- Email: [email protected]
Patient-Related Risk Factors for Periprosthetic Ankle Joint Infection: An Analysis of 6977 Total Ankle Arthroplasties
ARTICLE IN PRESS The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
Contents lists available at ScienceDirect
The Journal of Foot & Ankle Surgery j o u r n a l h o m e p a g e : w w w. j f a s . o r g
Original Research
Patient-Related Risk Factors for Periprosthetic Ankle Joint Infection: An Analysis of 6977 Total Ankle Arthroplasties Alyssa Althoff, BS 1, Jourdan M. Cancienne, MD 2, Minton T. Cooper, MD 3, Brian C. Werner, MD 3 1Medical
Student, Medical University of South Carolina, Charleston, SC Resident Physician, Department of Orthopaedic Surgery, University of Virginia, Charlottesville, VA 3 Assistant Professor, Department of Orthopaedic Surgery, University of Virginia, Charlottesville, VA 2
A R T I C L E
I N F O
Level of Clinical Evidence: 3
Keywords: complication periprosthetic joint infection risk factors total ankle arthroplasty
A B S T R A C T
Periprosthetic joint infection (PJI) after total ankle arthroplasty (TAA) is a devastating complication that often results in explantation to resolve the infection. The purpose of the present investigation was to determine the patient-related risk factors for PJI after TAA. A national insurance database was queried for patients undergoing TAA using the Current Procedural Terminology and International Classification of Diseases, ninth revision, procedure codes from 2005 to 2012. Patients undergoing TAA with concomitant fusion procedures or more complex forefoot procedures were excluded. PJI within 6 months was then assessed using the International Classification of Diseases, ninth revision, codes for diagnosis or treatment of postoperative PJI. Multivariate binomial logistic regression analysis was performed to evaluate the patient-related risk factors for PJI. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for each risk factor, with p < .05 considered statistically significant. A total of 6977 patients were included in the present study. Of these 6977 patients, 294 (4%) had a diagnosis of, or had undergone a procedure for, PJI. The independent risk factors for PJI included age <65 years (OR 1.44; p = .036), body mass index <19 kg/m2 (OR 3.35; p = .013), body mass index >30 kg/m2 (OR 1.49; p = .034), tobacco use (OR 1.59; p = .002), diabetes mellitus (OR 1.36; p = .017), inflammatory arthritis (OR 2.38; p < .0001), peripheral vascular disease (OR 1.64; p < .0001), chronic lung disease (OR 1.37; p = .022), and hypothyroidism (OR 1.32; p = .022). The independent patient-related risk factors identified in the present study should help guide physicians and patients considering elective TAA and develop risk stratification algorithms that could decrease the risk of deep, postoperative infection. © 2017 by the American College of Foot and Ankle Surgeons. All rights reserved.
End-stage ankle arthritis is a debilitating condition that affects comfort, function, and health-related quality of life (1). Ankle osteoarthritis is an increasingly prevalent condition, affecting approximately 1% of the population (1). Posttraumatic osteoarthritis accounts for ~12% of symptomatic lower extremity osteoarthritis, correlating to 5.6 million individuals in the United States and increasing (2). Unlike hip or knee osteoarthritis, the development of symptomatic ankle arthritis is predominately posttraumatic, occurring after an ankle sprain, fracture, or ligament instability (3–5). Conservative management of endstage ankle arthritis often involves physical therapy, administration of nonsteroidal antiinflammatory drugs, intraarticular corticosteroid injections, and modification of physical activity and footwear (6).
Financial Disclosure: None reported. Conflict of Interest: None reported. Address correspondence to: Jourdan M. Cancienne, MD, Department of Orthopaedic Surgery, University of Virginia, P.O. Box 800159, Charlottesville, VA 22908. E-mail address: [email protected] (J.M. Cancienne).
However, the surgical treatment of end-stage ankle arthritis is indicated when conservative treatment does not alleviate the symptoms (7). Ankle arthrodesis has previously been widely accepted as the reference standard for surgical treatment of ankle arthritis; however, total ankle arthroplasty (TAA) has gained favor in recent years as an effective alternative that preserves some tibiotalar motion (2,8–11). The goal of TAA is to restore functional range of motion, eliminate pain, and improve quality of life (12). During the past 10 years, the number of TAAs performed has dramatically increased in total volume and hospital usage (11,13), and the indications for surgery have expanded (12,14,15). Subsequent generations of implants since the initial introduction in the 1970s have reportedly decreased both the mechanical failure rate and the risk of infection (7,16). Despite these improvements, periprosthetic ankle joint infection remains a significant complication after TAA, often resulting in implant failure (16). The reported rate of infection ranges from 2.4% to 8.9% (6,17,18), greater than the rate of infection after total knee and hip replacements (19). Thus, it is critical to identify the patient-related risk factors associated with periprosthetic infection after TAA.
1067-2516/$ - see front matter © 2017 by the American College of Foot and Ankle Surgeons. All rights reserved. https://doi.org/10.1053/j.jfas.2017.09.006
ARTICLE IN PRESS 2
J.M. Cancienne, et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
Although published reports have identified some risk factors associated with postoperative infection after TAA within small cohorts (17), to the best of our knowledge, no studies have yet demonstrated the pervasiveness of these factors in a larger population. Accordingly, the purposes of the present study were to use a national database to determine the incidence of postoperative infection and identify independent variables that increase patients’ risk of periprosthetic infection after TAA. Materials and Methods Database The PearlDiver Patient Records Database (Fort Wayne, IN; available at: www.pearldiverinc.com), an insurance claim-based database of patient records, was used for the present study. The database contains data from several different insurers, including both Medicare and private insurers. The 100% Medicare Standard Analytical Files database within PearlDiver was chosen for the present study, because it contains data from the largest number of patients, and most patients undergoing TAA are of Medicare age. The PearlDiver Medicare database contains patient demographic and procedure volume data, among numerous other data for patients with International Classification of Diseases, ninth revision (ICD-9), diagnosis and procedure codes or Current Procedural Terminology (CPT) codes. The database covers patients insured from 2005 through 2012. Overall, the database contains the data from ~100 million patients. All data have been de-identified and anonymous and were thus deemed exempt by our institutional review board. Study Cohort Patients who had undergone TAA were identified using ICD-9 procedure code 81.56 (total ankle replacement) and/or CPT code 27702 (arthroplasty, ankle; with implant, total ankle). Patients with a concomitant revision arthroplasty code (ICD-9 code 81.59 and/or CPT codes 27703 and 27704) were excluded. Covariates and Predictors The covariates included in the present study were identified using specific ICD-9 