Abstract
Background
Patients with diabetes are known to be at greater risk for complications after arthroplasty than are patients without diabetes. However, we do not know whether there are important differences in the risk of perioperative complications between patients with diabetes who are insulin-dependent (Type 1 or 2) and those who are not insulin-dependent.
Questions/purposes
The purpose of our study was to compare (1) medical complications (including death), (2) surgical complications, and (3) readmissions within 30 days between patients with insulin-dependent and noninsulin-dependent diabetes, and with patients who do not have diabetes.
Methods
A total of 43,299 patients undergoing THA or TKA between 2005 and 2011 were selected from the American College of Surgeon’s National Surgical Quality Improvement Program’s (ACS-NSQIP®) database. Generalized linear models were used to assess the relationship between diabetes status and outcomes (no diabetes [n = 36,574], insulin dependent [n = 1552], and noninsulin dependent [n = 5173]). Multivariate models were established adjusting for confounders including age, sex, race, BMI, smoking, steroid use, hypertension, chronic obstructive pulmonary disease, and anesthesia type. Post hoc comparisons between patient groups were made using a Bonferroni correction.
Results
Patients who were insulin dependent had increased odds of experiencing a medical complication (OR, 1.6; 95% CI, 1.2–2.0; p < 0.001), as did patients who were noninsulin dependent (OR, 1.2; 95% CI, 1.1–1.4; p< 0.001). An increased likelihood of 30-day mortality was found only for patients who were insulin dependent (OR, 3.74; 95% CI, 1.6–8.5; p = 0.007). However, neither diabetic state was associated with surgical complications. Finally, readmission was found to be independently associated with insulin-dependent diabetes (OR, 1.6; 95% CI, 1.1–2.1; p = 0.023).
Conclusions
Patients with insulin-dependent diabetes are most likely to have a medical complication or be readmitted within 30 days after total joint replacement. However, patients who are insulin dependent or noninsulin dependent are no more likely than patients without diabetes to have a surgical complication. Physicians and hospitals should keep these issues in mind when counseling patients and generating risk-adjusted outcome reports.
Level of Evidence
Level III, therapeutic study. See the Instructions for Authors for a complete description of levels of evidence.
Electronic supplementary material
The online version of this article (doi:10.1007/s11999-014-3891-4) contains supplementary material, which is available to authorized users.
Introduction
The prevalence of diabetes mellitus is high in older patients, with 26.9% of US residents older than 65 years with the disease [3]. In addition, an estimated 79 million Americans 20 years or older currently are considered prediabetic [3]. Diabetes has been implicated as a risk factor for increased rates of surgical site infections, urinary tract infections, pneumonia, and stroke after lower extremity arthroplasty [2, 8–14, 17]. However, other studies challenge the role of diabetes as a risk factor, even for classic associations such as diabetes and wound infections [1, 4, 5].
Moreover, diabetes is a heterogeneous state and can be classified using many parameters. Data suggest that perioperative hemoglobin A1c and serum glucose levels may better predict infection after arthroplasty [8, 9, 12]. Diabetic state also may be classified using the method of diabetic control. For example, Iorio et al. [8] and Meding et al. [13] reported that patients with insulin-dependent diabetes had higher rates of infections after arthroplasty compared with patients with noninsulin-dependent diabetes. Thus, patients with insulin-dependent diabetes and patients with noninsulin-dependent diabetes may differ in terms of their surgical risk profiles, but this relationship remains poorly characterized.
We therefore used a large, multiinstitutional patient registry to determine whether distinct codeable, diabetic disease states (insulin-dependent diabetes [including Types 1 and 2] and noninsulin dependent diabetes) portend different risks for surgical and medical complications after TKA and THA. We also sought to determine whether these two groups have different risks for 30-day readmission when compared with patients without diabetes.
