Catherine L. Chen, MD, MPH
Associate Professor
Department of Anesthesia and Perioperative Care
Philip R. Lee Institute for Health Policy Studies
University of California, San Francisco
Contemporary Sedation
Practices for Cataract Surgery
in the United States
2025 MPOG Retreat
October 10, 2025
Disclosures
▪ Funded by the NIH (K23 AG072035 PI: Chen) “Frailty and
Monitored Anesthesia Care for Cataract Surgery in Older
Adults,” the Patricia Sander Award and UCSF Anesthesia
Research
▪ No conflicts of interest
2
The big picture
In 2021, 56% of
anesthesiologists
were age 55 or older
All baby boomers
will be over age 65
by 2030
US Census Bureau February 2020; National Eye Institute. Eye Health Data and Statistics; American Medical Association Physician Masterfile (Dec. 31, 2021)
38.7 million
cataract patients
by 2030
3
JAMA Intern Med. 2022;182(11):1171-1180. doi:10.1001/jamainternmed.2022.4333
Prevalence of anesthesia care for
selected low-risk procedures in
the 2017 Medicare 5% sample
4
5
Evidence from the Multicenter
Perioperative Outcomes Group
(MPOG) Registry
Methods
Retrospective observational cohort study
• Adult patients undergoing cataract surgery from 2018-2021
• Exclusion criteria: <18 years of age, emergency surgery, cataract surgery combined
with another procedure; institutions with low surgical volume or who did not report
data in all study years
Study cohort
• Anesthetic approach
Exposure
• Primary: Perioperative events requiring intervention by an anesthesia professional
• Sensitivity analyses: Propensity adjusted model, Difference in initial vs. final
sedation cohort assignment
Outcomes
What anesthetic approaches are being used for cataract
surgery patients in the US?
7
Everyone got
“Monitored Anesthesia
Care” (MAC), but 80%
of patients received
nothing or just IV
fentanyl and midazolam
during surgery
No IV sedative
medications
8%
GA
2%
Other IV sedative
medications
18%
Fentanyl/Midazolam only
72%
Total N=123,009
Proportion of patients receiving each anesthetic approach stratified
by institution
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Proportion of Cases
Anonymized Institution ID
General Anesthesia
Other Sedation
No Sedation
Fentanyl/Midazolam
Institutional intraoperative IV medication patterns for MAC cases only
’
6 (N=6,538) 11 (N=1,196) 1 (N=8,643) 3 (N=19,615) 8 (N=11,957) 5 (N=5,616) 9 (N=4,251) 2 (N=5,814)
10 (N=7,252) 4 (N=1,535) 7 (N=8,216) 14 (N=4,696) 13 (N=9,125) 12 (N=3,809)
17 (N=706) 16 (N=6,409) 15 (N=3,459)
19 (N=2,974) 18 (N=1,286)
20 (N=4,430)
4 (N=2,701)
Cluster 1
Cluster 2
Cluster 3
Cluster 4
Cluster 5
Cluster 6
Perioperative events requiring intervention by an anesthesia
professional
Event
Hypertension
Hypotension
Tachycardia
-blockers
Bradycardia
Hyperglycemia
Hypoglycemia
Opioid or Benzodiazepine Overdose
Bronchospasm
Hypoxemia
Proportion of patients within each group experiencing an event requiring
intervention by an anesthesia professional
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
GA
(n=2,744)
Other Sedation
(n=22,182)
Fentanyl/Midazolam
(n=88,073)
No Sedation
(n=10,010)
Proportion of patients
Anesthetic Approach
Hypotension
Bradycardia
Tachycardia
Bronchospasm
Hypertension
Hypoxemia
Hyperglycemia
Overdose
Hypoglycemia
8,149 (6.6%) of
patients had at
least one event
Association between anesthetic approach and events requiring intervention
by an anesthesia professional
Sedation Approach
Unadjusted OR (95% CI) Adjusted OR (95% CI)*
No Sedation
ref. ref.