diagnostic codes. These included age, sex, tobacco use, alcohol abuse, inflammatory arthritis, depression, diabetes mellitus, hyperlipidemia, hypertension, peripheral vascular disease, congestive heart failure, coronary artery disease, chronic kidney disease, chronic lung disease, chronic liver disease, hypercoagulable state, hypothyroidism, preoperative chronic anemia, body mass index (BMI) <19 kg/m2, and current hemodialysis use. The specific BMI was also included in the analysis. Separate ICD-9 diagnostic codes are used for obesity (BMI 30 to 40 kg/m2) and morbid obesity (BMI >40 kg/m2), which thus could be evaluated as 2 separate covariates. This method has been used in previous studies of administrative databases (20,21). Several of these covariates are also included in several comorbidity indexes, including the Charlson comorbidity score, Elixhauser comorbidity score, and the combined comorbidity score (22). In addition, several of these covariates have been evaluated in previous studies examining the risk factors for deep infection after TAA (17,18,23). Infection The study endpoint of interest was infection within 3 to 6 months after TAA. Although most of the current data have focused on the occurrence of early infection after TAA, studies have reported the development of chronic deep infections occurring >6 months postoperatively (17,18,23). After an institutional review of >900 total ankle arthroplasties, Patton et al (23) reported that deep infection after surgery was diagnosed <3 months after the index surgery in 38% of patients, 3 to 12 months in 38% of patients, and >12 months in 24% of patients. Thus, it is important to identify the factors that contribute to infection both within 3 months and >6 months postoperatively. However, many of the infection codes are not specific to TAA. Thus, a period >6 months was not appropriate, because any infection could have resulted from another procedure or event as the interval between the infection and TAA increased. Infection was defined as the diagnosis of periprosthetic infection or septic ankle and/or a procedure for postoperative infection or septic ankle arthritis. The diagnosis codes included ICD-9 codes 998.51, 998.59, 996.66-7, 996.69, 711.07, 711.47, 711.87, and 711.97. The CPT procedure codes included 10180, 20005, 27603, 27607, and 27610. Statistical Analysis The overall infection rate from 2005 to 2012 and the annual infection rate were calculated from the database output. Linear regression analysis was used to examine any
Table 1 Distribution of patient variables and infection rates (N = 6977 patients) Variable
Population
Infection at 3 mo
Infection at 6 mo
Age <65 y Obesity (BMI 30 to 40 kg/m2) Morbid obesity (BMI ≥40 kg/m2) Tobacco use Alcohol abuse Inflammatory arthritis Depression Hypercoagulable disorder Diabetes mellitus Hyperlipidemia Hypertension Peripheral vascular disease Congestive heart failure Coronary artery disease Chronic kidney disease Chronic lung disease Chronic liver disease Current hemodialysis use Hypothyroidism Low BMI (<19 kg/m2) Preoperative anemia
1105 (15.8) 1322 (18.9) 602 (8.6) 980 (14.0) 331 (4.7) 1093 (15.7) 1947 (27.9) 326 (4.7) 2013 (28.9) 5291 (75.8) 5857 (83.9) 793 (11.4) 1196 (17.1) 2401 (34.4) 1172 (16.8) 1505 (21.6) 396 (5.7) 122 (1.7) 2010 (28.8) 39 (0.6) 1020 (14.6)
43 (3.9) 46 (3.5) 21 (3.5) 47 (4.8) 15 (4.5) 41 (3.8) 72 (3.7) 9 (2.8) 66 (3.3) 137 (2.6) 159 (2.7) 50 (6.3) 49 (4.1) 66 (2.7) 44 (3.8) 61 (4.1) 11 (2.8) 2 (1.6) 63 (3.1) 2 (5.1) 45 (4.4)
72 (6.5) 70 (5.3) 37 (6.1) 70 (7.1) 23 (6.9) 74 (6.8) 108 (5.5) 12 (3.7) 124 (6.2) 234 (4.4) 270 (4.6) 82 (10.3) 87 (7.3) 132 (5.5) 83 (7.1) 111 (7.4) 15 (3.8) 7 (5.7) 106 (5.3) 5 (12.8) 62 (6.1)
Data presented as n (%). Abbreviation: BMI, body mass index.
change in the infection rate over time. A descriptive analysis was conducted at 3 and 6 months detailing the distribution of patient variables and infection rates. Multivariable binomial logistic regression analysis was then performed to examine the independent effect of the included covariates on the incidence of infection after TAA. The odds ratios (ORs), 95% confidence intervals (CIs), and p values were calculated for all variables; p < .05 was considered statistically significant for all statistical tests. All statistical testing was performed within the PearlDiver system using R, an open source statistical package (R Foundation, Vienna, Austria; available at: https://www.r-project.org/).
Results A total of 6977 patients who had undergone TAA were included in the present study. During the study period, 294 patients developed infection within 6 months postoperatively, for an overall infection rate of 4.2%. The rate of infection ranged from 3.0% to 4.8% from 2005 to 2012; however, no significant increasing or decreasing trend was found (p = .547; Table 1). The distribution of the covariates and infection rates for the population studied are listed in Table 1. Numerous independent risk factors for infection were identified using logistic regression analysis at both 3 and 6 months. Within 3 months of surgery, the significant risk factors for deep infection included age <65 years (OR 1.5, 95% CI 1.08 to 1.87), obesity (OR 1.48, 95% CI 1.08 to 2.02), low BMI (OR 2.12, 95% CI 1.78 to 2.53), tobacco use (OR 1.79, 95% CI 1.25 to 2.56), inflammatory arthritis (OR 1.42, 95% CI 1.01 to 2.01), peripheral vascular disease (OR 3.14, 95% CI 2.22 to 4.42), hypothyroid disorder (OR 1.27, 95% CI 1.02 to 1.58), and preoperative anemia (OR 1.51, 95% CI 1.07 to 2.13; Table 2). Within 6 months of surgery, the significant risk factors for deep infection included age <65 years (OR 1.61, 95% CI 1.20 to 2.16), low BMI (OR 2.67, 95% CI 1.07 to 6.67), obesity (OR 1.47, 95% CI 1.15 to 1.87), tobacco use (OR 1.44, 95% CI 1.08 to 1.92), diabetes mellitus (OR 1.35, 95% CI 1.05 to 1.73), inflammatory arthritis (OR 1.67, 95% CI 1.28 to 2.18), peripheral vascular disease (OR 2.46, 95% CI 1.87 to 3.22), chronic kidney disease (OR 1.39, 95% CI 1.05 to 1.84), chronic lung disease (OR 1.51, 95% CI 1.16 to 1.96), and hypothyroidism (OR 1.32, 95% CI 1.03 to 1.69; Table 3). Discussion The use of TAA has continued to gain favor over ankle arthrodesis as an effective alternative surgical treatment of end-stage ankle
ARTICLE IN PRESS J.M. Cancienne, et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
Table 2 Covariates for infection after total ankle arthroplasty within 3 months (N = 6977 patients) Variable Demographic data Age <65 y Male sex Procedure in 2005 to 2008 Comorbidities Obesity (BMI >30 kg/m2) Low BMI (<19 kg/m2) Tobacco use Alcohol abuse Inflammatory arthritis Depression Hypercoagulable disorder Diabetes mellitus Hyperlipidemia Hypertension Peripheral vascular disease Congestive heart failure Coronary artery disease Chronic kidney disease Lung disease Liver disease Current hemodialysis use Hypothyroidism Preoperative anemia
OR
95% CI
p Value
1.44 1.07 0.98
1.02 to 2.03 0.77 to 1.47 0.72 to 1.33
.688 .025 .678
1.48 2.12 1.79 1.16 1.42 1.16 0.73 1.41 0.98 1.08 3.14 1.24 0.63 1.17 1.29 0.63 0.32 1.27 1.51
1.08 to 2.02 1.78 to 2.53 1.25 to 2.56 0.66 to 2.05 1.01 to 2.01 0.84 to 1.61 0.38 to 1.42 1.01 to 1.95 0.66 to 1.44 0.68 to 1.72 2.22 to 4.42 0.84 to 1.84 0.44 to 0.91 0.81 to 1.70 0.92 to 1.82 0.34 to 1.19 0.08 to 1.34 1.02 to 1.58 1.07 to 2.13
.017 .009* .001* .600 .046* .359 .350 .033* .900 .747 < .0001* .279 .013* .408 .146 .155 .117 .018* .018*
Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio. * Statistically significant.