Patients and Methods
The design and statistical methods of the American College of Surgeon’s National Surgical Quality Improvement Program (ACS-NSQIP®) was described by Henderson and Daley [7]. Briefly, the ACS-NSQIP® collects demographic information, intraoperative variables, and 30-day postoperative complications for patients who underwent major surgical procedures. These data are obtained from more than 250 sites and are retrieved directly from patients’ medical records. Specifically, the merged Participant Use Data Files from 2005 to 2011 contain 135 variables from 1,777,035 deidentified cases. Of those, a total of 46,913 patients underwent either primary THA (CPT code, 27130) or TKA (CPT code, 27447); 43,299 had osteoarthritis (ICD-9 code 715). Other patients were excluded from this analysis.
As most of the demographic variables are categorical, they are presented as counts and percentages and compared across diabetic status groups using chi-square tests. BMI was normally distributed, and therefore was compared using ANOVA tests and presented as mean and SD. Total Work Relative Value Unit (RVU) was calculated based on the RVU of the primary procedure and any concurrent or other procedure listed. Age and total RVU were not normally distributed, so distributions were compared using Wilcoxon rank sum tests and are presented as median and 25th and 75th percentiles.
Diabetic status was among the variables collected and was grouped into three categories: insulin, noninsulin, and nondiabetic. The insulin category included any patients with diabetes currently using insulin (Types 1 and 2), whereas the noninsulin category included patients with diabetes using only noninsulin medications to control their disease. Age, sex, and race differed slightly between the insulin-dependent, noninsulin-dependent, and nondiabetic cohorts (Table 1). Mean BMI was highest among patients with insulin-dependent diabetes mellitus (35 ± 6 insulin dependent versus 34 ± 6 noninsulin dependent versus 31± 6 nondiabetic; p < 0.001). The proportion of patients classified as American Society of Anesthesiologists’ (ASA) Class 3 provides a succinct summary of how the comorbidity burden differed among the cohorts (77.5% with insulin-dependent diabetes mellitus versus 66.4% with noninsulin-dependent diabetes mellitus versus 40.0% nondiabetic; p < 0.001). The same trends were seen in the knee (Appendix 1. Supplemental material is available with the online version of CORR.) and hip (Appendix 2. Supplemental material is available with the online version of CORR.) subgroup analyses.
Table 1.
Preoperative characteristics
| Characteristic | Without diabetes | Noninsulin dependent | Insulin dependent | p value |
|---|---|---|---|---|
| Number of patients | 36,574 | 5173 | 1552 | – |
| Age (years) (median [25th,75th percentiles]) | 66 (59, 74) | 68 (62, 75)† | 67 (61, 74)‡ | < 0.001 |
| BMI | 30.9 (6.2) | 34.0 (6.2)† | 35.2 (6.3)†,‡ | < 0.001 |
| Sex (male) | 14,041 (38.6%) | 2157 (41.9%)† | 674 (43.5%)† | < 0.001 |
| Race | † | †,‡ | < 0.001 | |
| White | 30,264 (83.7%) | 4033 (78.8%) | 1234 (80.6%) | |
| Black | 2153 (6.0%) | 444 (8.7%) | 160 (10.5%) | |
| Asian | 486 (1.3%) | 141 (2.8%) | 25 (1.6%) | |
| Hispanic | 123 (0.3%) | 23 (0.5%) | 9 (0.6%) | |
| Other/unknown | 3148 (8.7%) | 475 (9.3%) | 103 (6.7%) | |
| Active smoker | 3653 (10.0%) | 438 (8.5%)† | 143 (9.2%) | 0.002 |
| Outpatient | 294 (0.8%) | 43 (0.8%) | 18 (1.2%) | 0.312 |
| Steroid use | 858 (2.4%) | 104 (2.0%) | 62 (4.0%)†,‡ | < 0.001 |
| Comorbidities | ||||
| Dyspnea | 2606 (7.1%) | 585 (11.3%)† | 228 (14.7%)†,‡ | < 0.001 |
| Hypertension | 22415 (61.3%) | 4472 (86.5%)† | 1389 (89.5%)†,‡ | < 0.001 |
