Journal of Plastic, Reconstructive & Aesthetic Surgery 85 (2023) 393–400

An enhanced recovery after surgery

]]
]]]]]]
]]

protocol for facial feminization surgery

reduces perioperative opioid usage, pain,
and hospital stay☆
Meiwand Bedar a, Dillon Dejam a, Rachel M. Caprini a,
Kelly X. Huang a, Brendan J. Cronin a, Sumun Khetpal a,
⁎
Katarina B.J. Morgan b, Justine C. Lee a,c,

a
Division of Plastic and Reconstructive Surgery, University of California Los Angeles, David Geffen School of
Medicine, Los Angeles, CA, United States
b
T.H. Chan School of Public Health, Harvard University, Boston, MA, United States
c
UCLA Gender Health Program, University of California Los Angeles, David Geffen School of Medicine, Los
Angeles, CA, United States

Received 17 February 2023; Accepted 18 July 2023

KEYWORDS Summary Background: Prescription drug misuse in transgender individuals is estimated to be

Gender-affirming three times higher than that of the general population in the United States, suggesting that
surgery; opioid-reduction strategies deserve significant consideration in gender-affirming surgeries. In
Facial feminization this work, we describe the implementation of an enhanced recovery after surgery (ERAS)
surgery; protocol to reduce opioid use after facial feminization surgery.
Enhanced recovery Methods: A total of 79 patients who underwent single-stage facial feminization surgery before
after surgery (n = 38) or after (n = 41) ERAS protocol implementation were included. Primary outcomes
protocol; assessed were perioperative opioid consumption (morphine equivalent dose/kilogram, MED/
Opioid use; kg), average patient-reported pain scores, and length of hospital stay. Comparisons between
Tramadol groups and multivariable linear regression analyses were conducted to define the contribution
of the ERAS protocol to each of the three primary outcomes.
Results: Age, body mass index, mental health diagnoses, and length of surgery did not differ
between pre-ERAS and ERAS groups. Compared to pre-ERAS patients, patients treated under
the ERAS protocol consumed less opioids (median [interquartile range, IQR], 0.8 [0.5–1.1]

☆
This work was presented at the Plastic Surgery Research Council’s 67th Annual Meeting and the 100th Annual Meeting of the American
Association of Plastic Surgeons.
⁎
Correspondence to: Division of Plastic and Reconstructive Surgery, University of California, Los Angeles, 200 UCLA Medical Plaza, Suite
460, Los Angeles, CA 90095-6960, United States.
E-mail address: [email protected] (J.C. Lee).

https://doi.org/10.1016/j.bjps.2023.07.044
1748-6815/© 2023 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. This is an open access
article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
 M. Bedar, D. Dejam, R.M. Caprini et al.

versus 1.5 [1.0–2.1] MED/kg, p < 0.001), reported lower pain scores (2.5 ± 1.8 versus
3.7 ± 1.6, p = 0.002), and required a shorter hospital stay (median [IQR], 27.3 [26.3–49.8]
versus 32.4 [24.8–39.1] h, p < 0.001). When controlling for other contributing variables such as
previous gender-affirming surgeries, mental health diagnoses, and length of surgery using
multivariable linear regression analyses, ERAS protocol implementation independently pre­
dicted reduced opioid use, lower pain scores, and shorter hospital stay after facial feminization
surgery.
Conclusions: The current work details an ERAS protocol for facial feminization surgery that
reduces perioperative opioid consumption, patient-reported pain scores, and hospital stays.
© 2023 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by
Elsevier Ltd. This is an open access article under the CC BY license (http://creative­
commons.org/licenses/by/4.0/).