arthritis in appropriately selected patients (11). Although first- and second-generation arthroplasty systems were associated with high rates of mechanical failure, more recent advances in surgical technique and implant design have significantly improved the survivorship and longevity of TAA (7,16,24,25). Thus, the focus on periprosthetic joint infections (PJIs) as a cause of implant failure and revision has been increasing (17). However, the limited patient numbers in previous studies have made it difficult to identify patient-specific risk factors
Table 3 Covariates for infection after total ankle arthroplasty within 6 months (N = 6977 patients) Variable Demographic data Age <65 y Male sex Procedure in 2005 to 2008 Comorbidities Obesity (BMI >30 kg/m2) Low BMI (<19 kg/m2) Tobacco use Alcohol abuse Inflammatory arthritis Depression Hypercoagulable disorder Diabetes mellitus Hyperlipidemia Hypertension Peripheral vascular disease Congestive heart failure Coronary artery disease Chronic kidney disease Lung disease Liver disease Current hemodialysis use Hypothyroidism Preoperative anemia
OR
95% CI
p Value
1.61 1.23 0.88
1.20 to 2.16 0.96 to 1.58 0.69 to 1.13
.002 .109 .317
1.47 2.67 1.44 1.09 1.67 0.97 0.51 1.35 0.90 1.33 2.46 1.21 0.94 1.39 1.51 1.46 1.28 1.32 1.05
1.15 to 1.87 1.07 to 6.67 1.08 to 1.92 0.69 to 1.74 1.28 to 2.18 0.75 to 1.25 0.28 to 0.91 1.05 to 1.73 0.66 to 1.23 0.89 to 1.98 1.87 to 3.22 0.89 to 1.62 0.71 to 1.23 1.05 to 1.84 1.16 to 1.96 0.85 to 2.50 0.62 to 2.65 1.03 to 1.69 0.78 to 1.40
.014 .019* .013* .701 < .0001* .825 .022 .012* .516 .158 < .0001* .219 .642 .022* .002* .246 .511 .028 .752
Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio. * Statistically significant.
3
for PJI after TAA. In the present study, we are able to sufficiently power an analysis of >6000 patients who had undergone TAA to identify several significant risk factors for periprosthetic ankle joint infection after TAA. These risk factors included age <65 years, low BMI, obesity, diabetes, inflammatory arthritis, peripheral vascular disease, and hypothyroidism. Surgical site infection is a high-grade complication that, when associated with TAA, has been reported to result in implant failure in 80.6% of cases (16). In studies reporting postoperative infection, the rates have ranged from 2% to 8.5% (6,8,17,18,23,26). The findings from the present study revealed an infection rate of 4.2% within 6 months postoperatively. However, we found no significant differences in the infection rates from 2005 to 2012. This unchanged risk of infection demonstrates the importance of further investigation into the patientrelated risk factors associated with infection. The independent demographic risk factors at 3 and 6 months were largely similar and included age <65 years, low BMI, obesity, tobacco use, inflammatory arthritis, peripheral vascular disease, and hypothyroid disease. Previous studies of the risk factors for deep postoperative infection after TAA have been limited by the small cohorts and have largely been insufficiently powered to identify independent, patient-related risk factors for infection (6,18,23,26). Kessler et al (17) conducted a matched case-control study and reported a number of risk factors that were associated with infection. In agreement with our reported rate, 26 of 408 patients who had undergone primary TAA developed periprosthetic ankle joint infection (4.7%). They also identified a low American Orthopaedic Foot and Ankle Society ankle-hindfoot score, previous surgery at the infection site before arthroplasty, and prolonged operative time as significant risk factors for postoperative PJI (17). Limited by the small, singleinstitutional sample size with age- and sex-matched controls, their study was likely underpowered to identify several other comorbid conditions that were identified in the present study as statistically insignificant (17). The present study has provided a larger, multiinstitutional retrospective review, with adequately powered patient-related demographic and comorbidity data, in particular, a low BMI (OR 3.35) and inflammatory arthritis (OR 2.30). Similarly, Patton et al (23), in a single-institutional review conducted during a 22year period, reported infection in 15 of 619 patients who had undergone primary arthroplasty (2.5%). The significant comorbidities associated with infection included diabetes and hepatitis C (23). Similar to the present study, age and sex were not associated with an increased risk of infection. However, in contrast, Patton et al (23) did not identify tobacco use, inflammatory arthroplasty, or low BMI as predictors of infection. The reported rate of infection after TAA exceeds the rate after total knee and hip replacements (19). The current surgical treatments are very similar and likely incur similar morbidity, such as long-term antibiotic use and the morbidity of subsequent surgery and hospital admission. In contrast to the current data evaluating PJI of the foot and ankle, the risk factors for joint infection after hip and knee arthroplasty have been well studied (19,27). Similar risk factors, such as obesity, young age, diabetes, thyroid disease, and lung disease, have been established (28). Although similar risk factors would be expected, TAA is often performed in a different cohort, with a greater proportion of patients undergoing TAA because of trauma rather than for osteoarthritis (2–4). Owing to the differing cohorts, the increasing incidence of TAA, and the subsequently increased burden of PJI, the findings from the present study add clinically significant data to better inform patients of the associated risks of postoperative infection. Although the present study has provided further insight into the patient-specific risk factors for infection after TAA, it was subject to certain limitations, many inherent to using an administrative
ARTICLE IN PRESS J.M. Cancienne, et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
4
database (29). Other risk factors, such as operative time, surgical approach, tourniquet use, and surgeon technique could not be accounted for owing to the nature of the present retrospective database study. Studies accounting for these factors have often been limited by small cohorts and single-institutional reviews that were inadequately powered to examine the relationship between patient demographic and comorbidity data and the occurrence of infection (17,23). Data limitations, such as electronic miscoding or the use of nonspecific codes for infection, are inherent to the use of a national patient database. In addition, race is not indexed in the standard analytical files of the Medicare database within the PearlDiver data set and therefore could not be included in the present analysis. Therefore, this could have represented a source of confounding that could not be accounted for. To limit the influence of nonspecific codes or infection resulting from an unrelated subsequent procedure, a study endpoint of 6 months was used to maximize the likelihood that any infection that developed was related to the primary TAA. The designated study endpoint might have excluded a small cohort of postoperative infections; however, a period >6 months was not deemed appropriate, because such infections could have resulted from a subsequent procedure or secondary cause. In addition, the present study analyzed the Medicare insured population, which might not be representative of the entire US population. The most common item for Medicare qualification for those aged <65 years is usage of disability insurance for ≥24 consecutive months or the diagnosis of end-stage renal disease. Therefore, evaluation of postoperative infection in a Medicare population might not necessarily have excluded a younger cohort. In conclusion, given the high rate of implant failure and morbidity associated with PJI after TAA, it is imperative that patient-related risk factors for postoperative infection are identified. The present study found that age <65 years, low BMI, obesity, diabetes, inflammatory arthritis, peripheral vascular disease, and hypothyroidism are strongly associated with an increased risk of postoperative infection after TAA. References 1. Glazebrook M, Daniels T, Younger A, Foote CJ, Penner M, Wing K, Lau J, Leighton R, Dunbar M. Comparison of health-related quality of life between patients with end-stage ankle and hip arthrosis. J Bone Joint Surg Am 90:499–505, 2008. 2. Easley ME, Adams SB, Hembree WC, DeOrio JK. Results of total ankle arthroplasty. J Bone Joint Surg Am 93:1455–1468, 2011. 3. Demetriades L, Strauss E, Gallina J. Osteoarthritis of the ankle. Clin Orthop Relat Res 467:28–42, 1998. 4. Saltzman CL, Salamon ML, Blanchard GM, Huff T, Hayes A, Buckwalter JA, Amendola A. Epidemiology of ankle arthritis: report of a consecutive series of 639 patients from a tertiary orthopaedic center. Iowa Orthop J 25:44–46, 2005. 5. Taga I, Shino K, Inoue M, Nakata K, Maeda A. Articular cartilage lesions in ankles with lateral ligament injury: an arthroscopic study. Am J Sports Med 21:120–126– 127, 1993.