Fentanyl/Midazolam
1.00 (0.70, 1.42) 1.03 (0.74, 1.44)
Other Sedation
1.81 (1.18, 2.78) 1.96 (1.34, 2.89)
GA
45.38 (30.30, 67.96) 51.89 (34.59, 77.86)
*Adjusted for age, gender, race, ASA score, selected comorbidities
Sensitivity Analyses
Sensitivity Analysis #1: Account for Treatment Assignment
Association between anesthetic approach and events adjusted for the
propensity to receive that anesthetic approach in the first place
Original Model 4-Way Propensity Adjusted Model
Sedation Approach
Unadjusted OR
(95% CI)
Adjusted OR
(95% CI)*
IPTW-adjusted OR
(95% CI)
IPTW-adjusted OR
(95% CI)*
No Sedation
ref. ref. ref. ref.
Fentanyl/Midazolam
1.00 (0.70, 1.42) 1.03 (0.74, 1.44) 1.07 (0.75, 1.52) 1.07 (0.75, 1.51)
Other Sedation
1.81 (1.18, 2.78) 1.96 (1.34, 2.89) 1.99 (1.31, 3.02) 2.00 (1.35, 2.95)
GA
45.38 (30.30, 67.96)
51.89 (34.59, 77.86)
46.22 (29.68, 71.97)
49.74 (32.21, 76.79)
*Adjusted for age, gender, race, ASA score, and selected Elixhauser comorbidities
Sensitivity Analyses #2: Account for Misclassification Bias
Add 251 additional patients to the “No Sedation” group and 1,683
additional patients to the “Fentanyl/Midazolam” group
To account for possibility that some patients in the “Other Sedation” or
“GA” groups could have started in “No Sedation” or “Fentanyl/Midazolam”
1. Within “Other Sedation” and “GA” groups,
review timing of medications and airway
management relative to surgery start time
2. Flag anyone who began receiving medications
requiring anesthesia training or placement of an
advanced airway only AFTER surgery start time
Sensitivity Analysis #2: Account for Misclassification bias
Association between anesthetic approach and events after re-categorizing “Other Sedation” or
“GA” patients whose initial anesthetic approach was ”No Sedation” or “Fentanyl/Midazolam”
Original Model
Model Adjusted for
“Conversion” Mid-
Procedure
Sedation Approach
Unadjusted OR
(95% CI)
Adjusted OR
(95% CI)*
Adjusted OR (95% CI)*
No Sedation
ref. ref. ref.
Fentanyl/Midazolam
1.00 (0.70, 1.42) 1.03 (0.74, 1.44) 0.97 (0.70, 1.33)
Other Sedation
1.81 (1.18, 2.78) 1.96 (1.34, 2.89) 1.79 (1.23, 2.60)
GA
45.38 (30.30, 67.96) 51.89 (34.59, 77.86) 47.57 (31.91, 70.91)
*Adjusted for age, gender, race, ASA score, and selected Elixhauser comorbidities
17
Adjusted OR
(95% CI)
Age 0.97 (0.97, 0.98)
Gender
Male ref.
Female 0.96 (0.91, 1.02)
Race/Ethnicity
White ref.
Asian/Pacific Islander 0.90 (0.63, 1.29)
Black 1.79 (1.15, 2.78)
Hispanic 1.30 (0.79, 2.13)
Other 0.44 (0.23, 0.86)
Unknown race 1.18 (0.68, 2.04)
ASA Class
1 ref.
2 1.19 (0.92, 1.54)
3 1.44 (1.04, 1.99)
4 1.66 (0.91, 3.02)
Comorbidities
*
Hypertension 0.67 (0.53, 0.84)
Diabetes 0.98 (0.75, 1.28)
Congestive heart failure 1.23 (0.84, 1.81)
Cardiac arrhythmia 0.74 (0.59, 0.94)
Chronic Pulmonary Disease 0.89 (0.68, 1.17)
Renal Failure 0.98 (0.72, 1.32)
Obesity 0.87 (0.63, 1.21)
Substance abuse 1.39 (0.71, 2.70)
*Reference group for each comorbidity is patients who did not have the comorbidity.