A combination of population and insurance claims-based Methods

analyses has placed the prevalence of prescription drug
misuse in transgender individuals at roughly three times Patients
higher than that of the general population of the United
States.1–3 A recent secondary analysis of the US Transgender Consecutive patients undergoing FFS by a single surgeon (JCL)
Survey administered to 26,689 individuals reported that prior to ERAS protocol implementation (2019–2021) were com­
16.5% of participants indicated prescription drug misuse pared to patients treated after ERAS implementation
within 12 months of the survey,1 with regional reports that (2021–2022). Patients were excluded from the analysis if they
may range significantly higher.4,5 Comparatively, the 2020 had a history of chronic pain or received prior FFS. Patients
US National Survey on Drug Use and Health estimated that were included in the analysis if they underwent de novo, single-
5.8% and 3.3% of the national population misused pre­ stage FFS that included soft tissue reconstructions and
scription drugs and prescription opioids, respectively, over at least one of the following osseous procedures: forehead re­
the preceding 12 months.3 The differences in the pre­ contouring, mandibular recontouring, or rhinoplasty.
valence of prescription drug misuse corroborate with the Retrospective chart review was performed to collect demo­
reported differences in diagnoses of substance use disorders graphic information, medical and surgical history, and FFS
between the two populations. Using insurance claims data characteristics (University of California Los Angeles IRB protocol
from the Optum Labs Data Warehouse, a direct comparison 11–000925). Additionally, information regarding intraoperative
of cisgender and transgender adults demonstrated that pain medication, duration of surgery, postoperative pain med­
transgender patients were more than three times as likely ication usage, average pain scores, and length of hospital stay
to carry a substance use disorder diagnosis compared to were collected. Intraoperative and postoperative opioid doses
cisgender patients.2 Of all prescription drugs, opioid misuse were converted to oral morphine equivalent dose (MED) using
is particularly concerning and germane to reconstructive previously reported conversion factors (Supplementary Table 1)
surgeons performing gender-affirming procedures. Although and divided by patient weight in kilograms (kg) at the time of
the causes are multifactorial, the transgender patient po­ surgery (MED/kg).18 Similarly, total acetaminophen and ketor­
pulation carries several known risk factors for opioid use olac use were expressed in milligrams per kg (mg/kg) body
including a higher proportion of mental health diagnoses weight. Pain scores were assessed by the visual analog scale
such as anxiety and depression as well as multiple re­ and average pain scores were calculated by averaging all in­
constructive surgical encounters for transitioning.6,7 Thus, patient postoperative pain scores for each patient recorded by
efforts to decrease perioperative opioid use after gender- the nursing staff as part of routine inpatient clinical care. The
affirming surgeries are important risk-reduction measures length of hospital stay was defined as the number of hours
that require further exploration. between the start of surgery and discharge from the hospital.
One of the core tenets of enhanced recovery after sur­
gery (ERAS) protocols is opioid use reduction via pre­
operative patient education and perioperative multimodal Pre-ERAS protocol
analgesia.8–10 The intentional diversification of pain medi­
cations in ERAS protocols has been associated with better Prior to ERAS protocol implementation, intraoperative pain
perioperative pain control, a reduced length of hospital control consisted of local anesthetic injection of 0.25% bu­
stay, and a reduction in outpatient narcotic usage.10–15 pivacaine with 1:200,000 epinephrine at each incision
While ERAS protocols have emerged in breast reconstruc­ 10 min or more prior to incision at each anatomic location
tion as well as in other surgical specialties, few have sequentially as well as intraoperative opioids as needed.
been developed for gender-affirming surgeries.16,17 In this Postoperatively, patients received pro re nata analgesia
work, we describe an ERAS protocol for gender-affirming that included primarily oral oxycodone and acetaminophen
facial feminization surgery (FFS) and its effects on perio­ with low-dose intravenous hydromorphone for severe pain.
perative opioid consumption, patient-reported pain scores, Upon discharge, oral oxycodone was prescribed for use as
and length of hospital stay. needed for severe pain and patients were encouraged to use

