;20(2):345e352.
34. Lundstrom M, Barry P, Henry Y, et al. Visual outcome of
cataract surgery; study from the European Registry of Quality
Outcomes for Cataract and Refractive Surgery. J Cataract
Refract Surg. 2013;39(5):673e679.
35. Norregaard JC. Results from the International Cataract
Surgery Outcomes Study. Acta Ophthalmol. 2007;85(2):
5e32.
36. Johnston RL, Taylor H, Smith R, Sparrow JM. The Cataract
National Dataset electronic multi-centre audit of 55,567 op-
erations: variation in posterior capsule rupture rates between
surgeons. Eye (Lond). 2010;24(5):888e893.
Analysis and interpretation: Cox, Munoz, Friedman, Ravindran
Data collection: Subburaman, Ravindran
Obtained funding: Study was performed as part of regular employment
duties at the Dana Center for Preventive Ophthalmology at the Wilmer Eye
Institute of Johns Hopkins Hospital and the Aravind Eye Hospital. No
additional funding was provided.
Overall responsibility: Cox, Friedman, Ravindran
Abbreviations and Acronyms:
AECS ¼ Aravind Eye Care System; BCVA ¼ best-corrected visual acuity;
DM ¼ Descemet membrane; ETDRS ¼ Early Treatment Diabetic Reti-
nopathy Study; IOL ¼ intraocular lens; LMIC ¼ low- and middle-income
country; logMAR ¼ logarithm of the minimum angle of resolution;
PECE ¼ phacoemulsification cataract extraction; SD ¼ standard deviation;
SICS ¼ small-incision cataract surgery; UCVA ¼ uncorrected visual
acuity.
Correspondence:
Ravilla D. Ravindran, MS, DO, Aravind Eye Hospital, 1, Anna Nagar,
Madurai, Tamil Nadu 625020, India. E-mail: rdr@aravind.org.