Foot & Ankle International - September 2019 - 1044

1044	
many patients are first exposed to opioids.2,7,21,22 Orthopedic
surgeons rank fourth among the highest opioid-prescribing
medical specialties.33,36 Recent studies have analyzed new
persistent opioid use among opioid-naïve patients following
surgery, and these studies have demonstrated that persistent
use is linked less to poorly controlled postsurgical pain and
more to identifiable patient risk factors and prescribing
patterns.2,3,7,13,19,21,35
Ankle fractures are one of the most common orthopedic
injuries and represent the second most common fracture of
the lower extremity behind proximal femur fractures.12,17,29,32
Ankle fractures encompass a spectrum of injury patterns to
the distal fibula and/or distal tibia malleoli, which in conjunction with the ligaments of the ankle maintain congruence of the tibiotalar joint. Stable ankle fractures represent
fractures in which tibiotalar congruence is maintained and
are treated nonoperatively.12 Most commonly, these fractures represent isolated lateral malleolus fractures in which
stability is maintained by the medial ankle ligaments.1
Unstable injuries represent bi- or trimalleolar fractures or
fractures with disruption of the medial ankle ligaments or
syndesmosis and are most often treated with open reduction
and internal fixation to restore tibiotalar congruence.12
We sought to examine the rates of new persistent opioid
use among opioid-naïve patients who underwent operative
treatment of unstable ankle fracture patterns as well as nonoperative treatment of a stable ankle fracture and to identify
how injury severity, procedure type, prescribing patterns,
and patient-related factors may be associated with persistent opioid use.

Methods
Dataset and Patient Cohort
Using the Clinformatics DataMart Database (OptumInsight,
Eden Prarie, MN), a de-identified claims database from a
national insurance provider, we identified patients in the
United States who underwent surgical treatment or closed
treatment of ankle fractures between January 2009 and June
2016. Patients who had additional procedures on the same
date or were admitted for multiple trauma (Appendix 1)
were excluded. Insurance claims were examined from
January 2008 to December 2016, among adults aged 18 to
64 years, to capture opioid-prescribing data in a period from
1 year pretreatment to 6 months posttreatment. Only patients
with continuous medical insurance coverage and prescription drug coverage were included. Patients with missing
insurance enrollee identification numbers were excluded,
and we deleted pharmacy claims records with quantity or
day supply equal to zero or if no National Drug Code was
available. Patients younger than 18 years and older than 64
years were also excluded, the latter due to the incomplete
data capture of Medicare Part D prescription claims data.

Foot & Ankle International 40(9)
The study was evaluated and determined exempt from
review by the Institutional Review Board.
Four Current Procedural Terminology, 4th Edition (CPT4) codes were chosen to identify patients who underwent
open surgical treatment of the most common unstable ankle
fracture subtypes during the study period. Additionally, a
CPT-4 code for closed treatment of distal fibula fractures
was used to identify a control, nonoperative group. This
group was chosen to represent a stable fracture pattern
group based on the diagnosis and associated treatment. The
included procedure codes are shown in Table 1. We sought
to compare new persistent opioid use among opioid-naïve
patients who underwent operative or nonoperative management of ankle fractures and who filled an opioid prescription during the peritreatment period of 14 days prior to and
7 days following either surgery or initiation of nonoperative
management. This period was chosen to capture those who
may have been treated for pain control between the date of
injury and the date of definitive treatment. Consistent with
prior similar studies of opioid-naïve patients,2,4,7,13,16,19,35
patients who filled an opioid prescription in the period of 12
months to 15 days prior to the peritreatment period were
excluded. Those who underwent any additional surgical
procedures in the 6 months following the date of treatment
of their ankle fracture were also excluded by identifying
CPT-4 codes for anesthesia (Appendix 2) which were billed
during that period.

Outcomes and Exposure Variables
The primary outcome measure of this study was new persistent opioid use, defined by fulfillment of an opioid prescription between 91 and 180 days after operative or nonoperative
treatment of an ankle fracture. A 90-day recovery period
was chosen, as this has been previously defined as the
period for normal healing.26 Exposure variables were treatment type while controlling for the total amount of opioids
prescribed during the peritreatment period, patient sociodemographic information, and other patient-specific factors.
Patient factors included age, sex, region of residence,
median household income of the metropolitan statistical
area where a patient resided, tobacco use defined by health
risk assessment data, and comorbid conditions using a
Charlson Comorbidity Index (CCI) score.11 Patients were
sorted by categorical variables including age group, by
household income, and into 1 of the 4 CCI score groups
(CCI = 0, 1-2, 3-4, ≥5). A few patients had no valid values
for education, geographic region, and household income
and were assigned as "unknown." Additionally, because of
their known association with pain perception and long-term
opioid use, coexisting mental health diagnoses23 and pain
disorders (Appendix 3) were identified.8,20,28 Diagnoses of
neuropathic pain conditions following treatment were also
evaluated to account for continued opioid use.
Foot & Ankle International - September 2019

Table of Contents for the Digital Edition of Foot & Ankle International - September 2019