394
 Journal of Plastic, Reconstructive & Aesthetic Surgery 85 (2023) 393–400

over-the-counter acetaminophen as needed. Nonsteroidal protocol and 38 patients (48.1%) were included in the pre-
anti-inflammatory drugs such as ketorolac and ibuprofen ERAS cohort. Age at surgery, body mass index, and medical
were not included in the regimen. comorbidities did not differ significantly between cohorts
(Table 1).
Approximately 73% of patients in each cohort had re­
FFS ERAS protocol
ceived other gender-affirming surgeries prior to FFS, with
no differences in the types of procedures between groups.
Our FFS ERAS protocol was implemented in January 2021
All patients had previously been treated with hormone
with the primary goal of reducing postoperative opioid use.
therapy and the median (interquartile range, IQR) length of
Preoperatively, patients were counseled and educated re­
hormone therapy was 3.0 (1.6–5.0) years in both cohorts.
garding the postoperative pain management protocol.
Mental health diagnoses were present in 60 patients
Intraoperatively, patients received a multimodal analgesia
(75.9%) in the cohort, with no differences between the pre-
regimen consisting of both opioid and nonopioid analgesics
ERAS and ERAS groups. Anxiety and depression were the
including local anesthetic injection of 0.25% bupivacaine
most frequently diagnosed mental health conditions, ac­
with 1:200,000 epinephrine at each incision 10 min or more
counting for 64.6% and 63.3% of all patients, respectively.
prior to incision at each anatomic location sequentially,
8.9% of patients had a history of substance abuse with no
intravenous opioids as needed, intravenous ketorolac
differences between the pre-ERAS and ERAS groups. Of
(15–30 mg), and intravenous acetaminophen (1000 mg).
note, many patients had more than one mental health di­
Postoperatively, patients received scheduled intravenous
agnosis.
ketorolac (15 mg) and oral acetaminophen (650 mg) every
6 h, until discharge. Breakthrough pain was addressed with
oral tramadol (50 mg). In severe pain, oral oxycodone (5 mg) Surgical characteristics
or intravenous hydromorphone was administered based on
clinical judgment. At discharge, patients were prescribed Overall, 93.7% of patients received one-stage FFS with no
oral acetaminophen (650 mg) and oral ibuprofen (400 mg) difference between pre-ERAS and ERAS groups (Table 2).
every 6 h alternating until no longer needed and were given Patients in each cohort underwent an average of 7.1 ± 1.9
a 5-day supply of tramadol (50 mg) for breakthrough pain. procedures, also with no differences between the pre-ERAS
and ERAS groups. With respect to the types of surgeries,
more patients within the pre-ERAS group received hairline
Statistical analyses lowering (81.6% vs. 58.5%; p = 0.03) and upper lip lift (55.3%
vs. 29.3%; p = 0.02) compared to the ERAS group, whereas
Descriptive statistics, chi-square analyses, and independent more patients in the ERAS group received mandibular re­
samples t-tests were used to compare pre-ERAS and ERAS construction (95.1% vs. 78.9%, p = 0.03). Despite the dif­
groups. The Shapiro–Wilk test was used to examine the ferences in surgical procedures, no difference in overall
normality of continuous data. Nonnormally distributed length of surgery was found between groups with a median
continuous data were compared using Mann–Whitney U (IQR) time of 7.8 (6.7–8.9) h for the entire cohort.
tests. Multivariable linear regression analyses were con­
ducted to determine predictors of total perioperative
Comparison of perioperative analgesic medications,
opioid use, mean pain scores, and length of hospital stay.
Independent variables included for all three models were:
patient-reported pain scores, and length of hospital
implementation of the ERAS protocol, previous gender-af­ stay between Pre-ERAS and ERAS groups
firming surgeries, mental health diagnoses, and duration of
surgery. Additionally, length of hospital stay was included in The primary outcomes assessed in this work were total in­
the models for perioperative opioid use and pain scores. All patient opioid use, patient-reported pain scores, and length
of the following assumptions were met for each model: of hospital stay (Table 3).
linearity and homoscedasticity using a residual versus pre­
dicted values plot, collinearity using a variance inflation Perioperative analgesic medications
factor < 2 for all predictors for each model, independence In the total cohort, a combination of nonopioid and opioid
of residuals using Durbin–Watson criteria of 1.5–2.5, and analgesic medications were used in the perioperative, in­
normality of residuals using a visual inspection of the P-P patient period with a median (IQR) of 1.1 (0.8–1.6) MED/kg
plot. Statistical significance was defined as a p-value less of total opioids and 33.2 (22.6–47.6) mg/kg of acet­
than 0.05. Statistical analyses were performed using SPSS aminophen. The total opioid use differed between pre-
Statistics Version 27 (IBM Corp., Armonk, NY). ERAS and ERAS groups (median [IQR] 1.5 [1.0–2.1] vs. 0.8
[0.5–1.1] MED/kg, p < 0.001) in a manner that was driven
by the postoperative opioid medications administered,
with no differences in intraoperative opioid medications
Results between the groups. Postoperatively, the replacement of
oxycodone with tramadol as the first-line oral narcotic
Patient characteristics reduced MED/kg in the ERAS group significantly.
No differences between the groups were seen in the usage
A total of 96 patients were identified through retrospective of intravenous hydromorphone in the postoperative
chart review, and 79 patients met the inclusion criteria. A period. For nonopioid medications, ketorolac was one
total of 41 patients (51.9%) were treated with the ERAS of the mainstays of the ERAS protocol and, hence, was

395
 M. Bedar, D. Dejam, R.M. Caprini et al.

Table 1 Patient characteristics.