6. Schmid T, Krause FG. Conservative treatment of asymmetric ankle osteoarthritis. Foot Ankle Clin 18:437–448, 2013. 7. Saltzman CL, Mann RA, Ahrens JE, Amendola A, Anderson RB, Berlet GC, Brodsky JW, Chou LB, Clanton TO, Deland JT, DeOrio JK, Horton GA, Lee TH, Mann JA, Nunley JA, Thordarson DB, Walling AK, Wapner KL, Coughlin MJ. Prospective controlled trial of STAR total ankle replacement versus ankle fusion: initial results. Foot Ankle Int 30:579–596, 2009. 8. Gougoulias N, Khanna A, Maffulli N. How successful are current ankle replacements? A systematic review of the literature. Clin Orthop Relat Res 468:199–208, 2010. 9. Jiang JJ, Schipper ON, Whyte N, Koh JL, Toolan BC. Comparison of perioperative complications and hospitalization outcomes after ankle arthrodesis versus total ankle arthroplasty from 2002 to 2011. Foot Ankle Int 36:360–368, 2015. 10. Raikin SM, Rasouli MR, Espandar R, Maltenfort MG. Trends in treatment of advanced ankle arthropathy by total ankle replacement or ankle fusion. Foot Ankle Int 35:216–224, 2014. 11. Stavrakis AI, SooHoo NF. Trends in complication rates following ankle arthrodesis and total ankle replacement. J Bone Joint Surg 98:1453–1458, 2016. 12. Santos ALG, Demange MK, Prado MP, Fernandes TD, Giglio PN, Hintermann B. Cartilage lesions and ankle osteoarthrosis: review of the literature and treatment algorithm (English ed.). Rev Bras Ortop 49:565–572, 2014. 13. Pugely AJ, Lu X, Amendola A, Callaghan JJ, Martin CT, Cram P. Trends in the use of total ankle replacement and ankle arthrodesis in the United States Medicare population. Foot Ankle Int 35:207–215, 2014. 14. Daniels TR, Mayich DJ, Penner MJ. Intermediate to long-term outcomes of total ankle replacement with the Scandinavian total ankle replacement (STAR). J Bone Joint Surg Am 97:895–903, 2015. 15. LaMothe J, Seaworth CM, Do HT, Kunas GC, Ellis SJ. Analysis of total ankle arthroplasty survival in the United States using multiple state databases. Foot Ankle Spec 9:336–341, 2016. 16. Glazebrook MA, Arsenault K, Dunbar M. Evidence-based classification of complications in total ankle arthroplasty. Foot Ankle Int 30:945–949, 2009. 17. Kessler B, Sendi P, Graber P, Knupp M, Zwicky L, Hintermann B, Zimmerli W. Risk factors for periprosthetic ankle joint infection: a case-control study. J Bone Joint Surg Am 94:1871–1876, 2012. 18. Myerson MS, Shariff R, Zonno AJ. The management of infection following total ankle replacement: demographics and treatment. Foot Ankle Int 35:855–862, 2014. 19. Zimmerli W, Trampuz A, Ochsner PE. Prosthetic-joint infections. N Engl J Med 351:1645–1654, 2004. 20. Cancienne JM, Brockmeier SF, Gulotta LV, Dines DM, Werner BC. Ambulatory total shoulder arthroplasty. J Bone Joint Surg 99:629–637, 2017. 21. Cancienne JM, Mahon HS, Dempsey IJ, Miller MD, Werner BC. Patient-related risk factors for infection following knee arthroscopy: an analysis of over 700,000 patients from two large databases. Knee 24:594–600, 2017. 22. Mehta HB, Dimou F, Adhikari D, Tamirisa NP, Sieloff E, Williams TP, Kuo Y-F, Riall TS. Comparison of comorbidity scores in predicting surgical outcomes. Med Care 54:180–187, 2016. 23. Patton D, Kiewiet N, Brage M. Infected total ankle arthroplasty: risk factors and treatment options. Foot Ankle Int 36:626–634, 2015. 24. Jastifer JR, Coughlin MJ. Long-term follow-up of mobile bearing total ankle arthroplasty in the United States. Foot Ankle Int 36:143–150, 2015. 25. Mann JA, Mann RA, Horton E. STAR™ ankle: long-term results. Foot Ankle Int 32:473–484, 2011. 26. Doets HC. Total ankle arthroplasty in inflammatory joint disease with use of two mobile-bearing designs. J Bone Joint Surg 88:1272–1284, 2006. 27. Pulido L, Ghanem E, Joshi A, Purtill JJ, Parvizi J. Periprosthetic joint infection: the incidence, timing, and predisposing factors. Clin Orthop Relat Res 466:1710–1715, 2008. 28. Lee QJ, Mak WP, Wong YC. Risk factors for periprosthetic joint infection in total knee arthroplasty. J Orthop Surg (Hong Kong) 23:282–286, 2015. 29. Cancienne JM, Dempsey IJ, Holzgrefe RE, Brockmeier SF, Werner BC. Is hepatitis C infection associated with a higher risk of complications after total shoulder arthroplasty? Clin Orthop Relat Res 474:2664–2669, 2016.