| Chronic obstructive pulmonary disease | 1233 (3.4%) | 240 (4.6%)† | 106 (6.8%)†,‡ | < 0.001 |
| Congestive heart failure < 30 days | 55 (0.2%) | 12 (0.2%) | 10 (0.6%)†,‡ | < 0.001 |
| Bleeding disorders | 884 (2.4%) | 161 (3.1%)† | 56 (3.6%)† | < 0.001 |
| Previous percutaneous coronary Intervention | 1347 (5.5%) | 359 (10.0%)† | 173 (16.5%)†,‡ | < 0.001 |
| Previous cardiac surgery | 974 (4.0%) | 246 (6.8%)† | 126 (12.0%)†,‡ | < 0.001 |
| Previous stroke with neurologic deficit | 236 (1.0%) | 46 (1.3%) | 33 (3.1%)†,‡ | < 0.001 |
| Previous stroke without neurologic deficit | 402 (1.6%) | 96 (2.7%)† | 44 (4.2%)†,‡ | < 0.001 |
| ASA class | † | †,‡ | < 0.001 | |
| 1–2 | 21,453 (58.7%) | 1602 (31.0%) | 257 (16.6%) | |
| 3 | 14,608 (40.0%) | 3425 (66.4%) | 1203 (77.5%) | |
| 4–5 | 461 (1.3%) | 135 (2.6%) | 92 (5.9%) | |
| Total relative value units (median [25th,75th percentiles]) | 23.0 (21.8, 23.3) | 23.3 (21.8, 23.3)† | 23.3 (23.0, 23.3)† | < 0.001 |
†Post hoc test, significantly different from patients without diabetes; ‡post hoc test, significantly different from patients with noninsulin-dependent diabetes mellitus.
Our primary comparisons included rates of medical complications (including death), surgical complications, and readmission. Endpoints of interest were compared across diabetic status groups using generalized linear mixed models, using logit, or log links, depending on the distribution of the outcome (logistic regression for binary outcomes or negative binomial for total length of hospital stay). Unadjusted analyses are presented as multivariable analyses that were adjusted for potential confounders (ASA classification, age, sex, race, BMI, hypertension, steroid use, history of chronic obstructive pulmonary disease, preoperative laboratory values, anesthetic type, and total RVU). Odds ratios, least-squared estimated means, and hazard ratios are presented with corresponding Bonferroni adjustments for post hoc comparisons across diabetic status groups. Secondary analyses stratified all analyses by procedure (THA or TKA). All analyses were run at a type I error rate of 5%, and all analyses were run in SASv9.2 (Cary, NC, USA).
Results
Patients with diabetes were more likely to have medical complications after arthroplasty than patients without diabetes, and patients with insulin-dependent diabetes were at greater risk for medical complications than those with noninsulin-dependent diabetes. In the combined hip and knee analysis, the baseline rate for a medical complication changed with diabetic status (7.0% insulin dependent versus 5.0% noninsulin dependent versus 3.4% nondiabetic; p < 0.001) (Table 2). Insulin-dependent diabetes mellitus was associated with an increased likelihood of experiencing a medical complication when compared with noninsulin-dependent diabetes and no diabetes (OR, 1.6, 95% CI, 1.2–2.1). In addition, insulin-dependent diabetes mellitus was associated with increased 30-day mortality (OR, 3.7; 95% CI, 1.4–10.3). Regarding specific complications, patients with insulin-dependent diabetes mellitus had greater odds of experiencing a cerebrovascular accident or pneumonia when compared with patients without diabetes (OR, 4.2; 95% CI, 1.3–13.4; and OR, 2.3; 95% CI, 1.1–4.9, respectively). Finally, noninsulin-dependent diabetes mellitus conferred an increased likelihood of medical complications, but to a smaller degree than insulin-dependent diabetes mellitus (OR, 1.22; 95% CI, 1.01–1.48). Secondary subgroup analyses (Appendices 3 and 4. Supplemental material is available with the online version of CORR.) revealed that insulin dependence was strongly associated with medical complications and mortality especially after THA (OR, 1.9; 95% CI, 1.2–3.2; and OR, 12.2; 95% CI, 3.1–47.1, respectively).