Total Cohort n = 79 Pre-ERAS ERAS p
n = 38 n = 41
Age, median (IQR) years 29.0 (25.0–38.0) 29.5 (25.0–45.3) 29.0 (25.0–37.5) NS
BMI, median (IQR) kg/m2 23.5 (20.9–27.7) 23.7 (20.8–26.0) 22.9 (20.8–30.8) NS
ASA score, n (%) NS
I 37 (46.8) 14 (36.8) 23 (56.0)
II 40 (50.6) 23 (60.5) 17 (41.5)
III 2 (2.5) 1 (2.6) 1 (2.4)
Comorbidities, n (%)
Hypertension 2 (2.5) 0 (0) 2 (4.9) NS
Diabetes Mellitus 1 (1.3) 0 (0) 1 (2.4) NS
HIV 5 (6.3) 3 (7.9) 2 (4.9) NS
Smoking history, n (%) NS
Never 56 (70.9) 28 (73.7) 28 (68.3)
Former 23 (29.1) 10 (26.3) 13 (31.7)
Other Gender-Affirming Surgeries, n (%) 58 (73.4) 28 (73.7) 30 (73.2) NS
Chest Reconstruction 27 (34.2) 17 (44.7) 10 (24.4) NS
Genital Reconstruction 24 (30.4) 13 (34.2) 11 (26.8) NS
Chondrolaryngoplasty 34 (43.0) 19 (50.0) 15 (36.6) NS
Other (body contouring, voice surgery) 5 (6.3) 4 (10.5) 1 (2.4) NS
Hormone therapy, median (IQR) years 3.0 (1.6–5.0) 3.0 (1.4–5.4) 3.0 (1.8–4.6) NS
Mental health diagnoses, n (%) 60 (75.9) 28 (73.7) 32 (78.0) NS
Anxiety 51 (64.6) 27 (71.1) 24 (58.5) NS
Depression 50 (63.3) 21 (55.3) 29 (70.7) NS
Attention deficit/hyperactivity 15 (22.8) 10 (26.3) 5 (12.2) NS
Bipolar disorder 6 (7.6) 5 (13.2) 1 (2.4) NS
Post-traumatic stress disorder 3 (3.8) 1 (2.6) 2 (4.9) NS
Personality disorder 3 (3.8) 3 (7.9) 0 (0) NS
Substance abuse 7 (8.9) 5 (13.2) 2 (4.9) NS
Eating disorder 5 (6.3) 1 (2.6) 4 (9.8) NS
ERAS, enhanced recovery after surgery; IQR, interquartile range; NS, not significant; BMI, body mass index; ASA, American Society of
Anesthesiologists Physical Status Classification; HIV, human immunodeficiency virus.

Table 2 Surgical characteristics.

Total Cohort n = 79 Pre-ERAS ERAS p
n = 38 n = 41
One-Stage FFS, n (%) 74 (93.7) 35 (92.1) 39 (95.1) NS
Number of procedures per patient (mean ± SD) 7.1 ± 1.9 7.3 ± 1.9 6.9 ± 2.0 NS
Procedures, n (%)
Hairline lowering 55 (69.6) 31 (81.6) 24 (58.5) 0.03
Brow lift 70 (88.6) 35 (92.1) 35 (85.4) NS
Fronto-orbital reconstruction 72 (91.1) 37 (97.4) 35 (85.4) NS
Fat graft 74 (93.7) 35 (92.1) 39 (95.1) NS
Rhinoplasty 73 (92.4) 34 (89.5) 39 (95.1) NS
Upper lip lift 33 (41.8) 21 (55.3) 12 (29.3) 0.02
Lip augmentation 51 (64.6) 28 (73.7) 23 (56.1) NS
Mandibular reconstruction 69 (86.3) 30 (78.9) 39 (95.1) 0.03
Length of surgery, median (IQR) hours 7.8 (6.7–8.9) 8.0 (7.0–9.4) 7.5 (6.4–8.8) NS
ERAS, enhanced recovery after surgery; NS, not significant; SD, standard deviation; IQR, interquartile range.

significantly higher compared to the pre-ERAS group where Patient-reported pain scores and length of hospital stay
no ketorolac was used. No differences were seen between In addition to the primary outcome of opioid usage, patient-
the groups for acetaminophen usage. In combination, reported pain scores and the length of hospital stay were
these data indicated that a reduction in perioperative evaluated. Compared to the pre-ERAS group, average pa­
opioid use occurred with the implementation of the ERAS tient-reported pain scores during the course of the hospital
protocol. stay were significantly lower in the ERAS group (2.5 ± 1.8

396
 Journal of Plastic, Reconstructive & Aesthetic Surgery 85 (2023) 393–400

Table 3 Analgesic medications, patient-reported pain scores, and length of hospital stay between Pre-ERAS and ERAS groups.
Total Cohort n = 79 Pre-ERAS ERAS p
n = 38 n = 41
Analgesic Medications
Total Opioids, median (IQR) MED/kg 1.1 (0.8–1.6) 1.5 (1.0–2.1) 0.8 (0.5–1.1) < 0.001
Intraoperative Total Opioids 0.6 (0.4–0.8) 0.6 (0.4–0.9) 0.6 (0.4–0.8) NS
Fentanyl, IV 0.3 (0.1–0.4) 0.2 (0.1–0.4) 0.3 (0.1–0.4) NS
Hydromorphone, IV 0.3 (0.2–0.4) 0.4 (0.2–0.5) 0.3 (0.2–0.4) NS
Postoperative Total Opioids 0.4 (0.1–0.9) 0.9 (0.4–1.5) 0.2 (0–0.5) < 0.001
Oxycodone, PO 0 (0–0.7) 0.8 (0.2–1.3) 0 (0–0) < 0.001
Tramadol, PO 0 (0–0.2) 0 (0) 0.1 (0–0.3) < 0.001
Hydromorphone, IV 0 (0–0.1) 0 (0–0.1) 0 (0–0.1) NS
Non-opioids, median (IQR) mg/kg
Ketorolac 0 (0–0.6) 0 (0–0) 0.6 (0.4–0.8) < 0.001
Acetaminophen 33.2 (22.6–47.6) 40.6 (21.9–57.6) 32.7 (24.8–39.1) NS
Pain Scores, (mean ± SD) 3.1 ± 1.8 3.7 ± 1.6 2.5 ± 1.8 0.002
Length of stay, hours (median, IQR) 29.4 (26.3–49.8) 32.4 (28.7–51.9) 27.3 (25.3–30.4) < 0.001
ERAS, enhanced recovery after surgery; NS, not significant; NA, not applicable; MED, morphine equivalent dose; IQR, interquartile range;
SD, standard deviation.