Contents lists available at ScienceDirect
The Journal of Foot & Ankle Surgery j o u r n a l h o m e p a g e : w w w. j f a s . o r g
Original Research
Patient-Related Risk Factors for Periprosthetic Ankle Joint Infection: An Analysis of 6977 Total Ankle Arthroplasties Alyssa Althoff, BS 1, Jourdan M. Cancienne, MD 2, Minton T. Cooper, MD 3, Brian C. Werner, MD 3 1Medical
Student, Medical University of South Carolina, Charleston, SC Resident Physician, Department of Orthopaedic Surgery, University of Virginia, Charlottesville, VA 3 Assistant Professor, Department of Orthopaedic Surgery, University of Virginia, Charlottesville, VA 2
A R T I C L E
I N F O
Level of Clinical Evidence: 3
Keywords: complication periprosthetic joint infection risk factors total ankle arthroplasty
A B S T R A C T
Periprosthetic joint infection (PJI) after total ankle arthroplasty (TAA) is a devastating complication that often results in explantation to resolve the infection. The purpose of the present investigation was to determine the patient-related risk factors for PJI after TAA. A national insurance database was queried for patients undergoing TAA using the Current Procedural Terminology and International Classification of Diseases, ninth revision, procedure codes from 2005 to 2012. Patients undergoing TAA with concomitant fusion procedures or more complex forefoot procedures were excluded. PJI within 6 months was then assessed using the International Classification of Diseases, ninth revision, codes for diagnosis or treatment of postoperative PJI. Multivariate binomial logistic regression analysis was performed to evaluate the patient-related risk factors for PJI. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for each risk factor, with p < .05 considered statistically significant. A total of 6977 patients were included in the present study. Of these 6977 patients, 294 (4%) had a diagnosis of, or had undergone a procedure for, PJI. The independent risk factors for PJI included age <65 years (OR 1.44; p = .036), body mass index <19 kg/m2 (OR 3.35; p = .013), body mass index >30 kg/m2 (OR 1.49; p = .034), tobacco use (OR 1.59; p = .002), diabetes mellitus (OR 1.36; p = .017), inflammatory arthritis (OR 2.38; p < .0001), peripheral vascular disease (OR 1.64; p < .0001), chronic lung disease (OR 1.37; p = .022), and hypothyroidism (OR 1.32; p = .022). The independent patient-related risk factors identified in the present study should help guide physicians and patients considering elective TAA and develop risk stratification algorithms that could decrease the risk of deep, postoperative infection. © 2017 by the American College of Foot and Ankle Surgeons. All rights reserved.
End-stage ankle arthritis is a debilitating condition that affects comfort, function, and health-related quality of life (1). Ankle osteoarthritis is an increasingly prevalent condition, affecting approximately 1% of the population (1). Posttraumatic osteoarthritis accounts for ~12% of symptomatic lower extremity osteoarthritis, correlating to 5.6 million individuals in the United States and increasing (2). Unlike hip or knee osteoarthritis, the development of symptomatic ankle arthritis is predominately posttraumatic, occurring after an ankle sprain, fracture, or ligament instability (3–5). Conservative management of endstage ankle arthritis often involves physical therapy, administration of nonsteroidal antiinflammatory drugs, intraarticular corticosteroid injections, and modification of physical activity and footwear (6).
Financial Disclosure: None reported. Conflict of Interest: None reported. Address correspondence to: Jourdan M. Cancienne, MD, Department of Orthopaedic Surgery, University of Virginia, P.O. Box 800159, Charlottesville, VA 22908. E-mail address: [email protected] (J.M. Cancienne).
However, the surgical treatment of end-stage ankle arthritis is indicated when conservative treatment does not alleviate the symptoms (7). Ankle arthrodesis has previously been widely accepted as the reference standard for surgical treatment of ankle arthritis; however, total ankle arthroplasty (TAA) has gained favor in recent years as an effective alternative that preserves some tibiotalar motion (2,8–11). The goal of TAA is to restore functional range of motion, eliminate pain, and improve quality of life (12). During the past 10 years, the number of TAAs performed has dramatically increased in total volume and hospital usage (11,13), and the indications for surgery have expanded (12,14,15). Subsequent generations of implants since the initial introduction in the 1970s have reportedly decreased both the mechanical failure rate and the risk of infection (7,16). Despite these improvements, periprosthetic ankle joint infection remains a significant complication after TAA, often resulting in implant failure (16). The reported rate of infection ranges from 2.4% to 8.9% (6,17,18), greater than the rate of infection after total knee and hip replacements (19). Thus, it is critical to identify the patient-related risk factors associated with periprosthetic infection after TAA.
1067-2516/$ - see front matter © 2017 by the American College of Foot and Ankle Surgeons. All rights reserved. https://doi.org/10.1053/j.jfas.2017.09.006
ARTICLE IN PRESS 2
J.M. Cancienne, et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
Although published reports have identified some risk factors associated with postoperative infection after TAA within small cohorts (17), to the best of our knowledge, no studies have yet demonstrated the pervasiveness of these factors in a larger population. Accordingly, the purposes of the present study were to use a national database to determine the incidence of postoperative infection and identify independent variables that increase patients’ risk of periprosthetic infection after TAA. Materials and Methods Database The PearlDiver Patient Records Database (Fort Wayne, IN; available at: www.pearldiverinc.com), an insurance claim-based database of patient records, was used for the present study. The database contains data from several different insurers, including both Medicare and private insurers. The 100% Medicare Standard Analytical Files database within PearlDiver was chosen for the present study, because it contains data from the largest number of patients, and most patients undergoing TAA are of Medicare age. The PearlDiver Medicare database contains patient demographic and procedure volume data, among numerous other data for patients with International Classification of Diseases, ninth revision (ICD-9), diagnosis and procedure codes or Current Procedural Terminology (CPT) codes. The database covers patients insured from 2005 through 2012. Overall, the database contains the data from ~100 million patients. All data have been de-identified and anonymous and were thus deemed exempt by our institutional review board. Study Cohort Patients who had undergone TAA were identified using ICD-9 procedure code 81.56 (total ankle replacement) and/or CPT code 27702 (arthroplasty, ankle; with implant, total ankle). Patients with a concomitant revision arthroplasty code (ICD-9 code 81.59 and/or CPT codes 27703 and 27704) were excluded. Covariates and Predictors The covariates included in the present study were identified using specific ICD-9 diagnostic codes. These included age, sex, tobacco use, alcohol abuse, inflammatory arthritis, depression, diabetes