Table 2.
Outcomes after lower extremity arthroplasty
| Outcomes | Unadjusted rates and odds ratios | Adjusted odds ratios* | ||||||
|---|---|---|---|---|---|---|---|---|
| Without diabetes | Noninsulin-dependent diabetes mellitus | Insulin-dependent diabetes mellitus | p value | Without diabetes | Noninsulin-dependent diabetes mellitus | Insulin-dependent diabetes mellitus | p value | |
| Overall complication rate | 4.5% | 6.3%† 1.43 (1.23, 1.67) | 8.8%†,‡ 2.05 (1.64, 2.57) | < 0.001 | – | 1.16 (0.98, 1.38) | 1.48 (1.15, 1.90)† | < 0.001 |
| Surgical complications | 1.2% | 1.5% 1.28 (0.95, 1.72) | 2.3%† 1.92 (1.26, 2.94) | < 0.001 | – | 0.93 (0.67, 1.31) | 1.27 (0.78, 2.07) | 0.395 |
| Superficial wound | 0.7% | 1.0% 1.43 (1.00, 2.05) | 1.3%† 1.77 (1.01, 3.09) | 0.005 | – | 1.16 (0.77, 1.73) | 1.21 (0.63, 2.31) | 0.588 |
| Deep incision/organ space | 0.3% | 0.4% 1.29 (0.73, 2.28) | 0.6% 2.06 (0.93, 4.54) | 0.065 | – | 0.86 (0.44, 1.68) | 1.40 (0.59, 3.36) | 0.507 |
| Medical complications** | 3.4% | 5.0%† 1.47 (1.24, 1.74) | 7.0%†,‡ 2.13 (1.66, 2.72) | < 0.001 | – | 1.22 (1.01, 1.48)† | 1.56 (1.18, 2.06) † | < 0.001 |
| Pneumonia | 0.3 | 0.4 1.29 (0.71, 2.35) | 0.9†,‡ 3.19 (1.61, 6.33) | < 0.001 | – | 0.81 (0.40, 1.65) | 2.32 (1.11, 4.85) †,‡ | 0.011 |
| Unplanned intubation | 0.2% | 0.3%† 2.01 (1.04, 3.88) | 0.7%† 4.34 (1.98, 9.55) | < 0.001 | – | 1.50 (0.72, 3.11) | 2.63 (1.08, 6.42) † | 0.027 |
| On ventilation > 48 hours | 0.1% | 0.1% 1.36 (0.42, 4.38) | 0.3%† 4.54 (1.41, 14.65) | 0.003 | – | 0.98 (0.29, 3.32) | 2.01 (0.51, 8.01) | 0.456 |
| Renal insufficiency | 0.1% | 0.3%† 3.20 (1.48, 6.92) | 0.6%† 7.65 (3.19, 18.30) | < 0.001 | – | 1.67 (0.71, 3.94) | 3.33 (1.29, 8.60) † | 0.009 |
| Urinary tract infection | 1.2% | 1.9%† 1.56 (1.19, 2.04) | 2.4%† 2.07 (1.37, 3.11) | < 0.001 | – | 1.33 (0.99, 1.80) | 1.53 (0.97, 2.42) | 0.014 |
| Stroke/cerebrovascular accident | 0.1% | 0.1% 1.18 (0.44, 3.13) | 0.5%† 3.94 (1.48, 10.49) | 0.004 | – | 0.97 (0.30, 3.15) | 4.23 (1.34, 13.37) †,‡ | 0.009 |
| Cardiac arrest | 0.1% | 0.2% 2.36 (0.94, 5.93) | 0.3% 3.50 (0.97, 12.63) | 0.012 | – | 1.68 (0.53, 5.38) | 2.38 (0.50,11.40) | 0.309 |
| Myocardial infarction | 0.2% | 0.5%† 3.11 (1.80, 5.35) | 0.8%† 4.45 (2.09, 9.46) | < 0.001 | – | 3.07 (1.69, 5.57) | 3.76 (1.60, 8.84) | < 0.001 |
| Sepsis or sepsis shock | 0.4% | 0.8%† 2.09 (1.37, 3.18) | 1.4%† 3.49 (1.99, 6.14) | < 0.001 | – | 1.61 (1.02, 2.56) † | 1.94 (1.02, 3.69) † | 0.007 |
| Return to operating room | 1.4% | 1.7% 1.17 (0.88, 1.54) | 2.1% 1.44 (0.92, 2.23) | 0.075 | – | 0.85 (0.62, 1.18) | 0.97 (0.58, 1.60) | 0.501 |
| Total length of stay (mean days, 95% CI) | 3.4 (3.4, 3.5) | 3.7 (3.6, 3.7)† | 4.0 (3.9, 4.1) †,‡ | < 0.001 | 4.3 (4.1, 4.4) | 4.4 (4.2, 4.5) | 4.6 (4.4, 4.8) †,‡ | < 0.001 |