vs. 3.7 ± 1.6; p = 0.002). The reduction in pain scores was Total perioperative opioid use
accompanied by a decrease in the length of hospital stay The regression model explained 42.4% of the variance in
from a median (IQR) time of 32.4 (28.7–51.9) h in the pre- total opioid use in MED/kg [F(5,73) = 10.736, p < 0.001].
ERAS group to 27.3 (25.3–30.4) h in the ERAS group Among the independent variables included in the model,
(p < 0.001). the ERAS protocol independently predicted reduced opioid
use (β = −0.25, p = 0.01), whereas both previous gender-
Implementation of the FFS protocol as a predictor affirming surgeries (β = 0.21, p = 0.03) and a longer length
of hospital stay (β = 0.45, p < 0.001) independently pre­
of perioperative opioid use, patient-reported pain
dicted increased opioid use.
scores, and length of hospital stay

Next, we constructed multivariable linear regression Mean pain scores

models to identify whether implementation of the ERAS The model for mean pain scores explained 22.4% of the
protocol independently predicted perioperative opioid use, variance [F(5,73) = 4.225, p = 0.002]. Similar to the opioid
pain scores, or length of hospital stay (Table 4). In­ use model, implementation of the ERAS protocol in­
dependent variables in each of the models included length dependently predicted lower pain scores (β = −0.23,
of surgery to account for the contribution of the complexity p = 0.04) and a longer length of hospital stay predicted
of the procedure, previous gender-affirming surgery before higher pain scores (β = 0.24, p = 0.04).
FFS to account for the potential for opioid tolerance from
previous surgeries, mental health diagnoses to account for
the contribution of anxiety, depression, and previous sub­ Length of hospital stay
stance abuse as known contributors to postsurgical opioid The regression model for length of hospital stay explained
use.7 For the models evaluating the variance of the total 21.6% of the variance among patients [F(4,74) = 5.095,
perioperative opioid use and patient-reported pain scores, p = 0.001]. Implementation of the ERAS protocol in­
the total length of hospital stay was also included. dependently predicted a shorter hospital stay (β = −0.37,

Table 4 Implementation of the ERAS protocol as a predictor for perioperative opioid use, pain scores, and length of hospital
stay in multivariable linear regression models.
Opioid use Mean pain scores Length of hospital stay
R2 = 0.424 R2 = 0.224 R2 = 0.216
F (5,73) = 10.736, F (5,73) = 4.225, F (4,74) = 5.095,
P < 0.001 P = 0.002 P = 0.001

Predictors β p β p β p
ERAS protocol −0.25 0.01 −0.23 0.04 −0.37 < 0.001
Previous gender-affirming surgeries 0.21 0.03 0.09 NS 0.12 NS
Mental health diagnoses 0.04 NS 0.11 NS 0.05 NS
Length of surgery 0.04 NS 0.13 NS 0.20 NS
Length of hospital stay 0.45 < 0.001 0.24 0.04
NS, not significant; ERAS, enhanced recovery after surgery.