mellitus, hyperlipidemia, hypertension, peripheral vascular disease, congestive heart failure, coronary artery disease, chronic kidney disease, chronic lung disease, chronic liver disease, hypercoagulable state, hypothyroidism, preoperative chronic anemia, body mass index (BMI) <19 kg/m2, and current hemodialysis use. The specific BMI was also included in the analysis. Separate ICD-9 diagnostic codes are used for obesity (BMI 30 to 40 kg/m2) and morbid obesity (BMI >40 kg/m2), which thus could be evaluated as 2 separate covariates. This method has been used in previous studies of administrative databases (20,21). Several of these covariates are also included in several comorbidity indexes, including the Charlson comorbidity score, Elixhauser comorbidity score, and the combined comorbidity score (22). In addition, several of these covariates have been evaluated in previous studies examining the risk factors for deep infection after TAA (17,18,23). Infection The study endpoint of interest was infection within 3 to 6 months after TAA. Although most of the current data have focused on the occurrence of early infection after TAA, studies have reported the development of chronic deep infections occurring >6 months postoperatively (17,18,23). After an institutional review of >900 total ankle arthroplasties, Patton et al (23) reported that deep infection after surgery was diagnosed <3 months after the index surgery in 38% of patients, 3 to 12 months in 38% of patients, and >12 months in 24% of patients. Thus, it is important to identify the factors that contribute to infection both within 3 months and >6 months postoperatively. However, many of the infection codes are not specific to TAA. Thus, a period >6 months was not appropriate, because any infection could have resulted from another procedure or event as the interval between the infection and TAA increased. Infection was defined as the diagnosis of periprosthetic infection or septic ankle and/or a procedure for postoperative infection or septic ankle arthritis. The diagnosis codes included ICD-9 codes 998.51, 998.59, 996.66-7, 996.69, 711.07, 711.47, 711.87, and 711.97. The CPT procedure codes included 10180, 20005, 27603, 27607, and 27610. Statistical Analysis The overall infection rate from 2005 to 2012 and the annual infection rate were calculated from the database output. Linear regression analysis was used to examine any
Table 1 Distribution of patient variables and infection rates (N = 6977 patients) Variable
Population
Infection at 3 mo
Infection at 6 mo
Age <65 y Obesity (BMI 30 to 40 kg/m2) Morbid obesity (BMI ≥40 kg/m2) Tobacco use Alcohol abuse Inflammatory arthritis Depression Hypercoagulable disorder Diabetes mellitus Hyperlipidemia Hypertension Peripheral vascular disease Congestive heart failure Coronary artery disease Chronic kidney disease Chronic lung disease Chronic liver disease Current hemodialysis use Hypothyroidism Low BMI (<19 kg/m2) Preoperative anemia
1105 (15.8) 1322 (18.9) 602 (8.6) 980 (14.0) 331 (4.7) 1093 (15.7) 1947 (27.9) 326 (4.7) 2013 (28.9) 5291 (75.8) 5857 (83.9) 793 (11.4) 1196 (17.1) 2401 (34.4) 1172 (16.8) 1505 (21.6) 396 (5.7) 122 (1.7) 2010 (28.8) 39 (0.6) 1020 (14.6)
43 (3.9) 46 (3.5) 21 (3.5) 47 (4.8) 15 (4.5) 41 (3.8) 72 (3.7) 9 (2.8) 66 (3.3) 137 (2.6) 159 (2.7) 50 (6.3) 49 (4.1) 66 (2.7) 44 (3.8) 61 (4.1) 11 (2.8) 2 (1.6) 63 (3.1) 2 (5.1) 45 (4.4)
72 (6.5) 70 (5.3) 37 (6.1) 70 (7.1) 23 (6.9) 74 (6.8) 108 (5.5) 12 (3.7) 124 (6.2) 234 (4.4) 270 (4.6) 82 (10.3) 87 (7.3) 132 (5.5) 83 (7.1) 111 (7.4) 15 (3.8) 7 (5.7) 106 (5.3) 5 (12.8) 62 (6.1)
Data presented as n (%). Abbreviation: BMI, body mass index.
change in the infection rate over time. A descriptive analysis was conducted at 3 and 6 months detailing the distribution of patient variables and infection rates. Multivariable binomial logistic regression analysis was then performed to examine the independent effect of the included covariates on the incidence of infection after TAA. The odds ratios (ORs), 95% confidence intervals (CIs), and p values were calculated for all variables; p < .05 was considered statistically significant for all statistical tests. All statistical testing was performed within the PearlDiver system using R, an open source statistical package (R Foundation, Vienna, Austria; available at: https://www.r-project.org/).
Results A total of 6977 patients who had undergone TAA were included in the present study. During the study period, 294 patients developed infection within 6 months postoperatively, for an overall infection rate of 4.2%. The rate of infection ranged from 3.0% to 4.8% from 2005 to 2012; however, no significant increasing or decreasing trend was found (p = .547; Table 1). The distribution of the covariates and infection rates for the population studied are listed in Table 1. Numerous independent risk factors for infection were identified using logistic regression analysis at both 3 and 6 months. Within 3 months of surgery, the significant risk factors for deep infection included age <65 years (OR 1.5, 95% CI 1.08 to 1.87), obesity (OR 1.48, 95% CI 1.08 to 2.02), low BMI (OR 2.12, 95% CI 1.78 to 2.53), tobacco use (OR 1.79, 95% CI 1.25 to 2.56), inflammatory arthritis (OR 1.42, 95% CI 1.01 to 2.01), peripheral vascular disease (OR 3.14, 95% CI 2.22 to 4.42), hypothyroid disorder (OR 1.27, 95% CI 1.02 to 1.58), and preoperative anemia (OR 1.51, 95% CI 1.07 to 2.13; Table 2). Within 6 months of surgery, the significant risk factors for deep infection included age <65 years (OR 1.61, 95% CI 1.20 to 2.16), low BMI (OR 2.67, 95% CI 1.07 to 6.67), obesity (OR 1.47, 95% CI 1.15 to 1.87), tobacco use (OR 1.44, 95% CI 1.08 to 1.92), diabetes mellitus (OR 1.35, 95% CI 1.05 to 1.73), inflammatory arthritis (OR 1.67, 95% CI 1.28 to 2.18), peripheral vascular disease (OR 2.46, 95% CI 1.87 to 3.22), chronic kidney disease (OR 1.39, 95% CI 1.05 to 1.84), chronic lung disease (OR 1.51, 95% CI 1.16 to 1.96), and hypothyroidism (OR 1.32, 95% CI 1.03 to 1.69; Table 3). Discussion The use of TAA has continued to gain favor over ankle arthrodesis as an effective alternative surgical treatment of end-stage ankle
ARTICLE IN PRESS J.M. Cancienne, et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
Table 2 Covariates for infection after total ankle arthroplasty within 3 months (N = 6977 patients) Variable Demographic data Age <65 y Male sex Procedure in 2005 to 2008 Comorbidities Obesity (BMI >30 kg/m2) Low BMI (<19 kg/m2) Tobacco use Alcohol abuse Inflammatory arthritis Depression Hypercoagulable disorder Diabetes mellitus Hyperlipidemia Hypertension Peripheral vascular disease Congestive heart failure Coronary artery disease Chronic kidney disease Lung disease Liver disease Current hemodialysis use Hypothyroidism Preoperative anemia
OR
95% CI
p Value
1.44 1.07 0.98
1.02 to 2.03 0.77 to 1.47 0.72 to 1.33
.688 .025 .678
1.48 2.12 1.79 1.16 1.42 1.16 0.73 1.41 0.98 1.08 3.14 1.24 0.63 1.17 1.29 0.63 0.32 1.27 1.51
1.08 to 2.02 1.78 to 2.53 1.25 to 2.56 0.66 to 2.05 1.01 to 2.01 0.84 to 1.61 0.38 to 1.42 1.01 to 1.95 0.66 to 1.44 0.68 to 1.72 2.22 to 4.42 0.84 to 1.84 0.44 to 0.91 0.81 to 1.70 0.92 to 1.82 0.34 to 1.19 0.08 to 1.34 1.02 to 1.58 1.07 to 2.13
.017 .009* .001* .600 .046* .359 .350 .033* .900 .747 < .0001* .279 .013* .408 .146 .155 .117 .018* .018*
Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio. * Statistically significant.