| Time from operating room to discharge (hazard ratio) | – | 0.89 (0.87, 0.92)† | 0.81 (0.77, 0.86) †,‡ | < 0.001 | – | 0.97 (0.94, 1.00) | 0.91 (0.86, 0.96) † | 0.001 |
| Readmission | 3.9% | 4.9% 1.28 (0.99, 1.66) | 8.0%†,‡ 2.15 (1.51, 3.06) | < 0.001 | – | 1.01 (0.75, 1.34) | 1.56 (1.06, 2.30) † | 0.023 |
| 30-day mortality rate | 0.1% | 0.3% 1.92 (0.91, 4.06) | 0.6%† 4.94 (2.14, 11.37) | < 0.001 | – | 1.50 (0.62, 3.63) | 3.74 (1.37,10.25) † | 0.007 |
*Adjusted for American Society of Anesthesiologists class, smoking status, steroid use, hypertension medications, history of chronic obstructive pulmonary disease, creatinine, platelet counts, white blood cell counts, anesthesia type (general or not), race, age, sex, BMI, and total relative value units; **specific medical complication rates with differences between diabetes study groups < 0.01; †post hoc test significantly different from that of patients without diabetes; ‡post hoc test significantly different from that of patients with noninsulin-dependent diabetes.
In contrast, neither patients with insulin-dependent nor noninsulin-dependent diabetes mellitus had greater odds of a surgical complication (insulin dependent: OR, 1.3, 95% CI, 0.8–2.1; noninsulin-dependent: OR, 1.2, 95% CI, 0.98–1.4) compared with patients without diabetes. However, secondary subgroup analyses (Appendices 3 and 4. Supplemental material is available with the online version of CORR.) showed that patients with insulin-dependent diabetes had an increased risk of deep infection after TKA compared with the cohorts with noninsulin-dependent diabetes and without diabetes (OR, 3.1; 95% CI, 1.1–8.5). Additionally, patients with insulin-dependent diabetes showed an increased odds of superficial wound infection after THA (OR, 3.4; 95% CI, 4.0–8.1) compared with patients without diabetes.
Patients with insulin-dependent diabetes were more likely to be readmitted within 30 days, while patients with noninsulin-dependent diabetes had the same likelihood of readmission as patients without diabetes (Table 2). In the combined analysis, baseline readmission rates were greatest for patients with insulin-dependent diabetes mellitus (8.0%; p < 0.001). Insulin-dependent diabetes mellitus showed an independent association with readmission in the combined group (OR, 1.6; 95% CI, 1.1–2.3). However, noninsulin-dependent diabetes was not associated with readmission (OR, 1.0; 95% CI, 0.8–1.3). In the TKA subgroup analysis, readmission was not associated with either form of diabetes (insulin-dependent diabetes mellitus: OR, 1.37; 95% CI, 0.84–2.22; noninsulin-dependent diabetes mellitus: OR, 0.98; 95% CI, 0.69–1.39).