397
 M. Bedar, D. Dejam, R.M. Caprini et al.

p < 0.001), while no other predictors were found to affect thoughtful opioid-reduction strategies are highly important
the length of hospital stay. in this patient population.
In the extant literature, only two opioid-sparing perio­
perative protocols have been described for gender-affirming
surgeries, both for gender-affirming vaginoplasty. Salgado
Discussion et al. reported that the addition of epidural anesthesia
improved pain control16 and Tirrell et al. reported that the
In this work, we described the implementation and out­ replacement of intravenous patient-controlled opioid an­
comes of an ERAS protocol for gender-affirming FFS to re­ algesia with pro re nata oral opioid medications resulted in
duce postoperative opioid use in an at-risk patient a reduced total quantity of opioid administration as well as
population. The two main differences in patients treated a reduced hospital stay.17 Both former reports differ sig­
under the ERAS protocol compared to pre-ERAS were: (1) nificantly from the current report due to differences in
the addition of scheduled ketorolac and (2) the replacement anatomic location. Unlike vaginoplasty surgery, FFS is fre­
of postoperative oral oxycodone with oral tramadol as the quently associated with sensory neuropraxia in the V1 and
primary pro re nata narcotic oral analgesic. The effects of V3 distributions of the trigeminal nerve due to the extensive
the ERAS protocol were then assessed using three primary fronto-orbital and mandibular osseous reconstruction.
outcome measures: total perioperative opioid usage, pa­ Hence, standard trigeminal nerve blocks are unlikely to
tient-reported pain scores, and total length of inpatient assist in postoperative pain control. In contrast to the latter
stay. All three measures demonstrated a reduction in the study, the length of hospital stay for facial gender-affirming
ERAS cohort compared to the pre-ERAS cohort. surgery patients is significantly shorter and the need for
Furthermore, when controlling for other potential factors opioid analgesics is also lower overall. Hence, the oral
that may influence any of the three outcomes such as pre­ opioid used in the current ERAS protocol, tramadol, was
vious gender-affirming surgeries, mental health diagnoses, chosen specifically due to its potency, approximately 20%
and complexity of the specific surgery, the implementation that of oral morphine18 as well as its lower risk of misuse.27
of the ERAS protocol independently predicted decreased In contrast to the majority of reported ERAS protocols,
opioid use, decreased patient-reported pain scores, and we consciously chose against incorporating gabapentin as
decreased length of hospital stay. part of the multimodal analgesia regimen.12,14,17 The ra­
Over the past decade, the contribution of postsurgical tionale to omit gabapentin was, in part, to limit the number
prescription opioids to opioid misuse has been increasingly of simultaneous changes in the perioperative pain regimen
clarified. Following elective surgical procedures, persistent as well as to prevent the potential for gabapentin misuse,
opioid use has been estimated by Waljee, Brummett, and an increasingly recognized entity.28–30 Motivations for ga­
colleagues to be approximately 6%−13% of all opioid-naïve bapentinoid misuse have been attributed to a myriad of
adult patients and approximately 5% of all opioid-naïve effects, such as auditory hallucinations, euphoric mood,
adolescent patients with variations based on the type of and dissociation.31 The alarming effect of gabapentin as a
surgery.7,19–21 These effects are potentially cumulative as potentiator of opioid-induced deaths28,32 has resulted in its
Agarwal et al. demonstrated that both prior noncontinuous reclassification to a Class C controlled substance in the
and continuous opioid exposure for 3 months or longer United Kingdom as well as tighter regulation at the state
predicted higher odds of persistent opioid use.22 Among the level in the United States. The hazards of gabapentinoid
transgender patient population, the risks of opioid use fol­ misuse suggest that re-evaluation for its inclusion in ERAS
lowing surgery may be higher secondary to other existing protocols overall may be indicated.
risk factors for substance use.7,23–25 While improvement in pain outcomes during the perio­
In addition to behavioral and mental health-related risk perative period is encouraging, ultimately, assessment of
factors, the quantity and frequency of gender-affirming the current ERAS protocol requires long-term evaluation of
surgeries may also place transgender individuals at an in­ opioid misuse in the outpatient setting. The high prevalence
creased risk of opioid misuse compared to their cisgender of mental health diagnoses in the transgender patient po­
peers. The US Transgender Survey data indicated that many pulation overall,24,25 as well as in the current cohort
transgender individuals seek to undergo more than one (75.9%), suggests that the risk profile for persistent opioid
gender-affirming surgery, and the perceived psychosocial use after gender-affirming surgery may be uniquely higher.7
benefits of these surgeries may encourage patients to un­ Further analysis using statewide prescription tracking da­
dergo multiple procedures in a short time span to alleviate tabases similar to the report by Chao and colleagues on
dysphoria-related stressors.26 An uptick in health insurance breast reconstruction14 will be necessary to determine the
coverage for various gender-affirming procedures may also success of the current ERAS protocol.
alleviate financial barriers to surgery, thus allowing trans­
gender patients to undergo more procedures with con­
sequent opioid exposures and less recovery time between Conclusions
procedures. While the relationship of gender-affirming sur­
geries to opioid misuse has not been well-studied, the cur­ We describe an ERAS protocol to reduce opioid usage in
rent work suggests that prior gender-affirming surgeries patients undergoing gender-affirming FFS. Our analysis
predicted higher quantities of total perioperative opioid suggests that this regimen reduced perioperative opioid
consumption after FFS (Table 4). In combination, these lines consumption, decreased patient-reported pain, and shor­
of evidence would suggest that vigilance in opioid admin­ tened the time to discharge after surgery. The implications
istration following gender-affirming surgeries as well as of these findings are particularly important among