arthritis in appropriately selected patients (11). Although first- and second-generation arthroplasty systems were associated with high rates of mechanical failure, more recent advances in surgical technique and implant design have significantly improved the survivorship and longevity of TAA (7,16,24,25). Thus, the focus on periprosthetic joint infections (PJIs) as a cause of implant failure and revision has been increasing (17). However, the limited patient numbers in previous studies have made it difficult to identify patient-specific risk factors
Table 3 Covariates for infection after total ankle arthroplasty within 6 months (N = 6977 patients) Variable Demographic data Age <65 y Male sex Procedure in 2005 to 2008 Comorbidities Obesity (BMI >30 kg/m2) Low BMI (<19 kg/m2) Tobacco use Alcohol abuse Inflammatory arthritis Depression Hypercoagulable disorder Diabetes mellitus Hyperlipidemia Hypertension Peripheral vascular disease Congestive heart failure Coronary artery disease Chronic kidney disease Lung disease Liver disease Current hemodialysis use Hypothyroidism Preoperative anemia
OR
95% CI
p Value
1.61 1.23 0.88
1.20 to 2.16 0.96 to 1.58 0.69 to 1.13
.002 .109 .317
1.47 2.67 1.44 1.09 1.67 0.97 0.51 1.35 0.90 1.33 2.46 1.21 0.94 1.39 1.51 1.46 1.28 1.32 1.05
1.15 to 1.87 1.07 to 6.67 1.08 to 1.92 0.69 to 1.74 1.28 to 2.18 0.75 to 1.25 0.28 to 0.91 1.05 to 1.73 0.66 to 1.23 0.89 to 1.98 1.87 to 3.22 0.89 to 1.62 0.71 to 1.23 1.05 to 1.84 1.16 to 1.96 0.85 to 2.50 0.62 to 2.65 1.03 to 1.69 0.78 to 1.40
.014 .019* .013* .701 < .0001* .825 .022 .012* .516 .158 < .0001* .219 .642 .022* .002* .246 .511 .028 .752
Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio. * Statistically significant.
3
for PJI after TAA. In the present study, we are able to sufficiently power an analysis of >6000 patients who had undergone TAA to identify several significant risk factors for periprosthetic ankle joint infection after TAA. These risk factors included age <65 years, low BMI, obesity, diabetes, inflammatory arthritis, peripheral vascular disease, and hypothyroidism. Surgical site infection is a high-grade complication that, when associated with TAA, has been reported to result in implant failure in 80.6% of cases (16). In studies reporting postoperative infection, the rates have ranged from 2% to 8.5% (6,8,17,18,23,26). The findings from the present study revealed an infection rate of 4.2% within 6 months postoperatively. However, we found no significant differences in the infection rates from 2005 to 2012. This unchanged risk of infection demonstrates the importance of further investigation into the patientrelated risk factors associated with infection. The independent demographic risk factors at 3 and 6 months were largely similar and included age <65 years, low BMI, obesity, tobacco use, inflammatory arthritis, peripheral vascular disease, and hypothyroid disease. Previous studies of the risk factors for deep postoperative infection after TAA have been limited by the small cohorts and have largely been insufficiently powered to identify independent, patient-related risk factors for infection (6,18,23,26). Kessler et al (17) conducted a matched case-control study and reported a number of risk factors that were associated with infection. In agreement with our reported rate, 26 of 408 patients who had undergone primary TAA developed periprosthetic ankle joint infection (4.7%). They also identified a low American Orthopaedic Foot and Ankle Society ankle-hindfoot score, previous surgery at the infection site before arthroplasty, and prolonged operative time as significant risk factors for postoperative PJI (17). Limited by the small, singleinstitutional sample size with age- and sex-matched controls, their study was likely underpowered to identify several other comorbid conditions that were identified in the present study as statistically insignificant (17). The present study has provided a larger, multiinstitutional retrospective review, with adequately powered patient-related demographic and comorbidity data, in particular, a low BMI (OR 3.35) and inflammatory arthritis (OR 2.30). Similarly, Patton et al (23), in a single-institutional review conducted during a 22year period, reported infection in 15 of 619 patients who had undergone primary arthroplasty (2.5%). The significant comorbidities associated with infection included diabetes and hepatitis C (23). Similar to the present study, age and sex were not associated with an increased risk of infection. However, in contrast, Patton et al (23) did not identify tobacco use, inflammatory arthroplasty, or low BMI as predictors of infection. The reported rate of infection after TAA exceeds the rate after total knee and hip replacements (19). The current surgical treatments are very similar and likely incur similar morbidity, such as long-term antibiotic use and the morbidity of subsequent surgery and hospital admission. In contrast to the current data evaluating PJI of the foot and ankle, the risk factors for joint infection after hip and knee arthroplasty have been well studied (19,27). Similar risk factors, such as obesity, young age, diabetes, thyroid disease, and lung disease, have been established (28). Although similar risk factors would be expected, TAA is often performed in a different cohort, with a greater proportion of patients undergoing TAA because of trauma rather than for osteoarthritis (2–4). Owing to the differing cohorts, the increasing incidence of TAA, and the subsequently increased burden of PJI, the findings from the present study add clinically significant data to better inform patients of the associated risks of postoperative infection. Although the present study has provided further insight into the patient-specific risk factors for infection after TAA, it was subject to certain limitations, many inherent to using an administrative
ARTICLE IN PRESS J.M. Cancienne, et al. / The Journal of Foot & Ankle Surgery ■■ (2017) ■■–■■
4
database (29). Other risk factors, such as operative time, surgical approach, tourniquet use, and surgeon technique could not be accounted for owing to the nature of the present retrospective database study. Studies accounting for these factors have often been limited by small cohorts and single-institutional reviews that were inadequately powered to examine the relationship between patient demographic and comorbidity data and the occurrence of infection (17,23). Data limitations, such as electronic miscoding or the use of nonspecific codes for infection, are inherent to the use of a national patient database. In addition, race is not indexed in the standard analytical files of the Medicare database within the PearlDiver data set and therefore could not be included in the present analysis. Therefore, this could have represented a source of confounding that could not be accounted for. To limit the influence of nonspecific codes or infection resulting from an unrelated subsequent procedure, a study endpoint of 6 months was used to maximize the likelihood that any infection that developed was related to the primary TAA. The designated study endpoint might have excluded a small cohort of postoperative infections; however, a period >6 months was not deemed appropriate, because such infections could have resulted from a subsequent procedure or secondary cause. In addition, the present study analyzed the Medicare insured population, which might not be representative of the entire US population. The most common item for Medicare qualification for those aged <65 years is usage of disability insurance for ≥24 consecutive months or the diagnosis of end-stage renal disease. Therefore, evaluation of postoperative infection in a Medicare population might not necessarily have excluded a younger cohort. In conclusion, given the high rate of implant failure and morbidity associated with PJI after TAA, it is imperative that patient-related risk factors for postoperative infection are identified. The present study found that age <65 years, low BMI, obesity, diabetes, inflammatory arthritis, peripheral vascular disease, and hypothyroidism are strongly associated with an increased risk of postoperative infection after TAA. References 1. Glazebrook M, Daniels T, Younger A, Foote CJ, Penner M, Wing K, Lau J, Leighton R, Dunbar M. Comparison of health-related quality of life between patients with end-stage ankle and hip arthrosis. J Bone Joint Surg Am 90:499–505, 2008. 2. Easley ME, Adams SB, Hembree WC, DeOrio JK. Results of total ankle arthroplasty. J Bone Joint Surg Am 93:1455–1468, 2011. 3. Demetriades L, Strauss E, Gallina J. Osteoarthritis of the ankle. Clin Orthop Relat Res 467:28–42, 1998. 4. Saltzman CL, Salamon ML, Blanchard GM, Huff T, Hayes A, Buckwalter JA, Amendola A. Epidemiology of ankle arthritis: report of a consecutive series of 639 patients from a tertiary orthopaedic center. Iowa Orthop J 25:44–46, 2005. 5. Taga I, Shino K, Inoue M, Nakata K, Maeda A. Articular cartilage lesions in ankles with lateral ligament injury: an arthroscopic study. Am J Sports Med 21:120–126– 127, 1993.