Discussion
Clinicians generally believe that diabetes increases the risk of perioperative complications in patients having total joint arthroplasties, but the exact risk is not well defined [1, 2, 4, 5, 8–14, 17]. The effect of diabetic disease state on risk of complication after THA and TKA is less well understood [8, 13]. We therefore sought to use a large, multiinstitutional patient registry to determine whether patients with insulin-dependent diabetes and noninsulin-dependent diabetes have different likelihoods for medical and surgical complications after TKA and THA (including death). We also investigated whether patients with diabetes are more likely to experience 30-day readmissions.
There were limitations to our study. Although the ASC-NSQIP® is a clinical database, it does not necessarily capture details of each patient’s clinical course. Furthermore, some variables are more subject to coding problems than others, a consideration that must be made in judging the accuracy of our conclusions. For example, superficial wound infections are difficult to diagnose clinically, while death or readmission is an easy variable to define and track. The second limitation of our study is that certain variables were not available to us through the ACS-NSQIP®. For example, perioperative serum hemoglobin A1c and glucose levels, which reflect varying levels of diabetic control, are not available in the ASC-NSQIP® database. Furthermore, the duration of disease, presence of peripheral neuropathy, previous foot and/or leg infection, or other late manifestations of diabetes are unavailable in the database. However, we believe the practical applications of our findings are worthwhile, especially in the development of risk assessment tools for patients undergoing surgery. Third, we were unable to control for factors such as hospital type, surgeon volume, or insurance status, owing to the deidentified nature of the registry. Although these factors may influence patient outcomes, we believe that the variety of participating hospitals in the ACS-NSQIP® leads to generalizable findings. Fourth, the cohorts were shown to have significant differences in BMI, severe chronic obstructive pulmonary disease, congestive heart failure, steroid use, and hypertension—factors that may influence rate of complications and wound healing [5, 14]. We also noted differences in age, sex, and race between our groups. Although we used logistic regression to control for these factors, statistical techniques may be limited in their ability to dissect the complex interplay of patient comorbidities in their effect on changing complication risk. Fifth, the ACS-NSQIP® database has a 30-day followup period, which affects some findings more than others. We believe that the trends seen in wound complication rates in patients with insulin-dependent diabetes mellitus are good approximations of the increase in postoperative risk of infection. However, reoperations most likely would be greater if followup were longer. Finally, second-order comparisons made in studies with a large number of simultaneous comparisons must be considered preliminary. Although we attempted to control for multiple comparisons using Bonferroni corrections, differences in specific surgical complications (for example, surgical site infections) and medical complications (for example, pneumonia) will require confirmation by future studies.
We found that classifying patients with diabetes by insulin dependence results in two separate risk profiles, with patients with insulin-dependent diabetes at the greatest risk for medical complications. The association with complications is more pronounced after THA, suggesting that diabetic disease state matters even more in these patients. Previous studies [9, 12] used controlled versus uncontrolled diabetes as a measure of diabetic disease state when assessing risk of medical complications after lower extremity arthroplasty. Marchant et al. [12] found that uncontrolled diabetes was associated with more stroke, urinary tract infections, and death when compared with controlled diabetes. Given that hemoglobin A1c levels of 7.0 are not always feasible for patients with diabetes undergoing total joint arthroplasty [6], we hope that this novel way of looking at diabetic disease state may provide a complementary mode of risk stratification before joint replacement.