398
 Journal of Plastic, Reconstructive & Aesthetic Surgery 85 (2023) 393–400

transgender and gender-diverse patients, populations that 5. Rowe C, Santos GM, McFarland W, Wilson EC. Prevalence and
are vulnerable to higher rates of substance abuse than the correlates of substance use among trans female youth ages 16-
general population. 24 years in the San Francisco bay area. Drug Alcohol Depend
2015;147:160–6. https://doi.org/10.1016/j.drugalcdep.2014.
11.023.
6. Hanna B, Desai R, Parekh T, Guirguis E, Kumar G, Sachdeva R.
Funding Psychiatric disorders in the US transgender population. Ann
Epidemiol 2019;39(1–7):e1.
7. Brummett CM, Waljee JF, Goesling J, et al. New persistent
This work was supported by the Bernard G. Sarnat opioid use after minor and major surgical procedures in US
Endowment for Craniofacial Biology (JCL) and the Jean adults. JAMA Surg 2017;152(6):e170504. https://doi.org/10.
Perkins Foundation (JCL). JCL is additionally supported by 1001/jamasurg.2017.0504.
the National Institutes of Health/National Institute of 8. Bartlett EL, Zavlin D, Friedman JD, Abdollahi A, Rappaport NH.
Dental and Craniofacial Research R01 DE029234 and R01 Enhanced recovery after surgery: the plastic surgery paradigm
DE028098. shift. Aesthet Surg J 2017;38(6):676–85. https://doi.org/10.
1093/asj/sjx217.
9. Ljungqvist O, Scott M, Fearon KC. Enhanced recovery after
surgery: a review. JAMA Surg 2017;152(3):292–8. https://doi.
Ethical approval org/10.1001/jamasurg.2016.4952.
10. Sebai ME, Siotos C, Payne RM, et al. Enhanced recovery
We have obtained ethical approval for this study involving after surgery pathway for microsurgical breast reconstruc­
human subjects through the University of California Los tion: a systematic review and meta-analysis. Plast Reconstr
Angeles Institutional Review Board protocol 11-000925. Surg 2019;143(3):655–66. https://doi.org/10.1097/prs.
0000000000005300.
11. Newman A, Davison SP. An analgesic regimen for opioid re­
duction in elective plastic surgery: a randomized prospective
Financial disclosure statement study. 02 01 Plast Reconstr Surg 2021;147(2):325e–30e.
https://doi.org/10.1097/PRS.0000000000007592.
All authors have no financial interests, including in pro­ 12. Dietz N, Sharma M, Adams S, et al. Enhanced Recovery After
Surgery (ERAS) for spine surgery: a systematic review. World
ducts, devices, or drugs associated with this manuscript.
Neurosurg 2019;130:415–26. https://doi.org/10.1016/j.wneu.
JCL is a medical education consultant for Stryker. All
2019.06.181.
sources of funds supporting the completion of this manu­ 13. Johnson AC, Colakoglu S, Reddy A, et al. Perioperative blocks
script are under the auspices of the University of California, for decreasing postoperative narcotics in breast reconstruc­
Los Angeles. tion. Anesth Pain Med 2020;10(5):e105686. https://doi.org/
10.5812/aapm.105686.
14. Rendon JL, Hodson T, Skoracki RJ, Humeidan M, Chao AH.
Enhanced recovery after surgery protocols decrease outpatient
Appendix A. Supporting information opioid use in patients undergoing abdominally based microsurgical
breast reconstruction. Plast Reconstr Surg 2020;145(3):645–51.
Supplementary data associated with this article can be https://doi.org/10.1097/prs.0000000000006546.
found in the online version at doi:10.1016/j.bjps.2023. 15. Lillemoe HA, Marcus RK, Day RW, et al. Enhanced recovery in
07.044. liver surgery decreases postoperative outpatient use of
opioids. Surgery 2019;166(1):22–7. https://doi.org/10.1016/j.
surg.2019.02.008.
16. Salgado CJ, Gonzalez DI, Wolfe EM, et al. Combined general
References and epidural anesthesia is associated with decreased opioid
consumption and enhanced pain control after penile inversion
1. Kidd JD, Goetz TG, Shea EA, Bockting WO. Prevalence and vaginoplasty in transwomen. Ann Plast Surg 2019;83(6):681–6.
minority-stress correlates of past 12-month prescription drug https://doi.org/10.1097/SAP.0000000000001921.
misuse in a national sample of transgender and gender non­ 17. Tirrell AR, Abu El Hawa AA, Bekeny JC, Fan KL, Del Corall D.
binary adults: results from the U.S. transgender survey. Drug Minimizing opioid consumption by eliminating patient-con­
Alcohol Depend 2021;219:108474. https://doi.org/10.1016/j. trolled analgesia after penile inversion vaginoplasty with
drugalcdep.2020.108474. Enhanced Recovery After Surgery (ERAS) protocol. Andrology
2. Hughto JMW, Quinn EK, Dunbar MS, Rose AJ, Shireman TI, Jasuja 2022;10(2):232–40. https://doi.org/10.1111/andr.13095.
GK. Prevalence and co-occurrence of alcohol, nicotine, and other 18. Nielsen S, Degenhardt L, Hoban B, Gisev N. A synthesis of oral
substance use disorder diagnoses among US transgender and cis­ morphine equivalents (OME) for opioid utilisation studies.
gender adults. JAMA Netw Open 2021;4(2):e2036512. https://doi. Pharmacoepidemiol Drug Saf 2016;25(6):733–7. https://doi.
org/10.1001/jamanetworkopen.2020.36512. org/10.1002/pds.3945.
3. NIDA. What is the scope of prescription drug misuse in the 19. Bennett KG, Kelley BP, Vick AD, et al. Persistent opioid use and
United States? National Institute on Drug Abuse; 2022. high-risk prescribing in body contouring patients. Plast
〈https://nida.nih.gov/publications/research-reports/misuse- Reconstr Surg 2019;143(1):87–96. https://doi.org/10.1097/
prescription-drugs/what-scope-prescription-drug-misuse〉. PRS.0000000000005084.
[Accessed May 4, 2022]. 20. Harbaugh CM, Lee JS, Hu HM, et al. Persistent opioid use
4. Benotsch EG, Zimmerman R, Cathers L, et al. Non-medical use among pediatric patients after surgery. Pediatrics
of prescription drugs, polysubstance use, and mental health in 2018;141(1). https://doi.org/10.1542/peds.2017-2439.
transgender adults. Drug Alcohol Depend 2013;132(1–2):391–4. 21. Johnson SP, Chung KC, Zhong L, et al. Risk of prolonged opioid
https://doi.org/10.1016/j.drugalcdep.2013.02.027. use among opioid-naïve patients following Common hand

