Abstract
Total knee arthroplasty (TKA) is a common and effective treatment of knee osteoarthritis.
As the amount of TKAs performed increases, so does the number of TKA failures and
subsequent revisions. Diabetes mellitus (DM) has been shown to increase complications
following orthopaedic procedures. For these reasons, it is important to understand
the association between severity of DM and the risk of postoperative adverse events
following revision TKA. A retrospective cohort study was conducted using the American
College of Surgeons' National Surgical Quality Improvement Program database. Patients
who underwent revision TKAs between 2007 and 2016 were identified and recorded as
having noninsulin-dependent DM (NIDDM), insulin-dependent DM (IDDM), or no DM. Univariate
and multivariate analysis were used to evaluate the incidence of multiple adverse
events within 30 days after revision TKA. A total of 13,246 patients who underwent
revision TKA were selected (without DM = 10,381 [78.4%]; NIDDM = 1,890 [14.3%]; IDDM = 975
[7.4%]). Patients with NIDDM were found to have an increased risk of developing renal
insufficiency and urinary tract infection (UTI) compared with patients without DM,
while patients with IDDM were found to have an increased risk of developing 10 of
20 adverse events compared with patients without DM. NIDDM is an independent risk
factor for UTI and IDDM is an independent factor for development of three complications
compared with no DM. Insulin dependency is an independent factor for septic shock,
postoperative blood transfusion, and extended postoperative hospital stay. Relative
to patients with NIDDM, those with IDDM have a greater likelihood of developing more
adverse perioperative outcomes than patients without DM. Although complication rates
remain relatively low, orthopaedic surgeons must consider the implications of diabetes
and insulin dependence on patient selection, preoperative risk stratification, and
postoperative outcomes.
Keywords
revision - knee - arthroplasty - diabetes mellitus - insulin