Previous studies have classified patients with diabetes by insulin status when assessing their risk of surgical infection. In one series, Meding et al. [13] reported a higher rate of deep infections after knee arthroplasty in patients who were insulin dependent compared with patients who were noninsulin dependent. Iorio et al. [8] reported an infection rate of 11.1% in the cohort that was insulin dependent and 1.4% in the cohort that was noninsulin dependent who were undergoing lower extremity arthroplasty. We found that neither noninsulin-dependent nor insulin-dependent diabetes mellitus was independently associated with overall surgical complication risk. However, examining specific complications in the subgroups provided preliminary evidence that patients with insulin-dependent diabetes mellitus were more likely to experience a superficial infection after THA or a deep infection after TKA. If the followup time were greater in our study, our rates of observed infection might more closely approximate those reported by Iorio et al. [8].
Insulin-dependent diabetes was associated with a higher risk of readmission during the study period. The most common reasons for readmission after arthroplasty relate to complications after surgery [15, 16, 18], so higher complication rates in patients who are insulin dependent would logically explain our results. Although a secondary finding, we noted that neither noninsulin-dependent nor insulin-dependent diabetes was associated with increased rates of reoperation. However, increasing followup time might influence this observation.
The prevalence of diabetes in the population undergoing arthroplasty is high and may increase in the future. We showed that adverse outcomes are greatest for patients with diabetes whose disease is either severe or chronic enough to merit insulin use. We believe that our findings support the simple and pragmatic use of a diabetic glucose control method as a variable for risk in the preoperative assessment of patients with diabetes who are undergoing arthroplasty. Our findings help clinicians better define perioperative risk when counseling patients, and provide a rationale for using simple codeable data when generating risk adjustment schemes for public reporting and reimbursement.
Electronic supplementary material
Footnotes
One of the authors certifies that he (DM) has or may receive payments or benefits, during the study period, an amount of USD 100,001-USD 1,000,000 from Biomet (Warsaw, IN, USA) and USD 0 from Medacta (Castel San Pietro, Switzerland).
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research editors and board members are on file with the publication and can be viewed on request.
Each author certifies that his or her institution approved or waived approval for the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research.
References
- 1.Adams AL, Paxton EW, Wang JQ, Johnson ES, Bayliss EA, Ferrara A, Nakasato C, Bini SA, Namba RS. Surgical outcomes of total knee replacement according to diabetes status and glycemic control, 2001 to 2009. J Bone Joint Surg Am. 2013;95:481–487. doi: 10.2106/JBJS.L.00109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bolognesi MP, Marchant MH, Jr, Viens NA, Cook C, Pietrobon R, Vail TP. The impact of diabetes on perioperative patient outcomes after total hip and total knee arthroplasty in the United States. J Arthroplasty. 2008;23(6 suppl 1):92–98. doi: 10.1016/j.arth.2008.05.012. [DOI] [PubMed] [Google Scholar]
- 3.Centers for Disease Control and Prevention. National diabetes fact sheet: national estimates and general information on diabetes and prediabetes in the United States, 2011. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention. Available at: http://www.cdc.gov/diabetes/pubs/pdf/ndfs_2011.pdf . Accessed November 11, 2011.