6. Schmid T, Krause FG. Conservative treatment of asymmetric ankle osteoarthritis. Foot Ankle Clin 18:437–448, 2013. 7. Saltzman CL, Mann RA, Ahrens JE, Amendola A, Anderson RB, Berlet GC, Brodsky JW, Chou LB, Clanton TO, Deland JT, DeOrio JK, Horton GA, Lee TH, Mann JA, Nunley JA, Thordarson DB, Walling AK, Wapner KL, Coughlin MJ. Prospective controlled trial of STAR total ankle replacement versus ankle fusion: initial results. Foot Ankle Int 30:579–596, 2009. 8. Gougoulias N, Khanna A, Maffulli N. How successful are current ankle replacements? A systematic review of the literature. Clin Orthop Relat Res 468:199–208, 2010. 9. Jiang JJ, Schipper ON, Whyte N, Koh JL, Toolan BC. Comparison of perioperative complications and hospitalization outcomes after ankle arthrodesis versus total ankle arthroplasty from 2002 to 2011. Foot Ankle Int 36:360–368, 2015. 10. Raikin SM, Rasouli MR, Espandar R, Maltenfort MG. Trends in treatment of advanced ankle arthropathy by total ankle replacement or ankle fusion. Foot Ankle Int 35:216–224, 2014. 11. Stavrakis AI, SooHoo NF. Trends in complication rates following ankle arthrodesis and total ankle replacement. J Bone Joint Surg 98:1453–1458, 2016. 12. Santos ALG, Demange MK, Prado MP, Fernandes TD, Giglio PN, Hintermann B. Cartilage lesions and ankle osteoarthrosis: review of the literature and treatment algorithm (English ed.). Rev Bras Ortop 49:565–572, 2014. 13. Pugely AJ, Lu X, Amendola A, Callaghan JJ, Martin CT, Cram P. Trends in the use of total ankle replacement and ankle arthrodesis in the United States Medicare population. Foot Ankle Int 35:207–215, 2014. 14. Daniels TR, Mayich DJ, Penner MJ. Intermediate to long-term outcomes of total ankle replacement with the Scandinavian total ankle replacement (STAR). J Bone Joint Surg Am 97:895–903, 2015. 15. LaMothe J, Seaworth CM, Do HT, Kunas GC, Ellis SJ. Analysis of total ankle arthroplasty survival in the United States using multiple state databases. Foot Ankle Spec 9:336–341, 2016. 16. Glazebrook MA, Arsenault K, Dunbar M. Evidence-based classification of complications in total ankle arthroplasty. Foot Ankle Int 30:945–949, 2009. 17. Kessler B, Sendi P, Graber P, Knupp M, Zwicky L, Hintermann B, Zimmerli W. Risk factors for periprosthetic ankle joint infection: a case-control study. J Bone Joint Surg Am 94:1871–1876, 2012. 18. Myerson MS, Shariff R, Zonno AJ. The management of infection following total ankle replacement: demographics and treatment. Foot Ankle Int 35:855–862, 2014. 19. Zimmerli W, Trampuz A, Ochsner PE. Prosthetic-joint infections. N Engl J Med 351:1645–1654, 2004. 20. Cancienne JM, Brockmeier SF, Gulotta LV, Dines DM, Werner BC. Ambulatory total shoulder arthroplasty. J Bone Joint Surg 99:629–637, 2017. 21. Cancienne JM, Mahon HS, Dempsey IJ, Miller MD, Werner BC. Patient-related risk factors for infection following knee arthroscopy: an analysis of over 700,000 patients from two large databases. Knee 24:594–600, 2017. 22. Mehta HB, Dimou F, Adhikari D, Tamirisa NP, Sieloff E, Williams TP, Kuo Y-F, Riall TS. Comparison of comorbidity scores in predicting surgical outcomes. Med Care 54:180–187, 2016. 23. Patton D, Kiewiet N, Brage M. Infected total ankle arthroplasty: risk factors and treatment options. Foot Ankle Int 36:626–634, 2015. 24. Jastifer JR, Coughlin MJ. Long-term follow-up of mobile bearing total ankle arthroplasty in the United States. Foot Ankle Int 36:143–150, 2015. 25. Mann JA, Mann RA, Horton E. STAR™ ankle: long-term results. Foot Ankle Int 32:473–484, 2011. 26. Doets HC. Total ankle arthroplasty in inflammatory joint disease with use of two mobile-bearing designs. J Bone Joint Surg 88:1272–1284, 2006. 27. Pulido L, Ghanem E, Joshi A, Purtill JJ, Parvizi J. Periprosthetic joint infection: the incidence, timing, and predisposing factors. Clin Orthop Relat Res 466:1710–1715, 2008. 28. Lee QJ, Mak WP, Wong YC. Risk factors for periprosthetic joint infection in total knee arthroplasty. J Orthop Surg (Hong Kong) 23:282–286, 2015. 29. Cancienne JM, Dempsey IJ, Holzgrefe RE, Brockmeier SF, Werner BC. Is hepatitis C infection associated with a higher risk of complications after total shoulder arthroplasty? Clin Orthop Relat Res 474:2664–2669, 2016.