399
 M. Bedar, D. Dejam, R.M. Caprini et al.

surgery procedures. J Hand Surg Am 2016;41(10):947–57. humans. Syst Rev Front Psychiatry 2019;10:704. https://doi.
https://doi.org/10.1016/j.jhsa.2016.07.113. org/10.3389/fpsyt.2019.00704.
22. Agarwal S, Shah A, Gunaseelan V, et al. New persistent opioid 28. Kalk NJ, Chiu CT, Sadoughi R, et al. Fatalities associated
use after surgery in patients with a history of remote opioid with gabapentinoids in England (2004-2020). Br J Clin
use. Surgery 2022;171(6):1635–41. https://doi.org/10.1016/j. Pharmacol 2022;88(8):3911–7. https://doi.org/10.1111/bcp.
surg.2021.11.008. 15352.
23. Bouman WP, Claes L, Brewin N, et al. Transgender and anxiety: 29. Karavolis ZA, Su AB, Peckham AM. State-level response to ga­
acomparative study between transgender people and the bapentin misuse in the United States: implications and future
general population. Int J Transgenderism 2017;18(1):16–26. direction. Am J Health Syst Pharm 2022;79(9):e143–8. https://
https://doi.org/10.1080/15532739.2016.1258352. doi.org/10.1093/ajhp/zxab486.
24. Veale JF, Watson RJ, Peter T, Saewyc EM. Mental health dis­ 30. Smith RV, Havens JR, Walsh SL. Gabapentin misuse, abuse and
parities among Canadian transgender youth. J Adolesc Health diversion: a systematic review. Addiction 2016;111(7):
2017;60(1):44–9. https://doi.org/10.1016/j.jadohealth.2016. 1160–74. https://doi.org/10.1111/add.13324.
09.014. 31. Vickers-Smith R, Sun J, Charnigo RJ, Lofwall MR, Walsh SL,
25. Streed Jr. CG, McCarthy EP, Haas JS. Association between gender Havens JR. Gabapentin drug misuse signals: a pharmacov­
minority status and self-reported physical and mental health in the igilance assessment using the FDA adverse event reporting
United States. JAMA Intern Med 2017;177(8):1210–2. https://doi. system. Drug Alcohol Depend 2020;206:107709. https://doi.
org/10.1001/jamainternmed.2017.1460. org/10.1016/j.drugalcdep.2019.107709.
26. James SE, Herman JL, Rankin S, et al. The report of the 2015 32. Gomes T, Juurlink DN, Antoniou T, Mamdani MM, Paterson JM,
U.S. transgender survey. Washington, DC: National Center for van den Brink W. Gabapentin, opioids, and the risk of opioid-
Transgender Equality; 2016. related death: a population-based nested case-control study.
27. Dunn KE, Bergeria CL, Huhn AS, Strain EC. A systematic review PLoS Med 2017;14(10):e1002396. https://doi.org/10.1371/
of laboratory evidence for the abuse potential of tramadol in journal.pmed.1002396.

400