- 4.Cordero-Ampuero J, de Dios M. What are the risk factors for infection in hemiarthroplasties and total hip arthroplasties? Clin Orthop Relat Res. 2010;468:3268–3277. doi: 10.1007/s11999-010-1411-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Dowsey MM, Choong PF. Obese diabetic patients are at substantial risk for deep infection after primary TKA. Clin Orthop Relat Res. 2009;467:1577–1581. doi: 10.1007/s11999-008-0551-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Giori NJ, Ellerbe LS, Bowe T, Gupta S, Harris AH. Many diabetic total joint arthroplasty candidates are unable to achieve a preoperative hemoglobin A1c goal of 7% or less. J Bone Joint Surg Am. 2014;96:500–504. doi: 10.2106/JBJS.L.01631. [DOI] [PubMed] [Google Scholar]
- 7.Henderson WG, Daley J. Design and statistical methodology of the National Surgical Quality Improvement Program: why is it what it is? Am J Surg. 2009;198(5 suppl):S19–S27. doi: 10.1016/j.amjsurg.2009.07.025. [DOI] [PubMed] [Google Scholar]
- 8.Iorio R, Williams KM, Marcantonio AJ, Specht LM, Tilzey JF, Healy WL. Diabetes mellitus, hemoglobin A1C, and the incidence of total joint arthroplasty infection. J Arthroplasty. 2012;27(726–729):e721. doi: 10.1016/j.arth.2011.09.013. [DOI] [PubMed] [Google Scholar]
- 9.Jämsen E, Nevalainen P, Eskelinen A, Huotari K, Kalliovalkama J, Moilanen T. Obesity, diabetes, and preoperative hyperglycemia as predictors of periprosthetic joint infection: a single-center analysis of 7181 primary hip and knee replacements for osteoarthritis. J Bone Joint Surg Am. 2012;94:e101. doi: 10.2106/JBJS.J.01935. [DOI] [PubMed] [Google Scholar]
- 10.Jämsen E, Varonen M, Huhtala H, Lehto MU, Lumio J, Konttinen YT, Moilanen T. Incidence of prosthetic joint infections after primary knee arthroplasty. J Arthroplasty. 2010;25:87–92. doi: 10.1016/j.arth.2008.10.013. [DOI] [PubMed] [Google Scholar]
- 11.Lai K, Bohm ER, Burnell C, Hedden DR. Presence of medical comorbidities in patients with infected primary hip or knee arthroplasties. J Arthroplasty. 2007;22:651–656. doi: 10.1016/j.arth.2006.09.002. [DOI] [PubMed] [Google Scholar]
- 12.Marchant MH, Jr, Viens NA, Cook C, Vail TP, Bolognesi MP. The impact of glycemic control and diabetes mellitus on perioperative outcomes after total joint arthroplasty. J Bone Joint Surg Am. 2009;91:1621–1629. doi: 10.2106/JBJS.H.00116. [DOI] [PubMed] [Google Scholar]
- 13.Meding JB, Reddleman K, Keating ME, Klay A, Ritter MA, Faris PM, Berend ME. Total knee replacement in patients with diabetes mellitus. Clin Orthop Relat Res. 2003;416:208–216. doi: 10.1097/01.blo.0000093002.90435.56. [DOI] [PubMed] [Google Scholar]
- 14.Namba RS, Inacio MC, Paxton EW. Risk factors associated with deep surgical site infections after primary total knee arthroplasty: an analysis of 56,216 knees. J Bone Joint Surg Am. 2013;95:775–782. doi: 10.2106/JBJS.L.00211. [DOI] [PubMed] [Google Scholar]
- 15.Saucedo JM, Marecek GS, Wanke TR, Lee J, Stulberg SD, Puri L. Understanding readmission after primary total hip and knee arthroplasty: who’s at risk? J Arthroplasty. 2014;29:256–260. doi: 10.1016/j.arth.2013.06.003. [DOI] [PubMed] [Google Scholar]
- 16.Schairer WW, Sing DC, Vail TP, Bozic KJ. Causes and frequency of unplanned hospital readmission after total hip arthroplasty. Clin Orthop Relat Res. 2014;472:464–470. doi: 10.1007/s11999-013-3121-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Tsang ST, Gaston P. Adverse peri-operative outcomes following elective total hip replacement in diabetes mellitus: a systematic review and meta-analysis of cohort studies. Bone Joint J. 2013;95:1474–1479. doi: 10.1302/0301-620X.95B11.31716. [DOI] [PubMed] [Google Scholar]
- 18.Zmistowski B, Restrepo C, Hess J, Adibi D, Cangoz S, Parvizi J. Unplanned readmission after total joint arthroplasty: rates, reasons, and risk factors. J Bone Joint Surg Am. 2013;95:1869–1876. doi: 10.2106/JBJS.L.00679. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.