Factors associated with conversion from “No Sedation” or “Fentanyl/Midazolam”
groups to “Other Sedation” or “GA” groups after surgery start time
Take home points
There is wide variation
in sedation approaches
for cataract surgery.
We have an opportunity
to consider whether
some cases could be
done without routine
anesthesia care.
More research is
needed to determine
which patients would be
safe to proceed without
anesthesia care.
Acknowledgments
UCSF Department of Anesthesia & Perioperative
Care: Michael Gropper, Judith Hellman, Adrian Gelb,
Mervyn Maze, Jackie Leung, Nora Lyang, Rachel
Schwartz, Jacob Flores, Julie Leong and many others
Philip R. Lee Institute for Health Policy Studies at
UCSF: Claire Brindis, Joanne Spetz, Dan Dohan,
Yanting Luo, Beth Thew
UCSF Pepper Center: Ken Covinsky, Louise Walter, Sei
Lee, John Boscardin, Siqi Gan
University of Minnesota Dept. of Medicine: R. Adams
Dudley
MPOG: Sachin Kheterpal, Mike Matthis, Doug
Colquhoun, Robert Coleman and the rest of the PCRC
Kaiser SF Dept. of Anesthesiology: Andrea Olmos
My contact info:
Email:
catherine.chen@ucsf.edu
19
21
Monitored Anesthesia Care
All
(N=123,009)
General
Anesthesia
(N=2,744)
No Sedation
(N=10,010)
Fentanyl/ Midazolam
(N=88,073)
Other Sedation
(N=22,182)
Age 69.5 (10.2) 61.4 (15.9) 71.5 (9.7) 69.6 (9.7) 69.3 (10.8)
Gender
Male 52,075 (42.3%) 1,279 (46.6%) 4,695 (46.9%) 36,783 (41.8%) 9,318 (42.0%)
Female 70,934 (57.7%) 1,465 (53.4%) 5,315 (53.1%) 51,290 (58.2%) 12,864 (58.0%)
Race/Ethnicity
White 84,595 (68.8%) 1,904 (69.4%) 7,103 (71%) 59,737 (67.8%) 15,851 (71.5%)
Asian or Pacific Islander 8,429 (6.9%) 106 (3.9%) 928 (9.3%) 6,165 (7.0%) 1,230 (5.6%)
Black 13,475 (11.0%) 393 (14.3%) 1,027 (10.3%) 9,568 (10.9%) 2,487 (11.2%)
Hispanic 1,948 (1.6%) 47 (1.7%) 135 (1.4%) 1,241 (1.4%) 525 (2.4%)
Other 1,731 (1.4%) 36 (1.3%) 66 (0.7%) 1551 (1.8%) 78 (0.4%)
Unknown race 12,831 (10.4%) 258 (9.4%) 751 (7.5%) 9,811 (11.1%) 2,011 (9.1%)
ASA Class
1 4,110 (3.3%) 76 (2.8%) 318 (3.2%) 2,929 (3.3%) 787 (3.6%)
2 58,215 (47.3%) 939 (34.2%) 4,333 (43.3%) 41,732 (47.4%) 11,211 (50.5%)
3 57,670 (46.9%) 1,602 (58.4%) 4,913 (49.1%) 41,441 (47.1%) 9,714 (43.8%)
4 3,014 (2.5%) 127 (4.6%) 446 (4.5%) 1,971 (2.2%) 470 (2.1%)
Comorbidities*
Hypertension 45,672 (37.1%) 864 (31.5%) 3,185 (31.8%) 35,583 (40.4%) 6,040 (27.2%)
Diabetes 18,577 (15.1%) 429 (15.63%) 1,176 (11.8%) 14,461 (16.4%) 2,511 (11.3%)
Congestive heart failure 5,144 (4.2%) 112 (4.1%) 475 (4.8%) 3,729 (4.2%) 828 (3.7%)
Cardiac arrhythmia 9,725 (7.9%) 189 (6.9%) 808 (8.1%) 7,358 (8.4%) 1,370 (6.2%)
Chronic Pulmonary Disease 12,224 (9.9%) 307 (11.2%) 849 (8.5%) 9,312 (10.6%) 1,756 (7.9%)
Renal Failure 8,319 (6.8%) 199 (7.3%) 626 (6.3%) 6,186 (7.0%) 1,308 (5.9%)
Obesity 39,389 (32.0%) 1,010 (36.8%) 3,025 (30.2%) 28,329 (32.2%) 7,025 (31.7%)
Substance abuse 407 (0.3%) 18 (0.7%) 21 (0.2%) 294 (0.3%) 74 (0.33%)
*Less than one percent of patients missing data in each comorbidity category reported
Baseline Characteristics of Patients Undergoing Cataract Surgery
Proportion of patients within each group experiencing an event requiring
intervention by an anesthesia professional
0 %
10 %
20 %
30 %
40 %
50 %
60 %
70 %
80 %
90 %
100 %
GA Other Sedation Fentanyl/Midazolam No Sedation
Proportion of patients
Anesthetic Approach
Association between anesthetic approach and events requiring intervention
by an anesthesia professional (full model results)
Unadjusted OR (95% CI) Adjusted OR (95% CI)
Sedation
None ref. ref.
GA 45.38 (30.30, 67.96) 51.89 (34.59, 77.86)
MAC 1.81 (1.18, 2.78) 1.96 (1.34, 2.89)
Nurse sedation 1.00 (0.70, 1.42) 1.03 (0.74, 1.44)
Age 0.99 (0.98, 1.00) 1.01 (1.00, 1.02)
Gender
Male ref. ref.
Female 0.87 (0.79, 0.96) 0.92 (0.83, 1.04)
Race/Ethnicity
White ref. ref.
Asian or Pacific Islander 1.18 (0.90, 1.56) 1.53 (1.15, 2.03)
Black 1.84 (1.47, 2.32) 1.77 (1.46, 2.14)
Hispanic 1.40 (1.03, 1.90) 1.44 (1.07, 1.94)
Other 0.91 (0.53, 1.57) 1.02 (0.68, 1.54)
Unknown 1.32 (0.94, 1.86) 1.53 (1.12, 2.09)
ASA Class
1 ref. ref.
2 1.35 (1.06, 1.73) 1.37 (1.16, 1.62)
3 2.53 (1.88, 3.40) 2.27 (1.88, 2.75)
4 3.93 (2.83, 5.46) 3.02 (2.25, 4.04)
Comorbidities*
Hypertension 1.16 (0.96, 1.40) 1.00 (0.82, 1.21)
Diabetes 1.38 (1.20, 1.59) 1.12 (0.99, 1.26)
Congestive heart failure 1.70 (1.45, 1.98) 1.17 (0.98, 1.39)
Cardiac arrhythmia 1.34 (1.07, 1.55) 1.14 (1.01, 1.29)
Chronic Pulmonary Disease 1.16 (0.97, 1.40) 0.97 (0.84, 1.12)
Renal Failure 1.68 (1.45, 1.94) 1.23 (1.14, 1.32)
Obesity 1.15 (0.97, 1.37) 0.98 (0.82, 1.17)
Substance abuse 1.98 (1.34, 2.92) 1.47 (0.84, 2.57)
List of perioperative events requiring intervention by anesthesia professionals
No. (%)
2,894 (2.4%)
2,664 (2.2%)
1,641 (1.3%)
852 (0.7%)
645 (0.5%)
73 (0.06%)
64 (0.05%)
55 (0.04%)
166 (0.1%)
8,149 (6.6%)
Median fentanyl and midazolam doses administered during cataract surgery
in Cluster 1 institutions
JAMA Intern Med. 2022;182(11):1171-1180. doi:10.1001/jamainternmed.2022.4333
26
Death and hospitalization after
selected low-risk procedures in the
2017 Medicare 5% sample
JAMA Intern Med. 2022;182(11):1171-1180. doi:10.1001/jamainternmed.2022.4333
27
Systemic complications after
selected low-risk procedures in the
2017 Medicare 5% sample